|
PHARMACEUTICAL MANUFACTURING HANDBOOK
Regulations and Quality
SHAYNE COX GAD, PH.D., D.A.B.T.
Gad Consulting Services
Cary, North Carolina
A JOHN WILEY & SONS, INC., PUBLICATION
CONTENTS
SECTION 1 GOOD MANUFACTURING PRACTICES (GMP) AND
OTHER FDA GUIDELINES 1
1.1 Good Manufacturing Practices (GMPs) and Related FDA
Guidelines 3
James R. Harris
1.2 Enforcement of Current Good Manufacturing Practices 45
Kenneth J. Nolan
1.3 Scale-Up and Postapproval Changes (SUPAC) Regulations 67
Puneet Sharma, Srinivas Ganta, and Sanjay Garg
1.4 GMP-Compliant Propagation of Human Multipotent Mesenchymal
Stromal Cells 97
Eva Rohde, Katharina Schallmoser, Christina Bartmann, Andreas Reinisch,
and Dirk Strunk
SECTION 2 INTERNATIONAL REGULATIONS OF GOOD
MANUFACTURING PRACTICES 117
2.1 National GMP Regulations and Codes and International GMP
Guides and Guildelines: Correspondences and Differences 119
Marko Narhi and Katrina Nordstrom
x CONTENTS
SECTION 3 QUALITY 163
3.1 Analytical and Computational Methods and Examples for Designing
and Controlling Total Quality Management Pharmaceutical
Manufacturing Systems 165
Paul G. Ranky, Gregory N. Ranky, Richard G. Ranky, and Ashley John
3.2 Role of Quality Systems and Audits in Phatmaceutical
Manufacturing Environment 201
Evan B. Siegel and James M. Barquest
3.3 Creating and Managing a Quality Management System 239
Edward R. Arling, Michelle E. Dowling, and Paul A. Frankel
3.4 Quality Process Improvement 287
Jyh-hone Wang
SECTION 4 PROCESS ANALYTICAL TECHNOLOGY (PAT) 311
4.1 Case for Process Analytical Technology: Regulatory and Industrial
Perspectives 313
Robert P. Cogdill
4.2 Process Analytical Technology 353
Michel Ulmschneider and Yves Roggo
4.3 Chemical Imaging and Chemometrics: Useful Tools for Process
Analytical Technology 411
Yves Roggo and Michel Ulmschneider
SECTION 5 PERSONNEL 433
5.1 Personnel Training in Pharmaceutical Manufacturing 435
David A. Gallup, Katherine V. Domenick, and Marge Gillis
SECTION 6 CONTAMINATION AND CONTAMINATION
CONTROL 455
6.1 Origin of Contamination 457
Denise Bohrer
6.2 Quantitation of Markers for Gram-Negative and Gram-Positive
Endotoxins in Work Environment and as Contaminants in
Pharmaceutical Products Using Gas Chromatography–Tandem
Mass Spectrometry 533
Alvin Fox
6.3 Microbiology of Nonsterile Pharmaceutical Manufacturing 543
Ranga Velagaleti
CONTENTS xi
SECTION 7 DRUG STABILITY 557
7.1 Stability and Shelf Life of Pharmaceutical Products 559
Ranga Velagaleti
7.2 Drug Stability 583
Nazario D. Ramirez-Beltran, Harry Rodriguez, and L. Antonio Estevez
7.3 Effect of Packaging on Stability of Drugs and Drug Products 641
Emmanuel O. Akala
7.4 Pharmaceutical Product Stability 687
Andrew A. Webster
7.5 Alternative Accelerated Methods for Studying Drug Stability:
Variable-Parameter Kinetics 701
Giuseppe Alibrandi
SECTION 8 VALIDATION 725
8.1 Analytical Method Validation: Principles and Practices 727
Chung Chow Chan
8.2 Analytical Method Validation and Quality Assurance 743
Isabel Taverniers, Erik Van Bockstaele, and Marc De Loose
8.3 Validation of Laboratory Instruments 791
Herman Lam
8.4 Pharmaceutical Manufacturing Validation Principles 811
E. B. Souto T. Vasconcelos D. C. Ferreira, and B. Sarmento
INDEX 839
PREFACE
This Handbook of Manufacturing: Regulations and Quality focuses on all regulatory
aspects and requirements that govern how drugs are produced for evaluation (and,
later, sale to and use in) humans. The coverage ranges from what the issues are at
the early stages (when the amounts are small and the materials of limited sophistication)
up to until the issue is reproducibly and continuously making large volumes
of a highly sophisticated manufactured product. These 25 chapters cover the full
range from preformulation of a product (the early exploratory work that allows us
to understand how to formulate and deliver the drug) to identifi cation of sources
of contamination and assessment of stability.
The Handbook of Manufacturing: Regulations and Quality seeks to cover the
entire range of available approaches to satisfying the wide range of regulatory
requirements for making a highly defi ned product that constitutes a successful new
drug and how to do so in as effective and as effi cient a manner as possible.
Thanks to the persistent efforts of Michael Leventhal, these 25 chapters, which
are written by leading practitioners in each of these areas, provide coverage of the
primary approaches to the fundamental regulatory challenges that must be overcome
to manufacture successfully a deliverable and stable new drug.
GOOD MANUFACTURING PRACTICES
( GMP ) AND OTHER FDA GUIDELINES
SECTION 1
3
1.1
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
GOOD MANUFACTURING
PRACTICES ( GMP ) AND RELATED
FDA GUIDELINES
James R. Harris
James Harris Associates, Inc., Durham, North Carolina
Contents
1.1.1 FDA Regulations: Real and Imagined
1.1.2 21 CFR 210 and 211: Current Good Manufacturing Practice for Finished
Pharmaceuticals
1.1.3 Guidance for Industry: Quality Systems Approach to Pharmaceutical Current Good
Manufacturing Practice Regulations
1.1.3.1 CGMPS and the Concepts of Modern Quality Systems
1.1.3.2 Quality Systems Model
1.1.4 Guidance for Industry: PAT — Framework for Innovative Pharmaceutical Development,
Manufacturing, and Quality Assurance
1.1.4.1 PAT Framework
1.1.5 Guidance for Industry: Part 11. Electronic Records; Electronic Signatures — Scope and
Application
1.1.6 Guidance for Industry and FDA: Current Good Manufacturing Practice for Combination
Products
1.1.7 Guidance for Industry: Powder Blends and Finished Dosage Units — Stratifi ed In -
Process Dosage Unit Sampling and Assessment
1.1.7.1 Validation of Batch Powder Mix Homogeneity
1.1.7.2 Verifi cation of Manufacturing Criteria
1.1.8 Guidance for Industry: Immediate - Release Solid Oral Dosage Forms Scale - Up and
Postapproval Changes (SUPAC) — Chemistry, Manufacturing and Controls, In Vitro
Dissolution Testing, and In Vivo Bioequivalence Documentation
1.1.9 Other GMP - Related Guidance Documents
4 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
1.1.1 FDA REGULATIONS: REAL AND IMAGINED
A regulation is a law. In the United States, all federal laws have been arranged or
codifi ed in a manner that makes it easier to fi nd a specifi c law. The Code of Federal
Regulations (CFR) is a compilation of all federal laws published in the Federal
Register by the executive departments and agencies of the federal government. This
code is divided into 50 titles which represent broad areas of federal regulation. Each
title is further divided into chapters. The chapters are then subdivided into parts
covering specifi c regulatory areas. Changes and additions are fi rst published in the
Federal Register . Both the coded law and the Federal Register must be used to determine
the latest version of any rule. All food - and drug - related laws are contained
in Title 21 of the CFR. Each title of the CFR is updated annually. Title 21 is updated
as of April 1 of each year.
Because virtually all of the drug regulations are written to state what should be
done but do not tell how to do it, the Food and Drug Administration (FDA) also
publishes guidance documents. These documents are intended to provide precisely
what the name implies — guidance. In this context, guidance documents are not law
and do not bind the FDA or the public . Manufacturers are not required to use the
techniques or approaches appearing in the guidance document. In fact, FDA representatives
have repeatedly stated that the regulations were not written to suggest
how something should be done in order to encourage innovation. While following
the recommendations contained in the guidance documents will probably assure
acceptance (agency philosophy and interpretation may have changed since the guidance
document was published), other approaches are encouraged. No matter how
they choose to proceed, manufacturers should be prepared to show that their
methods achieve the desired results.
A method used by the FDA to “ fl oat ” new ideas is to discuss them at industry
gatherings such as FDA - sponsored seminars or meetings of industry groups such as
the Pharmaceutical Manufacturers Association (PMA), the Parenteral Drug Association
(PDA), and the International Society of Pharmaceutical Engineering (ISPE).
Again, it must be remembered that while these comments refl ect current FDA
thinking, they are simply thoughts and recommendations. They are not law.
Several industry groups also publish comments, guidelines, and so on, that put
forth current thinking of the group writing the document. These publications are
interesting and often bring out valuable information. However, it is important to
remember that these publications are not regulations or even offi cial guidance documents.
If a fi rm chooses to follow the recommendations of such documents, they are
probably following good advice. However, since the advice comes from a nonoffi cial
source, fi rms should still be prepared to defend their actions with good scientifi c
reasoning.
1.1.2 21 CFR 210 AND 211: CURRENT GOOD MANUFACTURING
PRACTICE FOR FINISHED PHARMACEUTICALS
Parts 210 and 211 of CFR Title 21 are the laws defi ning good manufacturing practices
for fi nished pharmaceutical products. All manufacturers must follow these
regulations in order to market their products in the United States. When a fi rm fi les
an application to market a product in the United States through a New Drug Application
(NDA), abbreviated NDA, (ANDA), Biological License Application (BLA),
CURRENT GOOD MANUFACTURING PRACTICE 5
or other product application, one of the last steps in approving the application is a
preapproval inspection of the manufacturing facility. A major purpose of this inspection
is to assure adherence to the GMP regulations. Preapproval inspections are a
part of every application approval. Thus, if a fi rm has 10 applications pending, it
should expect 10 inspections. The fact that the manufacturing facility has already
been inspected will not alter the need for another inspection.
The FDA also has the right to visit and inspect any manufacturing facility that
produces a product or products sold in the United States. Such inspections are unannounced.
A manufacturer must admit an inspector when he or she appears at that
facility and must do so without undue delay.
GMP requirements for manufacturers of pharmaceutical dosage forms are discussed
below. This information should not be considered to be an exact statement
of the law. We have attempted to show intent and, occasionally, add some comments
that will clarify how that particular regulation is interpreted. For precise wording of
a regulation, refer to the CFR and then check the Federal Register to determine if
there have been any changes since the last update.
General Provisions
1. This section pertains to the manufacture of drug products for humans or
animals.
2. These requirements will not be enforced for over - the - counter (OTC) drug products
if the products and all their ingredients are ordinarily marketed and considered
as human foods and which products may also fall within the legal defi nition
of drugs by virtue of their intended use.
Organization and Personnel
1. Responsibilities of quality control unit
(a) A quality control unit must be a part of the facility organization.
(b) This unit must be given responsibility and authority to approve or reject all
components, drug product containers, closures, process materials, packaging
material, labeling, and drug products, and the authority to review production
records.
(c) Adequate laboratory facilities for testing and approval or rejection of the
above listed materials must be available.
(d) The quality control unit is responsible for approving or rejecting all procedures
or specifi cations that impact on the identity, strength, quality, and purity
of the drug product.
(e) Responsibilities and procedures applicable to the quality control unit must
be written and these procedures must be followed.
2. Personnel qualifi cations
(a) Every person involved in the manufacture, processing, packing, or holding of
a drug product must have education, training, and experience that enable
that individual to perform their duties. Employees must be trained in the
particular operations that they perform and in Current GMPs (CGMPs). The
GMP training must be conducted by qualifi ed individuals and with suffi cient
frequency to assure that workers remain familiar with the requirements
applicable to them.
6 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
(b) Persons responsible for supervision must have the education, training, and
experience to perform their assigned functions in such a manner as to assure
that the drug product has the safety, identity, strength, quality, and potency
that it is represented to possess.
(c) There must be an adequate number of qualifi ed personnel to perform the
needed tasks.
3. Personnel responsibilities
(a) Personnel shall wear clean clothing appropriate for the duties they perform.
Protective apparel must be worn as necessary.
(b) Personnel shall practice good sanitation and health habits.
(c) Only personnel authorized by supervisory personnel shall enter those areas
designated as limited - access areas.
(d) Any worker considered to have an apparent illness or open lesions that may
adversely affect safety or quality of drug products shall be excluded from
direct contact with product, components, or containers.
4. Consultants that advise on the manufacture, processing, packing, or holding of
drug products must have suffi cient education, training, and experience to advise
on the subject for which they are retained. The manufacturer must maintain
records of name, address, and qualifi cations of any consultants and the type of
service they provide.
Buildings and Facilities
1. Design and construction features
(a) Buildings should be of suitable size, construction location to facilitate cleaning,
maintenance, and proper operations.
(b) Space should be adequate for the orderly placement of equipment and materials
to prevent mix - ups between different components, drug product containers
and closures, labeling, in - process materials, or drug products and to
prevent contamination.
(c) The movement of components and product through the building must be
designed to prevent contamination.
(d) Operations should be performed within specifi cally defi ned areas having
adequate control systems to prevent contamination or mix - ups during each
of the following procedures:
(i) Receipt, identifi cation, storage, and withholding from use of components,
drug product containers, closures, and labeling, pending the
appropriate sampling, testing, and release for manufacturing or
packaging.
(ii) Holding rejected materials listed in (a) above.
(iii) Storage of released components, drug product containers, closures, and
labeling.
(iv) Storage of in - process materials.
(v) Manufacturing and processing operations.
(vi) Packaging and labeling operations.
(vii) Quarantine storage before release of drug products.
(viii) Storage of drug products after release.
(ix) Control and laboratory operations.
CURRENT GOOD MANUFACTURING PRACTICE 7
(x) Aseptic processing, which includes:
(1) Floors, walls, and ceilings of smooth, hard surfaces that are easily
cleanable.
(2) Temperature and humidity controls.
(3) An air supply fi ltered through High - Effi ciency Particulate Air
(HEPA) fi lters under positive pressure regardless of whether fl ow
is laminar or nonlaminar.
(4) A system for monitoring environmental conditions.
(5) A system for cleaning and disinfecting the room and equipment to
produce aseptic conditions.
(6) A system for maintaining any equipment used to control the aseptic
conditions.
(e) Operations relating to the manufacture, processing, and packing of penicillin
must be performed in facilities separate from those used for other drug
products for humans. Note : For all purposes of these GMP regulations, the
FDA considers cephalosporins to be penicillin.
2. Adequate lighting should be provided in all areas.
3. Heating, ventilation, and air conditioning (HVAC)
(a) Adequate ventilation is required in all areas.
(b) Equipment for adequate control over air pressure, microorganisms, dust,
humidity, and temperature must be provided when appropriate for the manufacture,
processing, packing, or holding of a drug product.
(c) When appropriate, air supplied to production areas should be fi ltered to
avoid any possibility of contamination or cross - contamination.
(d) Air - handling systems for the manufacture, processing, and packing of penicillin
shall be completely separate from those for other drug products for
humans.
4. Plumbing
(a) Potable water should be supplied in a continuous positive - pressure system
free from defects that could contribute to contamination of any drug
product.
(b) Potable water must meet the standards prescribed in the Environmental
Protection Agency (EPA) Primary Drinking Water Regulations defi ned in
40 CFR Part 141.
(c) Drainage must be of adequate size. Where connected directly to a sewer, an
air break or other suitable mechanical device must be provided to prevent
back - siphonage.
5. Sewage, trash, and other refuse in and from the building and immediate premises
must be disposed of in a safe and sanitary manner.
6. Adequate washing facilities should be provided. This is to include hot and cold
water, soap or detergent, air driers or single - service towels, and clean toilet facilities
easily accessible to all work areas.
7. Sanitation
(a) Any building used for manufacture, processing, packing, or holding of a drug
product should be maintained in a clean and sanitary condition. Such buildings
should be free of infestation by rodents, birds, insects, and other vermin.
(b) Trash and organic waste matter should be held and disposed of in a timely
and sanitary manner.
8 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
(c) Written procedures assigning responsibility for sanitation and describing in
suffi cient detail the cleaning schedules, methods, equipment, and materials
to be used in cleaning the buildings and facilities are required. Such procedures
must be followed.
(d) Written procedures for use of suitable rodenticides, insecticides, fungicides,
fumigating agents, and cleaning and sanitizing agents are required and must
be followed. These written procedures should be designed to prevent the
contamination of equipment, components, product containers, closures, packaging,
labeling materials, or drug products. Agent may not be used unless
registered and used in accordance with the Federal Insecticide, Fungicide,
and Rodenticide Act (7 U.S.C. 135).
(e) All sanitation procedures apply equally to contractors or temporary employees
as to regular employees.
8. All buildings used for GMP - related purposes must be maintained in a good state
of repair.
Equipment
1. Equipment should be of appropriate design, adequate size, and suitably located
to facilitate operations for its intended use and for cleaning and maintenance.
2. Equipment construction
(a) Equipment should be constructed so that surfaces that contact components,
in - process materials, or drug products should not be reactive, additive, or
absorptive so as to alter the safety, identity, strength, quality, or purity of the
drug product beyond offi cial or other established requirements.
(b) Any substance required for operation such as lubricants or coolants shall not
come into contact with drug products, containers, and so on, so as to alter
the safety, identity, strength, quality, or purity of the drug product beyond
established requirements.
3. Equipment cleaning and maintenance
(a) Equipment and utensils should be cleaned, maintained, and sanitized at
appropriate intervals to prevent malfunctions or contamination that would
alter the drug product beyond the offi cial requirements.
(b) Written procedures must be established and followed for cleaning and
maintenance of equipment and utensils used in the processing of a drug
product. These procedures must include but are not limited to the
following:
(i) Assignment of responsibility for cleaning and maintaining equipment.
(ii) Maintenance and cleaning schedules, including sanitizing schedules if
appropriate.
(iii) A suffi ciently detailed description of the methods, equipment, and
materials used in cleaning and maintenance operations and the methods
of disassembling and reassembling equipment as a part of cleaning and
maintenance.
(iv) Removal or obliteration of previous batch identifi cation.
(v) Protection of clean equipment from contamination prior to use.
(vi) Inspection of equipment for cleanliness immediately before use.
CURRENT GOOD MANUFACTURING PRACTICE 9
(vii) Records should be kept of maintenance, cleaning, sanitizing, and inspection
of all processing equipment.
4. Automatic, mechanical, and electronic equipment
(a) All such equipment, including computers or related systems that will perform
a function to be used in any GMP - related activity, must be routinely calibrated,
inspected, or checked according to a written program designed to
assure proper performance. Written records must be maintained for all such
activities.
(b) Appropriate controls should be exercised to assure that changes in master
production and control records or other similar records are made only by
authorized personnel. Input to and output from such systems should be
checked for accuracy.
A backup fi le of data entered into a computer - related system must be
maintained except where certain data such as calculations performed in connection
with laboratory analysis are eliminated by computerization or other
automated processes. In this situation, a written record of the program should
be maintained along with validation data.
5. Filters for liquid fi ltration used as a part of the manufacture, processing, or
packing of injectable drug products intended for human use must not release
fi bers into such products. Fiber - releasing fi lters may not be used unless it is not
possible to manufacture the product without the use of such a fi lter. In this situation,
an additional non - fi ber - releasing fi lter of 0.22 . m maximum must be used
after the fi ber - releasing fi ltration. Use of an asbestos - containing fi lter is permissible
only upon submission of proof to the appropriate FDA bureau that use of
a non - fi ber - releasing fi lter will compromise the safety or effectiveness of the drug
product.
Control of Components and Drug Product Containers and Closures
1. General requirements
(a) There must be written procedures describing in suffi cient detail the receipt,
identifi cation, storage, handling, sampling, testing, and approval or rejection
of product components, containers, and closures. Of course, all such procedures
must be followed. It is quite common and even more embarrassing to
be cited for not following your own written procedures. Note: For the rest of
this discussion, the term components will mean product ingredients, containers,
closures, and so on.
(b) All components listed above must be handled and stored in a manner that
will prevent contamination.
(c) Bagged or boxed components should be stored off the fl oor. Spacing should
allow cleaning and inspection.
(d) Every container of components must be identifi ed with a distinctive code or
lot number for each receival of that product. Even if the next receival is the
same vendor lot number, it must be a new identifying number by the pharmaceutical
manufacturer. Each lot must be appropriately identifi ed as to its
status (quarantined, approved, or rejected).
10 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
2. Receipt and storage of untested components
(a) Upon receipt each container of components must be visually examined for
appropriate labeling and any damage or contamination to the component
container.
(b) Components must be stored under quarantine until they have been tested as
appropriate and released for use.
3. Testing and approval or rejection of components
(a) Each lot of components shall be withheld from use until it has been sampled,
tested, and released by the quality control unit.
(b) Representative samples must be taken from every receival of every component.
The number or amount of component to be sampled should be based
on component appearance, statistical confi dence levels, the past history of
the supplier, and the quantity needed to analyze and reserve samples if
required.
(c) Sampling procedures
(i) The component containers should be cleaned where necessary.
(ii) The containers should be opened, sampled, and resealed in a manner
designed to prevent contamination of the sample and remaining contents
of the container.
(iii) If appropriate, sterile equipment and aseptic sampling techniques should
be used.
(iv) Where sampling is done from various parts of a container, samples
should not be composited for testing.
(v) Containers from which samples have been taken must be marked to
show that samples have been removed.
(d) Examination and testing of samples
(i) At least one test should be conducted on each lot of component drug
product to verify identity.
(ii) Each component must be tested for conformity with all appropriate
written specifi cations for purity, strength, and quality if an ingredient or
for conformity with written specifi cations for containers or closures.
(iii) In lieu of the above testing by the manufacturer, a report of analysis
may be accepted from the supplier provided that at least one specifi c
identity test is conducted on the component by the manufacturer and
provided that the manufacturer has established the reliability of the
supplier ’ s analyses through appropriate validation.
(iv) When appropriate, components should be examined microscopically.
(v) Each lot of a component that is liable to contamination with dirt, insect
infestation, or other extraneous adulterant should be examined against
established specifi cations for such contamination.
(vi) Each lot of a component that is subject to microbial contamination that
is contrary to its intended use should be subjected to microbiological
tests before use.
(e) If a lot of components meets the written specifi cations, it may be approved
and released for use. Any lot of such material that does not meet such speci-
fi cations must be rejected.
4. Use of approved components (including drug product containers and closures)
must be rotated to assure that the oldest approved stock is used fi rst.
CURRENT GOOD MANUFACTURING PRACTICE 11
5. Components must be retested and/or reexamined after storage for a long period
of time or after exposure to the atmosphere, heat, or other condition that might
adversely affect the component.
6. Rejected components should be identifi ed and controlled under a quarantine
system designed to prevent their use in manufacturing or processing.
7. Containers and closures
(a) Containers and closures must not be reactive, additive, or absorbent so as to
alter the drug beyond established acceptance criteria.
(b) Container closure systems must provide adequate protection against foreseeable
external factors in storage that can cause deterioration or contamination
of the product.
(c) Containers and closures should be clean and, if necessary, sterile and processed
to remove pyrogens.
(d) Standards or specifi cation, methods of testing, and, if appropriate, sterilization
and depyrogenation must be written and followed.
Production and Process Controls
1. Written procedures and procedure deviations
(a) Written procedures for production and process control must be written and
followed. These procedures should be designed to assure that the drug products
have the identity, strength, quality, and purity they are represented to
possess. These procedures must include all requirements given below and
must be drafted, reviewed, and approved by the affected organizational units
and reviewed and approved by the quality control unit.
(b) When following the above identifi ed procedures, all actions must be documented
at the time of performance. Any deviations from the written procedure
must be recorded and justifi ed.
2. Charge - in of components — Written production and control procedures must
include the following, which are designed to assure that the drug products produced
meet all specifi cations and standards.
(a) The batch must be formulated with the intent to provide not less than 100%
of the labeled amount of active ingredient.
(b) Components used must be weighed, measured, or subdivided appropriately.
If a component is removed from its original container and placed in
another, the new container should be identifi ed with the following
information:
(i) Component name and/or item code.
(ii) Receiving or control number.
(iii) Weight or measure of material in the new container.
(iv) Batch or lot number for which the component was dispensed, including
its product name, strength, and lot number.
(c) Weighing, measuring, or subdividing operations for all components must be
adequately supervised. Each container of component dispensed to manufacturing
must be examined by a second person to assure that:
(i) The component was released by the quality control unit.
(ii) The weight or measure is correct as stated in the batch production
records.
12 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
(iii) The containers are properly identifi ed and contain the quantity stated
on the label.
(d) Addition of each component must be performed by one person and verifi ed
by a second person.
3. Actual yield and percentage of theoretical yield should be determined at the
completion of each appropriate phase of manufacturing, processing, packaging,
or holding. These calculations should be performed by one person and independently
verifi ed by a second individual.
4. Equipment identifi cation
(a) All compounding and storage containers, processing lines, and major equipment
used during the production of a batch of a drug product must be properly
identifi ed at all times to indicate their contents and the phase of processing
of the batch.
(b) Major equipment should be identifi ed by a distinctive identifi cation that shall
be recorded in the batch production record to indicate the specifi c equipment
used. In cases where only one of a particular type of equipment exists in a
given manufacturing facility, the name of the equipment may be used instead
of creating a distinctive identifi cation.
5. Sampling and testing of in - process materials and drug products
(a) To assure batch uniformity and integrity, it is necessary to write and follow
procedures that describe the in - process controls and tests or examinations
that will be conducted on samples taken according to procedure. Procedures
should be written to monitor the output and to validate the performance of
those manufacturing processes that may be responsible for causing variability
in the product being manufactured. These control procedures should
include but are not limited to the following:
(i) Tablet or capsule weight variation.
(ii) Disintegraton time.
(iii) Adequacy of mixing or blending to assure uniformity and
homogeneity.
(iv) Dissolution time and rate.
(v) Clarity of solutions.
(vi) pH of solutions.
(b) In - process specifi cations for all characteristics must be consistent with the
drug product fi nal specifi cations and must be developed from previous
acceptable product average and process variability data.
(c) In - process materials should be tested for identity, strength, quality, and purity
as appropriate. As a part of the production process, they must be approved
for continued use or rejected by the quality control unit before production
continues.
(d) Rejected in - process materials must be identifi ed and controlled under a
quarantine system designed to prevent their use in manufacturing operations
for which they have been found to be unsuitable.
6. When appropriate, time limits should be established for the completion of each
phase of production. The purpose of this is to assure the quality of the drug
product. Deviation from the established time limits may be acceptable if this
deviation does not compromise the quality of the product. Any deviation must
be documented, including the justifi cation for such deviation.
CURRENT GOOD MANUFACTURING PRACTICE 13
7. Control of microbial contamination
(a) To prevent the growth of objectionable microorganisms in products not
required to be sterile, appropriate written procedures designed to prevent
such growth should be written and followed.
(b) If sterilization is a part of any procedure described in (a) above, this procedure
must be validated.
8. Reprocessing
(a) Written procedures describing any system used to reprocess batches that do
not conform to the established standards must be written and followed.
(b) Reprocessing must not be performed without the review and approval of the
quality control unit.
Packaging and Labeling Control
1. Materials examination and usage criteria
(a) Written procedures describing in detail the receipt, identifi cation, storage,
handling, sampling, examination, and/or testing of labeling and packaging
materials must be developed, approved, and followed. These materials must
be representatively sampled, examined, or tested on receipt and accepted by
the quality control unit before use.
(b) Any materials that do not fully meet acceptance criteria must be rejected to
prevent their use.
(c) Records of each receival of each different label and packaging material must
be maintained indicating receipt, examination or testing, and whether
accepted or rejected.
(d) Labels and other labeling materials for each different drug product, strength,
dosage form, or quantity of contents must be stored separately with suitable
identifi cation. Access to the storage area must be limited to authorized
personnel.
(e) Obsolete and outdated labels, labeling, and other packaging materials must
be quarantined and destroyed.
(f) The use of gang - printed labels for different drug products or different
strengths or different net contents is prohibited. The only exception to this
rule is if labels from gang - printed sheets are adequately differentiated by
size, shape, or color that will prevent mixing of labels.
(g) If cut labeling is used, packaging and labeling operations must include one
or more of the following special control procedures:
(i) Dedication of a labeling and packaging line to each different strength
of each different drug product.
(ii) Use of appropriate electronic or electromechanical equipment to
conduct a 100% examination for correct labeling during or after completion
of the fi nishing operation.
(iii) Use of visual inspection to conduct a 100% examination for correct
labeling. If visual inspection is used, the inspection should be performed
by one person and independently verifi ed by a second individual.
(h) Printing devices on or associated with the manufacturing line used to imprint
labeling upon the drug product unit label or case must be monitored to assure
14 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
that the printing conforms to the print specifi ed in the batch production
record.
2. Issuance of labeling
(a) Strict control should be exercised over the issuance of labeling for use in
drug product labeling operations.
(b) Labeling materials issued for a batch must be carefully examined for identity
and conformity to the labeling specifi ed in the batch production record.
(c) Procedures should be written and followed for reconciliation of the quantities
of labeling issued, used, destroyed, and returned. Procedures should
require evaluation of discrepancies found between the number of packages
fi nished and the amount of labeling issued if discrepancies outside narrow
preset limits occur. Limits should be established on the basis of historical
operating data. Labeling reconciliation is waived for either cut or roll labeling
if a 100% examination for correct labeling is performed.
(d) All excess labeling bearing a lot or control number must be destroyed.
(e) Returned labeling should be maintained and stored in a manner to prevent
mix - ups.
(f) Written procedures should describe the control procedures used for the issuance
of labeling.
3. There must be written procedures designed to assure that correct labels, labeling,
and packaging materials are used. These procedures should incorporate the following
features:
(a) Prevention of mix - ups and cross - contamination by physical or spatial separation
of operations on other drug products.
(b) Identifi cation and handling of fi lled drug product containers that are set
aside and held in unlabeled condition for future labeling operations. Such
procedures should be designed to prevent mislabeling individual containers,
lots, or portions of lots. It is not necessary to apply identifi cation to each
individual container, but the procedure should be adequate to determine the
name, strength, quantity of contents, and lot or control number of each
container.
(c) Identifi cation of the drug product with a lot or control number that permits
determination of the history of the manufacture and control of the batch.
(d) Examination of packaging and labeling materials for suitability and correctness
before issuing for use and before packaging operations. These examinations
must be documented in the batch production record.
(e) Inspection of the packaging and labeling facility immediately before use to
assure that all drug products and labeling materials from the previous operation
have been removed. Inspection results must be documented in the batch
production record.
4. Tamper - evident packaging requirements for OTC human drug products
(a) An OTC product (with the exception of a dermatological, dentifrice, insulin,
or lozenge product) intended for retail sale is considered adulterated or
misbranded or both if it is not packaged in a tamper - resistant package.
(b) Requirements for a tamper - evident package
(i) With the exceptions listed above, all OTC products must be packaged
in a tamper - evident package if the product is accessible to the
public while being held for sale. A tamper - evident package must have
CURRENT GOOD MANUFACTURING PRACTICE 15
one or more indicators or barriers to entry which, if breached or missing,
can reasonably be expected to provide visible evidence to consumers
that tampering has occurred: A tamper - evident package may involve an
immediate container and closure system or a secondary container or
carton system or a combination of systems intended to provide a visual
indication of package integrity. The tamper - evident feature must be
designed to and shall remain intact when handled in a reasonable
manner during manufacture, distribution, and retail display.
(ii) In addition to the tamper - evident packaging feature described above,
any two - piece hard gelatin capsule covered by this regulation must be
produced using an acceptable tamper - evident technology.
(c) Labeling
(i) In order to alert consumers to the specifi c tamper - evident features used,
each retained package of an OTC drug product covered by this regulation
is required to bear a statement that:
(1) Identifi es all tamper - evident features and any capsule - sealing
technologies.
(2) Is prominently placed on the package.
(3) Is so placed that it will be unaffected if the tamper - evident feature
of the package is breached or missing.
(ii) If the tamper - evident feature chosen to meet the requirement uses an
identifying characteristic, that characteristic is required to be referred
to in the labeling statement. For example, the labeling statement on a
bottle with a shrink band could say For your protection, this bottle has
an imprinted seal around the neck .
(d) A manufacturer or packer may request an exemption from the tamper -
evident requirement. A request for exemption is required to be submitted in
the form of a petition and should be clearly identifi ed on the envelope as a
“ Request for Exemption from the Tamper - Evident Packaging Rule. ” This
petition is required to contain the following:
(i) The name of the drug product or, if the petition seeks an exemption for
a drug class, the name of the drug class and a list of products within that
class.
(ii) The reasons that the drug product ’ s compliance with the tamper - evident
packaging and labeling requirements is unnecessary or cannot be
achieved.
(iii) A description of alternative steps that are available or that the petitioner
has already taken to reduce the likelihood that the product or
drug class will be the subject of malicious adulteration.
(iv) Other information justifying an exemption.
(e) Holders of approved new drug applications for OTC drug products are
required to provide the FDA with notifi cation of changes in packaging and
labeling to comply with the requirements of this section. Changes in packaging
and labeling required by the regulation may be made before FDA
approval. Manufacturing changes by which capsules are to be sealed require
prior FDA approval.
(f) This section does not affect any requirements for “ special packaging ” as
required under the Poison Prevention Packaging Act of 1970.
16 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
5. Drug product inspection
(a) Packaged and labeled products must be examined during fi nishing operations
to provide assurance that containers and packages in the lot have the
correct label.
(b) A representative sample of units should be collected at the completion of
fi nishing operations and should be visually examined for correct labeling.
(c) Results of these examinations must be recorded in the batch production
records.
6. Expiration dating
(a) All packaged drug products must carry an expiration date that has been
determined from appropriate stability testing.
(b) Expiration dates must be related to the recommended storage conditions
stated on the label as determined by stability studies.
(c) If the drug product is to be reconstituted at the time of dispensing, its label
must carry expiration information for both the reconstituted and unreconstituted
forms.
(d) Expiration dates must appear on labeling in accordance with the requirements
stated elsewhere in this regulation.
(e) Homeopathic drug products are exempt from the requirements of this
section.
(f) Allergenic extracts that are labeled “ No U.S. Standard of Potency ” are
exempt.
(g) New drug products for investigational use are exempt provided that they
meet appropriate standards or specifi cations as demonstrated by stability
studies during their use in clinical investigations. If new drug products for
investigational use are to be reconstituted at the time of dispensing, their
labeling must bear expiration information for the reconstituted product.
(h) Pending consideration of a proposed exemption published in the Federal
Register , September 29, 1978, the requirements in this section will not be
enforced for human drug products if their labeling does not bear dosage
limitations and they are stable at least three years as supported by stability
data.
Holding and Distribution
1. Warehousing procedures
(a) Written procedures describing the warehousing of drug products must be
written and followed. These procedures should include:
(i) Quarantine of drug products before release by the quality control
unit.
(ii) Storage of drug products under appropriate conditions of temperature,
humidity, and light so that the quality of the drug products is not affected.
2. Distribution procedures
(a) Written procedures concerning the distribution of drug products must be
established and followed. These procedures should include:
(i) A procedure that assures the distribution of the oldest approved stock
fi rst. Deviation from this procedure is acceptable if it is temporary and
appropriate.
CURRENT GOOD MANUFACTURING PRACTICE 17
(ii) A system for documenting distribution so that distribution of each lot
of drug product can be readily determined to facilitate its recall if
required.
Laboratory Controls
1. General requirements
(a) The establishment of any specifi cations, standards, sampling plans, test processes,
or other laboratory control mechanism required by this part of the
regulation, including any changes to the above must be drafted by the appropriate
organizational unit and reviewed and approved by the quality control
unit. All actions must be documented at the time of performance and any
deviation must be recorded and justifi ed.
(b) Laboratory controls must include the establishment of scientifi cally
sound and appropriate specifi cations, standards, sampling plans, and test
procedures designed to assure that all materials conform to appropriate
standards of identity, strength, quality, and purity. Laboratory controls should
include:
(i) Determination of conformance to written specifi cations for the acceptance
of each lot within each shipment of raw materials. The specifi cations
should include a description of the sampling and testing procedures
used. Samples must be representative and adequately identifi ed. These
procedures must also require appropriate retesting of any material that
is subject to deterioration.
(ii) Determination of conformance to written specifi cations and a description
of sampling and testing procedures for in - process materials.
(iii) The calibration of instruments, apparatus, gauges, and recording devices
at specifi ed intervals in accordance with an established written program
containing specifi c directions, schedules, limits for accuracy and precision,
and provisions for remedial action in the event that the limits are
not met. Any such devices that do not meet the established specifi cations
must not be used.
2. Testing and release for distribution
(a) Laboratory testing of each lot of drug product must be conducted to establish
conformance to fi nal specifi cations for the product. Testing must include
identity and strength of each active ingredient. Where sterility and/or pyrogen
testing are required on short - lived radiopharmaceuticals, batches may be
released prior to completion of this testing provided that such testing is
completed as soon as possible.
(b) Each batch of product required to be free of objectionable microorganisms
must be tested appropriately.
(c) All sampling and testing plans must be described in written procedures that
include the method of sampling and the number of units to be tested.
(d) Acceptance criteria for the sampling and testing conducted by the quality
control unit must be adequate to assure that the batch being tested meets all
specifi cations. Appropriate statistical quality control criteria should be used.
The statistical quality control criteria must include acceptance levels and/or
rejection levels.
18 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
(e) The accuracy, sensitivity, specifi city, and reproducibility of test methods used
must be established and documented. Validation and documentation must
be accomplished in accordance with this regulation.
(f) Drug products failing to meet established standards or specifi cations and any
relevant quality control criteria must be rejected. Reprocessing may be performed,
however, prior to acceptance and use, and reprocessed material must
meet all standards, specifi cations, and other relevant criteria.
3. Stability testing
(a) There must be a written testing program designed to assess the stability
characteristics of every drug product. The results of such testing must be used
to determine appropriate storage conditions and expiration dates. The written
program must include:
(i) Sample size and test intervals based on statistical criteria for each attribute
examined.
(ii) Storage conditions for sampled retained for testing.
(iii) Reliable, meaningful and specifi c test methods.
(iv) Testing of the product in the same container - closure system as the one
in which the product is to be marketed.
(v) Testing of drug products for reconstitution at the time of dispensing as
well as after they are reconstituted.
(b) An adequate number of batches of each drug product must be tested to
determine appropriate expiration date. A record of such data must be maintained.
Accelerated studies, combined with basic stability information on the
components and drug product in its container - closure system may be used
to project a tentative expiration date that is beyond the date supported by
shelf life studies. However, there must be stability studies conducted including
drug product testing at appropriate intervals until the tentative expiration
date is verifi ed.
(c) The requirements for homeopathic drug products are as follows:
(i) There must be a written assessment of stability based on testing or
examination of the drug product for compatibility of the ingredients,
and based on marketing experience with the drug product to indicate
that there is no degradation of the product for the normal or expected
period of use.
(ii) Evaluation of stability must be based on the same container - closure
system as the one in which the drug product is to be marketed.
(d) Allergenic extracts that are labeled “ No U.S. Standard of Potency ” are exempt
from the requirements of this section.
4. Special testing requirements
(a) For each batch of drug product claimed to be sterile and/or pyrogen
free, there must be appropriate laboratory testing to establish conformance
to this claim. The test procedures must be in writing and must be followed.
(b) For each batch of ophthalmic ointment, there must be appropriate testing to
determine conformance to specifi cations regarding the presence of foreign
particles and harsh or abrasive substances. The test procedures must be in
writing and must be followed.
(c) For each batch of controlled - release dosage form, there must be appropriate
laboratory testing to determine conformance to the specifi cations for the rate
CURRENT GOOD MANUFACTURING PRACTICE 19
of release of each active ingredient. The test procedures must be in writing
and must be followed.
5. Reserve samples
(a) An identifi ed reserve sample that is representative of each lot or of each
shipment of each active ingredient must be retained. This reserve sample
should contain at least twice the quantity needed for all tests required to
determine whether the active ingredient meets its established specifi cations
with the exception of sterility and pyrogen testing. The required retention
time is as follows:
(i) For an active ingredient in a drug product other than those described
in paragraphs (b) and (c) below, the reserve sample must be retained
for one year after the expiration date of the last lot of drug product
containing that lot of active ingredient.
(b) For an active ingredient in a radioactive drug product except for nonradioactive
reagent kits, the reserve sample must be retained for:
(i) Three months after the expiration date of the last lot of the drug product
containing that lot of active ingredient if the expiration dating period
of the drug product is 30 days or less.
(ii) Six months after the expiration date of the last lot of the drug product
containing that lot of active ingredient if the expiration dating period
of the drug product is more than 30 days.
(c) For an active ingredient in an OTC drug product that is exempt from bearing
an expiration date, the reserve sample must be retained for three years after
distribution of the last lot of drug product containing that lot of active
ingredient.
(d) A properly identifi ed reserve sample that is representative of each batch
of drug product must be retained and stored under conditions consistent
with the product labeling. The reserve sample must be stored in the
same immediate container closure system in which the drug product is
marketed or in one that has essentially the same characteristics. The
reserve sample consists of at least twice the quantity needed to perform
all the required tests except those for sterility and pyrogens. Reserve samples
from representative sample lots or batches selected by acceptable statistical
procedures must be examined visually at least once a year for evidence of
deterioration unless visual examination would affect the integrity of
the reserve sample. Any evidence of reserve sample deterioration must be
investigated. The results of the examination must be recorded and maintained
with stability data concerning that drug product. Retention times are
as follows:
(i) For a drug product other than the exceptions noted above, the reserve
sample must be retained for one year after the expiration date of the
drug product.
(ii) For a radioactive drug product, except for nonradioactive reagent kits,
the retention sample must be retained for:
(1) three months after the expiration date of the drug product if the
expiration date is 30 days or less or
(2) six months after the expiration date of the drug product if the expiration
date is more than 30 days.
20 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
(iii) For an OTC drug product that is exempt from bearing an expiration
date, the reserve sample must be retained for three years after the batch
of drug product is fully distributed.
6. Animals used in testing components, in - process materials, or drug products for
compliance with established specifi cations must be maintained and controlled in
a manner that assures their suitability for their intended use. They must be identi-
fi ed and adequate records must be maintained showing the history of their use.
7. If a reasonable possibility exists that a nonpenicillin drug product has been
exposed to cross - contamination with penicillin, the nonpenicillin drug product
must be tested for the presence of penicillin. The drug product may not be
marketed if a detectable level of penicillin is found when tested according to
procedures specifi ed in “ Procedures for Detecting and Measuring Penicillin contamination
in Drugs ” which is incorporated in the regulation by reference.
Records and Reports
1. General Requirements
(a) Any production, control, or distribution record that is associated with a batch
of a drug must be retained for at least one year after the expiration date of
the batch OR, for OTC drug products that do not have expiration dates, three
years after complete distribution of the batch.
(b) Records must be retained for all components, containers, closures, and labeling
for the same time periods shown in (a) above.
(c) All retained records or copies of these records must be readily available for
authorized inspection at any time in the required retention period. Records
must be available for inspection where the activities described therein
occurred. Photocopying or similar reproduction by investigators must be
permitted.
(d) Retained records may be original records or true copies such as photocopies,
microfi lm, microfi che, or other accurate reproduction of the original.
(e) Written records that must be retained must be maintained so that data contained
therein can be used for evaluating the quality standards of each drug
product to determine the need for changes in drug product specifi cations or
manufacturing or control procedures. Such reviews should be conducted at
least annually. Written procedures must be established and followed for these
evaluations and must include provisions for:
(i) A review of a representative number of batches, whether approved or
rejected, and records associated with the batch.
(ii) A review of complaints, recalls, returned or salvaged drug products, and
investigations conducted under Section 211.192 of the GMP regulations
for each drug product.
(f) Procedures must be established to assure that the responsible offi cials of the
fi rm are notifi ed in writing of any investigations conducted under Sections
211.198, 211.204, or 211.208 of any recalls, reports of inspectional observations
issued by the FDA, or any regulatory actions relating to GMP brought
by the FDA.
2. A written record of major equipment cleaning, maintenance (except routine
maintenance), and use must be included in individual equipment logs that show
CURRENT GOOD MANUFACTURING PRACTICE 21
the date, time, product, and lot number of each batch processed. The persons
performing and double checking the cleaning and maintenance should date and
sign or initial the log indicating that the work was performed. Entries in the log
must be in chronological order.
3. Component, drug product container, closure, and labeling records must include
the following:
(a) The identity and quantity of each shipment of each lot of components, drug
product containers, closures, and labeling. Also required are the identity of
the supplier, the supplier ’ s lot number(s), the receiving code, the date of
receipt, and name and location of the prime manufacturer if different from
the supplier.
(b) The results of any test or examination performed and any conclusions derived
from these results.
(c) An individual inventory record of each component and a reconciliation of
the use of each lot of such component. The inventory record must contain
suffi cient information to allow determination of any batch or lot of drug
product associated with the use of each component.
(d) Documentation of the examination and review of labels and labeling for
conformance with established specifi cations.
(e) The disposition of rejected materials.
4. Master production and control records
Batch production and control records should be prepared for each batch of
drug product produced and must include complete information about the production
and control of that batch. These records must include:
(a) A full and complete reproduction of the appropriate master production or
control record. The copy must be checked for accuracy, dated, and signed.
(b) Documentation that each signifi cant step in the manufacture, processing,
packaging, and holding of the batch was accomplished as prescribed,
including:
(i) Dates.
(ii) Identity of individual major equipment used. This includes packaging
lines.
(iii) Complete and specifi c identifi cation of each batch of component or
in - process material used.
(iv) Weight and measures of components used in the course of
processing.
(v) In - process and laboratory control results.
(vi) Inspection of the packaging and labeling area before and after use.
(vii) Documentation of the actual yield and the percentage of theoretical
yield that this represents at critical stages of processing.
(viii) Complete labeling control records, including specimens or copies of all
labeling used.
(ix) A description of drug product containers and closures.
(x) Any sampling performed.
(xi) Identifi cation of the persons performing and directly supervising or
checking signifi cant steps in the operation.
(xii) Any investigations conducted.
(xiii) Results of examinations made.
22 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
5. All drug product production and control records, including those for packaging
and labeling, must be reviewed and approved by the quality control unit to determine
compliance with all established written procedures before a batch is released
or distributed. Any unexplained discrepancy or the failure of a batch or any of
its components to meet any of the established specifi cations must be thoroughly
investigated. The investigation must be extended to other batches of the same
drug product and other drug products that may have been associated with the
specifi c fault or discrepancy. A written record of the investigation must be made
and include the conclusions and any required follow - up.
6. Laboratory records
(a) Laboratory records must include complete data derived from all tests needed
to assure compliance with established specifi cations and standards. This
includes examinations and assays as follows:
(i) A description of the sample received for testing with identifi cation of
source. For example, location where the sample was obtained, quantity,
lot number or other distinctive code, date the sample was taken, and
the date that it was received for testing.
(ii) A statement of each method used in the testing of the sample. The
statement must indicate the location of data that establish that the
methods used in the testing of the sample meet proper standards of
accuracy and reliability as applied to the product tested. (If the method
used is in the current revision of the U.S. Pharmacopeia (USP), National
Formulary (NF), or other recognized standard reference or if it is
detailed in an approved NDA, this statement will not be required.)
(iii) A statement of the weight or measure of sample used for each test.
(iv) A complete record of all data secured in the course of each test, including
all graphs, charts, and spectra from laboratory instrumentation
properly identifi ed to the specifi c component and lot tested.
(v) A record of all calculations performed in connection with the test,
including units of measure, conversion factors, and equivalency
factors.
(vi) A statement of the results of tests and how the results compare with
established standards of identity, strength, quality, and purity for the
component tested.
(vii) The initials or signature of the person who performed each test and
the date the tests were performed.
(viii) The initials or signature of a second person showing that the original
records have been reviewed for accuracy, completeness, and compliance
with established standards.
(b) Complete records must be maintained of any modifi cation of an established
method employed in testing. These records must include the reason for the
modifi cation and verify that the modifi cation produced results that are at
least as accurate and reliable for the material being tested as the established
method.
(c) Complete records must be maintained of any testing and standardization of
laboratory reference standards, reagents, and standard solutions.
(d) Complete records must be maintained of the periodic calibration of laboratory
instruments, apparatus, gauges, and recording devices.
CURRENT GOOD MANUFACTURING PRACTICE 23
(e) Complete records must be maintained of all stability testing performed in
accordance with Section 211.166 of the regulation.
7. Distribution records must contain the name and strength of the product and
description of the dosage form, name and address of the consignee, date and
quantity shipped, and lot or control number of drug product. For compressed
medical gas products, distribution records are not required to contain lot or
control numbers.
8. Complaint fi les
(a) Written procedures describing the handling of all written and oral complaints
regarding a drug product must be established and followed. These procedures
must include provisions for review by the quality control unit of any complaint
involving the possible failure of a drug product to meet any of its
specifi cations and a determination as to the need for an investigation. These
procedures must include provisions for review to determine whether the
complaint represents a serious and unexpected adverse drug experience
which is required to be reported to the FDA.
(b) A written record of each complaint must be maintained in a fi le designated
for product complaints. The fi le may be maintained at another facility if the
written records of such fi les are readily available for inspection at that other
facility. Written reports involving a drug product must be maintained until
at least one year after the expiration date of the drug product or one year
after the date that the complaint was received, whichever is longer. In the
case of certain OTC drug products lacking expiration dating because they
meet the criteria for exemption, such written records must be maintained for
three years after distribution of the drug product.
(i) The written record must include the following information
where known: the name and strength of the drug product, lot number,
name of complainant, nature or complaint, and reply to the
complainant.
(ii) Where an investigation is conducted, the written record must include
the fi ndings of the investigation and follow - up. The record or a copy
of the record of investigation must be maintained at the location where
the investigation occurred.
(iii) Where an investigation is not conducted, the written record must include
the reason that an investigation was not considered to be necessary and
the name of the responsible person making the determination.
Returned and Salvaged Drug Products
1. Returned drug products — Returned drug products must be identifi ed as such
and held. If the conditions under which returned drug products have been held,
stored, or shipped before or during the return or the condition of the drug product,
its container, carton, or labeling is a result of storage or shipping casts doubt on the
safety, identity, strength, quality, or purity of the drug product, the returned drug
product must be destroyed unless examination testing or other investigation proves
the drug product meets appropriate standards. Records of returned drug products
must be maintained and must include the name and label potency of the drug
24 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
product dosage lot number, reason for the return, quantity returned, date of disposition,
and ultimate disposition of the returned product. If the reason for a drug
product being returned implicates associated batches, an investigation must be
conducted. Procedures for the holding, testing, and reprocessing of returned drug
products must be in writing and must be followed.
2. Drug product salvaging — Drug products that have been subjected to improper
storage conditions, including extremes in temperature, humidity, smoke, fumes, pressure,
age, or radiation due to natural disasters, fi res, accidents, or equipment failures,
must not be salvaged and returned to the marketplace. Whenever there is a question
whether drug products have been subjected to such conditions, salvaging operations
may be conducted only if there is (a) evidence from laboratory tests and assays that
the drug products meet all applicable standards of identity, strength, quality, and
purity and (b) evidence from inspection of the premises that the drug products and
associated packaging were not subjected to improper storage conditions as a result
of the disaster or accident. Organoleptic examinations are acceptable only as supplemental
evidence that the drug products meet appropriate standards of identity,
strength, quality, and purity. Records including name, lot number, and disposition
must be maintained for drug products subject to this section.
1.1.3 GUIDANCE FOR INDUSTRY: QUALITY SYSTEMS APPROACH
TO PHARMACEUTICAL CURRENT GOOD MANUFACTURING
PRACTICE REGULATIONS
This guidance document was written by the FDA to help manufacturers implement
what they consider to be modern quality systems and risk management approaches
that will meet the requirements of the FDA ’ s GMP regulations. The guidance
describes what the FDA considers a comprehensive quality systems (QS) model. It
also explains how manufacturers can be in full compliance with the GMP regulations
by implementing such quality systems. The FDA does not intend this guidance
to place new expectations on manufacturers nor does this replace the GMPs.
As is true with all guidance documents, this document does not establish legally
enforceable responsibilities, but rather it describes the FDA ’ s current thinking. Thus,
this guidance should be viewed as a set of recommendations unless a regulation is
cited.
The objective of this guidance is to describe a quality systems model and demonstrate
how and where the elements of this model can fi t within the requirements
of the CGMP regulations. The philosophy being put forward is that quality should
be build into the product, and testing alone cannot be relied on to ensure product
quality .
1.1.3.1 CGMPS and the Concepts of Modern Quality Systems
The FDA believes that several key concepts are critical for any discussion of modern
quality systems. The following concepts are used throughout this guidance as they
relate to the manufacture of pharmaceutical dosage forms:
CURRENT GOOD MANUFACTURING PRACTICE 25
Quality For the purposes of this guidance, the phrase achieving quality means
achieving the identity, strength, purity, and other quality characteristics
designed to ensure safety and effectiveness.
Quality by Design and Product Development This means designing and developing
a product and its associated manufacturing processes that will be used
to ensure that the product consistently attains a predefi ned quality at the end
of the manufacturing process.
Quality Risk Management This component of a quality systems framework can
help guide the setting of specifi cations and process parameters for dosage form
manufacturing, assess and mitigate the risk of changing a process or specifi cation,
and determine the extent of discrepancy investigations and corrective
actions.
Corrective and Preventative Action (CAPA) This is a regulatory concept that
focuses on investigating, understanding, and correcting discrepancies while
attempting to prevent their recurrence. This model separates CAPA into three
separate concepts:
• Remedial corrections of an identifi ed problem
• Root cause analysis with corrective action to help understand the cause of
the deviation and prevent recurrence of a similar problem
• Preventative action to prevent recurrence of similar problems
Change Control This process focuses on managing change to prevent unintended
consequences.
Quality Unit While the GMPs refer to a quality unit, current industry practice
is to divide the responsibilities of this unit between two groups:
• Quality control (QC) usually involves (a) assessing the suitability of incoming
components and the fi nished products, (b) evaluating the performance
of the manufacturing process, and (c) determining the acceptability of each
batch for release and distribution
• Quality assurance (QA) involves (a) review and approval of all procedures
related to manufacturing and maintenance, (b) review of records, and
(c) auditing and performing/evaluating trend analyses.
Six - System Inspection Model The FDA ’ s instruction manual for its investigators
is a systems - based approach to inspection consistent with this guidance. The
FDA defi nes six interlocked systems: (1) the quality system which encompasses
all the other systems, (2) a materials system, (3) a production system,
(4) a packaging and labeling system, (5) a facilities and equipment system, and
(6) a laboratory controls system. The agency believes that use of this overall
system approach will help fi rms achieve better control.
1.1.3.2 Quality Systems Model
This section was written to describe a model for use in pharmaceutical manufacturing
that can supply the controls to consistently produce a product of acceptable
quality. The model is described by four major factors:
26 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
• Management responsibilities
• Resources
• Manufacturing operations
• Evaluation
Management Responsibilities The FDA feels that a robust quality system model
calls for management to play a key role in the design, implementation, and
management of the quality system.
Resources Suffi cient resources should be provided to create a robust quality
system that complies with the GMP regulations. Senior management or a
designee should be responsible for providing adequate resources.
Facilities and Equipment The technical experts who have an understanding of
pharmaceutical science, risk factors, and manufacturing processes related to
the product are responsible for defi ning specifi c facility and equipment requirements.
The equipment must be qualifi ed, calibrated, cleaned, and maintained
to prevent contamination and product mix - ups. It is important to remember
that the GMPs place as much emphasis on process equipment as on testing
equipment while most quality systems focus only on testing equipment.
Control Outsourced Operations Quality systems call for contracts with outside
suppliers that clearly describe the materials or service, quality specifi cation
responsibilities, and communication mechanisms.
Manufacturing There is an overlap between the elements of a quality system
and the GMP regulation requirements for manufacturing operations. One
should always remember that the FDA ’ s enforcement programs and inspectional
coverage are based on the GMPs. The FDA feels that the following
factors are essential in a manufacturing quality system:
1. Design, develop, and document product and processes
2. Examine inputs
3. Perform and monitor operations
4. Address nonconformities
Evaluation Activities This includes the following activities:
1. Analyze data for trends
2. Conduct internal audits
3. Quality risk management
4. Corrective action
5. Preventative action
6. Promote improvements
1.1.4 GUIDANCE FOR INDUSTRY: PAT — FRAMEWORK FOR
INNOVATIVE PHARMACEUTICAL DEVELOPMENT,
MANUFACTURING, AND QUALITY ASSURANCE
This guidance is intended to describe a regulatory framework that the FDA chooses
to call process analytical technology , or PAT. It is the FDA ’ s hope that this will
encourage the voluntary development and implementation of innovative pharmaceutical
development, manufacturing, and quality assurance. The FDA intended this
guidance for a broad audience in different organizational units. To a large extent,
the guidance discusses principles with the goal of highlighting opportunities and
developing regulatory processes that encourage innovation.
Conventional pharmaceutical manufacturing is usually accomplished using batch
processing with laboratory testing of samples at various stages of manufacturing to
evaluate quality. The FDA believes that opportunities exist for improving the development,
manufacturing, and quality assurance steps through innovation in product
and process development, process control, and analysis.
Typically, the pharmaceutical industry has been reluctant to try something new
due to the fear that the new approach will not fi nd favor with the FDA. An FDA
rejection would result in costly delays and processing revisions that industry is
unwilling to risk. The FDA now says that this hesitancy is undesirable from a public
health perspective and it would like to see more innovation introduced. According
to the FDA, pharmaceutical manufacturing should be based on:
• The design of effective and effi cient manufacturing manufacturing processes
• Product and process specifi cations based on an understanding of how formulation
and process factors affect product performance
• Continuous real - time quality assurance
• Relevant regulatory policies and procedures tailored to accommodate the most
current level of scientifi c knowledge
• Risk - based regulatory approaches that recognize:
The level of scientifi c understanding of how formulation and manufacturing
process factors affect product quality and performance
The capability of process control strategies to prevent or mitigate the risk of
producing a poor - quality product
It is the intent of this guidance to facilitate progress to this state. So far, the FDA ’ s
stated goal is not being met. FDA representatives have stated the agency ’ s concern
about the failure of industry to rush to implement change. However, the economies
of change continue to favor the status quo.
1.1.4.1 PAT Framework
Quality should be built into pharmaceutical products through a comprehensive
understanding of:
• Intended therapeutic objectives, patient population, route of administration,
and pharmacokinetic characteristics of a drug
• Chemical, physical, and biopharmaceutic characteristics of a drug
• Design of a product and selection of product components and packaging based
on drug attributes
• Design of manufacturing processes using principles of engineering, material
science, and quality assurance to ensure acceptable and reproducible product
quality and performance throughout a product ’ s shelf life
GUIDANCE FOR INDUSTRY 27
0
0
28 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
Process Understanding A process is considered to be well understood when all
critical sources of variability are identifi ed and explained, variability is managed by
the process, and product quality attributes can be accurately and reliably predicted.
Principles and Tools Pharmaceutical manufacturing often consists of a series of
unit operations, each of which is intended to change certain properties of the materials
being processed. To assure these changes are acceptable and reproducible, consideration
should be given to the quality attributes of incoming materials and their
acceptability for the given unit operation. Most current pharmaceutical processes
are based on time - defi ned endpoints such as “ blend for ten minutes. ” In some cases,
these time - defi ned endpoints do not consider the effects of physical differences in
raw materials. Processing diffi culties can arise that result in the failure of a product
to meet specifi cations even if the raw materials conform to established specifi cations.
Use of PAT tools and principles can provide relevant information relating to
physical, chemical, and biological attributes. The process understanding gained from
this information will enable process control and optimization, address the limitation
of the time - defi ned endpoints, and improve effi ciency.
PAT Tools There are many tools available that enable process understanding.
These tools, when used within a system, can provide effective and effi cient means
for acquiring information to facilitate process understanding, continuous improvement,
and development of risk mitigation strategies. Such tools are categorized as
follows:
• Multivariate tools for design, data acquisition, and analysis
• Process analyzers
• Process control tools
• Continuous improvement and knowledge management tools
Strategy for Implementation To enable successful implementation of PAT, fl exibility,
coordination, and communication with manufacturers are critical. The FDA
believes that current regulations are suffi ciently broad to accommodate these strategies.
In the course of implementing the PAT framework, manufacturers may want
to evaluate the suitability of a tool on experimental and/or production equipment
and processes. It is recommended that risk analysis of the impact on product quality
be conducted before installation. This can be accomplished within the facility ’ s
quality system without prior notifi cation to the agency. Data collected using an
experimental tool should be considered research data. If conducted in a production
facility, it should be done under the facility ’ s quality system. The FDA does not
intend to inspect research data collected on an existing product for the purpose of
evaluating the suitability of an experimental PAT tool. Its routine inspection of a
fi rm ’ s manufacturing process that incorporates a PAT tool for research purposes
will be based on current regulatory standards.
The FDA has posted much of the information that fi rms will need in order to
implement a PAT program on the Web at http://www.fda.gov/cder/ops/pat.htm .
All marketing applications, amendments, or supplements to an application should
be submitted to the appropriate Center for Drug Evaluation and Research (CDER)
or Center for Veterinary Medicine (CVM) division in the usual manner. In general,
PAT implementation plans should be risk based. The FDA has suggested the following
possible implementation plans, where appropriate:
• PAT can be implemented under the facility ’ s own quality system. CGMP inspections
by the PAT team or PAT - certifi ed investigator can precede or follow PAT
implementation.
• A supplement [Changes Being Expected (CBE), Changes Being Expected in 30
Days (CBE - 30), or Prior Approval Supplement (PAS)] can be submitted to the
agency prior to implementation, and, if necessary, an inspection can be performed
by a PAT team or PAT certifi ed investigator before implementation.
• A comparability protocol can be submitted to the agency outlining PAT research,
validation and implementation strategies, and time lines. Following approval of
this comparability protocol by the agency, one or a combination of the above
regulatory pathways can be adopted for implementation.
To facilitate adoption or approval of a PAT process, manufacturers may request
a preoperational review of a PAT manufacturing facility and process by the PAT
team by contacting the FDA Process Analytical Technology Team at PAT@cder.fda.
gov . It should be noted that when certain PAT implementation plans neither affect
the current process nor require a change in specifi cations, several options can be
considered. Manufacturers should evaluate and discuss with the agency the most
appropriate option for their situation.
1.1.5 GUIDANCE FOR INDUSTRY: PART 11. ELECTRONIC RECORDS;
ELECTRONIC SIGNATURES — SCOPE AND APPLICATION
Of the many regulations written by the FDA, the least understood is undoubtedly 21
CFR Part 11. Rather than review the regulation itself, which is under review and possible
revision, we will review the guidance for industry that FDA published in August
2003 to “ aid ” industry in their puzzlement. Depending on the source, it appears to be
questionable as to whether this guidance document aids or confuses. It exists, however,
and like it or not, understand it or not, the regulation must be followed.
The guidance indicates that the FDA ’ s approach is based on three main
components:
• The regulation will be interpreted narrowly. Fewer records will be considered
subject to Part 11.
• Those records that are considered subject to Part 11 will be subject to enforcement
discretion with regard to the requirements for validation, audit trails,
record retention, and record copying in the manner described and with regard
to all Part 11 requirements for systems that were operational before the effective
date of this regulation.
• All predicate rule requirements will be enforced. This includes record and
record - keeping requirements.
The FDA does intend to enforce all other provisions of Part 11, including certain controls
for closed systems. The following controls and requirements will be enforced:
GUIDANCE FOR INDUSTRY 29
30 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
• Limiting system access to authorized individuals
• Use of operational system checks
• Use of authority checks
• Use of device checks
• Determination that persons who develop, maintain, or use electronic systems
have the education, training, and experience to perform their assigned tasks
• Establishment of and adherence to written policies that hold individuals
accountable for actions initiated under their electronic signatures
• Appropriate controls over systems documentation
• Controls for open systems corresponding to controls for closed systems
• Requirements related to electronic signatures
Part 11 Records Under the narrow interpretation, the FDA considers Part 11 to
be applicable to the following records or signatures in electronic format:
1. Records that are required to be maintained under predicate rule requirements
and that are maintained in electronic format in place of paper format.
2. Records that are required to be maintainer under predicate rules, that are
maintained in electronic format in addition to paper format, and that are relied
on to perform regulated activities.
3. Records submitted to the FDA under predicate rules in electronic format.
However, a record that is not itself submitted but is used in generating a submission
is not a Part 11 record.
4. Electronic signatures that are intended to be the equivalent of handwritten
signatures, initials, and other general signings required.
FDA ’s Approach to Specifi c Part 11 Requirements
1. Validation With respect to validation, the agency intends to exercise enforcement
discretion regarding specifi c Part 11 requirements. However, compliance
with all applicable predicate rules for validation is still expected. The FDA
suggests an approach to validation be based on a justifi ed and documented
risk assessment and a determination of the potential of the system to affect
product quality, safety, and record integrity.
2. Audit Trail The agency also intends to exercise enforcement discretion
regarding specifi c requirements related to computer - generated, time - stamped
audit trails and any corresponding requirements in Part 11. Compliance with
all applicable predicate rule requirements related to documentation of date,
time, or sequencing of events is still expected. It is also required to comply
with rules for ensuring that changes to records do not obscure previous
entries.
3. Legacy Systems The FDA intends to exercise enforcement discretion with
respect to all Part 11 requirements for systems that otherwise were operational
prior to August 20, 1997. Thus they do not intend to take enforcement action
to enforce compliance with any Part 11 requirements if all of the following
criteria are met for a specifi c system:
• The system was operational before the effective date.
• The system met all applicable predicate rule requirements before the effective
date.
• The system currently meets all applicable predicate rule requirements.
• There is documented evidence and justifi cation that the system is fi t for its
intended use.
4. Copies of Records Enforcement discretion will be applied with respect to
specifi c Part 11 requirements for generating copies of records and any corresponding
requirements in this part. An investigator should be provided with
reasonable and useful access to records during an inspection. All records held
by a manufacturer are subject to inspection.
5. Record Retention The FDA intends to exercise enforcement discretion with
regard to the Part 11 requirements for the protection of records to enable their
accurate and ready retrieval at any time throughout the records retention
period.
1.1.6 GUIDANCE FOR INDUSTRY AND FDA : CURRENT GOOD
MANUFACTURING PRACTICE FOR COMBINATION PRODUCTS
This document discusses the applicability of GMPs to combination products as
defi ned under 21 CFR 3.2(e). Manufacturers must ensure that the product is not
adulterated; the product possesses adequate strength, quality, identity, and purity;
and the product complies with performance standards as appropriate. This guidance
does not address technical manufacturing methods or make recommendations for
manufacturers ’ selection of facilities used in manufacturing.
A combination product is a product composed of a drug and a device, a biological
product and a device, a drug and a biological product, or a drug, a device, and a
biological product. For the purposes of this document, a constituent part of a combination
product is an article in a combination product that can be distinguished by
its regulatory identity as a drug, device, or biological product.
For regulatory purposes, a combination product is assigned to an agency center
or alternative organizational component that will have primary jurisdiction for its
premarket review and regulation. Manufacturers will be required to use the applicable
GMP for their products. Regulations that may apply are:
• GMP regulations for fi nished pharmaceuticals (21 CFR Parts 210 and 211).
• Quality system regulations for devices (21 CFR Part 820).
• The biological product regulations (21 CFR Parts 600 – 680) may also apply to
the manufacture of drugs that are also biological products along with the drug
provisions.
There are no GMP regulations specifi cally for combination products. Until such
regulations are promulgated, the manufacture of each constituent part is governed
by the regulations for that component.
The Offi ce of Combination Products is available as a resource to sponsors
throughout the lifecycle of a combination product. This offi ce can be reached at
GUIDANCE FOR INDUSTRY AND FDA 31
32 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
(301) 427 - 1934 or by E - mail at combination@fda.cov . Updated guidance documents
are available at the offi ce ’ s Internet website, http://www.fda/gov/oc/combination .
1.1.7 GUIDANCE FOR INDUSTRY: POWDER BLENDS AND FINISHED
DOSAGE UNITS — STRATIFIED IN - PROCESS DOSAGE UNIT SAMPLING
AND ASSESSMENT
This guidance is intended to assist manufacturers in meeting the GMP requirements
for demonstrating the adequacy of mixing to ensure uniformity of in - process powder
blends and fi nished dosage units.
Stratifi ed Sampling In this process dosage units are sampled at predefi ned intervals
and representative samples collected from specifi cally targeted locations in the
compression/fi lling operations that have the greatest potential to yield extremes of
drug concentration.
This guidance describes methods of sampling that might be used to demonstrate
active ingredient homogeneity. These methods are put forward as suggestions and
are not intended to be the only methods for meeting FDA requirements for demonstration
of the adequacy of a powder mix.
Assessment of Powder Mix Uniformity The following procedures are
recommended:
1. Conduct blend analysis on batches by extensively sampling the mix in the
blender and/or intermediate bulk containers.
2. Identify appropriate blending time and speed ranges, dead spots in blenders,
and locations of segregation in intermediate bulk containers (IBCs).
3. Defi ne the effects of sample size (1 – 10 times the dosage unit range) while
developing a technique capable of measuring the true uniformity of the blend.
Sample quantities larger than 3 times the dosage size can be used with adequate
scientifi c justifi cation.
4. Design blend - sampling plans and evaluate them using appropriate statistical
analyses.
5. Quantitatively measure any variability that is present among the samples.
Attribute the sample variability to either lack of uniformity of the blend or
sampling error. Signifi cant variances in the blend data within a given location
can be an indication of one factor or a combination of factors such as inadequacy
of blend mix, sampling error, or agglomeration. Signifi cant between -
location variance can indicate that the blending operation is inadequate.
Correlation of Powder Mix Uniformity with Stratifi ed In -Process Dosage Unit
Data The following steps are recommended for correlation:
1. Conduct periodic sampling and testing of the in - process dosage units by sampling
them at defi ned intervals and locations throughout the compression or
fi lling process. Use a minimum of 20 appropriately spaced in - process dosage
unit sampling points. There should be at least 7 samples taken from each of
these locations for a total minimum of at least 140 samples.
2. Take 7 samples from each additional location to further assess each signifi cant
event, such as fi lling or emptying of hoppers and IBCs, start and end of the
compression or fi lling process, and equipment shutdown. This may be accomplished
by using process development batches, validation batches, or routine
manufacturing batches for approved products.
3. Signifi cant events may also include observations or changes from one batch
to another (e.g., batch scale - up and observations of undesirable trends in previous
batch data).
4. Prepare a summary of the data and analysis used to correlate the stratifi ed
sampling locations with signifi cant events in the blending process.
5. Compare the powder mix uniformity with the in - process dosage unit data
described above.
6. Investigate any discrepancies observed between powder mix and dosage
unit data and establish root causes. At least one troubleshooting guide is
available that may be helpful with this task. Possible corrections may range
from going back to formulation development to improve powder characteristics
to process optimization. Sampling problems may also be negated by use
of alternate state - of - the - art methods of in situ real - time sampling and
analysis.
Correlation of Stratifi ed In -Process Samples with Finished Product The following
steps are recommended:
1. Conduct testing for uniform content of the fi nished product using an appropriate
procedure or as specifi ed in the ANDA or the NDA for approved
products.
2. Compare the results of stratifi ed in - process dosage unit analysis with uniform
content of the fi nished dosage units from the previous step. This analysis
should be done without weight correction.
3. Prepare a summary of the data and analysis used to conclude that the stratifi ed
in - process sampling provides assurance of uniform content of the fi nished
product.
1.1.7.1 Validation of Batch Powder Mix Homogeneity
This section describes sampling and testing the powder mix of demonstration and
process validation batches used to support implementing the stratifi ed sampling
method described in this guidance.
The guidance document recommends that during the manufacture of demonstration
and process validation batches, the following uniformity characteristics be
assessed: (1) the powder blend, (2) the in - process dosage units, and (3) the fi nished
product. Each attribute should be determined independently. It is further recommended
that the following steps be used to identify sampling locations and acceptance
criteria prior to the manufacture of the exhibit and/or validation batches:
GUIDANCE FOR INDUSTRY AND FDA 33
34 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
1. Carefully identify at least 10 sampling locations in the blender to represent
potential areas of poor blending. For example, in tumbling blenders (such as
V - blenders, double cones, or drum mixers), samples should be selected from
at least two depths along the axis of the blender. For convective blenders (such
as a ribbon blender), a special effort should be made to implement uniform
volumetric sampling to include the corners and discharge area (at least 20
locations are recommended to adequately validate convective blenders).
2. Collect at least three replicate samples from each location. Samples should
meet the following criteria:
• Assay one sample per location (number of samples n = 10, or n = 20 for
ribbon blender).
• RSD (relative standard deviation) of all individual results is 5.0%.
• All individual results are within 10.0% (absolute) of the mean of the
results.
It is also recommended that you not proceed any further with implementation
of the methods described in this guidance until the criteria are met.
Sampling errors may occur in some powder blends, sampling devices, and techniques
that make it impractical to evaluate adequacy of mix using only the blend
data. In such cases, it is recommended that in - process dosage unit data be used in
conjunction with blend sample data to evaluate blend uniformity.
Some powder blends may present an unacceptable safety risk when directly
sampled. The safety risk, once described, may justify an alternate procedure. In such
cases, process knowledge and data from indirect sampling combined with additional
in - process dosage unit data may be adequate to demonstrate the adequacy of the
powder mix. Data analysis used to justify using these alternate procedures
should be described in a summary report that is maintained at the manufacturing
facility.
1.1.7.2 Verifi cation of Manufacturing Criteria
The assessment of powder mix uniformity and correlation of stratifi ed in - process
dosage unit sampling development procedures should be completed before establishing
the criteria and controls for routine manufacturing. It is also recommend
that the normality be assessed and that the RSD be determined from the results of
stratifi ed in - process dosage unit sampling and testing that were developed. The RSD
value should be used to classify the testing results as either readily pass (RSD 4.0%) ,
marginally pass (RSD 6.0%), or inappropriate for demonstration of batch homogeneity
when RSD > 6.0%.
The FDA recommends that routine manufacturing batches be evaluated
against the following criteria after completing the procedures described above to
assess the adequacy of the powder mix and uniform content in the fi nished dosage
form:
1. Standard criteria method (SCM) — This method is recommended when either of
the following conditions is met:
1.1. Results of establishing initial criteria are classifi ed as readily pass .
1.2. Results of testing to the marginal criteria method (MCM) pass the criteria
for switching to the SCM.
1.2.1. Stage 1 Test To perform the stage 1 test, collect at least three dosage
units from each sampling location, assay one dosage unit from each
location, weight correct the results, and compare the results with the
following criteria:
1.2.1.1. RSD of all individual results is less than 5%.
1.2.1.2. Mean of all results is 90 – 110% of target assay.
If the results pass these criteria and the adequacy of mix and uniformity
of dosage unit content for the batch are adequate, the SCM
can be used for the next batch. If test results fail stage 1 criteria,
extended testing to stage 2 is required.
1.2.2. Stage 2 Test To perform the stage 2 test, assay the remaining two
dosage units from stage 1 for each sampling location and compute the
mean and RSD of data combined from both stage 1 and stage 2.
Compare the results with the following criteria:
1.2.2.1. For all individual results, the RSD should be less than 5.0%.
1.2.2.2. Mean of all results is 90 – 110% of target assay.
If results pass the above criteria, the adequacy of mix and uniformity
of content for the batch are adequate and stage 1 can be used for
the next batch. If test results fail the criteria, use the MCM described
in the section below.
2. Marginal criteria method — The MCM can be used when either of the following
conditions is met:
2.1. Results of initial criteria establishment qualifi ed as marginally pass .
2.2. Results of initial criteria establishment qualifi ed as readily pass or a batch
was tested according to SCM and the test results failed both stage 1 and
stage 2 criteria.
2.3. If either of the above two criteria apply, use the weight corrected
results from the stage 2 SCM analysis and compare this with the MCM
criteria:
2.3.1. For al individual results, the RSD is less than 6.0% .
2.3.2. The mean of all results is 90.0 – 110.0% of target assay.
2.4. It is acceptable to switch to the SCM when fi ve consecutive batches pass the
MCM criteria and result in RSD of less than 5.0%.
1.1.8 GUIDANCE FOR INDUSTRY: IMMEDIATE - RELEASE SOLID
ORAL DOSAGE FORMS SCALE - UP AND POSTAPPROVAL CHANGES
( SUPAC ) — CHEMISTRY, MANUFACTURING AND CONTROLS,
IN VITRO DISSOLUTION TESTING, AND IN VIVO
BIOEQUIVALENCE DOCUMENTATION
This guidance provides recommendations to NDA and ANDA sponsors who intend
to make changes to the product during the postapproval period. Changes include
any change in components or composition of the product, the site of manufacture,
the scale - up/scale - down of batch size, and/or the manufacturing process and/or
equipment of an immediate - release oral formulation.
GUIDANCE FOR INDUSTRY 35
36 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
Changes in Components (Excipients) and Composition Changes in the amount
or source of drug substance are not addressed by this guidance. Changes in components
or composition that have the effect of adding a new excipient or deleting an
excipient are defi ned at level 3 except as described below:
1. Level 1 changes
1.1. Level 1 changes are those that are unlikely to have any detectable impact
on formulation quality and performance.
1.2. Allowed changes (changes that can be made without prior FDA
approval) are shown below. This is based on the assumption that the drug
substance in the product is formulated to 100% of label potency. To be considered
a level 1 change, the total additive effect of all excipient changes
should not be more than 5% relative to the target dosage form weight.
Excipient
Percentage of Excipient (W/W)
Out of Total Target Dosage
Form Weight
Filler ± 5
Disintegrant
Starch ± 3
Other ± 1
Binder ± 0.5
Lubricant
Calcium or magnesium Stearate ± 0.25
Other ± 1
Glidant
Talc ± 1
Other ± 0.1
Film coat ± 1
1.3. Test documentation
1.3.1. Chemistry — Application/compendial release requirements and stability
testing. For stability testing, one batch should be on long - term stability
testing with data being reported in the annual report.
1.3.2. Filing documentation — All information must be included in the annual
report (including long - term stability data).
2. Level 2 changes
2.1. Level 2 changes are those that could have a signifi cant impact on formulation
quality and performance. Tests and fi ling documentation for a level 2 change
depend on three factors: (1) therapeutic range, (2) solubility, and (3) permeability.
Therapeutic range is defi ned as either narrow or nonnarrow. Drug
solubility and drug permeability are defi ned as either low or high. Changes
in excipients, expressed as percent (w/w) of total formulation, greater than
those listed for a level 1 change but less than or equal to the following
percent ranges are acceptable level 2 changes:
Excipient
Percentage of Excipient (w/w) of Total Target
Dosage Form Weight
Filler ± 10
Disintegrant
Starch ± 6
Other ± 2
Binder ± 1
Lubricant
Ca or Mg stearate ± 0.5
Other ± 2
Glidant
Talc ± 2
Other ± 0.2
Film coat ± 2
These percentages are based on the assumption that the drug substance in
the fi nished product is formulated to 100% of labeled potency. The total
additive effect of all excipient changes should not change by more than
10%.
All components in the formulation should have numerical targets that
represent the nominal composition of the product on which any future
changes in the composition of the product are based. Allowable changes in
the composition should be based on the approved target composition and
not on the composition based on previous level 1 or level 2 changes.
2.2. Test documentation
2.2.1. Chemistry
2.2.1.1. Application/compendial release requirements and batch
records.
2.2.1.2. Stability testing — Test one batch with three months of accelerated
stability data in supplement and on batch on long - term
stability.
2.2.2. dissolution
2.2.2.1. High - permeability, high - solubility drugs — Dissolution of 85%
in 15 min in 900 mL of 0/1 N HCl. If a drug product fails to
meet this criterion, tests in 2.2.2.2 or 2.2.2.3 below should be
performed.
2.2.2.2. Low - permeability, high - solubility drugs — Multipoint dissolution
profi le should be performed in the application/compendial
medium at 15, 30, 45, 60, and 120 min or until an asymptote
is reached. The dissolution profi le of the proposed and currently
used product formulations should be similar.
2.2.2.3. High - permeability, low - solubility drugs — Multipoint dissolution
profi les should be performed in water, 0.1 N HCl, and
USP buffer media at pH 4.5, 6.5, and 7.5 (fi ve different pro-
fi les) for the proposed and currently accepted formulations.
Adequate sampling should be performed at 15, 30, 45, 60, and
120 min until either 90% of drug from the drug product is
GUIDANCE FOR INDUSTRY 37
38 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
dissolved or an asymptote is reached. A surfactant may be
used, but only with appropriate justifi cation. The dissolution
profi le of the proposed and currently used product formulations
should be similar.
2.2.3. In vivo bioequivalence documentation is not required for level 2. If
the product does not meet any of the level 1 cases above, refer to level
3 changes.
2.2.4. Filing documentation — A prior approval supplement with all data
including the accelerated stability data is required. This change should
also be documented in the annual report along with the long - term
stability data.
2.3. Level 3 changes
2.3.1. Level 3 changes are those that are likely to have a signifi cant impact
on formulation quality and performance. Tests and fi ling documentation
vary depending on the following three factors: therapeutic range,
solubility, and permeability. For example:
2.3.1.1. Any qualitative and quantitative excipient changes to a narrow
therapeutic drug beyond the ranges specifi ed in the level 1
table.
2.3.1.2. All other drugs not meeting the dissolution criteria under level
2.
2.3.1.3. Changes in the excipient ranges of low - solubility, low -
permeability drugs beyond those listed in level 1.
2.3.1.4. Changes in the excipient ranges of all drugs beyond those
listed in the level 2 table.
2.3.2. Test documentation
2.3.2.1. Chemical
(a) Application/compendial release requirements and batch
records:
• Information available — One batch with three months
accelerated stability data reported in a supplement and
one batch on long - term stability reported in the annual
report.
• Information NOT available — Up to three batches with
three months accelerated stability data reported in the
supplement and one batch on long - term stability data
reported in annual report.
(b) Dissolution documentation — Case B dissolution profi le as
described in the table for level 2.
(c) In vivo bioequivalence documentation — Full bioequivalence
study. This requirement may be waived with a veri-
fi ed acceptable in vivo/in vitro correlation.
2.3.2.2. Filing documentation — Prior approval supplement including
accelerated stability data plus an annual report showing long -
term stability data.
Site Changes Site changes are changes in the location of manufacture for both
company - owned and contract manufacturing facilities. A site change does not
include, for example, scale - up changes, changes in manufacturing equipment or a
manufacturing process, and changes in Standard Operating Procedures (SOPs) or
environmental changes. Each change must be considered separately.
1. Level 1 changes — A level 1 change consists of a site change within a single facility
where the same equipment, SOPs, environmental conditions, and personnel are
used and where no changes are made to the manufacturing batch records other
than location of the facility and administrative changes.
1.1. Required documentation — No documentation is required beyond the usual
application/compendial requirements. No in vivo bioequivalence documentation
is required.
1.2. Filing requirements — Annual report.
2. Level 2 changes — A level 2 change is a site change within a contiguous campus
or between facilities in adjacent city blocks where the same equipment, SOPs,
environmental conditions and controls, and personnel common to both manufacturing
sites are used. There must be no changes to the manufacturing batch
records except for administrative information and the location of the facility.
2.1. Required documentation
2.1.1. Chemistry—Identify location of new site and updated batch records.
No other documentation is required beyond application/compendial
release requirements, although one batch produced at the new site
should be placed on long - term stability and the data should be reported
in the annual report. Dissolution data other than normal release requirements
are not required nor is in vivo bioequivalence testing required.
2.1.2. Filing documentation — A supplement should be fi led showing the
changes being effected. Long - term stability test data should be included
in the annual report.
3. Level 3 changes — A level 3 change is a change in manufacturing site to a different
campus. However, the same equipment, SOPs, environmental conditions, and
controls should be used in the manufacturing process at the new site. No changes
may be made to the manufacturing batch records except for administrative information,
location, and language translation if needed.
3.1. Documentation
3.1.1. Chemistry — Location of new site and updated batch records.
3.1.2. Stability
3.1.2.1. If a signifi cant body of data is available, one batch with three
months accelerated stability data must be reported in a supplement.
One batch should be on long - term stability with the
stability data reported in the annual report.
3.1.2.2. If a signifi cant body of data is not available, up to three batches
with three months accelerated stability data should be reported
in the supplement. Up to three batches should be on long - term
stability with these data being reported in the annual report.
3.1.3. Dissolution — A multipoint dissolution profi le should be performed in
the application/compendial medium at 15, 30, 45, 60, and 120 min or
until an asymptote is reached. The dissolution profi le of the drug
product at the current and proposed site should be similar.
3.1.4. In vivo bioequivalence — None required.
GUIDANCE FOR INDUSTRY 39
40 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
3.2. Filing documentation required — Changes being effected should be identifi ed
in a supplement. Long - term stability data are reported in the annual
report.
Changes in Batch Size Postapproval changes in the size of a batch from the pilot
scale used to manufacture product for clinical trials to larger or smaller commercial
batch sizes require submission of additional information in the application. Scale -
down below 100,000 dosage units is not covered by this guidance. All scale - up
changes should be properly validated and, where needed, inspected by appropriate
FDA personnel.
1. Level 1 changes — A change in batch size, up to and including a factor of 10 times
the size of the pilot batch, is considered a level 1 change. However, (1) the equipment
used must be of the same design and operating principles, (2) the product
is manufactured in full compliance with the prevailing GMPs, and (3) the same
formulation and manufacturing procedures are used as well as the same SOPs
and controls.
1.1. Chemistry documentation — (1) Application/compendial release requirements,
(2) notifi cation of change to the FDA and submission of updated
batch records in the annual report, and (3) one batch should be on long - term
stability with results being provided in the annual report.
1.2. Dissolution documentation — None beyond application/compendial release
requirements.
1.3. In vivo bioequivalence — None.
1.4. Filing documentation — Annual report with long - term stability data.
2. Level 2 changes — Level 2 consists of changes in batch size beyond a factor of 10
times the size of the pilot batch where (1) the equipment used to produce the
pilot batches is of the same design and operating principles, (2) the product is
manufactured in full compliance with the prevailing GMPs, and (3) the same
formulation and manufacturing procedures are used as well as the same SOPs
and controls.
2.1. Chemistry — Application/compendial release requirements. Notifi cation of
change in batch size and submission of updated batch records to the FDA.
One batch must be placed on accelerated stability testing and one on long -
term stability.
2.2. Dissolution — None beyond application/compendial release requirements.
2.3. In vivo bioequivalence — None.
2.4. Filing requirements — Must submit changes being effected in the supplement.
Long - term stability data are reported in the annual report.
Manufacturing Manufacturing changes may be either the equipment used in the
manufacturing process or the process itself:
1. Equipment
1.1. Level 1 equipment changes — This category includes change from the use of
nonautomated or nonmechanical equipment to automated or mechanical
equipment to move ingredients and a change to alternative equipment of
the same design and operating principles of the same or different capacity.
1.1.1. Chemistry documentation — Application/compendial release requirements,
notifi cation of change, and submission of updated batch records.
One batch should be placed on long - term stability.
1.1.2. Dissolution documentation — None other than application/compendial
release requirements.
1.1.3. In vivo bioequivalence documentation — None.
1.1.4. Filing documentation—Annual report with long-term stability data.
1.2. Level 2 equipment changes — This type of change involves a change in equipment
to a different design and different operating principles.
1.2.1. Chemistry documentation — Application/compendial release requirements,
notifi cation of change, and submission of updated batch
records.
1.2.1.1. If a signifi cant body of data are available, one batch with three
months of accelerated stability data reported in the supplement
and one batch on long - term stability with data reported
in the annual report.
1.2.1.2. If a signifi cant body of data are not available, submit up to
three batches with three months accelerated stability data in
the supplement and up to three batches on long - term stability
with data reported in the annual report.
1.2.2. Dissolution documentation — A multipoint dissolution profi le should
be performed in the application/compendial medium at 15, 30, 45, 60,
and 120 min or until an asymptote is reached. The dissolution profi le
of the drug product at the current and proposed site should be
similar.
1.2.3. In vivo bioequivalence documentation — None.
1.2.4. Filing documentation — Prior approval supplement with justifi cation
for change; long - term stability data must be reported in the annual
report.
2. Process changes
2.1. Level 1 process changes — This includes process changes such as changes in
mixing times and operating speeds within application/validation ranges.
2.1.1. Chemistry documentation — None beyond application/compendial
release requirements.
2.1.2. Dissolution documentation — None beyond application/compendial
release requirements.
2.1.3. In vivo bioequivalence documentation — None.
2.1.4. Filing documentation — Annual report.
2.2. Level 2 process changes — Level 2 changes include process changes such as
mixing times and operating speeds outside of application/validation ranges.
2.2.1. Chemistry documentation — Application/compendial release requirements;
notifi cation of change and submission of updated batch records.
One batch on long - term stability.
2.2.2. Dissolution documentation — A multipoint dissolution profi le should
be performed in the application/compendial medium at 15, 30, 45, 60,
and 120 min or until an asymptote is reached. The dissolution profi le
of the drug product at the current and proposed site should be
similar.
GUIDANCE FOR INDUSTRY 41
42 GOOD MANUFACTURING PRACTICES & RELATED FDA GUIDELINES
2.2.3. In vivo bioequivalence documentation — None.
2.2.4. Filing documentation — A supplement with changes being effected.
Long - term stability data should be reported in the annual report.
2.3. Level 3 process changes — Level 3 includes change in the type of process used
in the manufacture of the product, such as a change from wet granulation to
direct compression.
2.3.1. Chemistry documentation — Application/compendial release requirements.
Notifi cation of change and submission of updated batch records.
Stability testing varies depending on the amount of data available:
2.3.1.1. Signifi cant body of data available — One batch with three
months accelerated stability data should be reported in the
supplement; one batch should also be put on long - term stability
with data being reported in the annual report.
2.3.1.2. No signifi cant body of data available — Up to three batches
with three months accelerated stability data should be reported
in the supplement. Up to three batches should be on long - term
stability with data being reported in the annual report.
2.3.2. Dissolution documentation — A multipoint dissolution profi le should
be performed in the application/compendial medium at 15, 30, 45, 60,
and 120 min or until an asymptote is reached. The dissolution profi le
of the drug product at the current and proposed site should be
similar.
2.3.3. In vivo bioequivalence documentation — An in vivo bioequivalence
study should be performed. This may be waived if a suitable in vivo/in
vitro correlation has been verifi ed.
2.3.4. Filing documentation — A prior approval supplement must be fi led
with justifi cation for the change. Long - term stability data should be
submitted in the annual report.
1.1.9 OTHER GMP - RELATED GUIDANCE DOCUMENTS
This chapter has discussed the CGMP regulations and some of the more important
guidances. There have been a number of additional guidance documents related to
GMPs published by the FDA. These documents are all posted on the FDA website.
They are listed below along with their URL:
• Current good manufacturing practice for combination products: http://www.
fda.gov/cder/guidance/OCLove1dft.pdf
• Questions and answers on current good manufacturing practices (cGMP) for
drugs: http://www.fda.gov/cder/guidance/cGMPs/default.htm
• Powder blends and fi nished dosage units — Stratifi ed in - process dosage unit
sampling and assessment
• Sterile drug products produced by aseptic processing — Current good manufacturing
practice: http
• Current good manufacturing practice for medical gases: http://www.fda.gov/
cder/guidance/3823dft.pdf
• General principles of process validation
• SUPAC - IR: Immediate - release solid oral dosage forms: Scale - up and post -
approval changes: Chemistry, manufacturing and controls, in vitro dissolution
testing, and in vivo bioequivalence documentation: http://www.fda.gov/cder/
guidance/cmc5.pdf
• SUPAC - IR/MR: Immediate release and modifi ed release solid oral dosage
forms manufacturing equipment addendum
• SUPAC - MR: Modifi ed release solid oral dosage forms scale - up and postapproval
changes: Chemistry, manufacturing, and controls; in vitro dissolution
testing and in vivo bioequivalence documentation: http://www.fda.gov/cder/
guidance/1214fnl.pdf
• SUPAC - SS: Nonsterile semisolid dosage forms; scale - up and post - approval
changes: Chemistry, manufacturing and controls; in vitro release testing and in
vivo bioequivalence documentation
• SUPAC - SS: Nonsterile semisolid dosage forms manufacturing equipment
addendum
OTHER GMP-RELATED GUIDANCE DOCUMENTS 43
45
1.2
ENFORCEMENT OF CURRENT GOOD
MANUFACTURING PRACTICES
Kenneth J. Nolan
Nolan & Auerbach, P. A., Fort Lauderdale, Florida
Contents
1.2.1 Introduction and Background
1.2.2 Enforcement Players
1.2.3 FDA Enforcement Techniques
1.2.3.1 Inspections
1.2.3.2 After the Inspection: Form 483
1.2.3.3 Recalls
1.2.3.4 Warning Letter
1.2.4 Judicial Enforcement: Beyond the Warning Letter
1.2.4.1 Introduction
1.2.4.2 Civil Proceedings
1.2.4.3 Criminal Proceedings
1.2.5 Conclusion
1.2.1 INTRODUCTION AND BACKGROUND
The legal authority for the Food and Drug Administration (FDA) to impose
minimum manufacturing standards is set forth in the federal Food and Drug and
Cosmetic Act (FDCA), 21 U.S.C. sec. 301 et seq. Section 351(a)(2)(B) of 21 U.S.C.
requires manufacturers of drugs to operate in conformance with manufacturing
regulations established by the FDA. The regulations are primarily contained in Title
21 of the U.S. Code of Federal Regulations (CFR), Parts 210 and 211, and are called
the current good manufacturing practice (cGMP) regulations.
The cGMP regulations stem from congressional concern that impure and otherwise
adulterated drugs might escape detection under a system predicated only on
seizure of drugs shown to be in fact adulterated. That is, the U.S. Congress desired
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
46 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
to require manufacturers to utilize manufacturing practices designed to prevent
pharmaceuticals from such defects as contamination, nonconforming bioavailability,
or potency defects.
Congress stated the rationale for imposing cGMP on the pharmaceutical industry
this way 1 :
The manufacturing of drugs is a business that requires highly qualifi ed and trained
personnel, and special laboratory and other facilities and most careful internal manufacturing,
packaging, and labeling controls. These requirements are necessary to the
assurance that the drugs will be safe for the user and will have, and so far as possible
retain, the identity, strength, quality, purity, and effectiveness that they purport to
have.
The purpose of the cGMP requirement is to prevent injury and death “ by building
quality into the design and production of pharmaceuticals, ” 2 so that substandard
prescription drugs do not jeopardize the health and safety of the patients.
The cGMPs require manufacturers to have adequately equipped manufacturing
facilities, adequately trained personnel, precisely controlled manufacturing processes,
appropriate laboratory controls, complete and accurate records and reports,
appropriate fi nished product examination, and so on. Current GMPs are not “ best
practices ” ; rather, they establish threshold or minimum standards which must be
satisfi ed in order for a pharmaceutical manufacturing operation to be compliant.
The cGMPs were modifi ed only once between 1963 and 2002 with changes made
in 1978 to update them in light of the current technology and also to describe the
requirements more explicitly and with more specifi city. Meanwhile, the intervening
decades saw myriad advances in manufacturing science, engineering, and technology,
including the development of better quality systems. These advances, combined with
the desire to harmonize manufacturing standards in an increasingly globalized production
environment, created the impetus to revamp the cGMPs again.
In August 2002, the FDA announced a comprehensive review of the pharmaceutical
cGMPs. The agency identifi ed its cGMP initiative “ Pharmaceutical cGMPs for
the 21st Century: A Risk - Based Approach. ” The FDA ’ s articulated goals for the
initiative, relevant to enforcement, were:
• The submission review program and the inspection program operate in a coordinated
and synergistic manner.
• Regulation and manufacturing standards are applied consistently.
• FDA resources are used most effectively and effi ciently to address the most
signifi cant health risks.
One of the major products of the cGMP initiative was issued by the FDA in
September 2006 in a document entitled “ Guidance for Industry — Quality Systems
Approach to Pharmaceutical cGMP Regulations. ” 3 The FDA described the guid-
1 H. R. Rep. No. 2464, 87th Cong., 2d Sess. 2 (1962). See also 1962 U.S. Cong. and Admin. News , p. 2884.
2 FDA, Pharmaceutical cGMPs for the 21st century: A risk - based approach, Rockville, MD, August 21,
2002.
3 The FDA ’ s guidance documents advise the reader that they “ do not establish legally enforceable responsibilities.
Instead, guidances describe the [the FDA ’ s] current thinking on a topic and should be viewed
only as recommendations, unless specifi c regulatory or statutory requirements are cited. ”
ance as a comprehensive quality systems model which, if followed, would improve
quality control and satisfy the requirements of the cGMP regulations. Quality
systems and quality assurance are important parts of the cGMP modernization
process because quality assurance problems have been the cGMP issues most frequently
cited by FDA investigators in recent years.
Drugs which are manufactured not in accordance with any cGMP requirement,
including the quality control and quality process mandates, are “ adulterated ” under
the FDCA. Section 351 of 21 U.S.C. defi nes a drug as adulterated
[if] the methods used in, or the facilities or controls used for, its manufacture, processing,
packing, or holding do not conform to or are not operated or administered in
conformity with current good manufacturing practice to assure that such drug meets
the requirement of the act as to safety and has the identity and strength, and meets the
quality and purity characteristics, which it purports or is represented to possess.
1.2.2 ENFORCEMENT PLAYERS
The FDA is obviously one of the most important regulatory agencies in the United
States. It may also be characterized as the most important consumer protection
agency in the world. Its decisions involving approval of drugs have a direct effect
on testing, approval, access, and distribution of prescription drugs worldwide. As a
regulatory agency in a largely scientifi c role, it is involved in shaping pharmaceutical
science and drug access throughout the world. As a scientifi c agency, the FDA
employs physicians, pharmacists, biologists, biochemists, engineers, biostatisticians,
and other highly educated and specialized professionals.
But the FDA also has very important law enforcement responsibilities. The
agency employs civil and criminal investigators, auditors, attorneys, and other
enforcement professionals. One of the FDA ’ s many enforcement functions is investigation,
remediation, and prosecution of cGMP violations.
The FDA district offi ces operate under the auspices of the agency ’ s Offi ce of
Regulatory Affairs (ORA). The ORA fi eld organization is divided into fi ve regional
offi ces (northeast, central, southeast, southwest, Pacifi c). Each region includes district
offi ces, of which there are 20 nationwide. Most district offi ces have three or
four branches, including either a compliance branch or an enforcement branch. The
branch offi ces are the primary regulatory contacts within the districts and act as the
“ eyes and ears ” for FDA headquarters.
The FDA ’ s Offi ce of Criminal Investigations (OCI) is responsible for reviewing
allegations which if proven would violate the U.S. criminal code, including potential
violations of the cGMPs. The OCI investigators conduct such investigations as is
deemed appropriate, sometimes in connection with other federal investigative
agencies, including the FBI and the Offi ce of Inspector General of the Department
of Health and Human Services. If the OCI chooses not to recommend to the Department
of Justice (DOJ) 4 that criminal indictment be pursued, then the district offi ce
is at liberty to pursue the matter through administrative or civil proceedings.
4 The DOJ is under the direction of the attorney general of the United States. Its mission, relevant to this
chapter, is to enforce federal statutes and uphold the rule of the law. It pursues violations brought to its
attention by the FDA as well as other federal agencies.
ENFORCEMENT PLAYERS 47
48 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
Although the FDA ’ s Offi ce of General Counsel is involved with enforcement of
both civil and criminal matters, cases involving court enforcement are handled by
assistant U.S. attorneys (AUSAs), who are located in U.S. attorneys ’ offi ces located
across the United States. U.S. attorneys are the local representatives of the DOJ;
they are appointed by and serve at the discretion of the president, with advice
and consent of the Senate. There are 93 U.S. attorneys, and they are located (by
district) across the United States and its territories. Each U.S. attorney is the chief
federal law enforcement offi cer of the United States within his or her particular
district.
The AUSAs are the principal trial attorneys for the U.S. government. Each U.S.
attorney exercises wide discretion in the use of his or her resources to further the
priorities of the local jurisdiction. Discretion and expertise are big factors in case
decisions. There may be signifi cant disparity in the experience, interest, and capability
of U.S. attorneys ’ offi ces with respect to their pursuit of cGMP violations.
The impact of this disparity is mitigated or eliminated by the expertise of the
DOJ ’ s Offi ce of Consumer Litigation (OCL), which is charged with coordinating
and supporting FDCA prosecutions nationwide. The OCR ’ s attorneys exercise considerable
infl uence over and discretion in deciding what to and what not to prosecute,
thus fostering consistent prosecutive decision making. Many civil actions,
particularly those seeking injunctive relief, cannot be brought by a U.S. attorney
without OCL approval, minimizing the risk that an inconsistent policy position is
taken by a U.S. attorney ’ s offi ce.
1.2.3 FDA ENFORCEMENT TECHNIQUES
1.2.3.1 Inspections
The FDA has the right to conduct surveillance inspections of manufacturing facilities
for the purpose of enforcement. The goal of inspections is “ to minimize consumers
exposure to adulterated products. ” 5
The FDCA, 21 U.S.C. 374, provides that the FDA is authorized to enter and “ to
inspect, at reasonable times and within reasonable limits and in a reasonable manner
… all pertinent equipment, fi nished and unfi nished materials, containers, and labeling
” in the manufacturing or related facility. This statute further authorizes the
inspection to “ extend to all things therein (including records, fi les, papers, processes,
controls, and facilities) ” as long as the records, for example, are relevant to any
potential adulteration or misbranding 6 or other FDCA violations. The statute denies
the agency the right to review “ fi nancial data, sales data, pricing data, personnel data
(other than data as to qualifi cations of technical and professional personnel performing
functions) ” and certain other types of documents.
Inspectors are required to notify the company that the inspection is occurring
but need not provide their reasons. They may take samples and photographs related
5 Compliance Program Guidance Manual for FDA Staff: Drug Manufacturing Inspection Program ,
7356.002, available: www.fda.gov .
6 Misbranding involves labeling a pharmaceutical product in a misleading way. See 21 U.S.C. 331(k).
FDA ENFORCEMENT TECHNIQUES 49
to the subject of the inspection. It is a criminal offense to deny entry to FDA inspectors
or other offi cials who have appropriately made attempts to conduct an inspection.
[21 U.S.C. 331(f)]
In addition to the for - cause inspections, the FDCA mandates that the FDA routinely
inspect a manufacturer ’ s facilities for cGMP compliance every two years.
This applies to domestic and foreign facilities which manufacture drugs for sale
within the United States. 7 Unfortunately, this two - year mandate is rarely satisfi ed
because the FDA ’ s district offi ces, which are charged with the responsibility for the
inspections, lack suffi cient resources to conduct regular cGMP compliance
inspections.
The FDA conducts two categories of facility inspections — surveillance inspections
and compliance inspections. Surveillance inspections are periodic. Whether
and when to inspect a particular manufacturing facility is decided in part by application
of an analytical model to determine high risk sites. In late 2004, the FDA issued
a report entitled “ Risk - Based Method for Prioritizing cGMP Inspections of Pharmaceutical
Manufacturing Sites — A Pilot Risk Ranking Model, ” which allows the
agency to rank manufacturing plants ’ risk of noncompliance by using an analytical
process to (1) pose a risk question, (2) identify potential hazards and risks, (3)
characterize factors that can be used as variables for quantifying risk, and (4) mathematically
combine the variables to yield an overall risk score. Since the publication
of the report, the FDA has added adverse events reports data to the model. Surveillance
inspections are supposed to involve audit coverage of two or more systems, 8
with mandatory coverage of the quality system. 9
Compliance inspections are for the purpose of evaluating or verifying compliance
corrective actions after a problem has been identifi ed and regulatory action has
been taken. Compliance inspections cover the areas found defi cient and subjected
to corrective actions. One type of compliance inspection is a “ for - cause ” inspection,
which is conducted to investigate a specifi c problem that has come to the attention
of the FDA. The sources that trigger a compliance inspection include fi eld alert
reports, industry complaints, and recalls.
In fi scal year 2005, the FDA fi eld offi ce conducted 1437 cGMP inspections, resulting
in 15 warning letters, six injunctions, and one seizure. These enforcement actions
are discussed later in this chapter. Data for the years 2000 – 2005 are set forth in
Figures 1 and 2 .
1.2.3.2 After the Inspection: Form 483
If the inspector determines that there are deviations from cGMP, he will complete
a form FDA - 483 (Inspectional Observations) detailing the violations. The fi ndings
are presented to the manufacturer, which is given an opportunity to respond. The
FDA - 483 advises:
7 The other cGMP basic enforcement strategy is collection and analysis of drug samples during factory
inspections as well as collecting and analyzing drug products in distribution.
8 The FDA has separated the cGMP regulation into six systems: quality, facilities and equipment, production,
materials, packaging and labeling, and laboratory controls.
9 Compliance Program Guidance Manual , 7356.002, February 1, 2002.
50 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
This document lists observations made by the FDA representative(s) during the inspection
of your facility. They are inspectional observations, and do not represent a fi nal
Agency determination regarding your compliance. If you have an objection regarding
an observation, or have implemented, or plan to implement, corrective action in
response to an observation, you may discuss the objection or action with the FDA
representative(s) during the inspection or submit this information to FDA at the
address above. If you have any questions, please contact FDA at the phone number
and address above.
Most manufacturers provide a written response to the FDA - 483, either disputing
the fi ndings or addressing how they will correct the issues and how problems are to
be corrected. Negotiations typically proceed for months or years until the inspectional
problems and issues are resolved or the FDA elects to pursue elevated
enforcement. The agency retains discretion to pursue elevated enforcement if it
concludes that there is a signifi cant risk of harm to patients, with such action being
more likely where patient harm is more likely or more serious.
In addition to providing a form FDA - 483, FDA investigators prepare an establishment
inspection report (EIR), which is sent to FDA headquarters, which then
evaluates the report and determines the corrective action, if any. The FDA then
classifi es the inspection as “ no action indicated, ” “ voluntary action indicated, ” or
“ offi cial action indicated. ” The EIR contains much greater detail than contained in
the 483 and is not provided to the manufacturer until after the inspection is deemed
closed.
FIGURE 1 CDER fi ve - year Inspection data. ( Source : FDA .)
2610
2529
2585
2627
2682
2450
2500
2550
2600
2650
2700
2001 2002 2003 2004 2005
Inspections (Foreign & Domestic)
FIGURE 2 Surveillance activity. ( Source : FDA .)
2529 2585 2627 2600 2682
1982
1712
2087
1434
1548
174
362
180 260
1183
0
500
1000
1500
2000
2500
3000
2001 2002 2003 2004 2005
Inspections
Domestic Samples
Import Samples
FDA ENFORCEMENT TECHNIQUES 51
When the FDA conducted an analysis of past FDA - 483 reports, 10 the two most
reported violations were:
1. Violations of 21 CFR 211.100(b) (failure to follow and/or document production
and process control procedures), occurring in over half of all of the
483 ’ s
2. Violations of 21 CFR 122d (failure to create adequate, written responsibilities
and procedures for the quality control unit or failure to follow them), occurring
in 42% of the 483 ’ s
The next eight violations, in order of prevalence, were as follows:
• Failure to have written procedures for production and process controls.
• Failure to have testing and release of drug product for distribution for determination
of satisfactory conformance to the fi nal specifi cations/identity and
strength of each active ingredient prior to release.
• Batch production and control records were not prepared or are incomplete.
• Control procedures are not established to monitor the output/validate the
performance of manufacturing processes that may be responsible for causing
variability in the characteristics of the drug product.
• Employees were not given appropriate training.
• Laboratory controls do not include the establishment of scientifi cally sound
and appropriate specifi cations/standards/sampling plans/test procedures.
• Drug product production and control records are not certifi ed by the quality
control unit to assure compliance with all established, approved written procedures
before a batch is released or distributed.
• Procedures describing the handling of all written and oral complaints regarding
a drug product either not established or not followed.
1.2.3.3 Recalls
Chapter 7 of the Regulatory Procedures Manual (March 2007, available at www.fda.
gov ) provides detailed instructions to FDA personnel regarding recalls. The FDCA
does not authorize the FDA to “ order ” a manufacturer to recall a drug product. 11
In practice, however, the manufacturers or distributors of the drug products are
encouraged to implement and carry out recalls voluntarily to fulfi ll their responsibility
to protect the public. It is not uncommon for a company to discover that one of
its products is defective and recall it entirely on its own; or the FDA informs a
company of its fi ndings that one of its products is defective and suggests or requests
10 The data for the analysis were compiled by the FDA and derived from 614 Turbo EIR reports completed
from 2001 to 2003. (FDA investigators enter their inspections observations on the FDA ’ s Turbo
EIR system. The Turbo ’ s electronic format prompts investigators to select the specifi c cGMP violation
in question and then to explain their fi ndings uncovered during the inspection.)
11 The FDCA gives authority to the FDA to order a recall in some cases involving infant formulas,
biological products, and devices that present a “ serious hazard to health, ” but not involving
pharmaceuticals.
52 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
a recall. 12 Once a voluntary recall is initiated, the FDA generally follows the following
protocol (Figures 3 – 5 ):
1. Classify the Recall The FDA reviews relevant information and then assigns
a recall classifi cation according to the level of health risk involved:
Class I recalls involve drug products in which the reason for recall predictably
could cause serious health problems or death.
Class II recalls involve drug products which defect might cause a temporary
health problem or pose only a slight threat of a serious nature.
Class III recalls involve products that are unlikely to cause any adverse
health reaction but that violate FDA labeling or manufacturing
regulations.
2. Monitor and Audit the Recall The FDA oversees a recall depending upon the
health risk involved. For a class I recall, the FDA checks to make sure that the
defective product has been recalled in full. In contrast, for a class III recall,
FDA oversight may be to simply spot - check.
FIGURE 3 2005 recall by class. [ Source : Centre for Drug Evaluation and Research (CDER)
2005 Report to the Nation. ]
Class I: 18
Class II: 314
Class III 170
TOTAL: 502
FIGURE 4 Drug recalls. One fi rm had over 100 recalls in 2005, which caused a spike in the
2005 recall fi gures. ( Source : CDER 2005 Report to the Nation .)
191
226
248
176
352
316
248
354
254
215
401
60 53
34
88
72
156
72
83 88
71
101
0
50
100
150
200
250
300
350
400
450
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Number of Recalls
Rx
OTC
12 If the company does not comply, then FDA can seek judicial enforcement under the FDCA.
FDA ENFORCEMENT TECHNIQUES 53
3. Notifi cation and Public Warning Class I recalls almost always warrant a
press release to the media. Classes II and III are not necessarily announced
in the media, but all of them are included in the FDA ’ s weekly enforcement
report, posted at www.fda.gov/opacom/Enforce.html on the FDA ’ s
website.
4. Termination The FDA provides written notice to the recalling manufacturer
on when the recall should be terminated.
5. Noncompliance If applicable, the FDA will take appropriate legal action if
a manufacturer fails or refuses to timely complete a recall.
1.2.3.4 Warning Letter
A warning letter is intended to notify manufacturers about violations that the FDA
has documented during its inspections or investigations. A warning letter will notify
a responsible individual and/or fi rm that the FDA considers one or more products,
practices, processes, or other activities to be in violation of the cGMPs. Warning
letters should only be issued for violations of regulatory signifi cance, that is, those
that may actually lead to an enforcement action if the documented violations are
not promptly and adequately corrected. A warning letter is one of the FDA ’ s principal
means of achieving prompt voluntary compliance.
Examples of situations in which the FDA may be expected to issue a warning
letter include:
• An active pharmaceutical ingredient (API) batch fails to conform to established
specifi cations and yet the manufacturer distributed it anyway.
• Deliberately blending API batches to dilute or hide noxious contaminant or
fi lth or failing to determine actual yield and percentages of expected yields.
• Contamination of drugs with toxic chemicals, drug residues, airborne contaminants,
or fi lth.
• Failing to comply with commitments in drug applications.
• Combining a batch that does not conform with critical attributes with a batch
that does.
• Failing to demonstrate water used in the manufacturing process is suitable.
• Failing to validate water systems.
FIGURE 5 Top 10 reasons for drug recalls in fi scal year 2005. ( Source : FDA .)
• Miscellaneous cGMP deviations (other than below)
• Failed USP dissolution test requirements
• Microbial contamination of non-sterile products
• Lack of efficacy
• Impurities/degradation products
• Lack of assurance of sterility
• Lack of product stability
• Labeling: Label error on declared strength
• Misbranded: Promotional literature with unapproved therapeutic
claims
• Labeling: Correctly labeled product in incorrect carton or package
54 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
• Lacking a formal written program to validate an API validation process.
• Failing to demonstrate homogeneity of fi nal blending operations.
• Failing to keep adequate batch records.
• Failing to have a formal process change control system in place.
• Using inadequate or unvalidated laboratory test methods.
• Packaging and labeling processes that could introduce a signifi cant risk of
mislabeling.
• Failing to test for residues of organic or inorganic solvents that may carry over
to the API.
• Using incomplete stability studies to establish API stability for the intended
period of use.
Warning letters detailing cGMP violations typically conclude with the following:
“ The article(s), (DRUG NAME), is (are) adulterated within the meaning of Section
501(a)(2)(B) of the Act, 21 U.S.C. 351(a)(2)(B), in that the methods used in, or the
facilities or controls used for, its manufacture, processing, packing, or holding fails
to conform to, or is not operated or administered in conformity with, cGMP regulations
[21 CFR 210, 211]. ” The number of warning letters issued by the FDA concerning
prescription and over - the - counter drugs has ranged from 130 letters in 2000 to
79 letters in 2005.
A warning letter is distinguishable from a notice of violation, also called an
untitled letter. An untitled letter cites violations that do not meet the threshold of
regulatory signifi cance for a warning letter, but the FDA has a need nevertheless to
communicate. Unlike a warning letter, an untitled letter does not include a warning
statement that failure to take prompt correction may result in enforcement action
and does not evoke a mandated FDA follow - up. Further, the untitled letter requests
(rather than requires) a written response (from the manufacturer) within a reasonable
amount of time (e.g., “ Please respond within 45 days ” ).
1.2.4 JUDICIAL ENFORCEMENT: BEYOND THE WARNING LETTER
1.2.4.1 Introduction
The FDA is likely to bypass sending a Warning Letter in certain circumstances.
According to Chapter 4 of the FDA Regulatory Procedures Manual , the following
violations are likely to result in an enforcement action without necessarily issuing
a warning letter:
1. The violation refl ects a history of repeated or continual conduct of a similar
or substantially similar nature during which time the individual and/or fi rm
has been notifi ed of a similar or substantially similar violation.
2. The violation is intentional or fl agrant.
3. The violation presents a reasonable possibility of injury or death.
4. Adequate notice has been given by other means and the violations have not
been corrected or are continuing.
5. The violations, under Title 18 U.S.C. 1001, are intentional and willful acts that
once having occurred cannot be retracted. Also, such a felony violation does
not require prior notice. Therefore, Title 18 U.S.C. 1001 violations are not suitable
for inclusion in warning letters.
In addition, actively deceiving the FDA is almost guaranteed to bring judicial
enforcement actions. This includes false representations in the written record -
keeping requirements or in written communication with the FDA. Manufacturing
record - keeping requirements which give exposure to fraud liability are summarized
in Figure 6 . Potential violations include the following:
FIGURE 6 Written record highlights.
Written records are required to be kept as set forth in 211.180 to 211.208. Highlights are as follows:
§ 211.182 Equipment cleaning and use log.
A written record of major equipment cleaning, maintenance (except routine
maintenance such as lubrication and adjustments), and use shall be included in
individual equipment logs that show the date, time, product, and lot number of each
batch processed.
§ 211.184 Component, drug product container, closure, and labeling records.
These records shall include the following:
(a) The identity and quantity of each shipment of each lot of components, drug
product containers, closures, and labeling; the name of the supplier; the supplier's
lot number(s) if known; the receiving code as specified in § 211.80; and the date
of receipt. The name and location of the prime manufacturer, if different from the
supplier, shall be listed if known.
(b) The results of any test or examination performed (including those performed
as required by § 211.82(a), § 211.84(d), or §211.122(a)) and the conclusions
derived therefrom.
§ 211.186 Master production and control records.
To assure uniformity from batch to batch, master production and control records
for each drug product, including each batch size thereof, shall be prepared, dated,
and signed (full signature, handwritten) by one person and independently checked,
dated, and signed by a second person. The preparation of master production and
control records shall be described in a written procedure and such written
procedure shall be followed.
§ 211.188 Batch production and control records.
Batch production and control records shall be prepared for each batch of drug
product produced and shall include complete information relating to the production
and control of each batch.
§ 211.194 Laboratory records.
Laboratory records shall include complete data derived from all tests necessary to
assure compliance with established specifications and standards, including
examinations and assays ...
§ 211.198 Complaint files.
A written record of each complaint shall be maintained in a file designated for drug
product complaints
JUDICIAL ENFORCEMENT: BEYOND THE WARNING LETTER 55
56 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
(a) Accepting and validating drug products that failed to meet established standards
or specifi cations and any other relevant quality control criteria (i.e.,
dissolution rates, content uniformity, purity, potency) and then falsely recording
the untruthful data as if the drug products did not fail
(b) Accepting and validating the stability characteristics of drug products and
then falsely recording the untruthful data as if the drug products did not
fail
(c) Documenting the examination and review of labels, when in truth and fact
no review occurred (which results in inaccurate labels distributed with
drugs)
(d) Falsely documenting any components of master production and central
records
(e) Falsely documenting any component of the batch production and control
records
(f) Falsely describing testing methods when no (or inadequate) testing methods
were performed
(g) Failing to accurately make a written record of all written and oral complaints
regarding a drug product and/or certifying that investigations were performed
when they were not, falsely certifying that the fi ndings were negative
when they were not, and so on
(h) Falsifying records which would indicate manufacturing changes which require
approval by the FDA
(i) False representations that contain statements of fact in correspondence sent
to the FDA addressing violations in an inspector ’ s form 483
The FDA typically initiates progressive enforcement, as described in Figure 7 .
Once the FDA and DOJ decide to bring enforcement action, the U.S. courts have
held that the FDA ’ s interpretation of its cGMPs is entitled to substantial deference.
As long as the FDA ’ s interpretation of its regulations are “ reasonable ” and “ sensibly
conforms to the purpose and wording of the regulations, ” courts are required to
follow the FDA ’ s interpretations.
1.2.4.2 Civil Proceedings
Seizures If during an inspection of a facility the FDA inspector or employee
making the inspection has reason to believe that a drug found in such facility is
adulterated, such inspector or employee may order the drug detained for a reasonable
period which may not exceed 20 days (unless the FDA institutes an action
under Subsection 334(a) or an injunction, in which case a longer detention period
may be authorized).
The FDCA expressly permits administrative seizure on the basis of an ex parte
showing of reasonable belief [21 U.S.C. 334 (g)]. Seizure of a company ’ s inventory
deprives the company of both capital investment and potential profi t.
If the FDA pursues relief beyond detainment, the United States can fi le a complaint
for forfeiture directing the U.S. marshall to “ seize ” the pharmaceuticals (or
take possession or place in constructive custody of the court). The theory in a complaint
for forfeiture is that there is a violation of the law by the pharmaceutical
product itself. Accordingly, the government asks the court to condemn the article
and declare forfeiture. Upon fi ling of the complaint, the clerk automatically issues
a warrant. Thus, the FDA is able to obtain a warrant without review by a judicial
offi cer or even a fi nding of probable cause.
There are three types of seizures: mass, open ended, and lot specifi c. A mass
seizure is the seizure of all FDA - regulated products at an establishment/facility.
Mass seizures might be conducted when all of the products are produced under the
same conditions (e.g., nonconformance with cGMPs). An open - ended seizure is the
seizure of all units of a specifi c product or products, regardless of lot or batch
number, when the violation is expected to be continuous. An open - ended seizure
may be conducted when a specifi c product extends to all lots or batches of a product
but not to all of the products in the facility.
Following seizure of its drugs a manufacturer has three courses of action. First,
it may do nothing, in which case the drug will be disposed of. Second, it can enter
into a consent decree, admitting the violation, agreeing to pay costs, and seeking to
destroy or rehabilitate the article. The consent decree will typically provide for
(1) condemnation of the article as being in violation of the law; (2) a penal bond in
approximately twice the retail value of the article under seizure; (3) provisions for
payment of costs for storage and handling by the U.S. marshall and for supervision
by the FDA before release of the product; and (4) a provision that the manufacturer
will attempt to bring the article into compliance under the supervision of and to the
satisfaction of the FDA. 13
Third, it can contest the action. If the manufacturer contests the action, the case
is then treated like any other civil case under the federal rules of civil procedure,
and the government must prove its case by a preponderance of the evidence. The
government must produce evidence, in support of its allegations, including proof of
interstate shipment of the drug or its components. FDA employees may testify, but
FIGURE 7 Progressive enforcement.
FDA enforcement mechanisms are often utilized progressively. A good example is the
enforcement action against Glaxo SmithKline (“GSK”), which began in July 2002, identifying
numerous significant cGMP violations found during a February/April 2002 inspection. A Warning
Letter requested that the violations be corrected and stated that failure to correct the violations may
result in regulatory action, including seizure and/or injunction. Although a limited follow-up FDA
inspection in October 2002, found that some specific corrections were acceptable, the subsequent
FDA inspections in November/December 2003 and September/November 2004, revealed continuing
significant cGMP violations. FDA concluded that the firm’s data and corrective plans were not
adequate to correct the cGMP violations. GSK also initiated recall of some, but not all, lots of the
two products. On March 4, 2005, in response to ongoing concerns about manufacturing quality,
FDA and the DOJ initiated seizures of two GSK pharmaceuticals. The Agency initiated these
seizures actions based on concerns that GSK’s violation of manufacturing standards may have
resulted in the production of poor quality drug products that could potentially pose risks to
consumers. On April 28, 2005, FDA announced that GSK had signed a Consent Decree with FDA
to correct manufacturing deficiencies at its Cidra, Puerto Rico, facility. The Consent Decree was
initiated based on FDA’s continued concerns that GSK’s violation of manufacturing standards may
have resulted in the production of drug products that could potentially pose risks to consumers.
13 Regulatory Procedures Manual , Chapter 6 - 1 - 11, March 2007. This contemplates that seizure of a specifi c
product(s) is the sole issue. More complex consent decrees are described hereinafter.
JUDICIAL ENFORCEMENT: BEYOND THE WARNING LETTER 57
58 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
also outside experts testify such as to the signifi cance of failure to comply with
cGMP requirements. If a decree of condemnation is entered (either after trial or by
consent), the court may direct disposition of the article by destruction.
Injunctions The FDCA expressly authorizes the courts to restrain and enjoin acts
that are in violation of 21 U.S.C. 331, which includes prohibition of adulterated
products. FDA policy provides that an injunction action is appropriate where:
(a) there is a current and defi nite health hazard or a gross consumer deception
requiring immediate action to stop the violative practice;
(b) there are signifi cant amounts of violative products owned by the same person
in many locations, voluntary recall by the fi rm was refused or is signifi cantly
inadequate to protect the public, and seizures are impractical or uneconomical;
or
(c) there are long - standing (chronic) violative practices that have not produced
a health hazard or gross consumer fraud, but which have not been corrected
through use of voluntary or other regulatory approaches. 14
A complaint for injunction is typically accompanied by a motion for preliminary
injunction. 15 The court schedules a court hearing to determine whether to grant a
preliminary injunction, often very quickly and on short notice. The government ’ s
main focus at this preliminary stage will be to prove that there is a “ substantial
likelihood ” that the defendant has been producing adulterated drugs in violation
of 21 U.S.C. 331, by substantial noncompliance with the cGMPs. The government
will also typically present evidence, if applicable, that the defendant has had
a history of prior noncompliance with the FDCA and implementing regulations.
No specifi c fi nding of irreparable harm is necessary as is required in the typical
injunction, because the passage of the statute proscribing adulterated products has
15 The government may also apply for a temporary restraining order (TRO) seeking immediate, temporary
relief (for a period of 10 days, which may be extended for 10 additional days) prior to the hearing
for preliminary injunction. The FDA will typically recommend a TRO when it believes that the violation
is so serious that it must be controlled immediately .
14 Regulatory Procedures Manual , March 2007.
FIGURE 8 Disgorgement.
Major recent consent decrees are United States v. Abbott Labs., Consent Decree of
Permanent Injunction filed Nov. 2, 1999; United States v. Various Articles of Drug Identified in
Attachment A & Wyeth-Ayerst Labs., Consent Decree of Condemnation and Permanent Injunction
filed Oct. 4, 2000; and United States v. Schering-Plough Corp., Consent Decree of Permanent
Injunction filed May 20, 2002. To avoid giving manufacturers the wrong message by allowing them
to keep on the market what FDA had determined to be produced in violation of the cGMPs, the FDA
included three separate types of “disgorgement” payments in the Abbott, Wyeth and Schering
consent decrees: (1) a lump sum payment (Abbott, $100 million; Wyeth, $30 million; and Schering,
$500 million) (2) if the remedial work was not achieved by the deadline established in the decree,
(i) a percentage of sales (Abbott, 16%; Wyeth, 18.5%; and Schering, 24.6%) and (ii) daily payments
of a certain flat amount. Both were to be paid until compliance was achieved.
been held itself to be an implied fi nding by Congress that violations will harm the
public.
United States courts are imbued with authority to enjoin present and future
violations of Section 331 based upon proof by the FDA that such violations have
occurred and could recur. Factors that courts consider when determining whether
there is a reasonable chance of future infractions include (1) the degree of scienter
involved on the part of the defendant; (2) the isolated or recurrent nature of the
infraction; (3) the defendant ’ s recognition of the wrongful nature of his or her
conduct; (4) the sincerity of the defendant ’ s assurances against future violations;
and (5) the nature of the defendant ’ s violation. The court also considers whether
the defendant voluntary ceased the challenged conduct, the genuineness of the
defendant ’ s efforts to conform to the law, the defendant ’ s progress toward improvement,
and the defendant ’ s compliance with any recommendations made by the
government.
Good faith is not a defense to the issuance of an injunction. Nor may a defendant
successfully defend against the issuance of an injunction by asserting that the injunction
would drive it out of business.
Consent Decrees and Disgorgement A consent decree is a judgment (legal order)
issued by the court that has been agreed to by the parties whereby the defendant
agrees to stop illegal or improper activity as alleged by the government. Once court
approval is obtained, the seizure or injunctive lawsuit, for instance, is dropped, and
the government ’ s remedy is then based upon any breach of the consent decree, itself
which is enforceable by the court.
Consent decrees typically involve a defendant agreeing to address the areas of
noncompliance in a manner satisfactory to the FDA within a certain amount of time.
It can also provide for the hiring of an expert consultant to certify in detailed reports
that the manufacturing facility, at periodic dates, is in full compliance with the
cGMPs, and has adequate adverse - event controls, adequate training, and adequate
recall procedures. It may also require the payment of money to the U.S. Treasury
such as under the equitable remedy of “ disgorgement, ” as described in Figure 8 .
As part of a court action, the FDA will sometimes pursue “ disgorgement. ” The
purpose of disgorgement is to deprive the wrongdoer of ill - gotten gains as well as
provide deterrence. The amount of disgorgement is not necessarily directly tied to
restitution. In practice, the amount the FDA exacts is supposed to be enough to
send a message but certainly does not provide for full disgorgement of profi ts of
the drug product(s) at issue.
False Claims Act The U.S. Civil False Claims Act, 31 U.S.C. 3729 et seq., is the
government ’ s principal means of redressing fraud by government contractors. The
act has implications for cGMP violations because the United States (funding as it
does the Medicare program, the state Medicaid programs, the Veterans Administration,
the TRICARE program, and others) is the world ’ s largest purchaser of prescription
medications.
Nonetheless, the government has yet to bring a False Claims Act case which seeks
damages for cGMP. One reason may be that the government has multiple other
remedies within which to recover damages from noncompliant manufacturers, such
as criminal fi nes and penalties and disgorgement.
JUDICIAL ENFORCEMENT: BEYOND THE WARNING LETTER 59
60 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
Qui tam whistleblowers, 16 however, have already begun bringing such cases.
Because the False Claims Act imposes liability on any government contractor which
knowingly submits false claims to the United States or which uses false documents
to get a false claim paid, a pharmaceutical manufacturer which knew or was recklessly
indifferent to the fact that the manufacturing process was compromised by
cGMP violations is in the same position as any other contractor which is required
to conform to contractual or regulatory standards. The basis of liability under the
False Claims Act is that false records have been generated which caused (false)
claims for drugs to be paid by the United States. 17 The monetary damages result
because the payor (in this case, the United States) is potentially paying for substandard
drugs due to the cGMP violations — later covered up by false statements in
documents required to be completed under the cGMP.
It makes sense, too: The cGMPs are a set of regulations which, by their very
nature, are designed to ensure that drugs are manufactured in such a way that they
meet the requirements of the federal Food, Drug and Cosmetic Act as to safety and
have the identity and strength and meet the purity characteristics that they purport
or are represented to possess. The major federally funded government health care
programs, Medicare and Medicaid, operate under the express provisions that they
will only pay for medical services and products that are “ reasonable and necessary. ”
Unsafe or ineffective drug products are neither reasonable nor necessary. Accordingly,
as the theory goes, the United States suffers monetary damages if Medicare
and Medicaid programs pay for unsafe or less effective products. These and other
federally funded health care programs spend billions of dollars every year on
pharmaceuticals.
False representations concerning minor or technical violations will not be the
basis for FCA liability. Distribution of products that are not totally cGMP compliant
(but have been falsely documented to be) does not necessarily result in unsafe (or
subpotent) products. Substantial violations of the cGMP, later covered up in writing,
however, could very well be the basis for FCA liability. The common thread through
each violation is that the violation is severe enough so that the drug product that
16 Qui tam is shorthand for the Latin phrase, qui tam pro domino rege quam pro seipso , meaning “ He
who is as much for the king as for himself. ” Qui tam statutes date back to thirteenth - century England.
The actions were a means of enabling private parties to allege the king ’ s interest and therefore gain
access to the royal courts.
The qui tam provisions of the federal False Claims Act allow any citizen who has knowledge of fraud
that has taken place against the government to bring a civil action in federal court in the name of the
United States. In return for his or her efforts, the citizen is entitled to share in the proceeds of the recovery.
The qui tam provisions raise the incentive for insiders to put the spotlight on the criminals, thereby
providing the government with tangible and detailed evidence upon which to base an investigation and
prosecution.
In 1986, Congress enacted amendments to the False Claims Act which strengthened the law and
increased monetary awards. When hearings were held in 1985 and 1986, the climate was favorable for
strengthened antifraud legislation, and Congress expected that most qui tam cases would involve defense
contractor fraud. In the last decade, the majority of cases have instead been against the health care
industry.
17 Even so, factual questions will be raised, including: (1) Even with the false representations, was a false
claim “ caused ” to be submitted? (2) Had the FDA known about the falsities, would it have enjoined the
manufacturer from any further production, etc? (3) What about the false record or statement made the
claims for such drugs false?
fi nally reaches the public is foreseeably and substantially less safe or less effective
than if the cGMPs were not violated.
1.2.4.3 Criminal Proceedings
Introduction Criminal prosecutions of violations of the FDCA are intended to
further the goal of protecting the health and safety of the public. The FDA historically
has not pursued criminal charges unless the defendant shows a continuous or
repetitive course of violative conduct, with the exception of intentional violations,
fraud, or danger to health. 18
While the FDCA contains various prohibitions and restrictions which a drug
company could violate, the most common FDCA violation arising out of cGMPs
is charged by using 21 U.S.C. 331(a), which specifi cally prohibits introducing an
adulterated 19 drug into interstate commerce. In addition to introducing an adulterated
drug into interstate commerce, some other acts prohibited by Section 331(a)
which could be involved in manufacturing violations include 331(e), which prohibits
the refusal to allow access to records mandated elsewhere in the act and 331(f),
which prohibits the refusal to allow inspection of production facilities (21 U.S.C.
374).
Commission of any act prohibited by Section 331 is a federal misdemeanor (21
U.S.C. 333). However, violations of Section 331(a) may be charged as felonies where
there is intent to defraud or mislead or where the defendant previously has been
convicted of a misdemeanor under the FDCA [21 U.S.C. 333(b)]. Federal misdemeanor
charges are typically resolved in proceedings before U.S. magistrate judges,
and federal felonies are resolved by U.S. district judges.
Individual versus Corporate Liability Introducing an adulterated product into
interstate commerce is a strict liability crime that can be enforced against individuals
in positions of suffi cient authority and responsibility as well as their company.
Persons at risk are those who, at minimum, 20 fail to take adequate measures to
prevent the cGMP violations. As such, warning letters and other communications
are often directed at presidents and CEOs as well as their companies. As stated by
the U.S. Supreme Court 21 in 1964, just two years after the FDCA as we know it was
passed:
Food and drug legislation, concerned as it is with protecting the lives and health of
human beings, under circumstances in which they might be unable to protect themselves,
often “ dispenses with the conventional requirement for criminal conduct —
awareness of some wrongdoing. In the interest of the larger good it puts the burden of
acting at hazard upon a person otherwise innocent but standing in responsible relation
to a public danger. . . . ”
18 A government review of recent FDA enforcement has suggested that adequate FDA enforcement
activity is lacking. See “ Prescription for Harm: The Decline in FDA Enforcement Activity, ” House Committee
on Government Reform, June 2006.
19 Failure to follow cGMP is the most common form of violating the prohibition against introducing an
adulterated drug into interstate commerce.
20 They may also be directly implicated in fraud and cover - ups.
21 United States v. Wiesenfeld Warehouse Co. , 376 U.S. 86, 91 (1964).
JUDICIAL ENFORCEMENT: BEYOND THE WARNING LETTER 61
62 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
In 1975, the Supreme Court made clear that individual responsibility is very
important 22 :
The [FDCA] imposes not only a positive duty to seek out and remedy violations when
they occur but also, and primarily, a duty to implement measures that will insure that
violations will not occur. The requirements of foresight and vigilance imposed on
responsible corporate agents are beyond question demanding, and perhaps onerous,
but they are no more stringent than the public has a right to expect of those who voluntarily
assume positions of authority in business enterprises whose services and products
affect the health and well - being of the public that supports them.
Manufacturing executives therefore carry a great liability burden. The government
only need establish that the individual defendant failed to act on his or
her own authority and that such an action could have prevented or corrected
the violation. The individual need not have formed any intent to break any laws in
order to be found guilty. What is relevant is, did the executive have the power to
prevent the acts or omissions complained of? This includes a consideration of
whether the executive could have prevented the acts or omission by the systems
and processes alone. The job is not made easier to the extent that the cGMP regulations
are open to varying interpretations or that the technology is constantly
changing.
Section 305 Proceedings Due to the nature of the inspection process, a company
that the FDA deems is in violation of the cGMPs should not be surprised when a
warning letter or more elevated enforcement techniques are implemented. Even so,
the FDA sometimes issues a formal form of notice that criminal charges will be
brought by what is called a Section 305 notice.
Section 305 of 21 U.S.C., the statutory basis for a 305 notice, seemingly requires
that before any violation of the FDCA is reported to the DOJ for institution of a
criminal proceeding, the target defendant must “ be given appropriate notice and an
opportunity to present his views, either orally or in writing, with regard to such
contemplated proceeding. ” The U.S. Supreme Court has watered down this provision,
holding that a notice under Section 305 is not a legal prerequisite to government
prosecution.
In practice, then, the FDA only sometimes issues a 305 notice and conducts a 305
hearing when it is considering a misdemeanor prosecution. A very informal process,
the manufacturer can approach it with as much or as little of a defense as counsel
deems appropriate, as there are pros and cons to providing the government with
the company ’ s full defense at that juncture.
A prototype Section 305 notice appears in Figure 9 .
Grand Jury Proceedings If the government will be pursuing felony criminal
charges against a manufacturing facility or persons associated with such facility, it
will proceed by grand jury. The Fifth Amendment to the U.S. constitution requires
that charges for all capital or “ infamous ” crimes be brought by an indictment
22 United States v. Park , 421 U.S. 658 (1975).
returned by a grand jury. This has been interpreted by the U.S. courts to require that
an indictment be used to charge federal felonies.
The activities, deliberations, or matters occurring before a grand jury are secret. 23
Strict adherence to grand jury secrecy is important to the integrity of the investigative
process and ensures that the grand jury will be able to deliberate without
outside pressure, to encourage people with information about a crime to come
forward without fear of disclosure, and to protect the rights of the accused, specifi -
cally the innocent accused, from disclosure of the fact that he or she or it was
investigated. Other than attorneys for the government, only the witness, interpreters
when needed, and a court reporter are authorized persons permitted to be present
while a grand jury is in session.
A grand jury ’ s function is to determine whether there is probable cause to believe
that a certain person(s) or company(ies) have committed a federal offense. 24 Prosecutors
are permitted to appear before the grand jury and, in practice, conduct the
grand jury proceedings. In general, the prosecutor is the one who makes the decision
FIGURE 9 Section 305 notice.
In reply refer to:
Sample No.
Product
Firm Name and Individual Date
Street Address
City, State, Zip
Investigation by this Administration indicates your responsibility for violations of the
Federal Food, Drug, and Cosmetic Act, and other Federal Laws, as described in the
attached Charge Sheet, with respect to the following:
[describes specifics of cGMP violations]
A meeting has been scheduled for (day, date, time) at (location), to give you an
opportunity to present your views on this matter. The enclosed INFORMATION SHEET
explains the purpose and nature of the meeting, and how you may reply. If no response
is received on or before the date set, our decision on whether to refer the matter to the
Department of Justice for prosecution will be based on the evidence in hand.
By direction of the Secretary of the Department of Health and Human Services:
Compliance Officer
Enclosures:
Legal Status Sheet (3)
Charge Sheet
Information Sheet
Regulations
23 See Rule 6(e) Fed. R. Crim. Pro.
24 The grand jury system is not presently used by countries outside the United States. The United
Kingdom, New Zealand, Canada, and Australia, for instance, all have abolished the use of grand juries.
See http://enwikipedia.org/wiki/Grand_jury .
JUDICIAL ENFORCEMENT: BEYOND THE WARNING LETTER 63
64 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
as to which witnesses to call and what evidence should come before the grand jury.
The prosecutor asks the witness questions and subsequently members of the grand
jury may also question witnesses directly or through the prosecutor.
During the course of a grand jury investigation regarding cGMP violations/adulterated
product allegations, for instance, the grand jury may hear testimony from
not only federal investigators, federal agents, and federal inspectors but also former
employees of the company (or current if a custodian of records) and/or experts in
the pharmaceutical manufacturing fi eld. These persons are considered witnesses.
Witnesses are typically subpoenaed and may not refuse to appear before the grand
jury or be subject to contempt charges.
In the federal grand jury system, a witness is not permitted to bring an attorney
into the grand jury room. However, a witness is permitted to consult with his attorney
outside the grand jury room even interrupting his own testimony. 25 It is typical
for corporations such as pharmaceutical manufacturing companies to provide an
attorney for any and all employees subpoenaed by a grand jury of which the manufacturer
is a target.
A target is the person who is the focus of the grand jury investigation and is likely
to be indicted. This company or person may receive a “ target letter ” from the grand
jury which offi cially advises them of their jeopardy and serves as a formal warning
of their status.
In practice, if a manufacturer is the target, the government will likely attempt
to develop evidence by subpoena of persons and materials which will help
prove culpability. It is likely the subpoenas will ask for correspondence, notes,
and memos during a particular time period and involving a particular subject
matter.
The grand jury may also issue a subpoena to the manufacturer ’ s designated
“ custodian of records ” for specifi c document production. The description of subpoenaed
documents can include statements or charts of an organization,
announcements, statements of policy and procedure, diaries, records of email,
manufacturing logs, emails, travel vouchers, fi nancial records and statements,
correspondence, notes of conversations, and any other documents that relate to the
manufacturing of certain drugs. A document subpoena may also request every
writing or record of whatever type and description in the possession, custody, or
control of the company that relates to a particular element of the criminal violation
the grand jury is investigating. The request typically includes all handwritten, typed,
printed, recorded, or transcribed records, including computer records tapes or
disks.
The burden is on the government to prove that the crime was committed in the
district in which the prosecution is brought. The grand jury should not consider a
case unless venue lies in the district where the grand jury is sitting. 26 In the case of
adulterated drugs, courts have generally held that it is proper to have venue in a
district from which the defendant caused the unlawful introduction of goods into
commerce, even though the physical shipment commenced from a different
district.
25 See Rule 6(d) Fed. R. Crim. Pro.; 28 U.S.C. secs. 515, 542, 547.
26 See Rule 18 Fed. R. Crim. Pro.
Form of Charges and Penalties A grand jury investigation may culminate in the
return of an indictment. This means that the grand jury found probable cause to
believe that a violation of law occurred. While the focus of the initial inquiry can
surround adulterated drugs by virtue of failure to abide by the cGMPs, additional
criminal violations may be charged, as, for instance, where there are actions to evade
or mislead a grand jury. The end result could, for instance, include accusations of
making false statements to the FDA and obstructing the FDA ’ s or DOJ ’ s investigation,
in addition to the “ adulterated ” drug charges.
An indictment consists of a statement describing the time, place, and manner
through which the defendant violated the law. Each violation of the law is set out
in a separate count. A defendant charged by an indictment is entitled to a trial by
jury, although this right can be waived. A defendant has the right to a trial by
jury for any criminal offense punishable by imprisonment for more than six
months. 27
If the matter only involves a misdemeanor violation, the prosecutor charges
“ by information. ” The information is often referred to as a complaint. An information,
like an indictment, is simply a pleading that accuses the defendant of committing
crimes. The distinction between an information and an indictment is that a
prosecutor can issue and fi le an information without the grand jury ’ s participation
or fi nding of probable cause but a grand jury must approve and return an
indictment. 28
The penalty for a violation of 21 U.S.C. 331(b) by violating the cGMPs resulting
in the adulteration of drug products in interstate commerce is set forth in 21 U.S.C.
sec. 333(a). Each separate count for violating the cGMPs (where a misdemeanor is
charged) carries with it a possible imprisonment of not more than one year or a fi ne
not to exceed $ 1000 or both. If the government charges a felony for violation of the
cGMPs, then the penalties are imprisonment of not more than three years or a fi ne
not to exceed $ 10,000 or both. Of course, there may be other charges with greater
or lesser penalties which are not related to the adulterated drug charges.
The criminal fi ne amounts (but not the imprisonment durations), however, are
superceded by the criminal fi ne amounts contained in a different federal statute
enacted later. Section 3571 of 18 U.S.C. provides for much greater fi nes than those
provided for within the FDCA itself. For a manufacturer convicted of a felony, the
fi ne can be as much as $ 500,000; for a misdemeanor (not resulting in death), it could
be $ 200,000. Fines for individuals include a maximum up to $ 150,000 for a felony
and an amount up to $ 100,000 for a conviction of a misdemeanor not resulting in
death. The statute also provides for a multiplier of 2 based upon a fi nding that the
defendant derived a pecuniary gain from the offense. The FDA and DOJ are able
to elevate the monetary recoveries against the manufacturers for violations of the
cGMP under a civil disgorgement theory explained infra. Recoveries have been in
the hundreds of millions in recent years, typically agreed to in a negotiated consent
decree.
27 See Sixth Amendment, U.S. Constitution.
28 Sometimes, prosecutors are in communication with defendants and their counsel during the investigatory
stage. If there are negotiations concerning a plea to a felony and they are successful, a defendant
can waive his or her right to be indicted by a grand jury, and the prosecutor can charge them by
information.
JUDICIAL ENFORCEMENT: BEYOND THE WARNING LETTER 65
66 ENFORCEMENT OF CURRENT GOOD MANUFACTURING PRACTICES
1.2.5 CONCLUSION
The diligent enforcement of good manufacturing practices is a cornerstone of the
safety net for drugs in the United States. Congress, the courts and, manufacturers,
most importantly, expect a degree of consistency and responsibility in enforcement
policy over a statute as powerful and central to public health and safety as the
FDCA. To the extent there is consistency and effective and evenhanded enforcement,
it not only protects the public, but it provides a level playing fi eld for those
manufacturers who operate in accordance with the cGMPs.
67
1.3
SCALE - UP AND POSTAPPROVAL
CHANGES (SUPAC) REGULATIONS
Puneet Sharma , Srinivas Ganta , and Sanjay Garg
University of Auckland, Auckland, New Zealand
Contents
1.3.1 Introduction
1.3.2 Scientifi c and Regulatory Rationale for SUPAC
1.3.2.1 Supporting Documents and Extent of Change
1.3.2.2 Supporting Documents for Change in Specifi cations
1.3.2.3 Comparability Protocols
1.3.2.4 In Vitro – In Vivo Requirements
1.3.3 Regulatory Agencies and Guidelines
1.3.3.1 FDA SUPAC Regulations
1.3.3.2 Regulatory Guidance on SUPAC by Pharmaceutical Unit of EU
1.3.3.3 Regulatory Guidance on SUPAC by Agencia Nacional de Vigilancia
Sanitaria
1.3.4 Harmonization
1.3.5 GMP Issues: Change Control and Process Validation
1.3.5.1 Change Control
1.3.5.2 Process Validation
1.3.6 Conclusion
1.3.1 INTRODUCTION
Product development aims at formulating active drug ingredient in a palatable form.
Technology transfer of a pharmaceutical product from research to the production
fl oor (referred to as “ shop fl oor ” ) with simultaneous increase in production outputs
is commonly known as scale - up. In simple terms, the process of increasing batch size
is termed as scale - up. Conversely, scale - down refers to decrease in batch size in
response to reduced market requirements.
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
68 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
Often, changing of scale from the research lab to the shop fl oor is fraught with
problems. The basic reason for such problems is the usage of different processing
equipment in research and on the shop fl oor. Moreover, insuffi cient information
about the equipment, various requirements of process control, complexity of a particular
pharmaceutical process which may have a several different unit operations,
limited information about the behavior of ingredients at different scales, and adoption
of trial - and - error methodology also add signifi cantly to scale - up issues. Every
product coming from research should be manufacturable and the process should be
capable to demonstrate its ruggedness at the shop fl oor level. This statement points
toward the criticality and signifi cance of scale - up and technology transfer in a pharmaceutical
development process. After successful accomplishment of technology
transfer and validation activity, a product usually has a smooth run on large - scale
production machines. Changes are being made in the manufacturing process and
chemistry of a drug product following approval and continue throughout its life.
Depending upon foreseen (or unforeseen) requirements, there can be changes in
the raw materials, process, equipment or manufacturing site, and batch size which
ultimately affect quality attributes of a drug or fi nished product. Therefore, there is
a need to anticipate and fully evaluate the impact of any kind of change on the
quality of a drug or fi nished product. There can be several reasons for these changes,
such as changed market requirement affecting batch size, new source of raw material,
change in manufacturing process, upgrades of packaging material, or shifting
to a new analytical methodology.
The intensity of the adverse effect produced by a particular change depends on
the type of dosage form. For example, a change in the inactive ingredient beyond a
certain range will have more effect on a modifi ed - release (MR) dosage form than
it would on an immediate - release (IR) dosage form, where bioavailability is not rate
limiting. Likewise, a change in the primary packaging of liquid parenteral may have
more pronounced effect on its effectiveness than it would have on a solid dosage
form. Hence, depending upon the intensity of change or the adverse effect it may
have on the critical parameters of a dosage form, reporting requirements to regulatory
authorities also vary.
A drug or drug product may experience many changes during its life cycle. These
changes may have an adverse effect on the overall safety and effectiveness of the
drug or drug product. After a number of changes over a long time period, the
product coming to market may be completely different from the one that was
approved. Hence, data submitted to regulatory authorities in support of a change
must have a comparison record of the drug or drug product to the one that was
approved initially. Documentation generated in support of any change to the
approved drug or drug product is submitted to regulatory authority for review, and
based on the benefi t - to - risk ratio, the drug or drug product is approved. Depending
upon the intensity of change, supporting documents are provided to the regulatory
agency.
Regulatory authorities such as the U.S. Food and Drug Administration (FDA),
the European Commission, the Agencia Nacional de Vigilancia Sanitaria (ANVISA)
(in english the National Health Surveillance Agency — Brazil, and others require the
pharmaceutical industries in respective countries to follow guidelines on scale - up
and postapproval changes (SUPAC) to maintain the quality of the pharmaceutical
produced. From time to time these guidelines are assessed so as to keep pace with
SCIENTIFIC AND REGULATORY RATIONALE FOR SUPAC 69
the technological advances and new guidelines are developed to reduce the burden
on the pharmaceutical industry and regulatory authorities. Apart from these guidelines,
there are other checkpoints within an industry to assure production of quality
products, such as change control and validation exercises, which will be discussed in
detail in this chapter. These operations are controlled through the principles of good
manufacturing practices issued by regulatory authorities.
This chapter describes the regulations imposed by different regulatory authorities
and measures taken by a pharmaceutical industry to assure quality and performance
of pharmaceuticals. The FDA guidelines, being most descriptive, have been
discussed at length. Other guidelines have been described in general terms and the
interested reader is referred to the references or the regulatory websites for more
specifi c details.
1.3.2 SCIENTIFIC AND REGULATORY RATIONALE FOR SUPAC
Guidelines pertaining to postapproval changes classify these changes in various
categories depending upon the effect a particular change may have on the quality
and performance of a drug or drug product. Irrespective of the terminologies used
by regulatory agencies, in general terms, changes can be described as mild, moderate ,
and major and the extent of supporting document varies with the nature of the
change. For example, U.S. FDA guideline “ Changes to an Approved NDA or ANDA ”
describe these changes as mild changes that can be implemented immediately and
fi led in the next periodic report, moderate changes that can be implemented immediately,
moderate changes that require 30 days notice before implementation, and
major changes that require FDA approval before implementation [1] . Similarly, any
changes in an approved drug or drug product under European Union (EU) domain
type I (type IA and type IB) and II variation are fi led prior to marketing products
[2] . The therapeutics Good Administration — Australia (TGA) describes postapproval
changes in three categories: nonassessable, self - assessable, and changes
requiring prior approval [3] .
1.3.2.1 Supporting Documents and Extent of Change
As per FDA guidelines, changes in excipients (%w/w) of total formulation not
greater than 5% are considered minor and all information is provided in the annual
report. However, changes likely to have signifi cant effect on the quality and performance
of a drug product calls for submission of a prior approval supplement on all
information (in vitro dissolution and in vivo dissolution), including accelerated stability
and long - term stability testing in the annual report [4] . Similarly, in EU guidelines,
a change in the batch size of the fi nished product up to10 - fold compared to
the original batch size approved at the grant of the marketing authorization (or
downscaling to 10 - fold) has been defi ned as type IA and requires batch analysis
data (in a comparative tabulated format) on a minimum of one production batch
manufactured to both the currently approved and the proposed sizes. Batch data
on the next two full production batches should be made available upon request and
reported by the marketing authorization holder if outside specifi cations (with proposed
action). However, for type IB (more than 10 - fold), in addition to the above
70 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
data, a copy of an approved release and end - of - shelf - life specifi cations as well as
the batch numbers ( . 3) used in the validation study should be indicated or a validation
protocol (scheme) be submitted and the number of batches used in the stability
studies should be indicated.
1.3.2.2 Supporting Documents for Change in Specifi cations
Changes in any type of specifi cation also need to be supported by documentation.
In all the guidelines, relaxing an acceptance criterion or deleting any part of the
specifi cation is classifi ed as a major change and hence extensive documentation is
required, for example, submission of a prior approval supplement to the FDA or
comparative table of current and proposed specifi cations and details of any new
analytical method and validation data and batch analysis data on two production
batches of the fi nished product for all tests in the new specifi cation to EU. The
specifi cations are benchmarks for comparison of performance of any product. For
example, content uniformity specifi cation of 90 – 110% assay limit of a 20 - mg (average
weight) tablet of a potent drug signifi es the challenge in maintaining the uniformity
of such a low - dose drug during the blending operation. Any relaxation in specifi cation
of this potent drug should be justifi ed with extensive documentation to assure
the performance. However, tightening of an acceptance criterion is considered as a
minor level change and to have minimal potential for an adverse effect on the
identity, quality, purity, or potency of a product.
1.3.2.3 Comparability Protocols
The FDA has introduced the concept of comparability protocols to expedite the
process of approval after submission of supporting document for a particular change
[5] . The protocol covers anticipated changes a product may experience during
it shelf life. Its recently published draft guidance “ Comparability Protocols —
Chemistry, Manufacturing, and Controls (CMC) Information ” describes the general
principles and procedures to prepare comparability protocols. The FDA suggests a
less stringent reporting category for any future change, where appropriate. Additionally,
if a detailed comparability protocol is provided, the FDA is less likely to
request additional supporting documents while comparing pre - and postapproval
change, and this could also help in implementing a particular CMC change, thereby
moving the product in the distribution line sooner. According to the FDA:
A comparability protocol is a well - defi ned, detailed, written plan for assessing the effect
of specifi c CMC changes in the identity, strength, quality, purity, and potency of a specifi
c drug product as these factors relate to the safety and effectiveness of the product.
A comparability protocol describes the changes that are covered under the protocol
and specifi es the tests and studies that will be performed, including the analytical procedures
that will be used, and acceptance criteria that will be achieved to demonstrate
that specifi ed CMC changes do not adversely affect the product. The submission of a
comparability protocol is optional.
A comparability protocol may be submitted with a new drug application (NDA),
abbreviated new drug application (ANDA), or supplements to these applications.
SCIENTIFIC AND REGULATORY RATIONALE FOR SUPAC 71
Comparability protocols can have single or multiple changes provided that each
change is discrete and specifi cation of the acceptance criteria for a change is well
defi ned.
1.3.2.4 In Vitro – In Vivo Requirements
Stability of a drug product, in vitro dissolution, and in vivo bioequivalence are prerequisites
for performance of a drug product and play a key role in establishing the
quality of a drug product after a postapproval change has been implemented. Any
type of major change, for example, in the manufacturing process from dry granulation
to wet granulation could affect the bioavailability and stability of a drug
product. Careful selection of the dissolution condition can obviate the need for a
costly bioequivalence study. Guidelines by the FDA [4] and ANVISA [6] take into
consideration the solubility and permeability of a drug substance for selection of
dissolution criteria for a particular drug product (immediate release or modifi ed
release) whereas guidelines by the EU and TGA recommend submitting comparison
records between a particular number of manufacturing batches pre - and
postapproval.
While categorizing a change or variation for its effect, suffi cient consideration
should be given to those parameters of a drug product which could affect its bioavailability.
Critical parameters like the particle size of active ingredient or excipients,
solid - state characteristics, and surface wettability may change during the
process variation and could adversely affect product performance resulting in an
altered dissolution profi le. The effect would be more pronounced in drug products
containing poorly soluble potent drugs and could have a deleterious effect on bioavailability.
The FDA guideline considers recommendation of the Biopharmaceutic
Classifi cation System (BCS) regarding solubility and permeability characteristics to
see whether any in vivo bioequivalence study is needed along with an in vitro dissolution
study. In the same pattern, ANVISA places drugs in three categories for
solid TM dosage form: case A, active substances with high permeability and high
solubility; case B, active substances with low permeability and high solubility; and
case C, active substances with high permeability and low solubility. As per the
guideline, for alteration of registration due to excipient change, for level 2 alterations
(that could cause signifi cant impact on quality and performance) the following
requirements should be met:
Case A “ The required documentation must include the undertaking of the
technical report and assessment of the results of the dissolution test, carried
out as described in the Brazilian Pharmacopoeia and, in its absence, other
codes authorized by the legislation in force. There must be dissolution of at
least 85% of the active substance in up to 15 minutes, using 900 ml of HCl
0.1 M . In case this criterion is not complied with, the tests described for Cases
B or C must be carried out. ”
Case B “ The required documentation must include the undertaking of the
technical report and assessment of the results of the dissolution profi le employing
Pharmacopeial conditions and removing samples from the medium at
appropriate time points until the plateau is reached. The dissolution profi le
obtained must be similar to the profi le of the unaltered formulation. ”
72 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
Case C “ The required documentation must include the undertaking of the
technical report and assessment of the results of the dissolution profi le in fi ve
different conditions: distilled water, HCl 0.1 M and phosphate buffer pH 4.5,
6.5 and 7.5 for the proposed formulation and the previous formulation, without
change. Samples of the dissolution medium must be removed at appropriate
time points until 90% of the active substance is dissolved or the plateau is
reached. A tensoactive may be used only when appropriately justifi ed. The
profi le obtained must be similar to the profi le of the unaltered formulation. ”
In addition, for level 2 change, no additional bioequivalence study is required if
the proposed alteration matches with the situation for cases A, B, and C. However,
if there is any deviation, then documentation containing the results and assessment
of a new bioequivalence and/or bioavailability study [if proper in vitro/in vivo correlation
( ivivc ) has not been established] should be submitted as per the conditions
mentioned in the level 3 alteration.
1.3.3 REGULATORY AGENCIES AND GUIDELINES
1.3.3.1 FDA SUPAC Regulations
The Food and Drug Administration Modernization Act (FDAMA) of 1997 (the
Modernization Act) was passed on November 21. With FDAMA in effect, another
Section 506A was added to the federal Food, Drug, and Cosmetic Act (the act) and
Section 314.70 (21 CFR 314.70) and the section included recommendations for
reporting categories (in terms of defi ned words) for any type of manufacturing
changes to an approved application (NDA or ANDA). In accordance with the act,
the FDA issued “ Guidance for Industry: Changes to an Approved NDA or ANDA ”
(fi nalized in 2004). This guidance is a current standard for pharmaceutical manufacturers
for making and reporting manufacturing changes to an approved application
and for distributing a drug product made with such changes.
“ SUPAC - IR: Immediate - Release Solid Oral Dosage Forms: Scale - Up and Post -
Approval Changes: Chemistry, Manufacturing and Controls, In vitro Dissolution
Testing, and In vivo Bioequivalence Documentation ” (issued 1995) was the fi rst
attempt to provide the pharmaceutical industry with a clear - cut guideline covering
the requirements for notifi cation and submission of documentation to regulatory
authorities pertaining to postapproval changes. This guideline was an outcome of
(a) a workshop on the scale - up of IR products conducted by the American Association
of Pharmaceutical Scientists with the U.S. Pharmacopoeia (USP) Convention
and the FDA; (b) research conducted by the University of Maryland at Baltimore
on the CMC of IR products; (c) drug categorization research on the permeability
of drug substances at the University of Michigan and the University of Uppsala;
and (d) SUPAC task force set up by the Centre for Drug Evaluation and Research
(CDER) CMC coordination committee. Following the issuance, it became a benchmark
for the industry. Two more guidances have been published on the same format
(level of changes as defi ned by SUPAC IR) as of SUPAC for MR drug products
(issued in 1997) [7] and nonsterile semisolid drug products (issued in 1997) [8] .
“ Guideline for Changes to Approved NDA or ANDA ” supersedes any previous
guidelines which have information on reporting categories that is inconsistent with
this guideline.
Guideline to Industry: Changes to Approved NDA or ANDA “ Guideline for
Industry: Changes to Approved NDA or ANDA ” provided reporting categories for
various postapproval changes and relaxed certain requirements that were considered
to have minimal or no impact on the drug product [1] . Moreover, it lessened
the burden on regulatory authorities and companies as well. Four reporting categories
provided in this guideline are as follows:
1. Prior Approval Supplement For a major change (substantial potential to
have effect on quality and performance), a supplement has to be submitted to
the FDA for approval before a product made with the change is distributed.
There is also a provision for “ Prior Approval Supplement: Expedite Review
Requested ” for public health reasons and if the delay in approval may cause
any substantial concerns for the applicant.
2. Supplement: Changes Being Effected (CBE) in 30 Days For a moderate
change (moderate potential to have effect on quality and performance), a
supplement has to be submitted to the FDA for approval 30 days before a
product made with the change is distributed.
3. Supplement: Changes Being Effected in 0 Days For some changes a supplement
has to be submitted to the FDA and simultaneously the product made
with the change can be distributed.
4. Annual Report For a minor change (minimal potential to have effect on
quality and performance), all information has to be submitted to the FDA in
the next annual review and the product made with the change can be
distributed.
All three types of changes under this guidance have been categorized as
follows:
(a) Changes in Components and Composition Any qualitative or quantitative
changes in the components and composition of a drug product is considered
as major changes. The current Guideline to Industry: Changes to Approved
NDA or ANDA does not mention these in detail because of the complexity
involved in the recommendations and therefore the SUPAC guideline has to
be followed for any such type of changes and regarding documentation
requirements for regulatory submission.
(b) Changes in Manufacturing Sites A change in a manufacturing site (for
manufacturing, packaging, labeling of drug products, testing components,
drug product containers, closures, packaging materials), either owned or
contract site, of drug products from the one that is approved requires prior
approval from the CDER. A prior approval supplement has to be submitted
for a change to a site that does not have a satisfactory CGMP inspection for
the type of operation to be performed. Further, changes in sites related to
operations like labeling, secondary packaging, and testing are considered to
have effect independent of drug product dosage form and therefore the
REGULATORY AGENCIES AND GUIDELINES 73
74 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
reporting categories for any of type of manufacturing site changes will be the
same. However, changes in sites related to operations like manufacturing and
primary packaging are considered to have effect that is dependent on dosage
form and hence reporting categories may be different.
(c) Changes in Manufacturing Process Changes in the manufacturing process
can have substantial effect on the identity, strength, quality, purity, or potency
of a drug product and there may be a change in the effi cacy of the drug
product regardless of the testing of drug product for conformance for the
approved specifi cation.
(d) Changes in Specifi cations Specifi cation, acceptance criteria, and regulatory
analytical procedure are a part of every dossier submitted to regulatory
agencies. Specifi cations are the standards, acceptance criteria are the limits
for specifi cations, and the regulatory analytical procedure is used for testing
a specifi cation ’ s acceptance criteria for the test substance that is approved
by the regulatory authority. Alternative analytical procedure may be
included in the application simultaneously with the main analytical
procedure.
(e) Changes in the Container Closure System Effects related to changes in the
container closure system are largely dependent on route on administration,
the operation in which the container closure system is involved, and contact
with the drug product. In some cases, there may be an effect in spite of the
conformance of drug product with the approved specifi cation.
(f) Changes in Labeling Changes in the package insert and package container
label are included in the labeling changes and applicant must immediately
revise all promotional labeling and drug advertisement in accordance with
the change in the approved labeling.
(g) Miscellaneous Changes Apart from categories mentioned above, changes
like stability protocol, expiration period, and addition of stability protocol
or comparability protocol have been included in the miscellaneous
category.
(h) Multiple Related Changes One change may lead to advertent or inadvertent
incorporation of another change, for example, a change in the manufacturing
site may lead to a change in the manufacturing equipment and manufacturing
process or changes in packaging material may cause changes in stability
protocol. For such combination changes, the CDER recommends submitting
documents in accordance with the most stringent reporting category for the
individual change.
Scale - up and Postapproval Changes: Immediate - Release and Modifi ed - Release
Dosage Forms SUPAC guidelines categorized postapproval changes in terms of
“ levels ” [4] . Three levels were defi ned depending upon the intensity of the adverse
effect on the formulation. Level 1 signifi es that the resulting effect on the quality
would be minimal and less extensive documentation should be presented to the
FDA in an annual review. Changes in accordance with level 2 could have signifi cant
effect on the quality and performance of the dosage form. Level 3 changes are most
likely to affect the quality and performance of the dosage form and hence extensive
documentation justifying those changes should be submitted to the FDA prior to
distribution of the products made with these changes. Apart from describing these
levels, recommendations were also made on the extent of CMC documentation, in
vitro dissolution, and in vivo bioequivalence tests that need to be submitted. Each
section in the guideline [(a) components and compositions, (b) site change, (c) scale -
up/scale - down; and (d) manufacturing equipment and process] was categorized in
terms of these three levels. Further, SUPAC IR also takes into consideration the
therapeutic range, solubility, and permeability of the drug for defi ning any particular
change. As per the guideline, three cases have been defi ned for the dissolution
testing (as mentioned in Table 1 ). Moreover, changes in excipient limits for a narrow
therapeutic range drug beyond that mentioned in level 1 have been recommended
as level 3 changes and extensive documentation is required for justifi cation. In the
SUPAC guideline for MR dosage forms, changes have been described at the same
three levels [7] .
However, dissolution conditions have been distinguished quite reasonably
between extended - and delayed - release dosage form (Table 2 ). For reporting any
level 3 change, three - month accelerated stability data of three batches (signifi cant
body of information not available) or three - month accelerated stability data for one
batch (signifi cant body of information available) have to be submitted in a supplement
along with long - term stability data for one batch in an annual review. A signifi
cant body of information has been defi ned in the guideline as availability of
suffi cient stability information of the product (stability data of fi ve commercial
batches). To provide a comparative outline, the guidelines for MR and IR dosage
forms are described in Tables 3 – 8 :
TABLE 1 Different Cases and Respective Dissolution Conditions for Immediate - Release
Solid Dosage Form
Case A a Case B b Case B c
Dissolution of 85% in
15 min in 900 mL of 0.1 N
HCl. If a drug product fails
to meet this criterion, the
applicant should perform
the tests described for case
B or C.
Multipoint dissolution
profi le should be
performed in the
application/compendial
medium at 15, 30, 45, 60,
and 120 min or until an
asymptote is reached.
Multipoint dissolution profi les
should be performed in water,
0.1 N HCl, and USP buffer
media at pH 4.5, 6.5, and 7.5
(fi ve separate profi les) for the
proposed and currently accepted
formulations. Adequate sampling
should be performed at 15, 30,
45, 60, and 120 min until either
90% of drug from the drug
product is dissolved or an
asymptote is reached. A
surfactant may be used, but only
with appropriate justifi cation.
a High - permeability, high - solubility drugs.
b Low - permeability, high - solubility drugs.
c High - permeability, low - solubility drugs.
REGULATORY AGENCIES AND GUIDELINES 75
76 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
TABLE 2 Dissolution Conditions for Modifi ed - Release Dosage Form
Extended Release Delayed Release
In addition to application/compendial
release requirements, multipoint
dissolution profi les should be
obtained in three other media, for
example, in water, 0.1 N HCl, and
USP buffer media at pH 4.5 and 6.8
for the changed drug product and the
biobatch or marketed batch
(unchanged drug product). Adequate
sampling should be performed, for
example, at 1, 2, and 4 h and every 2
hours thereafter until either 80% of
the drug from the drug product is
released or an asymptote is reached.
A surfactant may be used with
appropriate justifi cation.
In addition to application/compendial release
requirements, dissolution tests should be
performed in 0.1 N HCl for 2 h (acid stage)
followed by testing in USP buffer media, in
the range of pH 4.5 – 7.5 (buffer stage) under
standard (application/compendial) test
conditions and two additional agitation speeds
using the application/compendial test apparatus
(three additional test conditions). Multipoint
dissolution profi les should be obtained during
the buffer stage of testing. Adequate sampling
should be performed, for example, at 15, 30, 45,
60, and 120 min (following the time from which
the dosage form is placed in the buffer) until
either 80% of the drug from the drug product is
released or an asymptote is reached. The above
dissolution testing should be performed using
the changed drug product and the biobatch or
marketed batch (unchanged drug product).
(a) Changes in Components and Compositions (Table 3 ) The guideline for
changes to approved NDA or ANDA does not defi ne these changes in detail, and
thus the SUPAC guideline has to be followed for reference and reporting. Changes
in excipient levels are submitted as a prior approval supplement (with accelerated
stability data) whereas any changes in the levels of colors or fl avors are submitted
in an annual review (long - term stability data). In MR dosage forms these changes
have been logically categorized as (a) changes in excipient levels not affecting the
release profi le and (b) changes in excipient levels affecting the release profi le. In
level 2 changes for IR product and MR dosage forms for a non - narrow therapeutic
drugs, three - month accelerated stability data of one batch (in MR dosage form for
narrow therapeutic drugs three - month accelerated stability data of three batches)
in a supplement and long - term stability data of one batch in an annual review should
be submitted. Additionally, for delayed - release MR dosage forms of a narrow therapeutic
range drug, the multipoint dissolution profi le in the buffer stage of testing
should be generated for changed and commercial product using the medium that is
approved or in pharmacopeia. For extended - release MR dosage forms of a narrow
therapeutic range drug, the multipoint dissolution profi le should be generated for
changed and commercial product using the medium that is approved or in
pharmacopeia.
(b) Changes in Manufacturing Site (Table 5 ) A change in the manufacturing
or packaging site (or a contract manufacturing location) that has been approved by
the FDA in the original application has to be evaluated for its effect on the product
quality and performance. These changes have been described in detail in current
guideline changes to approved NDA or ANDA.
TABLE 3 Changes in Nonrelease Controlling Components and Composition
Level Classifi cation
Therapeutic
Range/Type of
Drug Test Documnetation
Filing
Documentation
I
Complete or partial
deletion
of color/fl avor
Change in inks,
imprints
SUPAC - IR level 1
excipient ranges
No other changes
All drugs Stability
Application/compendial requirements
No biostudy
Annual report
II
Change in technical
grade and/or
specifi cations
Higher than
SUPAC - IR level
1 but less than
level 2 excipient
ranges
No other changes
All drugs for
MR
Depending upon
therapeutic
range
solubility and
permeability
(as per BCS)
for IR
MR (ER):
Notifi cation and updated
batch record
Stability
Application/
compendial
requirements plus
multipoint dissolution
profi les in three other
media (e.g., water,
0.1
N HCl, and USP
buffer media at pH 4.5
and 6.8) until . 80% of
drug released or an
asymptote is reached
Apply some statistical
test (f2 test) for
comparing dissolution
profi les
No biostudy
MR (DR):
Notifi cation and updated
batch record
Stability
Application/compendial
requirements
plus multipoint
dissolution profi les in
additional buffer stage
testing (e.g., USP
buffer media at pH
4.5 – 7.5) under
standard and increased
agitation conditions
until . 80% of drug
released or an
asymptote is reached
Apply some statistical
test (f2 test) for
comparing dissolution
profi les
No biostudy
IR:
Notifi cation and
updated batch
record
Stability
Dissolution
requirements:
case A, case B, or
case C
Apply some
statistical test (f2
test) for comparing
dissolution profi les
No biostudy
Prior approval
supplement
Change in technical
grade and/or
specifi cations
Higher than
SUPAC - IR
level 1
No other changes
77
Level Classifi cation
Therapeutic
Range/Type of
Drug Test Documnetation
Filing
Documentation
III
Higher than
SUPAC - IR level
2 excipient ranges
for MR and IR,
change in
excipient range
for low solubility
and low
permeability
drugs beyond
level 1
All drugs for
MR and all
drugs failing
dissolution
criteria for
level 2 for IR
Updated batch record
Application/compendial (profi le) requirements and as
mentioned for level II
Stability
Biostudy or ivivc
Updated batch
record
Dissolution profi le as
for level II
Stability
Biostudy or ivivc
Prior approval
supplement
Note :
MR,
Modifi ed - release dosage form; ER, extended - release dosage form; DR, delayed - release dosage form; IR, immediate - release dosage form.
TABLE 3 Continued
78
TABLE 4 Changes in Release Controlling Components and Composition
Level Classifi
cation Therapeutic Range Test Documentation
Filing Documentation
I
. 5% w/w change based
on total release
controlling excipient
(e.g., controlled -
release polymer,
plasticizer) content
No other changes
All drugs Stability
Application/compendial requirements
No biostudy
Annual report
II
Change in technical
grade and/or
specifi cations
. 10% w/w change based
on total release
controlling excipient
(e.g., controlled -
release polymer,
plasticizer) content
No other changes
Nonnarrow
MR (ER):
MR (DR):
Prior approval
supplement Notifi cation and updated
batch record
Stability
Application/compendial
requirements plus
multipoint dissolution
profi les in three other
media (e.g., water, 0.1 N
HCl, and USP buffer media
at pH 4.5 and 6.8) until
. 80% of drug released or
an asymptote is reached
Apply some statistical test (f2
test) for comparing
dissolution profi les
No biostudy
Notifi cation and updated
batch record
Stability
Application/compendial
requirements plus
multipoint dissolution
profi les in additional
buffer stage testing (e.g.,
USP buffer media at pH
4.5 – 7.5) under standard
and increased agitation
conditions until . 80% of
drug released or an
asymptote is reached
Apply some statistical test
(f2 test) for comparing
dissolution profi les
No biostudy
Narrow Updated batch record
Stability
Application/compendial (profi le) requirements
Biostudy or ivivc
Prior approval
supplement
III
> 10% w/w change
based on total release
controlling excipient
(e.g., controlled -
release polymer,
plasticizer) content
All drugs Updated batch record and stability
Application/compendial (profi le) requirements
Biostudy or ivivc
Prior approval
supplement
79
TABLE 5 Site Changes
Level Classifi cation
Therapeutic
Range Test Documentation Filing Documentation
I Single facility
Common
Personnel
No other changes
All drugs
Application/compendial requirements
No biostudy
Annual report
II
Same contiguous
campus
Common personnel
No other changes
All drugs MR (ER):
MR (DR):
IR:
Changes being effected
supplement
(accelerated stability
data for MR and no
stability data for IR)
Annual report (long -
term stability data for
MR and IR)
Identifi cation and
description of site
change and updated
batch record
Notifi cation of site
change
Stability
Application/
compendial
requirements plus
multipoint
dissolution profi les
in three other media
(e.g., water, 0.1 N
HCl, and USP buffer
media at pH 4.5 and
6.8) until . 80% of
drug released or an
asymptote is reached
Apply some statistical
test (f2 test) for
comparing
dissolution profi les
No biostudy
Identifi cation and
description of site
change, and updated
batch record
Notifi cation of site
change
Stability
Application/
compendial
requirements plus
multipoint
dissolution profi les
in additional buffer
stage testing (e.g.,
USP buffer media at
pH 4.5 – 7.5) under
standard and
increased agitation
conditions until
. 80% of drug
released or an
asymptote is reached
Apply some statistical
test (f2 test) for
comparing
dissolution profi les
No biostudy
Identifi cation and
description of site
change and
updated batch
record
Notifi cation of site
change
Stability
Application/
compendial
requirements
No biostudy
III
Different campus
Different personnel
All drugs Notifi cation of site change
Updated batch record
Application/compendial (profi le) requirements
(as for level II)
Stability
Biostudy or ivivc
Notifi cation of site
change
Updated batch
record
Case B dissolution as
for excipient
change (level II)
Stability
No biostudy
Prior approval
supplement
(accelerated stability
data) for MR and
changes being
effected supplement
for IR
Annual report
80
TABLE 6 Changes in Batch Size:
Scale - Up/Scale - Down
Level Classifi
cation Change Test Documentation
Filing
Documentation
I
Scale - up of biobatch(s)
or pivotal clinical
batch(s)
No other changes
. 10 . (all drugs) Updated batch record
Stability
Application/compendial requirements
No biostudy
Annual report
II
Scale - up of biobatch(s)
or pivotal clinical
batch(s)
No other changes
> 10 . (all drugs)
MR (ER):
MR (DR):
IR:
Changes being
effected
supplement
(accelerated
stability data)
Annual report
(long - term
stability data)
Updated batch record
Stability
Application/
compendial
requirements plus
multipoint
dissolution profi les
in three other
media (e.g., water,
0.1
N
HCl,
and
USP buffer media
at pH 4.5 and 6.8)
until . 80% of drug
released or an
asymptote is
reached
Apply some statistical
test (f2 test) for
comparing
dissolution profi les
No biostudy
Updated batch record
Stability
Application/compendial
multipoint dissolution
profi les in additional
buffer stage testing
(e.g., USP buffer
media at p H
4.5 – 7.5)
under standard and
increased agitation
conditions until . 80%
of drug released or an
asymptote is reached
Apply some statistical
test (f2 test) for
comparing dissolution
profi les
No biostudy
Updated batch
record
Stability
Case B dissolution
as for excipient
change (level II)
No biostudy
81
TABLE 7 Changes in Manufacturing: equipment
Level Classifi cation Change Test Documentation Filing Documentation
I Equipment
changes
No other changes
(all drugs)
Alternate
equipment of
same design
and principle
Automated
equipment
Updated batch record
Stability
Application/compendial requirements
No biostudy
Annual report
II
Equipment
changes
No other changes
(all drugs)
Change to
equipment of
a different
design and
operating
principle
MR (ER):
MR (DR):
IR:
Prior approval
supplement
(accelerated
stability data)
Annual report (long -
term stability data)
Updated batch record
Stability
Application/
compendial
requirements plus
multipoint
dissolution profi les
in three other media
(e.g., water, 0.1 N
HCl, and USP
buffer media at pH
4.5 and 6.8) until
. 80% of drug
released or an
asymptote is
reached
Apply some statistical
test (f2 test) for
comparing
dissolution profi les
No biostudy
Updated batch
record
Stability
Application/
compendial
requirements plus
multipoint
dissolution
profi les in
additional buffer
stage testing (e.g.,
USP buffer media
at ph 4.5 – 7.5)
under standard
and increased
agitation
conditions until
. 80% of drug
released or an
asymptote is
reached
Apply some
statistical test
(f2 test) for
comparing
dissolution
profi les
No biostudy
Updated batch
record
Stability
Case C dissolution
as for excipient
change (level II)
No biostudy
82
TABLE 8 Changes in Manufacturing: Processes
Level Classifi cation Change Test Documentation
Filing
Documentation
I
Processing changes
affecting the
nonrelease/release
controlling excipients
for MR
Changes within
validation ranges (IR)
No other changes
Adjustment of
equipment
operating
conditions (mixing
times, operating
speeds) — within
approved
application ranges
Updated batch record
Application/compendial requirements
No biostudy
Annual report
II
Processing changes
affecting the
nonrelease controlling
excipients and/or the
release controlling
excipients
Processing changes
outside validation
ranges for IR
No other changes
Adjustment of
equipment
operating
conditions (e.g.
mixing times,
operating speeds,
etc.)
Beyond approved
application ranges
MR (ER):
MR (DR):
IR:
Changes being
effected
supplement
(accelerated
stability data
for MR)
Annual report
(long - term
stability data
for MR and
IR)
Updated batch record
Stability
Application/compendial
requirements plus
multipoint dissolution
profi les in three other
media (e.g. water,
0.1
N HCl, and USP
buffer media at pH
4.5 and 6.8) until
. 80% of drug
released or an
asymptote is reached
Apply some statistical
test (f2 test) for
comparing dissolution
profi les
No biostudy
Updated batch record
Stability
Application/compendial
requirements plus
multipoint dissolution
profi les in additional
buffer stage testing (e.
g., USP buffer media at
pH 4.5 – 7.5) under
standard and increased
agitation conditions
until . 80% of drug
released or an
asymptote is reached
Apply some statistical test
(f2 test) for comparing
dissolution profi les
No biostudy
Notifi cation of change
Updated batch record
Stability
Case B dissolution as
for excipient
change (level II)
No biostudy
III
Processing changes
affecting the
nonrelease controlling
excipients and/or the
release controlling
excipients
Change in the type
of process used
(e.g. from wet
granulation to
dry)
Updated batch record
Stability
Application/compendial (profi le) requirements
Biostudy or ivivc
Updated batch
record — stability
Case B dissolution as
for excipient
change (level II)
No biostudy
Prior approval
supplement
(accelerated
stability data
for MR)
Annual report
(long - term
stability data
for MR and
IR)
83
84 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
The FDA should be notifi ed of the new location. For any type of moderate
changes, accelerated stability data of one batch should be submitted with the CBE
supplement for MR dosage forms and long - term stability data of one batch should
be submitted in the annual review for IR as well as MR dosage forms. Additionally,
the CBE should be submitted to the FDA in case of any moderate change. Stability
requirements for any major change are the same as those mentioned in the previous
section, that is, three - month accelerated stability data of three batches (signifi cant
body of information not available) or three - month accelerated stability data for one
batch (signifi cant body of information available) have to be submitted in a prior
approval supplement along with long - term stability data for one batch in the annual
review.
(c) Scale - Up/Scale-Down (Table 6 ) A change in the batch size of a drug product,
either scale - up or scale - down, is likely to induce some changes in the operation
parameters. This in turn can adversely affect product quality.
(d) Changes in Manufacturing Equipment (Table 7 ) and Process (Table 8 ) Any
manufacturing changes in equipment and process are included in this section. For
example, a change in the blending equipment from octagonal blender to double
cone blender or a change in the granulation process from wet to dry granulation
calls for submission of proper validation documentation for FDA approval. All
these changes along with reporting categories have been described in current guidelines
for changes to approved NDA or ANDA.
Biowaivers In vitro and in vivo approaches are commonly used for establishment
of bioavailability and bioequivalence. Dissolution studies are used as in vitro
approaches and also serve as quality control tools for pharmaceuticals. Under
certain circumstances, in vitro dissolution may also act as a surrogate marker
for in vivo biostudy and enable the establishment of in vitro and in vivo
bioequivalence.
“ CDER Guidance for Industry: Waiver of In Vivo Bioavailability and Bioequivalence
Studies for IR Solid Dosage Form Based on Biopharmaceutic Classifi cation
System (BCS) ” recommends waiving an in vivo biostudy under specifi c circumstances.
For example, a waiver of the in vivo biostudy of one or more lower strengths
is acceptable based on the correlation data and in vivo bioequivalence of the higher
strength, provided all strengths are proportionally equivalent in terms of active and
inactive ingredients. A biostudy on a lower strength may also be requested based
on safety reasons (as for mitrazapine tablets) and a biowaiver for highest strength
is acceptable provided elimination kinetics is linear over a dose range, strengths are
proportional, and comparative dissolution data of all strengths are acceptable.
The BCS classifi es drugs in four classes:
Class I: high solubility, high permeability
Class II: low solubility, high permeability
Class III: high solubility, low permeability
Class IV: low solubility, low permeability
Dissolution, solubility, and permeability are the three fractors that control the
bioavailability of a drug for an IR drug product. Provided the inactive excipient
does not control or modify the release and absorption of the active ingredient, the
biostudy may be waived. According to the guideline, the solubility class is determined
for the highest dose strength of a drug product for which a biowaiver has
been requested. When the highest dose strength of a solid dosage form is soluble in
250 mL of water or less across a pH range of 1 – 7.5, it is considered as highly soluble.
For determination of permeability class various in vivo methods like mass balance,
absolute bioavailability, and intestinal perfusion approaches and in vitro methods
like permeation studies using excised tissue or monolayer of cultured epithelial cells
are used. When extent of absorption is greater than 90% of the administered dose
in humans, it is considered as highly permeable. For a dissolution study, drug release
should be evaluated in three media that are 0.1 N HCl or USP - simulated gastric
fl uid without enzymes, pH 4.5 buffer, and pH 6.8 buffer or USP - simulated intestinal
fl uid without enzymes. Rapidly dissolving drug products are those that dissolve
more than 80% in 900 mL of the above - mentioned media in less than 30 min using
USP apparatus at 100 rpm (or USP II apparatus of 50 rpm).
A biowaiver can be requested for the postchange products if it falls under class
I of the BCS and displays a rapidly dissolving profi le and there is a similarity (as
determined by f2 test) between the pre - and postchanged drug product in all three
media. For BCS class II drugs, a meaningful correlation (level A, B, or C correlation)
between in vitro drug release and in vivo absorption also may be used for requesting
the biowaiver. Deconvolution techniques are used for prediction of in vivo dissolution
and absorption.
1.3.3.2 Regulations Guidance on SUPAC by Pharmaceutical Unit of EU
The pharmaceutical market in European countries is one of the largest in the world.
To ensure that the EU promotes pharmaceutical trade and ensures safety, effi cacy,
and quality of medicinal products within the European member states, the pharmaceutical
unit of the EU runs a series of information and communication projects,
collectively called EUDRA projects. Out of these projects, the EUDRALEX pharmaceutical
unit is responsible for making community pharmaceutical legislation,
guidelines, and notices for applicants [9] . Under Volume 2, Section C, of Regulatory
Guidelines (Pharmaceutical Legislation: Notice to Applicants) of Eudralex, “ Guideline
on Dossier Requirements for Type IA and Type IB Notifi cations ” has been
provided [2] .
Regulations were introduced to lessen the administrative load on the authority
and to simplify the procedure for granting a postapproval variation without negotiating
any quality attribute of drug product [10] . Under these regulations, type IA
and type IB were defi ned; also, clearcut terms were introduced for extension application,
parallel/consequential notifi cation/variation, and urgent safety restriction.
For streamline operation of these regulations, four documents have been
prepared:
(a) A procedural guidance for the member states (reference or concerned) and
the applicant for notifi cations/variations in the mutual recognition procedure
(b) A procedural guidance for the applicant for notifi cations/variations in the
centralized procedure
REGULATORY AGENCIES AND GUIDELINES 85
86 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
(c) A common application form which may be used for type IA and type IB
notifi cations or type II variations in both the centralized and mutual recognition
procedures
(d) A guideline on the documentation to be submitted for type IA and type IB
notifi cations
All member states of the EU follow the same regulations for a change or “ variation
” in an already approved medical product. As per the guidance, three types of
variations or changes have been identifi ed — type I variation, which is further classi-
fi ed into types IA and IB and type II variations. The guidelines classify some specifi c
changes in type IA or IB. It also provides specifi c data analysis required for variation
and the types of document that need to be submitted to the regulatory authority.
Any change that is not listed in this section is classifi ed as type II variation.
According to Commission Regulation (EC, No. 1084/2003), type I variation has
been defi ned as “ A ‘ minor variation ’ of type IA or type IB means a variation listed
in Annex I, which fulfi ls the conditions, set out therein. ” Annex I of the regulation
provides a list of changes and conditions (to be satisfi ed) to be classifi ed as type IA
or type IB variation and Annex II provides changes falling under the extension
application category. Type II variations in proposed documentation are not type I
or extension application. There is also a provision for “ urgent safety restrictions. ”
These are any temporary or provisional changes in the product summary characteristics,
such as indications, posology, contraindications, warnings, target species, and
withdrawal periods, as result of a new information that may cause signifi cant safety
concerns about the medicinal product [11] .
Any change arising from the primary change has to be notifi ed separately. Consequential
changes form part of the same notifi cation whereas parallel changes do
not. A consequential change to type IA can only be another type IA whereas a
consequential change to type IB can be type IA or type IB. All other variations
should be submitted as Type II variations. “ Guideline on Dossier Requirements for
Type IA and Type IB Notifi cations ” provides a complete list of all changes, conditions
required to be met for the particular change, and documentation required by
the regulatory authority [10] .
1.3.3.3 Regulatory Guidance on SUPAC by Agencia Nacional de
Vigilancia Sanitaria
Agencia Nacional de Vigilancia Sanitara (ANVISA) issues Brazil ’ s generic drug
policy. Under legislation for industry, Resolution RE N ° 893, of May 29, 2003, is
described in “ Guide for Making Post - Registration Alterations, Inclusions and Noti-
fi cations of Drug Products ” [6] . This guideline describes postregistration changes as
“ alterations ” and “ inclusions ” and also tells about the documentation and assays
that need to be submitted in support of any type of change. As per the guideline,
each type of alteration or inclusion has to be submitted separately and approved
by the ANVISA before it can be implemented. Table 9 presents some examples for
each category.
Under each category, certain requirements have to be met before its implementation.
For example, for inclusion in the batch size, the company should notify, in
alteration, if the included batch size is more than 10 times. The documentation that
needs to be submitted includes the original proof of payment of fee or of exemption;
a copy of the certifi cate of good manufacturing and control practices (CBPFC)
issued by ANVISA; technical justifi cation; production and quality control records
of one batch of each strength of the product; a technical report; and a technical
report and assessment of the dissolution profi le.
1.3.4 HARMONIZATION
It is essential to evaluate the safety and quality of new or changed medical products
before they reach the market. However, the need to set specifi c guidelines has been
recognized at different times in different countries. For example, in the United
States a tragic incident with a junior paracetamol formulation was the alarm to initiate
guidelines for authorization of medical products. European countries followed
this trend in the 1960s after the thalidomide incident. Since then there have been a
large number of guidelines that have been put into place to evaluate medical products
in terms of their quality, safety, and effi cacy [12] . However, with the pharmaceutical
industries becoming international and aiming for a worldwide market, there
is a move toward internationally accepted guidelines and approval systems. In order
for medical products to be marketed internationally, companies have found it necessary
to duplicate many tests and studies that are time consuming and broad.
TABLE 9 Examples of Different Categories
Postregistration
Alterations
Postregistration
Inclusions
Postregistration
Notifi cations
Postregistration
Cancellation
Labeling alteration Inclusion of new
commercial
presentation
Temporary
suspension of
manufacture
Cancellation upon
request of
registration of
drug
presentation
Alteration of corporate
name
Inclusion of new
packing
Resumption
of drug
manufacture
Cancellation of
drug registration
Alteration of date of
expiry
Inclusion of new
concentration
already approved in
country
Alteration of
preservation
conditions
Inclusion of new
dosage form already
approved in country
Alteration of synthesis
path of drug
Inclusion of new
therapeutic
indication in country
Alteration of
manufacturer of
drug
Inclusion of
manufacture site
Alteration of
manufacturing site
Inclusion of
manufacturer of
drug
Alteration of excipient Inclusion in the batch
size
HARMONIZATION 87
88 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
Table 10 shows examples of documentation required by different countries when
a postapproval change is made during the manufacturing process of medical products.
When there are changes in the specifi cation of an excipient, the documents
required by the TGA and European Agency for Evaluation of Medicinal Products
(EMEA) are variable. Furthermore this would indicate that the benefi t of a patent/
medical product might not reach globally. There are also many chances of making
an error. For example, the 2003 recall of clotrihexal 100 - mg vaginal tablets in New
Zealand pharmacies was due to the fact that clotrihexal was packed according to
TGA guidelines and thus its sale was prohibited in New Zealand. This resulted in
much confusion and problems among patients and medical professionals. Harmonization
is the process by which the pharmaceutical industries worldwide adopt
the same laws and regulations. Harmonization is intended to assure the safety,
quality, and effi cacy of a medical product globally. The main goal of harmonization
is to recognize and minimize the differences in the scientifi c requirements for
medical product development within different regulatory agencies in different
countries.
Harmonization activities focus on reducing and simplifying the types of studies
that the pharmaceutical industries need to carry out in order to register a medical
product in another country, protocols to be followed when performing these studies,
techniques used to validate supporting data, and techniques used to perform risk
assessment.
Harmonization reduces replications and unnecessary production and registration
of new and changed products. The concept of harmonization was explored by
European countries in the 1980s. The success of harmonization in these countries
has demonstrated that it is practical and possible. Following harmonization in
Europe the International Conference on Harmonization (ICH) was set up in 1990
[13] . Table 11 shows some of the harmonized rules that have been successfully
developed by ICH.
TABLE 10 Changes in Specifi cation of Excipients (Addition of New Test Limit):
Comparison between Guidelines
Guidelines Documentation Type of Change
TGA Details of the test method must be provided.
Appropriate validation data have been generated for
the test method.
The limits proposed are based on batch analytical data
and are in compliance with offi cial standard and/or
relevant accepted guidelines if applicable.
Self - assessable
changes
EMEA Comparative table of current and proposed
specifi cations.
Batch analysis data on two production batches for all
tests in the new specifi cation.
Where appropriate, comparative dissolution profi le
data for the fi nished product on at least one pilot
batch containing the excipient complying with the
current and proposed specifi cation. For herbal
medicinal products, comparative disintegration data
may be acceptable.
Minor change
type IB
requires
approval
1.3.5 GMP ISSUES: CHANGE CONTROL AND PROCESS VALIDATION
Changes are unavoidable in a manufacturing setup. Manufacturers make changes
at some stage of manufacturing during and after approval of a product. However,
consistent quality of a drug product can only be assured through well - defi ned validation
procedures. When a change is made in the manufacturing process of a drug
product, sponsors are responsible for evaluating the effect of any change on the
safety, effi cacy, quality, stability, and potency of a drug product and ensuring that
these properties are not infl uenced by the change. In a manufacturing setup,
various disciplines like sales, marketing, medical, regulatory affairs, manufacturing,
electrical, and technical services work together. Hence, any kind of change in one
discipline will have direct consequences on other disciplines. Each company should
have a procedure with regard to handling a change. Quality control and quality
assurance departments usually keep track of various changes occurring in a
GMP environment. Therefore, it is required that personnel performing the job are
trained enough to assess the effect of any kind of change or variation and take
appropriate action for its evaluation or control. Supporting data should be generated
and once evaluated can confi rm whether further clinical or nonclinical studies
are required.
1.3.5.1 Change Control
When a change is made in a manufacturing setup, it is important to assess its impact.
As a change can have impact on regulatory fi ling, manufacturing parameters, speci-
fi cations, and technical services, it is important to consider the concerns and objections
of various disciplines involved and only through well - defi ned standard
operating procedures should it be properly validated, evaluated, and fi nally implemented.
A properly defi ned order of evaluation of a change with strategic input of
trained personnel is key to delivering a consistent quality product (Figure 1 ).
When a change is the processed, the manufacturer should have protocols in place
with regard to assessing the change. Therefore, “ control of change ” is important.
Control can be implemented effectively only through well - defi ned standard operating
procedures. The main purpose of “ change control ” exercise is to have a
TABLE 11 Example of Quality Guidelines Harmonized by ICH
Quality Topic Example of Guideline
Q1: Stability Q1B: Photostability testing
Q2: Validation of analytical procedure Q2A: Methodology
Q3: Impurity testing Q3A: Impurities in new drug
substances
Q4: Pharmacopoeias Q4: Pharmacopoeial harmonization
Q5: Quality of biotechnological products Q5A: Viral safety evaluation of
biotechnological products
Q6: Specifi cations for new drug substance and
products
Q6A: Acceptance criteria for new drug
substances
Q7: GMP for pharmaceutical ingredients Q7A: GMP for active pharmaceutical
ingredients
GMP ISSUES: CHANGE CONTROL AND PROCESS VALIDATION 89
90 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
systematic process in place to accurately evaluate a change using specifi c tests.
Moreover, it aims to measure the effects on quality safety and effi cacy before a
change is implanted. Change control and its evaluation through proper documentation
should include [14] :
(a) Description and purpose of change
(b) Inputs from research and development (R & D) department
(c) Evaluation steps for impact assessment, such as evaluation of stability, validation
requirements, and in vivo bioequivalence requirement
(d) Need and extent of regulatory documentation and approval
(e) Implementation schedule
(f) Clear defi nition of personnel authorized for change approval
(g) Monitoring protocol for change implementation and periodic review of
impact
Following the informal proposal of a change, it should be reviewed by the responsible
initiator, who will then generate a formal proposal [15] . The proposal should
describe accurately what the change is concerned with, how to validate the change,
and the time frame within which the change should be implemented. The fi nal proposal
should be reviewed and assessed by all functional groups involved. Once the
change is approved, it can be implemented and the change cycle is completed. Figure
2 describes responsibilities of different departments of a pharmaceutical company,
FIGURE 1 Change control cycle for change in manufacturing process.
Comments
And Signature
Comments
And Signature
Comments
And Signature
Comments
And Signature
Change
Implement
Initiator
Approved by
Quality
Assurance
Approved by
Production
Department
Approved by
Regulatory
Department
Approved by
Head of
Department
Change Control Cycle
in the change control procedure. Standard operating procedures (SOPs) for change
control are an important part of any GMP audit. Hence it is important that it is
implemented by trained and qualifi ed personnel from appropriate disciplines.
After a change has been approved by all functional groups within the manufacturing
setup and if it has no regulatory concerns, it can be implemented immediately.
However, if the impact comes under any regulatory domain, the company may have
to wait for regulatory approval.
1.3.5.2 Process Validation
Process validation is an important part in the implementation of a postapproval
change. It establishes the documented evidence of conformance of a pharmaceutical
operation in accordance with specifi cations. FDA “ Guideline on General Principles
of Process Validation ” describes in detail the principles and practices of process
validation and documentation required by the regulatory authority [13] . In general
terms, process validation may be defi ned as the procedure which generates suffi cient
assurance and documented evidence that a particular operation is operating
and producing drug products in accordance with the specifi cations and process
controls.
FIGURE 2 Responsibilities of different disciplines of a pharmaceutical company in a
change control procedure ( modifi ed from ref. 15 ).
INITIATION OF
CHANGE CONTROL
PROOFREADING
CONFORMANCE TO CGMP AND
APPLICABILITY TO OTHER SYSTEMS
REVIEW AND APPROVAL
REGULATORY IMPACT AND
WORLDWIDE FILING STRATEGY
VALIDATION
Quality assurance, Quality control,
Manufacturing, Process Engineering,
Technical services, Regulatory affairs,
Owner of system or procedure being
changes
Quality assurance
Quality assurance
Regulatory affairs
Quality assurance, Quality control,
Manufacturing, Process Engineering,
Technical services, Regulatory affairs,
Owner of system or procedure being
changes
Quality assurance
GMP ISSUES: CHANGE CONTROL AND PROCESS VALIDATION 91
92 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
Prospective validation, retrospective validation, concurrent validation, and revalidation
are the four validation components. Prospective validation is performed
before the distribution of drug products in the market or after the manufacturing
of a drug product using revised changes that can affect product quality and characteristics.
Retrospective validation is conducted for an established drug product
whose manufacturing process is stable to ensure that the current pharmaceutical
operation is performing as per the protocols and specifi cation and yielding satisfactory
product. Concurrent validation is conducted by monitoring in - process critical
manufacturing parameters and end - product testing to ensure that the current manufacturing
process is per the in - process control specifi cations. Revalidation is performed
after changes to an approved drug product are implemented to ascertain
that there is no adverse effect on the quality and performance of a drug product
[16] .
During a validation process, the products and processes are subjected to testing
at extreme conditions of in - process limits and their performance is evaluated against
the acceptance criteria. The parameters of different pharmaceutical operations are
varied and product properties are recorded and evaluated (Figure 3 ). When it is
found that adjustment is required, necessary actions are taken in consultation with
R & D personnel. Generally, validation data of three production scale batches are
compared to generate a high level of quality assurance.
Systematic documentation of the effect on the product attributes by varying
various process parameters is very important in the validation process. The product
development team, engineering and technical services, and production and regulatory
departments are also consulted while making any process change or before
fi nalizing any validation protocol or report. Depending on the “ level ” of change or
degree of effect to be produced, the extent of the validation is determined.
Based on the validation requirements, samples are collected at different stages
and submitted for analysis per the validation protocol. The data are fi nally compiled
in the form of a validation report. A systematic validation protocol and validation
report are the backbone of the validation process. Table 12 gives key components
of any validation activity [16] . These protocols and reports should be verifi ed and
approved by the relevant functions.
Some changes are often made in the manufacturing process without prior notifi -
cation, and hence it is advisable to consider revalidation at predetermined frequencies
(or whenever an unusual behavior is noted).
When new equipment is purchased or there is a change in the manufacturing site,
qualifi cation exercises are performed as part of the validation process. Qualifi cation
(installation qualifi cation, operation qualifi cation, and performance qualifi cation)
for any equipment or facility is an extreme process which involves testing, verifi cation,
and documentation to assure that the particular equipment or facility is per
the specifi cation and meets the appropriate standards as defi ned by vendor and
required by manufacturing and engineering personnel [14] .
1.3.6 CONCLUSION
The global pharmaceutical industry is continuously growing in a rapidly changing
and dynamic environment of the health care sector. New drugs and delivery systems
FIGURE 3 Various process parameters and product characteristics associated with validation
activity of typical coated tablet.
Wet Granulation
Drying
Milling
Blending
Tabletting
Coating
Process Parameters Product Characteristics
Premix time, Binder addition
time, Impeller/Chopper Speed,
Inlet air temperature, Bed Temperature,
Airflow rate, Raking frequency
Screen size, Hammer/knives
direction and speed
Compression speed, force (hardness
and thickness), tablet weight
Pan capacity, inlet/exhaust
temperatures, Pan Speed, spray
rate, Air flow rate, Bed temperature
Capacity of Blender, Mixing
time, mixing speed
Granule hardness and size
distribution
Moisture content of Granules,
Amount of residual solvent
Size distribution, Bulk/ Tapped
density of granules
Content uniformity in blender
and drum, Bulk/ Tapped density
of final blend
Tablet weight, hardness, thickness,
content uniformity, friability,
dissolution, disintegration, Assay/
potency
Dissolution, disintegration, coating
weight gain, mottling, assay/ potency
CONCLUSION 93
TABLE 12 Key Components of Validation Activity
Validation Protocol Validation Report
Purpose of study Aim of study
Personnel responsibility List of raw material used in study
Critical process steps List of manufacturing equipment
Critical process parameters Critical steps studied
Critical product parameters Collected data and its analysis
Sampling plan Acceptance criteria evaluation
Testing plan Statistical analysis
Acceptance criteria Recommendations by validation department
94 SCALE-UP AND POSTAPPROVAL CHANGES (SUPAC) REGULATIONS
surface each year in the market. To maintain the quality of new and existing drugs
and delivery technologies, pharmaceutical operations are controlled by regulatory
guidelines. The purpose of developing guidelines is to keep the health and safety of
a person on the highest priority by delivering quality pharmaceuticals. Implementation
of these guidelines and systematic follow - up of the effect of postapproval
changes in the form of documentation are essential to safeguard against any possible
failure of the whole system. Change control and validation ensure that there is no
deleterious impact on the drug product characteristics. Anticipated changes incorporated
in comparability protocols reduce signifi cant risk of experiencing unpredictable
adverse effects and help to introduce the product in less time. When an impact
is anticipated, it should be properly discussed with R & D, process development,
and other concerned departments for appropriate regulatory fi ling by following
regulatory guidelines. Provided that these guidelines are followed properly, quality
and performance of a drug product can be ensured.
REFERENCES
1. U.S. Department of Health and Human Services, Food and Drug Administration, Centre
for Drug Evaluation and Research (CDER) , Guidance for industry: Changes to an
approved NDA or ANDA, available,
accessed Apr. 15, 2006 .
2. European Commission , Guideline on dossier requirements for type IA and type IB
notifi cations: Pharmaceuticals: Regulatory framework and market authorizations, available:
http://ec.europa.eu/enterprise/pharmaceuticals/eudralex/vol - 2/c/gdvartypiab_rev0_
200307.pdf , accessed Apr. 20, 2006 .
3. Department of Health and Ageing , Therapeutic Goods Administration, Australian regulatory
guidelines for prescription medicines. Appendix 12: Changes to the quality information
of registered medicines: Notifi cation. Self - assessment and prior approval, available:
http://www.tga.gov.au/pmeds/argpmap12.pdf , accessed Apr. 12, 2006 .
4. Food and Drug Administration, Centre for Drug Evaluation and Research (CDER) ,
Guidance for industry: SUPAC - IR: Immediate - release solid oral dosage forms: Scale - up
and post - approval changes: Chemistry, manufacturing and controls, in vitro dissolution
testing, and in vivo bioequivalence documentation, available: http://www.fda.gov/cder/
guidance/cmc5.pdf , accessed May 11, 2006 .
5. Food and Drug Administration, Centre for Drug Evaluation and Research (CDER) ,
Guidance for industry: Comparability protocols — Chemistry, manufacturing, and controls
information (draft), available, accessed Apr. 15, 2006 .
6. Brazilian Sanitary Surveillance Agency (ANVISA) , Resolution: Guide for making post -
registration alterations, inclusions and notifi cations of drug products, Brazil ’ s generic drug
policy, industry legislation, available: http://www.anvisa.gov.br/hotsite/genericos/legis/
resolucoes/893_03re_e.htm , accessed Apr. 11, 2006 .
7. U.S. Department of Health and Human Services, Food and Drug Administration, Centre
for Drug Evaluation and Research (CDER) , Guidance for industry: SUPAC - MR: Modi-
fi ed release solid oral dosage forms scale - up and postapproval changes: Chemistry,
manufacturing, and controls; in vitro dissolution testing and in vivo bioequivalence documentation,
, accessed May 5,
2006 .
8. U.S. Department of Health and Human Services, Food and Drug Administration, Centre
for Drug Evaluation and Research (CDER) , Guidance for industry: SUPAC - SS: Nonsterile
semisolid dosage forms; scale - up and post - approval changes: Chemistry, manufacturing
and controls; in vitro release testing and in vivo bioequivalence documentation.
9. Pharmaceutical Unit of European Commission at EUROPA, available: http://ec.europa.
eu/enterprise/pharmaceuticals/pharmacos/docs/brochure/pharmaeu.pdf , accessed May
21, 2006 .
10. Variations. Pharmaceuticals: Regulatory framework and market authorizations, Chapter
5, in Procedures for Marketing Authorisation , Vol. 2A, European Commission, available:
http://ec.europa.eu/enterprise/pharmaceuticals/eudralex/vol - 2/a/v2a_chap5_r1_2004 - 02.
pdf , accessed Apr. 16, 2006 .
11. Commission regulation (EC) No. 1084/2003 of June 3, 2003, concerning the examination
of variations to the terms of a marketing authorisation for medicinal products for human
use and veterinary medicinal products granted by a competent authority of a member
state (Offi cial Journal L 159, 27/6/2003, pp. 1 – 23), available: http://ec.europa.eu/
enterprise/pharmaceuticals/eudralex/homev1.htm , accessed Apr. 10, 2006 .
12. The ICH process for harmonisation of guidelines, available: http://www.ich.org/cache/
compo/276 - 254 - 1.html , accessed May 15, 2006 .
13. Food and Drug Administration, Centre for Drug Evaluation and Research (CDER) ,
Guideline on general principles of process validation, available: http://www.fda.gov/cder/
guidance/pv.htm , accessed Apr. 26, 2006 .
14. Willig , S. H. ( 2001 ), Production and process controls , in Swarbrick , J. , Ed., Good Manufacturing
Practices for Pharmaceuticals: A Plan for Total Quality Control from Manufacturer
to Consumer , Marcel Dekker , New York , pp. 99 – 138 .
15. Waterland , N. H. , and Kowtna , C. C. ( 2003 ), Change control and SUPAC , in Nash , R. A. ,
and Wachter , A. H. , Eds., Pharmaceutical Process Validation , Marcel Dekker , New York ,
pp. 699 – 748 .
16. Ahmed , S. U. , Naini , V. , and Wadgaonkar , D. ( 2005 ), Scale - up, process validation and
technology transfer , in Shargel , L. , and Kanfer , I. , Eds., Generic Drug Product Development:
Solid Oral Dosage Form , Marcel Dekker , New York , pp. 95 – 136 .
REFERENCES 95
97
1.4
GMP - COMPLIANT PROPAGATION
OF HUMAN MULTIPOTENT
MESENCHYMAL STROMAL CELLS
Eva Rohde , Katharina Schallmoser , Christina Bartmann ,
Andreas Reinisch , and Dirk Strunk
Medical University of Graz, Graz, Austria
Contents
1.4.1 Introduction
1.4.2 Acronyms and Defi nitions
1.4.2.1 Mesenchymal Stromal Cells
1.4.2.2 Somatic Stem Cell Therapy
1.4.2.3 Good Manufacturing Practice
1.4.2.4 Cell - Based Medicinal Products
1.4.2.5 Human Platelet Lysate
1.4.3 Approaches
1.4.3.1 Adherence to Principles of GMP in a Preclinical Developmental Process
1.4.3.2 Effi cient Standardized MSC Propagation Using Low Cell Seeding Density
1.4.3.3 Superior MSC Proliferation Resulting from HPL - Driven as Compared to
FBS - Driven Cultures
1.4.3.4 Contamination Risks Can Be Minimized in Rational MSC Propagation
Procedures
1.4.4 Testing Methods
1.4.4.1 Safety and Effi cacy of CBMP in Preclinical Stage
1.4.4.2 Quality Controls During Cell Culture (In - Process Controls) and Final Product
Release Criteria
1.4.4.3 MSC Functionality and Potency Assays
1.4.5 Conclusion
References
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
98 GMP-COMPLIANT PROPAGATION
1.4.1 INTRODUCTION
Somatic stem cell therapy (SCT) is a rapidly growing fi eld that opens a broad spectrum
of therapeutic options. The concept of regenerative SCT is based on the
assumption that transplantation of adult human stem cells may support organ regeneration,
modulate immunity, and regulate hematopoiesis. Transplantation of bone
marrow (BM) – derived hematopoietic stem cells (SCs) for blood and immune system
regeneration has been a clinical reality for almost 40 years. The existence of detectable
numbers of mesenchymal and endothelial progenitors within blood and BM
has promoted the readily harvestable hematopoietic tissue as a source of SCs for
nonhematopoietic regenerative SCT (Figure 1 ).
Multipotent mesenchymal stromal cells (MSCs) are currently undergoing evaluation
in a number of clinical trials ( www.clinicaltrials.gov ). These nonhematopoietic
cells have been fi rst described by Friedenstein et al. in a fi broblast colony - forming
unit assay (CFU - F) based on low - density culture of adherent BM - derived cells
[1 – 3] . Alternative sources for MSCs have been identifi ed in a number of studies
showing the successful isolation of fi broblast precursors from umbilical cord blood,
placenta, umbilical cord, amniotic fl uid, and adipose tissue [4 – 13] . To date, most
experimental and clinical experience has been accumulated with BM - MSC [14 – 21] .
Ex vivo expansion of these rare BM constituents (representing less than 1% of
aspirated BM nucleated cells) is a prerequisite to achieve a reasonable MSC application
dose of at least 2 . 10 6 MSCs/kg of the recipients ’ body weight. The majority
of expansion procedures are currently based on the use of fetal bovine serum (FBS),
which carries the risk of xenoimmunization and transmission of known (e.g., prions
transmitting bovine spongiforme encephalopathia, BSE) and unknown pathogens.
These risks could be avoided by developing MSC expansion protocols that use
human alternatives which replace FBS.
The preclinical development of medicinal products in general bears high complexity
due to the lack of fi xed routines. Long term manipulations of cell - based
medicinal products (CBMPs) may enhance the risks for undesirable effects in the
course of ex vivo cell expansions. Safety concerns regarding the clinical application
of ex vivo generated MSCs require a logistic environment providing an established
good manufacturing practice (GMP) background embedded in a highly effective
quality system. Demonstration of manufacturing and product consistency is achieved
by applying rational in - process controls. Release criteria should ideally emerge from
successful product development and optionally include the sterility, safety, purity,
identity, and potency. 1 They must to be rapid, sensitive, and reliable and should
retain some fl exibility in type and timing of testing. The complexity and function of
different CBMPs require an array of analytical procedures to adequately characterize
the particular product (potency assays). Personalized (patient - specifi c) CBMPs
differ from large drug batches in the pharmaceutical industry in terms of practicability
in fi nal product release in that they may require process - oriented rather than
single - product potency testing. U.S. regulations demand that “ tests for potency
shall consist of either in vitro or in vivo tests, or both, which have been specifi cally
1 U.S. legislation: 21 CFR 610, General biological products standards, CFR 610.10 Potency, CFR
600.3(s).
FIGURE 1 Hematopoietic tissue - derived SC and progenitors. Hematopoietic tissue contains
( a ) mesenchymal and ( b ) endothelial in addition to ( c ) hematopoietic progenitor cells.
( a ) Adult human BM - derived MSCs were stained to visualize the actin cytoskeleton, mitochondria,
and nuclei. ( b ) The periphery of an umbilical cord blood – derived endothelial progenitor
cell (EPC) colony is depicted demonstrating typical cobble stone – like morphology.
The entire colony was derived from a single UCB - EPC indicating impressive proliferation
potential (more than 70,000 cells were obtained by harvesting single EPC - derived colonies
indicating the completion of at least 16 population doublings). Less than 10 mL of adult BM
[( a ) and ( c )] but at least 40 mL of UCB ( b ) were suffi cient to generate appropriate numbers
of cells for therapeutic purposes.
(c)
(b)
(a)
INTRODUCTION 99
100 GMP-COMPLIANT PROPAGATION
designed for each product so as to indicate its potency in a manner adequate to
satisfy the interpretation of potency given by the defi nition in 21 CFR 600.3(s). ”
Functional analyses accompanying the expansion process development leading to
a full product characterization and optimization of manufacturing steps are prerequisites
that allow for the creation of a safe and effective CBMP.
This chapter demonstrates that rapid and standardized expansion of human
MSCs to achieve a reasonable cell dose (i.e., . 2 . 10 6 /kg body weight of a 75 - kg
person corresponding to . 1.5 . 10 8 MSCs) is feasible within less than four weeks.
Replacing FBS with human platelet lysate (HPL) provides one strategy toward a
safer CBMP (Figure 2 ). Appropriate preclinical development adhering to GMP
principles will enhance safety in the course of a consecutive clinical evaluation of
MSCs as a therapeutic agent.
1.4.2 ACRONYMS AND DEFINITIONS
1.4.2.1 Mesenchymal Stromal Cells
Adhesion of mononuclear cells from human bone marrow aspirates (BM - MNC) to
tissue culture plastic and removal of nonadherent cells during the fi rst days of
culture selects for a population of proliferating spindle - shaped fi broblast - like non-
FIGURE 2 GMP - compliant propagation of human MSCs. The summary of a
two - step MSC production procedure shows seeding and harvest numbers of BM - MNC and
resulting numbers of MSC HPL as compared to MSC FBS . ( Reproduced with permission from
ref. 23 .)
4 x 2.5mL heparinized BM Aspiration diluted
immediately (without density gradient) in
.-MEM / 10% FBS .-MEM / 10% HPL
1 x 107 MNC / 60mL / 225cm 2 in
10 - 20 x T225 or 1 -2 CF-4 ( < 105 BM-MNC/cm 2)
8 x 106 MSC / 225cm 2 1 x 106 MSC / 225cm 2
STORE: n x 3x10 5
MSCHPL aliquots
STORE: n x 1x10 6
MSCFBS aliquots
.-MEM / 10% HPL
3 x 105 MSC / 1m 2
.-MEM / 10% FBS
3 x 105 MSC / 1m 2
STEP II
STEP I
1° SEEDING
day 0
1° HARVEST
. 10 - 16 days
BM aspiration
day 0
2° HARVEST
. 11 - 15 days
2° SEEDING
day 0
3.0 - 5.4 x 10 8 MSCHPL 0.5 – 1.1 x 10 8 MSCFBS
hematopoietic multipotent MSCs. Mesenchymal stromal cells can also be obtained
from umbilical cord blood, umbilical cord, placenta, adipose tissue, and several fetal
tissues. The minimum criteria for MSCs are defi ned in an ISCT (International
Society for Cellular Therapy) position paper published in 2006 [22] . The MSCs have
a high self - renewal potential and the capacity to be differentiated in vitro into
progeny displaying an osteo - , chondro - , or adipogenic phenotype.
1.4.2.2 Somatic Stem Cell Therapy
The concept of regenerative SCT is based on experimental and early clinical observations
indicating that the application of adult stem cells can improve organ regeneration
after ischemic, toxic, or metabolic injury. Bone marrow harbors hematopoietic
and mesenchymal stem cells and endothelial progenitor cells and is an easily accessible
but not the sole, source of candidate cells to promote organ repair after systemic
or local application. Regulation of hematopoiesis and immune modulation
are the two established applications in the broad fi eld of SCT with autologous and
allogeneic stem and progenitor cells.
1.4.2.3 Good Manufacturing Practice
Good manufacturing practice is that part of the quality management system (QMS)
that is concerned with the production and quality control of medicinal products
(drugs) for human and veterinary use. It includes documentation, personnel training,
facility, equipment, and process controls for the manufacture of pharmaceuticals.
1.4.2.4 Cell - Based Medicinal Products
Medicinal products containing viable cells are summarized under the umbrella term
cell - based medicinal products . The term CBMP does not cover products containing
nonviable cells or cellular fragments. The CBMPs may have much potential in the
treatment of various diseases that to date have no cure. They are heterogeneous in
terms of origin and type of cells and with regard to the complexity of the product.
Cells may be self - renewing stem cells, more committed progenitors, or terminally
differentiated cells exerting a specifi c regenerative function. Cells may be of autologous
or allogeneic origin. Cells may be used alone or in combination with biomolecules,
chemical substances, or structural materials that possibly potentiate their
desired effects.
1.4.2.5 Human Platelet Lysate
Human platelet lysate can be obtained from buffy coat – derived platelet rich plasma.
The platelet fraction is separated from the plasma and the white and red blood cell
fraction by centrifugation steps and concentrated to a density of at least 1 . 10 9 platelets/
mL. Platelets can either be activated with thrombin or lysed by repeated freeze –
thaw cycles. Both mechanisms result in the release of growth factors and mitogens
that are stored in intact platelets. Mediators released from platelets include, among
others, epidermal growth factor (EGF), basic fi broblast growth factor (bFGF), platelet
- derived growth factors (PDGFs), transforming growth factor (TGF - . 1), and
insulinlike growth factor (IGF) [23, 24] . Perhaps HPL may replace FBS in many
ACRONYMS AND DEFINITIONS 101
102 GMP-COMPLIANT PROPAGATION
cell culture systems that have previously been thought to strictly depend on the
presence of FBS.
1.4.3 APPROACHES
1.4.3.1 Adherence to Principles of GMP in a Preclinical Developmental Process
The standardized MSC propagation should be conducted as a well - planned, consistently
documented, and optimized procedure that also minimizes risks of microbiological,
particulate and pyrogen contamination by reducing manipulation steps and
manipulation time. According to current European legislation, 2 the principles of
GMP should be applied to CBMP when they are manufactured for use in human
subjects in phase 1 studies. These requirements do not apply to cellular or tissue -
based medicinal products used in phase 1 studies according to U.S. legislation 3 or
to products in the preclinical developmental phase. If it is expected that preclinical
fi ndings are to be translated into clinical use rather rapidly, it may be recommended
to establish GMP - compliant technology during the preclinical developmental
phase of any cell product. As a result, this ensures that products are consistently
produced and controlled to meet the quality standards appropriate for their intended
use or product specifi cation. The GMP requirements are well described in “ PIC/S
Guide to Good Manufacturing Practice for Medicinal Products ” and include
the implementation of an effi ciently running quality management system, dedi -
cated areas for manufacture of sterile medicinal products complying with GMP,
appropriately qualifi ed and trained personnel, suitable equipment, correct materials,
containers and labels, approved procedures and instructions, suitable storage and
transport facilities, and a record - keeping system that allows the complete history of
a medicinal product to be traced (See http://www.picscheme.org ). It is a challenge
to conduct preclinical research and development complying to GMP as procedures
routinely turn out to be much more time and cost intensive than common laboratory
- scale research. These circumstances can advance either the developmental
progress at the expense of quality standards or vice versa. It should therefore be
decided on a case - by - case basis how closely to adhere to GMP standards depending
on the more or less stringent time schedule for the considered clinical use of a
CBMP.
1.4.3.2 Effi cient Standardized MSC Propagation Using Low Cell
Seeding Density
The future use of MSCs in clinical studies may require very high absolute MSC
numbers to gain appropriate cell doses ( > 5 . 10 6 /kg body weight) per patient compared
to in vivo experimental models with small animals [20] . It is consequently
advantageous to develop large - scale MSC expansion protocols that allow for the
2 European legislation: Directive 65/65/EEC, Directive 75/318/EEC, Directive 75/318/EEC, Commission
Communication on the Community marketing authorisation procedures for medicinal products
(98/C229/03); Directive 2001/20/EC, EMEA/CHMP/410869/2006.
3 U.S. legislation: 21 CFR 210; 21 CFR 211; 21 CFR 312.21; 21 CFR 312.22(a) and 21 CFR
312.23(a)(7)(i).
APPROACHES 103
generation of up to 5 . 10 8 – 10 . 10 8 MSCs from the limited starting volume of
primary material.
The cell seeding density is of critical importance for the expansion rate of MSCs
and must be defi ned for the primary seeding and the following passaging steps. Most
experimental and clinical expansions described to date were started with a
high seeding density of more than 1 . 10 5 BM - MNC/cm 2 [2, 14, 16] . For further passages
pioneering studies showed that a very low seeding density between 0.5 and
10 MSCs/cm 2 selects for the expansion of a rapidly proliferating subpopulation of
recycling SCs, termed RS cells [25 – 28] . This seeding density, referred to as “ clonal
density, ” would necessitate a theoretical growth area of from 2,000,000 to 100,000 cm 2
(from 200 to 10 m 2 ) to obtain a clinical quantity of > 1 . 10 8 MSCs from 1 . 10 6 starting
MSCs within one passage. Plating 30 – 100 MSCs/cm 2 therefore is a reasonable
compromise density requiring a more realistic growth area between 10,000 and
25,000 cm 2 (1 and 2.5 m 2 ). We have recently shown that the primary seeding of only
10 mL bone marrow aspirates on approximately 0.2 m 2 culture area for two weeks
(culture step 1; BM diluted immediately after aspiration in culture medium without
density gradient separation; removal of nonadherent cells at day 3) followed by an
expansion on 2.5 m 2 (step 2) is suffi cient to consistently generate at least 1.5 .
10 8 MSCs in FBS - supplemented medium within less than four weeks (Figure 2 ) [29] .
This study furthermore corroborated earlier data on the inverse correlation of the
seeding density to MSC proliferation (Figure 3 ) [25 – 28] .
1.4.3.3 Superior MSC Proliferation Resulting from HPL - Driven as Compared
to FBS - Driven Cultures
The most commonly used basic cell culture medium compositions for MSC propagation
are minimum essential medium alpha ( . - MEM) and low - glucose (1 g/L) Dulbecco
’ s modifi ed Eagle medium (DMEM - LG) supplemented with l - glutamin,
antibiotics, and 5 – 20% FBS [14, 16, 19, 24, 25, 30] . Our experience with MSC propagation
relates to the use of . - MEM supplemented with either FBS or HPL. In
contrast to HPL that has been recognized only recently as a potent culture medium
supplement [24] , FBS is a well - known key medium supplement for cell culture and
its role has been unchallenged for more than 50 years [31] . The common use of FBS
in MSC cultures as a source of growth factors and mitogens bears the risk of transmission
of known and unknown pathogens as well as xenoimmunization against
bovine pathogens and should therefore be avoided for clinical use [32, 33] .
In a recent study we analyzed the capacity of HPL to replace FBS in large - scale
(clinical) MSC expansions and were able to demonstrate a superior propagation of
MSC cultured with HPL (MSC HPL ) as compared to MSC derived from FBS - driven
cultures (MSC FBS ) [23] . Figure 4 illustrates superior MSC proliferation at low plating
density and higher population doublings (PDs) with HPL after a culture period of
less than 14 days.
1.4.3.4 Contamination Risks Can Be Minimized in
Rational MSC Propagation Procedures
Cell expansion is mainly performed according to labor - intensive time - consuming
protocols using open systems that increase the risks of microbiological or particulate
contamination and supplementation with potent antibiotics to control these prob
104 GMP-COMPLIANT PROPAGATION
lems. Avoiding the use of penicillin during clinical - scale cell propagation follows the
rationale to reduce the risk of sensitization as well as anaphylactic precipitation.
Thus, it may be worthwhile not using other antibiotics for the GMP - compliant MSC
propagation. One approach to minimize potential contamination risks is to rigorously
reduce handling in the course of MSC propagation to the absolute minimum
of necessary steps. In our experience, the commonly used density gradient centrifugation
step can be skipped prior to the primary cell seeding of the bone marrow
aspirate. Immediate dilution of limited volumes (e.g., 10 – 20 mL) of heparinized BM
aspirate into supplemented . - MEM medium for direct cell seeding does not result
in a loss of MSC recovery [23] . Furthermore, the aforementioned low cell seeding
density and the employment of an increased growth area in a simplifi ed procedure
together with the use of HPL in fact allow for an effi cient production of high MSC
numbers within one to two harvest - replating cycles. The relatively short ex vivo
expansion time of less than three to four weeks may be helpful in reducing the
cumulative risk of contamination.
1.4.4 TESTING METHODS
1.4.4.1 Safety and Effi cacy of CBMP in Preclinical Stage
The preclinical developmental period should be used for the extensive characterization
of the CBMP. Release criteria have to be defi ned and reasonable time frames
1000/cm2
day 1
day 3
day 5
day 10
100/cm2 10/cm2 1/cm2
P+1 P+1
FIGURE 3 Inverse correlation of seeding density to MSC proliferation. BM - derived MSCs
derived from passage 2 were seeded at log fold deescalated density of 1000, 100, 10, and 1 cm . 2 .
Photographs were taken after 1, 3, 5, and 10 days of culture in . - MEM/10% FBS (original
magnifi cation 40 . ). In the case of MSC seeded at 100 and 1000 cells/cm 2 confl uence necessitated
trypsinisation between days 5 and 10 followed by reseeding at 100 and 1000 cells/cm 2 ,
respectively, and is therefore indicated as P + 1.
must be set to allow for a high safety and quality standard of the fi nal cellular
product. On the other hand, the logistic background should allow for a rapid release
of the CBMP within a few hours due to the potential short shelf life of many cellular
products. Ranges of cell purity, sterility, and absence of pyrogens and endotoxins
are factors of utmost importance which must be determined. It is an inherent feature
of CBMPs that product specifi cations must be adapted to the individual application.
The challenge in the preclinical developmental phase is to fi nd satisfactory answers
to unresolved questions in terms of cell type, source, dose, and mode of application
according to the particular target disease. Thus, in - process controls and defi nitive
release criteria must be met by each CBMP. Since many CBMPs are personalized
medicine, potency assays must be performed for selected representative products
(e.g., before initiating a study and consecutively once per year).
1.4.4.2 Quality Controls During Cell Culture (In - Process Controls) and Final
Product Release Criteria
General Safety According the Food and Drug Administration (FDA), cellular
therapy products are exempt from general safety testing [21 CFR 610.11(g)(1)].
Cell Dose The preclinical stage can be used to determine the specifi cations for the
minimum effective and maximum tolerable number of viable and functional cells.
The optimum dose of cells to be administered still needs to be established [20] .
FIGURE 4 MSC proliferation capacity depends on seeding density in xenogeneic FBS and
HPL - supplemented cultures. The inverse correlation of MSC proliferation to their seeding
density resulted in the formation of a confl uent MSC layer in cultures starting with 1 MSC/cm 2
in . - MEM/10% FBS and cultures starting with 1 – 10 MSCs/cm 2 in . - MEM/10% HPL but not
when initiating cultures with the respective higher seeding densities within less than two
weeks. The calculated fold increase of the cell number and corresponding population doublings
from a representative experiment harvested at day 13 are shown.
0
2
4
6
8
10
1/cm2 10/cm2 100/cm2
FBS HPL
FBS HPL
1/cm2 1/cm2
10/cm2 10/cm2
100/cm2 100/cm2
0
200
400
600
800
1/cm2 10/cm2 100/cm2
FBS HPL
PD (d 13) Fold increase
MSC cultured for 13 days
TESTING METHODS 105
106 GMP-COMPLIANT PROPAGATION
Viability Viability of MSCs can easily be determined immediately after trypsinization
via trypan blue or 7 - amino - actinomycin D (7 - AAD) exclusion. According
to the specifi cations developed from our cell culture studies, viability should be
> 90%. In selected exceptional cases a lower limit of 70% viability of total harvested
cells may be acceptable.
Microbiological Testing Sterility testing that detects fungal, anaerobic, and aerobic
bacterial and mycoplasma contamination should be performed after each critical
manipulation step during MSC culture that is prone to microbiological contamination
[34] . The crucial bacterial sterility check at the end of last harvesting step cannot
be evaluated prior to in vivo application if MSCs need to be applied immediately
after propagation due to the duration of the cultures. Mycoplasma polymerase chain
reaction (PCR) results can be obtained at the day of harvest within less than 6 h.
MycoAlert ® results are available within less than 1 h at the day of harvest. Defi nitive
culture results to exclude mycoplasma contamination are available within two to
three weeks and therefore are not applicable for CBMPs with a short shelf life that
are planned to be administered immediately after production.
Endotoxin and Pyrogenicity Testing Endotoxin measurement using the Limulus
amebocyte lysate (LAL) assay is typically done as an alternative to pyrogenicity
testing for early phase trials. For any parenteral drugs, except those administered
intrathecally, the FDA recommends that the upper limit for endotoxin be 5 EU/kg
body weight/dose. The LAL assay method can be applied to the safety evaluation
of biological preparations according to existing regulations. 4 We use the LAL assay
to substitute for the lengthy delay in microbiological data availability to obtain
results prior to the clinical application of the fi nal product within less than 2 h after
harvest.
Phenotypic Identity of MSC In addition to morphological identifi cation by microscopy,
the immunophenotypic characterization of MSCs can be done using a broad
panel of fl uorescence - conjugated antibodies directed against surface molecules. To
date there is no specifi c marker uniquely defi ning MSCs. Therefore a profi le is used
to show the expression of certain markers and to exclude the contamination by cells
expressing other marker profi les. Flow cytometry is recommended by the ISCT to
reveal that MSCs stain positive for CD73, CD90, and CD105 and negative for HLA -
DR, CD14, CD31, CD34, and CD45 (Figure 5 ) [22, 23] . Much more extensive phenotypic
analyses have been performed without retrieving additional information
about MSC type or function [35] . Gene expression profi ling will hopefully result in
a better defi nition of human MSCs [35 – 42] .
1.4.4.3 MSC Functionality and Potency Assays
Clonogenicity The self - renewal capacity of cells and the proportion of proliferating
cells within a heterogeneous cell mixture can be evaluated using the CFU assay.
4 Endotoxin testing, LAL, according to Eur. Pharm. 2.6.14 and Guideline on Validation of the Limulus
Amebocyte Lysate Test as an End - Product Endotoxin Test for Human and Animal Parenteral Drug,
Biological Products and Medical Devices, 1987, Sections I – IV, http://www.fda.gov/cber/gdlns/lal.pdf .
A tissue culture method allowing for the clone counting of cells was fi rst describend
in 1956 [43] . The introduction of bone marrow CFU assays led to the discovery of
hematopoietic stem cells [44] . Fibroblast precursors existing within the hematopoietic
system also have been evaluated with another specifi c CFU assay method
introduced by Friedenstein in 1974 (CFU - F) [2] . We analyzed the clonal expansion
capacity of MSC with the CFU - F method. Figure 6 shows differences in CFU - F
appearance between MSC HPL and MSC FBS . In the case of primary BM appropriate
dilution is necessary to determine the CFU - F frequency (Figure 7 ). Once MSCs are
enriched, the appropriate MSC seeding density recommended for CFU - F enumeration
may range from 1 to 5 MSCs/cm 2 [2, 29] .
Osteo - , Chondro - , and Adipogenic Differentiation Isolated BM - derived MSCs
were shown to differentiate along multiple mesechymal lineages in 1999 [45] . Evidence
suggests MSCs can also express phenotypic characteristics of endothelial,
neural, smooth muscle, skeletal myoblast, and cardiac myocyte cells [46] . The prototype
pathways of MSC differentiation occur along osteogenic, chondrogenic, and
adipogenic lineages and have been extensively demonstrated in a large number of
publications [47] . This kind of potency assay may be performed regularly if bone or
connective tissue repair is intended, although time limits do not enable immediate
product release.
FIGURE 5 Immune phenotype of human MSCs. Flow cytometric analysis of at least 10,000
viable MSCs was used to determine antibody reactivity (gray - fi lled histograms) compared to
appropriately diluted isotype controls (black line). Phenotypic criteria require positivity
( . 90%) for CD73, CD90, and CD105 and negativity ( . 2%) for HLA - DR, CD14 (or CD11b),
CD19 (or CD79 . ), CD34, and CD45. Absence of CD3+ T cells may be desirable in the case
of GvHD treatment. Depending on the culture conditions, MSCs share reactivity with the
anti-disialoganglioside antibody GD2 with neuroblastoma cells, melanoma, and small - cell
lung cancer cells.
HLA-AB CD 13 CD 29 CD 73 CD 90 CD105 CD 146
CD 45 CD 3 HLA-DR CD 31 CD 14 CD 34
FBS
FBS
HPL
HPL
GD-2
CD 19 CD 133
TESTING METHODS 107
108 GMP-COMPLIANT PROPAGATION
Immune Modulatory Effects Mesenchymal stromal cells inhibit T - cell alloreactivity
in mixed lymphocyte cultures (MLCs) or lymphocyte proliferation induced by
mitogens, such as phytohemaglutinin (PHA) or concanavalin A [29, 48 – 51] . It is of
note that high concentrations of MSCs (representing 10 – 40 MSCs per 100 responder
lymphocytes) have an inhibitory effect while low MSC concentrations (0.1 – 1%)
may stimulate lymphocyte proliferation in mixed lymphocyte cultures [50] . If MSCs
are used for immunosuppressive therapies, these fi ndings may imply that high doses
of MSCs are needed to inhibit T - cell proliferation in patients with graft - versus - host
disease following allogeneic bone marrow transplantation. The application of low
MSC numbers could stimulate lymphocyte proliferation in vivo and hence result in
an undesirable boost to graft - versus - host disease as an adverse reaction of the MSC
therapy. It is not clear so far whether the precise number of T cells in a
given MSC transplant needs to be determined to exclude a potential boost to
alloreactivity. Immune modulation can be measured with carboxyfl uorescein diacetate
N - succinimidyl ester (CFSE) labeling of cells to quantify proliferation in
response to allogeneic or mitogenic stimuli [52] . We analyzed the loss of CFSE fl uorescent
intensity indicating cell proliferation by fl ow cytometry after culturing
CFSE - labeled MNCs in the absence or presence of different numbers of MSCs [53] .
The immune regulatory capacity of MSC HPL and MSC FBS was studied by measurement
of allogenic MNC proliferation after co - culturing pairs of MNC from three
different donors with two independent MSC HPL and two other MSC FBS (Figure 8 ).
Hematopoiesis Regulation Regulation of the behavior of early hematopoietic
progenitor cells (HPCs) can be analyzed by MSC - HPC cocultures in vitro [54] .
FIGURE 6 Morphological evaluation of MSCs. CFU - F of MSC HPL compared to MSC FBS
differ in size, morphology, and density (scale bar identifi es magnifi cation in the upper panel;
colony photographs taken on day 12, 40 . original magnifi cation).
HPL
500.m
FBS
500.m
Liquid cultures of purifi ed CD34 + (HPC) with a preestablished MSC feeder layer
result in the expansion of CD34 + /38 + HPC and CD34 + /38 . hematopoietic SCs and
support the growth of mature hematopoietic total nucleated cell (TNC) progeny
(Figure 9 ).
Genetic Stability and Potential Tumorigenicity Genetic analysis of human MSCs
is not well established. The signifi cance of standard metaphase chromosome G
banding is limited due to the low number of metaphases recovered during standard
analyses. Advances in multicolor fl uorescence in situ hybridization (FISH) and high -
resolution array - based techniques may also soon be translated into practicable
diagnostic tools in relation to CBMP safety in regenerative medicine [55] .
Genetic instability can occur as a rare event after extended culture of mouse and
human MSCs in FBS - supplemented medium [56, 57] . To test for potential in vivo
tumor formation, MSCs derived from short - term clinical - scale expansions in FBS -
or HPL - supplemented media were injected into immunocompromised athymic
nude mice subcutaneously. Putative tumor formation was evaluated by histological
analyses three months after injection of 2 . 10 6 and 2 . 10 4 MSCs and compared to
controls that were injected 48 h prior to euthanasia. A primary cell deposit was
visible immediately and 48 h after injection and MSCs could be recovered by conventional
microscopic evaluation. However, none of 12 animals tested developed a
macroscopic or microscopic detectable tumor over the 90 - day observation period
[23] . In this situation, genetic testing may be encouraged for prospective data acqui-
FIGURE 7 CFU - F Evaluation of MSCs depends on BM seeding density. An appropriate
dilution of the heparinised BM aspiration is needed for accurate enumeration of the primary
CFU - F frequency as indicated in this representative experiment where whole heparinized
BM was seeded corresponding to the respective measured BM - MNC number per square
centimeters of growth area, cultured for 11 days at 37 ° C/humidifi ed atmosphere/3% O 2 /5%
CO2 . Nonadherent cells were removed at day 3. CFU - F are visualized by Harris hematoxylin
staining.
2.23 x 104 4.46 x 103 0.89 x 103 0.18 x 103
Seeded MNC / cm2
Day 11
TESTING METHODS 109
110 GMP-COMPLIANT PROPAGATION
FIGURE 8 MSC-mediated Immune Modulation. Allogeneic MNC proliferation (mean
cell number ± SEM) was measured after co - culturing pairs of MNC from three different
donors with two independent MSC HPL and two other MSC FBS as shown in the bar chart. MSC
were added in a 1 : 10 (3 . 10 4 MSC to 3 . 10 5 MNC/well; MNC[+PHA]:MSC = 10 : 1) or 1 : 100
(MNC[+PHA]:MSC = 10 : 1) ratio to test their infl uence on PHA - driven proliferation of
MNC. MNC numbers were measured by fl ow cytometric MNC count using BD Truecount TM
tubes. As a control numbers of mitogen stimulated MNC without additional MSC
(MNC[+PHA] only) and background proliferation without PHA stimulation (MNC w/o
PHA) are shown. MSC did not induce MNC proliferation (MNC + MSC w/o PHA). Signifi -
cant differences are marked by asterisks ( *p < 0.05 and * * p < 0.01). (Figure reproduced with
permission from reference 53 )
MNC[+PHA]:MSC = 100:1
MNC[+PHA] only
MNC[+PHA]:MSC = 10:1
MNC + MSC w/o PHA
MNC w/o PHA
0
2x105
4x105
6x105
8x105
1x106
1.2x106
1.4x106
CELL NUMBER
HPL FBS
200
CD34+ No
0
4x105
8x105
1,2x106
1,6x106
2x106
50
100
150
250
0
(b)
(c)
CD34+ FOLD INCREASE
0
3x106
6x106
8x106
1,2x107
1,5x107
300
1,8x107
2,1x107
TNC N o
FOLD INCREASE
20
5
10
15
25
0
CD34 only CD34
+
MSCFBS
(a)
FOLD INCREASE
CELL NUMBER
CD34
+
MSCHPL
CD34 only CD34
+
MSCFBS
CD34
+
MSCHPL
CD34+/CD38- No
0
1x105
2x105
3x105
4x105
12
6
8
10
14
0
4
2
CD34 only CD34
+
MSCFBS
CD34
+
MSCHPL
CD34+/CD38- FI vs. CTRL
FIGURE 9 MSC - mediated hematopoiesis regulation. ( a ) Umbilical cord blood (UCB) -
derived sorted CD34 + cells were expanded in cytokine - supplemented medium [Roswell Park
Memorial Institue (RMPI) - 1640/10% Fetal Bovine Serum (FBS)/Granulocyte and Macrophage
Colony Stimulating Factor (GM - CSF)/Interleukin 3(IL - 3)/Stem Cell Factor (SCF)/
FMS-like tyrosin kinase 3 ligand (Flt - 3L)] in the absence or presence of clinical - scale
expanded MSCs. Gray bars show harvested total nucleated cell number (TNC N ° .) and black
bars show the fold increase (FI) of the TNC N ° . compared to the starting CD34 + cell number.
( b ) Harvested number of CD34 + cells (gray bars) and fold increase (black bars) of CD34 +
cells after liquid culture with or without MSC support. ( c ) Harvested number (gray bars) and
fold increase (black bars) of CD34 + /CD38 . hematopoietic stem cells after liquid culture of
CD34 + cells with MSC FBS or MSC HPL support compared to cytokine - supplemented liquid
cultures in the absence of MSCs. (Mean ± Standard Error of the Mean (SEM) of two independent
expansions.) ( Reproduced with permission from ref. 53 .)
TESTING METHODS 111
112 GMP-COMPLIANT PROPAGATION
sition but is not considered as mandatory for product release in current MSC clinical
trials.
1.4.5 CONCLUSION
There are considerable limitations of common pharmacological techniques used in
determining the safety and effi cacy of CBMPs at the preclinical stage. Conventional
methods used in the pharmaceutical industry to develop pharmacological profi les
and to determine the acute toxicity of drugs in animals as well as toxicity studies
may not directly be translated to ex vivo generated cellular products. Nevertheless,
it is inevitable that preclinical research and development of cellular products will
be conducted under the guidance of either individual or consensus specifi cations
and defi nitions that will continuously be improved. This approach will be helpful in
developing successful therapeutic cellular agents.
ACKNOWLEDGMENTS
This work was supported in part by The Adult Stem Cell Research Foundation
(TASC RF ; C.B. and A.R.) and a young investigator fellowship of the Austrian Federal
Ministry for Education, Science and Culture, bm:bwk (A.R.). The Austrian Nano -
Initiative co - fi nanced this work as part of the Nano - Health project (no. 0200), the
sub-project NANO - STEM being fi nanced by the Austrian Science Fund (FWF
Project no. N211 - NAN).
REFERENCES
1. Luria , E. A. , Panasyuk , A. F. , and Friendenstein , A. Y. ( 1971 ), Fibroblast colony formation
from monolayer cultures of blood cells , Transfusion , 11 ( 6 ), 345 – 349 .
2. Friedenstein , A. J. , Deriglasova , U. F. , Kulagina , N. N. , et al. ( 1974 ), Precursors for fi broblasts
in different populations of hematopoietic cells as detected by the in vitro colony
assay method , Exp. Hematol. , 2 ( 2 ), 83 – 92 .
3. Kuznetsov , S. A. , Friedenstein , A. J. , and Robey , P. G. ( 1997 ), Factors required for bone
marrow stromal fi broblast colony formation in vitro , Br. J. Haematol. , 97 ( 3 ), 561 – 570 .
4. Erices , A. , Conget , P. , and Minguell , J. J. ( 2000 ), Mesenchymal progenitor cells in human
umbilical cord blood , Br. J. Haematol. , 109 ( 1 ), 235 – 242 .
5. Kogler , G. , Sensken , S. , Airey , J. A. , et al. ( 2004 ), A new human somatic stem cell from
placental cord blood with intrinsic pluripotent differentiation potential , J. Exp. Med. ,
200 ( 2 ), 123 – 135 .
6. Bieback , K. , Kern , S. , Kluter , H. , and Eichler , H. ( 2004 ), Critical parameters for the isolation
of mesenchymal stem cells from umbilical cord blood , Stem Cells , 22 ( 4 ), 625 – 634 .
7. Romanov , Y. A. , Svintsitskaya , V. A. , and Smirnov , V. N. ( 2003 ), Searching for alternative
sources of postnatal human mesenchymal stem cells: Candidate MSC - like cells from
umbilical cord , Stem Cells , 21 ( 1 ), 105 – 110 .
8. Miao , Z. , Jin , J. , Chen , L. , et al. ( 2006 ), Isolation of mesenchymal stem cells from human
placenta: Comparison with human bone marrow mesenchymal stem cells , Cell. Biol. Int. ,
30 ( 9 ), 681 – 687 .
9. In ‘ t Anker , P. S. , Scherjon , S. A. , Kleijburg - van der Keur , C. , et al. ( 2004 ), Isolation of
mesenchymal stem cells of fetal or maternal origin from human placenta , Stem Cells ,
22 ( 7 ), 1338 – 1345 .
10. In ‘ t Anker , P. S. , Scherjon , S. A. , Kleijburg - van der Keur , C. , et al. ( 2003 ), Amniotic fl uid
as a novel source of mesenchymal stem cells for therapeutic transplantation , Blood ,
102 ( 4 ), 1548 – 1549 .
11. Gronthos , S. , Franklin , D. M. , Leddy , H. A. , Robey , P. G. , Storms , R. W. , and Gimble , J. M.
( 2001 ), Surface protein characterization of human adipose tissue - derived stromal cells ,
J. Cell Physiol. , 189 ( 1 ), 54 – 63 .
12. Zuk , P. A. , Zhu , M. , Ashjian , P. , et al. ( 2002 ), Human adipose tissue is a source of multipotent
stem cells , Mol. Biol. Cell , 13 ( 12 ), 4279 – 4295 .
13. Kern , S. , Eichler , H. , Stoeve , J. , Kluter , H. , and Bieback , K. ( 2006 ), Comparative analysis
of mesenchymal stem cells from bone marrow, umbilical cord blood, or adipose tissue ,
Stem Cells , 24 ( 5 ), 1294 – 1301 .
14. Lazarus , H. M. , Haynesworth , S. E. , Gerson , S. L. , Rosenthal , N. S. , and Caplan , A. I. ( 1995 ),
Ex vivo expansion and subsequent infusion of human bone marrow - derived stromal
progenitor cells (mesenchymal progenitor cells): Implications for therapeutic use , Bone
Marrow Transplant. , 16 ( 4 ), 557 – 564 .
15. Koc , O. N. , Peters , C. , Aubourg , P. , et al. ( 1995 ), Bone marrow - derived mesenchymal stem
cells remain host - derived despite successful hematopoietic engraftment after allogeneic
transplantation in patients with lysosomal and peroxisomal storage diseases . Exp.
Hematol. , 27 ( 11 ), 1675 – 1681 .
16. Koc , O. N. , Gerson , S. L. , Cooper , B. W. , et al. ( 2000 ), Rapid hematopoietic recovery after
coinfusion of autologous - blood stem cells and culture - expanded marrow mesenchymal
stem cells in advanced breast cancer patients receiving high - dose chemotherapy , J. Clin.
Oncol. , 18 ( 2 ), 307 – 316 .
17. Lee , S. T. , Jang , J. H. , Cheong , J. W. , et al. ( 2002 ), Treatment of high - risk acute myelogenous
leukaemia by myeloablative chemoradiotherapy followed by co - infusion of T cell -
depleted haematopoietic stem cells and culture - expanded marrow mesenchymal stem
cells from a related donor with one fully mismatched human leucocyte antigen haplotype ,
Br. J. Haematol. , 118 ( 4 ), 1128 – 1131 .
18. Koc , O. N. Day , J. , Nieder , M. , Gerson , S. L. , Lazarus , H. M. , and Krivit , W. ( 2002 ),
Allogeneic mesenchymal stem cell infusion for treatment of metachromatic leukodystrophy
(MLD) and Hurler syndrome (MPS - IH) , Bone Marrow Transplant. , 30 ( 4 ),
215 – 222 .
19. Le Blanc , K. , Rasmusson, I. , Sundberg , B. , et al. (2004), Treatment of severe acute graft-
versus - host disease with third party haploidentical mesenchymal stem cells , Lancet , 1,
363 ( 9419 ), 1439 – 1441 .
20. Lazarus , H. M. , Koc , O. N. , Devine , S. M. , et al. ( 2005 ), Cotransplantation of HLA - identical
sibling culture - expanded mesenchymal stem cells and hematopoietic stem cells in hematologic
malignancy patients , Biol. Blood Marrow Transplant. , 11 ( 5 ), 389 – 398 .
21. Ringden , O. , Uzunel , M. , Rasmusson , I. , et al. ( 2006 ), Mesenchymal stem cells for
treatment of therapy - resistant graft - versus - host disease , Transplantation , 81 ( 10 ), 1390 –
1397 .
22. Dominici , M. , Le Blanc , K. , Mueler , I. , et al. ( 2006 ), Minimal criteria for defi ning multipotent
mesenchymal stromal cells. The International Society for Cellular Therapy position
statement , Cytotherapy , 8 ( 4 ), 315 – 317 .
23. Schallmoser , K. , Bartmann , C. , Rohde , E. , et al. ( 2007 ), Human platelet lysate can replace
fetal bovine serum for clinical scale expansion of functional MSC , Transfusion . 2007 Aug;
47 ( 8 ), 1436 – 1446 .
24. Doucet , C. , Ernou , I. , Zhang , Y. Z. , et al. ( 2005 ), Platelet lysates promote mesenchymal
stem cell expansion: A safety substitute for animal serum in cell - based therapy applications
. J. Cell. Physiol. , 205 ( 2 ), 228 – 236 .
REFERENCES 113
114 GMP-COMPLIANT PROPAGATION
25. Digirolamo , C. M. , Stokes , D. , Colter , D. , Phinney , D. G. , Class , R. , and Prockop , D. J.
( 1999 ), Propagation and senescence of human marro stromal cells in culture: A simple
colony - forming assay identifi es samples with the greatest potential to propagate and differentiate
, Br. J. Haematol. , 107 ( 2 ), 275 – 281 .
26. Colter , D. C. , Class , R. , DiGirolamo , C. M. , Prockop , D. J. ( 2000 ), Rapid expansion of
recycling stem cells in cultures of plastic - adherent cells from human bone marrow , Proc.
Nat. Acad. Sci. USA , 97 ( 7 ), 3213 – 3218 .
27. Colter , D. C. , Sekiya , I. , and Prockop , D. J. ( 2001 ), Identifi cation of a subpopulation of
rapidly self - renewing and multipotential adult stem cells in colonies of human marrow
stromal cells , Proc. Nat. Acad. Sci. USA , 98 ( 14 ), 7841 – 7845 .
28. Sekiya , I. , Larson , B. L. , Smith , J. R. , Pochampally , R. , Cui , J. G. , and Prockop , D. J. ( 2002 ),
Expansion of human adult stem cells from bone marrow stroma: Conditions that maximie
the yields of early progenitors and evaluate their quality , Stem Cells , 20 ( 6 ), 530 – 541 .
29. Bartmann , C. , Rohde , E. , Schallmoser , K. , et al. ( 2007 ), Two steps to functional MSC for
clinical application , Transfusion . 2007 Aug; 47 ( 8 ), 1426 – 1435 .
30. Horwitz , E. M. , Gordon , P. L. , Koo , W. K. , et al. ( 2002 ), Isolated allogeneic bone marrow -
derived mesenchymal cells engraft and stimulate growth in children with osteogenesis
imperfecta: Implications for cells therapy of bone , Proc. Natl. Acad. Sci. USA , 99 ( 13 ),
8932 – 8937 .
31. Pollard , J. W. ( 1989 ), Basic cell culture , in Pollard , J. W. , Ed., Animal Cell Culture , Human ,
Clifton, NJ , pp. 1 – 2 .
32. WHO ( 1997 ), Medicinal and other products and human and animal transmissible spongiform
encephalopathies: Memorandum from a WHO meeting , Bull. World Health Org. ,
75 ( 6 ), 505 – 513 .
33. European Union ( 2004 ), Note for guidance on minimising the risk of transmitting animal
spongiform encephalopathy agents via human and veterinary medicinal products
(EMEA/410/01 Rev. 2 — October 2003) adopted by the Committee for Proprietary Medicinal
Products (CPMP) and by the Committee for Veterinary Medicinal Products (CVMP) ,
Off. J. Europ. Union , C24/26 – C24/19 .
34. Schallmoser , K. , Rosin , C. , Vormittag , R. , et al. ( 2006 ), Specifi cities of platelet autoantibodies
and platelet activation in lupus anticoagulant patients: A relation to their history
of thromboembolic disease , Lupus , 15 ( 8 ), 507 – 514 .
35. Wagner , W. , Wein , F. , Seckinger , A. , et al. ( 2005 ), Comparative characteristics of mesenchymal
stem cells from human bone marrow, adipose tissue, and umbilical cord blood ,
Exp. Hematol. , 33 ( 11 ), 1402 – 1416 .
36. Ramalho - Santos , M. , Yoon , S. , Matsuzaki , Y. , Mulligan , R. C. , and Melton , D. A. ( 2002 ),
“ Stemness ” : Transcriptional profi ling of embryonic and adult stem cells , Science , 298 ( 5593 ),
597 – 600 .
37. Boquest , A. C. , Shahdadfar , A. , Fronsdal , K. , et al. ( 2005 ), Isolation and transcription
profi ling of purifi ed uncultured human stromal stem cells: Alteration of gene expression
after in vitro cells culture , Mol. Biol. Cell , 16 ( 3 ), 1131 – 1141 .
38. Shahdadfar , A. , Fronsdal , K. , Haug , T. , Reinholt , F. P. , and Brinchmann , J. E. ( 2005 ), In
vitro expansion of human mesenchymal stem cells: Choice of serum is a determinant of
cell proliferation, differentiation, gene expression, and transcriptome stability , Stem Cells ,
23 ( 9 ), 1357 – 1366 .
39. Silva , W. A. , Jr. , Covas , D. T. , Panepucci , R. A. , et al. ( 2003 ), The profi le of gene expression
of human marrow mesenchymal stem cells , Stem Cells , 21 ( 6 ), 661 – 669 .
40. Panepucci , R. A. , Siufi , J. L. , Silva , W. A. , Jr. , et al. ( 2004 ), Comparison of gene expression
of umbilical cord vein and bone marrow - derived mesenchymal stem cells , Stem Cells ,
22 ( 7 ), 1263 – 1278 .
41. Jeong , J. A. , Hong , S. H. , Gang , E. J. , et al. ( 2005 ), Differential gene expression profi ling
of human umbilical cord blood - derived mesenchymal stem cells by DNA microarray ,
Stem Cells , 23 ( 4 ), 584 – 593 .
42. Monticone , M. , Liu , Y. , Tonachini , L. , et al. ( 2004 ), Gene expression profi le of human bone
marrow stromal cells determined by restriction fragment differential display analysis ,
J. Cell. Biochem , 92 ( 4 ), 733 – 744 .
43. Puck , T. T. , and Marcus , P. I. ( 1956 ), Action of x - rays on mammalian cells . J. Exp. Med. ,
103 ( 5 ), 653 – 666 .
44. Tepperman , A. D. , Curtis , J. E. , and McCulloch , E. A. ( 1974 ), Erythropietic colonies in
cultures of human marrow , Blood , 44 ( 5 ), 659 – 669 .
45. Pittenger , M. F. , Mackay , A. M. , Beck , S. C. , et al. ( 1999 ), Multilneage potential of adult
human mesenchymal stem cells , Science , 284 ( 5411 ), 143 – 147 .
46. Pittenger , M. F. , and Martin , B. J. ( 2004 ), Mesenchymal stem cells and their potential as
cardiac therapeutics , Circ. Res. , 95 ( 1 ), 9 – 20 .
47. Delorme , B. C. S. , and Charbord , P. ( 2006 ), The concept of mesenchymal stem cells , Regenerative
Med. , 1 ( 4 ), 497 – 509 .
48. Di Nicola , M. , Carlo - Stella , C. , Magni , M. , et al. ( 2002 ), Human bone marrow stromal
cells suppress T - lymphocyte proliferation induced by cellular or nonspecifi c mitogenic
stimuli , Blood , 99 ( 10 ), 3838 – 3843 .
49. Bartholomew , A. , Sturgeon , C. , Siatskas , M. , et al. ( 2002 ), Mesenchymal stem cells suppress
lymphocyte proliferation in vitro and prolong skin graft survival in vivo , Exp.
Hematol. , 30 ( 1 ), 42 – 48 .
50. Le Blanc , K. , Tammik , L. , Sundberg , B. , Haynesworth , S. E. , and Ringden , O. ( 2003 ),
Mesenchymal stem cells inhibit and stimulate mixed lymphocyte cultures and mitogenic
responses independently of the major histocompatibility complex , Scand. J. Immunol. ,
57 ( 1 ), 11 – 20 .
51. Tse , W. T. , Pendleton , J. D. , Beyer , W. M. , Egalka , M. C. , and Guinan , E. C. ( 2003 ), Suppression
of allogeneic T - cell proliferation by human marrow stromal cells: Implications
in transplantation , Transplantation , 75 ( 3 ), 389 – 397 .
52. Muller , I. , Kordowich , S. , Holzwarth , C. , et al. ( 2006 ), Animal serum - free culture conditions
for isolation and expansion of multipotent mesenchymal stromal cells from human
BM , Cytotherapy , 8 ( 5 ), 437 – 444 .
53. Reinisch , A. , Bartmann , C. , Rohde , E. S. K. , Bjelic - Radisic , V. , Lanzer , G. , Linkesch , W. ,
and Strunk , D. ( 2007 ), A humanized system to propagate cord blood - derived mesenchymal
stem cells for clinical application , Regen. Med. 2007 Jul; 2 ( 4 ), 371 – 382 .
54. Robinson , S. N. , Ng , J. , Niu , T. , et al. ( 2006 ), Superior ex vivo cord blood expansion following
co - culture with bone marrow - derived mesenchymal stem cells , Bone Marrow
Transplant. , 37 ( 4 ), 359 – 366 .
55. Speicher , M. R. , and Carter , N. P. ( 2005 ), The new cytogenetics: Blurring the boundaries
with molecular biology , Nat. Rev. Genet. , 6 ( 10 ), 782 – 792 .
56. Rubio , D. , Garcia - Castro , J. , Martin , M. C. , et al. ( 2005 ), Spontaneous human adult stem
cell transformation , Cancer Res. , 65 ( 8 ), 3035 – 3039 .
57. Tolar , J. , Nauta , A. J. , Osborn , M. J. , et al. ( 2007 ), Sarcoma derived from cultured mesenchymal
stem cells , Stem Cells , 25 ( 2 ), 371 – 379 .
REFERENCES 115
INTERNATIONAL REGULATIONS OF
GOOD MANUFACTURING PRACTICES
SECTION 2
119
2.1
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
NATIONAL GMP REGULATIONS AND
CODES AND INTERNATIONAL GMP
GUIDES AND GUIDELINES:
CORRESPONDENCES AND
DIFFERENCES
Marko N a rhi and Katrina Nordstr o m
Helsinki University of Technology, Helsinki, Finland
Contents
2.1.1 Introduction
2.1.2 National GMP Regulations and Codes
2.1.2.1 United States
2.1.2.2 Canada
2.1.2.3 European Union
2.1.2.4 East Asian Countries
2.1.2.5 India
2.1.2.6 Australia
2.1.2.7 New Zealand
2.1.2.8 South Africa
2.1.3 International GMP Guides and Harmonization
2.1.3.1 World Health Organization
2.1.3.2 Pharmaceutical Inspection Cooperation Scheme
2.1.3.3 International Conference on Harmonization
2.1.3.4 Association of Southeast Asian Nations (ASEAN)
2.1.3.5 Mercado Comun del Sur (MERCOSUR)
2.1.4 Correspondences of the U.S. GMP Regulations with GMP Codes and Guidelines
2.1.4.1 General Issues
2.1.4.2 Organization and Personnel
2.1.4.3 Buildings and Facilities
2.1.4.4 Equipment
2.1.4.5 Control of Components and Drug Product Containers and Closures
2.1.4.6 Production and Process Controls
120 CORRESPONDENCES AND DIFFERENCES
2.1.4.7 Packaging and Labeling Control
2.1.4.8 Holding and Distribution
2.1.4.9 Laboratory Controls
2.1.4.10 Records and Reports
2.1.4.11 Returned and Salvaged Drug Products
References
2.1.1 INTRODUCTION
The fi rst predecessors of manufacturing and quality requirements, which later
evolved into good manufacturing practices (GMPs), were issued in the 1940s in the
United States by the Food and Drug Administration (FDA) [1] . In the general
meeting of the World Health Organization (WHO) held in 1969, the World Health
Assembly issued a recommendation for the introduction of GMPs [2] . Since then,
most industrialized countries have passed laws on control procedures essential for
the manufacture of drug products. In some countries GMPs are integrated into
national legislation as a part of laws or regulations on production, distribution,
marketing, and use of drug products (GMP regulations). In other countries, GMPs
are separate guidelines outside the national drug legislation (GMP codes). In addition
to national GMPs, also some international organizations and trade blocks have
issued their own international GMP guidelines to harmonize the requirements for
drug production in different countries. However, regardless of their origin, the main
purpose of GMPs is to ensure that manufactured drug products have the safety,
identity, potency, purity, and quality that they are presented to have [3] . To fulfi ll
this aim, most GMPs usually cover quality management, personnel, premises, equipment,
documentation, materials management, production and in - process controls,
packaging and labeling of intermediate and fi nished products, laboratory controls,
validation, and change controls [4] .
2.1.2 NATIONAL GMP REGULATIONS AND CODES
2.1.2.1 United States
In the United States the production of drug products is controlled under the federal
Food, Drug and Cosmetic Act, which states that a drug product will be deemed to
be adulterated unless the methods used in or the facilities or controls used for its
manufacture, processing, packaging, or holding conform to or are operated or
administered in conformity with current GMP [5] . The actual GMP regulations are
issued as a part of the Code of Federal Regulations and as such they are a federal
law. The current set of GMP regulations is based on the 1978 revision [6, 7] of the
original GMP regulations, which were fi rst promulgated in 1963. The GMP regulations
are updated every year in April [8] ; however, no major changes have been
implemented since 1978. As an addition to GMP regulations, the FDA also publishes
other GMP - related guidance documents covering various issues of drug manufacturing
[9] . On the other hand, although these documents refl ect current views and
NATIONAL GMP REGULATIONS AND CODES 121
expectations of the agency, they only provide guidance on principles and practices
that are not legal requirements [1] . As a member of the International Conference
on Harmonization of Technical Requirements for Registration of Pharmaceutical
for Human Use (ICH), the United States has adopted the ICH guidance document
Q7, Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients, and
published it as a guidance for industry document [10] .
The U.S. GMP regulations are divided into two parts: 210 [6] and 211 [7] . Part
210, “ Current Good Manufacturing Practice in Manufacturing, Processing, Packing
or Holding of Drugs — General, ” provides the framework for the regulations [6] , and
Part 211, “ Current Good Manufacturing Practice for Finished Pharmaceuticals, ”
states the actual requirements. Part 211 is further divided into 11 subparts, which
cover the requirements for personnel, premises, equipment, control of materials,
production and process controls, packaging and labeling control, holding and distribution,
laboratory controls, documentation, and returned and salvaged products [7] .
The contents of Part 211 are presented in Table 1 .
2.1.2.2 Canada
The production of drug products (drugs) in Canada is controlled under the Food
and Drugs Act, which states that distributors and importers are not allowed to sell
a drug product unless it has been manufactured according to the requirements of
GMP. The principles of GMP are laid down by Division 2 in Part C of the Food and
Drug Regulations, which is a part of the Food and Drugs Act [11] . The Health
Products and Food Branch Inspectorate has also issued a guidance document (GMP
code), which has been prepared to assist in the interpretation of GMP regulations.
The current set of the Canadian GMP code was issued in 2002 and has not been
revised since. It has been written with a view to harmonization with GMP standards
of other countries and international organizations [WHO, Pharmaceutical Inspection
Cooperation Scheme (PIC/S), ICH]. Canadian Healthcare authorities have also
published several annexes to the basic GMP code covering topics such as GMP for
medical gases, biological drug products, blood products, and production of investigational
new drugs. In addition to the GMP code and its annexes, the Canadian
TABLE 1 Contents of Part 211 of U . S . GMP Regulations [7]
Section Subject
Subpart A General provisions
Subpart B Organization and personnel
Subpart C Buildings and facilities
Subpart D Equipment
Subpart E Control of components and drug product containers and closures
Subpart F Production and process controls
Subpart G Packaging and labeling control
Subpart H Holding and distribution
Subpart I Laboratory controls
Subpart J Records and reports
Subpart K Returned and salvaged drug products
122 CORRESPONDENCES AND DIFFERENCES
authorities have also issued several other specifi c guidelines dealing with issues
related to GMP and manufacturing methods [12] .
As shown in Table 2 the Canadian GMP code can be divided into four chapters
and three annexes. The fi rst three chapters cover general issues such as scope and
applicability of the code, defi nitions of used terms, and issues concerning quality
management and GMP in general. GMP regulations and their application are presented
in the fourth chapter ( “ Regulation ” ), which is divided into 14 subchapters
covering the requirements for premises, equipment, personnel, sanitation, testing of
components and packaging materials, testing of fi nished product, production control,
quality control department, documentation, reserve samples, stability testing, and
manufacture of sterile drug products and medical gases. Each subchapter contains
the corresponding regulation according to regulations in Division 2 [11] issued with
a rationale and interpretation to assist in their application. The annexes include
requirements for batch certifi cation, application form for alternate sample retention
site, and references such as hyperlinks to Canadian laws concerning drug products
and other GMP - related national and international guidelines [12] .
2.1.2.3 European Union
The production of drug products (medicinal products) in the European Union (EU)
is controlled under Directive 2001/83/EC of the European parliament and of the
Council, which states that the holder of a manufacturing authorization for medicinal
products is obliged to comply with good manufacturing practices as laid down by
European Community law [13] . The principles and guidelines of GMP for medicinal
products are stated by the Commission directive 2003/94/EC, which provides the
TABLE 2 Contents of Canadian GMP Code [12]
Introduction
Quality management
Glossary of terms
Regulation
Premises
Equipment
Personnel
Sanitation
Raw material testing
Manufacturing control
Quality control department
Packaging material testing
Finished product testing
Records
Samples
Stability
Sterile products
Medical gases
Annex A: Internationally Harmonized Requirements for
Batch Certifi cation
Annex B: Application for Alternate Sample Retention
Annex C: References
NATIONAL GMP REGULATIONS AND CODES 123
legal basis for GMP in the EU [14] . The actual GMP code with detailed written
procedures is published in The Rules Governing Medicinal Products in the European
Union , volume 4. The current set of the EU GMP code was fi rst introduced in 1989
consisting of nine chapters covering the general requirements of GMP and one
annex on the manufacture of sterile drug products. Since then the EU GMP code
has been revised many times and several new annexes have been issued [15] . In
addition to the GMP code, the EU has also published several other guidelines concerning
the quality issues of drug production in The Rules Governing Medicinal
Products in the European Union , volume 3A [16] .
As shown in Tables 3 – 5 , the EU GMP code is presented in two parts of basic
requirements and 18 annexes. Part I, “ Basic Requirements for Medicinal Products, ”
covers GMP principles for the manufacture of drug products. It consists of nine
chapters covering the requirements for quality management and control, personnel,
premises, equipment, documentation, production, contract services, complaints,
product recall, and self - inspection. Part II, “ Basic Requirements for Active Substances
Used as Starting Materials, ” covers GMPs for active substances used as
starting materials. It is based on the ICH document Q7, Good Manufacturing Practice
Guide for Active Pharmaceutical Ingredients , and was originally introduced in
2001 as Annex 18 of the EU GMP code. In the restructured revision of the EU
GMP code issued in October 2005, Annex 18 was replaced with Part II. It consists
of 19 chapters, which cover basic GMP issues related to quality management, personnel,
premises, equipment, documentation, materials, production and process controls,
packaging and labeling, storage and distribution, laboratory controls, validation,
change control, complaints, recalls, contract services, co - operators, active pharmaceutical
ingredients (APIs) manufactured by cell culture/fermentation, and APIs
used in clinical trials. The annexes give more detailed specifi c guidance on the
manufacture of sterile drug products, biological drug products, radiopharmaceuticals,
veterinary drug products, medical gases, herbal drug products, oral liquids,
external preparations (creams, ointments), aerosols, investigational new drugs, and
blood and blood products. They also cover sampling of materials, computerized
systems, use of ionizing radiation, qualifi cation and validation, batch release, parametric
release, reference, and retention samples [15] .
TABLE 3 Contents of Part I of EU GMP Code Covering
Basic Requirements for Manufacture of Drug Products [15]
Section Subject
Introduction
Chapter 1 Quality management
Chapter 2 Personnel
Chapter 3 Premises and equipment
Chapter 4 Documentation
Chapter 5 Production
Chapter 6 Quality control
Chapter 7 Contract manufacture and analysis
Chapter 8 Complaints and product recall
Chapter 9 Self - inspection
Glossary
124 CORRESPONDENCES AND DIFFERENCES
TABLE 4 Contents of Part II of EU GMP Code Covering Basic Requirements for
Manufacture of Active Substances Used as Starting Materials [15]
Section Subject
1 Introduction
2 Quality management
3 Personnel
4 Buildings and facilities
5 Process equipment
6 Documentation and records
7 Materials management
8 Production and in - process controls
9 Packaging and identifi cation labeling of APIs and intermediates
10 Storage and distribution
11 Laboratory controls
12 Validation
13 Change control
14 Rejection and reuse of materials
15 Complaints and recalls
16 Contract manufacturers (including laboratories)
17 Agents, brokers, traders, distributors, repackers, and relabelers
18 Specifi c guidance for APIs manufactured by cell culture/fermentation
19 APIs for use in clinical trials
20 Glossary
TABLE 5 Annexes of EU GMP Code Covering Specifi c Guidance [15]
Section Subject
Annex 1 Manufacture of sterile medicinal products
Annex 2 Manufacture of biological medicinal products for human use
Annex 3 Manufacture of radiopharmaceuticals
Annex 4 Manufacture of veterinary medicinal products other than immunological
veterinary medicinal products
Annex 5 Manufacture of immunological veterinary medicinal products
Annex 6 Manufacture of medicinal gases
Annex 7 Manufacture of herbal medicinal products
Annex 8 Sampling of starting and packaging materials
Annex 9 Manufacture of liquids, creams, and ointments
Annex 10 Manufacture of pressurised metered - dose aerosol preparations for inhalation
Annex 11 Computerized systems
Annex 12 Use of ionizing radiation in manufacture of medicinal products
Annex 13 Manufacture of investigational medicinal products
Annex 14 Manufacture of products derived from human blood or human plasma
Annex 15 Qualifi cation and validation
Annex 16 Certifi cation by a qualifi ed person and batch release
Annex 17 Parametric release
Annex 19 Reference and retention samples
NATIONAL GMP REGULATIONS AND CODES 125
2.1.2.4 East Asian Countries
Japan In Japan the production of drug products (drugs) is regulated under the
Pharmaceuticals Affairs Law (PAL), which states that any drug manufacturer who
plans to manufacture a drug product for sale in Japan must have a Japanese drug
manufacturing license and comply with Japanese GMP requirements. The fi rst regulations
of Japanese GMP were introduced in 1974 as The Standards for Manufacturing
Control and Quality Control . In 1979 PAL was partially revised and GMPs
became legally binding [2] .
PAL is managed and enforced via ministerial ordinances and notices, which are
detailed regulations prepared by the Japanese government. The requirements for
premises for drug manufacture are given in Ministry of Health, Labor and Welfare
(MHLW) Ministerial Ordinance No. 73, 2005 Regulations for Buildings and Facilities
for Pharmacies, etc. [originally Ministry of Health and Welfare (MHW) Ministerial
Ordinance No. 2, 1961] [17] , and the requirements for manufacturing and quality
controls in MHLW Ministerial Ordinance No. 95, 2003 Regulations for Manufacturing
Control and Quality Control of Drugs (originally MHW Ministerial Ordinance
No. 3, 1994). As a member of the ICH Japan has adopted the ICH guidance
document Q7, Good Manufacturing Practice Guide for Active Pharmaceutical
Ingredients , and published it as Pharmaceutical and Food Safety Bureau (PFSB)
Director - General Notifi cation No. 1200, 2001 Guidelines on GMP for Drug Substances
, which states the requirements for the manufacture of APIs. The requirements
concerning imported drug products are given in MHLW Ministerial Ordinance
No. 97, 2003 Regulations for Importing/Retail Management and Quality Control of
Drugs and Quasi - Drugs (originally MHW Ministerial Ordinance No. 62, 1999). The
requirements specifying manufacture of investigational products are given in PAB
Notifi cation No. 480, 1997 Products and Standards for the Buildings and Facilities of
Manufacturing Plants for Investigational Products (Investigational Product GMP)
[2] .
South Korea The production of drug products (drugs) in South Korea is regulated
under the Pharmaceutical Affairs Law, which was fi rst enacted in 1953 and has since
been revised several times [18] . New drug approval and related activities are regulated
in much the same way as in the United States and Japan. Korean GMP, which
is often called KGMP, was initiated in 1984 and became mandatory in 1995 [19] . A
drug manufacturer who intends to manufacture a drug product for sale in Korea
must have approval from the Commissioner of the Korea Food and Drug Administration
(KFDA). In order to require the license for manufacturing business the
manufacturer has to prove the compliance of facility standards with KGMP [20] .
China China regulates the production of drug products (drugs) under the Drug
Administration Law of the People ’ s Republic of China, which states that a drug
manufacturer has to conduct drug manufacture according to the GMP for pharmaceutical
products formulated by the Drug Regulatory Department under the State
Council on the basis of the Drug Administration Law [21] . In June 2004 GMP
became mandatory in China and the State Drug Administration announced that
local drug manufacturing establishments lacking approved GMP certifi cation would
not be allowed to continue the production of pharmaceuticals [22] .
126 CORRESPONDENCES AND DIFFERENCES
2.1.2.5 India
The production of drug products (drugs) in India is controlled under the Drugs and
Cosmetics Rules (1945, last amended in 2005), which states that the holder of the
license to manufacture drugs has to comply with the requirements of GMP as laid
down in Schedule M [23] . Schedule M is a part of the Drugs and Cosmetics Rules
and embodies the Indian GMP regulations [24] , which are based on the 1982 version
of WHO GMP guidelines [25] .
As shown in Tables 6 – 8 the Indian GMP regulations consists of eight parts: I, IA,
IB, IC, ID, IE, IF, and II. Part I covers the general requirements of GMP. It is divided
into 29 chapters, which deal with the requirements for personnel, premises, equipment,
sanitation, production and process controls, materials, documentation, quality
management, validation, reserve samples, recalls, complaints, and self - inspection.
Parts IA to IE cover specifi c requirements for the manufacture of different dosage
forms regarding premises, equipment, and methods. Part IA deals with the require-
TABLE 6 Contents of Part I of Indian GMP Regulations
Covering Good Manufacturing Practices for Premises and
Materials [24]
Section Subject
1 General requirements
2 Warehousing area
3 Production area
4 Ancillary areas
5 Quality control area
6 Personnel
7 Health, clothing, and sanitation of workers
8 Manufacturing operations and control
9 Sanitation in the manufacturing premises
10 Raw materials
11 Equipment
12 Documentation and records
13 Labels and other printed materials
14 Quality assurance
15 Self - inspection and quality audit
16 Quality control system
17 Specifi cation
18 Master formula records
19 Packing records
20 Batch packaging records
21 Batch processing records
22 Standard operating procedures (SOPs) and records
23 Reference samples
24 Reprocessing and recoveries
25 Distribution records
26 Validation and process validation
27 Product recalls
28 Complaints and adverse reactions
29 Site master fi le
NATIONAL GMP REGULATIONS AND CODES 127
ments for the manufacture of parenteral preparations; Part IB with the requirements
for the manufacture of oral solid dosage forms such as tablets and capsules;
Part IC with the requirements for the manufacture of oral liquids such as syrups,
elixirs, emulsions, and suspensions; Part ID with the requirements for the manufacture
of external preparations such as creams, ointments, pastes, emulsions, and
lotions; and Part 1E with the requirements for the manufacture of inhalers. Part 1F
covers specifi c requirements for the manufacture of APIs regarding buildings and
facilities, utilities, equipment, controls, and containers. Part II of the Indian GMP
regulations consist of detailed recommendations for the process equipment to be
used in the manufacture of different dosage forms and requirements for the partition
of the production area [24] .
TABLE 7 Contents of Parts IA , IB , IC , ID , IE , and IF of Indian GMP Regulations
Covering Specifi c Guidance [24]
Section Subject
Part IA Specifi c requirements for manufacture of sterile products, parenteral
preparations (small - volume injectables and large - volume parenterals) and
sterile ophthalmic preparations
Part IB Specifi c requirements for manufacture of oral solid dosage forms (tablets and
capsules)
Part IC Specifi c requirements for manufacture of oral liquids (syrups, elixirs,
emulsions, and suspensions)
Part ID Specifi c requirements for manufacture of topical products, i.e., external
preparations (creams, ointments, pastes, emulsions, lotions, solutions, dusting
powders, and identical products)
Part IE Specifi c requirements for manufacture of metered - dose inhalers (MDIs)
Part IF Specifi c requirements of premises, plant, and materials for manufacture of
active pharmaceutical ingredients (bulk drugs)
TABLE 8 Contents of Part II of Indian GMP Regulations
Covering Requirements of Plant and Equipment [24]
Section Subject
1 External preparations
2 Oral liquid preparations
3 Tablets
4 Powders
5 Capsules
6 Surgical dressing
7 Ophthalmic preparations
8 Pessaries and suppositories
9 Inhalers and vitrallae
10 Repacking of drugs and pharmaceutical chemicals
11 Parenteral preparations
128 CORRESPONDENCES AND DIFFERENCES
2.1.2.6 Australia
In Australia the production of drug products (medicinal products) is controlled
under the Therapeutics Goods Act, which provides the Minister for Health and
Aged Care the right to determine written principles including codes of GMP to be
observed in the production of drug products for use in humans [26] . The Therapeutic
Goods (Manufacturing Principles) Determination No. 2 of 2002 given by the
minister states that drug products must be manufactured in compliance with the
Australian Code of Good Manufacturing Practice for Medicinal Products , dated
August 16, 2002 [27] . The current set of the Australian GMP code is based entirely
on the PIC/S GMP guide version PH 1/97 (Rev. 3) published in 2002 with some
minor modifi cations [28] .
As shown in Table 9 , the Australian GMP code consists of 9 chapters and 13
annexes. The chapters present the general requirements of GMP for the manufacture
of drug products, the requirements for quality management and control,
personnel, premises, equipment, documentation, production, contract services, complaints,
product recall, and self - inspection. The annexes give specifi c guidance on
the manufacture of sterile drug products, biological drug products, radiopharmaceuticals,
medical gases, herbal drug products, oral liquids, external preparations (creams,
ointments), aerosols, investigational new drugs, blood, and blood products. They also
TABLE 9 Contents of Australian GMP Code [28]
Section Subject
Introduction
Interpretation
Chapter 1 Quality management
Chapter 2 Personnel
Chapter 3 Premises and equipment
Chapter 4 Documentation
Chapter 5 Production
Chapter 6 Quality control
Chapter 7 Contract manufacture and analysis
Chapter 8 Complaints and product recall
Chapter 9 Self - inspection
Annex 1 Manufacture of sterile medicinal products
Annex 2 Manufacture of biological medicinal products for human use
Annex 3 Manufacture of radiopharmaceuticals
Annex 6 Manufacture of medicinal gases
Annex 7 Manufacture of herbal medicinal products
Annex 8 Sampling of starting and packaging materials
Annex 9 Manufacture of liquids, creams, and ointments
Annex 10 Manufacture of pressurised metered - dose aerosol preparations for
inhalation
Annex 11 Computerized systems
Annex 12 Use of ionizing radiation in the manufacture of medicinal products
Annex 13 Manufacture of investigational medicinal products
Annex 15 Qualifi cation and validation
Annex 17 Parametric release
Glossary
NATIONAL GMP REGULATIONS AND CODES 129
cover sampling of materials, computerized systems, use of ionizing radiation, quali-
fi cation, and validation and parametric release [28] .
Australia has not adopted Annexes 4, 5, 14, 16, and 18 of the PIC/S GMP guide.
Annexes 4 and 5 cover the manufacture of veterinary drug products. Annex 14
covers the manufacture of products derived from human blood or human plasma,
which is excluded from the Australian GMP code. Annex 16 is specifi c to the EU
GMP code and Annex 18 is the ICH GMP guide for the manufacture of APIs, which
Australia has adopted separately as a manufacturing principle [28] .
2.1.2.7 New Zealand
The production of drug products (medicines) in New Zealand is controlled under
the Medicines Act 1981, which states that a drug manufacturer is not allowed to
manufacture drug products without a manufacturing license issued by the licensing
authority. In order to obtain a manufacturing license the applicant must satisfy the
licensing authority with respect to the proposed manufacturing premises and equipment,
which must be suitable and adequate for the manufacture of drugs. Moreover,
the applicant must show that adequate arrangements have been made or are to be
made for the making, maintaining, and safekeeping of adequate records with reference
to the drug products that are to be manufactured [29] . The authorities (Medsafe)
require that any drug manufacturer who plans to manufacture drug products for
sale in New Zealand must deliver evidence of GMP compliance for the manufacturing
site. Copies of appropriate certifi cates, manufacturing licenses, or reports issued
by a regulatory authority whose competence is recognized by Medsafe are accepted
as proof of GMP compliance [30] .
As shown in Table 10 New Zealand ’ s own GMP code consists of fi ve parts. The
fi rst part covers the manufacture of drug products and the second part the manufacture
of blood products. Part 3 covers compounding and dispensing, including
compounding of sterile drug products. Part 4 deals with wholesaling and Part 5 with
product recalls. Parts 4 and 5 are combined in one document [31] .
2.1.2.8 South Africa
South Africa controls the production of drug products (medicines) under the Medicines
and Related Substances Control Act (Act 101 of 1965), which states that the
Medicines Control Council may issue to a drug manufacturer a license to manufacture
a drug product upon such conditions as to the application of such acceptable
TABLE 10 Contents of New Zealand ’ s GMP Code [31]
Section Subject
Part 1 Manufacture of pharmaceutical products
Part 2 Manufacture of blood and blood products
Part 3 Compounding and dispensing
Part 4 Wholesaling of medicines and medical devices
Part 5 Uniform recall procedure for medicines and
medical devices
130 CORRESPONDENCES AND DIFFERENCES
quality assurance principles and GMPs as the council may determine [32] . As a part
of the license application the manufacturer must provide acceptable documentary
proof of the ability to comply with GMP as determined by the council [33] . The
current set of South African GMP code determined by the council is entirely based
on the PIC/S GMP guide version PE 009 - 2, published in 2004 with some minor
modifi cations [34] .
As shown in Table 11 the South African GMP code consists of 9 chapters and 17
annexes. The chapters present the general requirements of GMP for the production
of drug products covering the requirements for quality management and control,
personnel, premises, equipment, documentation, production, contract services, complaints,
product recall, and self - inspection. The annexes give specifi c guidance on
the manufacture of sterile drug products, biological drug products, radiopharmaceuticals,
veterinary drug products, medical gases, herbal drug products, oral liquids,
external preparations (creams, ointments), aerosols, investigational new drugs, and
blood and blood products. They also cover sampling of materials, computerized
systems, use of ionizing radiation, qualifi cation and validation, organization, and
TABLE 11 Contents of South African GMP Code [34]
Section Subject
Introduction
Chapter 1 Quality management
Chapter 2 Personnel
Chapter 3 Premises and equipment
Chapter 4 Documentation
Chapter 5 Production
Chapter 6 Quality control
Chapter 7 Contract manufacture and analysis
Chapter 8 Complaints and product recall
Chapter 9 Self - inspection
Annex 1 Manufacture of sterile medicinal products
Annex 2 Manufacture of biological medicinal products for human use
Annex 3 Manufacture of radiopharmaceuticals
Annex 4 Manufacture of veterinary medicinal products other than
immunologicals
Annex 5 Manufacture of immunological veterinary medical products
Annex 6 Manufacture of medicinal gases
Annex 7 Manufacture of herbal medicinal products
Annex 8 Sampling of starting and packaging materials
Annex 9 Manufacture of liquids, creams, and ointments
Annex 10 Manufacture of pressurized metered - dose aerosol preparations for
inhalation
Annex 11 Computerized systems
Annex 12 Use of ionizing radiation in the manufacture of medicinal products
Annex 13 Manufacture of investigational medicinal products
Annex 14 Manufacture of products derived from human blood or human plasma
Annex 15 Qualifi cation and validation
Annex 16 Organisation and personnel
Annex 17 Parametric release
Glossary
personnel and parametric release. The original Annex 16, which is specifi c to the
EU GMP code, has been replaced in South African GMP code with an annex covering
organization and personnel. Nor has South Africa adopted Annex 18, which
covers the ICH GMP guide for the manufacture of APIs, as it has been adopted
separately as a manufacturing principle [34] .
2.1.3 INTERNATIONAL GMP GUIDES AND HARMONIZATION
2.1.3.1 World Health Organization
The WHO was established in 1948 as a specialized agency of the United Nations
(UN). Its purpose is to serve as the directing and coordinating authority for international
health matters and public health. One of the main functions of the WHO
is to provide objective and reliable information and advice in the fi eld of human
health, a task that it partly fulfi lls through WHO publications [35] . The fi rst WHO
draft text on GMP was prepared in 1967 and a revised version was published in
1968 as an annex of the twenty - second report of the WHO expert committee on
specifi cations for pharmaceutical preparations. Over the years the WHO has issued
several versions of its GMP guidelines as well as other guidelines related to the
GMP and quality issues of the production of therapeutic products. The latest version
of the WHO GMP guideline was published in 2003 as an annex of the WHO Technical
Report 908 [36] .
As shown in Table 12 the WHO GMP guideline is divided into fi ve parts: introduction,
general considerations, glossary, quality management in the drug industry,
and references. The actual GMP guidelines are presented in the fourth part, which
consists of 17 chapters covering the requirements for quality assurance and control,
personnel, premises, equipment, sanitation, materials, validation, documentation,
production, contract services, complaints, recalls, and self - inspection [36] . In addition
to this guideline laying down the main principles of GMP, the WHO has also published
several other guidelines covering specifi c requirements for components,
quality of water for pharmaceutical use, APIs, excipients, sterile drug products,
biological drug products, investigational drug products, herbal drug products, and
radiopharmaceuticals (Table 13 ).
2.1.3.2 Pharmaceutical Inspection Cooperation Scheme
The Pharmaceutical Inspection Convention (PIC), which is the predecessor of
PIC/S, was founded in 1970 by the European Free Trade Area (EFTA). The initial
members comprised of the 10 EFTA member countries at that time. From the beginning
one of the main goals has been the harmonization of GMP requirements as
well as the promotion of mutual recognition of inspections and uniformity of inspection
systems by training the inspectors, improving the exchange of information, and
mutual confi dence [46] . Originally PIC was a formal treaty between member countries
and as such it also had a legal status. When countries outside Europe were
seeking to join PIC, it became evident that, according to European law, individual
EU countries that were members of PIC were not permitted to sign agreements
with countries outside Europe. Only the European Commission, which itself was
INTERNATIONAL GMP GUIDES AND HARMONIZATION 131
132 CORRESPONDENCES AND DIFFERENCES
not a member of PIC, was permitted to sign agreements. Consequently, a less formal
and more fl exible PIC/S was developed to continue the work of PIC. The PIC/S,
which became operational in November 1995, is an informal arrangement without
legal status between regulatory authorities instead of countries. The PIC and the
PIC Scheme, operating together as PIC/S, provide an active and constructive cooperation
in the fi eld of GMP [47] .
The current members of PIC/S are Australia, Austria, Belgium, Canada, Czech
Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland,
TABLE 12 Contents of WHO GMP Guideline Covering General Requirements of GMP
for Manufacture of Drug Products [36]
Introduction
General considerations
Glossary
Quality management in the drug industry: philosophy and essential elements
Section Subject
1 Quality assurance
2 Good manufacturing practices for pharmaceutical products (GMP)
3 Sanitation and hygiene
4 Qualifi cation and validation
5 Complaints
6 Product recalls
7 Contract production and analysis
8 Self - inspection and quality audits
9 Personnel
10 Training
11 Personal hygiene
12 Premises
13 Equipment
14 Materials
15 Documentation
16 Good practices in production
17 Good practices in quality control
References
TABLE 13 GMP -Related WHO Documents Covering Specifi c Guidance
Document Subject
TRS 929, Annex 2 [37] Requirement for the sampling of starting materials
TRS 823, Annex 1 [38] Active pharmaceutical ingredients (bulk drug substances)
TRS 885, Annex 5 [39] Pharmaceutical excipients
TRS 902, Annex 6 [40] Sterile pharmaceutical products
TRS 834, Annex 3 [41] Biological products
TRS 863, Annex 7 [42] Investigational pharmaceutical products for clinical trials
in humans
TRS 863, Annex 8 [43] Herbal medicinal products
TRS 908, Annex 3 [44] Radiopharmaceutical products
TRS 929, Annex 3 [45] Water for pharmaceutical use
Italy, Latvia, Liechtenstein, Malaysia, Netherlands, Norway, Poland, Portugal,
Romania, Singapore, Slovak Republic, Spain, Sweden, Switzerland, and the United
Kingdom. In addition, Estonia, the European Agency for the Evaluation of Medicinal
Products (EMEA), UNICEF, and the WHO participate in PIC/S activities as
observers [48] . Also many other regulatory authorities have shown interest in joining
PIC/S, in particular Argentina, Brazil, Cyprus, Indonesia, Israel, Philippines,
Slovenia, Thailand, the United States, Bulgaria, Estonia, Lithuania, Oman, Russia,
South Africa, and the Ukraine [49] .
To become a PIC/S member, a joining regulatory authority is required to go
through a detailed assessment to prove that the authority has the arrangements and
competence necessary to apply an inspection system equivalent to inspection
systems of existing PIC/S members. To ensure that both new applicants and older
members fulfi ll the same requirements, also existing members are reassessed on a
regular basis. One of the main functions of PIC/S is to develop GMP guidance documents,
which it carries out in close cooperation with the EU and relevant agencies
thereof. Under this cooperation both parties have been able to adopt each others ’
documents, thus minimizing the duplication of effort in development of GMP -
related documents. Among other highly informative guides on various aspects of
GMP and quality issues [49] , PIC/S has also published its own GMP guide ( Guide
to Good Manufacturing Practice for Medicinal Products ), which is harmonized with
the EU GMP code [50] .
The latest revision of the PIC/S GMP guide (version PE 009 - 3) was issued in
January 2006. As shown in Table 14 , it consists of 9 chapters and 16 annexes. Chapters
present the general requirements of GMP for the production of drug products
covering the requirements for quality management and control, personnel, premises,
equipment, documentation, production, contract services, complaints, product recall,
and self - inspection. The annexes give specifi c guidance on the manufacture of sterile
drug products, biological drug products, radiopharmaceuticals, veterinary drug products,
medical gases, herbal drug products, oral liquids, external preparations (creams,
ointments), aerosols, investigational new drugs, and blood and blood products. In
addition, there are annexes covering the sampling of materials, computerized
systems, use of ionizing radiation, qualifi cation and validation, and parametric
release [50] .
Although the PIC/S GMP guide is harmonized with the EU GMP code and their
contents are similar, there are some minor differences between them. Instead of the
term qualifi ed person , the PIC/S GMP guide uses the term authorized person . Furthermore,
all references to EU directives have been deleted from the PIC/S GMP
guide. Moreover, PIC/S has not adopted Annexes 16 and 18 of the EU GMP code.
Annex 16 is specifi c to the EU GMP code covering the status of a qualifi ed person
in batch release and Annex 18 is the ICH GMP guide for the manufacture of APIs,
which the PIC/S Committee has adopted as a stand - alone document (PE 007)
[50] .
2.1.3.3 International Conference on Harmonization
The ICH was established in 1990. Its main aim is to improve the effi ciency of the
drug development process and the registration of new drug products in its member
countries through harmonization of national guidelines. This is a joint initiative
INTERNATIONAL GMP GUIDES AND HARMONIZATION 133
134 CORRESPONDENCES AND DIFFERENCES
involving both regulators and industry as equal partners. The founders and current
members of ICH, which represent the regulatory bodies and the research - based
industry in the member countries, are the EU, European Federation of Pharmaceutical
Industries and Associations (EFPIA), MHLW, Japan Pharmaceutical Manufacturers
Association (JPMA), FDA, and Pharmaceutical Research and Manufactures
of America (PhRMA). In addition to the actual member countries there are also
observers who act as a link between ICH and non - ICH countries and regions.
Current observers are the WHO, EFTA, Swissmedic (representing Switzerland), and
Health Canada (representing Canada) [51] .
Among other guidelines, ICH has also published a guide on GMP for APIs (Q7:
Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients ). It is
intended to provide guidance regarding GMP for the manufacture of APIs and to
help ensure that APIs meet the quality and purity requirements that they are presented
to possess. This covers APIs that are manufactured by chemical synthesis,
extraction, cell culture/fermentation, recovery from natural sources, or any combination
of these processes. Excluded are vaccines, medical gases, bulk - packaged drug
TABLE 14 Contents of PIC / S GMP Guide [50]
Section Subject
Introduction
Chapter 1 Quality management
Chapter 2 Personnel
Chapter 3 Premises and equipment
Chapter 4 Documentation
Chapter 5 Production
Chapter 6 Quality control
Chapter 7 Contract manufacture and analysis
Chapter 8 Complaints and product recall
Chapter 9 Self - inspection
Annex 1 Manufacture of sterile medicinal products
Annex 2 Manufacture of biological medicinal products for human use
Annex 3 Manufacture of radiopharmaceuticals
Annex 4 Manufacture of veterinary medicinal products other than
immunologicals
Annex 5 Manufacture of immunological veterinary medical products
Annex 6 Manufacture of medicinal gases
Annex 7 Manufacture of herbal medicinal products
Annex 8 Sampling of starting and packaging materials
Annex 9 Manufacture of liquids, creams, and ointments
Annex 10 Manufacture of pressurised metered - dose aerosol preparations for
inhalation
Annex 11 Computerized systems
Annex 12 Use of ionizing radiation in manufacture of medicinal products
Annex 13 Manufacture of investigational medicinal products
Annex 14 Manufacture of products derived from human blood or human plasma
Annex 15 Qualifi cation and validation
Annex 17 Parametric release
Glossary
products, radiopharmaceuticals, whole cells, whole blood and plasma, blood and
plasma derivatives, and gene therapy APIs. However, APIs that are produced
using blood or plasma as raw materials are included [52] . All ICH member countries
have adopted this guideline: the EU in November 2000, Japan in November
2001, and the United States in September 2001 [53] . In addition, it has also been
adopted by several other non - ICH countries such as Australia [28] and South Africa
[34] .
The basic structure of the ICH GMP guideline for API production is shown in
Table 15 . It consists of 19 chapters, which cover the requirements for quality management,
personnel, premises, equipment, documentation, materials, production and
process controls, packaging and labeling, storage and distribution, laboratory controls,
validation, change control, complaints, recalls, contract services, cooperators,
APIs manufactured by cell culture/fermentation, and APIs used in clinical trials
[52] .
2.1.3.4 Association of Southeast Asian Nations ( ASEAN )
ASEAN was established in 1967 by Indonesia, Malaysia, Philippines, Singapore, and
Thailand. Current members include also Brunei and Darussalam (joined in 1984),
Vietnam (joined in 1995), Laos and Myanmar (joined in 1997), and Cambodia
(joined in 1999). The aims and purposes of ASEAN involve cooperation in the
economic, social, cultural, technical, educational, and other fi elds [54] . Among other
cooperation schemes the ASEAN countries have also developed their own GMP
guidelines, which were issued in 1984 [55] .
TABLE 15 Contents of ICH GMP Guideline for API Production [52]
Section Subject
1 Introduction
2 Quality management
3 Personnel
4 Buildings and facilities
5 Process equipment
6 Documentation and records
7 Materials management
8 Production and in - process controls
9 Packaging and identifi cation labeling of APIs and intermediates
10 Storage and distribution
11 Laboratory controls
12 Validation
13 Change control
14 Rejection and reuse of materials
15 Complaints and recalls
16 Contract manufacturers (including laboratories)
17 Agents, brokers, traders, distributors, repackers, and relabelers
18 Specifi c guidance for APIs manufactured by cell culture/fermentation
19 APIs for use in clinical trials
20 Glossary
INTERNATIONAL GMP GUIDES AND HARMONIZATION 135
136 CORRESPONDENCES AND DIFFERENCES
2.1.3.5 Mercado Comun del Sur ( MERCOSUR )
MERCOSUR was established in 1991 by Argentina, Brazil, Paraguay, and Uruguay
to develop a common market between its member countries. Current members
include also Bolivia and Chile (joined in 1996). One of the original aims was to
harmonize the pharmaceutical legislation of the member countries. As a part of
these harmonization activities MERCOSUR has developed its own GMP guidelines,
which are based on WHO recommendations. In addition to the GMP guideline,
MERCOSUR has also issued other GMP - related guides covering inspections,
requirements for facilities, and quality control [56] .
2.1.4 CORRESPONDENCES OF THE U . S . GMP REGULATIONS WITH
GMP CODES AND GUIDELINES
The following sections deal with the correspondences and differences between the
U.S. GMP regulations and the Canadian and EU GMP codes and the WHO GMP
guideline. As the EU GMP code is harmonized with the PIC/S GMP guide, the
correspondences between the EU GMP code and the U.S. GMP regulations cover
also the correspondences between the U.S. GMP regulations and the PIC/S GMP
guide as well as all other national GMPs that are based on the PIC/S GMP guide.
Differences between the EU GMP code and the PIC/S GMP guide have been presented
in Section 2.1.3.2 .
2.1.4.1 General Issues
In the U.S. GMP regulations general issues related to the use and applicability of
GMP regulations are presented in Part 210 [6] , which consists of regulations 210.1,
210.2, and 210.3 and in Subpart A of Part 211 [7] , which consists of regulations 211.1
and 211.3. Contents of Part 210 and Subpart A of Part 211 are presented in Table
16 . Regulation 210.1 defi nes the status, 210.2 deals with the applicability, and 211.1
states the scope of the regulations. Defi nitions of terms used in the regulations are
provided in regulation 210.3 and in regulation 211.3, which states that the defi nitions
provided in regulation 210.3 apply also in Part 211.
Correspondences in Canadian GMP Code In the Canadian legislation general
issues related to the use and applicability of the GMP regulations and code are
covered in the introduction of the GMP code [12] and in Divisions 1A [57] and 2
TABLE 16 Contents of Part 210 and Subpart A of Part 211 of US GMP Regulations
Covering General Issues Related to Use and Applicability of Regulations [6, 7]
Section Subject
CFR 210.1 Status of current good manufacturing practice regulations
CFR 210.2 Applicability of current good manufacturing practice regulations
CFR 210.3 Defi nitions
CFR 211.1 Scope
CFR 211.3 Defi nitions
of the Part C of the Food and Drug Regulations [11] . Defi nitions for the GMP regulations
are covered in regulation C.01A.001 of Division 1A [57] and in regulation
C.02.002 of Division 2 [11] . Defi nitions for the GMP code are covered in the glossary
of terms of the code [12] .
Correspondences in EU GMP Code In the EU legislation general issues related
to the use and applicability of the GMP regulations and the code are covered in the
Commission Directive 2003/94/EC [14] and in the introduction of the GMP code
[15] . Defi nitions for the directive are covered in Article 2 of the directive [14] and
defi nitions for the code in the glossary of the GMP code [15] .
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
general issues related to the use and applicability of the GMP guide are covered in
section “ General Considerations. ” Defi nitions for the GMP guide are covered in the
glossary of the guideline.
2.1.4.2 Organization and Personnel
For GMP regulations in the United States issues related to organization and personnel
are covered in Subpart B [7] , which consists of regulations 211.22, 211.25, 211.28,
and 211.34. The contents of Subpart B is presented in Table 17 . Regulation 211.22
states the responsibilities and authorities of the quality control unit, including
requirements for the resources. Regulation 211.25 deals with personnel qualifi cations
covering the requirements for their education and experience and it also states
the requirements for the training of the personnel. Regulation 211.28 states the
responsibilities of personnel covering the requirements for the clothing and other
protective apparel, personal sanitation and health habits, as well as personal health
conditions. Furthermore, it states the requirements for the authorization for limited
access. Regulation 211.34 deals with consultants and lays down the requirements
for their education, training, and experience, including the requirements for
documentation.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to organization and personnel are mainly covered in the interpretation of
regulation C.02.006 (Personnel) and partly in the interpretations of regulations
C.02.004 (Premises), C.02.008 (Sanitation), C.02.011 (Manufacturing Control),
C.02.013 (Quality Control Department), C.02.015 (Quality Control Department),
and C.02.024 (Records). Correspondences to regulation 211.22 are covered in
TABLE 17 Contents of Subpart B of Part 211 of U. S .
GMP Regulations Covering Organization and Personnel [7]
Section Subject
CFR 211.22 Responsibilities of quality control unit
CFR 211.25 Personnel qualifi cations
CFR 211.28 Personnel responsibilities
CFR 211.34 Consultants
CORRESPONDENCES 137
138 CORRESPONDENCES AND DIFFERENCES
Sections 1 – 5 of the interpretation of regulation C.02.015 and in Section 2 of the
interpretation of regulation C.02.013. Sections 1 – 5 of the interpretation of regulation
C.02.015 state the responsibilities of the quality control unit (quality control
department), and Section 2 of the interpretation of regulation C.02.013 covers the
requirements for resources. Correspondences to regulation 211.25 stating the
requirements for the education, training, and experience of the personnel are
covered in Sections 1 – 5 of the interpretation of regulation C.02.006. Correspondences
to regulation 211.28 are covered in Section 6.3 of the interpretation of regulation
C.02.004, Sections 1 – 2 of the interpretation of regulation C.02.008, Section 8
of the interpretation of regulation C.02.011, and Section 4 of the interpretation of
regulation C.02.013. Section 1 of the interpretation of regulation C.02.008 states the
health requirements and Section 2 the requirements for clothing, other protective
apparel, and personal hygiene. Section 6.3 of the interpretation of regulation
C.02.004, Section 8 of the interpretation of regulation C.02.011, and Section 4 of the
interpretation of regulation C.02.013 cover the requirements regarding limited
access. Correspondences to regulation 211.34 are covered in Section 6 of the interpretation
of regulation C.02.006 and in Subsection 1.3.2 of the interpretation of
regulation C.02.024. Section 6 of the interpretation of regulation C.02.006 states the
requirements for the education, training, and experience of consultants and contractors
and Subsection 1.2.3 of the interpretation of regulation C.02.024 the requirements
for documentation.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
organization and personnel are mainly covered in Chapter 2 (Personnel) and partly
in Chapters 3 (Premises and Equipment), 5 (Production), and 6 (Quality Control).
Correspondences to regulation 211.22 are covered in Subchapters 2.6, 2.7, 6.1, and
6.2. Subchapters 2.6 and 2.7 deal with the responsibilities of the head of the quality
control unit (quality control department) and 6.2 with the responsibilities of
the quality control unit as a whole. Requirements for resources are covered in
Subchapter 6.1. Correspondences to regulation 211.25 are covered in Subchapters
2.1, 2.4, and 2.8 – 2.12. Subchapters 2.1 and 2.4 deal with the requirements for personnel
and Subchapters 2.8 – 2.12 with the requirements for their training. Correspondences
to regulation 211.28 are covered in Subchapters 2.15, 2.16, 3.5, 3.21, 5.16, and
6.4. Subchapter 2.15 deals with requirements for personal health conditions and 2.16
with requirements for clothing and protection. Access limitations are covered in
Subchapters 3.5, 3.21, 5.16, and 6.4. In the EU GMP code there is no correspondence
to regulation 211.34, which covers the requirements for the use of consultants.
However, Chapter 7 of the code deals with the requirements for the contract services
in general.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to organization and personnel are mainly covered in Chapter 9 (Personnel)
and partly in Chapters 10 (Training), 11 (Personal Hygiene), 16 (Good
Practices in Production), and 17 (Good Practices in Quality Control). Correspondences
to regulation 211.22 are covered in Subchapters 9.8, 9.10, 17.3, and 17.4.
Subchapters 9.8 and 9.10 state the responsibilities of the head of the quality control
unit and Subchapter 17.4 the responsibilities of the quality control unit as a whole.
Subchapter 17.3 covers the requirements for resources. Correspondences to regula
tion 211.25 are covered in Subchapters 9.2, 9.4, 9.7, and 10.1 – 10.4. Subchapters 9.2,
9.4, and 9.7 state the requirements for the personnel covering their education and
experience and Subchapters 10.1 – 10.4 the requirements for the training. Correspondences
to regulation 211.28 are covered in Subchapters 11.1 – 11.8, 9.5, and 16.7.
Subchapters 11.1 – 11.5 state the requirements for the health conditions and personal
hygiene and Subchapters 11.6 – 11.8 the requirements for the clothing and other
protective apparel. Subchapters 9.5 and 16.7 cover the requirements for the limited
access. Correspondences to regulation 211.34 are covered in Subchapter 10.6, which
covers the requirements for the use of consultants.
2.1.4.3 Buildings and Facilities
In the United States GMP regulations on issues related to buildings and facilities
are covered in Subpart C [7] , which consists of regulations 211.42, 211.44, 211.46,
211.48, 211.50, 211.52, 211.56, and 211.58. Contents of Subpart C are presented in
Table 18 . Regulation 211.42 deals with design and construction features covering
the requirements for the size, construction, and location of buildings used in the
production. Furthermore, it states the requirements for the placement of equipment
as well as the fl ow of materials and products and specifi es operations, which have
to be performed in separate or defi ned areas to prevent contamination or mix - ups.
It also covers the special requirements for the facilities used in aseptic processing
and facilities used in the production of penicillin. Regulation 211.44 states the
requirements for lighting and 211.46 for ventilation, including the requirements for
controls and air - handling systems. Furthermore, it states the special requirements
for ventilation in the production of penicillin. Regulation 211.48 deals with plumbing
covering requirements for the plumbing system, drains, and the quality of potable
water. Regulation 211.50 deals with sewage, trash, and other refuse stating the
requirements for their disposal. Regulation 211.52 covers the requirements for
washing and toilet facilities. Regulation 211.56 deals with sanitation stating the
requirements for the conditions to be maintained in the manufacturing facilities. It
also states the requirements for handling of trash and organic waste. Furthermore,
it states the requirements for the written procedures for sanitation operations
and use of biocides, fumigating, cleaning, and sanitizing agents. It also states the
TABLE 18 Contents of Subpart C of Part 211 of U. S .
GMP Regulations Covering Buildings and Facilities [7]
Section Subject
CFR 211.42 Design and construction features
CFR 211.44 Lighting
CFR 211.46 Ventilation, air fi ltration, air heating and
cooling
CFR 211.48 Plumbing
CFR 211.50 Sewage and refuse
CFR 211.52 Washing and toilet facilities
CFR 211.56 Sanitation
CFR 211.58 Maintenance
CORRESPONDENCES 139
140 CORRESPONDENCES AND DIFFERENCES
requirements for the use of biocides and the scope of sanitation procedures. Regulation
211.58 states the requirements for the maintenance of the buildings used in the
production.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to buildings and facilities are mainly covered in the interpretation of regulation
C.02.004 (Premises) and partly in the interpretations of regulations C.02.005
(Equipment), C.02.007 (Sanitation), C.02.009 (Raw Material Testing), C.02.011
(Manufacturing Control), and C.02.029 (Sterile Products). Correspondences to regulation
211.42 are covered in Sections 1, 2, 2.3, 6, 6.2, and 6.4 of the interpretation
of regulation C.02.004, in Section 15 of the interpretation of regulation C.02.011,
and in section “ Premises ” of the interpretation of regulation C.02.029. Sections 1
and 2 of the interpretation of regulation C.02.004 cover the requirements for the
size, construction, and location of buildings used in the production. Section 6.2 of
the interpretation of regulation C.02.004 and Section 15 of the interpretation of
regulation C.02.011 state the requirements for the placement of equipment. Requirements
for the fl ow of materials and products are covered in Section 6 of the interpretation
of regulation C.02.004 and operations, which have to be performed in
separate or defi ned areas in Sections 2.3 and 6.4 of the interpretation of regulation
C.02.004. The requirements for the facilities used in aseptic processing are covered
in section “ Premises ” of the interpretation of regulation C.02.029 and the requirements
for the facilities used in the production of penicillin in Section 11.1 of the
interpretation of regulation C.02.004. Correspondences to regulation 211.44 stating
the requirements for the lighting are covered in Section 6.5 of the interpretation of
regulation C.02.004. Correspondences to regulation 211.46 are covered in Sections
3.6 and 4 of the interpretation of regulation C.02.004. Section 3.6 of the interpretation
of regulation C.02.004 states the requirements for the air - handling systems and
Section 4 the requirements for the control of temperature and humidity. The specifi c
requirements regarding the production of penicillin are covered in Section 11.1.
Correspondences to regulation 211.48 are covered in Sections 3.5 and 7 of the
interpretation of regulation C.02.004, Section 3.7 of the interpretation of regulation
C.02.005, Section 4 of the interpretation of regulation C.02.009, and section “ Water
Treatment Systems ” of the interpretation of regulation C.02.029. Section 7 of the
interpretation of regulation C.02.004 states the requirements for the utilities and
support systems, including supplies of purifi ed water. Section 3.7 of the interpretation
of regulation C.02.005 states the requirements for the operation of water puri-
fi cation, storage, and distribution equipment. Requirements for the quality of water
are covered in Section 4 of the interpretation of regulation C.02.009 and in section
“ Water Treatment Systems ” of the interpretation of regulation C.02.029. The requirements
for drains are covered in Section 3.5 of the interpretation of regulation
C.02.004. Correspondences to regulation 211.50 stating the requirements for the
handling of sewage and refuse are covered in Section 2.6 of the interpretation of
regulation C.02.007. Correspondences to regulation 211.52 stating the requirements
for washing and toilet facilities are covered in Section 5 of the interpretation of
regulation C.02.004. Correspondences to regulation 211.56 stating the requirements
for sanitation are covered in Sections 1 and 2 of the interpretation of regulation
C.02.007. The Canadian GMP code does not state any separate requirements for
the handling of organic waste. General requirements for the handling of waste
materials are covered in Section 2.6 of the interpretation of regulation C.02.007.
Correspondences to regulation 211.58 stating the requirements for the maintenance
of the premises are covered in Section 9 of the interpretation of regulation
C.02.004.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
buildings and facilities are mainly covered in Chapter 3 (Premises and Equipment)
and partly in Annex 1 (Manufacture of Sterile Medicinal Products). Correspondences
to regulation 211.42 are covered in the foreword of Chapter 3 and in Subchapters
3.6 – 3.8, 3.13, 3.22, 3.23, 3.26, and 3.33. The requirements for the size,
construction and location of buildings used in the production are covered in the
foreword of Chapter 3. Subchapter 3.8 states the requirements for the placement
of equipment and Subchapter 3.7 for the fl ow of materials and products. Operations,
which have to be performed in separate or defi ned areas, are specifi ed in Subchapters
3.6, 3.13, 3.22, 3.23, 3.26, and 3.33. Annex 1 covers the requirements for facilities
used in aseptic processing and Subchapter 3.6 the requirements for the facilities
used in the production of penicillin. Correspondences to regulation 211.44 are
covered in Subchapters 3.3 and 3.16, which state the requirements for lighting. Correspondences
to regulation 211.46 are covered in Subchapters 3.3 and 3.12, which
state the requirements for ventilation. The specifi c requirements for the production
of penicillin are covered in Subchapter 3.6. Correspondences to regulation 211.48
are covered in Subchapters 3.10 and 3.11 and in Subsections 35 and 44 of Annex 1.
Subchapter 3.10 states the requirements for the plumbing and Subchapter 3.11 the
requirements for drains. Section 35 of Annex 1 covers the requirements for water
treatment plants and distribution systems and Section 44 the requirements for the
monitoring of water sources and water treatment equipment. More guidance on the
quality of water is given in the EU guidance document Note for Guidance on Quality
of Water for Pharmaceutical Use [58] . The EU GMP code does not have correspondence
to regulation 211.50, which covers the requirements for the handling of
sewage and other refuse. Correspondences to regulation 211.52 are covered in Subchapter
3.31, which covers the requirements for the facilities for washing and toilet
purposes. Correspondences to regulation 211.56 are covered in Subchapters 3.2, 3.4,
3.43, and 4.26. Subchapters 3.2 and 3.4 cover the requirements for the conditions to
be maintained in the manufacturing facilities. Subchapter 4.26 covers the procedures
for cleaning and sanitization and Subchapter 3.43 the requirements for the sanitization
of water pipes. The EU GMP code does not cover any separate requirements
for the handling of organic waste. Correspondences to regulation 211.58 are covered
in Subsection 3.2, which covers the requirements for the maintenance of the buildings
used in the production.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to buildings and facilities are mainly covered in Chapter 12 (Premises)
and partly in Chapters 3 (Sanitation and Hygiene), 14 (Materials), and 15 (Documentation).
Correspondences to regulation 211.42 are covered in Subchapters 12.1,
12.2, 12.4, 12.5, 12.10, 12.14, 12.17, 12.19, 12.22 – 12.26, and 12.33. The requirements
for the size, construction, and location of buildings are stated in Subchapters 12.1,
12.4, and 12.5. Subchapters 12.2 and 12.26 cover the requirements for the placement
of equipment and Subchapters 12.10 and 12.25 the requirements for the fl ow of
CORRESPONDENCES 141
142 CORRESPONDENCES AND DIFFERENCES
materials and products. Operations, which have to be performed in separate or
defi ned areas, are specifi ed in Subchapters 12.14, 12.17, 12.19, 12.22 – 12.24, and 12.33.
The requirements for the facilities used in aseptic processing are covered in Chapter
9 of Annex 6 of the WHO TRS 902 [40] and the requirements for the facilities used
in the manufacture of penicillin are in Subchapter 12.24. Correspondences to regulation
211.44 are covered in Subchapters 12.8 and 12.32, which state the requirements
for lighting. Correspondences to regulation 211.46 are covered in Subchapters 12.8
and 12.30, which state the requirements for ventilation. The specifi c requirements
for the production of penicillin are covered in Subchapter 12.24. Correspondences
to regulation 211.48 are covered in Subchapters 12.28, 12.29, and 14.6 and in Annex
3 of the WHO TRS 929 [45] . Subchapter 12.28 states the requirements for the
plumbing and Subchapter 14.6 for the quality of water used in the production of
drug products. More guidance on the quality of water is given in Annex 3 of the
WHO TRS 929 [45] . The requirements for the drains are stated in Subchapter 12.29.
Correspondences to regulation 211.50 are covered in Subchapters 14.44 and 14.45,
which state the requirements for the handling of sewage and other refuse. Correspondences
to regulation 211.52 are covered in Subchapter 12.12, which states the
requirements for the facilities for washing and toilet purposes. Correspondences to
regulation 211.56 are covered in Subchapters 3.1, 12.7, 12.9, 14.44 – 14.46, and 15.48.
Subchapters 12.7 and 12.9 state the requirements for the conditions to be maintained
in the manufacturing facilities and Subchapter 3.1 the general requirements
for sanitation and hygiene. In the WHO GMP guideline there is no separate guidance
on the handling of organic waste. General requirements for the handling of
waste materials are stated in Subchapters 14.44 and 14.45. Subchapter 15.48 states
the requirements for the written procedures for sanitation operations and Subchapter
14.46 for the use of rodenticides, insecticides, fumigating agents, and sanitizing
materials. Correspondences to regulation 211.58 are covered in Subchapter 12.6,
which states the requirements for the maintenance of the buildings used in drug
production.
2.1.4.4 Equipment
For GMP regulations in the United States issues related to equipment are covered
in Subpart D [7] , which consists of regulations 211.63, 211.65, 211.67, 211.68, and
211.72. Contents of Subpart D are presented in Table 19 . Regulation 211.63 states
the requirements for the production equipment covering design, size, and location.
Regulation 211.65 states the requirements for the construction of equipment cover-
TABLE 19 Contents of Subpart D of Part 211 of U. S .
GMP Regulations Covering Equipment [7]
Section Subject
CFR 211.63 Equipment design, size, and location
CFR 211.65 Equipment construction
CFR 211.67 Equipment cleaning and maintenance
CFR 211.68 Automatic, mechanical, and electronic
equipment
CFR 211.72 Filters
ing the characteristics of used materials and special requirements for the structure
of the equipment. Regulation 211.67 deals with cleaning, maintenance, and sanitizing
of equipment and utensils covering the requirements for the procedures for
cleaning and maintenance operations. Regulation 211.68 deals with automatic,
mechanical, and electronic equipment covering requirements for their calibration
and inspection, including the requirements for the documentation of checks and
inspections. Furthermore, it covers the requirements for the controls for computer
or related systems, including the requirements for the maintenance of backup data.
Regulation 211.72 covers the requirements for the fi lters for liquid fi ltration used
in the manufacture of injectable products, including the specifi c requirements for
the use of fi ber - releasing and asbestos - containing fi lters.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to equipment are mainly covered in the interpretation of regulation C.02.005
(Equipment) and partly in the interpretation of regulation C.02.007 (Sanitation)
and C.02.024 (Records). Correspondences to regulation 211.63 stating the requirements
for the design, construction, and location of equipment used in the manufacture
of drug products are covered in Sections 1 and 5 of the interpretation of
regulation C.02.005. Correspondences to regulation 211.65 stating the requirements
for the construction of equipment are covered in Sections 2.1 – 2.3 of the interpretation
of regulation C.02.005. Correspondences to regulation 211.67 stating the
requirements for the sanitation are covered in Sections 1, 2, and 3 of the interpretation
of regulation C.02.007. Correspondences to regulation 211.68 are covered in
Section 5.4 of the interpretation of regulation C.02.005 and in the foreword of the
interpretation of regulation C.02.024. Section 5.4 of the interpretation of regulation
C.02.005 states the requirements for the use of automatic, mechanical, and electronic
equipment, including computerized systems, and the foreword of the interpretation
of regulation C.02.024 the requirements for the maintenance of backup data. The
Canadian GMP code does not have correspondence to regulation 211.72, which
states the requirements for the fi lters for liquid fi ltration used in the manufacture
of injectable products. Nor does it cover requirements for the use of fi ber - releasing
or asbestos - containing fi lters.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
equipment are mainly covered in Chapter 3 (Premises and Equipment) and partly
in Chapter 4 (Documentation) and Annexes 1 (Manufacture of Sterile Medicinal
Products) and 11 (Computerised Systems). Correspondences to regulation 211.63
are covered in Subchapter 3.34, which states the requirements for the design and
location of equipment used in the manufacture of drug products. Correspondences
to regulation 211.65 are covered in Subchapters 3.38 and 3.39, which state the
requirements for the construction of equipment. Correspondences to regulation
211.67 are covered in Subchapters 3.36, 3.37, and 3.43, which cover the requirements
for cleaning and sanitizing the manufacturing equipment. Correspondences to regulation
211.68 are covered in Subchapters 3.41 and 4.9 and Annex 11. Subchapter
3.41 states the requirements for the maintenance of measuring, weighing, recording,
and control equipment and Subchapter 4.9 the requirements for the use of electronic
data processing systems and the maintenance of backup data. Additional
guidance on the use of computerized systems is given in Annex 11. Correspondences
CORRESPONDENCES 143
144 CORRESPONDENCES AND DIFFERENCES
to regulation 211.72 stating the requirements for fi lters for liquid fi ltration used in
the sterile fi ltration are covered in Sections 84 – 87 of Annex 1. The EU GMP code
does not have any separate guidance for the use of fi ber - releasing or asbestos -
containing fi lters.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to equipment are mainly covered in Chapter 13 (Equipment) and
partly in Chapters 14 (Materials), 15 (Documentation), and 16 (Good Practices in
Production). Correspondences to regulation 211.63 are covered in Subchapters 13.1
and 13.2, which state the requirements for the design, location, and installation of
equipment used in the manufacture of drug products. Correspondences to regulation
211.65 are covered in Subchapters 13.9 and 14.3, which state the requirements
for the construction of equipment. Correspondences to regulation 211.67 are covered
in Subchapters 13.6, 13.8, 13.12, 16.17, 16.18, and 16.22, which state the requirements
for cleaning and sanitizing the equipment. Correspondences to regulation 211.68
are covered in Subchapters 16.23 and 15.9. The requirements for the maintenance
of measuring, weighing, recording, and control equipment and instruments
are covered in Subchapter 16.23. Subchapter 15.9 states the requirements for the
use of electronic data - processing systems, including the requirements for the maintenance
of backup data. Correspondences to regulation 211.72 stating the requirements
for the use of fi lters are covered in Subchapters 7.6 – 7.9 of Annex 6 of the
WHO TRS 902 [40] . Subchapter 7.6 covers the requirements for asbestos - containing
fi lters.
2.1.4.5 Control of Components and Drug Product Containers and Closures
In the United States issues related to control of components and drug product
containers and closures are covered in Subpart E [7] , which consists of regulations
211.80, 211.82, 211.84, 211.86, 211.87, 211.89, and 211.94. Contents of Subpart E are
presented in Table 20 . Regulation 211.80 defi nes the requirements for the procedures
for the control of components, containers, and closures. It also states the
requirements for their handling, storing, and identifi cation. Regulation 211.82 covers
the requirements for receipt and storage of untested components, containers, and
TABLE 20 Contents of Subpart E of Part 211 of U . S . GMP Regulations Covering
Control of Components and Drug Product Containers and Closures [7]
Section Subject
CFR 211.80 General requirements
CFR 211.82 Receipt and storage of untested components, drug product containers,
and closures
CFR 211.84 Testing and approval or rejection of components, drug product
containers, and closures
CFR 211.86 Use of approved components, drug product containers, and closures
CFR 211.87 Retesting of approved components, drug product containers, and
closures
CFR 211.89 Rejected components, drug product containers, and closures
CFR 211.94 Drug product containers and closures
closures. Regulation 211.84 deals with testing and approval or rejection of components,
containers, and closures covering the requirements for sampling, testing, and
release. Regulation 211.86 deals with the use of approved components, containers,
and closures stating the requirements for the rotation of the storage. Regulation
211.87 states the requirements for the retesting of approved components, containers,
and closures. Regulation 211.89 covers the requirements for the handling of rejected
components, containers, and closures. Regulation 211.94 deals with drug product
containers and closures covering the requirements for materials and the cleanliness
of containers and closures. Furthermore, it states the requirements for container
closure systems, standards and methods.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to control of components and drug product containers and closures are
covered in interpretations of regulations C.02.009 (Raw Material Testing), C.02.010
(Raw Material Testing), C.02.011 (Manufacturing Control), C.02.014 (Quality
Control Department), C.02.016 (Packaging Material Testing), and C.02.017 (Packaging
Material Testing). Correspondences to regulation 211.80 stating the general
requirements for the handling, storing, and identifi cation of components (raw materials)
and drug product containers and closures (packaging materials) are covered
in Sections 1, 20, and 21 of the interpretation of regulation C.02.011. Correspondences
to regulation 211.82 stating the requirements for the receipt, testing, and
storage of untested components and drug product containers and closures are
covered in Sections 16, 18, and 19 of the interpretation of regulation C.02.011. Correspondences
to regulation 211.84 stating the requirements for testing and approval
of components and drug product containers and closures are covered in Sections 6
and 7 of the interpretation of regulation C.02.009, Sections 1 – 8 of the interpretation
of regulation C.02.010, Sections 1 and 2 of regulation C.02.016, Section 4 of its
interpretation, and Section 1 of the interpretation of regulation C.02.017. Interpretations
6 and 7 of regulation C.02.009 and interpretations 1 – 8 of regulation C.02.010
cover the requirements for components. Sections 1 and 2 and interpretation 4 of
regulation C.02.016 and interpretation 1 of regulation C.02.017 state the requirements
for drug product containers and closures. In the Canadian GMP code there
is no correspondence to regulation 211.86, which covers the requirements for the
rotation of the storage. Correspondences to regulation 211.87 stating the requirements
for the retesting of approved components are covered in Sections 8 – 10 of the
interpretation of regulation C.02.009. For the retesting of drug product containers
and closures the Canadian GMP code has no guidance. Correspondences to regulation
211.89 stating the requirements for the handling of rejected components and
drug product containers and closures are covered in Section 14 of the interpretation
of regulation C.02.011 and in Section 5 of the interpretation of regulation C.02.014.
The Canadian GMP code does not have correspondence to regulation 211.94, which
covers the requirements for containers and closure systems.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
control of components and drug product containers and closures are mainly covered
in Chapter 5 (Production) and partly in Chapter 6 (Quality Control). Correspondences
to regulation 211.80 are covered in Subchapters 5.2, 5.7, 5.10, 5.29, and
5.40 – 5.42, which cover the requirements for the handling, storing, and identifi cation
CORRESPONDENCES 145
146 CORRESPONDENCES AND DIFFERENCES
of components (starting materials) and drug product containers and closures
(primary packaging materials). Correspondences to regulation 211.82 are covered
in Subchapters 5.5, 5.27, and 5.40, which state the requirements for the receipt,
testing, and storage of untested components and drug product containers and closures.
Correspondences to regulation 211.84 are covered in Subchapters 5.31, 5.40
and 6.11 – 6.22 and Annex 8. The general requirements for sampling and testing are
covered in Subchapters 6.11 – 6.22. More guidance on sampling is given in Annex 8.
The requirements for the approved use of components and drug product containers
and closures are stated in Subchapters 5.31 and 5.40. Correspondences to regulation
211.86 are covered in Subchapter 5.7, which states the requirements for the storage
conditions and rotation. Correspondences to regulation 211.87 are covered in
Subchapters 5.29 and 5.40, which deal with the retesting of components and drug
product containers and closures. Correspondences to regulation 211.89 are covered
in Subchapter 5.61, which states the requirements for the handling of rejected components
and drug product containers and closures. Correspondences to regulation
211.94 are covered in Subchapter 5.48, which states the requirements for drug
product containers and closures.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to control of components and drug product containers and closures
are covered in Chapters 14 (Materials), 16 (Good Practices in Production), and 17
(Good Practices in Quality Control). Correspondences to regulation 211.80 are
covered in Subchapters, 14.5, 14.13, 14.14, 14.19 – 14.21, and 16.2, which state the
requirements for the handling, storing, and identifi cation of components (starting
materials) and drug product containers and closures (primary packaging materials).
Correspondences to regulation 211.82 are covered in Subchapters 14.4, 14.9 – 14.11,
and 14.19, which state the requirements for receipt, testing, identifi cation, and
storage of untested components and drug product containers and closures. Correspondences
to regulation 211.84 are covered in Subchapters 14.12, 14.15, and 17.7 –
17.17. The requirements for sampling and testing are covered in Subchapters
17.7 – 17.17 and 14.12. Subchapter 14.15 states the requirements for the approved
use of components and drug product containers and closures. Correspondences to
regulation 211.86 are covered in Subchapter 14.5, which states the requirements for
the storage conditions and the rotation of the storage. Correspondences to regulation
211.87 are covered in Subchapter 14.13, which states the requirements for the
retesting of approved components. The WHO GMP guideline does not cover the
requirements for the retesting of drug product containers and closures. Correspondences
to regulation 211.89 are covered in Subchapter 14.28, which states the
requirements for the handling of rejected components and drug product containers
and closures. Correspondences to regulation 211.94 are covered in Subchapter 16.19,
which states the requirements for the drug product containers and closures.
2.1.4.6 Production and Process Controls
In the United States GMP regulations on issues related to production and process
controls are covered in Subpart F [7] , which consists of regulations 211.100, 211.101,
211.103, 211.105, 211.110, 211.111, 211.113, and 211.115. Contents of Subpart F are
presented in Table 21 . Regulation 211.100 states the requirements for procedures
regarding production and process controls, including the requirements for the documentation
and handling of deviations. Regulation 211.101 deals with the requirements
for the charge - in of components. Regulation 211.103 states the requirements
for the determination of yields. Regulation 211.105 covers requirements for the
identifi cation of processing equipment such as containers, processing lines, and
major equipment used during manufacture. Regulation 211.110 states the requirements
for in - process controls, including the testing and approval of in - process
materials and handling of rejected in - process materials. Regulation 211.111 covers
the requirements for the time limitations on production, including the handling of
deviations from established limits. Regulation 211.113 covers the control of microbiological
contaminations. Regulation 211.115 states the requirements for the reprocessing
of batches that do not conform to standards or specifi cations.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to production and process controls are mainly covered in the interpretation
of regulation C.02.011 (Manufacturing Control) and partly in the interpretations of
regulations C.02.005 (Equipment), C.02.014 (Quality Control Department), and
C.02.029 (Sterile Products). Correspondences to regulation 211.100 are covered in
Sections 1 – 5 of the interpretation of regulation C.02.011. Interpretations 1 – 4 state
the requirements for manufacturing processes and interpretation 5 for the handling
of deviations. Correspondences to regulation 211.101 stating the requirements for
charge - in of components are covered in Section 22 of the interpretation of regulation
C.02.011. Correspondences to regulation 211.103 stating the requirements for
the determination of yields including the handling of deviations from the expected
yield are covered in Sections 6 and 7 of the interpretation of regulation C.02.011.
Correspondences to regulation 211.105 stating the requirements for the identifi cation
of piping, containers, equipment, and rooms used in the manufacturing of drug
products are covered in Section 3.5 of the interpretation of regulation C.02.005 and
in Section 13 of the interpretation of regulation C.02.011. Correspondences to regulation
211.110 are covered in Sections 11 and 14 of the interpretation of regulation
C.02.011 and in Section 5 of the interpretation of regulation C.02.014. Section 11 of
the interpretation of regulation C.02.011 states the requirements for the in - process
controls. The Canadian GMP code does not cover requirements for the testing of
in - process materials. The handling of rejected materials is covered in Section 14
of the interpretation of regulation C.02.011 and in Section 5 of the interpretation
TABLE 21 Contents of Subpart F of Part 211 of U . S . GMP Regulations Covering
Production and Process Controls [7]
Section Subject
CFR 211.100 Written procedures, deviations
CFR 211.101 Charge - in of components
CFR 211.103 Calculation of yield
CFR 211.105 Equipment identifi cation
CFR 211.110 Sampling and testing of in - process materials and drug products
CFR 211.111 Time limitations on production
CFR 211.113 Control of microbiological contamination
CFR 211.115 Reprocessing
CORRESPONDENCES 147
148 CORRESPONDENCES AND DIFFERENCES
of regulation C.02.014. Correspondences to regulation 211.111 dealing with the
requirements for the time limitations on production are covered in Section 24.7 of
the interpretation of regulation C.02.011. Correspondences to regulation 211.113
are covered in the interpretation of regulation C.02.029, which deals with the manufacture
of sterile products. Correspondences to regulation 211.115 stating the
requirements for the reprocessing of batches that do not conform to specifi cations
are covered in Sections 7 – 9 of the interpretation of regulation C.02.014.
Correspondences in EU GMP In the EU GMP code [15] issues related to production
and process controls are mainly covered in Chapter 5 (Production) and partly
in Chapters 3 (Premises and Equipment), 4 (Documentation), and 6 (Quality
Control). Correspondences to regulation 211.100 are covered in Subchapters 5.2,
5.15, and 5.22 – 5.24. The requirements for manufacturing processes are covered in
Subchapters 5.2 and 5.22 – 5.24. Subchapter 5.15 states the requirements for handling
of deviations from instructions or procedures. Correspondences to regulation
211.101 are covered in Subchapters 5.28 – 5.34, which state the requirements for the
charge - in of components. Correspondences to regulation 211.103 are covered in
Subchapters 5.8 and 5.39, which state the requirements for determination of yields,
including the handling of deviations from the expected yield. Correspondences to
regulation 211.105 are covered in Subchapters 3.42 and 5.12, which state the requirements
for identifi cation of piping, containers, equipment, and rooms used in the
manufacture of drug products. Correspondences to regulation 211.110 are covered
in Subchapters 3.17, 4.10, 4.12, 5.38, 5.61, and 6.18. The requirements for in - process
controls are covered in Subchapters 3.17, 5.38, and 6.18. Subchapters 4.10 and 4.12
state the requirements for the specifi cations for in - process materials (intermediate
products) and Subchapter 5.61 for handling of rejected materials. The EU GMP
code does not cover separate guidance on testing and approval of in - process materials.
General guidance on sampling and testing is given in Subchapters 6.11 – 6.22.
Correspondences to regulation 211.111, which deals with the time limitations on
production, are covered in Chapter 4.15. Correspondences to regulation 211.113 are
covered in Subchapter 5.10 and in Annex 1, which cover the requirements for the
control of microbiological contaminations. Correspondences to regulation 211.115
are covered in Subchapters 5.62 and 5.64, which state the requirements for the
reprocessing of rejected batches.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to production and process controls are covered in Chapters 13 (Equipment),
14 (Materials), 15 (Documentation), 16 (Good Practices in Production), and
17 (Good Practices in Quality Control). Correspondences to regulation 211.100 are
covered in Subchapters 16.1 – 16.3. Subchapters 16.1 and 16.2 state the requirements
for the manufacturing operations and Subchapter 16.3 for the handling of deviations
from instructions or procedures. Correspondences to regulation 211.101 are covered
in Subchapters 14.12 – 14.18, which state the requirements for the charge - in of components.
Correspondences to regulation 211.103 are covered in Subchapters 16.4
and 16.20, which state the requirements for the determination of yields, including
the handling of deviations from the expected yield. Correspondences to regulation
211.105 are covered in Subchapters 13.3, 13.4, and 16.6, which state the requirements
for the identifi cation of piping, containers, equipment, and rooms used during pro
duction. Correspondences to regulation 211.110 are covered in Subchapters 14.28,
15.20, 16.9, 16.16, and 17.8. Subchapters 16.9, 16.16, and 17.8 cover the requirements
for the in - process controls and Subchapter 15.20 for the specifi cations for in - process
materials (intermediate products). The requirements for the handling of rejected
materials are stated in Subchapter 14.28. Correspondences to regulation 211.111,
which deals with the time limitations on production, are covered in Chapter 15.23.
Correspondences to regulation 211.113 are covered in Subchapters 16.10 – 16.14 and
in Annex 6 of the WHO TRS 902 [40] . Subchapters 16.10 – 16.14 cover general
requirements for the prevention of cross - contamination and bacterial contamination
during production and Annex 6 general requirements for the manufacture of
sterile drug products. Correspondences to regulation 211.115 are covered in Subchapters
14.29, 14.31, and 15.40, which state the requirements for the reprocessing
of rejected batches.
2.1.4.7 Packaging and Labeling Control
For GMP regulations in the United States issues related to packaging and labeling
control are covered in Subpart G [7] , which consists of regulations 211.122, 211.125,
211.130, 211.132, 211.134, and 211.137. The contents of Subpart G is presented in
Table 22 . Regulation 211.122 deals with materials examination and usage criteria
covering the requirements for the receipt, identifi cation, storage, handling, sampling,
testing, and approval of labeling and packaging materials, including documentation.
Furthermore, it covers the requirements for the control of labeling, handling of
obsolete and outdated labeling and packaging materials, and special requirements
for different labeling methods. Regulation 211.125 states the requirements for the
labeling issuance covering the testing of labeling materials, the control of discrepancy
between the quantities of labeling issued, used, and returned, and the handling
of excess and returned labeling. Regulation 211.130 states the requirements for the
packaging and labeling operations covering the written procedures. Regulation
211.132 states the requirements for the tamper - evident packaging. Regulation
211.134 states the requirements for the inspections of packaged and labeled
products covering sampling, examination, and documentation. Regulation 211.137
states the requirements for the expiration dates, including exemptions from the
requirements.
TABLE 22 Contents of Subpart G of Part 211 of U . S . GMP Regulations Covering
Packaging and Labeling Control [7]
Section Subject
CFR 211.122 Materials examination and usage criteria
CFR 211.125 Labeling issuance
CFR 211.130 Packaging and labeling operations
CFR 211.132 Tamper - evident packaging requirements for over - the - counter (OTC)
human drug products
CFR 211.134 Drug product inspection
CFR 211.137 Expiration dating
CORRESPONDENCES 149
150 CORRESPONDENCES AND DIFFERENCES
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to packaging and labeling control are covered in the interpretations of regulations
C.02.011 (Manufacturing Control), C.02.017 (Packaging Material Testing),
C.02.016 (Packaging Material Testing), C.02.019 (Finished Product Testing), and
C.02.027 (Stability). Correspondences to regulation 211.122 are covered in Sections
1, 16, 40, and 43 – 48 of the interpretation of regulation C.02.011, Sections 1, 8, and 9
of the interpretation of regulation C.02.017, and Sections 1 and 4 – 7 of the interpretation
of regulation C.02.016. Sections 1, 16, 43, and 48 of the interpretation of regulation
C.02.011 state the general requirements for the handling of packaging and
labeling materials covering receipt and storage. Section 8 of the interpretation of
regulation C.02.017 states the requirements for the identifi cation of the packaging
and labeling materials. The requirements for the testing of the packaging and labeling
materials are covered in Sections 1 and 9 of the interpretation of regulation
C.02.017. Sections 1 and 4 of the interpretation of regulation C.02.016 and Sections
6 and 7 of the interpretation of regulation C.02.017 state the requirements for the
approval of packaging and labeling materials. Sections 44 – 47 of the interpretation
of regulation C.02.011 cover requirements for the use of roll - fed labels, cut labels,
gang printing, and the monitoring of the performance of printing. The requirements
for the handling of obsolete and outdated packaging and labeling materials are
covered in Section 40 of the interpretation of regulation C.02.011 and in Section 5
of the interpretation of regulation C.02.016. Correspondences to regulation 211.125
are covered in Sections 39 and 42 of the interpretation of regulation C.02.011
and in Section 8 of the interpretation of regulation C.02.017. Section 8 of the interpretation
of regulation C.02.017 states the requirements for the examination of
packaging and labeling materials. The requirements for the control and handling of
discrepancy between the quantities of labeling issued, used, and returned are covered
in Section 42 and the requirements for the handling of unused batch - coded packaging
and labeling materials in Section 39 of the interpretation of regulation C.02.011.
Correspondences to regulation 211.130 stating the requirements for the packaging
and labeling operations are covered in Sections 29 – 38 of the interpretation of regulation
C.02.011. In Canadian GMP code there is no correspondence to regulation
211.132 stating the requirements for the tamper - evident packaging. Correspondences
to regulation 211.134 stating the requirements for the inspections of packaged
and labeled products are covered in Section 1 of the interpretation of regulation
C.02.019. Correspondences to regulation 211.137 stating the requirements for the
expiration dates are covered in regulation C.02.027 and in Section 1 of its
interpretation.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
packaging and labeling control are mainly covered in Chapter 5 (Production) and
partly in Chapters 4 (Documentation) and 6 (Quality Control). Correspondences
to regulation 211.122 are covered in Subchapters 4.11, 4.19, 4.21 – 4.23, 5.2, 5.40 – 5.43,
and 5.50 – 5.52. Subchapters 4.19, 4.21 – 4.23, 5.2, and 5.40 – 5.42 cover the requirements
for purchase, handling, control, storage, and identifi cation of packaging and labeling
materials. Specifi cations for packaging and labeling materials are stated in Subchapter
4.11 and the requirements for handling of outdated or obsolete packaging and
labeling materials in Subchapter 5.43. Subchapter 5.51 covers the requirements for
the use of cut - labels, off - line overprinting, and roll - feed labels. The requirements for
the control of the printing and labeling operations are stated in Subchapters 5.50
and 5.52. Correspondences to regulation 211.125 are covered in Subchapters 5.2,
5.56, and 5.57. Subchapter 5.2 states the general requirements for the handling of
packaging and labeling materials. The requirements for the control of discrepancy
between the quantities of labeling issued, used, and returned are covered in Subchapter
5.56 and the requirements for the handling of unused batch - coded packaging
and labeling materials in Subchapter 5.57. Correspondences to regulation 211.130
are covered in Subchapters 5.2 and 5.44 – 5.49. Subchapter 5.2 states the general
requirements for the handling of packaging and labeling materials and Subchapters
5.44 – 5.49 cover the requirements for the packaging and labeling operations. The
European Community GMP code does not have correspondence to regulation
211.132, which covers the requirements for the tamper - evident packaging. Correspondences
to regulation 211.134 are covered in Subchapters 5.54 and 6.3, which
state the requirements for the control of packaged and labeled products. The EU
GMP code does not have correspondence to regulation 211.137, which covers the
requirements for expiration dates.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to packaging and labeling control are covered in Chapters 6 (Product
Recalls), 12 (Premises), 14 (Materials), 15 (Documentation), 16 (Good Practices in
Production), and 17 (Good Practices in Quality Control). Correspondences to regulation
211.122 are covered in Subchapters 12.21, 14.19 – 14.23, 15.18, 16.2, 17.14, and
17.16. Subchapters 6.2, 12.21, 14.19 – 14.21, 14.23, and 17.16 state the requirements
for the purchase, handling, control, storage, and identifi cation of packaging and
labeling materials. The requirements for the approval of packaging and labeling
materials are covered in Subchapters 17.14 and 15.18. Subchapter 14.20 states the
requirements for the use of roll - feed and cut labels and Subchapter 14.22 the
requirements for the handling of outdated and obsolete packaging and labeling
materials. Correspondences to regulation 211.125 are covered in Subchapters 16.2,
16.34, and 16.35. Subchapter 16.2 states the general requirements for the handling
of packaging and labeling materials and Subchapter 16.34 the requirements for the
handling of discrepancy between the quantities of labeling issued, used, and returned.
The requirements for the handling of unused batch - coded packaging and labeling
materials are covered in Subchapter 16.35. Correspondences to regulation 211.130
are covered in Subchapters 16.25 – 16.30, which state the requirements for the packaging
and labeling operations. The WHO GMP guideline does not cover correspondence
to regulation 211.132, which covers the requirements for tamper - evident
packaging. Correspondences to regulation 211.134 are covered in Subchapter 16.32,
which states the requirements for the control of packaged and labeled products.
Correspondences to regulation 211.137 are covered in Subchapter 17.24, which
states the requirements for the determination of expiration dates and shelf - life
specifi cations.
2.1.4.8 Holding and Distribution
In the United States GMP regulations on issues related to holding and distribution
are covered in Subpart H [7] , which consists of regulations 211.142 and 211.150. The
contents of Subpart H is presented in Table 23 . Regulation 211.142 states the
CORRESPONDENCES 151
152 CORRESPONDENCES AND DIFFERENCES
requirements for the warehousing procedures covering quarantine and storage and
regulation 211.150 for the distribution procedures covering distribution order and
recalls.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to holding and distribution are covered in the interpretations of regulations
C.02.004 (Premises), C.02.011 (Manufacturing Control), C.02.012 (Manufacturing
Control), and C.02.019 (Finished Product Testing). Correspondences to regulation
211.142 stating the requirements for the quarantine and storage of products are
covered in Sections 1 and 49 of the interpretation of regulation C.02.011, Section
11.4 of the interpretation of regulation C.02.004, and Section 2 of the interpretation
of regulation C.02.019. Correspondences to regulation 211.150 stating the requirements
for distribution and recalls are covered in Section 1 of the interpretation of
regulation C.02.011 and Section 1 of the interpretation of regulation C.02.012.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
holding and distribution are covered in Chapters 4 (Documentation), 5 (Production),
and 8 (Complaints and Product Recall). Correspondences to regulation
211.142 are covered in Subchapters 5.2, 5.58, and 5.60, which state the requirements
for the storage and quarantine of products. Correspondences to regulation 211.150
are covered in Subchapters 4.25, 5.2, and 8.8 – 8.15, which state the requirements for
distribution and recalls.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to holding and distribution are covered in Chapters 6 (Product
Recalls), 14 (Materials), 15 (Documentation), and 16 (Good Practices in Production).
Correspondences to regulation 211.142 are covered in Subchapters 14.4, 14.26,
and 16.2, which state the requirements for the storage and quarantine of products.
Correspondences to regulation 211.150 are covered in Subchapters 6.1 – 6.8, 15.45,
and 16.2, which state the requirements for distribution and recalls.
2.1.4.9 Laboratory Controls
In the United States GMP regulations [7] issues related to laboratory controls are
covered in Subpart I, which consists of regulations 211.160, 211.165, 211.166, 211.167,
211.170, 211.173, and 211.176. The contents of Subpart I is presented in Table 24 .
Regulation 211.160 states the requirements for the establishment of laboratory
controls such as specifi cations, standards, sampling plans, and test procedures.
Furthermore, it covers the requirements stated for the calibration of instruments,
apparatus, gauges, and recording devices. Regulation 211.165 states the require-
TABLE 23 Contents of Subpart H of Part 211 of U . S .
GMP Regulations Covering Holding and Distribution [7]
Section Subject
CFR 211.142 Warehousing procedures
CFR 211.150 Distribution procedures
ments for the laboratory testing of batches prior to release covering the requirements
for sampling, testing, and approval. Furthermore, it states the requirements
for the handling of rejected drug products. Regulation 211.166 states the requirements
for stability testing, including the requirements for the determination of
expiration dates and the requirements for stability testing of homeopathic drug
products. Regulation 211.167 deals with special testing requirements covering sterile
products, ophthalmic ointments, and controlled - release dosage forms. Regulation
211.170 states the requirements for reserve samples covering identifi cation, quantity,
retention time, and storage. Furthermore it covers the requirements for the deterioration
investigations. Regulation 211.173 deals with laboratory animals covering the
requirements for their maintenance and control. Regulation 211.176 states the
requirements for the testing of penicillin contamination and the handling of penicillin
contaminated drug product.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to laboratory controls are covered in the interpretations of regulations
C.02.004 (Premises), C.02.009 (Raw Material Testing), C.02.011 (Manufacturing
Control), C.02.014 (Quality Control Department), C.02.015 (Quality Control
Department), C.02.016 (Packaging Material Testing), C.02.017 (Packaging Material
Testing), C.02.018 (Finished Product Testing), C.02.025 (Samples), C.02.026
(Samples), C.02.027 (Stability), and C.02.028 (Stability). Correspondences to regulation
211.160 stating the general requirements for laboratory controls are covered in
regulation C.02.009 and Sections 1 – 3 and 5 – 6 of its interpretation, regulation
C.02.016 and Sections 1 – 3 of its interpretation, Section 1 of the interpretation of
regulation C.02.017, regulation C.02.018 and Sections 1 – 5 of its interpretation, and
Section 6.4 of the interpretation of regulation C.02.015. Correspondences to regulation
211.165 stating the requirements for the release for distribution including the
testing of fi nished drug products and the handling of rejected drug products are
covered in Sections 7 and 14 of the interpretation of regulation C.02.011, Sections
2 and 5 of the interpretation of regulation C.02.014, Section 3 of the interpretation
of regulation C.02.015, and Section 2 of regulation C.02.018 and Sections 1 and 4 of
its interpretation. Correspondences to regulation 211.166 stating the requirements
for stability testing are covered in Section 1 of the interpretation of regulation
C.02.027 and Sections 1 and 2 of the interpretation of regulation C.02.028. The
Canadian GMP code does not cover separate requirements for the stability testing
TABLE 24 Contents of Subpart I of Part 211 of U . S . GMP
Regulations Covering Laboratory Controls [7]
Section Subject
CFR 211.160 General requirements
CFR 211.165 Testing and release for distribution
CFR 211.166 Stability testing
CFR 211.167 Special testing requirements
CFR 211.170 Reserve samples
CFR 211.173 Laboratory animals
CFR 211.176 Penicillin contamination
CORRESPONDENCES 153
154 CORRESPONDENCES AND DIFFERENCES
of homeopathic drug products. Correspondences to regulation 211.167 stating the
requirements for sterility testing are covered in Sections 1 – 4 of the interpretation
of regulation C.02.029 (Sterile Products). The Canadian GMP code does not have
any guidance covering the testing of ophthalmic ointments and controlled - release
dosage forms. Correspondences to regulation 211.170 stating the requirements for
reserve samples are covered in Section 1 of regulation C.02.025 and in regulation
C.02.026 and Sections 1 and 3 – 5 of their interpretation. Correspondences to regulation
211.173 stating the requirements for laboratory animals are covered in Section
2.4 of the interpretation of regulation C.02.004. The Canadian GMP code does not
have correspondence to regulation 211.176, which covers the requirements for the
testing and handling of penicillin contamination.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
laboratory controls are covered in Chapters 1 (Quality Management), 4 (Documentation),
5 (Production), and 6 (Quality Control) and in Annexes 1 (Manufacture of
Sterile Medicinal Products), 9 (Manufacture of Liquids, Creams, and Ointments),
and 19 (Reference and Retention Samples). Correspondences to regulation 211.160
are covered in Subchapters 1.4, 4.2, 4.3, 4.10 – 4.13, 5.15, 6.7, and 6.18, which cover
the general requirements for laboratory controls. Correspondences to regulation
211.165 are covered in Subchapters 4.22, 4.23, 5.61, 5.62, 6.3, 6.11, and 6.15, which
state the requirements for the release for distribution, the testing of fi nished drug
products, and the handling of rejected drug products. The EU GMP code does not
have correspondence to regulation 211.166, which states the requirements for stability
testing. However, there is a separate guideline, Stability Testing on Active Ingredients
and Finished Products [59] , which provides guidance on issues related to
stability testing. Furthermore, Subchapters 6.23 – 6.33 cover the requirements for the
on - going stability program. Correspondences to regulation 211.167 are covered in
Annexes 1 and 9. Section 93 of Annex 1 covers the requirements for sterility testing
and Annex 9 the requirements for ointments. In the EU GMP code there is no
guidance on the testing of the controlled - release dosage forms. Correspondences to
regulation 211.170 are covered in Subchapters 1.4 and 6.12 and Annex 19, which
state the requirements for reserve samples. Correspondences to regulation 211.173
are covered in Subchapters 3.33 and 6.22, which state the requirements for the
maintenance of animals. The EU GMP code does not have correspondence to regulation
211.176, which covers the requirements for the testing and handling of penicillin
contaminations.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to laboratory controls are covered in Chapters 14 (Materials), 15
(Documentation), 16 (Good Practices in Production), and 17 (Good Practices in
Quality Control). Correspondences to regulation 211.160 are covered in Subchapters
15.14 – 15.16, 15.18 – 15.21, 16.3, and 16.23, which state the general requirements
for laboratory controls. Correspondences to regulation 211.165 are covered in
Subchapters 14.28, 14.29, 15.13, 15.42, 17.7 – 17.13, 17.19, and 17.20, which state the
requirements for the release for distribution covering the testing of fi nished drug
products and the handling of rejected drug products. Correspondences to regulation
211.166 are covered in Subchapters 17.23 – 17.26, which state the requirements for
stability testing. The WHO GMP guideline does not cover separate requirements
for the stability testing of homeopathic drug products. Correspondences to regulation
211.167 are covered in Annex 6 of the WHO TRS 902 [40] , which states the
requirements for sterility testing. The WHO GMP guideline does not cover any
requirements for the testing of ophthalmic ointments or controlled - release dosage
forms. Correspondences to regulation 211.170 are covered in Subchapter 17.22,
which states the requirements for reserve samples. The WHO GMP guideline does
not have correspondence to regulation 211.173, which covers the requirements
for the maintenance of laboratory animals. Nor does it have correspondence to
regulation 211.176, which covers the requirements for the testing of penicillin
contaminations.
2.1.4.10 Records and Reports
In the United States issues related to records and reports are covered in Subpart J
[7] , which consists of regulations 211.180, 211.182, 211.184, 211.186, 211.188, 211.192,
211.194, 211.196, and 211.198. The contents of Subpart J is presented in Table 25 .
Regulation 211.180 states the general requirements for documentation covering
maintenance, retention times, and availability of the records. Furthermore, it states
the requirements for the annual quality standards evaluation. Regulation 211.182
states the requirements for individual equipment logs. Regulation 211.184 states the
requirements for component, drug product container, closure, and labeling records.
Regulation 211.186 states the requirements for master production and control
records. Regulation 211.188 states the requirements for batch production and control
records. Regulation 211.192 states the requirements for the review and approval of
production and control records, including the requirements for the investigation of
any unexplained discrepancies. Regulation 211.194 states the requirements for laboratory
records, including the requirements for the documentation of modifi cations.
Furthermore, it covers the requirements for the documentation of the testing and
standardization of reference standards, reagents, and standard solutions; calibration
of laboratory instruments and recording devices; and stability tests. Regulation
211.196 states the requirements for the distribution records. Regulation 211.198
states the requirements for the handling of complaints, including the maintenance
and retention times of complaint fi les.
TABLE 25 Contents of Subpart J of Part 211 of U . S . GMP Regulations Covering
Records and Reports [7]
Section Subject
CFR 211.180 General requirements
CFR 211.182 Equipment cleaning and use log
CFR 211.184 Component, drug product container, closure, and labeling records
CFR 211.186 Master production and control records
CFR 211.188 Batch production and control records
CFR 211.192 Production record review
CFR 211.194 Laboratory records
CFR 211.196 Distribution records
CFR 211.198 Complaint fi les
CORRESPONDENCES 155
156 CORRESPONDENCES AND DIFFERENCES
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to records and reports are mainly covered in regulations C.02.021, C.02.022,
C.02.023, and C.02.024 (Records) and in their interpretations and partly in the
interpretations of regulations C.02.005 (Equipment), C.02.010 (Raw Material
Testing), C.02.011 (Manufacturing Control), C.02.012 (Manufacturing Control),
C.02.014 (Quality Control Department), C.02.015 (Quality Control Department),
and C.02.017 (Packaging Material Testing). Correspondences to regulation 211.180
stating the general requirements for the maintenance of records, including periodic
quality evaluation (self - inspection) and the retention time of the records are covered
in regulations C.02.021, C.02.022, C.02.023, and C.02.024 and their interpretations
and in Section 2 of the interpretation of regulation C.02.012. Correspondences
to regulation 211.182 stating the requirements for individual equipment logs are
covered in Section 5.5 of the interpretation of regulation C.02.005. Correspondences
to regulation 211.184 stating the requirements for records to be kept on components,
drug product containers, closures, and labeling are covered in Sections 4 and 5 of
the interpretation of regulations C.02.020 – 24, Section 5 of the interpretation of
regulation C.02.010, and Section 7 of the interpretation of regulation C.02.017.
Correspondences to regulation 211.186 stating the requirements for the master
production and control records (manufacturing and packaging master formulas) are
covered in Sections 23 – 25 of the interpretation of regulation C.02.011 and Section
1.1 of the interpretation of regulations C.02.020 – 24. Correspondences to regulation
211.188 stating the requirements for the batch production and control records
(manufacturing and packaging batch document) are covered in Sections 26, 27, 29,
and 30 of the interpretation of regulation C.02.011 and Section 1.2 of the interpretation
of regulations C.02.020 – 24. Correspondences to regulation 211.192 stating the
requirements for review and approval of production and control records including
investigation of batch deviations are covered in Section 2 of the interpretation of
regulation C.02.014. Correspondences to regulation 211.194 stating the requirements
for laboratory records are covered in Sections 6.4, 6.6, and 6.7 of the interpretation
of regulation C.02.015. Correspondences to regulation 211.196 stating the
requirements for distribution records are covered in Section 1.6 of the interpretation
of regulation C.02.012 and Section 2.1 of the interpretation of regulations
C.02.020 – 24. Correspondences to regulation 211.198 stating the requirements for
the maintenance of complaint fi les including retention times are covered in Section
4 of the interpretation of regulation C.02.015, Section 3.1 of the interpretation of
regulations C.02.020 – 24, and regulation C.02.023.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
records and reports are mainly covered in Chapter 4 (Documentation) and partly
in Chapters 1 (Quality Management), 5 (Production), 6 (Quality Control), 8 (Complaints
and Product Recall), and 9 (Self Inspection). Correspondences to regulation
211.180 are covered in Subchapters 4.1 – 4.9, 6.8, and 9.1 – 9.3, which state the general
requirements for the maintenance of the records, including periodic quality evaluation
(self - inspection) and retention times. Correspondences to regulation 211.182
are covered in Subchapters 4.28 and 4.29, which state the requirements for individual
equipment logs. Correspondences to regulation 211.184 are covered in Subchapters
4.19 and 4.20, which state the requirements for the records to be kept on the
receipt of components, drug product containers, closures, and labeling. Correspon
dences to regulation 211.186 are covered in Subchapters 4.14 – 4.16, which state the
requirements for the master production and control records (manufacturing formula,
processing, and packaging instructions). Correspondences to regulation 211.188 are
covered in Subchapters 4.17 and 4.18, which state the requirements for the batch
production and control records (batch processing and packaging record). Correspondences
to regulation 211.192 are covered in Subchapters 1.4, 4.3, 4.24, 5.8, and
5.39, which state the requirements for review and approval of production and
control records, including the investigation of unexplained discrepancies. Correspondences
to regulation 211.194 are covered in Subchapters 3.41, 6.7, 6.17, 6.20,
and 6.21, which state the requirements for laboratory records. Correspondences to
regulation 211.196 are covered in Subchapter 4.25, which states the requirements
for distribution records. Correspondences to regulation 211.198 are covered in
Subchapters 4.26 and 8.1 – 8.8, which state the requirements for the handling of
complaints.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to records and reports are covered in Chapters 5 (Complaints),
8 (Self - Inspection and Quality Audits), 13 (Equipment), 14 (Materials), 15
(Documentation), 16 (Good Practices in Production), and 17 (Good Practices in
Quality Control). Correspondences to regulation 211.180 are covered in Subchapters
8.1 – 8.6 and 15.1 – 15.9, which state the general requirements for the maintenance
of the records, including periodic quality evaluation (self - inspection) and retention
times. Correspondences to regulation 211.182 are covered in Subchapters 15.46
and 15.47, which state the requirements for individual equipment logs. Correspondences
to regulation 211.184 are covered in Subchapters 15.32 and 15.33, which state
the requirements for the records to be kept on the receipt of components,
drug product containers, closures, and labeling. Correspondences to regulation
211.186 are covered in Subchapters 15.22 – 15.24, which state the requirements for
the master production and control records (master formula and packaging instructions).
Correspondences to regulation 211.188 are covered in Subchapters 15.25 –
15.30, which state the requirements for the batch production and control records
(batch processing and packaging records). Correspondences to regulation 211.192
are covered in Subchapters 16.4, 16.20, and 17.21, which state the requirements for
review and approval of production and control records covering also the requirements
for the investigation of unexplained discrepancies. Correspondences to regulation
211.194 are covered in Subchapters 13.5, 14.34, 14.35, 14.41, 15.12, 15.42, 15.43,
and 16.23, which states the requirements for laboratory records. Correspondences
to regulation 211.196 are covered in Subchapter 15.45, which states the requirements
for the distribution records. Correspondences to regulation 211.198 are
covered in Subchapters 5.1 – 5.10, which state the requirements for the handling of
complaints.
2.1.4.11 Returned and Salvaged Drug Products
In the United States GMP regulation [7] issues related to returned and salvaged
drug products are covered in Subpart K, which consists of regulations 211.204 and
211.208. The contents of Subpart K is presented in Table 26 . Regulation 211.204
states the requirements for the handling of returned drug products, including repro-
CORRESPONDENCES 157
158 CORRESPONDENCES AND DIFFERENCES
cessing and documentation. Regulation 211.208 states the requirements for drug
product salvaging.
Correspondences in Canadian GMP Code In the Canadian GMP code [12] issues
related to returned and salvaged drug products are covered in the interpretation of
regulation C.02.014 (Quality Control Department). Correspondences to regulation
211.204 stating the requirements for the handling of returned drug products are
covered in Section 4 of the interpretation of regulation C.02.014. The Canadian
GMP code does not have correspondence to regulation 211.208, which covers the
requirements for drug product salvaging.
Correspondences in EU GMP Code In the EU GMP code [15] issues related to
returned and salvaged drug products are covered in Chapters 4 (Documentation)
and 5 (Production). Correspondences to regulation 211.204 are covered in Subchapters
4.26 and 5.26, which state the requirements for the handling of returned drug
products. The EU GMP code does not have correspondence to regulation 211.208,
which covers the requirements for drug product salvaging.
Correspondences in WHO GMP Guideline In the WHO GMP guideline [36]
issues related to returned and salvaged drug products are covered in Chapter 14
(Materials). Correspondences to regulation 211.204 are covered in Subchapter
14.33, which states the requirements for the handling of returned drug products. The
WHO GMP guideline does not have correspondence to regulation 211.208, which
covers the requirements for drug product salvaging.
REFERENCES
1. Immel , B. K. ( 2001 ), A brief history of the GMPs for pharmaceuticals , Pharm. Technol.
No. Am. , 25 ( 7 ), 44 – 48 .
2. Anonymous Pharmaceutical Administration and Regulations in Japan , Japan Pharmaceutical
Manufacturers Association, available: http://www.jpma.or.jp/english/library/pdf/2005.
pdf .
3. Vesper , J. L. ( 2003 ), So what are GMPs, anyway? BioProcess Int. , 1 ( 2 ), 24 – 29.
4. Rosin , L. J. ( 2006 ), Regulatory affairs: If you didn ’ t write it down, it didn ’ t happen ,
BioProcess Int. , 4 ( 3 , Suppl), 16 – 23 .
5. Anonymous ( 2002 ), Sec. 351: Adulterated drugs and devices, in United States Code ,
Title 21, Chapter 9, Subchapter V, Part A, U.S. Government Printing Offi ce, Washington,
DC, available: http://frwebgate.access.gpo.gov/cgi - bin/getdoc.cgi?dbname=browse_
usc&docid=Cite:+21USC351 .
TABLE 26 Contents of Subpart K of Part 211 of U . S .
GMP Regulations Covering Returned and Salvaged Drug
Products [7]
Section Subject
CFR 211.204 Returned drug products
CFR 211.208 Drug product salvaging
6. Anonymous ( 2005 ), Part 210: Current good manufacturing practice in manufacturing,
processing, packing, or holding of drugs: General, in Code of Federal Regulations , Title
21, Chapter I, U.S. Government Printing Offi ce, Washington, DC, pp. 118 – 119, available:
http://www.access.gpo.gov/nara/cfr/waisidx_05/21cfr210_05.html .
7. Anonymous ( 2005 ), Part 211: Current good manufacturing practice for fi nished pharmaceuticals,
in Code of Federal Regulations , Title 21, Chapter I, U.S. Government
Printing Offi ce, Washington, DC, pp. 120 – 141, available: http://www.access.gpo.govnara/
cfr/waisidx_05/21cfr211_05.html .
8. Grazal , J. G. , and Earl , D. S. ( 1997 ), EU and FDA regulations: Overview and comparison ,
Qual. Assur. J. , 2 , 55 – 60 .
9. Anonymous ( 2006 ), Guidance page, available: http://www.fda.gov/cder/guidance/index.
htm .
10. Anonymous ( 2001 ), Guidance for industry: Q7A Good manufacturing practice guidance
for active pharmaceutical ingredients, Offi ce of Training and Communications, available:
http://www.fda.gov/cber/gdlns/ichactive.pdf .
11. Anonymous ( 2005 ), Division 2: Good manufacturing practices, in Consolidated
Statutes and Regulations , Food and Drugs Act, Food and Drug Regulations, Part C,
Department of Justice, Canada, available: http://laws.justice.gc.ca/en/f - 27/c.r.c. - c.870/
230079.html .
12. Anonymous ( 2002 ), Good manufacturing practices guidelines, Version 2, Health Products
and Food Branch Inspectorate, available: http://www.hc - sc.gc.ca/dhp - mps/alt_formats/
hpfb - dgpsa/pdf/compli - conform/2002v2_e.pdf .
13. Anonymous ( 2001 ), Directive 2001/83/EC of the European Parliament and of the Council ,
Off. J. Eur. Union , 44 ( L311 ), 67 – 128 , available: http://europa.eu.int/eur - lex/pri/en/oj/
dat/2001/l_311/l_31120011128en00670128.pdf .
14. Anonymous ( 2003 ), Commission directive 2003/94/EC , Off. J. Eur. Union ,
46 ( L262 ), 22 – 26 , available: http://europa.eu.int/eur - lex/pri/en/oj/dat/2003/l_262/
l_26220031014en00220026.pdf .
15. Anonymous ( 2005 ), EU guidelines to good manufacturing practice, in The Rules Governing
Medicinal Products in the European Union , Vol. 4, European Commission Enterprise
and Industry Directorate - General, available: http://pharmacos.eudra.org/F2/eudralex/
vol - 4/home.htm .
16. Anonymous ( 1998 ), Quality and biotechnology, in The Rules Governing Medicinal Products
in the European Union , Vol. 3A, European Commission Directorate General III,
available: http://pharmacos.eudra.org/F2/eudralex/vol - 3/home.htm .
17. Anonymous ( 2005 ), Regulations for Buildings and Facilities of Pharmacies , etc., MHLW
Ministerial Ordinance, Yakuji Nippon Ltd. , Tokyo , No.73.
18. Kim , Y. - O. , Ha , K. - W. , and Choi , K. - S. ( 2001 ), Safety evaluation for new drug approval in
Korea , Drug Info. J. , 35 ( 1 ), 285 – 291 .
19. Shin , S. - G. ( 1998 ), Current status of clinical trials in the Republic of Korea , Drug Info. J. ,
32 (Suppl), 1217S – 1222S .
20. Anonymous ( 2000 ), Competition in the Pharmaceutical Industry — Republic of Korea ,
Working Party No. 2 on Competition and Regulation, Committee on Competition Law
and Policy, OECD Publishing , Paris .
21. Anonymous ( 2001 ), Drug Administration Law of the People ’ s Republic of China , Order
of the President of the People ’ s Republic of China No. 45, available: http://www.sfda.gov.
cn/cmsweb/webportal/W45649037/A48335975.html .
22. Deng , R. , and Kaitin , K. I. ( 2004 ), The regulation and approval of new drugs in China ,
Drug Info. J. , 38 ( 1 ), 29 – 39 .
REFERENCES 159
160 CORRESPONDENCES AND DIFFERENCES
23. Anonymous ( 2005 ), The Drugs and Cosmetics Act and Rules, Ministry of Health
and Family Welfare, Department of Health, available: http://cdsco.nic.in/html/
Drugs&CosmeticAct.pdf .
24. Anonymous ( 2005 ), Schedule M: Good manufacturing practices and requirements of
premises, plant and equipment for pharmaceutical products, in The Drugs and Cosmetics
Act and Rules, Ministry of Health and Family Welfare, Department of Health, pp. 386 –
436, available: http://cdsco.nic.in/html/Drugs&CosmeticAct.pdf .
25. Venkateswarlu , M. ( 2006 ), Why do we need revision of schedule M, available: http://www.
pharmabiz.com , 2006.
26. Anonymous ( 2006 ), Section 36: Manufacturing principles, in Therapeutic Goods Act 1989 ,
Offi ce of Legislative Drafting and Publishing, pp. 96 – 97, available: http://www.tga.gov.au/
legis/index.htm#instruments .
27. Slater , T. ( 2002 ), Therapeutic goods (manufacturing principles) determination no 2 of
2002 , Commonwealth Austral. Gaz. , GN 34, 2306 – 2307 .
28. Anonymous ( 2002 ), Australian code of good manufacturing practice for medicinal
products, Therapeutic Goods Administration, available: http://www.tga.gov.au/docs/pdf/
gmpcodau.pdf .
29. Anonymous ( 2005 ), Medicines Act 1981, Parliamentary Counsel Offi ce, available: http://
www.legislation.govt.nz/browse_vw.asp?content - set=pal_statutes .
30. Anonymous ( 2001 ), Guidance notes for applicants for consent to distribute new and
changed medicines and related products, in New Zealand Regulatory Guidelines for
Medicines , Vol. 1, 5th ed., MedSafe, available: http://www.medsafe.govt.nz/downloads/
vol1.doc .
31. Anonymous ( 2005 ), New Zealand code of good manufacturing practice for manufacture
and distribution of therapeutic goods, available: http://www.medsafe.govt.nz/Regulatory/
Guideline/code.htm .
32. Anonymous ( 2002 ), Medicines and Related Substances Control Act 101 of 1965,
Medicines Control Council, available: http://www.mccza.com/showdocument.
asp?Cat=27&Desc=Acts%20and%20Regulations .
33. Anonymous ( 2003 ), General regulations made in terms of the Medicines and Related
Substances Act 1965 (Act no. 101 of 1965) as Amended, Government Notice, Department
of Health, available: http://www.mccza.com/showdocument.asp?Cat=27&Desc=
Acts%20and%20Regulations .
34. Anonymous ( 2005 ), Guide to Good Manufacturing Practice for Medicines in South
Africa, Medicines Control Council, available: http://www.mccza.com/showdocument.
asp?Cat=21&Desc=Guidelines%20 - %20Good%20Manufacturing%20Practices .
35. Anonymous ( 2002 ), WHO Expert Committee on Specifi cations for Pharmaceutical Preparations:
37th Report, WHO Technical Report Series 908, World Health Organization,
Singapore, available: http://whqlibdoc.who.int/trs/WHO_TRS_908.pdf .
36. Anonymous ( 2002 ), Annex 4: Good manufacturing practices for pharmaceutical products:
Main principles, in WHO Expert Committee on Specifi cations for Pharmaceutical
Preparations: 37th Report, WHO Technical Report Series 908, World Health Organization,
Singapore, pp. 36 – 89, available: http://whqlibdoc.who.int/trs/WHO_TRS_908.pdf .
37. Anonymous ( 2005 ), Annex 2: Good manufacturing practices: Requirement for the sampling
of starting materials (Amendment), in WHO Expert Committee on Specifi cations
for Pharmaceutical Preparations: 39th Report, WHO Technical Report Series 929,
World Health Organization, Singapore, pp. 38 – 39, available: http://whqlibdoc.who.int/trs/
WHO_TRS_929_eng.pdf .
38. Anonymous ( 1992 ), Annex 1: Good manufacturing practices for pharmaceutical products:
Good manufacturing practices for active pharmaceutical ingredients (bulk drug
substances), in WHO Expert Committee on Specifi cations for Pharmaceutical Preparations:
32th Report, WHO Technical Report Series 823, World Health Organization,
Geneva, pp. 72 – 79, available: http://whqlibdoc.who.int/trs/WHO_TRS_823.pdf .
39. Anonymous ( 1998 ), Annex 5: Good manufacturing practices: Supplementary guidelines
for the manufacture of pharmaceutical excipients, in WHO Expert Committee on Speci-
fi cations for Pharmaceutical Preparations: 35th Report, WHO Technical Report Series
885, World Health Organization, Madrid, Spain, pp. 50 – 71, available: http://whqlibdoc.
who.int/trs/WHO_TRS_885.pdf .
40. Anonymous ( 2002 ), Annex 6: Good manufacturing practices for sterile pharmaceutical
products, in WHO Expert Committee on Specifi cations for Pharmaceutical Preparations:
36th Report, WHO Technical Report Series 902, World Health Organization, Singapore,
pp. 76 – 93, available: http://whqlibdoc.who.int/trs/WHO_TRS_902.pdf .
41. Anonymous ( 1993 ), Annex 3: Good manufacturing practices for biological products, in
WHO Expert Committee on Specifi cations for Pharmaceutical Preparations: 33th Report,
WHO Technical Report Series 834, World Health Organization, Geneva, pp. 20 – 30,
available: http://whqlibdoc.who.int/trs/WHO_TRS_834.pdf .
42. Anonymous ( 1995 ), Annex 7: Good manufacturing practices: Supplementary guidelines
for the manufacture of investigational pharmaceutical products for clinical trials in
humans, in WHO Expert Committee on Specifi cations for Pharmaceutical Preparations:
34th Report, WHO Technical Report Series 863, World Health Organization, Geneva, pp.
97 – 108, available: http://whqlibdoc.who.int/trs/WHO_TRS_863_(p1 - p98).pdf (pp. 97 – 98);
http://whqlibdoc.who.int/trs/WHO_TRS_863_(p99 - p194).pdf (pp. 99 – 108).
43. Anonymous ( 1995 ), Annex 8: Good manufacturing practices: Supplementary guidelines
for the manufacture of herbal medicinal products, in WHO Expert Committee on Speci-
fi cations for Pharmaceutical Preparations: 34th Report, WHO Technical Report Series
863, World Health Organization, Geneva, pp. 109 – 113, available: http://whqlibdoc.who.
int/trs/WHO_TRS_863_(p99 - p194).pdf .
44. Anonymous ( 2002 ), Annex 3: Guidelines on good manufacturing practices for radiopharmaceutical
products, in WHO Expert Committee on Specifi cations for Pharmaceutical
Preparations: 37th Report, WHO Technical Report Series 908, World Health
Organization, Singapore, pp. 26 – 35, available: http://whqlibdoc.who.int/trs/WHO_TRS_
908.pdf .
45. Anonymous ( 2005 ), Annex 3: WHO Good manufacturing practices: Water for pharmaceutical
use, in WHO Expert Committee on Specifi cations for Pharmaceutical Preparations:
39th Report, WHO Technical Report Series 929, World Health Organization,
Singapore, pp. 40 – 58, available: http://whqlibdoc.who.int/trs/WHO_TRS_929_eng.pdf .
46. Anonymous ( 2006 ), Background to PIC, available: http://www.picscheme.org/
indexnofl ash.php?p=backg .
47. Anonymous ( 2006 ), Introduction, available: http://www.picscheme.org/indexnofl ash.
php?p=intro .
48. Anonymous ( 2006 ), List of PIC/S participating authorities ( & observers), available: http://
www.picscheme.org/indexnofl ash.php?p=members .
49. Brunner , D. ( 2004 ), Pharmaceutical inspection co - operation scheme (PIC/S) , Qual. Assur.
J. , 8 , 207 – 211 .
50. Anonymous ( 2006 ), Guide to good manufacturing practice for medicinal products, PE
009 – 3, Pharmaceutical inspection co - operation scheme, available: http://www.picscheme.
org/guides.php# .
51. Anonymous ( 2006 ), Structure of ICH, available: http://www.ich.org/cache/html/
510 - 272 - 1.html .
REFERENCES 161
162 CORRESPONDENCES AND DIFFERENCES
52. Anonymous ( 2000 ), Good manufacturing practice guide for active pharmaceutical ingredients
Q7, ICH harmonised tripartite guideline, ICH Steering Committee, available:
http://www.ich.org/LOB/media/MEDIA433.pdf .
53. Anonymous ( 2006 ), Quality guidelines, available: http://www.ich.org/cache/compo/
363 - 272 - 1.html .
54. Anonymous ( 2006 ), The founding of ASEAN, available: http://www.aseansec.org/
7069.htm .
55. Anonymous ( 2006 ), Pharmaceuticals, available: http://www.aseansec.org/8657.htm .
56. Vernengo , M. J. ( 1998 ), Advances in pharmaceutical market integration in MERCOSUR
and other Latin American countries , Drug Info. J. , 32 ( 3 ), 831 – 839 .
57. Anonymous ( 2005 ), Division 1A establishment licences, in Consolidated Statutes and
Regulations , Food and Drugs Act, Food and Drug Regulations, Part C, Department of
Justice Canada, available: http://laws.justice.gc.ca/en/f - 27/c.r.c. - c.870/230049.html .
58. Anonymous ( 2002 ), Note for guidance on quality of water for pharmaceutical use, CPMP/
QWP/158/01, Committee for Proprietary Medicinal Products, Quality Working Party,
London, available: http://www.emea.eu.int/pdfs/human/qwp/015801en.pdf .
59. Anonymous ( 1998 ), Stability testing on active ingredients and fi nished products, in The
Rules Governing Medicinal Products in the European Union , Vol. 3A, European Commission
Directorate General III, pp. 143 – 151, available: http://pharmacos.eudra.org/F2/
eudralex/vol - 3/pdfs - en/3aq16aen.pdf .
QUALITY
SECTION 3
165
3.1
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
ANALYTICAL AND COMPUTATIONAL
METHODS AND EXAMPLES FOR
DESIGNING AND CONTROLLING
TOTAL QUALITY MANAGEMENT
PHARMACEUTICAL
MANUFACTURING SYSTEMS
Paul G. Ranky, 1 Gregory N. Ranky, 2 Richard G. Ranky, 1 and
Ashley John 1
1 New Jersey Institute of Technology, Newark, New Jersey
2 Public Research University of New Jersey, Newark, New Jersey
Contents
3.1.1 Introduction
3.1.2 Flexible Pharmaceutical Manufacturing and Assembly System Design
3.1.3 Flexible Manufacturing Model Integrated with Design
3.1.4 Real - Time Operation Control
3.1.5 Innovative Design
3.1.6 Open Innovation Architecture
3.1.7 Generic, Object - Oriented Innovation Process Modeling Method and Sample
Model
3.1.8 Systems Approach to Pharmaceutical Manufacturing Systems Management
3.1.9 Requirements Analysis for System Product, Process, and Service Design
Innovation
3.1.10 Innovation Risk Analysis and Opportunity Method and Tool with Pharmaceutical
Manufacturing System Applications
3.1.11 Open - Source Computational Statistical and Three - Dimensional Multimedia for
Pharmaceutical Manufacturing System Innovation and Project Communication
3.1.12 RFID Applications
3.1.13 RFID Examples
166 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
3.1.14 RFID Integration Models for Digital Pharmaceutical Manufacturing and Assembly
Supply Chains
3.1.15 Evaluation of Network Simulation Results
3.1.16 Summary
3.1.17 Complimentary Video on DVD
References
3.1.1 INTRODUCTION
Total quality management (TQM) and operation control in pharmaceutical manufacturing
system design engineering is essential. TQM - focused pharmaceutical
manufacturing system engineering involves the continual satisfaction of customer
requirements at lowest cost by harnessing the efforts of everybody in the company.
Quality assurance means sustaining a system that prevents defects. This includes
quality control and quality engineering. Quality control means establishing and
maintaining specifi ed quality standards of products; quality engineering is the establishment
and execution of tests to measure product quality and adherence to acceptance
criteria.
This chapter explains the importance of reducing variation for the purpose of
implementing total quality in every process of the pharmaceutical design and manufacturing
enterprise. Furthermore, it represents a modular product, process, service
design, implementation, and management approach to the introduction of various
TQM methods, tools, technologies, and their management issues within a variety of
small, medium, and large enterprises for the purpose of designing and controlling
pharmaceutical manufacturing systems.
These aspects are very important, clearly illustrated by the fact that the U.S. Food
and Drug Administration (FDA) has three classifi cation levels for medical
products:
• Class I products are passive devices that do not enter the patient ’ s body or
contact only the skin.
• Class II products are active devices or devices that are used to administer fl uids
to the patient ’ s body.
• Class III products are implanted inside the patient ’ s body.
The FDA is familiar with the complexity of designing pharmaceutical systems. To
support this activity, there are several software tools that help product/process and
system designers to achieve the above.
It should also be noted that the FDA expects design validation results to accompany
some submissions. This is particularly true of class II and III devices. The
agency expects such analysis results to match those obtained with established experimental
methods. A number of software tools, including fi nite element analysis
(FEA), motion and actuation simulation, computational fl uid dynamics (CFD), in
conjunction with the computer - aided design (CAD) used for the designs themselves
and other solutions are available that help today ’ s pharmaceutical/medical designer/
medical manufacturing/assembly system designer to meet the complex requirements
of the industry as well as the FDA. (The key, here, is to accept the important
principle that pharmaceutical design and manufacturing/assembly and even packaging
must be an integrated approach.)
The main problems when applying a traditional quality management philosophy
to any pharmaceutical design/manufacturing/assembly challenge include the
following:
• This philosophy focuses on correcting mistakes after they have been made,
rather than preventing them in the fi rst place.
• It allows mistakes to be made. It actually builds them into every aspect of the
system, typically costing around 20% of the turnover.
• It accepts that quality has to be sacrifi ced as the volume and the productivity
go up.
• As viewed by accountants, it is an expensive add on item of the value chain.
However, modern thinking claims that, because TQM involves every person,
aspect, and machine of the organization, it requires a total commitment. It is not a
“ test - and - fi x ” approach. It is a preventive system designed into every aspect of the
world - class design, manufacturing, and service enterprise, including product design,
manufacture, and management (and even in accounting terms costing somewhat
less than conventional quality systems, i.e., typically around 10% of the turnover).
The fundamental goal of TQM and TQC (total quality management and control)
is to program, measure, and keep process variability under control. Some of these
methods discussed in this chapter are as follows:
• Pharmaceutical manufacturing system design methods and tools with
examples
• Process modeling for designing and running pharmaceutical manufacturing
systems
• Requirements analysis modeling for pharmaceutical manufacturing systems
• Risk analysis modeling for pharmaceutical manufacturing systems
• Dynamic modeling and network simulation for globally distributed pharmaceutical
manufacturing systems and other methods and tools
3.1.2 FLEXIBLE PHARMACEUTICAL MANUFACTURING AND
ASSEMBLY SYSTEM DESIGN
A fl exible pharmaceutical manufacturing/assembly system, (FMS) is a highly automated,
distributed feedback - controlled system of data, information, and physical
processors, such as computer and manually controlled machines, cells, workstations,
and robots, in which decisions have to be made often in real time. This is only possible
if all information processors (including the human resources of such systems)
are “ well informed ” and lean/fl exible, meaning that they have the exact information
at the exact time, format, and mode they need to allow responsible decision making
within given time constraints. Note that this is a fundamentally different system
design concept than that of the transfer line, operating on a fi xed cycle time, and
designed for large batch production [1 – 8] .
When designing a fl exible manufacturing/assembly system (FMS/FAS), the design
team should consider the following steps:
FLEXIBLE PHARMACEUTICAL MANUFACTURING 167
168 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
1. Collect all current and possible future user and system requirements.
2. Analyze the system (i.e., the data processing and the FMS/ FAS hardware and
software constraints).
3. Design an appropriate data structure and database for describing processors
and their resources, such as machines, robots, and tools (and/or robot hands,
probes, sensory - based inspection and assembly tools, etc.).
4. Specify and design programs and query routines and dialogues that are capable
of accessing this database as well as communicating with the real - time production
planning and control system of the FMS/FAS.
5. Design and integrate the system with the rest of the hardware and software,
including on - line manuals, education, and training packages, preferably in
interactive, engineering multimedia format.
6. Maintain the system and continuously learn for the benefi t of the existing as
well as future system designs.
Probably the most important questions to be answered before starting to design
such a system are: Who is going to use it? For what purposes? With what data? How
will it be used?
As an example, consider that tooling data in FMSs will typically be used by
several subsystems as well as by human beings are as follows:
• Production planning subsystem
• Process control
• Part programming
• Tool preset and tool maintenance
• Tool assembly (manual or robotized)
• Stock control and material storage
By employing the above subsystems, the production planning system has to be
informed in real time about the availability of tools in stock as well as about the
current contents of the tool magazines of the machine tools (in the case of FASs
the robot hands in the end - of - arm - tool magazines); otherwise it will not be able to
generate a proper production schedule.
It must be noted that the real - time aspect is important because tools are changed
in the magazines of machines (or cells), not only because they wear, but also because
different part programs may need different sets of tools. (The actual tool - changing
operation is done in most cases by manipulators or by robots. The tool magazine
loading/unloading procedure is performed mostly by human operators, sometimes
by robots or special - purpose mechanisms, such as a tool shuttle.)
Both the process control and the production planning systems have to update
any changes and act in real time; otherwise the operation of the system can be
disrupted.
From the FMS/FAS tooling and tool management points of view one must
emphasize the links between the CAD system, in which the parts are designed
(using design for manufacturing principles), and the computer - aided manufacturing
(CAM) system, where the FMS part programs are written. Typically, an FMS part
programmer analyzes the CAD output (i.e., the design drawings of the pharmaceuti
cal products to be manufactured/assembled on the FMS), the fi xturing, the different
setup (i.e., work - mounting) tasks, as well as the necessary operations, their alternatives,
the required tools, and fi nally a precedence list of the resources (i.e., the possible
candidates of processing stations, or cells, or machines).
Real - time databases and software systems are also important, since they provide
the reports and status information that are needed for the smooth operation of the
FMS (in particular, its dynamic scheduler and other subsystems such as maintenance
should be emphasized here) [4, 9 – 14] .
3.1.3 A FLEXIBLE MANUFACTURING MODEL
INTEGRATED WITH DESIGN
The output of the CAM system is a production rule base. This is the knowledge the
FMS needs to produce each pharmaceutical product. In this production rule base,
among others, tools are assigned to each operation. The tool codes are selected by
the FMS process planner or automatically assigned by a process planning system
and are obtained from the tool database.
On the basis of the requested tools a list is sent via the network to the tool
preparation facility, or station, where the actual tools are prepared (i.e., assembled
and preset) and stored in an appropriate way such that the material - handling system
of the FMS can pick them up [12 – 21] .
The tool preparation station also deals with other activities, among which the
most important are as follows:
• Tool service and maintenance
• Tool assembly to orders (as it is necessary to replace worn tools)
• Tool preset, tool inspection and adjustment
• Real - time tool pickup and tool transportation organized to serve the needs of
the real - time FMS
The tool preparation station receives its orders, initially originated by the CAD
data processing system, via the FMS network and technically specifi ed by the CAM
system in the form of a production rule base. Order data arriving at the tool preparation
station include the following:
• Part orders (consisting of part codes and quantities). Note that this is a very
important data set for the real - time FMS dynamic scheduler too.
• Notifi cation of when the parts are physically available for FMS processing,
representing a due date for tool preparation.
• A priority order (note that this can change because of some real - time changes
in the system, and thus this station must be able to cope with this task too).
• The portion of the production rule base describing the requirements regarding
tool preparation.
The tool preparation station keeps in touch with the real - time FMS system, as well
as with the rest of the system, by feeding back important tooling system - related
data:
A FLEXIBLE MANUFACTURING MODEL INTEGRATED WITH DESIGN 169
170 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
• Stock reviews (regarding tools)
• FMS status report (regarding tools)
• Part priority status reports (in case dynamic changes must be performed in the
FMS which have an effect on tooling needs and tool preparation due dates)
3.1.4 REAL - TIME OPERATION CONTROL
The real - time part of the FMS operation control and management system must deal
with the following tasks:
• It must handle the application of tools for a variety of processes as defi ned in
the production rule base and assigned in real time to the FMS/FAS resources
by the dynamic scheduler.
• It must provide data to control the transportation of tools and tool magazines
within the FMS.
• It must provide information to perform and supervise tool changes and tool
magazine changes at all levels.
• It must be notifi ed of tool inspection results (e.g., if it fi nds a worn - out tool as
a result of an inspection procedure, it must generate a command that instructs
the tool magazine update system to change the tool in question in the appropriate
tool magazine).
• It must provide information in the case of emergency.
• It must provide the necessary interfaces and data to perform diagnostic/recovery
operations, preferably using diagnostic expert systems.
Finally, let us underline an important feedback loop starting at the real - time
system and ending at the tool preparation station, which contains the real - time tool
status, wear, and part priority information. These data are often useful to those
people and/or system software systems that deal with the generation of the production
rule base. It is also a very useful data set for FMS designers, since a lot of data
which would previously have been lost will be saved in this way.
The most important operation control activities in FMS/FAS identify three levels
at which simulation and optimization are required prior to or during FMS/FAS part
manufacturing:
1. The factory level or business level handled by the business system of the
computer integrated manufacturing (CIM) or, even broader, the enterprise
resource management system
2. The FMS off - line level representing scheduling, simulation, and optimization
activities prior to loading a batch or a single component on the FMS (handled
sometimes by the CAM system, sometimes by the FMS part programming
computer)
3. The real - time controlled level handled by the FMS/FAS operation control
system, a dynamic scheduler with integrated tool management and multimedia
support, representing a situation where the parts are already physically as well
as logically in the real - time controlled environment
Due to its complexity, a truly integrated approach is required in designing a
production rule base to provide the job description for the FMS dynamic scheduler.
This is because the dynamic system relies heavily on the knowledge base as represented
by the rule base, and an overly restrictive rule base will lead to ineffi cient,
at times even wrong, decisions. In other words, such a structure should represent all
the multilevel interactions and their possible precedence rules that relate to the
manufacturing process planning and processing decisions in an FMS. This turns out
to be a diffi cult task.
It should be underlined that the application of multimedia at this level is extremely
benefi cial in terms of part program preparation, teaching/training operators on
setting up parts, fi xtures, tools, machines, for troubleshooting, for regular maintenance,
at the computer numerical control (CNC) level programming, robot programming,
placement machine programming, programmable logic controller (PLC)
programming, quality control, maintenance, and other tasks.
Most FMSs have some part - buffering capability. This may be not for scheduling
reasons, but for technological, that is, process - planning, reasons (e.g., the part must
cool before an accurate inspection procedure is performed). Some level of buffering
is useful and necessary because of reliability reasons. (The actual number of buffer
store locations should be established on the basis of simulation and experience.)
Cells often have some buffers too. The reason for this is that, by providing a part
in the input queue of the cell just before the currently processed part is fi nished at
the particular cell, the cell is kept running at its highest effi ciency level, since time
is only “ wasted ” for part changing. The other important point to note is that well -
designed part buffers offer a direct access pickup/load facility, making the rescheduling
process in the queues short, simple, and dynamic [18, 19, 21 – 27] .
3.1.5 INNOVATIVE DESIGN
The key objective of this chapter is to describe a generic and systematic pharmaceutical
manufacturing/assembly system design method that includes product,
process, service systems, and even innovation project management architecture
aspects of such systems.
This architecture must be simultaneously novel as well as compliant with set
guidelines by the product/process design industry and the PMI (Project Management
Institute), following International Organization for Standardization (ISO)
9000:2000 quality standards. Our tested pharmaceutical manufacturing system
design solution integrates object - oriented process modeling, requirements and risk
analysis, statistical methods, design of experiments, and three - dimensional (3D)
interactive multimedia methods and tools which are 100% Web compatible.
Furthermore, our methods and software tools are generic in that they can be
applied not only to systems such as the pharmaceutical industries or automobile
manufacturing but also to processes such as the oil business or services such as
education.
A pharmaceutical manufacturing system design requires signifi cant level of innovation.
The broadest defi nition of innovation is the act of introducing something
new to a society or community, whether a product or process. This is often confused
with invention, which focuses more on specifi c objects. Within pharmaceuticals
innovation can therefore include new business structures within the company,
INNOVATIVE DESIGN 171
172 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
manufacturing processes and quality control for the medications, and product materials.
Process and service improvements can also qualify as innovation, but note that
in this case services are usually counted as processes [4, 13, 21, 23, 28 – 32] .
Discoveries such as the charting of new planets, land masses, or forms of life are
not classifi ed as innovations as they had existed before being observed by humans.
When new species are introduced into a society and fi nd a specifi c use, it can be
classifi ed as innovation. A pharmaceutical example of this is the antibiotic penicillin.
Although it had existed as a fungal secretion, it was only within the past century
that it was used to actively eliminate infectious bacteria. In initial analysis, however,
it was not thought to survive long enough within the human body to be effective.
This brings a vital aspect of innovation, namely the ability to recognize alternative
uses for existing processes or tools. This is diffi cult as unexpected changes within a
system are usually labeled as mistakes or anomalies. The development of the Post - It
note is an example of this. The original goal was to create a high - strength adhesive,
and an extremely weak one was created by accident. Nonetheless, instead of simply
disposing of it, the possibilities of this new substance were examined by technicians
and managers alike, allowing the use of easily placed reminders for everyday usage.
Possessing an ultraweak adhesive allows Post - It notes to be removed without damaging
the surfaces that they are placed on, and they are available in a variety of
colors and sizes.
The former example is a radical innovation, not only because it allowed signifi -
cant changes in message reminders and adhesives, but also because it was completely
unexpected. The diversifi cation of Post - It notes into different sizes and colors
is an example of incremental innovation, which involves step - by - step changes and
improvements to existing products or processes. Radical innovations are far less
common, though their effects are farther reaching over both society and history.
The general trend through human history has been one of learning to consciously
recognize and direct innovation, particularly through combining science and technology.
Human societies have often worked with certain processes even without
fully understanding their effects or underlying ideas. Metallurgy shows this clearly,
as iron, bronze, and gold have been used for millennia before the molecular structure
could be seen and analyzed. Note also that, although our ancestors could not
describe their chemical composition, these metals served a great many successful
purposes. In these cases, the goals of innovation are highly pragmatic, as successful
solutions are passed down and taught to future generations.
Those who innovate can therefore learn from working, viable solutions to begin
their own practices. Those who continuously work with a fi xed set of designs must
be willing to experiment, test, and diversify their practices to avoid stratifi cation, as
innovation not only allows survival but also encourages prosperity.
The ability to innovate also involves learning from past mistakes, not just one ’ s
own. Mistakes and errors in practices can be both costly and dangerous but can be
prevented from occurring successively if their causes are determined. This can be
diffi cult because a near - miss scenario can be seen either as an infrequent event or
as an averted disaster. The fi rst reaction to this is usually to continue without changing
current practices, allowing for similar mistakes to occur. Learning requires all
levels of an organization to participate and create channels of communication to
innovate effectively, as the inability to share experience denies new opportunities
[13, 29 – 40] .
3.1.6 OPEN INNOVATION ARCHITECTURE
Innovation as a process and the related research - and - development (R & D) project
management are considered to be two of the most complex information systems
and engineering architectures due to the large number of attributes, processes, and
dynamic changes projects go through during their life cycle.
Following our integrated and simultaneously open architectural approach, we
look at every innovation process and project as a system built of objects and classes
of objects.
Then we look at the way the components of these systems interact with each
other. Once we understand these behaviors, we follow our integrated system
approach in terms of looking at the project management system as processes, trying
to satisfy customer requirements and also representing risks.
We then embed this system model into a statistical analysis and 3D interactive
multimedia framework (Figure 1 ). We use statistical methods to capture processes
before they go out of control as well as to perform trend analysis, a great opportunity
for innovation, and use 3D interactive multimedia and 3D visualization methods
over the Web for communication purposes with global innovation team members.
The emphasis on collaboration in today ’ s competitive medical drug fi eld requires
these virtual environments to streamline team interaction. (Note that the active
FIGURE 1 When designing lean and fl exible pharmaceutical manufacturing/assembly/
packaging systems, one needs to analyze the required processes, customer, user, maintenance,
quality, reliability, fl exibility, lean, design requirements, and risks involved with any of the
listed processes, all in a statistical framework. (Note that our 3D interactive multimedia and
simulation framework supports integrated digital design and digital manufacturing system
design principles, meaning that one should test all designs and systems fi rst on the screen,
and only if everything looks fi ne, in the real world.)
Process analysis
Risk analysis
statistical analysis, design of experiments,
Web-base 3D interactive multimedia, DVD full-screen
vidoeos and just-in-time iPod videos for
knowledge managemet
Requirements analysis
OPEN INNOVATION ARCHITECTURE 173
174 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
code spreadsheets and 3D objects referred to in this presentation are all part of
Ranky ’ s eLibrary and are available at http://www.cimwareukandusa.com .)
To illustrate the importance of the “ openness ” of our architecture, consider
modern simulation/analysis tools by Parametric Technology Corporation (PTC)
(Figure 2 ), and PLM (product life - cycle management) tools, such as the IBM/Dassault
Systemes Delmia tools for pharmaceutical manufacturing system modeling
and design (Figure 3 ) with sensory feedback processing (Figure 4 ).
Since these models can be designed, edited, run, and driven even over the limits,
they can be extremely valuable sources for modeling in the digital domain, process
analysis, requirements modeling, risk analysis, and even collecting statistical data
and modeling breakdowns of complex systems.
Observe the FEA in Figure 2 a . This is a torsional test of a pharmaceutical manufacturing
machine element on the assembly line of a new medication packaging line.
The line is still being tested and improved in the virtual environment, which greatly
streamlines the refi nement process. As can be seen by the von Mises stress distribution,
the sharp edges of the shaft will need to be rounded with a fi llet. These would
also increase the distribution of the same stress and thus reduce the majority of the
red zones (high stress) to blue or even green (low stress). Without using the virtual
assembly line to test ideas before for the physical, the unexpected failure of this
part could create delays or contamination of product or even harm human
operators.
As can be seen, digital pharmaceutical manufacturing/assembly/packaging and
factory design tools include not only machines but also advanced sensors, actuators,
controls, material - handling systems, labeling machines, and even ergonomically realistic
human models and operators performing real - world tasks in extremely realistic
model factories. Simulations like these are not just pretty models; they actually save
huge investments because the factories are not built until the models are satisfactory.
Keep in mind that making changes in a physical factory costs time, money, and
possibly production effi ciency, even to just check a possible improvement. Virtual
models can be simultaneously run thousands of times over a period of days, with
hundreds of variables being optimized until the appropriate combination is chosen
[30 – 36, 38, 40 – 44] .
3.1.7 GENERIC, OBJECT - ORIENTED INNOVATION PROCESS
MODELING METHOD AND SAMPLE MODEL
Understanding, modeling, and then following processes, procedures, and best practice
reusable processes are essential for every business to stay at the top. The pharmaceutical
manufacturing system “ innovation business ” is not exception.
Major international product/process design standards written and reviewed by
thousands of leading researchers and companies around the world always help to
create a model for complex problem - solving challenges such as innovation. Therefore
this section discusses two of the eight quality management principles of the
ISO 9000:2000 international quality standard and the way these rules should be
applied to pharmaceutical manufacturing system designs.
We do this for the purpose of developing systematic innovation (with related
project modeling skills) and reusable, tested pharmaceutical system design
FIGURE 2 Finite element torsional test of pharmaceutical manufacturing machine element
on assembly line of new medication packaging line. The line is still being tested and improved
in the virtual environment, which greatly streamlines the refi nement process. As can be seen
by the von mises stress distribution, the sharp edges of the shaft will need to be rounded with
a fi llet. These would also increase the distribution of the same stress and thus reduce the
majority of the red zones (high stress) to blue or even green (low stress).
Von Mises stress distribution plot Displacement distribution plot
(a)
max_disp_mag
(mm)
P_Pass
Scale 1.0000E + 00
Loadset: LoadSet1
max_disp_mag
max_disp_mag
0.00
0.25
0.20
0.15
0.10
0.05
0 2 3
PLoop Pass
4 5 6
strain_energy
(mm N)
P_Pass
Scale 1.0000E + 00
Loadset: LoadSet1
strain_energy
strain_energy
2.00
4.00
6.00
10.00
8.00
12.00
14.00
16.00
18.00
0 2 3
PLoop Pass
4 5 6
max_strees_vm
(N/mm^2)
P_Pass
Scale 1.0000E + 00
Loadset: LoadSet1
max_strees_vm
max_strees_vm
40.00
50.00
60.00
70.00
80.00
90.00
100.00
110.00
120.00
130.00
140.00
0 2 3
PLoop Pass
4 5
(b)
176 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
FIGURE 3 Modern simulation/analysis and PLM tool: IBM/Dassault Systemes Delmia for
pharmaceutical manufacturing system modeling and design. The benefi ts are huge, since the
system can be built and tested in the digital domain. (Courtesy of IBM/Dassault Systemes
Delmia, Inc.)
processes. We use our own object - oriented process modeling method, called CIMpgr.
The nature of this method tansforms into UML models [the Unifi ed Modeling
Language of information technology (IT)] and also complies with international
process modeling standards used in complex system modeling environments.
First, we will discuss a few important defi nitions that closely relate to ISO
9000:2000 (quality process modeling) standard principle 4:
• A process, or activity, can be defi ned as a transfer function with one or more
inputs, outputs, controls, and resources that together all enable the variables to
gain data and then fi re.
• Transfer functions, when fi red, create a transformation process. A transformation
process in a project is made up of methods, steps, tasks, and various algorithms
and processes that acquire and manipulate data and then turn it into
system output(s). Note that the input data can describe material, human knowledge,
technological standing, fi scal information, and others.
• The output of the process is a product that consists of specifi c technical and/or
social products and services that conform to the sponsor ’ s requirements.
• Processes, in terms of quality project management, have visibility, documentation,
and traceability.
• In this context visibility, relates to whether we know and transparently (or
graphically) see what methods and techniques, system process steps, and technologies
are involved when creating the desired output. Do we know the
FIGURE 4 Advanced sensors working in pharmaceutical assembly systems help real - time
operation control and quality assurance system to test every product. (This is often referred
to as the zero - defect policy designed into a system.) The luminescence sensor illustrated will
detect a wide variety of invisible targets. This STEALTH - UV sensor was designed to sense
the presence of invisible fl uorescent materials contained in or added to many products. Users
can detect the most diffi cult targets, including clear tamper - proof seals, clear labels, and invisible
registration marks. This unique sensor is also ideal for solving many of today ’ s toughest
problems in product orientation, inspection, and verifi cation. (Courtesy of TRI - TRONICS
Co., Inc., www.ttco.com .)
sequence of these steps and the possible parallel process relationships? How
does one process affect the other?
• Documentation means that the methods, steps, processes, and technologies are
well specifi ed and recorded according to agreed - upon standard specifi cations.
• Traceability means that the process steps as well as the output(s) can be traced
back to actual customer requirements.
• Process capability can be defi ned as the ability of the production process to
meet certain specifi cations and tolerances.
• Process discrepancy is the deviation of process settings from specifi cations.
• Process variability is the variation in dimensional or other measurable characteristics
of output from a production process. (Note that in any project the
ultimate goal is to stay within the predefi ned limits of process variability and,
if possible and feasible, to reduce process variability, because this typically
reduces risk too.)
• Variability can be expressed in terms of average range of standard deviation.
• A process variable is a process parameter that fl uctuates in the manner of a
random variable and hence requires surveillance.
• Process management means getting the activities and procedures that highly
skilled and experienced managers carry in their heads into the open by means
of a well - documented model, often referred to as the process model [40 – 46] .
GENERIC, OBJECT-ORIENTED INNOVATION PROCESS MODELING METHOD 177
178 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
3.1.8 SYSTEMS APPROACH TO PHARMACEUTICAL
MANUFACTURING SYSTEMS MANAGEMENT
Identifying, understanding, and managing interrelated processes as a system contribute
to the organization ’ s effectiveness and effi ciency in achieving its objectives.
(Note that every one of the key drivers listed below embed one or more innovation
opportunities!)
Key drivers and achievable gains include the following:
• Processes that will achieve the desired results will become better integrated
and aligned.
• Management and process owners will have the ability to focus their efforts on
the key processes.
• Since the consistency, effectiveness, and effi ciency of the organization will grow,
the confi dence of interested parties and collaborators in the organization will
grow too.
• System structuring and fi ne - tuning will become possible to achieve the organization
’ s objectives in the most effective and effi cient way.
• Understanding the interdependencies between the processes of the system will
yield good results.
• Structured (and object/component - oriented) modeling approaches that harmonize
and integrate processes will become reality. Employees will understand
them, follow them, and therefore reduce waste and increase quality in every
process.
• The resistance created by cross - functional barriers will be reduced, providing
a better understanding of the roles and responsibilities necessary for achieving
common objectives.
• Organizational capabilities and the establishment of resource constraints prior
to action will be better understood by all involved (and mostly by all those who
have created the models).
• Targeting and defi ning how specifi c activities within a system should operate
will become reality.
• Continually improving the system through measurement and feedback -
controlled evaluation becomes possible due to the analytical and quantifi able
approach of the process models. (Note that at its ultimate level this will lead
to a real - time, feedback - controlled enterprise capable of reacting to dynamically
changing market needs.)
After this introduction, let us show our object - oriented system components,
following the above described ISO 9000:2000 principles, and how we can
model complex innovation and related project management processes using them
[40 – 54] .
As a simple example, consider, that you are packaging a pharmaceutical product
using a line that performs various process steps. Figure 5 a illustrates one of
these steps. It has input(s), output(s), control(s), and resource(s). These data
types help to identify under what conditions the process should be executed by
the pharmaceutical manufacturing system. (Defi nitions are offered in the diagrams.)
We can also see the way the CIMpgr process maps into a UML diagram.
This is important, since UML is the modeling language of the IT professionals
who will program the PLCs and control systems for the lines. Figure 5 b shows
how the CIMpgr process maps into a UML diagram. We can see in Figure 5 a how
multiple processes have to interact as we design a pharmaceutical assembly
system.
FIGURE 5 Object - oriented process modeling method (CIMpgr) as applicable to pharmaceutical
manufacturing/assembly/packaging system design.
This box represents the process. We can
identify a process by naming it A0 (as a
parent), and it’s children (a A1, A2, etc.)
CIMpgr Process Model A0
This is the control side to our process. This is where data
somehow limits, or controls the process. (As an example for
controls, imagine the international emission control
regulations that automotive designers must follow). We can
identify each control by a variable name, such as C1A0.
A0
DBI_A0:
This identifies a data
storage, a file, or a
database for the process.
This is the resource side to our process. This is where
data describes the available manpower, hardware,
software, and other available resources for executing the
process. We can identify each resource by a variable
name, such as R1A0.
This is the administrative
section of out model.
Purpose: Why are we doing this? What is the fundamental purpose of this model?
Viewpoint: ‘As is’ System Analyst, System Designer; ‘To be’ System Analyst, System Designer
Authoring Team Members: Ranky, with
Key Contact: Paul G. Ranky, Email: cimware@earthink.net, USA Tel: (201) 493 0521
Client Ref: Company ABC Inc.
Data: January 21, 2004, Version: ver. 1.0
Confidential! Public Release: OK; Object / Class inheritance: ON
R1A0:
R2A0:
C3A0:
C4A0:
C2A0:
C1A0:
I1A0:
I2A0:
I3A0:
I4A0:
I5A0:
I6A0:
C5A0:
C6A0:
R3A0:
R4A0:
R5A0:
O5A0:
O4A0:
O2A0:
O1A0:
This is the
output side to
our process.
This is where
data leaves
the process.
We can
identify each
output by a
variable
name, such
as O1A0.
This is the
input side
to our
process.
This is
where data
enters the
process.
We can
identify input
by a
variable
name, such
as I1A0.
(a)
PHARMACEUTICAL MANUFACTURING SYSTEMS MANAGEMENT 179
180 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
Class NameL ClassA0
Attributes in detail (also in CIMpgr model):
Input(s):
I1A0= ,
I2A0= ,
Output(s):
O1A0= ,
O2A0= ,
Control(s):
C1A0= ,
C2A0= ,
Resource(s):
R1A0= ,
R2A0= ,
Link to
Requirements Analysis Model Filename:
Risk Analysis Model Filename:
Attributes
Attributes
Operations
Operations
Operations (Mathematical TR functions, pseudo code for
proc. descr., etc., using the attributes, as above=CIMpgr
model attributes)
Class Name:
ClassA1
Attributes
Attributes
Operations
Operations
Class Name:
ClassA2
(b)
I1A1: New
need from
the customer
/project
sponsor
C1A1:
I2A1:
I1A2:
I3A1:
I4A1:
I5A1:
I6A1:
I7A1:
C6A1:
C1A3:
I1A3:
C2A3:
C2A4:
C1A4:
C1A5:
C2A5:
I1A4:
R2A4:
C2A4:
I1A4:
C2A2:
C1A2:
C5A1:
C2A1:
Project time, budget and quality control. (Note that there are many other types of control, that relate to environmental
issues, specific design, material, manufacturing, assembly, test, service, IT, and other processes and sub-systems.)
Conception, or Conceptual,
or Requirements Analysis
Phase, or Process A1
A1
A2
A3
A4
O1A1: The results of the requirements analysis study. (This data-set offers valuable information for this future projects,
for data mining, and for knowledge management.) O2A1: Project
specification: What?
When? How much?
etc.
R1A1: CORA
software tool,
and
experienced
CORA
consultant R2A1:
Requirements
analysis team
R3A1: PFRA
risk analysis
tool,project
time and cost
management
software and
consultant Definition, or Planning, or
System Analysis Phase, or
Process A2
O1A2: Project specification results (This is n important data-set in case of multiple projects with precedence constraints.
Also, this data-set offers valuable information for data mining, and for knowledge management.) O2A2:
Detailed
project
specification
DBI_A1:
Process A1
related
requirements
analysis
results are
stored here
(e.g. results
of a CORA
study)
DBI_A2:
The
system
analysis
documents
and data
are stored
here
R1A2: CIMpgr
process
model
drawing tools,
optional
dynamic
simulation
software, and
experienced
consultant
R2A2: Project
time
management
and budget
management
software, and
experienced
planning team Design, or Acquisition, or
System Design Phase, or
Process A3
O1A3 Project design result. (This data-set offers valuable information for
data mining, and for knowledge management.) O2A3:
Detailed
project
design
DBI_A3:
The
system
design
blueprints
are stored
here
R1A3: Project
time
management
and budget
management
software, and
experienced
project design
team
R2A3:
Specific
CORA
requirements
analysis and
PFRA risk
analysis, and
product/
process
experts
Operation, or Integration,
or System Implementation
and Test Phase, or
Process A4
A5
O1A4: Project implementation results. (This data-set
offers valuable information for data mining, and
for knowledge management.)
O2A4:
Operational
parameters of
the
implemented
project
Design-To-Analysis Feedback Loop
Design-To-Requirements Feedback Loop
Generic Project Management Model in CIMpgr for A1 to A5 Processes
Key Contact: Paul G. Ranky, Email: cimware@earthlink.net, USA Tel: (201)493 0521
Client Ref: Company ABC Inc., Date: June 09, 2004, Version: Ver. 5.0
Detailed Project Management eBook Template Build-up.
Object/Class Inheritance: ON
Notation: Example: ‘Conception, or Conceptual, or Requirements Analysis Phase, or
Process A1’, meaning, that this is a process, typically called by one or more of the listed
names, for our purposes meaning exactly the same.
DBI_A4:
The system
integration,
implementation
and test data
are stored here
R1A4: Project
time
management
and budget
management
software, and
experienced
project
implementation
team
DBI_A5: The
post project
review data is
stored here
based on longterm
test and
maintenance
results
Post Project Review, or
Long Term System Test, or
Maintenance phase, or
Process A5
O1A5:
Test
results
O2A5:
R2A5:
R1A5:
Continuous
improvement
team
Implementation-To-Design Feedback Loop
Long Term Test/Maintenance-To-Design Feedback Loop
(c)
FIGURE 5 Continued
3.1.9 REQUIREMENTS ANALYSIS FOR SYSTEM PRODUCT, PROCESS,
AND SERVICE DESIGN INNOVATION
Processes in a successful innovation project must satisfy requirements set by the
market, the sponsors, and/or the inventor ’ s own dreams. Requirements analysis is
considered to be one of the most important features of any innovative pharmaceutical
manufacturing system project, because if done professionally, it helps to specify,
research, and develop appropriate features and processes that customers need.
In our innovation project examples we have focused on generic needs and
requirements, and our associated “ customers ” are the pharmaceutical R & D team
members, managers, and operators in various industries.
In terms of our research approach, we have followed a proven method: Analyze
the needs and the requirements, the demonstrated processes, and the methods and
systems they try to or have to satisfy, and if you fi nd a “ gap ” , you have found an
innovation opportunity. Note that when we search for this gap, it will simultaneously
appear as a missing process in our CIMpgr model or as an existing process but
missing attributes as well as a requirement in our CORA model (component -
oriented requirements analysis) model:
• Analyze the actual methods presented. Find the core methodologies, the
mathematical models, and the underlying engineering and/or other science
foundation.
• Analyze the technologies involved. (How is science turned into a practical
solution/engineering and/or computing technology?) Is there a need for a new,
novel technology that has not been invented yet or applied in this fi eld?
• Analyze and review the actual processes and the way the process fl ow is integrated.
(Follow an object - oriented process analysis method, i.e., from concept
to product.) Focus on the attributes of the processes. Note that by adding a new
attribute you create new data types, with new information, and if your process
can reason over this in a new way, new knowledge; therefore your combined
CIMpgr and UML model becomes a new knowledge representation model too.
This is important because innovation is formalized this way and can be communicated
among global teams.
• Analyze potential alternative solutions. (A pharmaceutical manufacturing/
assembly/packaging system must be very fl exible these days, due to dynamically
changing customer requirements and even operating conditions.)
• Analyze the benefi ts and the disadvantages of each process/solution.
• Design alternative methods, processes based on what you have experienced/
seen and learned.
• Design an integrated system, based on what you have analyzed in this case.
• Work in a multidisciplinary team and exchange ideas.
• Understand the boundaries as well as the tremendous potential of new ideas
and developments by working on this case (realize that in order to survive and
win, you must add value) [54 – 57] .
After this short introduction, we demonstrate our CORA spreadsheet solution
with a real - world example (Figure 6 ).
REQUIREMENTS ANALYSIS 181
182 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
Object / Component Oriented Requirements Analysis Program for
Networked Lean Manufacturing by Paul G. Ranky © 1992–2006
© Paul G. Ranky. 2000–2002
Engineering / Software Solutions
Responding to Customer
Requirements
Lean Manufacturing Manager:
Customer Requirements
( Reflecting component object
behavior related to customer needs)
S.No Describe the Requirement
Reliability of data transfer for realtime
access should be high
Reliability of reporting process failure
to the line manager’s computer
Ease of integration into a system (plug
and play networking): important!
Ease of machine programming (CNC
machining / inspection)
Ease of changng CNC part programs
(locally, and via the network)
Ease of adding new sensors to a
workstation, CNC, or cell: high
safety of operation: critical!
Cost of change/extenaion/system
expansion should be low
Operator training needs and costs
should be low
Network installation complexity and
cost should be low
1
2
3
3
3 3 3 3
3
3 3
3
4
4
4
4
4
5
5
5
6
7
8
9
9 9 9 9 3 9
9 9 9 9
9 9 9 9
9
9 9 9 9 9 9
9 9 9 9
9 9 9
9
3 9
3
3
9
3 3
3 3
3 9
9
9 9
9
9 9 9 9
9 3 3
9 3 3
9 9 9
9 9 9
9 9 9
9 3 3
9 9 9 9 9
9 9 9
9 9 9
9 3 3 10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Importance Rating (1–5)
Fieldbus Network
Profibus Network
DeviceNet Network
Ethemet
Graphical CNC Progr.
On-site maint. Support
Off-site maint. Support
On-site Redundant Server
Off-site Redundant Server
Link to Factory prod. contr.
Link to Factory TQM/TQC
2D videos/3D multimedia
Cell-based web Camera
PC-based CNC Controller
PC based workkstation
contr.
PC based Cell controller
(a)
FIGURE 6 CORA method. This is a spreadsheet - based tool, designed to analyze customer
requirements. (A “ customer ” here can mean a pharmaceutical manufacturing line vendor,
user, operators, maintenance engineers, and many others.) The key approach is that we create
a correlation matrix and then evaluate the results using a quantitative, computational
approach. This is much more accurate than just a simple structured list. Our method offers
a list of all key requirements as well as the priorities for the pharmaceutical manufacturing
system design team they should follow during the design process. (For more about this software
tool, see http://www.cimwareukandusa.com .
0
Competitor C’s product
Competitor B’s product
Competitor A’s product
Our product
0
1
2
3
4
5
6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Enter Competitor C’s ratings (1-5); Graph
Enter Competitor B’s ratings (1-5); Graph
Enter Competitor A’s ratings (1-5); Graph
Enter Our Product ratings
(1 = low, 5 = High)
Relative Importance Rating
Absolute Importance Rating
3
3 3
3 3
3 3
3 3
3
3
3 3 3
3
3
5 5 5
5
5
5 5 5 5
2 3 3 2 2 2
2 2
2 2
2 1 1 4 4
4 4
4 4 4
4
4
4
4
4 4 4 4 4 4 4
4
4 4
4 4
4 4 5
416 416 416 489 207 147 111 186 107 144 135 87 87 567 0 0 0 0
0 0 0 0 13 8.5 13 1.9 1.9 3 3.2 2.4 4.1 2.5 3.3 4.6 11 9.3 9.3 9.3
381 588
Target Values (List here the parameters
that specify engineering solutions
accurately. If you don’t know the range of
the acceptable values, use our Taguchi
Calculator Program for Designing an
Experiment)
The system history database should
be on the network 20 5 3 3 3 3 3 3 3 3 9 9 9 9 9
Within 27 m sec
Within 27 m sec
Within 27 m sec
Within 27 m sec
GUI, iconized,
multimedia
Less than 3
minutes
Less than 24 hrsr
0 sec switch
Less than 30 sec
switch
Retresh every 2
minutes
Refresh every 2
minutes
320 240 pxels
or better
320 240 pxels
or better
Win, Linux,
Solans, or OSX
Response within
12 m sec
Response within
24 m sec
4 0
(b)
3.1.10 INNOVATION RISK ANALYSIS AND OPPORTUNITY METHOD
AND TOOL WITH PHARMACEUTICAL MANUFACTURING
SYSTEM APPLICATIONS
Our failure risk analysis and opportunity method and iterative software tool, as part
of our New Product & Process Innovation (NPPI) Tool Library, promotes systematic
collaboration and team - oriented engineering thinking when a new pharmaceutical
manufacturing system process and/or product are developed. (We call it “ opportunity
method ” too, since most risks, if not all, offer new opportunities for innovation.)
It is based on our generic process failure risk analysis method that could be applied
to literally any process that involves risk — and innovation is a very risky process.
We follow a rule - based method when we analyze risk objects and components
and their attributes. These plug - and - play rules can be different for different subjects,
research areas, and industries. They can be designed and standardized for different
industry sectors, enabling an analytical approach, systematic standardization, and
accurate and predictable results.
Our risk analysis method and tools help the engineering management team to
understand some of the following concerns:
• What could go wrong with the processes involved during the innovation
project?
• How badly might it go wrong and what could the fi nancial loss be?
• Which are the highest risk processes/operations when working on the product/
process/service - related innovative design and project?
• What needs to be done to prevent failures?
• Which processes must be changed to reduce the risk of failure?
• What tools and fi xtures are required to prevent failures and reduce the risk?
• What education is needed for participants, innovators, engineers, and process
owners, such as line management and operators, to reduce or prevent
failures?
After this introduction, we show the risk analysis system components, following
the already described ISO 9000:2000 principles, and how we can model complex
project management risks using them (Figure 7 ) (note that the active code spreadsheets
and 3D objects are part of Ranky ’ s eLibrary) [50 – 55] .
3.1.11 OPEN - SOURCE COMPUTATIONAL STATISTICAL
AND THREE - DIMENSIONAL MULTIMEDIA FOR
PHARMACEUTICAL MANUFACTURING SYSTEM INNOVATION
AND PROJECT COMMUNICATION
Since we follow an analytical, quantitative, and open - source computational approach,
our pharmaceutical product/process and project management method and software
toolset are implemented as (Internet browser readable) MS - Excel spreadsheets,
integrated with several hyperlinks to the rule base and to optional 2D video and
3D virtual - reality objects for visualization.
OPEN-SOURCE COMPUTATIONAL STATISTICAL 183
184 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
FIGURE 7 The process failure risk analysis (PFRA) tool is an analytical and computational
tool using rule bases for evaluating process risks. It is an ideal method and tool for reducing
costly failures. (For more about this software tool, see http://www.cimwareukandusa.com .
Rev.2.1.3. by Ranky
9/19/01
11/16/01
Ranky 111601/DFRA_Ver.5
List/Identify the Parts/Components Retrieved
in Each Disassembly Process Step
Painted metal PC cover (File: 3DMetalCover.
mov)
Floppy drive, hard drive, mounted in a solid
sheet metal bracket inside the PC
(File: 3DFloppyHDassy. mov)
Floppy drive (File: 3DFloppyDrive. mov)
Hard drive (File: 3DHDobj. mov)
Metal bracket holding the floppy and the hard
drive (File: 3DFloppyHDBracket. mov)
Disassembly Process Code
Engineering Rerease Date or Process Methodology
Type of Product Disassembled
Product Group Classifier
Engineering Release Date of the Product
Process
Time
Process
Cost
Accumulated
Process Cost
Ranky PC DisassyCode: 05/07/97
5/7/97
Electro-mechanical
Desktop PC
Estimated: 1993
The DFRA Team Describes/Illustrates the Potential Disassembly
Failure Mode and the Effect; the Risk of Failure
Failure Mode(s) and Effect(s)
Metal Cover scratched by slipped screwdriver
As PC Metal Cover is removed, internal parts are cratched
Floppy drive assy. screw removal can damage nother board
Movie, illustrating floppy/HD assy. removal risks
Can damage Floppy Drive if assy. Is dropped
Can damage Hard Drive if assembly is dropped
Can damage Hard Drive if assembly is dropped
Proc. ID
ID 5.1
ID 5.2
ID 5.3
ID 4.1
ID 4.2
ID 4.3
ID 3.1
ID 3.2
ID 3.3
ID 2.1
ID 2.2
ID 2.3
ID 1.1
ID 1.2
ID 1.3
[sec] [USD] [USD]
16.40
45 0.21
137 0.62
35 0.16
65 0.30
12 0.05
0.21
0.83
0.99
0.28
1.34
(a)
Ranky PC DisassyCode: 05/07/97
5/7/97 ss
Electro-mechanical
Desktop PC
Estimated: 1993 oduct
ed
ost
This DFRA Study Prepared By
DFRA Team
Responsible Organization/Department
Paul
NJIT
NJIT
RPN
Pr
Nu
Comments
The DFRA Team Describes/Illustrates the Potential Disassembly
Failure Mode and the Effect; the Risk of Failure
Proc.ID
ID 1.1
ID 1.2
ID 1.3
ID 2.1
ID 2.2
ID 2.3
ID 3.1
ID 3.2
ID 3.3
ID 4.1
ID 4.2
ID 4.3
ID 5.1
ID 5.2
ID 5.3
Metal Cover scratched by slipped screwdriver
As PC Metal Cover is removed, internal parts are cratched
Floppy drive assy, screw removal can damage mother board
Movie, illustrating floppy/HD assy. removal risks
Can damage Floppy Drive if assy. Is dropped
Can damage Hard Drive if assembly is dropped
Can damage Hard Drive if assembly is dropped
Severity
Rating
Detection
Rating
Occurrence
Rating
(1-10) (1-10) (1-10)
3
3
3
3
5
5
4
2
2
2
2
2
9
9
8
8
1
1
Failure Mode(s) and Effect(s)
(b)
The reason for this is because we would like to offer our users the opportunity
not just to understand the method and the coded logic, but also to be able to enjoy
the 3D interactive graphics, the digital videos, the color images, and most importantly
the active code spreadsheets. Along with any other imaginable visualization,
this can be executed and experimented with using their own data.
In terms of statistical methods, our NPPI Tool Library has several statistical
analysis tools to capture innovation opportunities at processes that are likely to drift
and become out of control or processes that execute with random failure.
This DFRA Study Prepared By
OFRA Team
Responsible Organization/Department
Comments
Paul G Ranky NJIT/MERC
NJIT/MERC CFRA Team
NJIT/MERC
Severity
Rating
Detection
Rating
Occurrence
Rating
RPN (Risk
Priority
Number)
Max. RPN Tooling
Factor
Clamping/
Fixturing
Factor
Skill Factor
Any Other
Factor You
Define
Accumulated
RPN
Risk
Associated
(1-10) (1-10) (1-10) 0.1-2.1=100% 0.1-2.1=100% 0.1-2.1=100% 0.1-2.1=100%
3
3
3
3
2
2
2
2
2
5
5
8
8
9
9
4 1
1
18
20 20
0
0
0
0
0
0
0
0
0
0
90
90
16
48 48
54
54
1.40
1.40
1.40
1.40
1.40
1.40
1.60
1.20 1.20
1.20
1.20
1.00 1.00
1.00
1.00
1.00
1.00 1.00 1.00 1.00
33.60
282.24
96.77
127.01
0.00
Low
Low
Low
Low
HIGH
Mc
Ha
(c)
(d)
FIGURE 7 Continued
OPEN-SOURCE COMPUTATIONAL STATISTICAL 185
186 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
For capturing such critical opportunities for innovation and process improvement,
we use a range of control charts for drifting data analysis, Taguchi DOE
(design of experiments) methods for developing the desired list of parameters for
our engineering solutions in our requirements analysis method, and Weibull methods
for process reliability analysis. As we progress, we plan to introduce further statistical
and other tools to our NPPI Tool Library [56 – 70] .
3.1.12 RFID APPLICATIONS
Radio - frequency identifi cation (RFID) technologies are being adopted in the
United States at a fast pace in pharmaceutical/assembly and packaging, in general
manufacturing, warehousing, distribution, and global supply chain management. The
market size for this technology is expected to rise from around $ 500 million in 2005
to about $ 4 billion in 2010. In this section we outline some of the main application
areas with a focus on the pharmaceutical applications.
We also deal with the R & D opportunities and some digital pharmaceutical
manufacturing systems with RFID information system modeling results . Furthermore
we offer a generic factory assembly and tracking digital model for RFID
integration, the most complicated task manufacturing systems engineers, industrial
engineers, and IT experts have faced due to the mixed real - time as well as global
traceability and messaging challenges one faces with RFID - tagged parts and shipments.
RFID opportunities are great since with the appropriate IT infrastructure
they help both major distributors and manufacturers as well as other logistics operations,
such as in the health care system, defense industries, and others, dealing with
complex, global supply chains in which products and product shipments must be
traced and identifi ed in a noncontact, wireless fashion using a computer network,
because of cost, security, or safety or because parts are subject to corrosion or medicine
is subject to quality degradation.
All of these requirements point to an automated, wireless - readable sensory -
based identifi cation method and network that offers more functionalities and is
signifi cantly “ smarter ” than the well - known bar code or the unifi ed product code
(UPC).
RFIDs are available as passive and/or active radio read/write sensor packages
with active read (and often write) capabilities in relatively large areas (e.g., a large
distribution center warehouse or a containership), all performed automatically,
supervised by computers, and communicated in a wireless fashion over secure
intranets. The attraction to a pharmaceutical assembly factory or a supply chain
manager is that when the RFID network is integrated with the factories ’ material
resource IT management systems, accurate information can be obtained on all
tagged parts in close to real time throughout the entire supply chain. This can
include the globally distributed factories as well as information about parts and
assemblies during shipment, including in transit. This is why RFID represents great
research and technology as well as huge business opportunities.
We introduce here some of the most important engineering and information
systems management principles and challenges that RFID researchers, implementers,
and users should keep in mind when developing such systems and/or planning
for such applications as well as offer an RFID digital factory integration model in
UML [60 – 64, 70, 73] .
3.1.13 RFID EXAMPLES
To set the scene, consider a large storage house for a variety of medications or their
distribution center with thousands of boxes, parts, and assemblies that range from
low cost to high value, on occasion even highly sensitive technology or perishable
drugs that must be kept in certain environmental conditions, such as temperature,
humidity, or pressure, for the entire period of the shipment and/or production/packaging
operations.
Just - in - time delivery in an environment like this means that in order to build an
order with variety or a new combination of treatments in a medical drug, every
component must be in place on time and in good condition, which is a very diffi cult
criterion to satisfy.
Obviously supply chains are global these days, and shipments are typically made
by a variety of means, including cargo ships, air, rail, and trucks; all of these can be
late or can get in trouble because of the weather, traffi c, industrial disputes, or other
reasons. Supply chain systems like this are very complex, because of the uncertainties
in deliveries, parts and shipments are lost and/or stolen, goods get damaged
during shipment, or the number of international ports and customs often take
unpredictable time to check shipments with different levels of safety/security, and
many other reasons.
There are many valid reasons why wireless, computer - networked, sensory - based
part identifi cation methods, tools, and technologies are being researched and
deployed in industry. The application fi elds and opportunities are vast. The key
driver is that even in chaotic, largely distributed, more stochastic than deterministic
business environments, adaptive organizations and enterprises must react to
demands quickly, else a competitor will take the business. Therefore they must
reduce waste and improve effi ciency at all fronts. The most important aspect of this
strategy is to know exactly what parts they have in stock, exactly where these parts
are, and in what condition/state of assembly or preparedness they are. Furthermore,
major distributors dealing with complex, global supply chains must be able to trace
their shipments in detail because of cost, security, safety, quality degradation (as in
the case of temperature - , humidity - , and/or shock - sensitive components or drugs),
or other reasons.
RFID technologies with the appropriate IT infrastructure help major distributors
and manufacturers as well as other logistics operations such as the health care
system, defense industries, and others deal with complex, global supply chains in
which products and product shipments must be traced and identifi ed in a noncontact,
wireless fashion using a computer network.
All of the above - listed requirements point to an automated, wireless - readable,
sensory - based identifi cation method and network that offer more functionalities,
and are signifi cantly “ smarter ” than the well - known bar code or the UPC — hence
the new popularity of RFID technology.
RFID tags carry a serialized tag data construct. As an example, a 64 - bit class 0
tag offered by a supplier includes 64 bits of total user memory on the tag itself,
RFID EXAMPLES 187
188 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
including a unique serial number. This number is encoded by the manufacturer and
uniquely identifi es up to 264 = 18,446,744,073,709,551,616 tagged items.
RFIDs are available as passive and/or active radio read/write sensor packages
with active read (and often write) capabilities in relatively large areas (like a large
distribution center warehouse or a containership), all performed automatically,
supervised by computers, and communicated in a wireless fashion. The attraction to
an assembly factory or a supply chain manager here is that when the RFID network
is integrated with the factories ’ material resource IT management systems, accurate
information can be obtained on all tagged parts at close to real time, all throughout
the entire supply chain. This can include the globally distributed factories as well as
information about parts and assemblies in shipment/in transit.
This is why RFID represents excellent research, technology, as well as big business
opportunities. To illustrate this, consider research challenges such as remotely
scanning and tracing products and parts in boxes on a cargo ship as it approaches
national waters from international waters; tracing parts that are subject to corrosion
and being used in agricultural or military equipment; medical drugs that are being
counterfeited and repackaged and then shipped and imported illegally; or laptop
computers that are dropped and damaged by accident. As a clear sign of the business
opportunity, consider that according to a U.S. Department of Defense published
presentation, RFID - enabled supply chain savings reached over U.S. $ 460
million in 2004 and the projections for 2010 are in excess of $ 4 billion!
3.1.14 RFID SYSTEM INTEGRATION MODELS FOR DIGITAL
PHARMACEUTICAL MANUFACTURING AND ASSEMBLY
SUPPLY CHAINS
In the U.S. manufacturing and assembly industry, many of the RFID pilot projects
focus on achieving 100% read rates at speeds set by the widely used bar code technology.
The focus for these projects is to achieve proper tag placement on cases and
pallets as well as the proper confi guration of pallets to enable 100% RFID tag
read rates. This is a huge issue in pharmaceutical manufacturing and assembly
in the fi ght to eliminate fake products and packages reaching the market! (See
Figures 8 – 15 .)
Based on the above - described requirements analysis, network planning, and
server balancing reasoning, for our purposes, we have decided to follow a simple
but powerful network architecture. In this architecture, we have included subnetworks.
In terms of the way OPNET IT - Guru handles subnetworks, a subnetwork
contains other network objects and abstracts them into a single object. A subnetwork
can encompass a set of nodes and links to represent a physical grouping of
objects (this can be a local - area network of CNC machines or robot PC controllers)
or it can contain other subnetworks (e.g., including the material - handling system
control of the line) [32, 34, 63, 66, 69 – 77] .
Subnetworks within other subnetworks form the hierarchy of the network model.
This hierarchy can then be extended as required to model the structure of the
network. A subnetwork is considered the parent of the objects inside of it, and
the objects are the children of the subnetwork. The highest level subnetwork in the
network hierarchy does not have a parent, and therefore it is the top subnetwork,
FIGURE 8 UML model segment illustrating the way the stock fi le is integrated with the
routing and tooling fi les, assuming that all parts and all tools are RFID tagged. UML models
like this should be used prior to any implementation work to assess requirements, technology
needs, and RFID integration challenges with the rest of the factory ’ s IT infrastructure.
FIGURE 9 Simulation network for distributed pharmaceutical manufacturing systems and
their warehouses in U.S., Europe, India, and Asia. Model focuses on information and data
management, the way the servers can cope with the task of tracking pharmaceutical product,
and RFID data on a world wide basis. As a modeling tool we use OPNET, a professional
network simulation tool.
RFID SYSTEM INTEGRATION MODELS FOR DIGITAL PHARMACEUTICAL 189
FIGURE 10 Segment of simulation model illustrating corporate headquarters in New York.
This is where we have our main servers in our distributed system. Modeling tool is OPNET.
FIGURE 11 Segment of simulation model illustrating New Delhi campus network. This is
where the business process outsourcing team and related servers in our distributed system
are located. Modeling tool is OPNET.
FIGURE 12 Pharmaceutical company portals as a wireless network of a pharmaceutical
manufacturing system. The power of the model is that we can simulate a shop - fl oor request,
comment, or warning throughout the entire international network of globally distributed
pharmaceutical companies, with all important functions and processes. This means that before
any pharmaceutical manufacturing system is actually built, we can simulate the entire system
in the digital domain, saving huge expense and time. Modeling tool is OPNET.
FIGURE 13 Simulation diagram illustrating and confi rming that the network system design
from an ATM variation – response time point of view can cope with the demand. Modeling
tool is OPNET.
192 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
or global subnetwork. Subnetworks can be created and interconnected within this
top level or within other subnetworks. Subnetworks provide a powerful mechanism
for manipulating complex networks by breaking down the system ’ s complexity
through abstraction.
Since in our pharmaceutical network simulation models we deal with packets, let
us explain a few aspects of packet formats. Packets carry information and can be
sent between transmitters and receivers. In our example, packets can carry robot
programs when uploaded from the design/programming offi ce servers to the robot
lines and then to the individual CNCs, or robots, or parts of them if there is a need
for an update, edit, quality control, production control, maintenance, and other data.
(Packets can include mission - critical, “ panic ” related real - time data between the
robot controller PCs and the line servers.)
Packets are data structures consisting of storage areas called fi elds and can either
be formatted or unformatted. Formatted packets have fi elds designed according to
a packet format which specifi es the packets ’ fi eld names, data types, sizes, and default
values. Formatted packets can be read by corresponding communication protocols
only. Unformatted packets have no predefi ned fi elds. In IT - Guru, packet formats
are predefi ned and typically named according to the model in which they are
intended to be used.
FIGURE 14 Simulation diagram illustrating and confi rming that the server balancing
aspects of the network system design can cope with the demand. Modeling tool is OPNET.
3.1.15 EVALUATION OF NETWORK SIMULATION RESULTS
The goal of most simulation scenarios is to evaluate some aspect of a system ’ s
behavior or performance and to quantify, typically in terms of statistics, the results
and then use the results for decisions. This requires a simulation environment with
software tools that provide insight into a model ’ s dynamic operation.
Based on IT - Guru ’ s in - depth analysis, the pharmaceutical manufacturing system
network engineering analyst can collect object, scenariowide, and global statistics
as follows:
• Object statistics are collected from individual objects. They allow the network
engineering analyst to evaluate the performance of specifi c network nodes or
links (a single hub ’ s Ethernet delay or a server balancing change, as in our
example).
• Scenariowide object statistics are collected from all relevant objects in a network
(e.g., Ethernet delay for every node). They allow the network engineering
analyst to easily monitor the performance of all objects of a specifi c type.
EVALUATION OF NETWORK SIMULATION RESULTS 193
FIGURE 15 Simulation diagram illustrating and confi rming that the network system design
from an object variation – response time point of view can cope with the demand. As an
example, this is important if a pharmaceutical manufacturing system line manager in India
wants to notify a manager in New York by sending an image object, a sound object, or a
multimedia object of a machine in the line for quality evaluation. Modeling tool is OPNET.
194 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
• Global statistics are collected from the entire network. They represent results
that apply to the network as a whole (such as global end - to - end delay) and let
the designers and management analyze aspects of the network ’ s overall
performance.
• More specifi cally, IT - Guru offers the following types of statistics when analyzing
networks:
Queue size
Available space
Overfl ow occurrences
Delay
Interarrival times
Packet sizes
Throughput
Utilization
Error rates
Collisions
Application - specifi c statistics defi ned by a model developer
Because there are many possible statistics to collect, the data fi les would quickly
grow past practical use if the simulation program recorded them all. Therefore, the
analyst must specifi cally select the statistics that are valuable for the particular study
before running a simulation [71 – 79].
3.1.16 SUMMARY
In this chapter, we have presented the foundations of an analytical and simultaneously
computational lean and fl exible pharmaceutical manufacturing system design
approach based on total quality standards. We have discussed why this approach is
essential for pharmaceutical product, process, and manufacturing system designs.
As illustrated, based on simulation results, using the plotted graphs and screens,
management can easily evaluate different design alternatives, machine and human
behavior models, control systems, sensory feedback processing, and the need of a
balanced server architecture, and even investigate “ what if ” scenarios further,
without committing to major upfront investment.
We can clearly state that the time has come when pharmaceutical manufacturing
systems can be designed and built in an entirely digital domain, saving huge amounts
of capital and other related cost, and simultaneously increasing quality.
3.1.17 COMPLIMENTARY VIDEO ON DVD
To show real - world high - technology examples of pharmaceutical product, process,
manufacturing, assembly, and packaging system designs, in action, something we
cannot do in static, printed books, we have created a supplementary video, in high
defi nition, and compressed onto a DVD. This professionally edited DVD supports
0
0
0
0
0
0
0
0
0
0
0
this chapter as an independent, self - contained publication illustrating advanced
pharmaceutical and medical product, process, and manufacturing system designs,
related quality assurance processes and solutions, and others, explained by industry
experts. To fi nd out more about this DVD, refer to Ranky, P. G., Ranky, G. N., and
Ranky, R. G. (2006), Design Principles and Examples of Pharmaceutical Manufacturing
Systems (Product, Process, Lean and Flexible Manufacturing, Assembly and
Packaging System Designs) , Video on DVD, available: www.cimwareukandusa.
com .
REFERENCES
1. Ranky , P. G. , A generic, analytical method to assess process - related risk with case studies,
The Project Management Institute (PMI) Risk SIG and the Institute for International
Research (IIR), paper presented at the Annual US National Project Risk Symposium,
Houston, TX, May 22 – May 26, 2006.
2. Ashley , S. ( 2004 ), Penny - wise smart labels , Sci. Am . 291 ( 2 ), 30 – 31 .
3. Ashton , P. , and Ranky , P. G. , ( 1999, Feb. ), An advanced concurrent engineering research
toolset and its application at Rolls - Royce motor cars, ADAM (Adv. Des. Manufacturing),
available: http://www.cimwareukandusa.com , listed and indexed by the Association of
Research Libraries , Washington, DC, and the Edinburgh Engineering Virtual Library,
United Kingdom, Vol. 1.
4. Bradbrook , R. ( 2004 ), Wal - Mart and RFID, folding carton industry , Printing News , 31 ( 4 ),
30 – 33 .
5. Bradbrook , R. ( 2004 ), Procter and Gamble aim to be among the world leaders in RFID
implementation , Int. Paper Board Ind ., 47 ( 8 ), 20 – 23 .
6. Brzozowski , C. ( 2004 ), Tags, tickets & labels: New technologies emerge , Printing News ,
153 ( 3 ).
7. Ranky , P. G. , An integrated PM approach, including: Process modeling, requirements
analysis, risk analysis, statistical tools and 3D multimedia, presented at the Project Management
Institute (PMI), New Jersey Chapter. Jan. 2006.
8. Ranky , P. G. , Focus on RFID (radio frequency identifi cation) methods, technologies and
education, presented as part of the NCME Mission (National Center for Manufacturing
Education), sponsored by NSF (National Science Foundation, USA) and industry, Jan.
2006.
9. Flaherty , M. , Ranky , P. G. , Ranky , M. F. , Sands , S. , and Stratful , S. ( 1999, Mar. ), An
engineering multimedia approach to servo pneumatic positioning, ADAM (Adv. Des.
Manufacturing), available: http://www.cimwareukandusa.com , listed and indexed by the
Association of Research Libraries, Washington DC, and the Edinburgh Engineering
Virtual Library, United Kingdom, Vol. 1.
10. Glidden , R. , Bockorick , C. , etal . ( 2004 ), Design of ultra - low - cost UHF RFID tags for
supply chain applications , IEEE Commun. Mag . 42 ( 8 ), 140 – 151 .
11. Graham-Rowe , D. (2004), Tags to banish forgetfulness , New Scientist , 183 ( 2460 ), 19 .
12. Graham - Rowe , D. ( 2004 ), Who ’ s keeping tabs on your tags? New Scientist , 183 ( 2462 ),
22 .
13. Ho , K. L. , and Ranky , P. G. ( 1999, Mar. ), An object oriented approach to fl exible
conveyor system design, ADAM (Adv. Des. Manufacturing), available: http://www.
cimwareukandusa.com , listed and indexed by the Association of Research Libraries, Washington
DC, and the Edinburgh Engineering Virtual Library, United Kingdom, Vol. 1.
REFERENCES 195
196 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
14. Knights , P. F. , Henderson , E. , and Daneshmend , L. K. ( 2004 ), Drawpoint control using
radio frequency identifi cation systems , CIM Bull ., 89 ( 1003 ), 53 – 58 .
15. Loose , D. C. , and Ranky , P. G. ( 2007 ), A Case - Based Introduction to IBM ’ s Telematics
Solutions , interactive multimedia eBook with 3D objects, text, and videos in a browser -
readable format on CD - ROM/intranet, available: http://www.cimwareukandusa.com ,
CIMware USA, Inc., and CIMware Ltd., United Kingdom; multimedia design and programming
by P. G. Ranky and M. F. Ranky.
16. Mongeon , D. G. ( 2005, Feb. ), RFID: A tool for the 21st century distribution, USA Department
of Defense Conference Presentation, RFID Media Briefi ng, Washington DC.
17. Nadler , S. F. , Ranky , P. G. , and Ranky , M. ( 2002 – 2003 ), A 3D multimedia approach to the
diagnosis of low back pain (Vol. 1, 18 - and 40 - year - old males), interactive 3D multimedia
presentation on CD - ROM with off - line Internet support (650 Mbytes, approx. 150 interactive
screens, 50 minutes of digital videos, 3D internal and external body tour, animation,
and 3DVR objects), by CIMware (IEE and IMechE Approved Professional Developer);
also in Multimedia design and programming by P. G. Ranky and M. F. Ranky.
18. Ranky , G. N. , and Ranky , P.G. ( 2005 ), Japanese service robot R & D trends and examples ,
Ind. Robot , 32 ( 6 ), 460 – 467 .
19. Ranky , P. G. , Caudill , R. J. , Limaye , K. , Alli , N. , ChamyVelumani , S. , Bhatia , A. , and
Lonkar , M. ( 2002, May ), A Web - enabled virtual disassembly manager (web - VDM) for
electronic product/process designers, disassembly line managers and operators, a UML
(Unifi ed Modeling Language) model of our generic digital factory, and some of our
electronic support system analysis tools, ADAM with IT (Adv. Des. Manufacturing),
available: http://www.cimwareukandusa.com , listed and indexed by the Association of
Research Libraries, Washington DC, and the Edinburgh Engineering Virtual Library,
United Kingdom, Vol. 3.
20. Ranky , P. G. ( 2006 ), Introduction to RFID — Radio frequency identifi cation methods and
solutions , Assembly Automation , 26 ( 1 ), 28 – 33 .
21. Ranky , P. G. ( 1999, Apr. ), New trends in fl exible, lean and agile manufacturing cells and
systems, ADAM (Adv. Des. Manufacturing) , available: http://www.cimwareukandusa.com ,
listed and indexed by the Association of Research Libraries, Washington DC, and the
Edinburgh Engineering Virtual Library, United Kingdom, Vol. 1.
22. Ranky , P. G. ( 2002 ), A method for planning industrial robot networks for automotive
welding and assembly lines , Ind. Robot: Int. J ., 29 ( 6 ), 530 – 537 .
23. Ranky , P. G. ( 2003 ), A real - time manufacturing/assembly system performance evaluation
and control model with integrated sensory feedback processing and visualization , Assembly
Automation .
24. Ranky , P. G. ( 2003 ), A simulation method and distributed server balancing results of networked
industrial robots for automotive welding and assembly lines , Ind. Robot: Int. J .,
30 ( 2 ), 192 – 197 .
25. Ranky , P. G. ( 2002 ), Advanced digital automotive sensor applications , Sensor Rev.: Int. J .,
22 ( 3 ), 213 – 217 .
26. Ranky , P. G. ( 2003 ), Advanced machine vision systems and application examples , Sensor
Rev.: Int. J ., 23 ( 3 ), 242 – 245 .
27. Ranky , P. G. ( 2003 ), Collaborative synchronous robots serving machines and cells , Ind.
Robot: Int. J ., 30 ( 3 ), 213 – 217 .
28. Ranky , P. G. ( 2004 ), Digital, Internet - enabled assembly line and factory modeling , Assembly
Automation , 24 ( 3 ), 247 – 253 .
29. Ranky , P. G. ( 2004 ), Novel automated inspection methods, tools and technologies , Assembly
Automation: Int. J ., 23 ( 3 ), 252 – 257 .
30. Ranky , P. G. ( 2003 ), Reconfi gurable robot tool designs and integration applications , Ind.
Robot: Int. J ., 30 ( 4 ), 338 – 344 .
31. Ranky , P. G. ( 2002 ), Smart sensors , Sensor Rev.: Int. J ., 22 ( 4 ), 312 – 318 .
32. Ranky , P. G. ( 1999, Jan. ,) Some generic algorithmic solutions to the problem of dynamic
scheduling in fl exible manufacturing systems that operate globally, ADAM (Adv. Des.
Manufacturing), available: http://www.cimwareukandusa.com , listed and indexed by the
Association of Research Libraries, Washington DC, and the Edinburgh Engineering
Virtual Library, United Kingdom, Vol. 1.
33. Ranky , P. G. ( 2001 ), Trends and R & D in virtual and robotized product disassembly , Ind.
Robot , 28 ( 6 ), 454 – 456 .
34. Ranky , P. G. ( 2000, May ), Some analytical considerations of engineering multimedia
system design within an object oriented architecture , Int. J. CIM , 13 ( 2 ), 204 – 214 .
35. Ranky , P. G. , and ChamyVelumani , S. ( 2003 ), A method, a tool (CORA), and application
examples for analyzing disassembly user interface design criteria , Int. J. CIM , 16 ( 4 – 5 ),
317 – 325 .
36. Ranky , P. G. , and ChamyVelumani , S. ( 2003 ), An analytical approach, a tool (DFRA) and
application examples for assessing process - related failure risks , Int. J. CIM , 16 ( 4 – 5 ),
326 – 333 .
37. Ranky , P. G. , and Nadler , S. F. , A novel multimedia approach to low back pain diagnosis
with internal and external 3D interactive body tours, paper presented at the 29th Annual
Northeast Bioengineering Conference, New Jersey Institute of Technology, University
Heights, Newark, NJ, Mar. 2003.
38. Ranky , P. G. , and Nadler , S. F. , A new, Web-enabled multimedia approach with 3D virtual
reality internal and external body tours to support low back pain diagnosis, paper presented
at the 4th Annual Faculty Best Practices Showcase in Kean University, NJ, Mar. 2003.
39. Ranky , P. G. , and Ranky , M. F. ( 2000 ), A Dynamic Operation control algorithm with
multimedia objects for fl exible manufacturing systems , Int. J. CIM , 13 ( 2 ), 245 – 263 .
40. Ranky , P. G. , Lonkar , M. , and ChamyVelumani , S. ( 2003 ), eTransition models of collaborating
design and manufacturing enterprises , Int. J. CIM , 16 ( 4 – 5 ), 255 – 266 .
41. Ranky , P. G. , Morales , C. , and Caudill , R. J. , Lean Disassembly line layout, process and
network simulation models and cases, based on real - world data, paper presented at the
IEEE (USA) International Symposium on Electronics and the Environment and the
IAER Electronics Recycling Summit, Boston, MA, May 19 – 22, 2003.
42. Ranky , P. G. , Ranky , G. N. , and Ranky , R.G. ( 2006 ), Examples of pharmaceutical product/
process/manufacturing/assembly and packaging system designs, video on DVD, available:
www.cimwareukandusa.com .
43. Ranky , P. G. , Subramanyam , M. , Caudill , R. J. , Limaye , K. , and Alli , N. , A dynamic scheduling
and balancing method and software tool for lean and reconfi gurable disassembly
lines, paper presented at the IEEE (USA) International Symposium on Electronics and
the Environment and the IAER Electronics Recycling Summit, Boston, MA, May 19 – 22,
2003.
44. Ranky , P. G. , 3D engineering multimedia cases. A customizable 3D Web - enabled library
with reusable objects, paper presented at the ASEE (American Society of Engineering
Educators) Mid - Atlantic Conference, Kean University, NJ, Apr. 2003.
45. Ranky , P. G. , A 3D multimedia approach to biomedical engineering: Low back analysis,
paper presented at the ASEE, American Society of Engineering Educators, U.S. National
Meeting, Biomedical Engineering Division, Nashville, TN, June 2003.
46. Ranky , P. G. , Ranky , G. N. , and Ranky , R. G. ( 2006 ), Design principles and examples of
pharmaceutical manufacturing systems (product, process, lean & fl exible manufacturing,
REFERENCES 197
198 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
assembly and packaging system designs), video on DVD, available: www.cimwareukandusa.
com .
47. Ranky , P. G. ( 2001 – 2006 ), A 3D multimedia case: Component oriented disassembly failure
risk analysis, an interactive multimedia publication with 3D objects, text and videos in a
browser - readable format on CD - ROM/intranet available: http://www.cimwareukandusa.
com , CIMware USA, Inc., and CIMware Ltd., United Kingdom; Multimedia design and
programming by P. G. Ranky and M. F. Ranky (published 6 volumes of this main title
with different risk analysis challenges explained).
48. Ranky , P. G. ( 2001 – 2005 ), A 3D multimedia case: component oriented disassembly user
requirements analysis, an interactive multimedia eBook publication with 3D objects, text
and videos in a browser - readable format on CD - ROM/intranet available: http://www.
cimwareukandusa.com , CIMware USA, Inc., and CIMware Ltd., United Kingdom, Multimedia
design and programming by P. G. Ranky and M. F. Ranky (published 7 volumes
of this main title with different requirements analysis challenges explained).
49. Ranky , P. G. , A 3D Web collaborative concurrent automotive engineering Method Based
on our “ distributed digital factory ” and “ digital car ” models, paper presented at the
Society of Automotive Engineers World Congress, Detroit, MI, Mar. 2003.
50. Ranky , P. G. ( 2003 ), A 3D Web - enabled, case based learning architecture and knowledge
documentation method for engineering, information technology, management, and
medical science/biomedical engineering , Int. J. CIM , 16 ( 4 – 5 ). 346 – 356 .
51. Ranky , P. G. , A Biomedical Engineering case with 3D lower back interactive virtual
anatomy tours inside and outside the human body with automated post - test student
assessment, paper presented at the ASEE (American Society of Engineering Educators)
Mid - Atlantic Conference, Kean University, NJ, Apr. 2003.
52. Ranky , P. G. , A new approach for teaching and learning about engineering process failure
risk analysis with IE (industrial engineering) case studies, paper presented at the ASEE,
American Society of Engineering Educators, US National Meeting, Industrial Engineering
Division, Nashville, TN, June 2003.
53. Ranky , P. G. , A novel 3D Internet - based multimedia method for teaching and learning
about engineering management requirements analysis, paper presented at the ASEE,
American Society of Engineering Educators, US National Meeting, Engineering Management
Education Division, Nashville, TN, June 2003.
54. Ranky , P. G. , An interactive 3D multimedia problem - based library for manufacturing
engineering technology education with Internet support, paper presented at the ASEE,
American Society of Engineering Educators, US National Meeting, Engineering Technology
Division, Nashville, TN, June 2003.
55. Ranky , P. G. ( 2003 – 2005 ), An introduction to alternative energy sources: Hybrid & fuel
cell vehicles; an interactive multimedia eBook publication with 3D objects, text, and
videos in a browser - readable format on CD - ROM/intranet, available: http://www.
cimwareukandusa.com , CIMware USA, Inc., and CIMware Ltd.; United Kingdom, Multimedia
design and programming by P. G. Ranky and M. F. Ranky, (2003 – 2005), Customer
needs, wants & requirements analysis: Automotive exterior rearview mirror, an interactive
multimedia eBook publication with 3D objects, text, and videos in a browser - readable
format on CD - ROM/intranet, available: http://www.cimwareukandusa.com , CIMware
USA, Inc., and CIMware Ltd., United Kingdom.
56. Ranky , P. G. ( 2003 – 2005 ), An introduction to digital factory & digital telematic car modeling
with R & D and industrial case studies, an interactive multimedia eBook publication
with 3D objects, text, and videos in a browser - readable format on CD - ROM/intranet,
available: http://www.cimwareukandusa.com , CIMware USA, Inc. and CIMware Ltd.,
United Kingdom, Multimedia design and programming by P. G. Ranky and M. F. Ranky.
57. Ranky , P. G. (2005), An introduction to RFID, radio frequency identifi cation methods and
applications, DVD video, available: www.cimwareukandusa.com (approximately 30 min).
58. Ranky , P. G. ( 2005 – 2006 ), An introduction to RFID, radio frequency identifi cation
methods and applications with a total quality management and control focus, interactive
browser - readable 3D eBook, available: www.cimwareukandusa.com , (approximately
30 min).
59. Ranky , P. G. ( 1999, Apr. ), An object oriented system analysis and design method (CIMpgr)
and an R & D case study, Adv. Des. Manufacturing, available: http://www.cimwareukan
dusa.com , listed and indexed by the Association of Research Libraries, Washington DC,
and the Edinburgh Engineering Virtual Library, United Kingdom, Vol. 1.
60. Ranky , P. G. , Computerized engineering assessment method based on 3D interactive
multimedia, That students enjoy, paper presented at the ASEE, American Society of
Engineering Educators, US National Meeting, Continuing Professional Development
Division, Nashville, TN, June 2003.
61. Ranky , P. G. ( 1999 ), Design, manufacturing and assembly automation trends and strategies
in China , Assembly Automation , 19 ( 4 ), 301 – 305 .
62. Ranky , P. G. ( 2003 , Feb.), Designing a lean infrastructure; advanced machining cell design
concepts, methods, architectures and cases , Manuf. Eng., J. IEE , 22 – 24 .
63. Ranky , P. G. ( 2000 ), Engineering multimedia in CIM (computer integrated manufacturing)
, Int. J. CIM , 13 ( 2 ), 169 – 171 .
64. Ranky , P. G. ( 2003 ), eTransition in the multi - lifecycle CIM (computer integrated manufacturing)
context , Int. J. CIM , 16 ( 4 – 5 ), 229 – 234 .
65. Ranky , P. G. , Interactive 3D multimedia cases for engineering education with Internet
support, ASEE, American Society of Engineering Educators, paper presented at the
U.S. National Meeting, Computers in Education Division, Nashville, TN, June 2003.
66. Ranky , P. G. , Interactive 3D multimedia cases for manufacturing engineering education
with Internet support, paper presented at the ASEE, American Society of Engineering
Educators, US National Meeting, Manufacturing Engineering Education Division,
Nashville, TN, June 2003.
67. Ranky , P. G. ( 2002, Dec. ), Introduction to concurrent engineering, an NSF (National
Science Foundation, USA) sponsored Gateway Coalition streamed multimedia narrated
web presentation, New Jersey Intitute of Technology, Public Research University, Newark,
New Jersey .
68. Ranky , P. G. ( 2003 – 2005 ), Key R & D and eTransition trends in US and international collaborative
design & manufacturing enterprises, an interactive multimedia eBook publication
with 3D objects, text, and videos in a browser - readable format on CD - ROM/intranet,
available: http://www.cimwareukandusa.com , CIMware USA, Inc., and CIMware Ltd.,
United Kingdom; Multimedia design and programming by P. G. Ranky and M. F. Ranky .
69. Ranky , P. G. ( 2000, Jan. ), Modular fi eldbus designs and applications , Assembly Automation
, 20 ( 1 ), 40 – 45 .
70. Ranky , P. G. ( 2003 ), Network simulation models of lean manufacturing systems in digital
factories and an intranet server balancing algorithm , Int. J. CIM , 16 ( 4 – 5 ), 267 – 282 .
71. Ranky , P. G. , Rapid prototyping cases for integrated design and manufacturing engineering
education with 3D Internet support, paper presented at the ASEE, American Society
of Engineering Educators, US National Meeting, Design in Engineering Education Division,
Nashville, TN, June 2003.
72. Roman , H. T. , and Ranky , P. G. ( 2003 – 2005 ), A case - based Introduction to Service robotics,
an interactive multimedia eBook publication with 3D objects, text, and videos in a
browser - readable format on CD - ROM/intranet, available: http://www.cimwareukandusa.
REFERENCES 199
200 ANALYTICAL AND COMPUTATIONAL METHODS AND EXAMPLES
com , CIMware USA, Inc., and CIMware Ltd., United Kingdom. Multimedia design and
programming by P. G. Ranky and M. F. Ranky.
73. Romero , C. , Department of logistics passive RFID initial implementation, paper presented
at the USA Department of Defense Conference, RFID Media Briefi ng,
Washington DC, Feb. 2005.
74. Sangoi , R. , Smith , C. G. , et al . ( 2004 ), Printing radio frequency identifi cation (RFID) tag
antennas using inks containing silver dispersions , J. Dispersion Sci. Technol . 25 ( 4 ),
513 – 521 .
75. Smith , K. , Enabling the supply chain, paper presented at the USA Department of Defense
Conference, RFID Media Briefi ng, Washington, DC, Feb. 2005.
76. Sugimoto , M. , Kusunoki , F. , Inagaki , S. , Takatoki , K. , and Yoshikawa , A. ( 2004 ), A system
for supporting collaborative learning with networked sensing boards , Syst. Comput. Jpn .,
35 ( 9 ), 39 – 50 .
77. Wilke , P. , and Braunl , T. ( 2001 ), Flexible wireless communication network for mobile
robot agents , Ind. Robot , 28 ( 3 ), 220 – 233 .
201
3.2
ROLE OF QUALITY SYSTEMS AND
AUDITS IN PHARMACEUTICAL
MANUFACTURING ENVIRONMENT
Evan B. Siegel and James M. Barquest
Ground Zero Pharmaceuticals, Inc., Irvine, California
Contents
3.2.1 cGMP Regulations
3.2.1.1 Duties of Quality Control Unit under cGMP Regulations
3.2.2 Quality Assurance Function
3.2.3 Quality Systems Approach
3.2.4 Management Responsibilities
3.2.5 Resources
3.2.6 Manufacturing Operations
3.2.6.1 Design, Develop, and Document Product and Processes
3.2.6.2 Inputs
3.2.6.3 Perform and Monitor Operations
3.2.6.4 Address Nonconformities
3.2.7 Evaluation Activities
3.2.7.1 Trend Analysis
3.2.7.2 Conduct Internal Audits
3.2.7.3 Quality Risk Management
3.2.7.4 Corrective and Preventive Actions
3.2.7.5 Promote Improvement
3.2.8 Transitioning to Quality Systems Approach
3.2.9 Audit Checklist for Drug Industry
3.2.9.1 Instructions for Using Audit Checklist
References
By regulation, appropriate practice, and common sense, quality assurance (QA) is
a critical function in the pharmaceutical manufacturing environment. The need for
an independent unit to audit and comment on the appropriate application of
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
202 ROLE OF QUALITY SYSTEMS AND AUDITS
standard operating procedures, master batch records, procedures approved in
product applications, and the proper functioning of the quality control (QC) unit is
paramount. This helps assure that products are manufactured reliably, with adherence
to approved specifi cations, and that current good manufacturing practices
(cGMP) are maintained in conformance to regulation, both in the facility in general
and the microenvironment of each product ’ s manufacturing sequence.
Quality assurance personnel must have the appropriate training, experience,
familiarization with the manufacturing facility and products, enforced independence
from the production chain of command, and the ability to review adherence to
procedures, policies, and agreed - upon approaches to manufacturing quality pharmaceuticals.
This helps to provide both an environment and a manufactured product
that can withstand Food and Drug Administration (FDA) inspection and support
a fi rm ’ s reputation for quality products.
The cGMP regulations establish requirements that are intended to provide a high
level of assurance that the pharmaceutical products produced satisfy the strength,
purity, potency, and other quality requirements established for the fi nished product
to assure that it is fi t for its intended use. Manufacturers must establish a quality
control unit that is responsible for many of the quality - related activities required
by the regulations. These regulations have not been substantially updated since 1978.
Since then, the science and practice of quality assurance have substantially evolved
to include the development of quality management systems and risk management
approaches to better assure product quality and fi tness for use. Pharmaceutical
product manufacturers are increasingly interested in implementing a comprehensive
quality management system (QMS) and employing risk management approaches
because they allow them to apply newer quality management principles that they
believe enable them to more effectively assure product quality and better allow
harmonization with evolving international regulatory quality system requirements.
The FDA has not changed the cGMP regulations but, as part of its Pharmaceutical
CGMPs for the 21st Century Initiative, encourages this quality systems approach to
cGMP compliance.
This Chapter describes outlines and discusses the regulations applicable to the
QA function and unit, structure, function, charter, and application of the unit in the
pharmaceutical manufacturing environment. In addition, it discusses additional
quality - related responsibilities that may result when manufacturers move toward a
quality systems approach to quality that incorporates current quality system models
to further improve quality and harmonize with international quality system
requirements.
The justifi cation for, and execution of, the QA audit are also described, including
preparation, key items of interest, a typical checklist of the audit itself, corrective
and preventive actions following the audit, and suggested measures for assuring
successful operation of the unit.
3.2.1 c GMP REGULATIONS
The cGMP regulations for the manufacture of pharmaceutical products are contained
in Parts 210 and 211 of Title 21 of the Code of Federal Regulations (CFR)
[1] . These regulations, as well as guidance documents and other FDA documents
pertaining to the regulation and FDA inspection of pharmaceutical product manufacturers,
may be accessed on the FDA website at www.fda.gov . Part 210 specifi es
the scope and applicability of the cGMP regulations and defi nes terms used in the
regulations. Part 210 also indicates that the regulations establish “ minimum ” cGMP
requirements and that products that are not manufactured under cGMP are adulterated.
Adulterated products and the persons responsible for the adulteration are
subject to regulatory action by the FDA.
Part 211 contains specifi c good manufacturing practice requirements for fi nished
pharmaceuticals and is divided into Subparts A – K as follows:
A. Scope
B. Organization and Personnel
C. Buildings and Facilities
D. Equipment
E. Control of Components and Drug Product Containers and Closures
F. Production and Process Controls
G. Packaging and Labeling Control
H. Holding and Distribution
I. Laboratory Controls
J. Records and Reports
K. Returned and Salvaged Drug Products
The cGMP regulations are written to address the primary potential sources of
product variability. Subpart B establishes the quality control unit and the duties of
that unit, establishes personnel requirements and addresses personnel practices
(e.g. sanitation) intended to reduce the likelihood of product contamination. Subparts
C and D establish requirements for buildings and facilities and equipment
used in the manufacture, processing, packing, or holding of a drug product. Subparts
E through H establish controls over the major processes associated with the production
of a fi nished and packaged drug product that is ready to be shipped for distribution
to users. Controls are established for incoming raw materials and components
and continue through manufacturing, packaging, labeling, holding, and distribution
of fi nished, packaged, labeled, and released drug product. Subpart I requires the
establishment of scientifi cally sound and appropriate specifi cations, standards, sampling
plans, and test procedures; requires instrument specifi cations and calibration;
and establishes lot or batch testing and release requirements. Subpart J establishes
documentation requirements including master and batch records, and Subpart K
addresses the control and disposition of returned drug products and places limitations
on the salvage of drug products that have been subjected to improper storage
conditions (e.g., smoke, heat, fi re, moisture).
3.2.1.1 Duties of Quality Control Unit under c GMP Regulations
The cGMP regulations assign specifi c duties to the quality control unit. The unit is
required to have the responsibility and authority to approve or reject all components,
drug product containers, closures, in - process materials, packaging material,
cGMP REGULATIONS 203
204 ROLE OF QUALITY SYSTEMS AND AUDITS
labeling, and drug products and the authority to review production records to assure
that no errors have occurred or, if errors have occurred, that they have been fully
investigated. The responsibilities of the unit extend to approving or rejecting drug
products manufactured, processed, packed, or held by contract manufacturers. The
organization must assure that the quality control unit has adequate laboratory
facilities for the testing and approval (or rejection) of components, drug product
containers, closures, packaging materials, in - process materials, and drug products.
In addition to duties associated with the approval of materials and fi nished products,
the unit is also responsible for approving or rejecting all procedures or speci-
fi cations impacting on the identity, strength, quality, and purity of the drug product.
This includes review and approval of procedures for production and process control,
including any changes to these procedures. These procedures, and the responsibilities
and procedures applicable to the quality control unit within the organization,
must be written and followed.
All specifi cations, standards, sampling plans, test procedures, or other laboratory
control mechanisms, including any changes, must be in writing and reviewed and
approved by the quality control unit.
Written procedures describing the handling of all written and oral complaints
regarding a drug product are required. The quality control unit is responsible for
reviewing any complaint involving the possible failure of a drug product to meet
any of its specifi cations and, for such drug products, making a determination as to
the need for an investigation in accordance with cGMP requirements. The review
should include a determination if the complaint represents a serious and unexpected
adverse drug experience, which is required to be reported to the FDA. A written
record of each complaint must be maintained in a complaint fi le.
3.2.2 QUALITY ASSURANCE FUNCTION
The term quality is used in many industries and in everyday life and can have various
meanings depending on context. For the purposes of discussion here quality means
the product requirements or attributes that have a bearing on the product ’ s specifi ed
requirements. Quality assurance activities are those processes and activities conducted
to assure that a product or service consistently satisfi es its requirements and
is fi t for its intended use. In the pharmaceutical manufacturing environment, this
means the activities conducted to assure that the pharmaceutical product ’ s identity,
strength, purity, potency, and other quality attributes conform to approved
specifi cations.
In the United States, cGMP requirements for the manufacture of drugs were
established by regulation in 1978 and have not been substantially updated since
then. The science and practice of quality assurance has substantially evolved since
then to include the development of quality systems [2, 3] and risk management
approaches [4] to better assure product quality and fi tness for use. Pharmaceutical
product manufacturers are increasingly interested in implementing these approaches
because they allow the manufactures to apply newer quality management principles
that they believe enable them to more effectively assure product quality and
better allow harmonization with evolving international regulatory quality system
requirements.
QUALITY SYSTEMS APPROACH 205
3.2.3 QUALITY SYSTEMS APPROACH
The systems approach to quality involves a coordinated approach to the management
of quality - related activities as processes that work in conjunction with one
another to provide assurance that the product meets its specifi ed requirements. It
involves:
• A management commitment to quality that is communicated throughout the
organization
• Identifying quality requirements using risk management and other methods as
appropriate
• Developing a quality policy, plan, objectives
• Establishing an organizational structure with identifi ed responsibilities and
authorities that allows quality objectives to be met
• Providing the resources needed to meet quality objectives
• Developing the required systems and processes
• Establishing methods for the ongoing objective evaluation of the performance
of systems and processes including quality auditing
• Initiating corrective and preventive actions as needed to assure that quality
objectives are consistently and reliably met
The use of risk management techniques in identifying product requirements, establishing
processes and process control and monitoring methods, evaluating quality
data, identifying appropriate corrective and preventive actions to address quality
problems, and for other quality - related activities can increase the overall effi ciency
and effectiveness of the quality system.
The FDA has recognized the value of and encourages a risk based quality systems
approach for the manufacture of pharmaceutical products. This is refl ected in its
Pharmaceutical CGMPs for the 21st Century Initiative. In association with this initiative
the FDA has published reports and guidance documents that collectively
provide information that can be used by pharmaceutical product manufacturers in
implementing a quality systems and risk management approach to pharmaceutical
cGMP regulations compliance [5 – 8] . In implementing this initiative, the FDA has
made it clear that it does not impose new regulatory requirements on manufacturers.
The FDA has provided information and guidance that is intended to serve as a
bridge between the 1978 regulations and current quality systems by explaining how
manufacturers implementing such systems can do so in full compliance with the
cGMP regulations. This approach differs from that used by the FDA when it updated
the cGMP regulations for medical devices to employ a quality systems approach.
The 1996 Quality System Regulation updated the GMP requirements for fi nished
medical device manufacturers to reduce the risk of inadequate device design and
to harmonize them with international quality system standards that were in effect
at that time [9] . These international standards have since been updated [10] ; however
the device quality system regulation remains consistent with modern quality system
models.
In a modern quality system, the organizational unit responsible for quality -
related activities within the organization generally has a central role in the
206 ROLE OF QUALITY SYSTEMS AND AUDITS
development and management of the overall quality system. These activities can
include quality control, quality assurance, quality planning, and quality improvement.
The cGMP regulations do not defi ne or employ these terms, but the activities
the regulations assign to the quality control unit fall within these defi nitions as currently
defi ned [2, 8, 11] .
Current quality system models involve quality - related activities and terms that
are not included in the cGMP regulations. Further, quality as a professional discipline
is evolving. It is, therefore, important for organizations adopting a quality
systems approach to unambiguously defi ne the terms and quality concepts they will
be using and to include these defi nitions as appropriate in training all staff in the
organization who will be involved in quality - related activities. This will help assure
effective communication throughout the organization and with vendors and others
(e.g., regulatory agencies, third - party auditors) who interact with the organization
on quality - related matters. Regulatory defi nitions should be recognized, and the use
of nonstandard or outdated terminology should be avoided to the extent possible.
Incorporating terms and defi nitions by reference from pertinent standards and FDA
guidance documents may be helpful. FDA guidance on the quality systems approach
to pharmaceutical cGMP regulations (pharmaceutical QS guidance) includes the
following defi nitions:
Quality Assurance (QA) Proactive and retrospective activities that provide con-
fi dence that requirements are fulfi lled.
Quality Control (QC) The steps taken during the generation of a product or
service to ensure that it meets requirements and that the product or service is
reproducible.
Quality Management (QM) Accountability for the successful implementation
of the quality system.
Quality System (QS) Formalized business practices that defi ne management
responsibilities for organizational structure, processes, procedures, and
resources needed to fulfi ll product/service requirements, customer satisfaction,
and continual improvement.
Quality Unit (QU) A group organized within an organization to promote quality
in general practice.
The FDA notes in its pharmaceutical QS guidance document that many current
quality system concepts correlate very closely with the cGMP regulations and that
the activities required by the regulations are generally consistent with a quality
systems approach. In this and other guidance documents, the FDA uses the
term quality unit rather than quality control unit as defi ned in the cGMP regulations
to refer to the organizational unit with responsibility for quality - related
activities. In a modern quality systems model these quality - related activities may go
beyond, but are not necessarily inconsistent with, those required by the cGMP
regulation.
Use of the term quality unit is consistent with current quality management system
models [2, 10] , which are intended to assure that the various operations associated
with all systems are appropriately planned, approved, conducted, and monitored,
and because the cGMP regulations specifi cally assign the QU the authority to create,
QUALITY SYSTEMS APPROACH 207
monitor, and implement a quality system. The FDA cautions that such activities do
not substitute for, or preclude, the daily responsibility of manufacturing personnel
to build quality into the product. The FDA has specifi cally indicated that the overarching
philosophy articulated in both the cGMP regulations and in robust modern
quality systems is that quality should be built into the product, and testing alone
cannot be relied on to ensure product quality.
Other cGMP - assigned responsibilities of the QU that are consistent with modern
quality system approaches include the following:
• Ensuring that controls are implemented and completed satisfactorily during
manufacturing operations
• Ensuring that developed procedures and specifi cations are appropriate and
followed, including those used by a fi rm under contract to the manufacturer
• Approving or rejecting incoming materials, in - process materials, and drug
products
• Reviewing production records and investigating any unexplained
discrepancies
The FDA has stressed that the release of the pharmaceutical QS guidance document
does not impose new regulatory requirements on manufacturers but encourages
manufactures to adopt a quality systems approach to cGMP compliance
because of the potential benefi ts. An appropriately designed and implemented
quality system can do the following:
• Reduce the number of (or prevent) recalls, returned or salvaged products, and
defective products entering the marketplace
• Harmonize the cGMP regulations to the extent possible with other widely used
quality management systems, which is desirable because of the globalization of
pharmaceutical manufacturing, and the increasing prevalence of drug – device
and biologic – device combination products
• When coupled with manufacturing process and product knowledge and the use
of effective risk management practices, handle many types of changes to facilities,
equipment, and processes without the need for prior approval regulatory
submissions
• Potentially result in shorter and fewer FDA inspections by lowering the risk of
manufacturing problems
• Provide the necessary framework for implementing quality by design (building
in quality from the product development phase and throughout a product ’ s life
cycle), continual improvement, and risk management in the drug manufacturing
process
This suggests that even without making changes in the cGMP regulations, the
FDA may be looking at them from a “ new ” quality systems perspective. The regulations
include terms such as adequate and appropriate that may be subject to interpretation
based on relevant technical or scientifi c capabilities and state - of - the - art
knowledge. As these improve, the interpretation of what is adequate or appropriate
208 ROLE OF QUALITY SYSTEMS AND AUDITS
can change as well. Practically, most manufacturers are more than willing to adopt
methods that can improve the quality and safety of their pharmaceutical products
because it is cost effective in the long run [11] but may be reluctant to do so for
fear of being considered out of compliance with the cGMP regulations. Current
FDA efforts in this regard should serve to allay manufacturers ’ concerns in this
area.
The major elements of the quality system model described in the FDA ’ s
pharmaceutical QS guidance document are consistent with existing quality
system standards. These elements are as follows:
• Management responsibilities
• Resources
• Manufacturing operations
• Evaluation activities
3.2.4 MANAGEMENT RESPONSIBILITIES
Current quality system models assign management a major role in the deployment
and operation of a successful quality system. In such systems, major management
responsibilities include the following:
• Provide leadership by establishing a commitment to quality that is supported
by all levels of management and is communicated throughout the
organization
• Create an organizational structure with clearly defi ned responsibilities and
authorities to perform quality functions associated with achieving quality
objectives
• Building and documenting a quality system to meet specifi ed quality and
regulatory requirements and achieve quality objectives
• Establishing a quality policy and objectives, and quality plans that are aligned
with the organization ’ s strategic plans and communicate this throughout the
organization
• Reviewing the system by establishing appropriate accountability systems within
the organization to monitor and report quality data and system status to management
and assure that appropriate corrective and preventive actions are
taken in response to quality problems using effective change control procedures
and documented
The cGMP does not specifi cally assign management responsibility for these
actions, although actions of this nature are required by the regulation. Table 1 from
the pharmaceutical QS guidance document shows this relationship.
Under a comprehensive quality system the QU can expect an expanded and more
visible role within the organization with greater accountability to and interaction
with upper management. The QU should ideally be independent of the other organizational
units to assure clear delineation of responsibility and authority and avoid
confl icts. In certain instances, such as auditing, independence or objectivity is central
to the effectiveness of the audit process, and auditors therefore should not have
direct responsibility over the areas being audited.
The cGMP regulations do not specify how the QU should be integrated into the
overall organization but, in general, the QU should be structured to refl ect management
’ s strong commitment to quality and to facilitate achieving quality objectives.
The structure (e.g., organizational relationship to other organizational units, reporting
relationships) should provide clear lines of responsibility and authority that
support the production, quality, and management activities necessary to achieve
quality objectives. Different organizations may accomplish this in different ways;
however, experience has been that placement of the quality function on the same
level within the organizational hierarchy as other major organizational units (e.g.,
production) sends a clear message both within and outside the organization that
top management has a strong commitment to quality.
The cGMP regulations require quality - related activities to be conducted during
all phases of manufacturing from the acceptance of raw materials through batch
release, packaging, and labeling. The regulations also require that all personnel,
including those engaged in quality - related activities, have suffi cient education, training,
and experience or any combination thereof to enable them to perform their
assigned functions. In a quality systems approach to cGMP compliance, the role of
quality personnel can be signifi cantly expanded to include internal quality auditing,
expanded review and analysis of quality data, investigation of nonconformance, root
cause analysis, risk analysis, and other quality - related activities. Many of these activities
are likely to be conducted with personnel from other organizational elements
such as manufacturing, material control, facilities, product development, or engineering
staff. Quality staff should have suffi cient scientifi c and technical knowledge
and training (e.g., statistical methods, risk analysis) and knowledge of the product
and manufacturing processes to effectively perform their assigned functions
TABLE 1 21 CFR cGMP Regulations Related to Management Responsibilities
Quality System Element Regulatory Citations
1. Leadership
2. Structure Establish quality function: § 211.22(a) [see defi nition
§ 210.3(b)(15)[
Notifi cation: § 211.180(f)
3. Build QS QU procedures: § 211.22(d)
QU procedures, specifi cations: § 211.22(c), with reinforcement
in: § § 211.100(a), 211.160(a)
QU control steps: § 211.22(a), with reinforcement in
§ § 211.42(c), 211.84(a), 211.87, 211.101(c)(1), 211.110(c),
211.115(b), 211.142, 211.165(d), 211.192
QU quality assurance; review/investigate: § § 211.22(a),
211.100(a – b) 211.180(f), 211.192, 211.198(a)
Record control: § § 211.180(a – d), 211.180(c), 211.180(d),
211.180(e), 211.186, 211.192, 211.194, 211.198(b)
4. Establish policies,
objectives, and plans
Procedures: § § 211.22(c – d), 211.100(a)
5. System review Record review: § § 211.100, 211.180(e), 211.192, 211.198(b)(2)
MANAGEMENT RESPONSIBILITIES 209
210 ROLE OF QUALITY SYSTEMS AND AUDITS
and competently interact with personnel from other organizational elements as
necessary.
3.2.5 RESOURCES
The appropriate assignment of resources is essential to the success of any endeavor,
and this is particularly critical in a pharmaceutical manufacturing environment.
Inadequate staffi ng, training, manufacturing equipment and facilities, environmental
controls, analytical equipment, and other resources can be sources of variability
leading to the production of product that does not meet specifi ed requirements.
Modern quality system standards specifi cally address the issue of resources by
requiring the organization to determine and provide the human, infrastructure, and
work environment resources necessary for the quality system. The cGMP regulations
address the resource issue in provisions that are intended to assure the
adequacy of personnel (including consultants), manufacturing facilities including
contract facilities, equipment, and laboratory facilities. The QU has signifi cant
responsibility in this regard.
The FDA, in its pharmaceutical QS guidance document, discusses the need for
adequate resources in developing, implementing, and managing a quality system
that complies with the cGMP regulations. Management is responsible for identifying
resource requirements and providing resources accordingly, including providing
training that is appropriate to the assigned activities. Personnel should understand
the impact of their activities on their assigned duties and be familiar with cGMP
requirements and the organization ’ s quality system. This is consistent with the
generally accepted idea that a culture of quality within an organization requires
personnel to understand quality concepts, the organization ’ s quality and regulatory
objectives, and how their assigned activities contribute to the achievement of these
objectives and fi t into the overall quality system. Management should establish a
working environment that encourages problem solving and communication in
identifying and acting upon quality - related issues. While the provision of resources
is generally considered a management function, the QU and other organizational
units should be involved in the identifi cation of the resources required to achieve
quality objectives, including regulatory compliance, the assessment of the adequacy
of existing resources, evaluating the effect of personnel, facility, product, process,
regulatory, and other changes on resource needs, and generally providing management
the information needed to make necessary and appropriate resource
decisions.
Current quality system models employ a risk - based and data - driven approach to
the development of QS system requirements to assure their adequacy. The FDA
notes that the cGMP regulations place as much emphasis on processing equipment
as testing equipment and contain specifi c requirements for the qualifi cation, calibration,
cleaning, and maintenance of production equipment that may be a higher
standard than most nonpharmaceutical quality system models. Organizations should
always keep in mind that, while the FDA may be encouraging the adoption of a
comprehensive quality system, any system developed must satisfy the requirements
of the cGMP regulations.
Under a quality system model, the specifi cation of facility and equipment requirements
may be performed by technical experts (e.g., engineers, development scientists)
who have an understanding of the pharmaceutical science, manufacturing
processes, and risk factors associated with the product and its manufacture. The
cGMP regulations require the QU to be responsible for reviewing and approving
all initial design criteria and procedures pertaining to facilities and equipment and
any subsequent changes. These requirements are not mutually exclusive; while they
place ultimate responsibility for review and approval of these activities with the QU,
the regulations do not preclude a cross - functional review involving persons with
relevant expertise from multiple areas of the organization. A requirement of both
the cGMP and current quality system models is that such review and approval be
conducted by persons who are qualifi ed by education, training and experience to
do so.
In the control of outsourced operations, the cGMP regulations require that the
QU approve or reject products or services provided under a contract. Under current
quality system models, the organization must follow a formal vendor qualifi cation
process to qualify outsource providers and verify through inspection or other appropriate
means that the provider is capable of meeting the requirements of the organization.
To comply with the regulation, these operations should be conducted by
the QU.
Table 2 compares the major elements of a quality systems approach to addressing
resource issues with corresponding requirements in the CGMP regulations.
3.2.6 MANUFACTURING OPERATIONS
There is signifi cant commonality between the requirements contained in current
quality system models such as ISO 9001 - 2000 and the cGMP regulation requirements
for manufacturing operations. The FDA has identifi ed four major elements
of a QS approach to manufacturing operations. These are identifi ed and compared
to the cGMP requirements in Table 3 .
TABLE 2 21 CFR cGMP Regulations Related to Resources
Quality System Element Regulatory Citation
1. General arrangements
2. Develop personnel Qualifi cations: § 211.25(a)
Staff number: § 211.25(c)
Staff training: § 211.25(a – b)
3. Facilities and equipment Buildings and facilities: § § 211.22(b), 211.28(c),
211.42 – 211.58, 211.173
Equipment: § § 211.63 – 211.72, 211.105, 211.160(b)(4),
211.182
Lab facilities: § 211.22(b)
4. Control outsourced operations Consultants: § 211.34
Outsourcing: § 211.22(a)
MANUFACTURING OPERATIONS 211
212 ROLE OF QUALITY SYSTEMS AND AUDITS
3.2.6.1 Design, Develop, and Document Product and Processes
In a modern quality systems manufacturing environment, the signifi cant characteristics
of the product being manufactured should be defi ned and verifi ed as meeting
requirements from design to delivery, and control should be exercised over all
changes. This is consistent with the requirements of the cGMP regulation that
require quality and manufacturing processes and procedures, and changes to them,
to be defi ned, approved, and controlled. The idea of controlling the design of both
product and process is consistent with concepts included in the FDA Pharmaceutical
cGMPs for the 21st Century Initiative to assure product safety that focus on the
entire product life cycle. No amount of “ downstream ” control and testing can compensate
for a design that results in a product or production process that is incapable
of meeting the requirements necessary to assure that the product is safe and
effective for its intended use. Documentation is required and can include
the following:
• Resources and facilities used
• Procedures to carry out the process
• Identifi cation of the process owner who will maintain and update the process
as needed
• Identifi cation and control of important variables
• Quality control measures, necessary data collection, monitoring, and appropriate
controls for the product and process
• Any validation activities, including operating ranges and acceptance criteria
• Effects on related process, functions, or personnel
The cGMP regulations include specifi c packaging and labeling controls, so packaging
and labeling requirement, processes, and controls should be included in a QS -
based approach to product and process design and development.
TABLE 3 21 CFR cGMP Regulations Related to Manufacturing Operations
Quality System Element Regulatory Citation
1. Design and develop product and
processes
Production: § 211.100(a)
2. Examine Inputs Materials: § § 210.3(b), 211.80 – 211.94, 211.101,
211.122, 211.125
3. Perform and monitor operations Production: § § 211.100, 211.103, 211.110,
211.111, 211.113
QC criteria: § § 211.22(a – c), 211.115(b),
211.160(a), 211.165(d), 211.188
QC checkpoints: § § 211.22 (a), 211.84(a),
211.87, 211.110(c)
4. Address nonconformities Discrepancy investigation: § § 211.22(a),
211.100, 211.115, 211.192, 211.198Recalls:
21 CFR Part 7
Manufacturers and the FDA have expressed concern that existing regulatory
requirements (e.g., the need to effect manufacturing process changes through
the regulatory submission process) may be excessively rigid and not conducive
to innovation regardless of the potential benefi ts. The FDA acknowledges that
the reluctance to pursue potentially innovative changes in pharmaceutical manufacturing
can be undesirable from a public health perspective and has published a
process analytical technology (PAT) guidance document that is intended to address
this by promoting the use of analytical tools to gain process understanding
and meet regulatory requirements for validating and controlling manufacturing
processes [7] .
The PAT guidance document describes a voluntary approach to the design,
analysis, and control of manufacturing processes that involves the timely (e.g., in -
process) measurement of critical quality and performance attributes of raw and
in - process materials and processes, with the goal of ensuring fi nal product quality.
The term analytical in PAT is broadly interpreted to include the integrated application
of chemical, physical, microbiological, mathematical, and risk analysis as
appropriate. One goal of PAT is to design and develop well - understood processes
that will consistently ensure predefi ned quality at the end of the manufacturing
process. This is consistent with a quality systems approach. PAT should ideally
be initiated during the development stage and is intended to be integrated into
existing regulatory processes with timely communication with the FDA a key
element. The FDA has published the guidance document and other pertinent PAT
information on its website at www.fda.gov . Companies interested in PAT methods
should contact the FDA. FDA internal implementation of PAT includes the
following:
• A PAT team approach of CMC review and cGMP inspections
• Joint training and certifi cation of FDA PAT review, inspection, and compliance
staff
• Scientifi c and technical support for the PAT review, inspection, and compliance
staff
Process analytical technology is consistent with the quality systems approach in
that it is based on science and engineering principles for assessing and mitigating
risks related to poor product and process quality. In the PAT guidance, the FDA
indicates that the desired state for pharmaceutical manufacturing may be characterized
as follows:
• Product quality and performance are ensured through the design of effective
and effi cient manufacturing processes
• Product and process specifi cations are based on a mechanistic understanding
of how formulation and process factors affect product performance
• Continuous real - time quality assurance
• Relevant regulatory policies and procedures are tailored to accommodate the
most current level of scientifi c knowledge
• Risk - based regulatory approaches recognize the following:
MANUFACTURING OPERATIONS 213
214 ROLE OF QUALITY SYSTEMS AND AUDITS
The level of scientifi c understanding of how formulation and manufacturing
process factors affect product quality and performance
The capability of process control strategies to prevent or mitigate the risk of
producing a poor - quality product
Process analytical technology is consistent with a modern risk - based data - driven
quality systems approach to cGMP compliance.
3.2.6.2 Inputs
Current QMS models adopt a process - oriented approach to the design and operation
of a QMS as a system of interrelated processes, each with inputs and outputs,
which are designed to function in a defi ned way. Some process outputs are inputs
to other processes. This concept is easily applied and understood within the manufacturing
environment because it is process oriented. Inputs to manufacturing processes
include any material that goes into the fi nal product, including materials
purchased from vendors for use in manufacturing and in - process materials. Manufacturing
operations generally involve multiple processes conducted in a defi ned
manner to produce the fi nished product. Each process has a set of inputs and produces
one or more outputs that may, in turn, be an input to a subsequent process.
Each process has an input – output relationship such that changes or variation in one
or more inputs will produce an attendant change in the output. Input specifi cations
are established to assure that the fi nal product meets its requirements. A robust
quality system will ensure that all inputs to the manufacturing process are suitable
for use by establishing quality controls for the receipt and acceptance from qualifi ed
vendors, production, storage, and use of all inputs.
The cGMP regulations require either testing or use of a certifi cate of analysis
(COA) plus an identity analysis for the release of materials for manufacturing. The
quality systems approach additionally calls for initial supplier qualifi cation based
on an objective evaluation and periodic auditing of suppliers based on risk assessment
to verify the adequacy of suppliers ’ quality systems. During the audit, a manufacturer
can observe the testing or examinations conducted by the supplier to help
determine the reliability of the supplier ’ s COA. Under a QMS model, the QU would
normally be responsible for auditing suppliers as part of its overall responsibility
for materials acceptance.
Change control involves the evaluation of proposed changes in a systematic way
to determine how they would affect process outputs and ultimately the fi nished
product and is an important element of current quality system models. The cGMP
regulations require the QU to approve specifi cations, and certain changes require
review and approval by the QU. Under a quality system model, changes to materials
(e.g., specifi cation, supplier, or materials handling) should be implemented through
a formal change control system involving the documented competent review and
approval of the proposed change prior to implementation and communication of
changes as appropriate throughout the organization. Manufacturers should also
consider how best to assure that changes made by suppliers in supplied materials
that may affect the quality of the fi nished product can be identifi ed and appropriately
evaluated by the manufacturer. Such provisions should be included in supplier
agreements where possible.
0
0
3.2.6.3 Perform and Monitor Operations
Both the cGMP regulations and quality system models call for the monitoring of
critical processes that may be responsible for causing variability during production.
The cGMP regulations require written production and process control procedures
and specify process control activities that must be performed and documented.
Current quality system models also require written procedures, process verifi cation
and validation as appropriate, the establishment of appropriate process control
measures and documentation. Risk analysis methods and design and development
data may be used to establish process control and monitoring requirements. A
quality systems approach allows the manufacturer to more effi ciently and effectively
validate, perform, and monitor operations and ensure that the controls are scientifi -
cally sound and appropriate. Production and process controls should be designed
to ensure that the fi nished products have the identity, strength, quality, and purity
they purport or are represented to possess. A systems approach will consider all
sources of variability from inputs, through manufacturing processes, packaging,
labeling, and shipping to assure that the product that is delivered to the user meets
quality requirements.
One important aspect of the quality systems approach is the ongoing collection
and analysis of quality data to continuously evaluate quality system effectiveness.
Historical data, process knowledge, and risk analysis methods can be applied to
identify specifi c data requirements. Trending and other data analysis methods can
allow identifi cation of actual and potential sources of nonconformity so that appropriate
corrective and preventive actions can be taken in accordance with established
change control procedures.
The entire product life cycle should be addressed by the establishment of monitoring
and continual improvement mechanisms in the quality system. Even well -
defi ned or mature manufacturing processes may “ drift ” due to a host of factors
including equipment and facility aging, changes or variation in raw materials, electrical
power fl uctuations, and environmental changes. Thus, process validation is not a
one - time event but an activity that continues throughout a product ’ s life. One
major quality system objective should be to identify emerging quality problems
before nonconformities occur. Trending of periodically collected environmental
monitoring data may, for example, identify a slow but steady increase in airborne
particulate levels that, if left unaddressed and the trend continues, could exceed a
fi rm ’ s internal environmental standards and adversely affect the product. Early
identifi cation of such problems allows an investigation to be initiated to identify the
cause so that appropriate corrective and preventive actions can be taken in accordance
with established change control procedures. After a change is implemented,
its effectiveness should be objectively verifi ed and affected processes revalidated if
necessary.
3.2.6.4 Address Nonconformities
A key component in any quality system is appropriately responding to nonconformities
(i.e., deviations from requirements established under the quality system for
in - process material or fi nal product quality attributes, process control parameters,
records, procedures, etc.). Nonconformities may be detected during any stage of the
MANUFACTURING OPERATIONS 215
216 ROLE OF QUALITY SYSTEMS AND AUDITS
manufacturing process or during quality control activities. The cGMP regulations
require an investigation to be initiated and that the investigation, conclusion, and
follow - up be documented. A primary objective of any manufacturing quality system
is to prevent nonconforming product from being produced and distributed. The
complete response to nonconformities should be risk based and can include the
following components:
• Assessment of how the nonconformity will affect the quality of the fi nished
product (i.e., determination if the nonconformity has resulted, or could result,
in product that does not meet its specifi ed purity, potency, and quality
characteristics).
• Determine any actions necessary to assure that product that does not meet its
specifi ed requirements is not produced and that appropriate steps are taken
with regard to any nonconforming product that has been produced to assure
that consumers are not harmed and that regulatory requirements are
satisfi ed.
• Determine the cause of the nonconformity.
• Identify any actions needed to correct the cause and to prevent recurrence.
• Document the investigation, fi ndings, and follow - up actions.
• Assess the effectiveness of follow - up actions.
• Repeat the cycle as needed.
A nonconformity may not result in the fi nished product failing to meets its
requirements; however, investigation of the nonconformity may identify process or
quality system defi ciencies that require attention. For example, a small but unexpected
deviation from a process control requirement (e.g., temperature, blending
time) may not exceed the limit for which the process was initially validated and
thus not be expected to adversely effect the fi nished product but could suggest an
emerging process control or equipment issue that if not corrected could result in
future product nonconformities. Similarly, nonconformities in the form of errors or
omissions in production records or deviations from written procedures may not
always result in product nonconformity but could suggest training, process design,
or other issues that ought to be addressed. Thus the response to nonconformities
should not be limited to a determination of the immediate impact on the fi nished
product, but also consider its implications regarding overall quality system
performance.
3.2.7 EVALUATION ACTIVITIES
The evaluation component of a QMS is intended to provide objective information
and data that allow the organization to assess the conformity of the product,
evaluate the performance of its quality system, and maintain and improve its effectiveness
[10] . The cGMP regulations similarly require evaluation activities as shown
in Table 4 .
3.2.7.1 Trend Analysis
The cGMP regulations require review and analysis of certain quality data annually
at least. Current quality system models emphasize data - based decision making and
the use of appropriate statistical analysis methods [2, 11] . Trend analysis is one statistical
tool specifi cally recommended by the FDA in its pharmaceutical QS guidance
document that can be very valuable in monitoring processes and quality system
performance to identify emerging problems and to assess the effectiveness of
improvement efforts. Traditional statistical process control and other methods also
provide valuable support in the objective and ongoing analysis of quality data and
can be helpful in implementing real - time quality assurance practices as recommended
by the FDA [7] .
3.2.7.2 Conduct Internal Audits
Internal auditing is not specifi cally required by the cGMP regulations, but manufacturers
have traditionally used internal audits as a self - assessment tool and to
prepare for FDA inspections. The FDA has for some time recognized the value of
internal auditing and encourages fi rms to conduct audits by, as a matter of policy,
not reviewing internal audit results during inspections [12] .
Current quality system models call for audits to be conducted at planned intervals
to evaluate effective implementation and maintenance of the quality system and to
determine if processes and products meet established parameters and specifi cations.
International standards provide guidance on auditing [13] . Audit procedures should
be developed and documented to ensure that the planned audit schedule takes into
account the relative risks of the various quality system activities. Factors that can
be incorporated into a risk - based approach to planning audit frequency and scope
include the following [6] :
• Existing legal requirements (e.g., cGMPs)
• Overall compliance status and history of the company or facility
• Robustness of a company ’ s quality risk management activities
• Complexity of the site
• Complexity of the manufacturing process
• Complexity of the product and its therapeutic signifi cance
• Number and signifi cance of quality defects (e.g., recall)
TABLE 4 21 CFR cGMP Regulations Related to
Evaluation Activities
Quality System Element Regulatory Citation
1. Analyze data for trends Annual review: § 211.180(e)
2. Conduct internal audits
3. Risk assessment
4. Corrective action Discrepancy investigation:
§ § 211.22(a), 211.192
5. Preventive action
6. Promote improvement § 211.110
EVALUATION ACTIVITIES 217
218 ROLE OF QUALITY SYSTEMS AND AUDITS
• Results of previous audits/inspections that can include prior internal audit
results as well as regulatory (e.g., state, federal, or other regulatory agencies)
and third - party audits
• Major changes of building, equipment, processes, and key personnel
• Experience with manufacturing of a product (e.g., frequency, volume, number
of batches)
• Test results of offi cial control laboratories
In general, auditors should not have direct responsibility over the matters being
audited. Auditors should be trained in auditing methods and have suffi cient technical
knowledge to be able to evaluate the systems being audited using objective audit
criteria [14] . Audit criteria may be based on applicable regulatory requirements,
standards to which the quality system is intended to conform (e.g., ISO 9001 - 2000),
and the specifi c requirements of the quality system being audited as indicated in
quality system documents. Auditing criteria should be defi ned prior to the initiation
of the audit.
Different audit approaches may be applied depending on the intended purpose
and scope of the audit. A top - down approach fi rst evaluates the overall structure of
the quality system and its subsystems. Selected subsystems may be chosen for review.
Systems identifi ed and developed by the FDA in a six - system inspection model for
the inspection of drug manufacturers [15] include the following:
• Overall quality system
• Facilities and equipment
• Materials system
• Production system
• Packaging and labeling
• Laboratory controls
Subsystems must be pertinent to the specifi c quality system being audited and may
coincide with major elements of a standard to which the quality system is intended
to conform or the major elements identifi ed in the FDA pharmaceutical QS guidance.
When using the top - down approach, the auditor will fi rst review each subsystem
to determine if the requirements that apply to that subsystem (e.g.,
regulatory requirements, the requirements of the standard) are met by defi ning,
documenting, and implementing appropriate procedures. Once the auditor has
verifi ed that the requisite procedures are in place, he or she will review the associated
records and other documents to verify that the procedures have been followed
and documented and that the quality system is functioning effectively as designed
and conforms to applicable regulatory requirements and standards. This approach
allows for a systematic evaluation of each subsystem and can be as detailed as
needed.
A bottom - up approach may be used to follow up on a specifi c quality problem
identifi ed from trend analyses, product nonconformities, adverse experiences, customer
complaints, or other sources of quality data. Starting with quality records
associated with the problem, the auditor will work his or her way up through the
quality system, examining the quality processes having a bearing on the quality
problem. This approach is helpful in identifying quality system issues that may be
associated with specifi c quality problems but does not readily allow evaluation of
the entire quality system.
A combination approach may also be used that employs elements of top - down
and bottom - up audits. This allows some level of assessment of the effectiveness of
the overall quality system while evaluating the cause of specifi c quality problems.
Auditors should select the audit method most appropriate for their intended
audit purpose. Initial quality system audits or regularly scheduled audits are likely
candidates for the top - down approach, while audits conducted as part of a root
cause analysis, for example, may best employ a bottom - up approach. The FDA
employs a similar approach to inspections. Regular scheduled biennial inspections
are more likely to employ a top - down methodology. For cause inspections conducted
in response to a specifi c product issue such as a recall are more likely to
employ a bottom - up approach. FDA investigators may employ a combination
approach during biennial inspections if investigators are aware of specifi c quality
problems that they wish to include in the inspection.
Auditing as described in QMS models is intended to assess the effectiveness of
the overall quality system as designed and conformance to applicable standards. The
overall quality system does not have to be covered in a single audit. Manufacturers
may choose to employ a rolling audit approach in which specifi cally identifi ed subsystems
are chosen for evaluation in accordance with an approved audit schedule.
Audit plans should be designed to effectively perform this assessment.
Compliance with cGMP requirements is also a major concern, and audit planning
should include assessment of conformance to cGMP requirements and readiness
for FDA inspections. Existing FDA guidance documents and compliance policy
guides describe FDA inspectional approaches and policy and can be used for reference
in developing audit plans [15 – 17] . It can be helpful to include mock FDA audits
as part of an overall auditing regimen. Some fi rms prefer to use outside auditors for
mock audits to better simulate the FDA inspection process. Mock audits are also
useful for training purposes to prepare the organization for FDA inspections.
The audit plan should be consistent with written quality auditing procedures
included in the quality manual or other quality system documentation. The plan
should include or refer to the objective criteria to be used to evaluate conformance
to requirements. The plan should include or refer to other documents that will be
used during the audit, including previous audit reports. If the audit is to include the
review of batch or production records, such review should be conducted in accordance
with a specifi ed sampling plan or other appropriate statistical rationale as
specifi ed in a fi rm ’ s quality system procedures.
Manufacturers implementing a quality system that conforms to an existing standard
may fi nd it helpful to create a table or some other document that shows the
relationship between cGMP requirements, requirements of the standard, and the
element(s) of the manufacturer ’ s quality system. Such a tool can help assure that
all pertinent requirements are covered in the quality system design and that audit
plans designs include assessment of all pertinent requirements.
Since current quality system models employ a systems approach, an audit checklist
that is organized by subsystem may be helpful, as described in Table 5 . The
form would include appropriate document control information such as form
EVALUATION ACTIVITIES 219
220 ROLE OF QUALITY SYSTEMS AND AUDITS
identifi cation, revision, and approval information. Companies may also wish to
include reference information used in planning the audit such as previous audit
reports, completed FDA Form 483 Inspectional Observations, third - party audit
reports, and pertinent internal QS documents (e.g., audit procedures). Depending
on the purpose of the audit, the subsystems may correspond to the six subsystems
identifi ed by the FDA for use by investigators in conducting cGMP inspections (i.e.,
quality, production, facilities and equipment, laboratory controls, materials, packaging
and labeling) or the major elements of a quality system standard. Cross references
between elements of the standard being used and the pertinent sections of
the cGMP regulations may be included as appropriate. The audit form should allow
entry of information regarding conformance or nonconformance to each requirement
and have space for a description of pertinent fi ndings.
The QMS models require periodic audits but do not specify audit frequency.
Audit frequency must be determined based on the risk associated with the matters
to be audited and other factors including results of previous audits and other quality
data. Periodic audits should be conducted over the entire product life cycle and
follow - up audits conducted as appropriate to verify that previously identifi ed quality
problems have been corrected in accordance with applicable quality system and
regulatory requirements.
3.2.7.3 Quality Risk Management
The FDA has endorsed quality risk management as part of an overall quality
systems approach to compliance with the cGMP regulations and achieving overall
TABLE 5 Example Audit Checklist
[Company Name] Quality System Audit Checklist
Form: Rev: Date: Approved:
Audit Date(s): Refs:
Auditor: Title Signature:
Requirement cGMP Section Cross
Reference
Conforms (Y/N/NA) Objective Evidence
and Comments
Subsystem 1
Requirement 1.1
Requirement 1.2
Subsystem 2
Requirement 2.1
Requirement 2.2
Subsystem 3
Requirement 3.1
Requirement 3.2
Subsystem N
quality system objectives [6] . Risk management methodologies permit management
to assign priorities to activities or actions based on an assessment of the risk including
both the probability of occurrence of harm and the severity of that harm.
Implementation of quality risk management includes assessing the risks, selecting
and implementing risk management controls commensurate with the level of risk,
and evaluating the results of the risk management efforts. In a manufacturing quality
systems environment, risk management is used as a tool in the development of
product specifi cations and critical process parameters. Used in conjunction with
process understanding, quality risk management helps manufacturers effectively
manage and control change.
A formal risk management process consists of several components:
• Risk assessment
Risk identifi cation
Risk analysis
Risk evaluation
• Risk control
Risk reduction
Risk acceptance
• Risk communication
• Risk review
Risk assessment starts with risk identifi cation , a systematic use of available information
to identify hazards (i.e., events or other conditions that have the potential
to cause harm). Information can be from a variety of sources including stakeholders,
historical data, information from the literature, and mathematical or scientifi c analyses.
Risk analysis is then conducted to estimate the degree of risk associated with
the identifi ed hazards. This is estimated based on the likelihood of occurrence and
resultant severity of harm. In some risk management tools, the ability to detect the
hazard may also be considered. If the hazard is readily detectable, this may be considered
a factor in the overall risk assessment. Risk evaluation determines if the risk
is acceptable based on specifi ed criteria. In a quality system environment, criteria
would include impact on the overall performance of the quality system and the
quality attributes of the fi nished product. The value of the risk assessment depends
on how robust the data used in the assessment process is judged to be. The risk
assessment process should take into account assumptions and reasonable sources
of uncertainty. Risk assessment activities should be documented.
Risk control starts with risk reduction, which includes any actions taken to eliminate
or reduce the risk. Actions taken should be commensurate with the signifi cance
of the risk. If the risk has been reduced to an acceptable level, an affi rmative decision
can be made to accept the risk (risk acceptance). One question to ask is if new
risks have been introduced as a result of the identifi ed risks being controlled. Risk
control measures should generally be conducted in accordance with change control
procedures and documented.
Risk communication involves the communication of appropriate information
about the risk to stakeholders (e.g., others involved in or affected by the quality
EVALUATION ACTIVITIES 221
0
0
0
0
0
222 ROLE OF QUALITY SYSTEMS AND AUDITS
system including management, users, regulatory agencies). Risk communication
should be documented. The included information might relate to the existence,
nature, form, probability, severity, acceptability, control, treatment, detectability, or
other aspects of risks to quality. Communication should be as appropriate and does
not necessarily need to be carried out for each and every risk acceptance.
Risk review should be conducted to evaluate the outputs of the risk management
process and repeated as necessary, based on new quality data or if there are process
or product changes.
The Q9 Quality Risk Management guidance document [6] identifi es a number
of risk management tools that manufacturers can apply, including failure mode
effects and criticality analysis (FMECA), hazard analysis and critical control
points (HAACP), and preliminary hazard analysis (PHA), and provides examples
of how quality risk management might be applied to quality management, development,
materials management, production, and other operations within the
organization.
3.2.7.4 Corrective and Preventive Actions
Corrective and preventive action (CAPA) is the term commonly used to describe
the subsystem of a comprehensive quality system that deals with the systematic
investigation, understanding, and response to quality issues including nonconformities.
A corrective or preventive action may be initiated based on review and analysis
of quality data from a variety of sources including adverse experiences, product
complaints, quality audits, FDA inspections, third - party inspections, nonconforming
materials reports, process control information, trend analyses, and other sources.
A corrective action is initiated to correct the cause of an identifi ed nonconformity
and to prevent it or similar problems from reoccurring. It may include initial and
follow - up actions (e.g., conducted after root cause analysis). Current quality system
models and the cGMP regulations emphasize corrective actions and require that
actions be documented. Under current quality system models, preventive actions
include actions taken in response to quality data to address the cause of potential
nonconformities to prevent their occurrence. An effective CAPA system therefore
includes both reactive and proactive components. The effectiveness of corrective
and preventive actions should be evaluated using objective criteria when possible
and the evaluation documented.
A fi rm ’ s CAPA system and processes should be designed to analyze and respond
to quality issues in a systematic way that is commensurate with the risk. The system
should provide for the verifi cation or validation of corrective and preventive actions
to assure their effectiveness and to assure that actions do not adversely affect the
fi nished product. The system should also assure that pertinent CAPA information
is appropriately disseminated throughout the organization as necessary to assure
the effective operation of the quality system and for management review.
3.2.7.5 Promote Improvement
Continual improvement is a requirement of existing quality system models such as
ISO 9001 - 2000 in which the organization is required to continually improve the
effectiveness of the quality management system through the use of the quality
policy, quality objectives, audit results, analysis of data, corrective and preventive
actions, and management review. In adapting the ISO 9001 - 2000 standard to serve
as a regulatory standard for medical device quality management systems, drafters
of the ISO 13485 standard altered the requirement slightly to require the organization
to “ identify and implement any changes necessary to ensure and maintain the
continued suitability and effectiveness of the quality management system through
the use of the quality policy, quality objectives, audit results, analysis of data, corrective
and preventive actions, and management review. ” The word improvement
was deleted as not an objective of current regulatory standards, but the concept of
continually monitoring the performance of the quality system and appropriately
responding to quality data was retained.
The cGMP regulation does not specifi cally require continual improvement;
however, the regulations are specifi c with regard to the sampling and testing of in -
process materials and drug products, and failure to take reasonable action to reduce
identifi ed sources of variability may be of concern to FDA investigators. The FDA
in its pharmaceutical QS guidance document encourages organizations to promote
improvement through quality system activities and notes that it is critical for senior
management to be involved. Process improvement, along with improvement of in -
process controls, can render a manufacturing process more effi cient and more
robust. The end result can reduce costs and further prevent product failures and
defects from occurring.
3.2.8 TRANSITIONING TO QUALITY SYSTEMS APPROACH
The cGMP regulations assign signifi cant responsibilities to the organizational unit
responsible for quality - related activities. Organizations implementing a quality
system model will be responsible for additional quality - related activities including,
but not necessarily limited to, conducting quality audits, analysis of quality data,
risk assessment, and preventive actions based on review and analysis of quality data
to prevent the occurrence of product nonconformities. In addition, management
is required to provide requisite leadership by actively participating in the
quality system and assuring that the quality system functions as intended. This is
accomplished by establishing a quality policy and associated objectives, planning
for quality, establishing an appropriate organization structure with designated
responsibilities and authorities to appropriately carry out quality system requirements,
providing appropriate resources and training, and periodically reviewing
quality information and data, and assuring that the organization responds
appropriately.
The organizational unit responsible for quality - related activities will in all likelihood
have an even greater role within the organization, and roles and responsibilities
throughout the organization are likely to change. Careful planning will be
required to assure that the transition is effected smoothly with no adverse impact
on product quality. Following are some points to consider in planning the
transition:
• Create a transition team: A cross - functional team should be developed involving
key managers and staff from throughout the organization to plan and
TRANSITIONING TO QUALITY SYSTEMS APPROACH 223
224 ROLE OF QUALITY SYSTEMS AND AUDITS
execute the transition. The transition team should have a clear understanding
of its mission and the organizational objectives associated with the transition.
• Train the transition team: The decision to make the transition must come from
management and management should assure that all individuals on the transition
team receive proper training on quality systems requirements, risk management,
and FDA ’ s recommended approach to quality systems.
• Develop a transition plan: A transition plan, based on clearly defi ned objectives,
should be developed by the transition team.
• Identify staffi ng requirements: The transition will likely affect individual job
descriptions and create additional duties that will have to be addressed through
the reassignment of staff, hiring new staff, and providing necessary training to
all affected staff.
• Identify other resource needs: The plan should include a defi nition of resource
requirements for planning and executing the plan.
• Defi ne roles and responsibilities: the plan should clearly defi ne the roles and
responsibilities of those responsible for development and execution of the plan
for quality system implementation as well as staff roles and responsibilities
under the quality system.
• Consider organizational structure requirements: In order to function properly,
persons responsible for quality - related activities must have the responsibility
and associated authority defi ned and appropriately communicated within the
organization.
• Conduct a gap analysis: The plan should conduct a gap analysis that identifi es
how the quality system model chosen can be effectively integrated with existing
processes to create a quality system that conforms to the organization ’ s quality
objectives, meets regulatory requirements, and is consistent with other organizational
requirements. The quality systems approach is intended to be somewhat
fl exible in application and can be tailored to specifi c organizational
requirements. In order to function properly the quality system must be effectively
integrated into the organization so that it is not viewed as an “ add - on ”
or a set of extra requirements that prevent the “ real ” work from getting
done.
• Consider benchmarking: If possible, arrange with other organizations that have
successfully made the transition to meet with them, review their system, and
discuss transition issues and how they were solved.
• Consult with experts: In addition to benchmarking, seeking assistance from
persons familiar with quality systems can be very helpful, particularly when
existing staff are relatively inexperienced with quality systems. It may be useful
for one or more outside experts to work with the transition team on a regular
basis as a coach or facilitator.
• Communicate regularly: Clear and ongoing communication within the transition
team and with management is essential to effectively coordinate plan
activities, report progress, resolve issues, and identify evolving resource
needs.
• Sell the system: Successful implementation of a QS requires the active and
informed participation of many individuals within the organization. Manage
ment commitment should be clearly communicated and training provided so
that affected staff understand basic quality system concepts and their role in
the quality system.
• Validate the system.
• Maintain regulatory compliance.
3.2.9 AUDIT CHECKLIST FOR DRUG INDUSTRY
The checklist provided in Table 6 [15] is intended to aid in the systematic GMP audit
of a facility that manufactures drug components or fi nished products.
The adequacy of any procedures is subject to the interpretation of the auditor.
Therefore, the author accepts no liability for any subsequent regulatory observations
or actions stemming from the use of this audit checklist.
3.2.9.1 Instructions for Using Audit Checklist
Before starting an on - site audit, plan the audit. Review past audits, note indications
of possible problem areas and items, if any, that were identifi ed for corrective action
in a previous audit. If you are not already familiar with this facility, learn the type
of product produced and how it is organized by personnel and function. What does
your “ customer, ” that is, your superior or senior facility management, expect to learn
from this audit?
1. The checklist is to be used with a notebook into which detailed entries can be
made during the audit.
2. While the checklist is to guide the auditor, it is not intended to be a substitute
for knowledge of the GMP regulations.
3. Although a single question may be included about any requirement, the
answer will usually be a multipart one since the auditor should determine the
audit trail for several products that may use many different components. Enter
details in you notebook and cross reference your comments with the
questions.
4. At least three production batches should be selected for thorough analysis to
include: (a) traceability of all components or materials used in the subject
batches, (b) documentation of raw material or component, in - process, and
fi nished goods testing for the subject product batches, and (c) warehousing
and distribution records as they would relate to a possible recall.
5. Responses entered on the checklist should be consistent. “ X ” is recommended
for “ No ” ; a checkmark for “ Yes ” ; “ N/A ” for not applicable to questions that
do not apply. An asterisk and notebook page number should be entered on
the checklist to identify where relevant comments or questions are recorded
in your notebook.
6. The notebook used should be a laboratory - type notebook with bound pages.
The notebook should be clearly labeled as to the audit type, date, and auditor(s).
Many auditors prefer to use a notebook for a single audit so it may be fi led
with the checklist and the fi nal report.
AUDIT CHECKLIST FOR DRUG INDUSTRY 225
226 ROLE OF QUALITY SYSTEMS AND AUDITS
TABLE 6 Audit Checklist
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
1.0 General Controls
Does the facility and its departments (organizational units)
operate in a state of control as defi ned by the GMP
regulations?
1.1 Organizational & Management Responsibilities
1.101 Does this facility/business unit operate under a facility or
corporate quality policy?
1.102 § 211.22(a) Does a Quality Assurance unit (department)
exist as a separate organizational entity?
1.103 § 211.22(a) Does the Quality Assurance unit alone have
both the authority and responsibility to approve or reject
all components, drug product containers and closures, in -
process materials, packaging materials, labeling, and drug
products?
1.104 § 211.22 Does the QA department or unit routinely review
production records to ensure that procedures were
followed and properly documented?
1.105 § 211.22(b) Are adequate laboratory space, equipment, and
qualifi ed personnel available for required testing?
1.106 If any portion of testing is performed by a contractor, has
the Quality Assurance unit inspected the contractor ’ s site
and verifi ed that the laboratory space, equipment,
qualifi ed personnel, and procedures are adequate?
1.107 Date of last inspection: —
1.108 § 211.22(c) Are all QA procedures in writing?
1.109 § 211.22(c) Are all QA responsibilities in writing?
1.110 Are all written QA procedures current and approved?
(Review log of procedures)
1.111 Are the procedures followed? (Examine records to ensure
consistent record - keeping that adequately documents
testing.)
1.112 § 211.25 Are QA supervisory personnel qualifi ed by way of
training and experience?
1.113 § 211.25 Are other QA personnel (e.g., chemists, analysts,
laboratory technicians) qualifi ed by way of training and
experience?
1.2 Document Control Program
1.201 § 211.22(a) Does the QA unit have a person or department
specifi cally charged with the responsibility of designing,
revising, and obtaining approval for production and
testing procedures, forms, and records?
1.202 § 211.22(d) Does a written SOP, which identifi es how the
form is to be completed and who signs and countersigns,
exist for each record or form?
1.203 § 211.165(a)(b)(c) Is the production batch record and
release test results reviewed for accuracy and
completeness before a batch/lot of fi nished product is
released?
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
1.3 Employee Orientation, Quality Awareness, and Job
Training
1.301 Circle the types of orientation provided to each new
employee: (1) Company brochure. (2) Literature
describing GMP regulations and stressing importance of
following instructions. (3) On - the - job training for each
function to be performed ( before the employee is allowed
to perform such tasks). (4) Other: enter in notebook.
1.302 § 211.25(a) Does each employee receive retraining on an
SOP (procedures) if critical changes have been made in
the procedure?
1.303 Indicate how ongoing, periodic GMP training is
accomplished.
1.304 § 211.25 is all training documented in writing that indicates
the date of the training, the type of training, and the
signature of both the employee and the trainer?
1.305 § 211.25 Are training records readily retrievable in a manner
that enables one to determine what training an employee
has received, which employees have been trained on a
particular procedure, or have attended a particular
training program?
1.306 Are GMP trainers qualifi ed through experience and
training?
1.307 § 211.25(a) Are supervisory personnel instructed to prohibit
any employee who, because of any physical condition
(as determined by medical examination or supervisory
observation) that may adversely affect the safety or
quality of drug products, from coming into direct contact
with any drug component or immediate containers for
fi nished product?
1.308 § 211.28(d) Are employees required to report to supervisory
personnel any health or physical condition that may have
an adverse effect on drug product safety and purity?
1.309 § 211.25(a) Are temporary employees given the same
orientation as permanent employees?
1.310 § 211.34 Are consultants, who are hired to advise on any
aspect of manufacture, processing, packing or holding, of
approval for release of drug products, asked to provide
evidence of their education, training, and experience?
1.311 § 211.34 Are written records maintained stating the name,
address, qualifi cations, and date of service for any
consultants and the type of service they provide?
1.4 Plant Safety and Security
1.401 Does this facility have a facility or corporate safety
program?
1.402 Are safety procedures written?
1.403 Are safety procedures current?
TABLE 6 Continued
AUDIT CHECKLIST FOR DRUG INDUSTRY 227
228 ROLE OF QUALITY SYSTEMS AND AUDITS
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
1.404 Do employees receive safety orientation before working in
the plant area?
1.405 Is safety training documented in a readily retrievable
manner that states the name of the employee, the type of
training, the date of the training, and the name of the
trainer and the signature of the trainer and the
participant?
1.406 Does this facility have a formal, written security policy?
1.407 Is access to the facility restricted?
1.408 Describe how entry is monitored/restricted:
1.409 Is a security person available 24 hours per day?
1.5 Internal Quality/GMP Audit Program
1.501 Does this business unit/facility have a written quality
policy?
1.502 Is a copy of this quality policy furnished to all employees?
1.503 If “ yes ” to above, when provided? —
1.504 Is training provided in quality improvement?
1.505 Does a formal auditing function exist in the Quality
Assurance department?
1.506 Does a written SOP specify who shall conduct audits and
qualifi cations (education, training, and experience) for
those who conduct audits?
1.507 Does a written SOP specify the scope and frequency of
audits and how such audits are to be documented?
1.508 Does a written SOP specify the distribution of the audit
report?
1.6 Quality Cost Program
1.601 Does this facility have a periodic and formal review of the
cost of quality?
1.602 Does this facility have the ability, through personnel,
software, and accounting records, to identify and capture
quality costs?
1.603 Does this facility make a conscious effort to reduce quality
costs?
2.0 Design control
Not directly related to the drug regulation
3.0 Facility control
3.1 Facility Design and Layout
3.101 § 211.42(a) Are all parts of the facility constructed in a way
that makes them suitable for the manufacture, testing,
and holding of drug products?
3.102 § 211.42(b) Is there suffi cient space in the facility for the
type of work and typical volume of production?
3.103 Does the layout and organization of the facility prevent
contamination?
3.2 Environmental Control Program
3.201 The facility is NOT situated in a location that potentially
subjects workers or product to particulate matter, fumes,
or infestations?
TABLE 6 Continued
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
3.202 Are grounds free of standing water?
3.203 § 211.44 Is lighting adequate in all areas?
3.204 § 211.46 Is adequate ventilation provided?
3.205 § 211.46 Is control of air pressure, dust, humidity, and
temperature adequate for the manufacture, processing,
storage, or testing of drug products?
3.206 § 211.46 If air fi lters are used, is there a written procedure
specifying the frequency of inspection and replacement?
3.207 Are drains and routine cleaning procedures suffi cient to
prevent standing water inside the facility?
3.208 § 211.42(d) Does the facility have separate air - handling
systems, if required, to prevent contamination?
(MANDATORY IF PENICILLIN IS PRESENT!)
3.3 Facility Maintenance and Good Housekeeping Program
3.301 § 211.56(a) Is this facility free from infestation by rodents,
birds, insects, and vermin?
3.302 § 211.56(c) Does this facility have written procedures for
the safe use of suitable (e.g., those that are properly
registered) rodenticides, insecticides, fungicides, and
fumigating agents?
3.303 Is this facility maintained in a clean and sanitary condition?
3.304 Does this facility have written procedures that describe
in suffi cient detail the cleaning schedule, methods,
equipment, and material?
3.305 Does this facility have written procedures for the safe and
correct use of cleaning and sanitizing agents?
3.306 § 211.58 Are all parts of the facility maintained in a good
state of repair?
3.307 § 211.52 Is sewage, trash, and other refuse disposed of in a
safe and sanitary manner (and with suffi cient frequency)?
3.4 Outside Contractor Control Program
3.401 § 211.56(d) Are contractors and temporary employees
required to perform their work under sanitary
conditions?
3.402 Are contractors qualifi ed by experience or training to
perform tasks that may infl uence the production,
packaging, or holding of drug products?
4.0 Equipment control
4.1 Equipment Design and Placement
4.101 § 211.63 Is all equipment used to manufacture, process, or
hold a drug product of appropriate design and size for its
intended use?
4.102 Are the following pieces of equipment suitable for their
purpose: blender(s), conveyor(s), tablet, presses, capsule
fi llers, bottle fi llers, other (specify)?
4.103 Are the following pieces of equipment suitable in their size/
capacity: blender(s), conveyor(s), tablet, presses, capsule
fi llers, bottle fi llers, other (specify)?
TABLE 6 Continued
AUDIT CHECKLIST FOR DRUG INDUSTRY 229
230 ROLE OF QUALITY SYSTEMS AND AUDITS
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
4.104 Are the following pieces of equipment suitable in their
design: blender(s), conveyor(s), tablet, presses, capsule
fi llers, bottle fi llers, other (specify)?
4.105 Are the locations in the facility of the following pieces of
equipment acceptable: blender(s), conveyor(s), tablet,
presses, capsule fi llers, bottle fi llers, other (specify)?
4.106 Are the following pieces of equipment properly installed:
blender(s), conveyor(s), tablet, presses, capsule fi llers,
bottle fi llers, other (specify)?
4.107 Is there adequate space for the following pieces of
equipment: blender(s), conveyor(s), tablet, presses,
capsule fi llers, bottle fi llers, other (specify)?
4.108 § 211.65(a) Are machine surfaces that contact materials
or fi nished goods nonreactive, nonabsorptive, and
nonadditive so as not to affect the product?
4.109 § 211.65(b) Are design and operating precautions taken to
ensure that lubricants or coolants or other operating
substances do NOT come into contact with drug
components or fi nished product?
4.110 § 211.72 Fiber - releasing fi lters are NOT used in the
production of injectable products.
4.111 § 211.72 Asbestos fi lters are NOT used in the production of
products.
4.112 Is each idle piece of equipment clearly marked “ needs
cleaning ” or “ cleaned; ready for service ” ?
4.113 Is equipment cleaned promptly after use?
4.114 Is idle equipment stored in a designated area?
4.115 § 211.67(a)(b) Are written procedures available for each
piece of equipment used in the manufacturing, processing,
or holding of components, in - process material, or fi nished
product?
4.116 Do cleaning instructions include disassembly and drainage
procedure, if required, to ensure that no cleaning solution
or rinse remains in the equipment?
4.117 Does the cleaning procedure or startup procedure ensure
that the equipment is systematically and thoroughly
cleaned?
4.2 Equipment Identifi cation
4.201 § 211.105 Are all pieces of equipment clearly identifi ed with
easily visible markings?
4.202 § 211.105(b) Are all pieces of equipment also marked with
an identifi cation number that corresponds with an entry
in an equipment log?
4.203 Does each piece of equipment have written instructions for
maintenance that includes a schedule for maintenance?
4.204 Is the maintenance log for each piece of equipment kept on
or near the equipment?
TABLE 6 Continued
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
4.3 Equipment Maintenance & Cleaning
4.301 § 211.67(b) Are written procedures established for the
cleaning and maintenance of equipment and utensils?
4.302 Are these procedures followed?
4.303 § 211.67(b)(1) Does a written procedure assign responsibility
for the cleaning and maintenance of equipment?
4.304 § 211.67(b)(2) Has a written schedule been established
and is it followed for the maintenance and cleaning of
equipment?
4.305 Has the cleaning procedure been properly validated?
4.306 § 211.67(b)(2) If appropriate, is the equipment sanitized
using a procedure written for this task?
4.307 § 211.67(b)(3) Has a suffi ciently detailed cleaning and
maintenance procedure been written for each different
piece of equipment to identify any necessary disassembly
and reassembly required to provide cleaning and
maintenance?
4.308 § 211.67(b)(3) Does the procedure specify the removal or
obliteration of production batch information from each
piece of equipment during its cleaning?
4.309 Is equipment cleaned promptly after use?
4.310 Is clean equipment clearly identifi ed as “ clean ” with a
cleaning date shown on the equipment?
4.311 § 211.67(b)(5) Is clean equipment adequately protected
against contamination prior to use?
4.312 § 211.67(b) Is equipment inspected immediately prior to
use?
4.313 § 211.67(c) Are written records maintained on equipment
cleaning, sanitizing, and maintenance on or near each
piece of equipment?
4.4 Measurement Equipment Calibration Program
4.401 § 211.68(a) Does the facility have approved written
procedures for checking and calibration of each piece of
measurement equipment? (Verify procedure and log for
each piece of equipment and note exceptions in notebook
with cross reference.)
4.402 § 211.68(a) Are records of calibration checks and inspections
maintained in a readily retrievable manner?
4.5 Equipment Qualifi cation Program
4.501 § 211.63 Verify that all pieces of equipment used in
production, packaging, and quality assurance are capable
of producing valid results.
4.502 § 211.68(a) When computers are used to automate
production or quality testing, have the computer and
software been validated?
4.503 Have on - site tests of successive production runs or tests
been used to qualify equipment?
4.504 Were tests repeated a suffi cient number of times to ensure
reliable results?
TABLE 6 Continued
AUDIT CHECKLIST FOR DRUG INDUSTRY 231
232 ROLE OF QUALITY SYSTEMS AND AUDITS
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
4.505 § 211.63 Is each piece of equipment identifi ed to its
minimum and maximum capacities and minimum and
maximum operating speeds for valid results?
4.506 Have performance characteristics been identifi ed for each
piece of equipment? (May be provided by the
manufacturer but must be verifi ed under typical
operations conditions.)
4.507 Have operating limits and tolerances for performance been
established from performance characteristics?
5.0 Material/component control
5.1 Material/Component Specifi cation and Purchasing Control
Although purchasing is not specifi cally addressed in the
current GMP regulation, incumbent upon user of
components and materials to ensure quality of product,
material, or component.
5.101 Has each supplier/vendor of material or component been
inspected/audited for proper manufacturing controls?
(Review suppliers and audits and enter names, material
supplied, and date last audited in notebook.)
5.2 Material/Component Receipt, Inspection, Sampling, and
Laboratory Testing
5.201 § 211.80(a) Does the facility have current written procedures
for acceptance/rejections of drug products, containers,
closures, labeling, and packaging materials? (List selected
materials and components in notebook and verify
procedures.)
5.202 § 211.80(d) Is each lot within each shipment of material or
components assigned a distinctive code so material or
component can be traced through manufacturing and
distribution?
5.203 § 211.82(a) Does inspection start with visual examination of
each shipping container for appropriate labeling, signs of
damage, or contamination?
5.204 § 211.82(b) Is the number of representative samples taken
from a container or lot based on statistical criteria and
experience with each type of material or component?
5.205 § 211.160(b) Is the sampling technique written and followed
for each type of sample collected?
5.206 Is the quantity of sample collected suffi cient for analysis
and reserve in case retesting or verifi cation is required?
Verify that the following steps are included in written
procedures unless more specifi c procedures are followed:
5.207 § 211.84(c)(2) Containers are cleaned before samples are
removed.
5.208 § 211.84(c)(4) Stratifi ed samples are not composited for
analysis.
5.209 § 211.84(c)(5) Containers from which samples have been
taken are so marked indicating date and approximate
amount taken.
TABLE 6 Continued
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
5.210 Each sample container is clearly identifi ed by material or
component name, lot number, date sample taken, name
of person taking sample, and original container
identifi cation.
5.211 § 211.84(d)(1)(2) At least one test is conducted to confi rm
the identity of a raw material (bulk chemical or
pharmaceutical) when a Certifi cate of Analysis is
provided by supplier and accepted by QA.
5.212 If a Certifi cate of Analysis is not accepted for a lot of
material, then additional testing is conducted by a written
protocol to determine suitability for purpose.
5.213 § 211.84(d)(6) Microbiological testing is conducted where
appropriate.
5.3 Material Component Storage and Handling
Verify that materials and components are stored and
handled in a way that prevents contamination, mixups,
and errors.
5.301 § 211.42(b) Are incoming material and components
quarantined until approved for use?
5.302 Are all materials handled in such a way to prevent
contamination?
5.303 Are all materials stored off the fl oor?
5.304 Are materials spaced to allow for cleaning and inspection?
5.305 § 211.122(d) Are labels for different products, strengths,
dosage forms, etc., stored separately with suitable
identifi cation?
5.306 Is label storage area limited to authorized personnel?
5.307 § 211.89 Are rejected components, material, and containers
quarantined and clearly marked to prevent their use?
5.4 Inventory Control Program
5.401 § 211.142 Are inventory control procedures written?
5.402 Does the program identify destruction dates for obsolete
or out - dated materials, components, and packaging
materials?
5.403 § 211.150(a) Is stock rotated to ensure that the oldest
approved product or material is used fi rst?
5.404 § 211.184(e) Is destruction of materials documented in a
way that clearly identifi es the material destroyed and the
date on which destruction took place?
5.5 Vendor (Supplier) Control Program
5.501 Are vendors periodically inspected according to a written
procedure?
5.502 Is the procedure for confi rming vendor test results written
and followed?
6.0 Operational control
TABLE 6 Continued
AUDIT CHECKLIST FOR DRUG INDUSTRY 233
234 ROLE OF QUALITY SYSTEMS AND AUDITS
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
6.1 Material/Component/Label Verifi cation, Storage, and
Handling
6.101 § 211.87 Do written procedures identify storage time beyond
which components, containers, and closures must be
reexamined before use?
6.102 § 211.87 Is release of retested material clearly identifi ed for
use?
6.103 Are retesting information supplements originally obtained?
6.104 Do written procedures identify steps in the dispensing of
material for production?
6.105 Do these procedures include (1) release by QC, (2)
documentation of correct weight or measure, and (3)
proper identifi cation of containers?
6.106 Does a second person observe weighing/measuring/
dispensing and verify accuracy with a second signature?
6.107 § 211.101(c) Is the addition of each component documented
by the person adding the material during manufacturing?
6.108 § 211.101(d) Does a second person observe each addition of
material and document verifi cation with a second
signature?
6.109 § 211.125(a) Does a written procedure specify who is
authorized to issue labels?
6.110 § 211.125(a) Does a written procedure specify how labels
are issued, used, reconciled with production, returned
when unused, and the specifi c steps for evaluation of any
discrepancies?
6.111 § 211.125(d) Do written procedures call for destruction of
excess labeling on which lot or control numbers have
been stamped or imprinted?
6.2 Equipment/Line/Area Cleaning, Preparation, and Clearance
6.201 § 211.67(b)(5) Do written procedures detail how equipment
is to be checked immediately prior to use for cleanliness,
removal of any labels, and labeling from prior print
operations?
6.202 § 211.67(b)(3) Do written procedures detail any
disconnection and reassembly required to verify readiness
for use?
6.3 Operational Process Validation and Production Change
Order Control
6.301 Have production procedures been validated? (Review
selected procedures for validation documentation.
Adequate?)
6.302 § 211.100(a) Does the process control address all issues to
ensure identity, strength, quality, and purity of product?
6.303 § § 211.101(a) Does the procedure include formulation that
is written to yield not less than 100% of established
amount of active ingredients?
TABLE 6 Continued
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
6.304 § 211.101(c) Are all weighing and measuring preformed by
one qualifi ed person and observed by a second person?
6.305 § 211.101(d) Have records indicated preceding policy been
followed by presence of two signatures?
6.306 § 211.103 Are actual yields calculated at the conclusion of
appropriate phases of the operation and at the end of the
process?
6.307 § 211.103 Are calculations performed by one person? Is
there independent verifi cation by a second person?
6.4 In - Process Inspection, Sampling, and Laboratory Control
6.401 § 211.110(a) Are written procedures established to monitor
output and validate the performance of manufacturing
procedures that may cause variability in characteristics of
in - process materials and fi nished drug products?
6.402 § 211.110(c) Are in - process materials tested at appropriate
phases for identity, strength, quality, purity, and are they
approved or rejected by Quality Control?
6.403 § 211.160(b) Are there laboratory controls including
sampling and testing procedures to assure conformance
of components, containers, closures, in - process materials,
and fi nished product specifi cations?
6.5 Reprocessing/Disposition of Materials
6.501 § 211.115(a) Do written procedures identify steps for
reprocessing batches?
6.502 § 211.115(b) Are quality control review and approval
required for any and all reprocessing of material?
6.503 Does testing confi rm that reprocessed batches conform to
established specifi cation?
6.504 Does a written procedure outline steps required to
reprocess returned drug products (if it can be determined
that such products have not been subjected to improper
storage conditions)?
6.505 Does Quality Control review such reprocessed returned
goods and test such material for conformance to
specifi cations before releasing such material for resale?
7.0 Finished product control
7.1 Finished Product Verifi cation, Storage, and Handling
7.101 § 211.30 Do written procedures indicate how and who
verifi es that correct containers and packages are used for
fi nished product during the fi nishing operation?
7.102 § 211.134(a) In addition, do written procedures require that
representative sample of units be visually examined upon
completion of packaging to verify correct labeling?
7.103 § 211.137(a) Are expiration dates stamped or imprinted on
labels?
7.104 § 211.137(b) Are expiration dates related to any storage
conditions stated on the label?
TABLE 6 Continued
AUDIT CHECKLIST FOR DRUG INDUSTRY 235
Question
Instructions/Questions
(note any exceptions and comments in notebook) Yes, No, or NA
7.105 § 211.142(a) Are all fi nished products held in quarantine
until QC has completed its testing and releases product
on a batch - to - batch basis for sale?
7.106 § 211.142(o) Is fi nished product stored under appropriate
conditions of temperature, humidity, light, etc.
7.2 Finished Product Inspection, Sampling, Testing, and
Release for Distribution
7.201 § 211.166 Has the formulation for each product been tested
for stability based on a written protocol? (Containers
must duplicate those used in fi nal product packaging.)
7.202 § 211.166 Are written sampling and testing procedures and
acceptance criteria available for each product to ensure
conformance to fi nished product specifi cations?
7.203 § 211.170(a) Is a quantity of samples equal to at least twice
the quantity needed for fi nished product release testing
maintained as a reserve sample?
7.204 § 211.167(a) Are sterility and pyrogen testing performed as
required?
7.205 § 211.167(b) Are specifi c tests for foreign particles or
abrasives included for any ophthalmic ointments?
7.206 § 211.167(c) Do controlled release or sustained release
products include tests to determine conformance to
release time specifi cation?
7.3 Distribution Controls
7.301 § 211.150(a) Does a written procedure manage stocks to
ensure that oldest approved product is sold fi rst?
7.302 § 211.150(a) Are deviations to the policy above
documented?
7.303 § 211.150(a) Does a written procedure identify the steps
required if a product recall is necessary?
7.304 Is the recall policy current and adequate?
7.4 Marketing Controls
7.401 The current regulation does not address marketing controls
per se except that all fi nished products must meet their
specifi cations.
7.5 Complaint Handling and Customer Satisfaction Program
7.501 § 211.198(a) Are complaints, whether received in oral or
written form, documented in writing, and retained in a
designated fi le?
7.502 § 211.198(a) Are complaints reviewed on a timely basis by
the Quality Control unit?
7.503 § 211.198(b)(1) Is the action taken in response to each
complaint documented?
7.504 § 211.198(b)(3) Are decisions not to investigate a complaint
also documented and the name of the responsible person
documented?
7.505 § 211.198(b)(2) Are complaint investigations documented
and do they include investigation steps, fi ndings, and
follow - up steps, if required? Are dates included for each
entry?
TABLE 6 Continued
236
7. The references to sections in the GMP regulation are for your convenience
should a question arise. In some instances, two or more sections within the
GMP regulation may have bearing on a specifi c subject. The headings in the
GMP regulation will usually offer some guidance on the areas covered in each
section.
8. A general suggestion for a successful audit is to spend most of your time on
major issues and a smaller portion of your time on small issues. There may be
observations that you may wish to point out to supervisory personnel that
deserve attention but do not belong in an audit report because they are relatively
insignifi cant. By the same token, too many small items suggests a trend
of noncompliance and deserve attention as such. When citing these, be
specifi c.
REFERENCES
1. U.S. Code of Federal Regulations (CFR) , Title 21, Part 211, Current good manufacturing
practice for fi nished pharmaceuticals, available: http://www.accessdata.fda.gov/scripts/
cdrh/cfdocs/cfcfr/CRFSearch.cfm?CFRPart=211 , accessed Dec. 5, 2006 .
2. American National Standards Institute (ANSI) ( 2000 ), Quality management system —
Requirements, ANSI/ISO/ASQ Q9001 - 2000, ANSI, New York.
3. American National Standards Institute (ANSI) ( 2000 ), Quality management system —
Fundamentals and vocabulary, ANSI/ISO/ASQ Q9000 - 2000, ANSI, New York.
4. International Organization for Standardization (ISO) , Application of risk management
of medical devices, ISO 14971:2000, ISO, Geneva.
5. U.S. Department of Health and Human services, U.S. Food and Drug Administration ,
Pharmaceutical cGMPs for the 21st century — A risk - based approach, Final Report —
Fall 2004, September 2004 , available: http://www.fda.gov/cder/gmp/gmp2004/GMP_
fi nalreport2004.htm , accessed Dec. 5, 2006.
6. U.S. Department of Health and Human Services (DHHS) , Food and Drug Administration
( 2006 , June), Guidance for industry: Q9 Quality risk management, DHHS, Rockville,
MD.
7. U.S. Department of Health and Human Services (DHHS) , Food and Drug Administration
( 2004 , Sept.), Guidance for industry: PAT — A framework for innovative pharmaceutical
development, manufacturing, and quality assurance, DHHS, Rockville, MD.
8. U.S. Department of Health and Human Services (DHHS) , Food and Drug Administration
( 2006 , Sept.), Guidance for industry: Quality systems approach to pharmaceutical
CGMP regulations, DHHS, Rockville, MD.
9. U.S. Code of Federal Regulations (CFR) , Title 21, Part 820, Quality system regulation
for medical devices, available: http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/
CFRSearch.cfm?CFRPart=820 , accessed Dec. 5, 2006 .
10. International Organization for Standardization (ISO) , ( 2003 ), Medical devices — Quality
management systems — Requirements for regulatory purpose, ISO 13485:2003, ISO,
Geneva.
11. Juran J. M. , and Godfrey , A. B. , Eds. ( 1999 ), Juran ’ s Quality Handbook , 5th ed. , McGraw -
Hill , New York .
12. FDA compliance policy guide section 130.000, FDA access to results of quality assurance
program audits and inspections (CPG 7151.02), available: http://www.fda.gov/ora/
compliance_ref/cpg/cpggenl/cpg130 - 300.html , accessed Dec. 5, 2006 .
REFERENCES 237
238 ROLE OF QUALITY SYSTEMS AND AUDITS
13. International Organization for Standardization (ISO) , ( 2002 ), Guidelines for quality
and/or environmental management systems auditing, ISO 19011:2002, ISO, Geneva.
14. The Global Harmonization Task Force, SG4, Training requirements for auditors (guidelines
for regulatory auditing of quality systems of medical device manufacturers — Part 1:
General requirements — Supplement 3), available: http://www.ghtf.org/sg4/inventorysg4/
trainingfi nal.pdf , accessed Dec. 5, 2006 .
15. FDA compliance program guidance manual for FDA staff: Drug manufacturing inspections
program (7356.002), 2/1/ 2002 , available: http://www.fda.gov/cder/dmpq/compliance_
guide.htm , accessed Dec. 5, 2006.
16. U.S. Department of Health and Human Services (DHHS) , Food and Drug Administration
( 2004 , Sept.), Guidance for industry: Sterile drug products produced by aseptic processing
— Current good manufacturing practice, DHHS, Rockville, MD.
17. U.S. Department of Health and Human Services (DHHS) , Food and Drug Administration
( 2001 , Aug.), Guidance for industry: Q7A good manufacturing practice guidance for
active pharmaceutical Ingredients, DHHS, Rockville, MD.
239
3.3
CREATING AND MANAGING A
QUALITY MANAGEMENT SYSTEM
Edward R. Arling , Michelle E. Dowling , and Paul A. Frankel
Amgen, Inc., Thousand Oaks, California
Contents
3.3.1 Introduction
3.3.2 Understanding a Quality Management System
3.3.2.1 Defi ning Quality Management Systems
3.3.2.2 Synthesis versus Analysis
3.3.2.3 System versus Process
3.3.2.4 Business Benefi ts of Establishing a Robust Quality Management System
3.3.2.5 Industry and Regulatory Expectations
3.3.3 Management and Staff: Leadership and Support
3.3.3.1 Outlining Benefi ts to the Enterprise
3.3.3.2 Speaking Management Language
3.3.3.3 Translating Benefi ts to Staff
3.3.3.4 Ensuring Staff Support and Management Leadership
3.3.3.5 Traps to Avoid
3.3.4 Establishing Quality Management System Scope
3.3.4.1 Defi ning Business Requirements
3.3.4.2 Integrating Quality Management System into Quality Plans
3.3.4.3 Determining Process Resolution Requirements
3.3.4.4 Scalability to Enterprise
3.3.5 System and Process Ownership: Roles and Responsibilities
3.3.5.1 Quality Management System Ownership and Management
3.3.5.2 Process Ownership
3.3.5.3 Process Owner Selection
3.3.5.4 Stakeholder/Process Owner Integration
3.3.5.5 Decision Authority
3.3.5.6 Industry Knowledge
3.3.5.7 Regulatory Inspection and Audit Lead
3.3.5.8 Subject Matter Expert
3.3.5.9 Metric Ownership
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
240 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
3.3.5.10 Documentation Ownership
3.3.5.11 Training
3.3.5.12 Risk Management
3.3.5.13 Continuous Improvement and Project Management
3.3.5.14 Nonconformance / CAPA / Planned Deviation Ownership
3.3.6 Change Management/Communication
3.3.6.1 Managing Organizational Change
3.3.6.2 Communication
3.3.6.3 Feedback and Alignment
3.3.6.4 Training
3.3.7 Measuring Success through Meaningful Metrics
3.3.7.1 Performance Metric Development
3.3.7.2 Metric Review
3.3.7.3 Maturity Model
3.3.7.4 Meeting Process Maturity Requirements
3.3.8 Driving Continuous Improvement: Projects
3.3.8.1 Process Improvements
3.3.8.2 Process Improvement Proposal
3.3.8.3 Task versus Project
3.3.8.4 Project Metrics
3.3.9 Ensuring Ongoing Success
3.3.9.1 Establishing Mutual Goals
3.3.9.2 Rewards and Recognition
3.3.9.3 Ensuring Ongoing Program Continuity
3.3.9.4 Program Institutionalization
References
3.3.1 INTRODUCTION
The world ’ s population continues to grow and the average life expectancy continues
to increase. Pharmaceutical and biopharmaceutical products are more in demand
as the population expands, requiring novel and specialized medications to treat
common and debilitating diseases. The industry is challenged to rapidly discover
and commercialize products to treat existing unmet medical needs and emerging
threats as viruses mutate into new diseases that threaten the stability of the world
as we know it.
At the same time, the global marketplace continues to increase its demand on
the industry. Government, consumer, and wholesale buying pressures demand lower
prices. Higher quality standards are expected by regulators and consumers. Competition
continues to increase from generic, biosimilar, and counterfeit producers.
Developing nations, with lower cost overheads, are developing economical production
capabilities. Meanwhile, research and development costs are increasing.
This chapter will outline the concepts, benefi ts, and practical implementation
steps for developing a comprehensive quality management system (QMS) that supports
pharmaceutical and biopharmaceutical manufacturing operations. The material
presented is universal in its utility, applicable to small and large companies,
development, and commercial enterprises. A QMS is a proactive, structured approach
UNDERSTANDING A QUALITY MANAGEMENT SYSTEM 241
to supporting development and manufacturing operations. It includes all processes,
metrics, management review, and continuous improvement activities. The QMS, as
described in this chapter, is further supported through an active change management
program and application of annual quality plans to ensure ongoing system
sustainability.
A well - designed QMS, with mature, developed processes, provides the required
infrastructure and support necessary for successful manufacturing operations. Integrated
processes, proactively managed, that can be quickly modifi ed to meet changing
business and regulatory demands will support ongoing manufacturing operations
and provide competitive advantage. This chapter provides guidance on creating and
managing a robust QMS that supports manufacturing operations in the pharmaceutical
and biopharmaceutical industry.
3.3.2 UNDERSTANDING A QUALITY MANAGEMENT SYSTEM
Every development, testing, manufacturing, packaging, warehouse, or distribution
facility has its own unique role in producing an output or product for consumption
by a customer somewhere in the pharmaceutical or biopharmaceutical supply chain.
Each facility and organization is critically dependent upon several different processes
that function interdependently producing the desired output. Organizations ’
survival and profi tability are directly linked to the effi ciency of design, execution,
performance, and interrelational attributes of these processes. Throughout a product
life cycle, from early discovery through development, scale up, clinical testing,
product technology transfer, registration, approval, commercialization, and eventually
product discontinuance, robust processes are the foundation supporting the
successful enterprise.
Manufacturing support processes are discrete in their output, but interrelated in
their overall effect. Weak or ill - defi ned processes have a diminishing overall effect
on the organization and its product. It manifests itself as increased rework, rejected
material, extended cycle times, delayed disposition, high nonconforming performance
metrics, complaints, recalls, or other inabilities to meet customer or market
demands. A comprehensive QMS may encompass all the processes supporting
development and manufacturing. It includes the standards, policies, and procedures
required to measure those processes for performance and maturity. It provides
metrics necessary for leadership to perform risk - based prioritization and focus
resources for business improvement and regulatory compliance.
Robust processes will have owners that have defi ned roles, responsibilities, and
accountabilities. These process owners must be fully dedicated to their process. They
must know their process capabilities and expectations, the interrelationship between
their process and other processes and manage them like a business unto themselves.
Functional management must support process owners, and leadership must understand
and lead the QMS effort as an ongoing program, treating it as the integral
part of the business that it is.
A QMS is an organizational approach consisting of people, interrelated processes,
process inputs and outputs, and structured review programs that lead to
ongoing continuous improvement. This complexity of processes requires a programmatic
organization and management to effectively interrelate its components. A
242 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
QMS program offi ce is required to provide the organizational benefi ts expected
from well - managed processes and should be one of the fi rst elements established
when instituting the program.
A QMS and the processes comprising it are not the sole responsibility of the
quality function or a single functional group. Inherently, these processes have no
bounds in the organization. The concept must be owned, managed, or executed by
all staff from leadership to the most entry - level manufacturing associate. A quality
mindset must be part of every employee that contributes to the discovery, manufacture,
packaging, testing, warehousing, and shipping of a product or output. A
culture of quality and understanding of the processes in which personnel work are
essential to advance the QMS to maximize benefi ts to the enterprise and remain
competitive.
Instituting a QMS through a holistic approach that supports manufacturing operations
has the potential to meet and exceed customer, patient, shareholder, and
employee expectations. It requires a cross - functional team approach, with proactive
management of all the processes responsible for manufacture, including functional
support from development, manufacturing, analytical, engineering, and quality
assurance. The development and maintenance of a tested, robust QMS requires time
and resources. Full maturation of processes and organizational culture change may
take, in some cases, years to fully implement and realize benefi ts, but worth the effort
and time. Signifi cant QMS issues should not be addressed as one - off fi xes. Rather,
action taken to remediate defi cient processes should be approached as long - term
corrections, addressing the root cause of the failed process, so they do not repeatedly
plague the organization.
The ultimate responsibility for a robust, functional QMS lies with top management.
The organization follows the leadership, and therefore, leadership must
support a QMS that is specifi cally designed for the organization, be aware of and
monitor its progress and contribution to the organization, and frequently support,
guide, and maintain it. Doing so ensures viability of the QMS, and in turn the QMS
will provide leadership the data and guidance necessary to effectively manage the
organization.
3.3.2.1 Defi ning Quality Management Systems
The term system or quality system is used with surprising inconsistency throughout
the pharmaceutical and biopharmaceutical industry and by government regulators.
Even within a single company or within a department, the terms can be nebulous
in their use and interpretation. System is often used to describe an individual process
or unit operation. Often, the term system is used so narrowly as to describe an
individual policy, standard, or even a single procedure.
Recent initiatives by global organizations such as ISO (International Organization
for Standardization, www.iso.org ) and ICH (International Conference on
Harmonization, www.ich.org ) are attempting to bring consistency in concept and
standardization in defi nition to the QMS. In 2004, the Pharmaceutical Inspection
Co - Operation Scheme (PIC/S, www.picscheme.org ) issued its recommendation on
Quality System Requirements for Pharmaceutical Inspectorates. The U.S. Food and
Drug Administration (FDA) initiated inspection surveillance approaches based
upon QMS organization and is another source of defi nition and interpretation.
UNDERSTANDING A QUALITY MANAGEMENT SYSTEM 243
Inconsistency in language and expectations continues to exist; however, efforts are
progressing to minimize distinctions and globally harmonize efforts, structure, and
language concerning quality systems.
According to Webster ’ s dictionary, system is defi ned as a regularly interacting or
interdependent group of items forming a unifi ed whole; a group of interacting
bodies under the infl uence of related forces . . . an assemblage of substances that is
in or tends to equilibrium . . . a group of organs that, when together, perform one or
more vital functions . . . an organization forming a network especially for distributing
something or serving a common purpose . . . an organized set of doctrines, ideas,
or principles usually intended to explain the arrangement or working of a systematic
whole [1] .
The vocabulary and defi nitions used in this chapter defi nes a quality management
system as the compilation of all the processes required to support the manufacture,
packaging, testing, release, and distribution of an active pharmaceutical ingredient
(API) or drug product. It is aligned with that of the FDA Center for Drug Evaluation
and Research (CDER) compliance program 7356.002, issued to investigators
for the inspection of pharmaceutical and biopharmaceutical manufacturing plants
( www.fda.gov/IOM 7356.002). The CDER inspection program subdivides the processes
comprising the QMS into six subsystems: quality, facilities/equipment, production,
materials control, laboratory controls, and packaging and labeling.
There are no specifi c CDER requirements as to which processes belong under
each subsystem; however, one can easily follow the outline provided in 21 CFR Part
211, the regulations applicable to human drug product manufacture, to aid in the
determination of processes likely to be inspected during a regulatory inspection
( www.fda.gov ). The FDA subdivides all the processes comprising a company ’ s QMS
into six subsystems to ensure adequate and varied coverage during inspections. See
Figure 1 . Using the same process organization structure and vocabulary as regulators
provides an enterprise the advantage of more effi cient inspection preparation
and avoidance of miscommunication during and after regulatory inspections.
The CDER subsystem organization provides regulators and management the
ability to focus attention to specifi c functional areas. Table 1 is an example of the
processes, organized under appropriate subsystems, supporting a typical API or drug
fi ll - and - fi nish operation. These subsystems are organized according requirements
found in regulations used by investigators during inspections, 21 CFR Part 210, 211,
and the unit operations and support processes necessary for production.
One size does not fi t all situations. Each enterprise has the responsibility and
latitude to design a QMS to meet its specifi c needs. Even facilities with very similar
manufacturing operations may require different processes to support the business.
Each manufacturing organization requires a customized set of processes which will
comprise its QMS. The management group responsible for the QMS should be able
to identify and justify the processes comprising the system. There is not a single set
of processes that can be universally applied to all operations, as each organization
is unique in its business, product output, organization, culture, as well as local and
global regulatory and customer requirements.
Processes identifi ed as part of the QMS can be organized into the appropriate
CDER subsystems for the purpose of aligning with the methodology used during
inspections. It also provides management the ability to determine areas of strength
or opportunities for improvement within the QMS. Regulators will always include
244 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
FIGURE 1 Subsystems and management relationship.
TABLE 1 Quality Management System Subsystems and Processes
Quality Facilities/equipment
Audits and inspections Facility and equipment design
Management review Equipment maintenance
Risk management Equipment cleaning
Organization and personnel Calibration
Training Materials control
Document management Supplier quality management
Change control Sampling and inspection
Nonconformances Receiving, warehouse, and storage
Corrective and preventative actions Inventory management
Biological product deviation Transport
Product disposition Return and salvage
Validation Laboratory controls
Production Laboratory testing
Manufacturing Sample management and sample plans
Process monitoring Stability program
Environmental and gowning monitoring Packaging and labeling
In - process controls Labeling controls and approvals
Gowning Package development
UNDERSTANDING A QUALITY MANAGEMENT SYSTEM 245
a focus on the processes within the quality subsystem. Other subsystems will be
reviewed during inspections based upon the type of inspection and compliance
history of the enterprise. More information on how the FDA focuses inspections
based on quality system and subsystem organization is available at the FDA website
( www.fda.gov ) or articles written on this subject [2] .
To maximize the effect of a QMS, it should be designed to be scalable and transferable
throughout the enterprise and easy to understand and execute. An adequately
designed QMS results in increased effi ciency, a compliant operation, and
staff satisfaction.
3.3.2.2 Synthesis versus Analysis
With systems thinking, the whole is greater than the sum of its parts. Systems rely
upon the interaction of several processes. An individual process has limited value
on its own, regardless of the level of development it has achieved. Processes provide
value to the system through synthesis with other processes.
In the early twentieth century, researchers began to recognize the existence of
interdependent relationships and organizational patterns among seemingly discrete
parts. It is the relationships that allow parts to function as a whole. The “ perceived
whole ” is a system. Systems thinking involves considering the parts in the context
of that whole. In systems thinking:
• Everything in a system is related to everything else in the system.
• The parts of a system work together to achieve the overall objective of the
whole system.
• In addition to the immediate effects of an action, there will be other consequences
that ripple through the system.
• Every change brings benefi ts and consequences.
• Changing or reinforcing patterns and relationships within a system is as necessary
to achieving the goals of the system as changing or retaining the parts of
the system.
• Systems are “ living ” entities that sustain themselves through self - regulating
dynamic equilibrium and organize to respond to externally imposed change.
Viewing a QMS in this context is benefi cial to organizational leadership and management
responsible for the system. It puts into perspective the overall effect on an
organization that is achievable by individual processes alone and what can be
achieved and sustained through active management and the interaction between
those processes.
3.3.2.3 System versus Process
Traditional industry paradigm has the Quality Department responsible for quality
and the Manufacturing Department responsible for producing product. Inherent
confl ict exists in this model due to competing functional priorities. By building
quality concepts and accountabilities into production processes responsible for
production, quality becomes infused into the organization. Both Quality and
246 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
Manufacturing therefore share the common goal of supplying high - quality product
through the effi cient execution of their processes.
Historically, very few processes were regarded as “ quality systems, ” and they were
viewed as something owned by the Quality Department. These “ systems ” were in
fact ill defi ned and nonrelated processes used to monitor or detect individual actions
and activities occurring in the manufacturing environment. These systems were
based on quality control (QC) type of responsibilities for testing quality into the
product. Examples include raw material testing, in - process and fi nished - product
testing, nonconforming material review, environmental monitoring, and release and
distribution. Few were interrelated with other processes, actively supported by
management, or reviewed by leadership for performance or compliance.
The QC monitoring processes described above, if supported, were limited in their
ability to support improvements and could only lead to action that was reactive in
nature. Process integration is weak or nonexistent. Neither process maturity and
development nor proactive system management is achievable. In the past, QMS
enhancement was viewed as an expense and not seen as a relational contributor to
the value chain. Aware management now realizes, through regulatory action, penalty
and fi nes, delayed product approvals, recalls, and the like that establishment of a
comprehensive QMS is essential to survive in the current regulatory environment
and remain competitive in the business environment.
With the advancement of quality assurance (QA) principles and concepts at the
end of the last century, QMSs have evolved to be more proactive to include change
control, supplier and internal auditing, risk management, lagging and leading metric
collection, and review. Review of predictive metrics has become the basis for preventive
action and continuous improvement programs. Today ’ s competitive environment
obligates leading manufacturers and world - class organizations to apply
proactive system thinking to expand their focus to include all processes that support
product quality, irregardless of the stage of development or manufacture. Early
implementation of appropriate processes supports quality - by - design concepts and
practice, within the framework of a QMS and ensures quality in all processes and
provides the foundation for good investigations and continuous improvement.
A QMS should be comprised of all the processes supporting that business and
include an effective management review of those process metrics. Management
needs to be aware of and understand process performance through structured
metrics review programs in order to take appropriate action, providing resources
and capital to improve the QMS. This hierarchy is illustrated in Figure 2 .
Processes supporting and applicable to pharmaceutical and biopharmaceutical
manufacturing are easily determined by examining the business needs of the organization
and the regulations governing them. A carefully designed QMS will consider
the needs of the enterprise as a whole, as well as that of the individual unit
operations comprising the enterprise. If the QMS design is comprehensive, it will
provide signifi cant value to global and local management. It will support staff by
standardizing processes, requirements, and expectations and provide leadership
meaningful and comparable metrics on system and process performance. Changes
can be quickly facilitated and implemented when process modifi cations are required.
A consistent representation of processes to regulators builds confi dence and trust
that the enterprise is capable to produce the product for which approval has been
granted.
UNDERSTANDING A QUALITY MANAGEMENT SYSTEM 247
3.3.2.4 Business Benefi ts of Establishing a Robust Quality Management System
The competitive nature of the pharmaceutical business demands capable and effi -
cient processes supporting discovery, development, technology transfer and scale -
up, and commercial manufacturing and distribution. Execution of effi cient processes
is the foundation for new and ongoing enterprises to be successful. It is the basis
for successful manufacture and the bedrock upon which management and regulators
can gauge the capability level of the enterprise. Providing patients with needed
medicines in a timely, cost - effi cient manner, without delay due to manufacturing or
compliance issues, should be a primary driving force behind the pharmaceutical and
biopharmaceutical industry.
Leadership may ask the question: Why implement a quality management system?
The answer is that a well - designed system is necessary to establish a state of control
to ensure that a high quality, safe, and effi cacious product is produced and available
for patients. Quality systems as described in the forthcoming ICH Q10 guidance is
the logical complement to its predecessors, ICH Q8 (Product Development) and
ICH Q9 (Risk Management) ( www.ich.org ). These three guidance documents build
upon each other from quality - by - design activities in development through the entire
product life cycle. When used together, the guidance documents maximize their
benefi ts to the enterprise through better process understanding, less regulatory
scrutiny, and increased freedom to operate. Together, these guidance ’ s support more
effi cient product life - cycle management from discovery through development and
commercialization.
Ineffi cient operations cost businesses untold amounts in fi nancial and human
capital. A poorly designed system coupled with ineffi cient processes may result in
rework of development and commercialization activities, data integrity issues, inef-
fi cient use of resources, and delay in approval. Poorly designed processes may also
FIGURE 2 Quality management system hierarchy.
248 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
lead to loss of future revenue with business partners and have a negative regulatory
consequence.
A recent study conducted by the Pharmaceutical Manufacturing Research
Project, a joint venture by Georgetown University and Washington University in St.
Louis business schools, collected data from 42 manufacturing facilities owned by 19
companies to determine factors that affected industry performance. The Final
Benchmarking Report assessed performance in terms of manufacturing times, frequency
of deviations from manufacturing standards, reasons for deviations, manufacturing
yield, and rates of improvement for those metrics.
The study determined that improvements in manufacturing process could save
industry more than $ 50 billion in manufacturing costs, which the researchers believe
could result in lower drug prices and more money for R & D. The report received no
industry or government funding [3] .
Leadership is both challenged and rewarded for supporting the development of
a robust QMS. On one hand, it takes time and resources to design and develop a
comprehensive program. Immediate return on this investment is not usually forthcoming.
Management is typically under pressure to deliver aggressive results in a
short time period, which is counterintuitive to careful planning and long - range
development. Conversely, proactively formalizing and supporting a robust QMS
will, in the long run, ensure the operations freedom to operate (regulatory compliance)
and deliver business effi ciencies.
In the pharmaceutical and biopharmaceutical manufacturing industry, the perception
of quality has dramatically changed over the past several years, and loss of
market capitalization can be a direct correlation to this perception. Large pharmaceutical
companies have gone from some of the world ’ s most admired companies to
losing signifi cant percentage of their value, based on consumer, media, and investor
perceptions of quality and ethics. Speaking at a recent Parenteral Drug Association
(PDA)/FDA joint regulatory conference, Daniel Diermeier, IBM Distinguished Professor
of Regulation and Competitive Practice, Northwestern University stated: “ The
perception of quality on the pharmaceutical value chain is greater than in other
industries (auto, furniture, etc.). Patients cannot assess the quality of drugs as they
can a car or hotel room. In healthcare, the ‘ value proposition ’ is higher than other
industries and the Quality [Management] System is a critical subset of that perception
” [4] . Dr. Diermeier goes on to suggest a QMS include processes for decision
and detection to further protect the “ value proposition ” of the enterprise.
Enterprises lacking individual capable processes experience degrees of negative
effects throughout the organization. This is true for processes that support discovery,
development, manufacturing, or marketing. Recent examples of fi nes imposed by
regulators for poor processes supporting the QMS are increasing (see Table 2 ).
These costs are only indicative of the fi ne itself and do not include lost revenue, cost
of consultancy for remediation, decreased shareholder value, and diminished staff
morale and support. These costs are typically an order of magnitude or more greater
than the fi ne itself.
A common misconception of pharmaceutical and especially smaller biopharmaceutical
companies is that the implementation of a robust QMS is not required in
areas other than commercial manufacturing. Small, biotech start - up companies also
tend to delay the implementation of well - designed processes until they near the
approval stage, focusing the organization instead for product approval or sale. This
UNDERSTANDING A QUALITY MANAGEMENT SYSTEM 249
can become a costly miscalculation, as speed to market and limited capital demand
processes supporting effi cient development, clinical and regulatory submission processes
be executed with minimal waste or rework.
Although QMSs are routinely identifi ed with commercial manufacturing, it is
critical to establish process parameters for discovery, development, and technology
transfer, including scale - up, characterization of process, analytical methodology, and
validation. Development activities are executed more effi ciently through the application
of robust processes and ultimately become the foundation for robust manufacturing.
Failed development studies, inadequate comparability reports, clinical
studies requiring repeated, or poorly supported analytical and process characterization
contribute to delayed submissions and weak regulatory submission and inspection
presentation. The identifi cation of processes supporting these activities, owner
identifi cation and accountability and support will ensure success of the enterprise
and reduce the anxiety and uncertainty that is inherent in development and approval
activities.
Several opportunities exist for pharmaceutical and biopharmaceutical manufacturing
plants to improve effi ciency and cost savings, which ultimately validate the
program ’ s benefi ts and supports leadership in achieving their fi nancial goals. Traditionally,
the industry environment is heavily regulated and has been very risk
adverse. These two elements combine to offer countless opportunities to improve
ineffi cient and ill - defi ned processes, clarify process scope, defi ne process owner
accountabilities and responsibilities, and remediate process duplication or gaps.
Performing ineffi cient processes for the sake of avoiding regulatory scrutiny or
attempting to defend poorly characterized processes without adequate data and
interpretation becomes self - defeating to the industry. Poor prioritization of work,
ill - defi ned process relationships, and functional management interference or neglect
may also contribute to ineffi ciency. Staff requires processes that are easy to execute,
well integrated, and result in value - added activities. This can only be accomplished
through the design and execution of effi cient processes that are interrelated, bringing
value to the enterprise, process owners, and stakeholders.
An example of a robust process is the design, development, and operation of a
nonconformance process. Regulations require an operational process to identify,
document, and correct nonconformances occurring in licensed pharmaceutical manufacturing
facilities for approved products. Companies spend signifi cant human
TABLE 2 Potential Financial Impacts
Company Compliance Issue Type of Impact
Cost to Business
( $ Mil)
A Failure to follow procedure
Inadequate training
Multiple 483
observations
< 1
B Inadequate process defi nition,
controls, and oversight
Warning letter > 1
C Repeat observations — direct product
impact Failure to meet warning
letter commitments
Consent decree > 100
D Plant shutdown Direct fi nes product
stock - out
> 500
250 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
capital identifying, documenting, and tracking nonconformances. But how much is
actually being done to remediate these nonconformances? Can the nonconformances
be related to previously completed development or commercialization
studies? Is the nonconformance process suffi ciently related to an effective corrective
or preventive action (CAPA) process? Does the preventive action interrelate
effi ciently with an effi cient change control process to ensure proposed changes
remain in compliance with registrations? Are the documentation and training processes
suffi cient to support approved changes? An adequately designed QMS will
ensure the supporting processes are present and that functional and interrelationships
established. A systems implementation provides a holistic approach, which
results in both building effective individual processes and interrelating those processes
to maximize their effect on the business, driving effi cient and science - based
activities.
Maintaining good manufacturing practice (GMP) compliance is essential for
pharmaceutical and biopharmaceutical companies. Results of noncompliance are
costly fi nes, loss of revenue, higher overhead costs, delayed approvals, and poor
customer and regulatory perceptions. Poor compliance results from an inadequately
designed QMS that lacked the processes and management review required to
support the enterprise. Processes supporting compliance include self - audits, change
control, document revision and approval, and staff training programs. Regular management
review of these processes will ensure resources are allocated to appropriate
initiatives and there should be no surprises during inspections. A well - designed
QMS should prevent negative regulatory consequences. Effi cient and compliant
processes support lean manufacturing efforts through the documentation and
understanding of processes. Management review of these processes ensures that
leadership awareness, support, and action is taken by the organization when
appropriate.
Figures 3 and 4 illustrate how a biennial document review process and document
processing cycle time metrics faltered in their early stages due to lack of process
ownership, defi nition, and management review. This situation presented a compliance
risk to the organization and resulted in poor business effi ciencies. Improve-
FIGURE 3 Biennial document review process.
11%
19%
15%
55%
87%
92%
84%
88%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Q1/05 Q2/05 Q3/05 Q4/05 Q1/06 Q2/06 Q3/06 Q4/06
Reviews complete
Actual
Target
UNDERSTANDING A QUALITY MANAGEMENT SYSTEM 251
ment was attained by assigning a process owner who defi ned and improved each
process, developed meaningful metrics, and presented those metrics to management.
Management became aware of process performance, understood the compliance
risk and business impact and took appropriate action to focus staff efforts to meet
process requirements. Results were improved document review cycles, proactive
compliance with internal procedures and regulatory requirements, and the satisfaction
of knowing that no additional effort was required to achieve better business
results and regulatory compliance.
3.3.2.5 Industry and Regulatory Expectations
While there are no requirements for a “ quality system ” in current FDA regulations
applicable to pharmaceutical and biopharmaceutical manufacturing, regulatory
agencies and industry trade organizations are increasingly recognizing the importance
of robust, functioning quality systems in support of manufacturing the world ’ s
medicinal products. The FDA realizes not all quality principles are represented in
current GMP regulations for drug products (21 CFR Part 211), which were last
updated in 1978.
Quality management system issues and their association with risk management
are common topics discussed in trade and regulatory seminars and conferences.
Recent guidelines such as FDA “ Quality Systems Approach to Current Good Manufacturing
Practice Regulations ” found on the FDA website and part of FDA ’ s initiative
titled “ GMP ’ s for the 21st Century ” was written to complement existing
regulations. While the FDA guidance may change or even become redundant with
the issuance of ICH Q10, there is common intent among industry and government
to advance quality management systems. According to Joe Famulare, Director
DMPQ, FDA, the “ FDA wanted to write a comprehensive Quality System model
that would support and correlate with CGMP regulations. The guidance is consistent
with defi ning a state of control; facilitate quality efforts, change control, Quality by
Design, and risk management ” [4] .
In discussing quality systems at a recent industry conference on GMPs, Chris
Joneckis of the FDA CBER (Center for Biological Evaluation and Research) had
FIGURE 4 Document review cycle time.
0
10
20
30
40
50
60
Q1/05 Q2/05 Q3/05 Q4/05 Q1/06 Q2/06 Q3/06 Q4/06
Month
Number of days
Actual
Target
252 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
this to say: “ A robust Quality Management System makes a strong case for quality
product. It is a win, win, win — for patient, industry and regulators. It benefi ts technology
transfer, process control, monitoring, capability, improves manufacturing,
fewer nonconformances and better quality of investigations. Regulatory benefi ts
include enhanced Chemistry, Manufacturing, Controls (CMC) review, change
control, and submission of postapproval changes ” [5] .
Regulatory and industry guidance documents have been generated in support of
developing and organizing quality systems. In the late 1990s, the system - based
inspection approach was formalized by the Center for Devices & Radiological
Health (CDRH) of the FDA [6] . These regulations were codifi ed as QSR, Quality
Systems Regulations, and are included in Part 820 of the Code of Federal Regulations
(CFR).
The CDER and CBER soon followed the CDRH approach and issued their own
Compliance Program Guidance Manuals, 7356.002 [7] and 7345.848 [8] , respectively,
which were modeled on the CDRH QSR approach. The CDER and CBER are
responsible for ensuring the biennial inspection of pharmaceutical and biopharmaceutical
manufacturing facilities. The guidelines listed here are used by investigators
during manufacturing inspections. Process owners and stakeholders as well as management
and leadership should be familiar with these compliance manuals and how
investigators plan to use them during inspections.
Current FDA inspectional surveillance, based on the models described above,
requires investigators evaluate the processes within the subsystems defi ned by the
QMS to determine compliance and risk to patient safety. This is different than the
traditional approach of reviewing individual products during inspections. There is
subtle, yet signifi cant advantage to both the regulating agencies and compliant
companies by using a system approach, as the inspections are designed to be faster
and cover many product types during one inspection. Companies with compliant
histories can benefi t with nominal inspections, whereas companies with noncompliant
histories will receive more regulatory scrutiny and possible regulatory action.
The movement by industry groups such as the ISO, which attempts to provide
recognized standards for many industries, was also grounded in a systems approach
with the publication and certifi cation of ISO 9000 series and later with ISO 2000:9004
( www.iso.org ), which is based on QMS establishment and eventually continuous
improvements once processes become stable.
The ICH, a joint regulatory – industry initivative on international harmonization
for drug development and approval, also recognizes the value and contribution of
a quality systems approach through its guidance development on this topic (ICH
Q10). The pharmaceutical and biopharmaceutical industry and regulatory agencies
are collaborating to fi nalize the guidance sometime in 2008. ICH Q10 is focused on
pharmaceuticals and is intended to align GMP requirements with a quality system
approach. It will be applicable to drug substance and drug product, large and small
molecule products, and harmonize one approach to quality systems. It also will
complement ICH Q8 and ICH Q9. ICH Q10 contains a pharmaceutical context
emphasizing a comprehensive approach; key elements included are management
response and continuous improvement. Several ICH guidance documents are
already adopted by regulatory agencies, such as ICH Q7A, for the manufacture of
APIs. As these guidance documents are adopted, they often become the basis for
regulatory expectations and inspections.
3.3.3 MANAGEMENT AND STAFF: LEADERSHIP AND SUPPORT
All manufacturing operations operate, to some extent, with elements and components
of a quality management system. Those elements and processes may not be
recognized or managed as though they are an integral part of a larger system and
may be primarily reactive in nature. Signifi cant time and resources are required to
change an organization ’ s culture and practices to move existing elements from a
fragmented, reactive program to a defi ned structure that is proactively managed.
The degree to which a program is proactively managed and supported by its leadership
is directly related to the benefi ts experienced by the organization.
Three distinct levels of support are required for successful implementation of a
QMS program: executive leadership, functional management, and operational staff.
All three levels of the organization must support the effort to attain success. Delivering
program understanding and benefi ts to each should be a priority to ensure
acceptance and continuity. Motivating staff and leadership, through benefi ts and
business results, is important to ongoing program sustainability.
Leadership requires capable and dedicated staff to design and maintain a dynamic
QMS program. Leadership must embrace the program and support it throughout
the organization. Functional management must understand the program in order to
support it and direct its staff in execution of the program. Staff must understand
what the program means to them and experience and realize the benefi ts in order
to support it.
The quality organization must be seen as a partner in assuring product quality,
not the department that disseminates quality. Within a QMS, certain processes are
owned by the quality function, just as manufacturing, engineering, development,
technical support, and facilities own processes within the system. All functional
groups should have defi ned roles and responsibilities to ensure quality product is
produced. Cross - functional support and delineation of responsibilities ensure quality
is built into every process, and each process owner is ultimately responsible for his
or her process output. Leadership that understands and embraces this concept will
support and infuse a culture of quality throughout the organization, maximizing the
probability of success and competitive advantage.
The organizations leadership, management, staff, and QMS program group must
work together to develop and progress the QMS. A successful program should detail
expected benefi ts for all stakeholders in the organization and provide ongoing
results demonstrating functionality and utility.
3.3.3.1 Outlining Benefi ts to the Enterprise
Establishing a formal, structured QMS for an organization requires leadership
approval, resources, and capital. Leadership support and approval is the place to
initiate the program to ensure all program efforts are supported and the proposed
system meets the business needs. This includes having dedicated resources that can
focus their efforts to design and manage the program and operate and manage the
processes.
Leadership has visibility to present business needs and budget and the vision and
insight for the organizations ’ future. Quality management system design needs to
fulfi ll present and future needs to be robust and value added. A gap analysis on
MANAGEMENT AND STAFF: LEADERSHIP AND SUPPORT 253
254 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
current business processes can help leadership understand where opportunities exist
for improving processes. These gaps can be determined by analyzing the purpose of
the organization and its ability to deliver quality results on time and on budget.
Manufacturing areas to examine for operational improvement are regulatory
compliance, audit fi ndings, rework, nonconformances, document revisions, disposition
timeliness, complaints received, inventory on hand, equipment failures, manufacturing
cycle times, employee turnover, and training opportunities. Additional
areas targeted for improvement may come from benchmarking key manufacturing
parameters against industry peers. Results of a gap analysis begin the dialogue
regarding process performance and the need for process improvements. Leadership
must be convinced there is opportunity for fi nancial and competitive gain, and the
resource investment to operate the QMS will be outweighed by the program bene-
fi ts received.
Management at the highest level in the organization must understand, support,
and lead the strategy to implement systems across the enterprise. More often than
not, this requires some level of business transformation, a cultural and behavioral
shift, and a certain level of risk. The risk associated in implementing change is
minimal compared to that of not having a robust system, as outlined in the benefi ts
section.
3.3.3.2 Speaking Management Language
Without upper management championing the establishment of systems, midlevel
management will not support the effort, dedicate the time required, nor practice the
behaviors essential to establish and maintain the processes. Leadership needs to be
cognizant of the benefi ts and consequences of nonimplementation and be clear and
unwavering in its support, delivering frequent consistent messaging to management
and staff. Leadership requires tangible and intangible benefi ts to be convinced that
the efforts are worthwhile and working and to regularly convey results to staff.
Tangible benefi ts should include metrics and improvements demonstrating
process and system cost savings, compliant inspections and customer audits, faster
product approvals and manufacturing throughput, less rejected material, reduced
nonconformance issues, and more effi cient continuous improvement and project
implementation. Intangible benefi ts include improved staff morale, faster, more
accurate transparent decision making, less employee turnover, increased staff
accountability, and an enhanced culture of quality throughout the organization. The
“ feeling ” conveyed by an organization that is reactive, stressed, and without well -
structured processes is much different than that of a proactive organization with
simple processes that are easily and successfully executed by trained staff.
Systems thinking allows decision making and process management to occur at
the process owner level, not the functional management level. This is a cultural shift
for many organizations but brings with it many benefi ts. Faster decision making, by
subject matter experts is valuable to organizations. It can benefi t both on a day - to -
day, lot - to - lot basis as well as provide long - term strategic direction to leadership.
Taking the burden off functional management and defi ning process owner responsibilities
allows functional management to manage resource and personnel issues
and not split time and attention between resources, personnel, technical, and process
issues.
3.3.3.3 Translating Benefi ts to Staff
Similar to leadership and management requirements regarding system understanding
and benefi ts, staff requires understanding prior to accepting the cultural changes
that a system - based approach will bring to the organization. Once the program is
initiated, tangible and intangible benefi ts must be realized and appreciated in order
for staff to continually support the program. Staff support, through benefi t realization
and management direction, will ensure program execution, ultimately delivering
the expected business results.
Transforming disparate processes into processes that are simple to understand,
easy to execute, and provide a sense of accomplishment meet one of management ’ s
obligations to staff. Staff interest lies in the ability to perform their work, contribute
to continuous improvement, and have a reasonable work – life balance. Finally, they
want to be able to contribute to their careers, have defi ned career paths, and have
attainable development goals for advancement. A well - designed quality management
system can contribute to provide all these employee benefi ts.
Staff benefi ts should be designed into the QMS. An outline of expected benefi ts
should be presented to staff to gain their support of the system initiative. Accomplishments
should be advertised and rewarded. Establishing well - defi ned processes
empowers employee involvement, participation, and contribution to the organization.
It reinforces a culture of quality throughout the organization, and provides a
conduit for their contribution.
3.3.3.4 Ensuring Staff Support and Management Leadership
Management ’ s responsibility includes providing staff robust tools and processes
necessary to accomplish their jobs effi ciently. Complex, missing, or fragmented
processes do not allow for easy operational execution, the ability to leave work at
reasonable times, and may result in poor - quality output or rework. This type of
environment quickly becomes dissatisfying to employees and results in poor morale,
low effi ciency, and ultimately lack of interest and loss of staff.
Staff empowerment allows pride in workmanship. Well - designed quality systems
make clear to staff where decision authority and process accountability lies, provide
clear expectations of the process and process owners, and provide personnel a clear
development path to process ownership.
Clearly identifi ed process attributes provide organizations more than tribal
knowledge to pass onto the next process owner. They provide clear structure, process,
and other attributes critical to the ongoing success of the enterprise. The organization
becomes reliant on their system and processes not people ’ s personal knowledge,
which can be lost with staff turnover.
Ensuring leadership and staff support requires that a well - defi ned plan be
designed and shared throughout the organization. A long - range plan, spanning
several years may benefi t the organization to maintain perspective and govern
expectations. An annual quality plan should encompass all aspects of the QMS and
contain detailed periodic goals and objectives. Progress against the quality plan
needs to be advertised and celebrated. Quality plan leadership should be recognized
for its efforts and accomplishments. Advertising wins and accomplishments in both
small group and large settings should be designed into the communication and
change management program.
MANAGEMENT AND STAFF: LEADERSHIP AND SUPPORT 255
256 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
Table 3 provides an outline of a long - term vision and goals for a quality management
system. A long - term strategy provides leadership, management, and staff with
an understanding of the program and anticipated timelines for implementation and
benefi t expectations. Annual quality plans become the short - term strategic milestone
vehicle to achieve the long - term strategic vision.
3.3.3.5 Traps to Avoid
Several challenges and requirements present themselves when establishing a formal
QMS. A primary requirement is a skilled team that understands the needs of the
organization, regulatory, and customer requirements. It should have the skill, experience,
and expertise to design a robust system and identify processes that support
the enterprise. A mismatch of team skills with enterprise needs may result in a
nonviable system that is not supported by leadership and staff, leading to failure
and disuse over time.
Quality management system design must be well thought out and tested. Pilot
programs are crucial to test system robustness and reliability, staff and management
acceptance, and the ability to produce the desired results. Time spent in system
design will pay dividends for years to come and increase staff support and critical
mass throughout the enterprise supporting the program efforts. Avoid implementing
any system or process design that has not been well thought out, does not have input
from the stakeholders using the system, or has not been piloted prior to a full - scale
implementation. Typically, a single opportunity exists to introduce a new program
before staff and management either accept it or reject the ideas and concepts.
Rebuilding interest and trust of a failed system is diffi cult. The probability for successful
reintroduction is minimized. Taking suffi cient precaution for correct implementation
the fi rst time is important.
TABLE 3 Long - Term Strategic Vision
Year 1 Year 2 Year 3 Year 4 Year 5
Gain
management
support
Create QMS
offi ce
Identify site
processes and
resources
Develop
communication
and change
management
plan
Implement
program
Train
management,
process
owners, QA,
and support
staff
Focus on
maturing
high - risk/
impact
processes
Reward and
recognize
QMS efforts
Indoctrinate
remaining
processes into
program
Document and
communicate
cost/resource
savings
Begin
integrating
processes
across the
organization
Focus on key
projects based
on QMS
portfolio and
management
review
Provide ongoing
training,
communications,
and change
management
Adapt to changing
business and
regulatory
environment
Provide leadership to
industry on QMS
paradigm
Change management is another very important consideration when implementing
a QMS because of the culture change required from the organization. Several
resources can assist in managing change, and these should be incorporated into the
system design. It is important to be cognizant that successful implementation
requires change at all three levels of the organization; leadership, functional management,
and operational staff. Each will need different messages, encouragement,
rewards, and benefi ts. Consideration to deliver both tangible and intangible benefi ts
to stakeholders is necessary.
Leadership support from the highest level is required. Middle management will
not support an effort that is not supported by its leadership. Leadership must
provide unwavering support, not provide mixed messages, continue to advertise and
celebrate success, and support the program through rough times. Consistency in
language and deeds from management supports understanding and appropriate risk
taking by management and staff.
Functional management must also support system efforts and long - term strategies,
to ensure that staff, who are critical to execution of the processes, know that
their support and efforts are expected. Functional management send powerful messages
to staff, and their support of the long - term plan and annual quality plan are
essential. Specifi c system objectives, included in leadership, management, and staff
goals reinforce the commitment and help ensure success of the program.
The system needs to remain fl exible. Having a long - term plan and vision is necessary
to provide a roadmap to the future. That roadmap may need to change as the
business environment and enterprise needs change. The long - term plan and vision
should be written at the level that it changes very little, but fl exibility is maintained
through the preparation of an annual quality plan that is capable of addressing
temporal issues and business needs.
Prior to implementing any QMS initiative, one must understand what leadership,
management, staff, and customers require. Knowing which processes are required
to support customer needs and the impact of those processes upon each other is
essential to system design. Developing process owners that understand their roles
and deliverables in the organization, eliminating constraints so they may meet their
goals is essential for success. Process owners must understand product and process
priorities so signifi cant benefi ts may be realized. These are important considerations
in designing system and processes that support the organization, produce meaningful
metrics, and demonstrate progress. Consideration of these important points
prevents system initiatives from failing and interpreted as another burden to the
already overburdened work and demands placed upon the organization.
3.3.4 ESTABLISHING QUALITY MANAGEMENT SYSTEM SCOPE
In many pharmaceutical and biopharmaceutical manufacturing operations, duplicity
exists in some processes and gaps are present between others. Often it is unclear
exactly what boundaries or scope constitute a process, the expected outputs, who
are the customers, who is the owner, and who is responsible for continuous improvement.
Duplicity is ineffi cient and costly. Examples include multiple layers of an
organization performing data reviews as documentation or information moves
through the value chain. Regulatory submissions for analytical validation are an
ESTABLISHING QUALITY MANAGEMENT SYSTEM SCOPE 257
258 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
example, where raw data may be checked at the laboratory, supervisor, quality assurance,
compliance, and regulatory group levels. On the other hand, gaps may exist
where each functional group listed above assumes data verifi cation is occurring with
another group, and in fact there are gaps in data integrity. In this case, the result can
be tremendously expensive if, upon regulatory inspection, errors are found and it
appears data integrity issues are ubiquitous in a submission.
This section will discuss the importance of defi ning business requirements to
ensure processes comprising the QMS are designed to support the enterprise, integrated
into a quality plan, suffi ciently defi ned to provide adequate resolution and
are transferable and scalable throughout the enterprise.
3.3.4.1 Defi ning Business Requirements
The QMS and the processes that comprise it must be custom designed for the needs
of the business. One size does not fi t all situations. The requirements of an enterprise
vary across sites and the phases of a product life cycle. A comprehensive system will
ensure a holistic programmatic approach in its support to the enterprise. This does
not mean that every phase of the product life cycle (discovery, development, commercial
manufacturing) will utilize all the processes that comprise the system. Nor
does it require that all commercial manufacturing sites will necessarily implement
all processes. It does, however, provide a common platform and expectation for all
processes, owners, metrics review programs, continuous improvement efforts, and
the like when they are implemented.
The fi rst step in designing a QMS is determining business needs and the processes
required to support the enterprise. Important consideration must be given to ensure
that all processes are included in the assessment. The assessment must include all
activities that affect product quality at corporate, business, manufacturing, distribution,
contractors, or joint venture sites. Processes controlling incoming materials
from vendors, laboratory services, contractual support, and other inputs should also
be included in the initial assessment.
Upon identifi cation of the processes required to support the enterprise, the next
step is to defi ne exactly what is in and out of scope for each process. Mapping all
the processes and their inter relationship with other processes will determine if any
gaps or duplication exists in the system. Duplication may be warranted or eliminated.
Gaps between processes require remediation. For example, a nonconformance
process should have direct linkage into a corrective action process. A
well - operating nonconformance process without an active, integrated corrective/
preventive action process will yield little benefi t to the organization and efforts
expended on the nonconformance process will be nominal in their overall positive
business impact.
This comprehensive approach allows for effi cient integration between processes,
different phases of product life cycle, and integration between different sites in the
supply chain. This integration provides opportunity for effi ciency in that process
owners are integrated with each other ’ s needs and expectations. Duplication of
effort is avoided and effi ciencies gained. Quality outputs from one process become
reliable inputs into the next process. Management and leadership will have access
and insight into compliance, infrastructure, and performance metrics of all processes
on a comparable basis. This provides leadership the opportunity for risk - based
resource allocation to appropriate areas of the enterprise.
Process mapping of the enterprise ’ s requirements to supply product enables
design of the processes required for the system. Staff, management, and leadership
input into the business needs provide additional guidance into processes
attributes.
3.3.4.2 Integrating Quality Management System into Quality Plans
A quality plan is required by the regulations governing medical devices (QSR) but
can readily be adopted as a useful tool for pharmaceutical and biopharmaceutical
manufacturing operations. A quality plan is the documented plan and goals for
enhancing and advancing the QMS. It can provide the outline and requirements of
the organization ’ s purpose, mission, product, and business practices used to produce
a quality product. A quality plan can detail the processes that comprise the QMS,
the maturity level required for each process, organizational structure, and other
requirements needed to meet the organization ’ s purpose. Included in the quality
plan are the elements of the business including location, size, products, and expectations.
It also includes its structure and support functions, values, and other attributes
of the organization.
An annual quality plan can be the detailed execution plan of the organization ’ s
long - term quality vision for the QMS. It provides staff and management the outline
and goals for improving the QMS. It enables employees to see the big picture, how
they fi t into the organization, and the organization ’ s expectations. Within the quality
plan attributes of the QMS should be described, including functional management
responsibility. This then becomes the foundation for further defi nition of processes,
description of management review and responsibility, and continuous improvement
programs. The preparation of a quality plan begins defi ning what is assumed to be
known by all levels of the organization. It is the mechanism for ensuring requirements
are addressed and gaps in the organization do not exist.
A quality plan may outline the organization ’ s long - term (several years) and
short - term (annual) goals through a risk - based approach to improving product
quality. It is the foundation for the manufacturing structure and support processes.
A quality plan ensures integration of personnel, their qualifi cations, product requirements,
quality management system, and regulatory and compliance infrastructure.
An example of an outline of a quality plan is in Table 4 . Leadership review and
approval of the quality plan is required to ensure that mission, scope, expectations,
and division of labor in the organization is consistent and supported.
In larger organizations, site or suborganization - based quality plans can be
designed to support the scaling of the QMS across all components of the enterprise.
The individual site plans provide focus on process challenges that are more critical
than at other sites due to variations in business and compliance environments. While
the specifi c plans emphasize goals based on site priorities, they also connect
the members of an organization to the mission of the greater QMS, as shown in
Figure 5 .
3.3.4.3 Determining Process Resolution Requirements
Leadership expects cost - effi cient reliable results from their manufacturing operations.
Management requires a capable workforce, equipment, facilities, and materials
to manufacture the product. Employees require robust processes that are easy to
ESTABLISHING QUALITY MANAGEMENT SYSTEM SCOPE 259
260 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
execute to perform their jobs. All this needs to be considered in the design of the
processes that comprise the quality management system.
Complex processes may need to be managed as distinct subprocesses in order to
provide process owners the ability to accomplish their work with specifi c focus and
expertise. Management and leadership may require data and metrics on specifi c
areas of the process that are not available if the process is too complex and large.
Dividing a complex process into simpler, more manageable processes also allows
for scalability and transferability throughout the organization.
Once processes have been defi ned for the enterprise, suffi cient system resolution
should be determined. This is accomplished by evaluating the ability of the process
owner to manage and execute the process requirements. Another factor in this
determination is the data and metrics needed from the process by management and
leadership. An example of a complex process that benefi ts the organization by being
managed through distinct subprocesses is validation.
Validation is a regulatory requirement and has become an industry standard for
ensuring product consistently meets quality attributes and regulatory requirements.
Validation requirements are woven throughout the manufacturing supply chain
encompassing many different subprocesses. The validation process may best be
TABLE 4 Elements of QMS Annual Plan
Element Defi nition
Introduction Purpose of plan and defi nitions for clarity
Plan Planned activities for the calendar year
Goals Specifi c/cascading goals of the site
Projects Major projects in support of the goals
Metrics Key metrics with defi ned targets
Approvals Site/plant management
FIGURE 5 Scaling the QMS through site quality plans.
Site 2
Quality
plan
Site 6
Quality
plan
Site 5
Quality
plan
Site 7
Quality
plan
Site 4
Quality
plan
Site 3
Quality
plan
Site 8
Quality
plan
Site 1
Quality
plan
Corp.
Quality
Plan
managed by dividing it into manageable subprocesses. This allows for effi cient management
and execution of the subprocesses, and the metrics reported for those
subprocesses are meaningful and specifi c. See Figure 6 , which illustrates one potential
organization of the validation subprocesses. Subprocesses contained within the
validation system could be cleaning, computers, automation, analytical, packaging,
process, transport validation, etc. Manufacturing is another example of a large,
complex process that may best be subdivided to support better management and
more meaningful metrics to management.
By dividing a larger process into manageable and specifi c subprocesses, management
can assign appropriate subject matter expertise to lead and manage each
subprocess. The metrics measuring subprocess performance can be uniquely
reviewed, evaluated, and compared to similar subprocess metrics at other sites or
companies. Valuable, meaningful comparisons can be obtained for process and subprocess
performance that would otherwise be blinded or diluted, if they were summarized
within the higher level process metrics.
An additional advantage of establishing subprocesses is that it affords the opportunity
for rapid assimilation and transfer of the subprocess at various sites within
the enterprise. An example of this is the comparison of a bulk manufacturing facility
with that of a distribution center. Both will need to implement aspects of the subprocess
“ transport validation, ” however, the distribution center will not need to
implement other subprocesses such as process or packaging validation. As the subprocess
transport validation is designed and implemented at one site, that infrastructure
and knowledge transfer to the other site is rapid, avoiding duplication of efforts.
Sharing of information and expectations of the two sites becomes a common goal
and format. Management can, therefore, compare transport validation needs and
maturity levels between sites equally.
Once all the processes and subprocesses supporting an operation have been
defi ned, another gap analysis may be conducted to ensure that there are no assumptions,
and all required processes and subprocesses required to support the business
are included in the scope of the QMS. This can easily be accomplished by listing all
the business drivers for an operation and comparing that against the processes
FIGURE 6 Validation subprocesses.
Computer
validation
Transport
validation
Packaging
validation
Analytical
validation
Automation
validation
Process
validation
Cleaning
validation
ESTABLISHING QUALITY MANAGEMENT SYSTEM SCOPE 261
262 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
established. Written defi nition of the process and subprocess scope is required.
Stakeholders, owners, and users of the processes should be involved to ensure clear
defi nition and understanding of process scope. Questions to be asked are: Are all
the business needs addressed? Have all our activities and operations been included
in the assessment? Is there any duplication in process expectations? Are there any
gaps between process outputs and inputs? Once this evaluation has been concluded,
it can be easily determined if any existing work process have been overlooked and
the system requires further modifi cation.
3.3.4.4 Scalability to Enterprise
Well - designed processes and subprocesses are scalable to the enterprise. A comprehensive
design will allow for replication and comparison of processes and subprocesses
between multiple sites. This allows for rapid implementation of new technology,
sharing of best practices, and comparison of similar metrics to determine compliance,
infrastructure, and performance. A comprehensive system allows for each unit
operation or site within the enterprise to have the fl exibility to apply applicable
processes and subprocesses, yet continue operating within the defi ned structure of
the QMS. For example, a manufacturing site may utilize almost all of the processes
discussed in the validation process, whereas a distribution site may only utilize the
transport process. Both sites, however, implement the same structure for the transport
process, allowing for meaningful comparison of data and metrics and rapid
implementation of any required changes to that process.
Well - designed quality management systems support structured organic growth
and are valuable in evaluating and integrating manufacturing acquisition opportunities.
Business and manufacturing management should utilize the QMS and its standards
whenever evaluating external facilities for appraisal, approval, integration, or
expansion. Meaningful metrics obtained from a QMS provides the standard to
make critical decisions affecting multiple internal or external manufacturing
capabilities.
Documented process structure provides rapid employee assimilations when
transferring employees between sites. New employees, replacing existing process
owners, are enabled to rapidly execute process responsibilities due to the abbreviated
learning curve when processes have been well defi ned and documented. Systems
designed as described here provide meaningful and comparable metrics for leadership
to evaluate progress, compliance, and performance.
3.3.5 SYSTEM AND PROCESS OWNERSHIP:
ROLES AND RESPONSIBILITIES
A well - designed QMS and the processes that comprise it require competent ownership
with defi ned roles and responsibilities for program success. This combination
ensures that the system and processes are established, maintained, improved, and
remain current with industry practices and business expectations. Operational
execution of the QMS and the processes comprising it will engage stakeholders,
management, and leadership, provide business results, and support and ensure
compliance.
3.3.5.1 Quality Management System Ownership and Management
The QMS is best owned at the highest level in the organization. At a minimum it
should be owned at a level in the organization above manufacturing and quality.
The owners ’ main responsibility is to champion the program and ensure organizational
alignment. Regulatory investigators expect processes supporting manufacture
are fully incorporated into the QMS. They also expect leadership to have signifi cant
knowledge of the operations and interact with investigators during inspections with
some degree of familiarity with the processes supporting manufacture. At the conclusion
of an inspection, regulators issue inspectional fi ndings and, if appropriate,
take regulatory action against the most senior member of the leadership group.
Through high - level leadership ’ s active involvement and ownership, the QMS
program and enterprise will be successful.
As mentioned previously, the QMS is best managed by a group dedicated to the
program. The QMS program offi ce should have defi ned roles and responsibilities.
In the FDA regulation 21 CFR Part 211.22, the responsibility of the quality unit is
described. It is the only functional group in a manufacturing organization that has
its job description codifi ed in federal regulations. These responsibilities should not
enable or dilute the responsibility for ensuring quality of other functional groups in
the organization. All functional groups supporting manufacture should be applying
their trade to the GMP world. Regulators expect the Quality Department to have
oversight and approval of all processes affecting product quality. Program management
is important because of the need for coordination and accountability to bring
individual processes, long - term system strategy, yearly quality plans, and goals
together to accomplish the program ’ s objectives. These activities and benefi ts cannot
be realized from individual process owners.
The program management of the QMS can be managed just as a process, with
predefi ned expectations, metric collection, and management review, culminating in
risk management application to continuous improvement programs. These metrics
and improvement initiatives need to be vetted through leadership review and input
to ensure alignment throughout the organization.
An outline for the roles and responsibilities for the QMS program offi ce is illustrated
in Table 5 . By establishing the roles and responsibilities of the program offi ce
a defi ned point of contact and accountability is established for program execution.
It establishes strong linkage and focus on the program objectives for process owners,
training, functional management, and leadership. Similar program management
structures are required at manufacturing sites and corporate functions to maximize
benefi ts of the program through establishing common and specifi c goals and
TABLE 5 QMS Program Offi ce Roles and Responsibilities
Subject matter expert for QMS program
Develop and execute communication plan
Initial and ongoing training
Facilitate management review process
Identify process maturity goals and metrics
Develop long - term strategic vision
Create and execute annual action plan
SYSTEM AND PROCESS OWNERSHIP: ROLES AND RESPONSIBILITIES 263
264 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
providing a platform for sharing best practices and knowledge. As organizations
grow in complexity, additional management may be required to ensure that the
elements of the QMS are integrated, functioning, and delivering the results
expected.
3.3.5.2 Process Ownership
Designing a QMS that mandates and assigns process ownership to designated individuals
is a signifi cant strategic decision in the establishment of a successful quality
management system. It provides effi ciency, expertise, dedication to the process, and
focused ownership for documentation, improvements, benchmarking, and compliance.
Without defi ned and assigned process ownership functional management
becomes the de facto process owner. This is problematic in that functional management
is already overburdened with personnel and business management issues,
unable to adequately focus and deliver the demanding process owner requirements
required in today ’ s manufacturing environment. Several processes typically are
organized under an individual functional manager, further diluting focus, attention,
and expertise if functional management is relied upon as a process owner.
3.3.5.3 Process Owner Selection
Process owner selection requires program management to establish defi ned criteria
for the selection process. Criteria include the capability to perform process owner
roles and responsibilities, including self - development and decision making. Empowered
process owners are accountable for maintaining and executing the processes
that management relies upon to deliver business results. This accountability ensures
that staff, management, and leadership know who to solicit for answers to process -
related questions and issues. It also provides the best representation to regulators,
clients, and customers. Effi ciencies are gained and current trends maintained with
an active owner, with defi ned responsibilities.
Selection criteria may include attributes of technical, interpersonal, and management
skills. The capabilities needed for different process will vary and should be
considered in the selection process. At the end of the selection process, functional
management and the process owner may consider inclusion of the process owner ’ s
roles and responsibilities into the process owner ’ s job description. Personal goals
and development activities should be based on improving the process owner ’ s capabilities
to manage the process and develop future process owners through active
mentoring and talent development programs.
Process owners need to be dedicated to their process. They must be empowered
and held accountable for all the attributes listed in their roles and responsibilities.
Process owners may have ownership of more than one process and may have other
job responsibilities, but it must be clear throughout the organization as to who has
full authority for the process.
Process owners require a defi ned set of responsibilities to maintain a vibrant and
effective process that continues to support product quality deliverables. Having
roles and responsibilities defi ned provides owners with the structure and parameters
needed to be effective. Examples of owner responsibilities include identifi cation of
stakeholders, defi ned decision authority, document ownership, nonconformance
ownership, knowledge of regulations and industry trends, subject matter expertise,
training content, metric ownership, and representation to internal auditors and
external regulators. Identifying, training, and development of the process owner on
his or her roles and responsibilities is similar to assembling the piece of a puzzle. If
one piece is missing, the effectiveness of the process owner will be minimized (see
Figure 7 ). Developing a sound methodology for process owner selection ensures
objectivity and is critical to the success of the program. A brief discussion on each
process owner roles and responsibilities follows.
3.3.5.4 Stakeholder/Process Owner Integration
Process owners must identify the stakeholders of their process and ensure design
and output of the process meets stakeholder needs. Regular communication, interaction,
and support are maintained with stakeholders through scheduled meetings
to discuss process status and improvements. Any changes to the process are vetted
through the stakeholder group. Typical stakeholders include the QA unit that is
responsible for review and approval of the process components, suppliers to and
receivers of the process (i.e., owners of other processes that interact with the
process), customers, management, and leadership. Each stakeholder roles and
responsibilities also require defi nition.
Including the key stakeholders in decisions affecting process design or changes
ensures efforts by the process owner are applied correctly. Process robustness is
dependent upon meeting business and customer needs, and the process owners
require input and support from the stakeholder group.
FIGURE 7 Process owner roles and responsibilities.
Accountable
ownership
Risk
management
Nonconformance
& corrective
action
ownership
Inter/Intra
process
expertise
Metrics
& process
improvements
Inspection
& audit
point of contact
Stakeholder
management
Document
& training
content
Decision
authority
SYSTEM AND PROCESS OWNERSHIP: ROLES AND RESPONSIBILITIES 265
266 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
3.3.5.5 Decision Authority
Each process owner requires a defi ned level of decision authority. This authority
level delineates the bounds of decision making granted by the organization to the
process owner. Business needs and risk assessment must be incorporated into the
design of the decision authority granted to a process owner. Table 6 is an example
of a decision authority matrix design for a process owner. It requires cross - functional
management support to be effective.
Preparing a decision matrix that is shared and agreed upon by the stakeholder
group and functional management ensures decisions are made and communicated
quickly by appropriate persons. It removes the burden of making every technical
process decision from functional management. It is important to outline the process
owner ’ s role in the decision - making process, as well as conditions for escalation.
Effective process management is realized when the culture of an organization can
support the outline of the decision matrix and not continually rely upon functional
TABLE 6 Decision Authority Matrix
Decision
Category Defi nition
Decision
Maker
Decision
Support
Required
Informed of
Decision
Company
standards
Process - related
global
standards
relevant to all
manufacturing
sites
Corporate
process
owner
Site process
owners
Site quality
assurance
counterpart
Process
stakeholders
Impacted staff
Standard
operating
procedures
SOPs related to
the specifi c
process
Site process
owner
Process
stakeholders
Site quality
assurance
counterpart
Management
review
Corporate
quality
assurance
counterpart
Impacted staff
Training Training on
processes or
procedures
Corporate and
site process
owners
Training
Technical
system
matter expert
Process
stakeholders
Corporate
quality
assurance
counterpart
Impacted staff
Site projects All projects
related to the
existing
process or the
projected
improved
state of the
process at a
specifi c site
Site head Process owner
Leadership
team
Site project
portfolio
manager
Process
stakeholders
Corporate
quality
assurance
counterpart
QMS offi ce
management. If functional management continues to be relied upon and seen as the
process decision makers, efforts and progress by the process owner will be nominal.
The organization ’ s culture must support the process owner at all levels of the enterprise
for the owners to be successful.
3.3.5.6 Industry Knowledge
The c in c GMP represents the notion of current industry practices. Process owners
must work to remain current with industry and regulatory trends affecting their
process and its overall effect on the QMS and the business. Awareness of process
capability and comparability with other like processes within and outside the pharmaceutical
and biopharmaceutical industry is essential. Regulators will compare an
owner ’ s process against other similar processes in which they have experience when
formulating value judgments. Benchmarking against similar processes provides
process owners the data needed to determine adequacy of their process with industry
peer groups.
Where technology, effi ciency, performance, or compliance can be enhanced, it
should be considered by an aware and informed process owner. Functional management
cannot keep pace with the changes occurring with all the processes supporting
manufacturing. Ensuring process owners dedicate suffi cient time to keeping current
with process - related external events will ensure process success. This may include
review of industry periodicals, attendance at seminars and regulatory presentations,
and routine self - evaluation and benchmarking against relative processes.
Often, the best examples of process effi ciency can be found outside the pharmaceutical
and biopharmaceutical industries. Other fi elds such as electronics, space,
and software industries have evolved their documentation, training, quality, and
change control systems to the point of best in class. These industries are more time
sensitive to get product to market and have often evolved their processes to be
effi cient and decision processes to be very quick. Process owners may expand their
knowledge by investigating other industries to fi nd best practices and apply them
internally.
3.3.5.7 Regulatory Inspection and Audit Lead
Process owners play a critical role during regulatory inspections and customer
audits. Process owners are the best choice to represent process attributes and performance
to interested parties. Process owners provide regulators and auditors with
a capable, knowledgeable resource to represent the process and answer detailed
questions. The process owner should be aware of process history, requirements,
operations, exceptions, changes, and nonconformances. The process owner will have
detailed knowledge of process operations, compliance, and be able to defend the
process. Providing an accurate answer the fi rst time to regulators and auditors is
essential in building trust and representing competence.
Each process owner is required to work closely with his or her QA counterpart.
This ensures design and operational issues are clearly reviewed and approved by a
representative from the quality assurance function, a regulatory expectation. The
quality assurance counterpart must be familiar with the process, understand documentation
supporting the process, and able to convey what approval the Quality
SYSTEM AND PROCESS OWNERSHIP: ROLES AND RESPONSIBILITIES 267
268 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
Department has conveyed on the process and meaning of that approval. The QA
counterpart to a process should also have defi ned and documented roles and
responsibilities.
When teamed together, the process owner and the quality assurance counterpart
for a process will make a favorable impression upon regulators, be able to
explain all operations involving the process, supporting documentation, and any
ongoing projects or process improvements. This pair is the best to evaluate and
consider any deviations to the process or recommendations for continuous
improvement.
In most all cases, the process owner and QA contact will posses more information
about the process than regulators and be capable to defend the process design and
operation. Should regulators have suggestions on process design or functionality,
the process owner may consider them. If appropriate, any recommendations or
observations made by regulators or auditors can be incorporated into the process
design. However, it is critical for the process owner and stakeholders to evaluate
proposed changes to avoid reactive management commitments, which could be
deleterious to the effi cient operation and output of the process.
3.3.5.8 Subject Matter Experts
Process owners, through their selection and development, become the subject matter
experts for the process. It is more effi cient for an enterprise to focus its expertise
on individuals that have the authority and accountability described in this section,
rather than dilute those attributes and accountabilities, thereby risking poor process
execution and management.
As a subject matter expert on the process, the owner has the capacity to deliver
results outlined in the list of owner responsibilities, mentor future process owners,
assist in staff development, and accurately guide management in its strategy related
to the process. The process owner ’ s personal development of his or her process
expertise is essential in delivering operational results and providing direction for
future strategic changes to the process.
3.3.5.9 Metric Ownership
Process owners ’ responsibilities include determining appropriate metrics for their
process. These metrics should include lagging and leading metrics that are meaningful
to the process owner and management in determining performance, compliance,
and infrastructure of the process.
The process owner should represent and interpret these metrics to the organizations
leadership. The metric output from a process is the basis for management
and leadership ’ s action in resource deployment and approval of continuous improvement
projects. Key operating parameters such as number of nonconformances and
regulatory observations against the process should be tracked and factored into the
maturity of the process.
Every process owner needs to base their continuous improvement plan for the
process based upon metrics collected from the process output. The metrics must be
designed to assist in these decisions and be readily available for review, presentation,
and interpretation.
3.3.5.10 Documentation Ownership
Process owners are the most appropriate owners for all documentation supporting
their processes. This includes having either direct ownership or controlling infl uence
over guidance and execution documentation such as corporate policies and standards,
local requirements and standard operating procedures (SOPs), logs, and
records.
For the manufacturing process owner, this means owning the master manufacturing
records and executed batch records, SOPs, use logs, and related training documents
for their process. Combining responsibilities for process management and
process ownership results in true accountability for the process owner. It also allows
for progress and continuous improvement of the QMS. Removing questions
of responsibility and accountability ensures integration between requirements
(standards, policies, procedures) and execution (training, performance, and
documentation).
3.3.5.11 Training
Assurance of adequate training for process users is an important responsibility of
a process owner. Process owners must have a clear understanding of the requirements
of their process and its operation. This understanding requires translation
into executable training. Users must be able to understand and apply the training.
Complicated processes coupled with ambiguous training will lead to confusion and
an inability to properly execute a process, which eventually constitutes failure for
the organization. A simple process, with easy to understand process steps, that are
consistent with instructions and documentation requirements will support success,
reduce production costs, minimize nonconforming events, and allow for employee
satisfaction.
Process owners are subject matter experts and should infl uence and provide
consulting for training on the process. They may also participate in training delivery.
Ensuring adequate training on a process is a key goal for system effi ciency and
regulatory compliance. Process owners, capable of explaining the reasoning behind
the process requirements, enhance the training experience for process users. Process
owners should include effective presentation and training skill development into
their personal development programs.
3.3.5.12 Risk Management
Process owners require basic understanding of risk management and its application
to process design and continuous improvement prioritization. Several industry and
regulatory resources exist, such as ICH Q9, that provide understanding on risk
assessment, identifi cation, control, methodology, and the overall risk management
process. Process owners should be familiar with risk management techniques and
tools and apply them to their process management when designing, executing, or
managing improvement efforts for their process.
Risk management is especially important for the presentation of process improvement
proposals to management where resources are required. The ability to quantify
risk and demonstrate continuous improvement benefi ts is essential to project
SYSTEM AND PROCESS OWNERSHIP: ROLES AND RESPONSIBILITIES 269
270 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
and resource approval. Risk analysis, management, and presentation constitute
guiding leadership to work on the right things at the right time and then improve
it.
3.3.5.13 Continuous Improvement and Project Management
Instituting quality - by - design efforts early in the design of a process should negate
the need for major process improvements. However, over time, due to business
needs, regulatory changes, or technology improvements, processes will require some
form of change to ensure compliance or performance enhancement. As part of
executing and maintaining their processes, owners need to collect and report performance
metrics to management and staff. These metrics will inevitably direct
attention to opportunities for improvement that require capital and human resources.
Process owners are the best leaders of continuous improvement projects due to their
intimate knowledge of the process and accountability for process output.
Managing or leading a continuous improvement project requires process owners
be knowledgeable in project management and team - leading skills. Improvement
projects typically require cross - functional support and expertise from areas such as
information systems, project management, manufacturing, engineering, and development.
It is essential that continuous improvement efforts name the process owner
as the project lead to ensure the required output of the project meets the process
owner, stakeholders, and enterprise needs. Often, projects are completed and
declared a success, delivering a substandard result that the process owner, users, and
stakeholders fi nd inadequate to meet process requirements.
Upon completion of continuous improvement projects, process owners ’ responsibilities
include monitoring the changes made to the process to determine the
impact of improvements. Metrics monitoring changes to the process, pre - and postimplementation,
should be incorporated into existing performance metrics and
reported during regular management reviews.
3.3.5.14 Non Conformance/ CAPA /Planned Deviation Ownership
An important barometer of process performance is the number of nonconformances,
corrective and preventive actions taken, and planned deviations initiated
against a process. These types of process artifacts must be known and owned by the
process owner and stakeholders. The process owner must consider these process
metrics for evaluation of and changes to process design, training, documentation,
and performance.
Nonconformances may fall into the category of manpower, machinery methods,
materials, etc. Employees not following procedures or unable to execute required
steps of the process indicate a poorly designed process requiring modifi cation and/or
improved training. Machinery failures often indicate poor qualifi cation, validation,
calibration, or maintenance programs. Unexpected results or outcomes are
indicative of poor process design, characterization, or a break down between
processes.
Although planned deviations are frowned upon by many in the industry and
regulators, there are times when temporary changes to a process must be employed
to support the business. Permanent changes must be made through a formal change
control process. When the use of a planned deviation is required, the affected
process owner should be aware of and own the change. This provides owner control
over the duration and extent of the change to the process and provides data for
possible consideration in making a permanent change to the process. A planned
deviation should be rare and monitored closely as it affects previously established
standards, expectations, and training.
Process owners must be capable to evaluate and interpret the effect of nonconformances
and planned deviations on their systems. Process owners can evaluate
the need and lead efforts for corrective or preventive action, ensuring adequate
corrections and improvements are implemented. An effective QMS ensures deviations
from approved processes are owned and adequately investigated by the process
owner ’ s and ultimately approved by their quality assurance counterpart. The knowledge
of these events is the basis and foundation for the process owners to make a
risk - based evaluation on whether or not process changes are required, documentation
or training require modifi cation, or continuous improvement efforts are
warranted.
A well - designed QMS will include identifi ed process owners with defi ned roles
and responsibilities. Process owners require support from management, their customers,
stakeholders, and quality assurance. Accountability and decision - making
parameters will empower process owners to drive execution and improvements to
their process, delivering the business results expected. Without these process owner
attributes and support, minimal results will be achieved, and functional management
will be burdened with and assume the responsibility for making decisions that
should be in the hands of capable process owners.
3.3.6 CHANGE MANAGEMENT/COMMUNICATION
Establishing and maintaining an effective QMS, as this chapter describes,
requires a signifi cant cultural shift. Many employees and functional management
will fi nd the business transformation of defi ning processes, assigning ownership,
delegating authority, and responsibility for process performance within the QMS
is a signifi cant change in business conduct. The most signifi cant change results
from the shift of control in process expertise and decision - making authority
from functional management to process owner. Signifi cant business transformation
may result by assigning responsibility and accountability to the process owner,
and management ’ s support of process owners who drive continuous process
improvement.
In The Second American Revolution , Rockefeller describes the conservatism of
organizations: “ An organization is a system, with a logic of its own, and all the weight
of tradition and inertia. The deck is stacked in favor of the tried and proven way
of doing things and against the taking of risks and striking out in new directions ”
[9 ]. If an organization is not already practicing principles of delegation, process
ownership, established metric collection, management review, and continuous
improvement, barriers within the organization will need to be addressed and broken
down in order to establish new behaviors. These barriers to change will exist within
and between functions, functional management and staff, and possibly between
companies and regulators.
CHANGE MANAGEMENT/COMMUNICATION 271
272 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
Although expected benefi ts are signifi cant when implementing a QMS and the
end result desirable for employees and management, describing the desired
state and motivating personnel to change and implement new behaviors contains
signifi cant challenges. A successful business transformation requires a robust
change management and communication plan that includes support for all staff
affected.
3.3.6.1 Managing Organizational Change
Integration of the skill sets of human resources, training, and change management
groups will signifi cantly augment efforts toward cultural change and acceptance.
Often personality profi ling tools are effective to gauge the organization ’ s preferences,
learning styles, and adoption tendencies. These types of tools should be considered
in the overall change management program, used where applicable, and
program modifi cations made based upon their results.
The fi rst and critical step in developing a successful change management plan is
to obtain initial support from the corporation ’ s leadership and functional management.
Without this support the QMS will not gain critical mass and may not deliver
the desired effects or changes. To acquire this support, the implementation team
must put together a strong business case that speaks to the leadership ’ s needs and
wants. The business case must include a risk assessment against compliance and the
benefi ts of the fi nancial gains. It is important to be honest, consider current system
status and future requirements, and include a long - term strategy that addresses costs
and benefi ts. The change management plan must include frequent and repetitive
communications, to all levels of the organization, of the cost/benefi ts and successes
expected and realized by the program.
Functional management support is also critical to the success of a new program
such as a quality management system. Any time a staff member is asked to take on
a new role or responsibility, he or she needs to be supported by the functional
manager as well as leadership within the organization. Corporations are resource
limited and necessarily need to continually prioritize where to allocate resources.
Staff will only take on roles or responsibilities that they believe are supported by
their functional manager in an effort to successfully meet their perceived immediate
goals. Quantifi able support from leadership and functional management can be
directly correlated to the success or failure of the QMS program.
Signifi cant work is involved in training new process owners, functional managers,
leadership, support organizations, and actualizing their new behaviors. A support
system must be in place for the process owners, stakeholders, and management to
guide and reinforce the new behaviors and maintain the process effectiveness. It is
preferable that this support system be established through a dedicated team
that can be fully attentive to all their needs. Without a single source to lead the
efforts, diversity in interpretation and implementation will dilute the program,
within different functions and sites, and its effectiveness and outcomes will be
diminished.
Establishing an organization to lead the systems initiative is important. That
organization requires management, standards, and parameters similar to managing
an individual quality process. It requires roles and responsibilities be established,
metrics be determined, collected, reviewed and acted upon, and receive management
and leadership visibility and support. The QMS program is best organized as
a function within the Quality Department and be regarded as an ongoing program,
not a short - term project or effort with limited shelf life. The group must be led by
competent persons who are familiar with quality concepts and applications, regulatory
expectations and requirements, needs of the enterprise, good communication
and infl uencing skills, and are fl exible and enduring.
3.3.6.2 Communication
Trying to get people to comprehend a vision of an alternative future is also a communications
challenge of a completely different magnitude from organizing them
to fulfi ll a short - term plan. It is much like the difference between a football quarterback
attempting to describe to his team the next two or three plays versus his
trying to explain to them a totally new approach to the game to be used in the
second half of the season. Aligning the organization to accept and implement a
system - based approach requires careful messaging coupled with management
support and results.
Messages are not necessarily accepted just because they are understood. Another
big challenge for leadership is credibility and getting people to believe the message.
Aligning words and deeds supports the worthiness and credibility of the messaging.
People have learned from experience that even if they correctly perceive important
external changes and then initiate appropriate actions, they are vulnerable to
someone higher up who does not like what they have done. Reprimands can take
many forms: “ That ’ s against policy, ” or “ We can ’ t afford it, ” or “ Shut up and do as
you ’ re told ” [10] .
Having established a dedicated team that provides overall program management,
it is imperative that the team outline a strategic plan for presentation to leadership.
Without a vision and long - term plan, which is supported by the enterprise leadership,
quality system initiatives will become diffi cult. The plan needs to be comprehensive
in nature, yet broad enough to convey purpose, mission, and benefi ts at a
high enough level to be understood and supported. An outline such as this provides
framework and direction for the program management team and leadership. It
also guides the program management team to developing annual goals and quality
plans that fi t into the overall strategy and provide momentum and results to the
organization.
Annual quality plans should be prepared by the QMS program offi ce that
address the long - term strategy and intermediate goals that come to surface during
program implementation. Training, changes in regulatory requirements, metric -
driven projects, and special circumstances warranting process changes such as implementation
of new technology or programs should be included into the annual
quality plan.
Long - term strategy documents and annual quality plans require leadership and
functional management support and approval. These documents must be reviewed
and discussed with the leadership of the organization, modifi ed to meet the business
and regulatory requirements, and then have full support through upper management
approval. In this way, the goals are being led by top leadership and management
and not any individual group in the organization. Once top leadership signs
CHANGE MANAGEMENT/COMMUNICATION 273
274 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
onto the program, it can be shared throughout the organization in a number of
ways.
If leadership can support the long - term strategy and annual quality plans to
accomplish the vision, then the foundation for change management and cultural
shift is in place. Leadership will need to continually discuss the need for systems
implementation, in front of a variety of audiences. This includes leadership staff
meetings, management, and employee meetings. The importance of leadership
support cannot be overlooked. Without consistent visible leadership and management
support process owners and staff will revert to old behaviors, become reactive,
and perhaps unrelated in their process integration efforts. Leadership needs to
require aspects of the program be included into functional management annual
objectives with defi ned deliverables outlined and evaluated. Likewise, functional
management should require staff to include appropriate aspects and objectives of
the QMS program into their individual goals and work to accomplish them.
3.3.6.3 Feedback and Alignment
Managing the changes required to fully implement a QMS can include several forms
of communication and feedback. A detailed annual communications plan can aid
the QMS ’ s group in identifying specifi c target groups, methods, and frequencies of
communications, messaging types, and feedback mechanisms to monitor progress
for program modifi cations. Table 7 is an example of an annual communications plan
that supports efforts to keep internal audiences informed, aligned, and engaged.
Each target audience requires specifi c messaging that connects with its needs.
Failure to get the appropriate message, that is, what is the program bringing to them,
will minimize support for the program. This plan should include face - to face and
written communications addressing multiple audiences and media types. Face - to -
face meetings can include presentations to steering committees, process owners,
functional departments, and all staff meetings. Written communications can include
sitewide communications, poster sessions, and newsletters. The communications
should speak to all audiences — “ what ’ s in it for me? ” Topics can include leadership ’ s
commitment (direct quotes or actions taken); spotlight on successes (real - life stories
from process owners); impact to the site (process improvements or risk mitigation);
and progress to the program (metrics and successes). The progress and success of
the QMS cannot be overcommunicated.
Another useful tool to help the message and modify the program is the use
of a feedback survey. If properly designed and distributed to a defi ned set of stake-
TABLE 7 Communication Plan
Vehicle Communication Type Frequency Date
Functional metrics meeting Face to face Monthly First week of month
Management interviews Face to face Annually January 1 – 31
QMS newsletter Written Quarterly First week of
quarter
All staff meeting Presentation Semiannually March & September
Poster session Written/face to face Annually July
holders and employees, the survey can provide valuable insight into how staff and
management view the program, its progress, and suggestions for modifi cations.
If surveys are distributed electronically and offer only one - way communication,
the benefi ts may be limited as the respondents are limited in their ability to fully
convey their impressions or offer effective feedback. An electronic feedback survey
may be a fi rst good step in understanding the thoughts and concerns of the
stakeholders.
Another suggestion or follow - up to the electronic survey is to utilize focus groups
that have the ability to interact with the program questioners. This two - way conversation,
verbal dialogue, allows further understanding of the program by the participants
that follows with more meaningful feedback to the program administrators.
Focus groups should be selected at different levels within the organization, including
process owners, stakeholders and users of the system, leadership, functional management,
and the general populace of employees. Focus groups provide valuable
input into programs that the program administrators may be unaware of and can
provide program redirection.
Once suggestions are received on the program, it is essential to consider and
incorporate those ideas and modifi cations that make sense to implement. Those
changes need to be communicated and seen by the focus group members to ensure
that their time and effort has not been wasted and their suggestions have been
heard. This is one of the best ways to spread the word about the QMS program and
garner grassroots support.
3.3.6.4 Training
A training plan should be developed to identify the needs of the staff and affected
functional areas required to support the successful implementation of a QMS. It is
the responsibility of the corporation to adequately support staff with training and
tools when staff is expected to take on new roles, responsibilities, or behaviors. The
training plan should consist of targeted training for general staff, process owners,
and functional management of the process owners.
At a minimum, all staff should be introduced to the purpose, goals, and requirements
of the QMS. This training should be a high level explanation of the program
looking to gain understanding and support for the program by communicating why
it is important and what are the risks of not adopting the program. This can be
accomplished by instructor - led training or an electronic, Web - based learning module
depending on the size of the corporation.
Process owners require more comprehensive levels of training to fully understand
their role and responsibilities within the program. Process owner training
should teach key concepts and tools that owners will need to evaluate and support
their processes. This training can be done in a phased approach to support the elevation
and advancement of a process within the organization ’ s chosen maturity
model.
Training should be provided to offer functional managers supervising process
owners a thorough understanding of the QMS. Training should address new roles
and responsibilities of staff, time demands on process owners, overall program
timelines, and impact to functional areas. The acceptance and support of functional
management is critical to successful implementation of a QMS.
CHANGE MANAGEMENT/COMMUNICATION 275
276 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
Managing organizational change demands a well - written strategy, skill set, and
resources to ensure changes that come with system implementation are understood,
supported, and maintained. Starting with high - level overviews of system design,
benefi ts and timelines for implementation are the foundation for management
understanding and support. Detailed annual quality plans can be the tactical vehicle
for program implementation. Leadership support through understanding and
approval of the annual quality plan, inclusion of program objectives into management
goals, and frequent verbal and visual support of the program are essential to
success. Building the program infrastructure is a signifi cant undertaking. Inclusion
of a comprehensive training, communication, and change management plan should
be built into the overall goals of the program and routinely evaluated and
delivered.
3.3.7 MEASURING SUCCESS THROUGH MEANINGFUL METRICS
Successful implementation of a comprehensive QMS can be determined by the
establishment of a meaningful metrics program. The purpose of a metrics program
is twofold: fi rst, to allow an organization to evaluate its progress toward meeting its
goals in an objective, data - driven manner and, second, to monitor the performance
of each process to ensure continuous improvement. By evaluating metrics for the
QMS and its processes, the enterprise has the knowledge and understanding of the
overall health of its system and processes and can develop strategies based on risk
for continuous improvement of the system and processes.
Once the metrics program is in place, the system and process metrics require
visibility to process owners, upper management, and stakeholders. Process owners
require understanding of the metrics ’ trends, issues, and associated risks. Stakeholders
must work with the process owner to identify and propose process improvement
opportunities. Leadership is accountable to understand the issues and associated
risks and responsibly apply resources for remediation efforts.
3.3.7.1 Performance Metric Development
Quality and business indicating metrics should also be reviewed on a routine basis.
These may include the following:
• Quality indicating: ability to meet quality standards and procedures
• Supply: ability to meet demand
• Cost: savings as well as avoidance
• Safety: near misses and incidents against process
The guiding principle of metric development is to have a stable system or process
to collect, review, and draw conclusions. All metrics should be developed with stakeholders
input taking into account the requirements and needs of the customers. This
includes the touch points of the downstream quality processes. Without this input
and understanding metrics may be developed within a silo and hold little value,
causing both frustration at the leadership as well as the staff level. Without proper
design, metrics may become a check box activity that results in minimal or no action
by management to support efforts by a process owner.
Metrics can either fall into one of two categories: lagging or leading indicators.
Both types are important to the process owner and management. Lagging indicators
are metrics that represent the process ’ s ability to deliver results or outputs. They
indicate the performance of the system in the past. They can assist process owners,
management, and leadership in determining if goals have been met, objectives
attained, or existing standards or expectations have been met. Leading metrics focus
on the inputs and suppliers of a process. These metrics are important indicators to
proactively allow owners and management to take action on a process prior to violating
a standard, objective, or goal. A successful process owner will understand the
relationship of leading metrics and their affect on the lagging metrics and process.
Metrics need to be designed to meet the needs of the organization, be simple to
track and present, and be regularly reviewed.
3.3.7.2 Metric Review
Ignorance of system and process performance leads to ineffi ciency, poor compliance,
and low employee morale. It is good business practice to have regular review of
process metrics to gauge the health and output of the system and processes that
drive the organization.
Process owners should be aware of all the metrics affecting their process and
have a conduit to present the critical metrics to upper management. There are
examples in the industry where process owners responsible for execution of a
process are not aware of the metrics being collected, if any are, and have no basis
for judging the adequacy of their process or its performance.
Regulatory agencies hold management accountable for the operations of an
organization. It is the fi duciary responsibility for process owners to share the output
and performance of the operation with management and be able to explain and
interpret those metrics. Management has the responsibility to know the operations,
its performance, and take appropriate action to ensure compliance with government,
industry, and company policies and regulations.
Regulations require an annual product review be conducted of pharmaceutical
products to determine and assess changes made to processes that may affect product
quality. However, good industry practices would mandate quarterly or monthly
review for faster detection, decision, and action. Reviews need to include metrics on
key operating parameters and critical quality attributes to ensure product safety and
effi cacy. Several other key business metrics also benefi t the organization and should
be included in the metrics review program. The metrics collected should easily
provide the process owner and management with an indication if the process is in
control and delivering the desired results. If not, the process owner needs to present
management a proposal to pursue continuous improvement opportunities and be
able to describe required changes necessary to realize process enhancement.
3.3.7.3 Maturity Model
A maturity model is a useful management tool to determine process status and
provides a standard in which to value processes. It provides a standard in determin-
MEASURING SUCCESS THROUGH MEANINGFUL METRICS 277
278 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
ing the overall robustness and progression of a process and assists in the determination
of resource prioritization. It provides the basic framework to apply risk
management in determination of process development. For example, development
of high - risk systems, such as aseptic fi lling where high patient and business risk exist,
should be developed to a higher maturity than other processes with less patient or
regulatory risk. Business demands placed on the pharmaceutical and biopharmaceutical
industry limit resources in development, quality, and manufacturing requiring
wise deployment of these resources to the areas that can best benefi t the
organization.
An example of a maturity model can be seen in Figure 8 . This example provides
the QMS program group and leadership the ability to evaluate processes based on
an objective standard. It is divided into fi ve general levels, moving from informal,
unstructured to best in class. It includes specifi c deliverables for each level of the
model to be completed before a process can be considered to have achieved that
level. This model can also be divided into distinct subcategories, for instance, infrastructure,
performance, and compliance, which are depicted in Table 8 . Each subcategory
can be designed to provide meaningful information to the process owners
FIGURE 8 Maturity model overview. ( Source : Adapted from Capability Maturity Model
Integration, www.sei.cmu.edu . )
TABLE 8 Example Maturity Model
Theme
Level 1 (No
Formal
Approach)
Level 2
(Process
Defi ned)
Level 3
(Proactively
Managed)
Level 4
(Continuous
Improvement)
Level 5 (Best
in Class)
Compliance
Infrastructure
Performance
Source : Adapted from Capability Maturity Model Integration, www.sei.cmu.org .
and management. The maturity model is an excellent metric to measure development
of the QMS and focus leadership in deployment of resources.
The subcategories of the model demonstrate, through defi ned attributes that
must be in place, specifi c areas required of a robust process. The infrastructure category
includes a capable owner of the process is in place and a quality assurance
counterpart is identifi ed, the process owner has a strong understanding of the
process fl ow, scope, process boundaries, suppliers, customers, and roles and responsibilities.
The goal is to develop a highly integrated process that is fully transferable
and scalable.
Compliance is a key process attribute for a process in the pharmaceutical and
biopharmaceutical industry. Process maturity determination related to compliance
can include documentation such as standards and SOPs, number of observations
written against the process from internal audits, supplier audits, regulatory inspections,
nonconformances, and a risk assessment on the process against patient safety
and effi cacy. Training programs are also required as part of the process compliance.
Audit and inspection observations written against a process are key metrics indicating
maturity. Processes that can meet high maturity level for compliance represents
a well - managed process that is consistently delivering a compliant and quality
output.
Performance metrics determine process performance, preferably against
predetermined standards or expectations. Performance metrics should be indicators
as to the health and robustness of the system. Performance metrics may include
cycle turn around time, time to disposition from end of manufacture, and a risk
assessment against business drivers. The purpose is to raise target performance
objectives, developing a strategic approach, reducing variability, and improving
effi ciencies. Effi ciencies gained in process performance contribute to the business
needs.
Advantages to utilizing a maturity model are that it provides a useful methodology
for the QMS program group and leadership to evaluate, grade and provide
process owners a goal for process development. Again, using a risk - based mindset,
the entire inventory of process can be evaluated by leadership to determine where
to place resources and to what maturity level each process is best positioned to
support the enterprise.
Maturity - level goals are best made by process owners, the QMS program group,
and leadership. It is recommended that all processes are assessed against risk to the
customer and the business. This allows the QMS to prioritize the processes and
identify which of the processes need to be elevated to a higher maturity level in the
maturity model. Upon completion of the risk assessment, results should be reviewed
by leadership to determine if the processes have been prioritized appropriately and
meet the corporation ’ s goals. This feedback forum will ensure that leadership supports
process owners as they endeavor to achieve a higher level of process maturity.
A well - designed QMS will allow for two - way conversation between the leadership
and the process owners. It is as important for the leadership to communicate priorities
to the process owners as well as having the process owners communicate issues
and concerns that need to be addressed to the leadership. This will improve the
alignment of priorities between the leadership and process owners. This integration
ensures that the corporation is working on the right things at the right time with
the right people.
MEASURING SUCCESS THROUGH MEANINGFUL METRICS 279
280 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
3.3.7.4 Meeting Process Maturity Requirements
A dedicated team or review board should be developed to review and approve all
maturity - level deliverables upon completion of the attributes for the current level.
This review board ’ s purpose is to ensure that all deliverables meet a consistent level
of quality and documentation. This board can provide feedback to process owners
or QMS program group to communicate best practices and lessons learned.
A well - designed metric review program is essential to the success of the QMS.
The program should include metrics for the QMS, process maturity - level assessment
and process performance, infrastructure, and compliance metrics. These metrics are
the basis for evaluating system progress against long - term vision and annual quality
plans. The metrics provide leadership and process owners specifi c and objective data
to determine program goal achievement. Leadership will have visibility and comparability
of process performance within and between sites and have risk - based data
to support their deployment of resources in addressing business issues.
3.3.8 DRIVING CONTINUOUS IMPROVEMENT: PROJECTS
Pharmaceutical and biopharmaceutical companies are under signifi cant pressure to
deliver consistent quality product as well as drive the overall product cost down.
The goal of implementing ICH Q8, Q9, and ultimately Q10 is to characterize processes
based on risk assessments and improve them through a well - designed QMS.
There are regulatory and business drivers to continually improve the QMS processes
by building in quality and improve process effi ciency. The regulatory agencies
are now focused on ensuring systems are in place that protect the public health by
assuring both the safety and effi cacy of products. Understanding manufacturing
processes, through well - designed characterization studies, is one of the most effi cient
and effective methods to ensure process effi ciency. To meet business and consumer
demands as well as regulatory guidance and expectations, the implementation of
continuous improvement through risk - managed evaluations of manufacturing processes
is expected.
3.3.8.1 Process Improvements
A quality management system ’ s process should follow a standard Six Sigma process
improvement life cycle that includes the following steps: defi ne (process and metrics),
measure and control (identify problems and issues), analyze (analyze problems and
issues), and improve (implement) circling back to measure and control [11] . An
example of a process improvement life cycle can be seen in Figure 9 .
The basic foundation of continuous improvement begins with a process owner
who fully understands the process and recognizes how the process impacts other
processes within the QMS. Understanding this cause - and - effect relationship between
processes requires close integration between process owners and stakeholders. This
integration is critical throughout the entire life cycle of a process, from design
through development and management.
Prior to process improvements the process must be well - defi ned and predictable.
This does not mean that the process or output is desirable but instead well under
stood and predictable. It is through metrics, trends, and risk assessments that issues
and concerns should be evident. Process owners can use the management review
forum to present a proposal for process improvements.
3.3.8.2 Process Improvement Proposal
The process owner with stakeholders will need to provide a process improvement
proposal if the issue or change requires prioritization due to funds or additional
resources from the enterprise. The proposal should include, at minimum, the problem
and or opportunity statement, impact to the site based on risk, and proposal of an
action and/or project, including both cost and resource requirements.
During the development of the proposal, the process owner should consider
requesting subject matter experts to assist with the development of the problem
statement, risk assessment, and cost avoidance or savings. Many times a process
owner ’ s core competencies align closely with the process but may lack business or
project management skills. The process owner may need assistance to clearly articulate
to the leadership what the benefi ts are to accept the proposed change versus
the risks for not adopting the proposal.
Risk assessment tools such as a nine - block risk assessment (Table 9 ) or a failure
mode and effect analysis (FMEA) are available to assist the process owner with
the evaluation of the process or issue to better understand and communicate the
FIGURE 9 Continuous improvement process.
Monitor /
control Analyze
problem
Improve /
implement
Identify
problems/
opportunities
Define process
and metrics
TABLE 9 Nine - Block Risk Assessment Matrix
Severity
Minor Major Severe
Frequency
Probable Medium High High
Occasional Low Medium High
Remote Low Low Medium
DRIVING CONTINUOUS IMPROVEMENT: PROJECTS 281
282 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
probability of failure and the severity of a process issue. These analyses assist with
the prioritization of issues and identifi cation of specifi c actions required to mitigate
risks or the identifi cation of contingency plans for issues that are not mitigated. In
a pharmaceutical and biopharmaceutical environment it is important that all risk
tools are completed assessing impact to product safety and effi cacy as well as the
business drivers. Any steps or issues in the process that can negatively impact the
safety and effi cacy of the product require immediate elevation to leadership and
must be addressed immediately. If the proposal requires prioritization, the process
owners should clearly identify potential cost savings or avoidance through a cost of
quality model to further engage senior leadership. The combination of risk and costs
is an effective way to gain leadership support and attention.
Leadership ’ s role in the process improvement proposal is to understand the issue
or opportunity, understand associated risk(s), and approve or redirect a proposed
action or project and provide appropriate funding and or resources. As action items
and proposals are approved and initiated, the progress should be monitored on a
routine basis to ensure appropriate progress is made.
3.3.8.3 Task versus Project
Process improvements may be conducted by the completion of a task or a project.
A task is an activity that can be completed by the process owner with minimal cost
and/or resources over a short period of time. A project is defi ned as temporary work
to provide a product or service that is beyond the process owner ’ s support. In
general, a project requires more than one full - time equivalent (FTE), crosses over
multiple functional organizations, and the duration of the effort spans over a longer
period of time. Improvement status, updates, and issues should be discussed on a
regular basis by a management forum or steering committee. Tasks and projects
should be prioritized based on the risk against patient safety and effi cacy and
compliance.
If the process improvement meets the requirements of a project, a project
manager should be identifi ed. Formal project management allows for a holistic and
integrated approach to the change. The project manager should not replace the
process owner but ensure that the issues are identifi ed, prioritized, and resources
are applied, milestones are met, issues escalated and resolved, and progress reported.
The process owner needs to be the project lead with the stakeholders or steering
committee, providing support and guidance. This allows the process owner to focus
on the issues and improvements (their core competencies) and allows the project
manager to move the project forward in a methodical manner. During the project
it is critical that success is defi ned and measured.
3.3.8.4 Project Metrics
Project metrics should be identifi ed to measure the actual benefi t of the change
versus the expected result following the implementation. Many times, corporations
implement a change and move on to the next project without fully understanding
whether or not the changes achieved the desired result. A project that does not
achieve the expected benefi ts can lead to an ineffective process, confl icts with associated
touch points with other processes, or frustration from staff and customers.
Applying a systems - based approach to continuous improvement of the QMS,
utilizing formal risk management tools benefi ts the overall effi ciency of the organization.
Process owners are accountable and empowered to drive continuous improvements.
Metrics are utilized to identify trends, issues, and opportunities. Stakeholders
are engaged throughout the process, and management is involved in the prioritization
and staffi ng of the task or project. The processes are continually managed and
evaluated. Continuous improvements based on risk allow the organization to apply
resources and money to the most critical projects that will make the most impact.
As process improvements are implemented, staff will benefi t from a predictable,
lean process allowing them to focus on the proactive nature of their work as
opposed to the high stress of reacting to the issue of the day. The process owner will
gain credibility as he or she demonstrate the ability to ensure that the right people
are making the right decisions in a timely manner and that process improvements
are addressing systemic process problems and not superfi cially addressing issues
that will resurface again.
3.3.9 ENSURING ONGOING SUCCESS
Building infrastructure to establish and maintain a quality management system
requires resources and resolve from leadership and staff. The current pharmaceutical
and biopharmaceutical global and regulatory environment requires an organization
invested in developing and maintaining a robust system and processes meeting
the organization ’ s requirements for producing quality product. Future competition,
shorter time to market, effi cient development, and fi rst - pass approval expectations
exacerbate the need for robust processes. The global marketplace continues its pressure
on industry to deliver lifesaving and life - style changing medicines faster and
cheaper.
3.3.9.1 Establishing Mutual Goals
Companies that have designed, developed, and established QMSs and processes
that are simple to execute, easy to understand, and deliver the business and regulatory
results will have competitive advantage over their industry peers. They will be
faster and more effi cient at adapting new technologies, assimilating new organizations
through merger and acquisitions, able to apply adequate resources to appropriate
business needs, and most importantly quickly modify and adapt to changing
marketplace demands. Dependence on people, fragmented procedures, or tribal
knowledge, rather than integrated, functional processes, will bring undesirable
results to all levels of the organization.
Ensuring ongoing success requires establishing mutual goals for the organization
from the beginning. These goals must satisfy the needs of the business, the employees,
and the shareholders. Well - designed processes with accountable ownership that
have been established through discussion, design, and support of leadership,
functional management, operational stakeholders, and general staff provide the
foundation for common shared needs (Figure 10 ). If anyone of these groups is not
considered, nominal support and eventually failure of the program can be
expected.
ENSURING ONGOING SUCCESS 283
284 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
These shared goals need to be memorialized through documentation of the
program. This includes outlining the long - term objectives of the program, benefi ts
required to be achieved for stakeholders, and annual quality plans for achieving
milestone goals and success. Program success comes from leadership support, robust
system design, adequate training for employees, and meaningful metrics to measure
performance and continuous improvement efforts.
Mechanisms to determine stakeholder feedback on program acceptance, clarity,
and improvement opportunities need incorporation into the ongoing maintenance
of the QMS. Focus groups are one method of obtaining this type of information.
Another opportunity exists with regulatory inspections and customer audits. Taking
appropriate action to implement program changes and enhancements while recognizing
contributors will ensure stakeholder support and participation.
Mutual goals will drive success of the program and provide the reference for why
the system approach is needed and the benefi ts it can bring. There is no better situation
than having an entire organization aligned around the business design, and
executing against it, while supporting each other.
3.3.9.2 Rewards and Recognition
Process owners ’ responsibilities are signifi cant. Owners need to be selected from
predetermined criteria that are discriminatory in nature. Process owners are the
drivers of the operations and therefore need to be recognized for their special
efforts and responsibilities. This recognition can take many forms. A signifi cant distinction
in base qualifi cations and rewards is a valuable incentive to becoming a
process owner. Ongoing development for process owners is another incentive and
reward for the process owner. In addition to the fi nancial and tangible rewards,
being recognized by the organization to have the confi dence of management is also
another form of reward and recognition. Inevitably, processes contain waste and
FIGURE 10 Process owner support model.
ineffi ciencies, thereby providing another opportunity for owners to improve their
process and be recognized for that improvement.
Public recognition of system program and process owner accomplishments is
essential. This can easily be accomplished through regular review sessions, at metric
review meetings, through staff meetings and updates, poster sessions, newsletters,
and departmental meetings. Simple recognition and small gifts are appreciated and
reinforce management ’ s support and commitment to the program. Process owners
and stakeholders are the most infl uential group to spread the word on the usefulness
of the program and must be cultured to ensure ongoing success.
Studies indicate fi nancial rewards alone cannot provide employee satisfaction
and retention. High employee turnover costs companies tremendous fi nancial and
competitive resources. Many employees faced with equal or higher pay but unsatisfying
work will move onto another company or position. A poorly integrated QMS
with complicated processes is often the foundation for that dissatisfaction. To repeat
work, lose valuable time, or deliver substandard product does not satisfy today ’ s
highly educated and competitive worker in the pharmaceutical and biopharmaceutical
industry. The cost to recruit, replace, relocate, and retrain employees is signifi -
cant. Avoidance of these costs can be used as a partial basis for support of the
program.
3.3.9.3 Ensuring Ongoing Program Continuity
Accomplishments of a comprehensive QMS program should be shared between
locations and be consistent. Common, competent leadership for the enterprise will
ensure consistency. A consistent QMS program also allows for transfer of staff
between sites with little or no training and assimilation requirements. Divergent
evolution will dilute the QMS effort and support. Flexibility to execute is important,
however, caution must be exercised to restrict diverging language, interpretation,
and philosophy. Within a short time of a global execution, effi ciencies will be quickly
realized. Ensuring consistency also increases the number of process users with
similar experiences and leverages focus for process improvements and therefore
support.
Regulators and customers require assurance in consistency of pharmaceutical
and biopharmaceutical manufacturing operations. Today ’ s manufacturing supply
chains require multiple sites in varying locations to produce a product. Quality
systems must be perceived as an integral part of the value chain. This requires that
all sites be compliant in their operations and systems. Strong areas in one location
do not make up for weak or absent systems in another location. Fines are levied
and business is made or lost based on the individual site or weakest link in the
supply chain. Management must have a mechanism to measure its processes, and a
comprehensive QMS is the mechanism to demonstrate capability.
3.3.9.4 Program Institutionalization
Program institutionalization is realized with time. All levels of the organization
need to recognize and verbalize that the quality management system approach is
the way business is conducted. This way of doing business will become part of
the culture to the point at which it is second nature to leadership, management,
ENSURING ONGOING SUCCESS 285
286 CREATING AND MANAGING A QUALITY MANAGEMENT SYSTEM
and staff. Regulators and customers will recognize the benefi ts, as do the shareholders
and patients.
REFERENCES
1. Webster ’ s New Collegiate Dictionary , ninth edition, 1986 , p. 1199 .
2. Arling , E. R. ( 2004 ), Integrating QSIT into quality plans , Biopharm. Int. , June, 44 – 46, 48 ,
50 – 52 .
3. Drug Industry Daily , Oct. 12, 2006, Vol. 5, No. 200, Washington Business Information.
4. 2006 PDA/FDA joint regulatory conference , Sept. 11, 2006, Washington, DC.
5. Joneckis , C. ( 2006 ), Ph.D. presentation at 11th Annual GMP by the Sea, Aug. 28 – 30,
Cambridge, MD.
6. Quality systems regulations CDRH , available: www.fda.gov .
7. CPGM 7356.002, CDER, available: www.fda.gov .
8. IOM biological inspections 7345.848 CBER, available: www.fda.gov .
9. Rockefeller , J. D. , III ( 1973 ), The Second Revolution , Harper - Row , New York .
10. Kotter , J. P. ( 2001 ), Leadership insights , Harvard Bus. Re. , p. 29 .
11. George , M. (2004), Lean Six Sigma Pocket Tool Book , McGraw - Hill Professional , New
York , p. 4 .
287
3.4
QUALITY PROCESS IMPROVEMENT
Jyh-hone Wang
University of Rhode Island, Kingston, Rhode Island
Contents
3.4.1 Diagnosing a Process
3.4.1.1 Introduction
3.4.1.2 Basic Tools for Diagnosing a Process
3.4.2 Stabilizing and Improving a Process
3.4.2.1 Introduction
3.4.2.2 Control Charts for Attributes
3.4.2.3 Control Charts for Variables
3.4.2.4 Special Control Charts
3.4.3 Improving Performance of a Process
3.4.3.1 Introduction
3.4.3.2 Process Capability and Improvement Studies
Bibliography
3.4.1 DIAGNOSING A PROCESS
3.4.1.1 Introduction
Quality process improvement starts with a diagnostic journey where problems are
identifi ed. Remedial activity will be taken and the process will be continuously
monitored afterward. The common activities taken in the diagnostic journey are
analyzing symptoms, formulating hypotheses, testing hypotheses, and identifying
causes. Table 1 describes basic tools for the diagnostic journey. A description of them
is given in Section 3.4.1.2 .
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
288 QUALITY PROCESS IMPROVEMENT
3.4.1.2 Basic Tools for Diagnosing a Process
Cause - and - Effect Diagram A cause - and - effect diagram relates potential causes
of a problem to their effects. This is a tool that could be very useful in diagnosing
a process. It focuses on the possible causes of a specifi c problem in a structured and
systematic way. The following steps are suggested for constructing a cause - and -
effect diagram:
1. Defi ne the problem (effect).
2. Write problem on the right side and draw an arrow from the left to the right
side.
3. Brainstorm the main categories of causes of problems and draw major branch
arrows to the main arrow.
4. For each major branch, detailed causal factors (subcauses) are drawn as
subbranches.
5. Write sub - subcauses branching off the subcauses.
6. Ensure all the items that may be causing the problem are indicated in the
diagram.
Figure 1 shows a cause - and - effect diagram which is used to identify causes to yield
a problem in a biopharmaceutical manufacturing process. Possible main causes and
subcauses are identifi ed. Once the causes are identifi ed, other tools are employed
to determine the contribution of various causes to the effect. Actions are taken to
eliminate or minimize the impact of these causes.
Pareto Chart The Pareto principle suggests a problem (effect) can be attributed
to relatively few causes. In quantitative terms, 80% of the problems come from 20%
of the causes (machines, raw materials, operators, etc.); therefore effort aimed at the
right 20% can solve 80% of the problems. A Pareto chart includes three basic elements:
(1) the causes to the total effect, ranked by the magnitude of the contribution;
(2) the frequency of each cause; and (3) the cumulative - percent - of - total effect of
TABLE 1 Basic Quality Process Improvement Tools during Process Diagnosis
Common Activities to
Diagnose Cause
Basic Tools for Quality Process Improvement
Cause – Effect
Diagram
Pareto Chart
Histogram
Scatter Diagram
Normal
Probability Plot
Flow Diagrams
Data Collection
Box Plot
Stratifi cation
Analyzing symptoms • • • • • • •
Formulating hypotheses • • • •
Testing hypotheses • • • • • •
Identifying cause(s) • • • • •
Note : ( • ) major; ( ) minor.
0 0
0 0 0 0 0
0 0 0
0 0 0 0
0
DIAGNOSING A PROCESS 289
the ranked causes. Figure 2 gives an example of a Pareto chart which exhibits errors
found in a pharmacy store chain in one month.
Histogram A histogram is a graphic summary of variation in a set of data. Data
are clustered into categories and the values of individual clusters are plotted to give
a series of bars. For illustration, Table 2 presents 40 observations on the shelf life of
a certain drug and their frequency distribution. Figure 3 gives a histogram for the
drug shelf life data.
Scatter Diagram A scatter diagram is a basic tool to identify the potential relationship
between two variables. Scatter diagrams are similar to line graphs in that
they use horizontal and vertical axes to plot data points. However, they have a very
specifi c purpose. Scatter diagrams show how much one variable is affected by
another. The relationship between two variables is called their correlation. The
FIGURE 1 Cause - and - effect diagram.
Agitation pH Coolant
flow
Concentration
Substrate
Flow rate
Feed
Temperature
Aeration
Oxygen Water
temperature
Yield
FIGURE 2 A Pareto chart showing pharmacy errors.
Count
Percent
Pharmacy error
Count
29.8 12.4 6.3 2.1 1.2
Cum % 48.1 78.0 90.4 96.7
2452
98.8 100.0
1520 632 320 108 62
Percent 48.1
Miss
ed
drug
allerg
ies
W
rong
patient
Mixing
up
prescriptions
Incorrect
label
Incorrect
dosing
Misread
pres
cription
5000
4000
3000
2000
1000
0
100
80
60
40
20
0
290 QUALITY PROCESS IMPROVEMENT
closer the data points come when plotted to making a straight line, the higher the
correlation between the two variables. If the data points make a straight line going
from the origin out to high x and y values, then the variables are said to have a
positive correlation. If the line goes from a high value on the y axis to a high value
on the x axis, the variables have a negative correlation. Figure 4 gives a few examples
of scatter diagrams.
Normal Probability Plot The normal probability plot is a graphical technique for
assessing whether or not a data set is approximately normally distributed. The data
are plotted against a theoretical normal distribution in such a way that the points
form an approximate straight line. Departures from this straight line indicate departures
from normality. The normal probability plot is important for quality process
improvement since many other tools require the normality assumption. A normal
TABLE 2 Drug Shelf Life (days)
102.2 104.1 103.5 104.5 103.2 103.7 103.0 102.6
103.4 101.6 103.1 103.3 103.8 103.1 104.7 103.7
102.5 104.3 103.4 103.6 102.9 103.3 103.9 103.1
103.3 103.1 103.7 104.4 103.2 104.1 101.9 103.4
104.7 103.8 103.2 102.6 103.9 103.0 104.2 103.5
Range Midpoint Frequency Cumutative %
101.5 . x < 102.0 101.75 2 5.00
102.0 . x < 102.5 102.25 2 10.00
102.5 . x < 103.0 102.75 5 22.50
103.0 . x < 103.5 103.25 15 60.00
103.5 . x < 104.0 103.75 8 80.00
104.0 . x < 104.5 104.25 6 95.00
104.5 . x < 105.0 104.75 2 100.00
FIGURE 3 Histogram of drug shelf life.
0
2
4
6
8
10
12
14
16
101.75 102.25 102.75 103.25 103.75 104.25 104.75
Shelf life
Frequenc
. 00%
20.00%
40.00%
60.00%
80.00%
100.00%
120.00%
Frequency
Cumulative %
DIAGNOSING A PROCESS 291
probability plot of the drug shelf life in Table 2 is given in Figure 5 . As seen from
the plot, the observations follow a straight line and are contained in the 95% confi -
dence interval. It can thus be said that the shelf life of this drug follows a normal
distribution.
Other Tools
Box Plot This plot is useful when analyzing the pattern of the data. It displays
several important features of data such as central tendency, variability, departure
from symmetry, and presence of outliers.
Flow Diagrams A process fl ow diagram can be used to study and understand
the process.
Data Collection Data are essential for making a proper evaluation of the current
process. Tools for data collection include checklists and data sheets.
FIGURE 4 Scatter diagrams.
X
Y
No correlation
X
Y
Negative correlation
X
Positive correlation
Y
292 QUALITY PROCESS IMPROVEMENT
Stratifi cation This technique is used to separate data into groups based on categories
or characteristics. It is the basis for the application of other tools or it
can be used with other data analysis tools such as scatter diagrams.
3.4.2 STABILIZING AND IMPROVING A PROCESS
3.4.2.1 Introduction
Basic Concepts of a Control Chart The control chart is one of the main tools for
quality process improvement. It is used to assess the nature of variation in a process
and to facilitate the forecasting and management of a process. Values of the quality
characteristic are plotted against the sample number or time, as shown in Figure 6 .
The centerline represents the process average. The upper and lower control limits
(UCL and LCL) are usually chosen as three standard deviations (SDs) above and
below the centerline so they can be used to detect “ out - of - control ” situations without
causing mamy false alarms. An out - of - control situation is usually signaled by a
plotted point falling outside the control limits or a cluster of plotted points forming
an abnormal pattern.
FIGURE 5 Normal probability plot of drug shelf life with 95% confi dence interval.
Shelf life
Percent
106.0 105.5 105.0 104.5 104.0 103.5 103.0 102.5 102.0 101.5 101.0
99
95
90
80
70
60
50
40
30
20
10
5
1
95% confidence interval
FIGURE 6 Normal curve - based control chart.
Upper control limit (UCL)
Center line
Lower control limit (LCL)
Sample number
X
68.26%
95.46%
99.73%
–3. –2. –1. X +1. +2. +3.
Plotted points on a control chart are usually based on data collected from samples
in a process. After a suffi cient number of samples are collected and the data are
plotted on a control chart, the stability of the process can be evaluated. A stable
process is “ in control ” while an out - of - control process is unstable. Depending on
the type of quality characteristic, control charts can be divided into two groups:
variable control charts and attribute control charts. Variable control charts are used
to monitor quality characteristic that are continuously varying in nature; attribute
control charts are used to monitor those quality characteristics that are not numerically
measurable. The determination of the centerline and control limits are described
in Sections 3.4.2.2 and 3.4.2.3 with respect to different types of control charts.
Applications of Control Charts Control charts serve to direct management attention
toward special causes of variation in a process when they appear. In evaluating
control charts, the following symptoms could indicate a process that is out -
of - control:
• Outlier One or more point(s) that fell outside the control limits.
• Run A series of plotted points above or below the centerline.
• Trend A continual rise or fall of plotted points.
• Cyclicity A pattern that repeats itself over time.
The following steps are usually followed in a control chart ’ s development and
application:
• Determine a “ base period ” for initial control chart development.
• Collect sample data from the base period.
• Calculate the parameters for the control chart, that is, centerline and control
limits.
• Plot collected sample points on the chart with the centerline and control
limits.
• Determine whether the chart parameters can be used to monitor the process;
revise the parameters if necessary.
• Collect ongoing samples and continue monitoring the process using the
developed control chart.
• Conduct periodic audits on the parameters of the control chart.
Variable control charts are widely applied in many manufacturing and nonmanufacturing
settings. They can be used to monitor, for example, the inside diameter of
an aircraft bearing, the moisture content of a drug tablet, the net weight of a pharmaceutical
product, the processing time of phone inquiries, and the satisfaction level
of customers. The latter two are examples of nonmanufacturing applications.
Attribute control charts are less used compared to variable control charts. When
it is not possible or practical to measure the quality characteristic of a product,
attribute control charts are often applied. Examples of their application include
monitoring the fraction of nonconforming of a certain sensor production, the number
of defective diodes in an electronic assembly, the number of imperfections in textile
STABILIZING AND IMPROVING A PROCESS 293
294 QUALITY PROCESS IMPROVEMENT
production, the fraction of defective batches in a biomedical manufacturing production,
and the number of errors found in a pharmacy store.
In most applications, the choice between a variable control chart and an attribute
control chart is clear - cut. In some cases, the choice will not be obvious. For instance,
if the quality characteristic is the softness of an item, such as the case of pillow
production, then either an actual measurement or a classifi cation of softness can be
used. Quality managers and engineers will have to consider several factors in the
choice of a control chart, including cost, effort, sensitivity, and sample size. Variable
control charts usually provide more information to analysts but cost more to implement
and use. Attribute control charts are less sensitive and expensive but usually
requires large samples to reach certain statistical signifi cance.
3.4.2.2 Control Charts for Attributes
Control charts based on attribute data include the p chart, np chart, c chart, and u
chart. The former two are applied when fraction nonconforming or number of nonconforming
is a concern, and the latter two are used to deal with the nonconformities.
Most pharmaceutical manufacturing industries employ one or more of these
charts.
p Control Chart A p chart can be used to monitor the fraction nonconforming of
a process. Fraction nonconforming is defi ned as the ratio of the number of nonconforming
items in a population to the total number of items in that population. In
pharmaceutical manufacturing, an item will be classifi ed as nonconforming if it fails
to conform to standards on one or more attributes, for example, fi ll volumes of vials,
moisture content, hardness, and solubility.
Let us suppose a random sample of n items is selected and examined from a
process running with a stable nonconforming rate p and D units of nonconforming
items are found; then D is a random variable following a binomial distribution
with parameters n and p . If the true fraction nonconforming, p , is known, then the
parameters of the p chart are
UCL
Centerline
LCL
= +
.
=
= .
.
p
p p
n
p
p
p p
n
3
1
3
1
( )
( )
(1)
In practice, the fraction nonconforming, p , is unknown most of the time and is thus
needed to be estimated from the sample data. An estimated p i can be calculated for
the i th sample collected and an average p. value can be obtained as an arithmetic
average of those individual p i found from the m samples:
p
D
mn
p
m
i
m
i i
m
i = = = = . . 1 1
(2)
The p. can then be used in place of p in Equation (1) in the application. It should be
noted that the p. value needs to be assessed periodically to assure its representativeness
of the average process fraction nonconforming.
np Control Chart An np chart is used to monitor the number of nonconforming
items produced in a process. Very similar to the p chart, the parameters of an np
chart are
UCL
Centerline
LCL
= + .
=
= . .
np np p
np
np np p
3 1
3 1
( )
( )
(3)
As in the p chart, if the actual p value is not available, p. can be used in the
calculation.
c Control Chart A c chart can be used to monitor the number of nonconformities
(defects) per inspection unit. An inspection unit can be a single unit of product, a
batch of multiple products, or a certain measured volume (weight) of product. Many
pharmaceutical manufacturing processes are lot based where raw material or semiproduct
passes from one process to the next. For example, an inappropriately coated
tablet in a coating process can be considered as a nonconformity (defect) where an
inspection unit might be defi ned as 1 kg of the tablet.
Suppose an inspection unit of a certain product is selected and examined from a
process running with a stable nonconformity rate c per inspection unit and X nonconformities
are found. Then X is a random variable following a Poisson distribution
with parameter c . If the true nonconformity level c is known, then the parameters
of the c chart are
UCL
Centerline
LCL
= +
=
= .
c c
c
c c
3
3
(4)
If the actual nonconformity level c is unknown, it can be estimated by using average
c values obtained from m inspection units collected in a base period:
c
C
m
i
m
i = = . 1
(5)
The c. can then be used in place of c in Equation (4) in the application. Since it is
possible to obtain a negative LCL using Equation (4) , a value of zero should be
used in that case.
u Control Chart A u chart is used to monitor the rate of nonconformities. The
rate of nonconformities ( u ) is the number of nonconformities ( x ) in an inspection
unit divided by the number of physical units ( n ) inspected (e.g., 100 ft of pipe, 100
items in a batch). Similar to the c chart, the parameters of a u chart are
STABILIZING AND IMPROVING A PROCESS 295
296 QUALITY PROCESS IMPROVEMENT
UCL
Centerline
LCL
= +
=
= .
u
u
n
u
u
u
n
3
3
(6)
If the actual u value is not available, can be used in Equation (6) .
Example 1 A medical device manufacturer is concerned about the nonconforming
(defective) and the nonconformity (defect) produced in its recently set - up production
line. Twenty batches of this medical device were randomly selected from the production
line. Each batch contained 100 units. Each unit is inspected and is classifi ed as
either “conforming” or “nonconforming. ” During the inspection, the number of nonconformities
(defects) was also counted. The data collected are shown in Table 3 .
3.4.2.3 Control Charts for Variables
Control charts based on variable sample data include the x. chart and the s chart.
When dealing with a numerically measurable quality characteristic, the x. chart is
usually employed to monitor the process average and the s chart is used to monitor
the process variability. When there is only one observation in each sample, the individual
measurement chart ( I chart) and moving range chart (MR chart) are used to
monitor the process average and variability. It should be noted that due to the poor
TABLE 3 Nonconforming and Nonconformity Counts of 20 Batches of Medical Device
Batch number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Nonconformings 3 2 4 2 5 2 1 2 0 5 2 4 1 3 6 0 1 2 3 2
Nonconformities 9 7 13 8 6 8 10 10 7 10 12 9 11 15 8 12 11 8 7 15
FIGURE 7 p Chart for medical device manufacturing example.
Sample no.
Fraction nonconforming
19 17 15 13 11 9 7 5 3 1
0.08
0.06
0.04
0.02
0.00
_
P=0.025
UCL=0.07184
LCL=0
Based on the data in Table 3 , the average fraction nonconforming, p. , can be
obtained as 2.5%; the average nonconformity per batch, c. , is 9.8; and the average
nonconformity per unit, , is 0.098. The resulting control charts are shown in Figures
7 – 10 . These charts indicate that the process is in control and thus the parameters
established here can be used to monitor future productions.
u
u
FIGURE 8 np Chart for medical device manufacturing example.
Sample no.
No. of nonconformities
19 17 15 13 11 9 7 5 3 1
8
6
4
2
0
__
np =2.5
UCL=7.184
LCL=0
FIGURE 9 c Chart for medical device manufacturing example.
Sample no.
No. of nonconformities
19 17 15 13 11 9 7 5 3 1
20
15
10
5
0
_
C=9.8
UCL=19.19
LCL=0.41
FIGURE 10 u Chart for medical device manufacturing example.
Sample no.
Nonconformity rate
19 17 15 13 11 9 7 5 3 1
0.20
0.15
0.10
0.05
0.00
_
U=0.098
UCL=0.1919
LCL=0.0041
sample size, the I and MR charts are less sensitive in detecting if the process is out
of control than the x. and s charts.
x. Control Chart An x. chart is used to monitor the process average. It is usually
used in pharmaceutical manufacturing where multiple units are collected in each
sample (e.g., a sample of multiple tablets formed by dry powders or wet granules.)
Due to contamination risk and cost of sampling (including product loss due to
STABILIZING AND IMPROVING A PROCESS 297
298 QUALITY PROCESS IMPROVEMENT
sample volumes and incurred labor cost of laboratory analysis), the sample size is
usually kept small.
Sample means x. are plotted on the x. chart. Assume that random samples of n
items are collected and examined from a stable process with a process mean . and
standard deviation . . Then x. can be considered as a random variable following a
normal distribution with mean . x. , and standard deviation . x. , where
. . .
.
x x n
= = and
(7)
If the true process mean . and standard deviation . are known, then the parameters
of the x. chart are
UCL
Centerline
LCL
= + = +
= =
= . = .
. . .
.
. .
. . .
.
x x
x
x x
n
n
3 3
3 3
(8)
Since . and . are not usually known, estimators of them can be obtained from the
sample means ( x. ) and sample standard deviations ( s ) of the m samples collected in
the base period:
Estimator of
Estimator of
.
.
= =
=
. .
=
=
=
.
.
x
x
m
s m
n n
s
c
i
m
i
i
n
i
1
1
4 4 1 4 3 ( )
(9)
Using the estimators, the parameters of the x. chart are now
UCL
Centerline
LCL
= + = +
=
= . = .
x
s c
n
x A s
x
x s c
n
x A s
3
3
4
3
4
3
(10)
The common values of constants c 4 and A 3 are tabulated in Table 4 for sample sizes
from 2 to 10. Like other control charts, the values of x and s. should be periodically
verifi ed to assure that they can be used to derive good estimators for the process
average and process standard deviation.
s Control Chart An s chart is used to monitor the process variability. Since it is
equally important to ascertain that the variability and the mean of a process are in
control, an s chart is usually used in conjunction with the x. chart. Sample standard
deviations are plotted on the s chart. Consider s as a random variable with mean . s
and standard deviation . s . Then the parameters of the s chart can be stated as
UCL
Centerline
LCL
= +
=
= .
. .
.
. .
s s
s
s s
3
3
(11)
In practice, the parameters of the s chart can be estimated using s. as
UCL
Centerline
LCL
= + . =
=
= . . =
s
s
c
c Bs
s
s
s
c
c Bs
3 1
3 1
4
4
2
4
4
4
2
3
(12)
If the LCL calculation results in a negative value, use zero as the LCL .
TABLE 4 Values of Constants in Variable Control Chart
Parameters
n A 3 c 4 B 3 B 4 d 2 d 3
2 2.659 0.798 0 3.267 1.128 0.853
3 1.954 0.886 0 2.568 1.693 0.888
4 1.628 0.921 0 2.266 2.059 0.880
5 1.427 0.940 0 2.089 2.326 0.864
6 1.287 0.952 0.030 1.970 2.534 0.848
7 1.182 0.959 0.118 1.882 2.704 0.833
8 1.099 0.965 0.185 1.815 2.847 0.820
9 1.032 0.969 0.239 1.761 2.970 0.808
10 0.975 0.973 0.284 1.716 3.078 0.797
Example 2 In a Pet Tabs (pet vitamin tablets) production, the pharmaceutical
manufacturer is using milling and micronizing machines to pulverize raw materials
into fi ne particles. These fi nished particles are combined and processed further in
mixing machines. The mixed ingredients are then pressed into tablets, dried, and
sealed in packages. A normally distributed quality characteristic, moisture content,
is monitored. Samples of n = 4 tablets are taken from the manufacturing process
every hour. The data after 25 samples have been collected are shown in Table 5 .
From these data, it is found that x = 10 254 . and s. = 0.926. Using Equations (10)
and (12) , the parameters of the x. and s charts are found as:
x. Chart s Chart
UCL 11.761 2.098
Centerline 10.254 0.926
LCL 8.747 0
The control charts are shown in Figure 11 . The x. and s charts show that the process
is in control and thus the parameters established here can be used to monitor future
productions.
STABILIZING AND IMPROVING A PROCESS 299
300 QUALITY PROCESS IMPROVEMENT
TABLE 5 Moisture Content (%) of 25 Samples of Pet Tabs
Sample Number
Observations
x. s 1 2 3 4
1 7.84 11.01 10.14 9.41 9.600 1.343
2 10.51 9.1 9.52 10.83 9.990 0.814
3 9.74 10.39 9.62 11.16 10.228 0.708
4 10.71 11.41 10.71 8.63 10.365 1.203
5 9.93 10.95 8.99 10.73 10.150 0.889
6 9.94 10.27 9.35 9.42 9.745 0.438
7 12.11 9.72 8.89 9.75 10.118 1.387
8 9.61 8.93 11.12 8.75 9.603 1.077
9 9.17 10.87 9.97 10.79 10.200 0.798
10 11.41 10.39 8.83 12.19 10.705 1.451
11 8.43 9.48 10.56 10.2 9.668 0.939
12 9.92 10.13 9.66 8.21 9.480 0.868
13 8.39 9.94 10.4 8.69 9.355 0.967
14 10.42 10.27 10.94 10.91 10.635 0.341
15 10.98 12.57 11.14 8.97 10.915 1.481
16 9.73 10.05 12.82 12.43 11.258 1.592
17 11.36 8.91 10.08 10.55 10.225 1.024
18 9.42 11.12 9.01 10.52 10.018 0.973
19 10.15 10.08 10.12 9.88 10.058 0.122
20 11.73 11.1 10.75 9.94 10.880 0.746
21 11.52 9.11 9.88 11 10.378 1.087
22 11.29 10.43 11.6 11.74 11.265 0.588
23 9.39 12.96 11.42 10.28 11.013 1.541
24 10.26 9.59 9.33 9.26 9.610 0.456
25 11.25 10.65 11.06 10.63 10.898 0.307
FIGURE 11 x. and s charts for Pet Tab manufacturing example.
Sample no.
Sample mean
25 23 21 19 17 15 13 11 9 7 5 3 1
12
11
10
9
__
X=10.254
UCL=11.761
LCL=8.747
Sample no.
Sample SD 25 23 21 19 17 15 13 11 9 7 5 3 1
2.0
1.5
1.0
0.5
0.0
_S
=0.926
UCL=2.098
LCL=0
Individuals Control Charts In some chemical and biopharmaceutical manufacturing
processes involving lengthy and expensive procedures, it is not feasible to form
a sample of size greater than one because only one product or one batch is available
each time. When the sample size used for statistical process monitoring is limited
to one, individual control charts, I and MR charts, are needed.
The I chart is serving the same function as the x. chart except that now x is the value
of the individual measurement. Assuming that x follows a normal distribution with
mean . and standard deviation . , the theoretical parameters of the I chart are
UCL
Centerline
LCL
= +
=
= .
. .
.
. .
3
3
(13)
The process average . can be estimated by x. , which is
.. = . = . x
x
m
i
m
i 1
(14)
Since only individual measurements are available, moving ranges need to be calculated
for the estimation of process standard deviation . . A k - point moving range,
MR k , can be calculated as
MR max( , . . . , min , . . . , k i i k i i k x x x x = . + + ) ( ) (15)
For m individual measurements, there are m . k MR k available, and the process
standard deviation . can be estimated as
.. = =
. =
. . MR MR k i
m k
ki
d
m k
d 2
1
2
(16)
The estimated process mean and standard deviation can be used to calculate the
practical parameters for the I chart in Equation (13) . The constant d 2 value is determined
by k and can be found by using k as n in Table 3 . Common k values can range
from 2 to 5.
The MR chart is used to monitor process variability. Considering MR k as a
random variable with a mean of .MRk and a standard deviation of .MRk, the theoretical
parameters of the MR chart can then be stated as
UCL
Centerline
LCL
MR MR
MR
MR MR
= +
=
= .
. .
.
. .
k k
k
k k
3
3
(17)
Since .MRk and .MRk are not usually available, they can be estimated as
. . . . MR MR MR and MR k k k k
d
d
= =3
2
(18)
STABILIZING AND IMPROVING A PROCESS 301
302 QUALITY PROCESS IMPROVEMENT
The constant value of d 3 can also be found in Table 3 . If a negative LCL was
obtained, use zero.
3.4.2.4 Special Control Charts
The control charts discussed earlier are very useful in the diagnostic aspects of
quality process improvement. They can be used to stabilize a process by identifying
out - of - control situations. After the process is stabilized and brought in control,
further improvement of the process can be achieved by using some special control
charts such as the cumulative sum (CUSUM) control chart and the exponentially
weighted moving average (EWMA) control chart. These control charts can be used
when “ small shifts ” in a process are of interest.
CUSUM Control Chart A CUSUM chart provides an effi cient way of detecting
small shifts in the mean of a process ( < 1/2 . ). For larger shifts ( > 1/2 . ), the x. chart
is usually used. The CUSUM chart incorporates information contained in a sequence
of sample points. It keeps track of the cumulative sum of the deviations between
each sample point (a sample mean) and a target value. Unlike the x. chart,
which often bases its out - of - control decision on just the most recently collected
sample, the CUSUM calculated for a sample point carries the “ history ” prior to
that sample. For example, a sequence of sample points above the centerline can
trigger an out - of - control signal although all of them stayed well below the UCLs of
the x. chart.
There are two forms of the CUSUM chart, the tabular form and the V - mask form.
Due to its practicality, the tabular form is more preferred in industrial settings. The
tabular CUSUM accumulates deviations from a target value (or a known process
mean . 0 ). Deviations above that target value are cumulated as a one - sided upper
CUSUM ( C + ) and deviations below the target value are cumulated as one - sided
lower CUSUM ( C . ):
C x k C
C k x C
i i x i
i x i i
+
.
+
.
.
.
= . + +
= . . +
max[0,
max ,
( ) ]
[ ( ) ]
. .
. .
0 1
0 1 0
(19)
where C C 0 0 0 + . = = .
The parameter k is called the allowance and is usually determined as the magnitude
of the shift to be detected in terms of . x. . If either Ci
+ or Ci
. exceeds a decision
interval h , the process is considered out of control. In other words, the value of h is
considered a UCL and . h is considered an LCL. Its centerline is always at zero. A
reasonable value for h is fi ve times the process standard deviation . .
EWMA Control Chart An EWMA control chart plots weighted moving average
values for variables data. A weighting factor is chosen by the user to determine how
older data points affect the mean value compared to more recent ones. Because the
EWMA chart uses information from all samples, it is a good alternative to the
CUSUM chart in detecting smaller process shifts.
The EWMA for sample i ( z i ) is plotted on the chart and is defi ned as z i = . x. i +
(1 . . ) z i . 1 , where z 0 = . 0 . The constant . defi nes the weight assigned to the current
sample (0 < . . 1) and 1 . . is the weight assigned to earlier samples. Parameters
of the EWMA are
UCL
Centerline
LCL
= +
.
. .
=
= .
.
. .
. .
.
.
.
.
. .
.
.
0
2
0
0
2
1 1
2
1 1
L
L
x
i
x
[ ( ) ]
[ ( .) ] 2i
(20)
where L is a design parameter that defi nes the width of the control limits. The choice
of L = 3 and 0.05 < . . 0.25 is reasonable. The control limits will become wider when
the sample number i is getting larger and fi nally reach constant values as
UCL
Centerline
LCL
= +
.
=
= .
.
. .
.
.
.
. .
.
.
0
0
0
2
2
L
L
x
x
(21)
Example 3 The data in Example 2 are now analyzed by CUSUM and EWMA
charts. Table 6 shows calculated CUSUM and EWMA values. The value of h in
CUSUM is chosen as 5 times the standard deviation of x. ( . . .x = 0 5027) and the value
of k is chosen as 0.5. The Ci
+ and Ci
. are calculated using a target value . 0 = 10.
The CUSUM chart is shown in Figure 12 . The value of . in EWMA is chosen as 0.2
and L is chosen as 3. The UCL and LCL for individual samples are shown in Table
6 and the EWMA chart is shown in Figure 13 . Although the x. and s charts in Figure
6 indicate that the process is in control, both CUSUM and EWMA gave out - of -
control signals at sample point 22. A small process shift has occurred after sample
21.
IMPROVING PERFORMANCE OF A PROCESS 303
3.4.3 IMPROVING PERFORMANCE OF A PROCESS
3.4.3.1 Introduction
Basic Concepts After a process is diagnosed, corrected, and brought into statistical
control, the next question is “ How can the performance of a process be improved? ”
To answer this question, quality managers and engineers need fi rst measure the
present process performance. This measurement can be achieved through a process
capability study which gauges the ability of a process to produce products according
to the specifi cations. A process can achieve a state of statistical control but still exhibit
a poor capability due to the variability in the process. It will be necessary to reduce
variability to improve the process capability. Designed experiments based on statistical
principles can offer helps toward reduction of variability and optimization of the
process. Employing designed experiments, intentional changes can be made in various
places in the process; results gathered from these experiments can lead to further
process improvement and bring it to the next level. This section presents commonly
304 QUALITY PROCESS IMPROVEMENT
TABLE 6 CUSUM and EWMA Values for Pet Tabs Example
Sample Number x.
CUSUM EWMA
Ci
+ Ci
. z i UCL LCL
1 9.600 0.000 0.149 9.920 10.302 9.698
2 9.990 0.000 0.000 9.934 10.386 9.614
3 10.228 0.000 0.000 9.993 10.432 9.568
4 10.365 0.114 0.000 10.067 10.458 9.542
5 10.150 0.012 0.000 10.084 10.475 9.525
6 9.745 0.000 0.004 10.016 10.485 9.515
7 10.118 0.000 0.000 10.036 10.491 9.509
8 9.603 0.000 0.146 9.950 10.495 9.505
9 10.200 0.000 0.000 10.000 10.498 9.502
10 10.705 0.454 0.000 10.141 10.500 9.500
11 9.668 0.000 0.081 10.046 10.501 9.499
12 9.480 0.000 0.350 9.933 10.501 9.499
13 9.355 0.000 0.744 9.817 10.502 9.498
14 10.635 0.384 0.000 9.981 10.502 9.498
15 10.915 1.047 0.000 10.168 10.502 9.498
16 11.258 2.054 0.000 10.386 10.502 9.498
17 10.225 2.027 0.000 10.354 10.502 9.498
18 10.018 1.794 0.000 10.286 10.502 9.498
19 10.058 1.600 0.000 10.241 10.502 9.498
20 10.880 2.229 0.000 10.368 10.502 9.498
21 10.378 2.355 0.000 10.370 10.503 9.497
22 11.265 3.369 0.000 10.549 10.503 9.497
23 11.013 4.130 0.000 10.642 10.503 9.497
24 9.610 3.489 0.139 10.435 10.503 9.497
25 10.898 4.135 0.000 10.528 10.503 9.497
h = 2.513 . = 0.2
k = 0.5 L = 3
FIGURE 12 CUSUM chart for Pet Tabs manufacturing example.
Sample no.
Cumulative sum
25 23 21 19 17 15 13 11 9 7 5 3 1
5
4
3
2
1
0
-1
-2
0
UCL=2.513
LCL=-2.513
FIGURE 13 EWMA chart for Pet Tabs manufacturing example.
Sample no.
EWMA
25 23 21 19 17 15 13 11 9 7 5 3 1
10.75
10.50
10.25
10.00
9.75
9.50
UCL=10.503
LCL=9.497
Target = 10
used methods in process capability studies. Design - of - experiment techniques can be
found elsewhere in this handbook and in many other textbooks.
Specifi cation Limits, Control Limits, and Natural Tolerance Limits To conduct
a process capability study, it is important to distinguish the specifi cation limits of a
product, the control limits of the process producing the product, and the natural
tolerance limits (NTLs) of the product. In general, specifi cation limits are given by
customers or prescribed by in - house design engineers before production. A product
that failed to meet the specifi cations is a nonconforming product. Control limits are
usually determined by samples collected from a process during a base period. A
sample point that fell outside the control limits will trigger an out - of - control state;
however, a product produced in the out - of - control state is not necessarily a nonconforming
product. It should also be noted that a sample point plotted in a control
chart usually represents a statistic of the sample such as the sample mean. In other
words, a single product that fell outside of the control limits will neither cause the
process to be out of control nor become nonconforming. The variability of products
produced can usually be described by its natural tolerance limits. It is commonly
acceptable that the ± 3 standard deviations from the process mean be used as the
natural tolerance limits.
Example 4 Following Example 2 , the specifi cation limits are specifi ed as 10.00 ±
2.00, where:
Nominal or target value ( . 0 ) = 10.00
Upper specifi cation limit (USL) = 10.00 + 2.00 = 12.00
Lower specifi cation limit (LSL) = 10.00 . 2.00 = 8.00
The control limits for the x. chart are:
Center line ( x ) = 10.254
Upper control limit (UCL) = 11.761
Lower control limit (LCL) = 8.747
IMPROVING PERFORMANCE OF A PROCESS 305
306 QUALITY PROCESS IMPROVEMENT
3.4.3.2 Process Capability and Improvement Studies
Process Capability Indices Process capability indices provide a quantitative measure
to assess the ability of a process to produce products that meet the specifi cations. A
commonly used process capability index, denoted as C p , can be calculated as
Cp =
. USL LSL
6.
(22)
where USL is the upper specifi cation limit, LSL is the lower specifi cation limit, and
. is the process standard deviation. Since . is not usually known, it can be estimated
by .. = s c / 4 . A C p = 1 means that the process is just capable. If the process is centered
at its nominal value, it will produce 2,700 nonconforming products out of one million
(PPM). The target value C p is usually set at 1.33 for an existing process and 1.50 for
a new process.
It should be noted that the C p value could not indicate the proper process capability
if the process is not centered since C p does not account for where the process
mean is with respect to the specifi cations. To alleviate this issue, another process
capability index, C pk , is used:
C C C pk pu pl = min( , )
Using the x ± 3.
natural tolerance limits, they can be obtained as:
Process mean ( x ) = 10.254
Upper natural tolerance limit (UNTL) = 13.270
Lower natural tolerance limit (LNTL) = 7.238
The relationships among the three sets of limits are illustrated in Figure 14 . As can
be seen from this fi gure, the current process is not centered at its nominal value and
its specifi cation limits are tighter than its natural tolerance limits. Due to this, a
portion of manufactured products ( . 5.4%) will not be able to conform to the
specifi cations.
FIGURE 14 Specifi cation limits, control limits, and natural tolerance limits for Pet Tabs
manufacturing example.
LSL m0 USL
LNTL LCL x UCL UNTL
where
C C pu pl =
.
=
. USL
and
LSL
3
.
.
.
. 3
(23)
The . value can be estimated by x and the . value can be estimated as discussed
earlier. In general, a process is considered “ centered ” at the nominal value of the
specifi cations when C p = C pk and “ off centered ” when C p < C pk . The relationships
between C p and C pk are further illustrated in Figure 15 where the process mean has
shifted from . 0 to . 0 + 2 . to . 0 + 4 . . As noted from the fi gure, C p remains the same
regardless of the shift but C pk is signifi cantly reduced.
FIGURE 15 Relationships between C p and C pk .
0 2 4 6 8 10 12 14 16 18 20
0 2 4 6 8 10 12 14 16 18 20 22
0 2 4 6 8 10 12 14 16 18 20 22 24
LSL m0
s = 2
USL
Cp = 1.667
Cpk = 1.667
Cp = 1.667
Cpk = 1.0
Cp = 1.667
Cpk = 0.333
IMPROVING PERFORMANCE OF A PROCESS 307
If one - sided specifi cations are used, one - sided process capability can also be
defi ned by Equation (23) where C pu is for upper specifi cation and C pl for lower
specifi cation.
Interpretation and Improvement of Process Capability Evaluation and interpretation
of process capability represent an important step in process quality
308 QUALITY PROCESS IMPROVEMENT
improvement. A process must have its source of instability eliminated before it can
be improved. Results obtained from process capability studies can help determine
whether the process is stable and meeting its specifi cations. It should be noted that
a valid process capability study is based on the normality assumption of the process.
The normality assumption will need to be checked before proceeding to the next
step.
Conclusions regarding whether the process is centered at the target and is meeting
the specifi cations can be drawn from the process capability study. When C p = C pk ,
the process is centered. When C p has a value of 1.0 or greater, the process is capable
of producing products meeting specifi cations; otherwise, it is not capable.
Example 5 Following Example 2 , C p and C pk are calculated as
Cp =
.
=
.
.
= USL LSL
6.
12 8
6 1 0054
0 6633
.
.
C C C pk pu pl = = = min( , min , ) (. . ) . 0 5790 0 7476 0 5790
where
Cpu =
.
=
.
.
= USL .
. 3
12 10 254
3 1 0054
0 5790
.
.
.
Cpl =
.
=
.
.
=
.
.
LSL
3
10 254 8
3 1 0054
0 7476
.
.
.
Figure 16 shows the histogram of the data in relation to the specifi cations. The x. and
s charts in Figure 11 show that the process is in statistical control. However, since
C p < C pk , the process is not centered. With a C pk value of 0.579, it is expected to have
53,711 nonconforming Pet Tabs manufactured out of one million parts in this production
line.
FIGURE 16 Process capability plot for Pet Tabs manufacturing example.
BIBLIOGRAPHY
Aft , L. S. ( 1997 ), Fundamentals of Industrial Quality Control , 3rd ed., CRC Press , Boca Raton,
FL .
DeVor , R. E. , Chang , T. H. , and Sutherland , J. W. ( 2006 ), Statistical Quality Design and Control ,
2nd ed., Prentice - Hall , Upper Saddle Brook, NJ .
Gitlow , H. , Gitlow , S. , Oppenheim , A. , and Oppenheim , R. ( 1989 ), Tools and Methods for the
Improvement of Quality , Irwin , Boston .
Grant, E. L. , and Leavenworth, R. S. (1996), Statistical Quality Control , 7th ed., McGraw-Hill,
New York .
Ishikawa , K. ( 1982 ), Guide to Quality Control , 2nd ed., Asian Productivity Organization ,
Tokyo, Japan .
Juran , J. M. , and Godfrey , A. B. ( 1998 ), Juran ’ s Quality Handbook , 5th ed., McGraw - Hill , New
York .
Ledolter , J. , and Burrill , C. W. ( 1998 ), Statistical Quality Control: Strategies and Tools for
Continual Improvement , Wiley , New York .
Montgomery , D. C. ( 2004 ), Design and Analysis of Experiments , 6th ed., Wiley , New York.
Montgomery , D. C. ( 2001 ), Introduction to Statistical Quality Control , 4th ed., Wiley , New
York .
Ryan , T. P. ( 2000 ), Statistical Methods for Quality Improvement , 2nd ed., Wiley , New York.
Smith , G. M. ( 2003 ), Statistical Process Control and Quality Improvement , 5th ed., Prentice -
Hall , Upper Saddle Brook, NJ .
Summers , D. C. S. ( 2006 ), Quality , 4th ed., Prentice - Hall , Upper Saddle Brook, NJ .
Tague , N. R. ( 2005 ), Quality Toolbox , ASQ Quality Press , Milwaukee .
Thompson , J. R. , and Koronacki , J. ( 2001 ), Statistical Process Control: The Deming Paradigm
and Beyond , 2nd ed., Chapman & Hall , New York .
To improve the process capability, the process needs to be centered fi rst. This
usually involves adjusting the process settings. Cause - and - effect diagrams, Pareto
charts, and other tools discussed earlier in the chapter can be employed to fi nd
causes to the “ off - centering ” problems. After the process is brought back to its
nominal (10), the total nonconforming Pet Tabs produced will be dropped to
46,673 PPM. This is still far from the 2700 PPM for a “ just capable ” process ( C p = 1),
not to mention reaching the goal of 63 PPM at C p = 1.33.
To further improve the process capability, the variability needs to be reduced.
This can be achieved by designed experiments. Design of experiment (DOE) is a
systematic approach that allows engineers and managers to make intentional
changes in some process settings and assess the effects of those changes. An experiment
can be designed in this example by varying a few key process settings such as
drying time, mixing time, and temperature. Through a series of experimentations,
optimum settings are found for these process variables and the variability of the
process is reduced by 50%. With this reduction in process variability, the process is
now exhibiting a C p of 1.265 with 694 PPM. This example highlights the benefi ts of
process improvement. The move from an off - centered state to a centered state
resulted in a reduction of process fall - out by 13.1%. With designed experiments, the
process variability was cut in half and the process fall - out was signifi cantly reduced
by 98.5%.
BIBLIOGRAPHY 309
PROCESS ANALYTICAL
TECHNOLOGY ( PAT )
SECTION 4
CASE FOR PROCESS ANALYTICAL
TECHNOLOGY: REGULATORY AND
INDUSTRIAL PERSPECTIVES
Robert P. Cogdill
Duquesne University, Center for Pharmaceutical Technology, Pittsburgh, Pennsylvania
Contents
4.1.1 Introduction
4.1.2 Basis for Process Analytical Technology
4.1.2.1 Process Analytical Chemistry
4.1.2.2 Quality Management
4.1.2.3 Lean Manufacturing
4.1.3 Historical Factors Limiting Implementation of PAT
4.1.3.1 Real and Perceived Technological Barriers
4.1.3.2 Lack of Economic Incentive
4.1.3.3 Regulatory Disincentives
4.1.4 FDA Twenty - First - Century cGMPs Initiative
4.1.4.1 Conception of the Initiative
4.1.4.2 Risk - Based Orientation
4.1.4.3 Quality Systems Approach
4.1.4.4 Science - Based Policies
4.1.4.5 International Collaboration
4.1.5 PAT Evolution in Pharmaceutical Manufacturing
4.1.5.1 Process Understanding
4.1.5.2 PAT Principles and Tools
4.1.5.3 Strategy for Implementation
4.1.6 PAT Implementation Process
4.1.6.1 Preparation
4.1.6.2 Assessment
4.1.6.3 Analyze
4.1.6.4 Control
4.1.6.5 Release Philosophy
4.1.6.6 Optimization
4.1
313
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
314 REGULATORY AND INDUSTRIAL PERSPECTIVES
4.1.7 Perspectives on the Impact of PAT
References
4.1.1 INTRODUCTION
The implementation of process analytical technology (PAT) is occurring in what is
perhaps the most exciting period of change in pharmaceutical manufacturing of the
past three decades. A host of technological, regulatory, and market forces have
converged during the last fi ve years, yielding new opportunities for innovation in
the development and operation of pharmaceutical production processes. A major
driving force for change is the Food and Drug Administration (FDA) initiative to
implement a modern, risk - based framework for regulation and oversight of pharmaceutical
manufacturing [1] . The objectives of this section are to outline the historical
background of process analytics, to provide an overview of PAT in the pharmaceutical
industry and the business drivers for change, to summarize the FDA ’ s new initiative
and the PAT guidance [2] , and to present a basic plan for PAT implementation.
While the focus of this chapter is PAT, it should be kept in mind that PAT is an
important part of the much broader and risk - based paradigm introduced by the
twenty - fi rst - century current good manufacturing practices (cGMPs) initiative.
4.1.2 BASIS FOR PROCESS ANALYTICAL TECHNOLOGY
Despite the fact that the FDA ’ s PAT framework (and guidance) began to take form
just ahead of the creation of the twenty - fi rst - century cGMPs initiative in 2001, it is
well known that several of the core concepts were pioneered decades ago by other
manufacturing industries such as fi ne chemicals, semiconductors, petroleum, and
consumer products. The main concepts that differentiate PAT from the traditional
industrial pharmacy skill set (including pharmaceutical and materials science, chemistry,
and engineering) are process analytical chemistry (PAC) and advanced manufacturing
science (Figure 1 ).
For the purpose of this discussion, the term manufacturing science is meant to
describe the science and technology related to modern innovations in the design
and management of manufacturing processes. Since it is neither practical nor necessary
to cover all aspects of modern pharmaceutical manufacturing science in detail,
the following sections are intended to introduce two specifi c topics which are popular
in the current industrial vernacular but are not covered in detail in the pharmaceutical
literature: quality management systems and “ lean ” manufacturing.
4.1.2.1 Process Analytical Chemistry
Process analytical chemistry generally describes the science and technology associated
with displacement of laboratory - based measurements with sensors and instrumentation
positioned closer to the site of operation. Although industrial process
analyzers have been in use for more than 60 years [3] , the modern period of PAC
essentially began with the formation of the Center for Process Analytical Chemistry
(CPAC) in 1984 [4] . As described by Callis, Illman, and Kowalski [5] , the goal of
BASIS FOR PROCESS ANALYTICAL TECHNOLOGY 315
PAC is to “ supply quantitative and qualitative information about a chemical process ”
for monitoring, control, and optimization: they went on to defi ne fi ve “ eras ” of PAC:
(1) off line, (2) at line, (3) online (4) inline, and (5) noninvasive, which describe the
evolution of sensor technologies. In addition, they discussed the importance of issues
beyond chemical sensing, such as sampling, extraction of information from data
(chemometrics), integration with process controls, as well as the sociological aspects
of PAC deployment (e.g., gaining the trust of plant operators).
The industrial PAC movement has been bolstered by two decades of advances
in materials science, electronics, and chemometrics. Since the inception of CPAC,
the pace of innovation in sensors, instrumentation, and analytics has quickened
dramatically. The development of more robust, sensitive photodetector materials,
microelectromechanical systems (MEMSs), and fi ber optics and the perpetual
advancement of computing power (as predicted by Moore ’ s law) have both increased
the performance and reduced the cost of PAC. As a result, PAC is now a critical
part of routine operations within the realm of industrial chemistry. Many general
reviews on the subject of PAC (and PAT) have been published [6 – 10] . A series of
literature reviews on the subject of PAC have been published regularly in Analytical
Chemistry .
The fi rst review [11] listed manuscripts published between 1987 and 1992, covering
seven specifi c topics (general PAC, chromatography, optical spectroscopy, fi ber
optics, mass spectrometry, chemometrics, and fl ow injection analysis), along with a
section on needs for the future of PAC; in all, the fi rst review included 507 references.
Subsequent reviews were published in 1995 [12] , 1999 [13] , 2001 [14] , 2003
[15] , and 2005 [16] . The review series is an essential resource for scientists seeking
information on specifi c PAC methods; in total, 2650 references covering more than
16 topics were catalogued by the authors.
Currently there are three major consortia involving university, government,
and industrial partners — CPAC, the Measurement & Control Engineering Center
(MCEC), and the Control Theory and Applications Centre (CTAC) — along with an
annual conference, the International Forum on Process Analytical Chemistry
(IFPAC), and numerous online resources that are devoted to issues related to
process analytics [16] . In parallel with the FDA ’ s initiative, the term process
FIGURE 1 Multidisciplinary components of PAT in pharmaceutical manufacturing.
316 REGULATORY AND INDUSTRIAL PERSPECTIVES
analytical chemistry is gradually being replaced in the industrial vernacular by
process analytical technology . This refl ects the expansion of the fi eld as the importance
of physical characterization, risk analysis, and manufacturing science is
recognized.
4.1.2.2 Quality Management
Many of the quality improvement goals for implementation of PAT in the pharmaceutical
industry have been achieved by companies in other industries, such as
automobile production and consumer electronics, as a direct result of adopting principles
of quality management. The lineage of modern quality management can be
traced to the work of Walter Shewhart, a statistician for Bell Laboratories in the
mid - 1920s [17] . His observation that statistical analysis of the dimensions of industrial
products over time could be used to control the quality of production laid the
foundation for modern control charts. Shewhart is considered to be the father of
statistical process control (SPC); his work provides the fi rst evidence of the transition
from product quality (by inspection) to the concept of quality processes [18, 19] .
Shewhart ’ s methodologies were adopted and expanded by W. Edwards Deming
[20] and Joseph M. Juran [21] , who are credited with the birth of the “ total quality ”
(TQC) approach in Japan following World War II. Successors to the total quality
movement include management by objectives (MBO) (1960 ’ s), Crosby ’ s zero -
defects (ZD) movement (1970s), the American incarnation of total quality management
(TQM) (1970s – 1980s), quality circles (1970s), quality function deployment
(QFD) (1980s), the International Organization for Standardization (ISO) 9000
series (1987), and the Malcolm Baldridge National Quality Award (1987 – present).
The most recent major quality management methodology, Six Sigma (6 . ) [22, 23] ,
pioneered by Motorola, has become immensely popular because of the litany of
corporate CEOs (e.g., Thomas Galvin, Jack Welch) who have openly credited their
internal 6 . initiatives for dramatic improvements in bottom - line performance. All
of these quality movements [24] , however, as well as PAT, are related to the principles
of Shewhart, Deming, Juran, Crosby, Taguchi [25, 26] , and others, in that they
are based on systematic methods for understanding the sources of variability in
processes and minimizing their impact on product quality.
The so - called DMAIC (defi ne, measure, analyze, improve, and control) methodology
is a common framework used by improvement teams in many industries to
apply the concepts of quality management to systematically identify, prioritize,
and eliminate the root cause of quality problems. A variant of DMAIC, known as
DMADV (defi ne, measure, analyze, design, and verify), is sometimes used when a
process or operation requires complete redesign to bring about the desired quality
improvement and is a central concept of the DFSS (design for six sigma) movement.
The origins of DMAIC, DMADV, DFSS, and other various quality management
cycles can be traced to the “ Shewhart cycle ” of (1) plan, (2) do, (3) study, and (4)
act [24] .
Arguably, the most important aspects of quality management for PAT are the
concepts of quantitative process performance characterization using process capability
indices as universal descriptors, which form the basis of the “ measure ” and
“ analyze ” portions of the DMAIC model. Process capability indices consider simultaneously
both process variability and process specifi cations to determine whether
BASIS FOR PROCESS ANALYTICAL TECHNOLOGY 317
the process is “ capable ” [27] . A process is said to be capable if the quality measurements
for nearly all samples are within the specifi cation limits. A common version
of the process capability index, Cpk , is calculated according to
Cpk =
. . ...
...
min
USL
,
LSL
3
.
.
.
. 3
where . and . are the mean and standard deviation and USL and LSL are the upper
and lower specifi cation limits, respectively, for a product quality measurement.
Process capability indices are useful for process improvement studies because they
transform diverse measures of quality (e.g., weight, concentration, rate) into dimensionless
units, thereby allowing investigators to pinpoint major sources of variation
in a process (operations which have the lowest Cpk scores) when many measurement
systems and quality attributes are involved.
The process capability index, Cpk , is related to the so - called “ process sigma ” such
that a 6 . process corresponds to a Cpk of exactly 2.00, or 2.0 defective parts per
billion (PPB), assuming .N (0, . ) quality variance distribution (can alternative calculation
for process sigma estimates 3.4 defective parts per million for a 6 . process).
Examples of the correspondence between Cpk , process sigma, and defect rate for
.N (0, . ) distributions are shown in Figure 2 . The process capability (based on
observed yield) of pharmaceutical manufacturers has been cited by some benchmark
studies to be roughly 0.7 (2.1 . ) [28] .
While industrial benchmarks clearly indicate that pharmaceutical manufacturers
have many opportunities to improve quality control, direct comparison with other
industries may be somewhat misleading. As opposed to such industries as semiconductor
manufacturing, where defective parts are often readily apparent at some
FIGURE 2 Graphical illustration of the correspondence between the defects per million
opportunity (DPMO) process capability (C pk ) and process sigma (assuming normally distributed
quality variation).
318 REGULATORY AND INDUSTRIAL PERSPECTIVES
point in the value chain (i.e., the device built from the part will fail), drug products
suffer from a high degree of ambiguity in their quality specifi cations.
For example, fi nished - product release specifi cations such as content uniformity
are rarely correlated to clinical evidence; rather, they are set according to compendial
test standards. Furthermore, the functional relationship between in - process
material characteristics and fi nished - product quality is seldom known at a high level;
hence, the assigned in - process specifi cations for some operations may over - or
underestimate the true level of process capability. As the level of process understanding
in the pharmaceutical industry increases, development of science - and
evidence - based in - process and release specifi cations will improve the reliability of
C pk as a tool for process characterization.
For further information, the NIST/SEMATECH Handbook of Engineering Statistics
, which is freely available online [23] , and the American Society for Quality
( www.ASQ.org ), are excellent sources for background information and technical
details related to quality management.
4.1.2.3 Lean Manufacturing
In contrast to quality management systems, which have clear parallels with PAT (i.e.,
reduction of quality variation), the links between PAT and lean manufacturing are
less direct. In fact, while quality management systems are concerned with process
analysis of quality variation, lean fl ow path management is concerned with process
analysis of production time variation. Furthermore, the core concepts of lean manufacturing,
however, provide the technology platform which the pharmaceutical
industry will use to derive gains in production effi ciency from the adoption of PAT.
Without considering the impact of PAT on production effi ciency [i.e., the return on
investment (ROI) from implementing PAT], industry would have very little impetus
to voluntarily embrace PAT. The following paragraphs are intended to provide a
brief introduction to lean manufacturing; later portions of this chapter will discuss
the business drivers for implementation of PAT.
Lean manufacturing, or “ lean, ” is often misunderstood (not unlike TQM or 6 . );
for some people, lean business initiatives conjure “ slash - and - burn ” management
tactics to reduce workforce levels or shut down low - productivity operations. In fact,
lean manufacturing has been characterized as “ an amalgam of methodologies
including industrial engineering, just - in - time (JIT) (Osadas ’ s) 5 - S ’ s, TQC, continuous
quality improvement (CQI), Visual Control, Total Productive Maintenance
(TPM), Quality Circles, and Kaizen ” [24] .
The origins of lean manufacturing are often ascribed to the creation of the Toyota
Production System (TPS) by the Toyota Motor Corporation. However, the history
of lean manufacturing can be traced back to industrial developments which occurred
more than 150 years before TPS. The foundation for modern manufacturing was
laid by Eli Whitney in 1798; while Whitney is best known for his invention of the
cotton gin, it is his invention of interchangeable parts and uniform production which
revolutionized mass production ( www.EliWhitney.org ).
Nearly a century later, Frederick W. Taylor introduced the concepts of time study
and standardized work, coining the term scientifi c management . It was not until 1908,
with Henry Ford ’ s introduction of the Model T, that the value of lean manufacturing
was recognized worldwide. Henry Ford is considered by some to be the fi rst practi
BASIS FOR PROCESS ANALYTICAL TECHNOLOGY 319
tioner of JIT manufacturing; furthermore, his manufacturing system has been
described as the inspiration for TPS [29] . More recently Ford Motor Company has
developed a modernized version of Henry Ford ’ s original system, the Ford Production
System [24] , which borrows heavily from TPS.
As a discipline of manufacturing science, lean manufacturing is a technical
philosophy focused on the reduction of seven types of waste, or “ muda, ” in manufacturing:
overproduction, waiting, transport, inappropriate processing, unnecessary
inventory, excess motion, and defects. The transformation of a process to lean operation
is accomplished using many tools and strategies. Arguably, the most important
mechanism for change is to replace traditional “ make to forecast ” or “ push ” production
scheduling with “ pull ” strategies, such as “ kanban ” cards. The principles of lean
have been applied with success in manufacturing and service industries, as well as
governmental entities. Not unlike quality management, there are literally hundreds
of books describing the various tools and techniques used to apply lean methodologies.
The Society of Manufacturing Engineers ( www.SME.org ) maintains publications,
conferences, and a technical community devoted to production management
and is a good fi rst source for more information on lean manufacturing.
Compared with other industries, pharmaceutical manufacturers have been relatively
late to adopt lean manufacturing; consequently, pharmaceutical cycle times
are extremely long when compared with other industries [30, 31] . By comparing the
ratio of total cost of goods sold (COGS) to inventory value for the top 22 publicly
traded branded, generic, and biotech pharmaceutical companies to the reported
fi gures for other process industries, a rough indication can be gained of how much
less effectively pharmaceutical manufacturers manage their supply chains (Figure
3 ). Furthermore, it would not be diffi cult for most industrial pharmaceutical scientists
to fi nd common examples of each of the “ seven wastes ” in a typical pharmaceutical
manufacturing facility.
Admittedly, there are some constraints intrinsic to the industry which may ultimately
prevent pharmaceutical manufacturers from achieving “ world - class ” supply
chain and manufacturing performance. Furthermore, application of lean and quality
management tools to pharmaceutical manufacturing is proving to be a unique challenge.
A recent survey of 1500 pharmaceutical manufacturing professionals indicated
that, while more than half of the companies surveyed have implemented lean,
6 . , or operational excellence, less than half of those programs have yielded satisfactory
results [32] .
While these data seem to suggest that lean manufacturing is not suited to pharmaceutical
manufacturing, it is important to consider that most lean methodologies
(e.g., TPS) were developed for high - volume production of uniform products.
Although many “ blockbuster ” drugs are produced in dedicated facilities or in plants
specializing in only a few products, it is quite common for pharmaceutical manufacturers
to produce many products in a single plant, having a high proportion of
shared equipment. Traditional lean methods, such as kanban cards, are diffi cult to
manage in a complex, “ high - mix ” production environment. In order to solve these
limitations, innovative software algorithms for “ fl ow path management ” [33] have
been developed to simulate, design, and optimize pharmaceutical production processes
according to lean manufacturing principles.
Furthermore, the effectiveness of lean manufacturing is limited by variability in
the cycle time (C/T) for individual unit operations as well as by the fi nite risk of
320 REGULATORY AND INDUSTRIAL PERSPECTIVES
batch failure during production; this is true regardless of the complexity of the fl ow
path or of the degree to which equipment is shared. Pharmaceutical manufacturers
cope with such risks by building up long production queues to accumulate work in
process (WIP) ahead of unit operations. While this helps to improve capacity utilization
and overall equipment effectiveness (OEE), it decreases effi ciency by consuming
working capital and increasing the intensity of overhead operations (required
to fi nance, transport, and warehouse WIP). In order to gainfully implement lean,
pharmaceutical manufacturers must fi rst minimize C/T variation and risks to product
quality.
Finally, it is well known that a signifi cant portion of the typical production C/T
reported by industry is consumed by the delay between completion of a unit operation,
sampling, analysis, reporting, and in - process or fi nished - product release. In
some scenarios, PAT will enable manufacturers to release fi nished products to the
market immediately, with no delay for manual, offl ine testing; this is the so - called
real - time release (RTR) benefi t of PAT. Without PAT and RTR, the effectiveness
of lean strategies in reducing C/T will be limited by the maximum rate of product
inspection and release. Thus, it is critical for pharmaceutical manufacturers to deploy
PAT and lean in parallel if real gains in process performance are to be realized. The
lean – PAT concept is quite similar to lean – 6 . , or “ fusion management ” [24] .
FIGURE 3 Ratio of total COGS to reported inventory value. The ratio of COGS to inventories
is a rough indicator of supply chain velocity. A large ratio implies that inventories are
small relative to COGS and are turned over frequently. Toyota Motors, for example, which
is well known for effective supply chain management, has a much higher ratio of COGS to
inventories than General Motors.
Budweiser
Pepsi
Microsoft
Tyson
Toyota
Kelloggs
Kraft foods
Groupe Danone
DOW
FMC
Coca Cola
Unilever
BASF
Proctor & Gamble
HJ Heinz
ConAgra
Celgene
Genzyme
Amgen
Biogen IDEC
Genentech
General Motors
Gilead
Watson Labs
Barr Laboratories
TEVA
Alpharma
Mylan
Forest labs
Johnson & Johnson
Merck
Bristol Myers Squibb
Astra Zeneca
Novartis
Sanofi Aventis
GSK
Eli Lilly
Pfizer
Wyeth
2.00
0.00
4.00
6.00
(COGS/Inventory) ratio
8.00
10.00
12.00
14.00
4.1.3 HISTORICAL FACTORS LIMITING IMPLEMENTATION OF PAT
Despite the evidence of fi scal and competitive benefi ts enjoyed by the various
industries which have embraced process analytics, pharmaceutical companies have
been notoriously restrained in their efforts to deploy PAT. Indeed, the pharmaceutical
industry has slipped so far behind peer industries that a well - known Wall Street
Journal article from 2003 [34] characterized the manufacturing prowess of drug
makers as lagging “ far behind potato - chip and laundry - soap makers. ” While the
declaration was shocking to many, it was, nonetheless, an accurate assessment.
Before indicting the industry for gross negligence, however, it is important to consider
the various factors which have acted over time to create the current state of
affairs.
Over the years, dozens of excuses have been provided for the industry ’ s lack of
manufacturing innovation; many of the reasons are well known and have been
published elsewhere [35] . For the sake of simplicity, the factors limiting the adoption
of PAT can be distilled into three categories: real and perceived technological barriers,
lack of economic incentive, and regulatory disincentives.
4.1.3.1 Real and Perceived Technological Barriers
Despite the fact that near infrared spectroscopy (NIR) has been used industrially
for decades [36] , there has been hesitance to accept and trust “ new ” process analytical
measurement technologies as equivalent or superior to traditional methods. For
example, when a discrepancy between online NIR and laboratory analyses is
observed, it is rare that the destructive reference methods are ever targeted as the
source of error, despite the fact that NIR is often the more precise method. The
hesitance to trust more advanced, multivariate tools (which are perhaps less directly
understood) has certainly been a detriment to progress in deploying PAT.
Similar concerns persist with regard to chemometrics (multivariate data analysis),
information technology (IT), and advanced controls. One reason for such behavior
may be the practice of calibrating and validating PAT sensors by correlating their
signals to traditional, laboratory - based reference methods and characterizing performance
in terms of prediction error [37 – 39] . It is a truism of statistics that, no
matter how sensitive or accurate the PAT sensor may be in detecting quality variation,
the performance of the reference method will always limit the level of perceived
accuracy. A much more accurate depiction of the performance of PAT sensors
compared to reference techniques would be to compare analytical fi gures of merit,
such as signal - to - noise ratio (S/N) or analytical sensitivity, which explicitly account
for measurement precision [40, 41] .
Even though the perceptions of PAT instrumentation have begun to improve,
companies continue to worry that the intensity of product quality sampling afforded
by PAT sensors will result in negative consequences, such as increased inspection
and investigations. In other words, many companies continue to “ fear what they will
fi nd ” if they begin to analyze their operations more closely. Prior to the introduction
of rapid, nondestructive quality monitoring tools, there were few alternatives for
effi cient quality assurance except to rely on batch release criteria, such as the well -
known U.S. Pharmacopeia (USP) . 905 . procedure, which were based on extremely
limited sampling (i.e., assay 10 individual dosage units from a 30 - unit sample of a
production - scale batch).
HISTORICAL FACTORS LIMITING IMPLEMENTATION OF PAT 321
322 REGULATORY AND INDUSTRIAL PERSPECTIVES
Despite the fact that the operating characteristic (OC) curve of the USP . 905 .
test guarantees a signifi cant portion of each batch will have poor quality before
batch rejection is probable [42, 43] , companies have become comfortable with their
odds. Process analytical monitoring tools such as NIR spectroscopy, which are
capable of high - speed sampling in line, online, or at line, have been perceived as an
additional burden on the rate of successful batch release.
By forgoing real - time, pervasive quality monitoring, however, companies incur
signifi cant opportunity costs in at least three ways. First, without continuous monitoring
there are few feasible opportunities for implementing RTR; time delays
related to offl ine release testing are one of the most signifi cant factors limiting
supply chain velocity in pharmaceutical manufacturing. Second, while there is some
potential for “ discovering ” a greater number of batches which do not meet release
criteria, statistical simulations suggest that potentially fewer batches will be rejected
when larger sample sizes are considered. In other words, when the impact of measurement
imprecision and the true distribution of quality characteristics are considered,
traditional release testing methods pose fi nite risks of failing passable batches
(which otherwise should have passed) because the limited sample does not adequately
represent the characteristics of the population (Figure 4 ). Finally, and
perhaps most importantly, traditional sampling techniques are an effective barrier
to continuous improvement; based on fundamentals of statistical theory, it can be
shown that samples of at least hundreds of individuals are required to detect incremental
changes in process capability (Figure 5 ). Hence, even if a company were to
investigate potential process improvements, only process capability changes of
improbable magnitude would be recognized with statistical confi dence.
FIGURE 4 Comparison of operational characteristic (OC) curves for the USP . 905 . ( a )
and PAT - based ( b ) release strategies generated by Monte Carlo simulation. The USP OC
curve ( a ) is based on the assumption of 2% RSD measurement precision; the PAT OC curve
( b ) assumes NIR measurement of 800 tablets with 0.9% measurement precision; both curves
were estimated using the same simulated populations of one million tablets having varying
levels of quality uniformity. Each curve consists of four regions: the regions above and below
the sigmoid curve correspond to proportions of batches accurately passed or rejected based
on the release criteria. Along the sigmoid curve are regions related to the rates of false batch
failure (lower side of curve) and false batch acceptance (upper side of curve). The jagged
nature of the curves is related to the limitations imposed by fi nite iterations. The slope of the
curves demonstrates the superior specifi city (or “ tunability ” ) of release tests optimized for
PAT systems.
100
90
80
70
60
50
40
30
20
10
0
Batches (%)
98 96 94 92 90 88
Within-batch true coverage (%)
False
fail lots
False pass
lots
(a) (b)
98 96 94 92 90 88
Within-batch true coverage (%)
4.1.3.2 Lack of Economic Incentive
A common refrain within the industry has been that there simply is not suffi cient
fi nancial return from investment in process analytics or manufacturing technology
upgrades to justify spending. In some respects, this is a valid argument. Historically,
many of the industries which have justifi ed signifi cant investment in process analytics
utilized continuous manufacturing; it is far more diffi cult to effi ciently control continuous
processes (relative batch production systems) without real - time process analytics
[35] . Hence, while the pharmaceutical industry has been able to choose, many
other manufacturers have been forced to integrate PAT into their operations.
Since pharmaceutical investment in PAT continues to be an option rather than
a priority for most companies, arguments justifying PAT spending are forced to
compete with other spending initiatives for capital. During each planning cycle,
company managers must decide whether to allocate additional capital toward
diverse opportunities, such as greater research and development (R & D), improvements
in manufacturing capabilities, or additional forces in sales and marketing [i.e.,
selling, general and administrative (SG & A)]. For any particular project to be funded,
expected returns must not only exceed the company ’ s cost of capital [i.e., weighted
average cost of capital (WACC)], winning projects may be required to exceed the
company ’ s expected return on invested capital (ROIC) or at least provide expected
returns in excess of other investment alternatives. A recent academic case study of
the potential fi nancial returns on investment (ROI) in PAT and lean manufacturing
in the pharmaceutical industry show, however, that many pharmaceutical manufacturers
could ultimately benefi t tremendously by improving manufacturing performance
[44] .
FIGURE 5 Relationship between sampling rate and effective resolution of process capability
assessment. The curve is based on the width of the confi dence intervals for estimation of
mean and variance. The relationship shown does not consider the effect of reference measurement
precision, which would further reduce the ability to discern changes in process
capability.
Detectable change in process capability (95% confidence)
as a function of sampling rate
Detectable change in Cpk (%)
Samples assayed (N)
10 100 200 300 400 500 600 700 800 900 1000
1
2
3
4 5
10
20
30
50
100
USP <905>, 30 Samples
HISTORICAL FACTORS LIMITING IMPLEMENTATION OF PAT 323
324 REGULATORY AND INDUSTRIAL PERSPECTIVES
Unfortunately, proponents of PAT are only just beginning to develop the methods
to quantify all of the potential opportunities for ROI. Furthermore, it is important
to consider the relative level of risk posed by investment in PAT (as opposed to
other alternatives). Unlike investments in sales or marketing, there remains
considerable uncertainty in the industry regarding the likelihood of achieving ROI
projections or the prospect of PAT investment creating new problems. For these
reasons, management teams have typically found it easier to justify spending in
R & D and marketing instead of PAT or manufacturing reforms.
Besides concerns over the likelihood and magnitude of returns on PAT investments,
it is often cited that manufacturing and optimizing the cost of production
have simply not been a priority in the industry; manufacturing has often been
viewed as a cost rather than a value - generating component. The distribution of
corporate expenditures has been provided as evidence in support of this theory
(Figure 6 ). Based on corporate annual income statements from 2005, the average
expenditure on R & D and SG & A among the top - 10 branded pharmaceutical companies
(by market capitalization, November 7, 2006) was nearly double their reported
cost of goods sold. Another take on this theory is that institutional and individual
investors (who own the pharmaceutical companies and supply the capital for their
operation) and the boards of directors elected by them look favorably on the expansion
of R & D and marketing investment while taking a more myopic view on the
importance of manufacturing. It has sometimes been said that Wall Street rewards
FIGURE 6 Distribution of the components of revenue (FY2005 annual data) for branded
( a ), generic ( b ), and biotech ( c ) drug manufacturers. Companies are arranged according to
market capitalization (as of November 2006).
0
10
20
30
40
50
60
70
80
90
100
Net profit
Tax, interest, other
Research & development
Net profit
Tax, interest, other
Research & development
Cost of goods sold
Cost of goods sold
Selling, general & administrative
Selling, general & administrative
JNJ
AMGN DNA GILD CELG GENZ BIIB
PFE GSK NVS SNY
Company (ticker symbol)
Company (ticker symbol)
MRK AZN WYE LLY BMK
(a)
% of revenues
0
10
20
30
40
50
60
70
80
90
100
Net profit
Tax, interest, other
Net profit
15%
Other Exp.
12%
R&D
15%
Branded
Pharma
COGS
25%
SG&A
33%
Net profit
16%
Other Exp.
10%
R&D
9%
Generics
COGS
40%
SG&A
25%
Net profit
19%
Other Exp.
14%
R&D
23%
Biotech
COGS
16%
SG&A
28%
R&d
Cost of goods sold
Selling, general & administrative
TEVA FRX BRL
Company (ticker symbol)
MYL WPI ALO
(b)
% of revenues
0
10
20
30
40
50
60
70
80
90
100
(c)
% of revenues
(pharmaceutical companies) for innovation in discovery and replication in manufacturing
[45] . It is not completely coincidence, for example, that Merck ’ s appointment
of its president of manufacturing, Richard T. Clark, to chief executive in May
2005, which, according to fi nancial journalists, “ disappointed investors ” who apparently
would have preferred someone with a “ research and development background
” [46] , marked the beginning of a nearly 25% loss in market capitalization
over the next six months.
While the various reasons discussed for the pharmaceutical industry ’ s tepid
approach to PAT and manufacturing reform are plausible, they are likely secondary
to the real and perceived risks posed by the regulatory uncertainty surrounding
innovation in manufacturing. For example, it is well known that many companies
were beginning to use PAT tools long before the FDA ’ s initiative, which suggests
that the economic benefi ts of process analytics have been recognized internally for
some time. In response to the fear that their use of new technologies would spur
additional investigations by the FDA, however, some of these companies operated
in a “ Don ’ t use, or don ’ t tell ” manner with regard to PAT [45] .
4.1.3.3 Regulatory Disincentives
The real and perceived fear of regulatory noncompliance has arguably been one of
the most important factors explaining the industry ’ s reluctance to pursue manufacturing
innovation [1, 2] . While the fi rst 25 years of pharmaceutical GMP have been
effective in ensuring the safety of prescription drug products for consumers, it has
been achieved at the expense of innovation and fl exibility. Without the ability to
adjust processes to account for changes in materials, operating conditions, or the
level of process understanding, process analytics are of nearly no value since there
is no capacity to act on new information (besides material/batch rejection).
Furthermore, companies who dared to make changes or implement new technologies,
whether conventional process improvements, new unit operations, or
process analytics, were met with extensive supplemental documentation, FDA
inspection, and the fi nite risk of production delays. Ultimately, the potential for
regulatory action stifl ed the industry ’ s desire to pursue technologies which might
have seemed extraordinary, such as real - time analytics or chemometrics. Finally,
without the benefi ts conferred by the PAT guidance and risk - based cGMPs initiative,
industry rarely had incentive to formally analyze the risk of established processes
out of fear that what they might discover would be used against them in
regulatory or legal actions.
4.1.4 FDA TWENTY - FIRST - CENTURY c GMP s INITIATIVE
The observation that the state of cGMP at the beginning of the twenty - fi rst - century
was stifl ing innovation in pharmaceutical manufacturing did not go unnoticed by
the FDA, which also saw opportunity in remodeling the regulatory framework.
Since many changes, even minor operational modifi cations, required prior approval
from the agency prior to implementation, regulators were swamped with thousands
of supplements every year. Resources were stretched between processing of supplements,
review and approval of new facilities, processes and documentation, and
inspection; all the while, the FDA was being squeezed by external constraints on
FDA TWENTY-FIRST-CENTURY cGMPs INITIATIVE 325
326 REGULATORY AND INDUSTRIAL PERSPECTIVES
budget growth (Figure 7 ). As of 2001, FDA regulators were so burdened that they
were unable to meet statutory biennial GMP inspections. Finally, the load of supplements,
reviews, and inspections were acting as a signifi cant drag on the advancement
to market of new pharmaceutical therapies.
4.1.4.1 Conception of the Initiative
The agency began a public dialogue on the state of pharmaceutical manufacturing
and FDA regulation during discussions with the Advisory Committee for Pharmaceutical
Science (ACPS) in July 2001, followed by further discussion within the FDA
Science Board meetings in November 2001 and April 2002 [47] . A signifi cant focus
of the discussions was the impact of the regulatory framework on innovation, quality,
and effi ciency as well as opportunities for change. A new, risk - based paradigm which
rewards innovative producers through opportunities for “ regulatory relief ” began to
take shape, displacing the notion of regulatory compliance as a force for innovation.
The new paradigm offered advantages to the FDA, as well, in that the level of inspection
resources could be prioritized and allocated according to risk, thereby easing
FIGURE 7 Trends in FDA workload and staffi ng resources. ( Adapted from L. X. Yu, Implementation
of quality - by - design: Question - based review, Drug Information Association (DIA)
42nd Annual Meeting, Philadelphia, PA, 2006 .)
1000
800
600
400
200
0
2001 2002 2003 2004 2005
ANDAs
Employees
4000
3500
3000
2500
2000
2001 2002 2003 2004
Supplements
the strain on FDA resources. These changes signaled an evolution of what seemed
to be an adversarial FDA – industry relationship toward greater cooperation.
While the pharmaceutical incarnation of the term PAT was formally introduced
during these meetings [48] , a signifi cant portion of the concepts which defi ne the
core of PAT in pharmaceutical science were presented by industrial and academic
scientists, many of whom had been building support for and working on these issues
within their organizations for years. Industrial and academic presentations included
topics such as total quality management [49] , new technologies for pharmaceutical
manufacturing [50] , and QbD [51] , among others.
In August 2002, the agency announced the Pharmaceutical cGMPs for the 21st
Century initiative (or “ the initiative ” ), which began a two - year effort undertaken
by a number of multidisciplinary working groups within the FDA, as well as the
cGMP steering committee, to assess the current regulatory structure and defi ne the
agency ’ s new vision for risk - based regulation of manufacturing and product quality.
The new initiative, which was intended to modernize the FDA ’ s regulation of
pharmaceutical quality for human, veterinary, and select human biological products,
sought to reform the pharmaceutical as well as the chemistry, manufacturing, and
controls (CMC) programs, with the following specifi c objectives:
• Encourage the early adoption of new technological advances by the pharmaceutical
industry.
• Facilitate industry application of modern quality management techniques,
including implementation of quality systems approaches, to all aspects of pharmaceutical
production and quality assurance.
• Encourage implementation of risk - based approaches that focus both industry
and agency attention on critical areas.
• Ensure that regulatory review, compliance, and inspection policies are based
on state - of - the - art pharmaceutical science.
• Enhance the consistency and coordination of the FDA ’ s drug quality regulatory
programs, in part, by further integrating enhanced quality systems approaches
into the agency ’ s business processes and regulatory policies concerning review
and inspection activities.
The result of the working groups ’ assessment enabled the development of the
new framework embodied by the fi nalized twenty - fi rst - century cGMPs as well as
the associated components, such as the PAT guidance. Throughout the assessment
and development, and continuing during the “ implementation phase ” of the initiative,
the following set of guiding principles has been maintained:
• Risk - based orientation
• Science - based policies and standards
• Integrated quality systems orientation
• International cooperation
• Strong public health protection
The fi nal report on the results and future plans for the initiative were released in
September 2004. The report effi ciently describes the motives, origins, development
FDA TWENTY-FIRST-CENTURY cGMPs INITIATIVE 327
328 REGULATORY AND INDUSTRIAL PERSPECTIVES
process, and mechanisms for implementing and evaluating the initiative and can be
found posted on the Center for Drug Evaluation and Research (CDER) Offi ce of
Pharmaceutical Science (OPS) website ( http://www.fda.gov/cder/OPS/ ).
Since it would be impractical to accurately describe all of the important aspects
of the report within this space, the following sections are intended to detail some
of the concepts and guiding principles of the initiative which are particularly important
for understanding PAT. The organization of this summary is intended to effi -
ciently describe selected concepts of the agency ’ s twenty - fi rst - century cGMPs and
is not intended to mirror the structure or totality of the associated FDA documentation.
All who are actively engaged in pharmaceutical manufacturing or are interested
in PAT are encouraged to read the fi nal report [1] , which should be considered
a primary source for direction.
4.1.4.2 Risk - Based Orientation
The FDA ’ s adoption of a risk - based orientation for regulation is the most important
aspect of the twenty - fi rst - century cGMPs. It is a common misconception that the
agency ’ s initiative describes a new set of practices for the industry. In fact, while the
FDA is committed to encouraging innovation in the industry, the twenty - fi rst -
century cGMPs initiative is entirely focused on changing the agency ’ s regulatory
framework so that quality and innovation are rewarded with reduced oversight.
Now that the agency has entered the implementation phase of the initiative, many
of the previous regulatory disincentives have been eliminated. In other words, pharmaceutical
companies are currently free to voluntarily choose whether or not to
pursue innovative changes in their development, operation, and quality assurance
of manufacturing processes such as PAT.
Risk - Based Prioritization of c GMP Inspections The mechanism by which the
FDA will encourage the industry to join in implementing the new methods is provided
by the risk - based algorithm for prioritizing cGMP inspections. Incidentally,
risk - based site selection is the same mechanism which will allow the agency to
optimally allocate its limited oversight resources to achieve the greatest public
health impact. Operational effi ciency is a major component of the FDA ’ s plans for
the future. The key to the risk - based site selection program is the agency ’ s risk -
ranking model, which has been deployed as a pilot program since the beginning of
its 2005 fi scal year.
The model is based on a hierarchical risk - ranking and risk - fi ltering method
whereby a site risk potential (SRP) is estimated as a function of the weighted
potentials for each of three top - level components of site risk — product, facility, and
process (Figure 8 ). The risk potential for each of the three top - level components
is calculated as a function of selected risk factors which are relevant to the component
(specifi c to the site). A set of subcategories are defi ned for each top - level
component; each subcategory is comprised of individual risk factors. The initial
model weights (the actual risk scores at the lowest level) were optimized using
a combination of empirical evidence and expert judgment. Examples of potential
risk factors for each top - level component (and associated subcategories) were provided
in a report which describes the fi rst iteration of pilot risk - ranking model in
detail [52] .
The results from the fi rst iteration of the risk - ranking model demonstrated the
capability of the model to spread SRP scores for the purpose of fi ltering. Future
iterations of the risk - ranking model will be generated by correlating predicted site
risk potentials with data gathered by traditional oversight activities (e.g., cGMP
compliance inspections) and adjusting the risk factor weights to maximize the effectiveness
of SRP prediction (similar to multivariate linear regression). The selection
of risk factors included in the fi rst iteration of the model was based on the availability
of data. Some proposals for future iterations of the model include incorporating
factors such as systems for continuous assessment of process capability as
indicators of the site ’ s level of process understanding and control. Certainly, as the
model is updated to capture the benefi ts of new best practices in manufacturing,
such as PAT, the risk ranking will begin to provide effective incentive for producers
to pursue innovation.
4.1.4.3 Quality Systems Approach
According to the FDA staff manual guide [53] , a quality system is a “ set of formal
and informal business practices and processes that focus on customer needs,
leadership vision, employee involvement, continual improvement, informed decision
making based on real - time data and mutually benefi cial relationships with
external business partners to achieve organizational outcomes. ” Based on this
description, PAT should be considered to be an important tool for supporting a
quality management system. As stated earlier, one of the FDA ’ s objectives in undertaking
the initiative was to integrate quality systems and risk management approaches
into its existing programs with the goal of encouraging industry to adopt modern
and innovative manufacturing technologies, including industrial deployment of
quality management systems such as those described earlier in this chapter (e.g.,
ISO 9000).
In September 2006, the FDA released “ Guidance for Industry: Quality Systems
Approach to Pharmaceutical cGMP Regulations ” [54] . The guidance is intended to
“ help manufacturers implementing modern quality systems and risk management
approaches to meet the requirements of the Agency ’ s cGMP regulations, ” in particular,
Parts 210 and 211. In developing the guidance, the Quality System Guidance
FIGURE 8 Schematic of FDA ’ s pilot risk - ranking model for calculation of site risk
potential.
CD1 CD2 CP1 CP2 CF1 CF2
Top-level
components
Categories of
risk factors
Risk factors
(quantitative or
qualitative
variables)
Site risk potential
Product Process Facility
FDA TWENTY-FIRST-CENTURY cGMPs INITIATIVE 329
330 REGULATORY AND INDUSTRIAL PERSPECTIVES
Development (QS) working group “ mapped ” the relationship between cGMP
regulations and various quality system models both internal and external to the
FDA. Their result is a comprehensive model which allows producers seeking to
implement their own quality management systems to quickly identify those aspects
of quality systems which are, and are not, correlated with cGMP.
The QS guidance begins by defi ning critical concepts of modern quality systems,
including quality, QbD and product development, quality risk management, corrective
and preventative action (CAPA), change control, the “ quality unit, ” and the
six - system inspection model. The discussion of the quality unit describes its relationship
with the concepts of quality control (QC) and quality assurance (QA) and the
relationship between the quality unit and the other units within the pharmaceutical
manufacturing organization. The six - system inspection model is described as a
blueprint for how compliance inspections will be organized under the new quality
systems approach and should be considered a template for internal verifi cation of
compliance within pharmaceutical organizations adopting quality management
systems (Figure 9 ).
The majority of the QS guidance is devoted to describing the essential components
of modern quality systems, including four major factors which must be
addressed: management responsibilities, resources, manufacturing operations, and
evaluation activities. Each factor is described in detail, including aspects which
overlap with cGMP regulations (for each factor there is a table listing the related
regulatory citations). In particular, the manufacturing section describes aspects of
quality systems (and related cGMPs) which are closely related to PAT, including
raw materials analysis, operations monitoring, and procedures for addressing nonconformities.
Finally, the guidance includes many important references and related
guidance documents which should be considered by companies seeking to implement
a quality management system.
FIGURE 9 FDA ’ s six - system inspection model.
4.1.4.4 Science - Based Policies
Continuous improvement, which the agency describes as an “ essential element in a
modern quality system, ” is aimed toward improving effi ciency by “ optimizing a
process and eliminating wasted efforts in production ” [1] . One of the unintended
consequences of the regulatory system (prior to the new initiative) had been the
suppression of nearly all opportunities for continuous improvement in manufacturing
once a pharmaceutical product has been approved for market. Changes to formulations
and processes needed to be justifi ed regarding their impact on product
quality, often requiring time - consuming postapproval supplements. Producers in
most other modern industries (many of whom deal with public safety risks on par
with or exceeding those managed by the pharmaceutical industry) make it a practice
to continuously fi ne tune and adjust their operations to maximize quality and effi -
ciency. Pharmaceutical manufacturers, on the other hand, have largely been constrained
to treat demonstrated processes as if they were set in stone.
While there is some logic to limiting the scope and pace at which changes can be
made to processes, there is obvious fallacy in the idea that the fi rst approved con-
fi guration for a drug manufacturing operation will be optimal, especially considering
the enormous fi nancial and ethical pressures on process development teams to
quickly bring new drug therapies to market. This realization spurred the agency to
begin the process of developing science - based policies and standards to facilitate
innovation , which currently includes three new updated guidance documents:
“ Sterile Drug Products Produced by Aseptic Processing — cGMP ” [55] , the PAT
guidance, and the draft guidance on comparability protocols. Each guidance document
encourages voluntary adoption of new technologies in pharmaceutical manufacturing
by defi ning modern, science - based regulatory mechanisms which enable
producers to implement strategic improvements with opportunities for more effi -
cient regulatory compliance.
Comparability Protocols In fact, pharmaceutical manufacturers have always had
the option to explore changes to their production processes. The difference between
the old regulatory paradigm and the twenty - fi rst - century cGMPs initiative is that
producers who seek to improve the quality and effi ciency of their processes will be
able to implement changes much more quickly while spending signifi cantly fewer
resources to maintain compliance. The key to achieving these benefi ts is demonstrating
that there is suffi cient understanding of the process and changes to be made and
that implementation of the improvements poses very little risk to consumers.
A new mechanism for implementing process changes, which refl ects the inclination
for science - based policies, is detailed in the FDA ’ s draft guidance “ Comparability
Protocols — Chemistry, Manufacturing, and Controls (CMC) Information ” . A comparability
protocol (CP) is a “well-defi ned, detailed, written plan for assessing the
effect of specifi c CMC changes in the identity, strength, quality, purity, and potency
of a specifi c drug product as these factors relate to the safety and effectiveness of
the product ” [56] . Submission of a CP by a producer is optional and may be used to
facilitate changes in a manufacturing process, analytical procedures, manufacturing
equipment or facilities, or container closure systems or for implementation of PAT.
The benefi t for producers submitting a CP is that, upon approval of a CP, “ the
FDA can designate, where appropriate, a reduced reporting category for future
FDA TWENTY-FIRST-CENTURY cGMPs INITIATIVE 331
332 REGULATORY AND INDUSTRIAL PERSPECTIVES
reporting of CMC changes covered by the approved CP ” . For example, changes that
otherwise would require submission, review, and acceptance of a postapproval supplement
(PAS) might be designated as annual report (AR) changes if they were provided
for in an approved CP. The CP is one of the mechanisms by which the FDA intends
to reduce the number of supplements requiring review. Additionally, the CP was
designed to facilitate free fl ow of communication with the agency, thereby reducing
the risk that process changes will lead to unexpected regulatory shutdown or delay.
Process Validation In agreement with the pursuit of science - based policies, the
FDA has begun to revise the 1987 “ Guideline of General Principles of Process Validation
” and in March 2004 released a revision of the compliance policy guide (CPG)
(Section 490.100) “ Process Validation Requirements for Drug Products and Active
Pharmaceutical Ingredients Subject to Pre - Market Approval ” [52] . The current revisions
are designed to support continuous improvement and replace the notion of
“ three - batch ” validation. The CPG describes the concept that, after having identi-
fi ed and established control of all critical sources of variability, conformance batches
are prepared to demonstrate that under normal conditions and operating parameters
the process results in the production of acceptable product. However, the CPG
does not describe how many conformance batches are required; rather, the manufacturer
is expected to provide “ sound rationale ” for the procedure they choose to
follow in demonstrating validation.
The ambiguity in the revised (CPG) regulations may seem to signify that manufacturers
would need to undertake even more extensive validation exercises when
in fact the CPG contains language providing a pathway for batch release to market
distribution concurrent with the manufacture of initial conformance batches or with
a single conformance batch [57] :
Advanced pharmaceutical science and engineering principles and manufacturing
control technologies can provide a high level of process understanding and control
capability. Use of these advanced principles and control technologies can provide a
high assurance of quality by continuously monitoring, evaluating, and adjusting every
batch using validated in - process measurements, tests, controls, and process endpoints.
For manufacturing processes developed and controlled in such a manner, it may not
be necessary for a fi rm to manufacture multiple conformance batches prior to initial
distribution.
Interpretation of the CPG suggests that implementation of PAT can be an important
consideration for streamlining process validation. Finally, a quotable interpretation
of the new science - based paradigm suggests that (instead of validating the process)
producers should “ control the process, and validate the controls. ” Beyond revision
of the CPG, FDA is expected in the near future to release draft guidance on process
validation, which will be closely aligned with concepts associated with PAT, QbD,
and the rest of the 21 st century cGMPs.
4.1.4.5 International Collaboration
Recognizing the current realities of the global marketplace, the FDA has made
coordination with international regulatory partners a priority of the twenty - fi rst -
century cGMPs initiative. By increasing its collaboration with international health
and regulatory partners, the FDA has been able to leverage its resources through
increased sharing of information and harmonization of activities. The International
Conference on Harmonization of the Technical Requirements for Registration of
Pharmaceuticals for Human Use (ICH) ( www.ich.org ) has been the dominant mechanism
for international cooperation among pharmaceutical regulatory authorities
in Europe, Japan, and the United States.
A consensus vision statement was drafted at the July 2003 ICH meeting with
regard to the objective of the ICH in harmonizing the efforts of regulatory bodies
to establish quality systems approaches in their operations: “ Develop a harmonized
pharmaceutical quality system applicable across the life cycle of the product emphasizing
an integrated approach to quality risk management and science. ”
Three consensus guidelines defi ne the core of the ICH ’ s involvement in harmonization
of pharmaceutical quality systems — Q8: Pharmaceutical Development, Q9:
Quality Risk Management, and Q10: Pharmaceutical Quality Systems (in addition,
each of the guidance documents cites critical areas of overlap with Q6A: Specifi cations:
Test Procedures and Acceptance Criteria for New Drug Substances and New
Drug Products: Chemical Substances).
Q 8: Pharmaceutical Development According to the ICH Q8 guideline [58] ,
the aim of pharmaceutical development is to “ design a quality product and the
manufacturing process to deliver the product in a reproducible manner. ” While
QbD is not specifi cally mentioned in the guideline, the intent of the ICH Q8 expert
working group (EWG) was to describe a system that would provide incentive for
manufacturers to incorporate aspects of QbD and continuous improvement throughout
the product life cycle. In achieving this goal, the guideline they produced
describes the suggested contents for Section 3.2.P.2 of a regulatory submission in
the ICH M4 common technical document (CTD) [59] and the FDA electronic
common technical document (eCTD) [60] .
The pharmaceutical development and quality overall summary (QOS) sections
of the CTD (Figure 10 ) provide pharmaceutical scientists with dedicated channels
to present regulators with the relevant knowledge and process understanding gathered
during the development of a new product (which can be updated to support
new knowledge gained over the life cycle of the product following approval). The
knowledge communicated within these sections are important considerations for
justifi cation of a lower site risk potential (i.e., SRP, with regard to risk - based inspection)
and for facilitation of effi cient, question - based review (QbR) [61] . Question -
based review is another mechanism by which the agency intends to streamline the
regulatory process as well as reward producers for adopting best practices in quality
management.
In addition to facilitating risk - based oversight, the content of the pharmaceutical
development and QOS sections of the CTD are critical to enabling continuous
improvement and fl exible operation. The information and knowledge communicated
within these sections provide scientifi c understanding to support the establishment
of a manufacturing design space, in - process and release specifi cations, and
manufacturing controls.
As described within the Q8 guideline, a design space is the “ multidimensional
combination and interaction of input variables and process parameters that have
been demonstrated to provide assurance of quality. ” So long as process control is
maintained within the bounds of the design space, operating parameters can be
adjusted to improve product quality or manufacturing effi ciency. Based on the
FDA TWENTY-FIRST-CENTURY cGMPs INITIATIVE 333
334 REGULATORY AND INDUSTRIAL PERSPECTIVES
current defi nition, operation outside of the established design space would initiate
a regulatory postapproval change process. Thus, complete and accurate communication
of the knowledge supporting a company ’ s design space is vital for a company
to maximize productivity while maintaining regulatory compliance. Furthermore,
with the new communication pathways in place, companies have incentive to pursue
manufacturing studies beyond marketing approval to expand their design space or
to update specifi cations and controls. In addition to the product under review, if
appropriate, experiences gained from the development (and manufacture) of similar
drug products may be included.
Q 9: Quality Risk Management The second working group (ICH Q9 EWG) is
trying to better defi ne the principles by which risk management will be integrated
into decisions by regulators and industry regarding quality, including cGMP compliance.
In November 2005, the Q9 EWG released the “ Step 4 ” version of the Q9
guideline which defi nes the two primary principles of quality risk management,
provides a model for the quality risk management process (Figure 11 ), and describes
the terminology and tools for risk assessment and management. In addition, the
document includes a concise reference list for more detailed information on risk
management methods, such as failure mode effect and criticality analysis (FMECA),
which are important tools for prioritized implementation of PAT. While it is not
intended to be a “ how to ” manual for risk management, the Q9 guideline is a valu-
FIGURE 10 Schematic illustration of the ICH M4 common technical document (CTD);
the contents of the Quality Overall Summary (2.3) and Quality (3) modules are most relative
to PAT.
Module 2
Module 1
Regional
Administrative
Information
1
1.1 Submission
Module 3 Module 4 Module 5
Not part of the CTD
CTD
CTD Table of Contents
2.1
CTD Introduction
2.2
Quality
Oveall
Summary
2.3
Quality
3
Nonclinical
Overview
2.4
Nonclinical
Study Reports
4
Clinical
Overview
2.5
Nonclinical Written
and Tabulated
Summaries
2.6
Clinical
Summary
2.7
Clinical
Study Reports
5
able information source for companies seeking to incorporate quality risk management
into their operations [62] .
Q 10: Pharmaceutical Quality Systems While the Step 2 document for the third
tripartite guideline, Q10: Pharmaceutical Quality Systems, has not yet been released,
the fi nal concept paper has been available since 2005 [63] . Similar to the manner by
which the FDA ’ s quality systems approach guidance mapped the relationship
between cGMPs and other industrial quality management systems, the Q10 guideline
is anticipated to serve as a bridge between the approaches to quality systems
taken by the different regional regulations, thereby helping to achieve global harmonization
of quality systems. The guideline is expected to strengthen and complement
issues covered in Q6A, Q8, and Q9 and will provide a foundation for a
pharmaceutical quality system based on elements from the ISO 9001 and 9004
standards. The guideline is also expected to develop harmonized defi nitions for
issues critical to PAT, including continuous improvement activities, data - gathering
methods, and the approach to measurement system validation.
4.1.5 PAT EVOLUTION IN PHARMACEUTICAL MANUFACTURING
Though it may be tempting to characterize PAT as a revolutionary change in pharmaceutical
manufacturing, history will likely show that the beginning of the twenty -
fi rst - century cGMPs initiative and the development of the PAT guidance mark the
FIGURE 11 Schematic of quality risk management process described within ICH Q9.
Initiate
Quality Risk Management Process
Risk Assessment
Risk Identification
Risk Analysis
Risk Acceptance
Risk Evaluation
Risk Control
Risk Reduction
Output / Result of the
Quality Risk Management Process
Risk Review
Review Events
unacceptable
Risk Management tools
Risk Communication
PAT EVOLUTION IN PHARMACEUTICAL MANUFACTURING 335
336 REGULATORY AND INDUSTRIAL PERSPECTIVES
beginning of a period of rapid evolution in pharmaceutical manufacturing which
will extend far into the future. Even though the twenty - fi rst - century cGMPs initiative
is more extensive (with regard to changing the relationship between the FDA
and the pharmaceutical industry), interest in the PAT guidance and the opportunities
it presents for the industry were initially much greater. More recently, perhaps
in parallel with some changes in leadership in the agency, there has been a palpable
shift of emphasis toward QbD, which was barely mentioned in many of the twenty -
fi rst - century cGMPs documents. It is important to keep in mind that, just as most
industries have seen a parade of “ new ” quality systems initiatives over the years
since Shewhart ’ s fi rst methods were published, the principles upon which PAT and
QbD are built, such as robust process design, quality monitoring, and effective controls,
will persist regardless of the name of the initiative. Furthermore, as with PAT,
QbD is not a new concept. Indeed, Dr. Genichi Taguchi, who has been credited by
some as the father of QbD, began applying QbD in pharmaceutical manufacturing
while working as a statistical consultant for Morinaga Pharmaceuticals Company of
Japan from 1947 – 1949 [25] .
The PAT guidance is unique when compared with typical FDA guidance documents
in that it is not instructive or limiting per se; rather, the guidance describes
the principles and tools upon which the PAT framework is built, with the goal of
“ highlighting opportunities and developing regulatory processes that encourage
innovation. ” The FDA ’ s goal in developing the PAT guidance was to eliminate the
specter of regulatory uncertainty which has been identifi ed as a major factor limiting
innovation in pharmaceutical manufacturing. The guidance works with existing
regulations and was designed to be consistent with the agency ’ s twenty - fi rst - century
cGMPs initiative. Furthermore, the guidance emphasizes that the decision on the
part of manufacturers to work with the agency to implement PAT is voluntary. Since
the guidance is not prescriptive in nature, it neither describes “ how to do PAT ” nor
identifi es any particular practice or technology as “ approved for PAT. ”
4.1.5.1 Process Understanding
The agency considers PAT to be a “ system for designing, analyzing, and controlling
manufacturing through timely measurements of critical quality and performance
attributes of raw and in - process materials and processes, with the goal of ensuring
fi nal product quality. ” Based on this defi nition, it would be practical to consider PAT
to be an expansion of PAC; PAT builds on the measurement and control aspects of
PAC by incorporating additional emphasis on QbD and process understanding.
According to the PAT guidance, a process is generally considered well understood
when:
1. All critical sources of variability are identifi ed and explained.
2. Variability is managed by the process.
3. Product quality attributes can be accurately and reliably predicted over the
design space established for materials used, process parameters, manufacturing,
environmental, and other conditions.
Furthermore, according to the guidance, the ability to predict “ refl ects a high degree
of process understanding. ”
Possession of a predictive model (for product quality attributes) alone does
not necessarily constitute process understanding, however. A relatively common
example would be prediction of material or product performance characteristics
using multivariate measurements, such as prediction of tablet dissolution rate using
NIR spectroscopy. Multiple researchers have demonstrated that (in some cases) it
is possible to predict drug release from tablets in vitro using nondestructive NIR
spectra by generating a calibration model for dissolution rate. Without demonstrating
at least mechanistic understanding of the physicochemical feature (correlated
to dissolution rate) being detected by NIR, the calibration model would constitute
nothing more than pattern recognition (Figure 12 ) [64] . While such a calibration
may be useful, without greater insight as to the basis for correlation, it would not
likely be a useful demonstration of process understanding.
Design Space and Quality by Design The concept of a multidimensional space of
acceptable operating conditions, or design space, is perhaps one of the most important
aspects of the twenty - fi rst - century cGMPs which facilitates continuous improvement.
In a PAT - enabled environment, the process design space must provide
evidence of QbD [65] and should be the mathematical medium by which process
understanding and real - time control decisions are communicated (Figure 13 ).
The current ICH Q8 defi nition of design space, unfortunately, offers little guidance
with regard to the aspects of a process design space which are required for
implementation. As a result, a variety of interpretations of what constitutes a suitable
process design space have recently surfaced among industry participants. One
of the most popular misconceptions is that an effective design space for a process
or unit operation can be determined by the common trajectory of PAT measurements
(i.e., “ process signature ” ) related to product batches known to have acceptable
quality (i.e., “ golden path ” ). While such data are useful for monitoring, they
are nothing more than a modern version of “ 3 - batch ” process validation. Golden
paths or process trajectories are not suffi cient for control since 1) the path itself is
not necessarily predictive and 2) such controls would imply that a process is limited
by its historical path in the space of process parameters. Originally, the term process
signature was defi ned as a multivariate process measurement, that is, NIR spectrum,
which contained features useful for describing the impact of the process on the
chemical and physical aspects of the processed material [38] .
FIGURE 12 Illustration of aspects of method understanding which must be in place to
justify product performance measurements using indirect and/or nondestructive analyses.
PAT EVOLUTION IN PHARMACEUTICAL MANUFACTURING 337
338 REGULATORY AND INDUSTRIAL PERSPECTIVES
While it is perhaps too early to posit a conclusive standard for pharmaceutical
process design space development, the following minimum criteria should be
achieved for a process design space to be suitable for process control:
• The process design space should be expressed in the form of a mathematical
model which quantitatively links process capability , quality of input materials,
and process operating parameters.
• Relevant critical - to - quality product attributes should be considered by the
design space model (e.g., content uniformity, bioavailability, stability).
• Borrowing from a famous quote by Albert Einstein, the (design space) model
should be as complex as necessary (for accurate prediction), but no less.
• Product attributes that are superfl uous or are not known to be critical to quality
should not be considered by the design space model (there should not be a
penalty for monitoring such parameters, however).
• In the same way that in vitro – in vivo correlation (IVIVC) is required to be
granted a biowaiver for implementation of postapproval changes, the ability of
the design space model to predict the quality of fi nished goods must be validated
prior to implementation.
• If the accuracy of the design space model cannot be established a priori with
statistical signifi cance within portions of the parameter hyperspace, operation
in such regimes should initiate supplementary quality assurance (inspection)
activities until the design space model can be updated and revalidated.
• If unacceptable product quality is observed during operation within a region
of the design space expected to yield acceptable quality, the design space should
be considered unsuitable for process control (due to drift or the appearance of
new factors in the parameter space) until the missing factor(s) can be identifi ed
and incorporated into the model and the model is revalidated.
If such a model - based process design space includes a suffi cient portion of the
factors affecting product quality variance, the process control space can be projected
to defi ne the bounds of normal operation. Based on this defi nition, the control
FIGURE 13 Interrelation between design space, PAT, and process control in a manufacturing
system based on quality - by - design. ( Source : R. C. Lyon, Process monitoring of pilot - scale
pharmaceutical blends by near - infrared chemical imaging and spectroscopy, Eastern Analytical
Symposium (EAS), Somerset, NJ, 2006 .)
model algorithm for each operation in the manufacturing process would be generated
from a subset of the control space spanned by the material qualities and processing
parameters which impact that operation. Each unit operation control model
seeks to adjust process parameters in a timely manner in response to changes in
raw material (feedforward) or fi nished - product (feedback) quality. In other words,
control the process and validate the controls.
The mathematical linkage of the design space, process, and control models
enables continuous optimization of product quality by seeking the optimal point
within the control space. As the level of process understanding increases or as
processing conditions evolve, factors might be added or removed from the design
space and the process and control models updated. Furthermore, by considering
other factors such as yield, effi ciency, or C/T as a function of the variables spanned
by the process design space, the process might be co - optimized for quality and
profi tability.
It is likely that many pharmaceutical manufacturing operations are not understood
in a way that product quality variance can be fully described in functional
form (e.g., transfer functions); attaining such a level of manufacturing knowledge
should be a goal for the industry. Using functional representations of process understanding
as the basis set for a process design space, rather than historical performance,
offers many operational advantages:
• Effi cient Process Development While the current defi nition of design space
does not preclude the incorporation of knowledge from other products and
processes, model - based knowledge representation offers a more robust framework
for incorporation of external or a priori information. Even though the
level of quality expected by a particular combination of input and process
parameters from another product is not likely to transfer to a new product or
process (in absolute terms), the functional relationships which predict quality
may be quite similar. Furthermore, model - based design space development
enables direct incorporation of fi rst principles and mechanistic knowledge,
which might signifi cantly reduce the complexity of experimental designs
required for process development since signifi cant terms may be identifi ed in
silico.
• Quality by Design The incorporation of functional relationships between
inputs, parameters, and product quality (or effi ciency), which inherently imply
magnitude and directionality, enables the use of a process design space as a
tool for multiobjective process optimization. Furthermore, the model - based
representation of knowledge is compatible with concepts of risk management,
enabling more fl exible operation since the risk associated with extrapolation
could be predicted.
• Control System Development Model - based design space development offers
an ideal segue between process and control development. Quite literally,
a model - based design space would provide the template for development
of feedforward process control models. Moreover, development of a process
design space using a model - based framework would facilitate control
system validation and identifi cation of science - based, in - process, and release
specifi cations.
PAT EVOLUTION IN PHARMACEUTICAL MANUFACTURING 339
340 REGULATORY AND INDUSTRIAL PERSPECTIVES
• Scaling and Technology Transfer Within the current system for process development,
it is common to use designed experiments (i.e., DOE) where some
input variables are product specifi c (e.g., excipient “ grade ” ) or process parameters
are device dependent (e.g., chopper speed, damper angle). In a model -
based paradigm, however, a process design space would ideally be generated
using product - and device - independent units which have more basic physical
meaning (e.g., modulus, viscosity, energy, or work). Designing and describing
production processes in fundamental terms or, perhaps, standardized dimensionless
units would facilitate scaling and transfer of design space and process
control models to similar manufacturing processes that are based on the same
physical operating principles.
Academic research is currently underway to further develop the model - based
design space concept. Working within the limits of the current system, though, producers
who are able to demonstrate process understanding or are willing to invest
in a PAT system to facilitate their development of process understanding can use
the tools and provisions of the framework to pursue innovation and continuous
improvement with more effi cient regulatory oversight (i.e., the ability to make
changes without supplemental review). The PAT framework is described as consisting
of two components: (1) a set of scientifi c principles and tools supporting innovation
and (2) a strategy for regulatory implementation that will accommodate
innovation. The following paragraphs will describe selected aspects of both components
in detail.
4.1.5.2 PAT Principles and Tools
Central to the PAT framework is the acceptance that certain physical and mechanical
attributes of pharmaceutical ingredients are not necessarily well understood and
that even processes which have achieved signifi cant process understanding are
subject to a fi nite level of stochastic variation. Thus, the core of the PAT guidance
is allocated to describing the principles and tools, such as process analyzers and risk
analysis, which producers can employ to augment process understanding and mitigate
latent risks to product quality.
PAT Tools The guidance describes four categories of PAT tools:
• Multivariate tools for design, data acquisition, and analysis
• Process analyzers
• Process control tools
• Continuous improvement and knowledge management tools
Since each of the four categories draws upon methods and technology which are
already established in other fi elds such as PAC, the discussion of each category
within the guidance is focused on aspects which are unique or signifi cant to pharmaceutical
manufacturing, such as process signature [2] . Furthermore, in keeping
with the spirit of the framework as a catalyst for innovation, the agency made an
effort to avoid mention of any particular tool or technology in the fi nal version of
the PAT guidance. The PAT tools section of the guidance does, however, include
cross - references to relevant portions of current regulations which should be considered
by a manufacturer developing a PAT strategy or system.
Standards for Pharmaceutical Applications of PAT During the early stages of
developing the PAT framework, the agency was aware that the lack of international
standards was a signifi cant impediment to regulatory coordination and implementation
of PAT in the global pharmaceutical industry. In 2003, the FDA ’ s PAT team
worked with ASTM International to form Technical Committee E55 on Pharmaceutical
Application of Process Analytical Technology. The E55 committee addresses
issues related to process control, design, and performance as well as quality acceptance/
assurance tests for the pharmaceutical manufacturing industry. Stakeholders
in the committee include manufacturers of pharmaceuticals and pharmaceutical
equipment, federal agencies, design professionals, professional societies, trade associations,
fi nancial organizations, and academia ( www.ASTM.org ).
As of mid - 2006, there were three subcommittees of E55: PAT system management,
PAT system implementation and practice, and PAT terminology. The PAT
team has been represented on E55 committees with a goal to ensure that standards
developed are aligned with the PAT guidance and acceptable to the FDA. To date,
one active standard has been published, while 16 additional standards have been
proposed. The ASTM International provides another venue for international cooperation
(consistent with the twenty - fi rst - century cGMPs initiative); the defi nitions
of PAT (in the FDA guidance and ASTM E55) as well as other concepts are being
incorporated into the ICH Q8 guidance.
Real - Time Release ( RTR ) The PAT guidance defi nes RTR as “ the ability to
evaluate and ensure the acceptable quality of in - process and/or fi nal product based
on process data. ” Whereas fi nished products are typically released for marketing
only after sampling, inspection (i.e., laboratory - based QC testing), and review,
implementation of an RTR system enables release of fi nished products concurrent
with the completion of manufacturing operations. Practically speaking, RTR is one
of the most signifi cant, tangible benefi ts for producers who implement PAT, because
it can facilitate dramatic reductions in process C/T.
Real - time release is considered by the guidance to be comparable to alternative
analytical procedures for fi nal product release and is defi ned within the guidance as
an extension of parametric release. The defi ning characteristic of RTR is that it
considers simultaneously the degree to which material attributes and process parameters
are measured and controlled during manufacturing. It was not intended that
RTR be implemented by simply installing a rapid measurement system at the end
of a manufacturing process; such uses for PAT tools would be tantamount to inspection
and would do nothing to improve quality management.
The guidance does suggest, however, that it may be feasible to implement RTR
without fi nished - product quality monitoring by using “ the combined process measurements
and other test data gathered during the manufacturing process. ” Similar
language is found in the USP general notices, where it is suggested that data derived
from “ [validation studies and] in - process controls may provide greater assurance
that a batch meets a particular monograph requirement than analytical data derived
from an examination of fi nished units drawn from that batch. ” It would not be
PAT EVOLUTION IN PHARMACEUTICAL MANUFACTURING 341
342 REGULATORY AND INDUSTRIAL PERSPECTIVES
diffi cult to create a system more capable of detecting quality variation than current
methods based on inspection. Recent statistical analyses [42] have demonstrated
that, for determining batch quality, the traditional USP . 905 . method of content
uniformity testing may indeed have little more statistical power than a coin toss
until more than 5% of the product exceeds specifi cation limits (corresponds to
within - batch C pk of approximately 0.65, only slightly worse than has been observed
in a recent industry benchmarking study [28] ).
On the other hand, deployment of an RTR system without fi nished - product
monitoring would require the manufacturer to demonstrate a very high level of
process understanding based on, for example, their development of a comprehensive
design space and/or a well - validated process model. Even though it may be
feasible to implement RTR without end - of - process monitoring, a well - designed PAT
system will typically include some form of fi nal product quality monitoring as a
means for mitigating latent risk and creating strategic redundancy in process controls
and as an additional tool to bolster process understanding.
4.1.5.3 Strategy for Implementation
One of the FDA ’ s goals for the PAT guidance is to “ tailor the Agency ’ s usual regulatory
scrutiny to meet the needs of PAT - based innovations that (1) improve the
scientifi c basis for establishing regulatory specifi cations, (2) promote continuous
improvement, and (3) improve manufacturing while maintaining or improving the
current level of product quality. ” Recognizing that the achievement of this goal
requires a unique interface between regulators and manufacturers seeking to implement
PAT, a strategy for implementation based on the integrated systems approach
was developed. An objective of the strategy for implementation is to facilitate clear,
effective, and meaningful communication between the agency and industry, for
example, in the form of meetings or informal communication.
In practice, the strategy breaks with traditional industry – FDA modes of communication;
whenever PAT is concerned, it is anticipated that regulators will communicate
directly with the pharmaceutical scientists and engineers involved with
development and operation of the PAT system rather than indirectly via a department
of regulatory affairs. The components of the agency ’ s regulatory strategy
include:
• A PAT team approach for CMC review and cGMP inspections
• Joint training and certifi cation of PAT review, inspection, and compliance
staff
• Scientifi c and technical support for the PAT review, inspection, and compliance
staff
• Recommendations provided within the PAT guidance
PAT Team Approach FDA ’ s assembly of the PAT team was one of the most signifi
cant incentives for the industry to pursue manufacturing innovation as described
in the twenty - fi rst - century cGMPs initiative and the PAT guidance. The PAT team
was put in place to ensure that industrial PAT applications were handled with expediency
and accuracy by scientists familiar with the most up - to - date PAT methods.
At one point the PAT team included more than 20 scientists, including investigators,
compliance offi cers, reviewers, training coordinators, and a policy development
team. More recently the agency has begun steps to “ sunset ” the PAT team, the duties
of which will ultimately be handled by FDA staff trained in PAT systems. A comprehensive
scientifi c training program was developed for the PAT team with guidance
from the ACPS PAT subcommittee. Initial training began in January 2006, with
plans for further training to be provided by faculty at Duquesne and Delaware
Universities [47] .
Research Data Provision In developing the PAT guidance, the FDA recognized
that, even with the guidance in place, manufacturers seeking to evaluate the suitability
or potential value of new technologies for process control may be hesitant,
fi guring that such data will be subject to cGMP inspection, thereby increasing their
liability with respect to regulatory actions. To allay these fears, the agency included
a statement which applies to investigational deployment of new technologies [2] :
Data collected using an experimental tool should be considered research data. If
research is conducted in a production facility, it should be under the facility ’ s own
quality system. . . . FDA does not intend to inspect research data collected on an existing
product for the purpose of evaluating the suitability of an experimental process
analyzer or other PAT tool. FDA ’ s routine inspection of a fi rm ’ s manufacturing process
that incorporates a PAT tool for research purposes will be based on current regulatory
standards (e.g., test results from currently approved or acceptable regulatory methods).
Any FDA decision to inspect research data would be based on exceptional situations
similar to those outlined in Compliance Policy Guide Sec. 130.300. Those data used to
support validation or regulatory submissions will be subject to inspection in the usual
manner.
4.1.6 PAT IMPLEMENTATION PROCESS
The PAT guidance identifi es three possible plans for companies seeking to implement
PAT:
• PAT can be implemented under the facility ’ s own quality system; cGMP inspections
by the PAT team or PAT - certifi ed investigator can precede or follow PAT
implementation.
• A changes being effected (CBE), CBE in 30 days (CBE - 30), or prior approval
(PAS) supplement can be submitted to the agency prior to implementation,
and, if necessary, an inspection can be performed by a PAT team or PAT -
certifi ed investigator before implementation.
• A comparability protocol (CP) can be submitted to the agency outlining PAT
research, validation, and implementation strategies and time lines. Following
approval of this comparability protocol by the agency, one or a combination of
the above regulatory pathways can be adopted for implementation.
Refl ecting its nonprescriptive nature, the three implementation plans are essentially
the only “ how to ” portions of the PAT guidance. This leaves industrial (and academic)
scientists and engineers with the burden of determining how best to proceed
PAT IMPLEMENTATION PROCESS 343
344 REGULATORY AND INDUSTRIAL PERSPECTIVES
in the deployment of a PAT system. Despite the fact that some pioneering companies
have been incorporating aspects of PAT in their operations since long before
the start of the FDA ’ s twenty - fi rst - century cGMPs initiative, there continues to be
signifi cant diversity in their approaches to implementation. While perhaps the ambiguity
(in how best to proceed) has slowed the uptake of PAT to some degree, in the
long run, the latitude is preferable since the optimal path of implementation will
likely be unique for most facilities.
With regard to drug manufacturers ’ implementation of PAT, a list of 10 questions
has been presented which provides an initial checklist for companies seeking
approval of their plans [10, 66] :
1. Is this a PAT system?
2. Does it have aspects of design, measurement, and manufacturing control?
3. Are PAT principles and tools used?
4. Which tools specifi cally are used for manufacturing control?
5. How are continuous improvement and knowledge management
performed?
6. What risk - based approach has the company taken — assessment, prevention,
and management?
7. How are the PAT systems integrated?
8. What kind of RTR is being proposed or used?
9. What regulatory process is being considered?
a. Can the companies ’ quality systems manage the PAT change?
b. Are the submission proposals appropriate and justifi ed?
10. What are the critical aspects that will be evaluated during site visits/
inspections?
Drawing from aspects of the DMAIC model, as well as the risk - based orientation
and quality systems approach espoused by the FDA ’ s twenty - fi rst - century cGMPs
initiative, the Duquesne University Center for Pharmaceutical Technology (DCPT)
has proposed a six - phase, iterative cycle for process improvement based on PAT
(Figure 14 ). While there are certainly many acceptable variants of this strategy, some
of which have begun to appear in conferences and the industrial literature, any successful
PAT deployment, large or small, will most likely include some combination
of these elements. In addition, each project phase will necessarily include one or
more modules of training. Finally, while the project phases are presented as being
discrete, most of the phases will overlap to some degree. In particular, consideration
of the objectives for control, release strategies, and plans for continuous improvement
should begin, along with management buy - in, early in the cycle.
4.1.6.1 Preparation
The preparation phase is arguably the most critical step in the path toward PAT
implementation. Process analytical technology projects are inherently multidisciplinary,
requiring acceptance and buy - in from corporate divisions which sometimes
FIGURE 14 PAT implementation cycle with examples of associated activities for each
phase.
operate with rather divergent goals and procedures. Most importantly, those who
are seeking to initiate a PAT project will need to obtain management buy - in at a
level high enough in the corporate structure to ensure suffi cient resources will be
available and that the company will be committed to positive change. During the
preparation phase, a PAT team having a diverse background and critical skills
should be assembled, and formal planning of the project should begin, including
selection of the product and process to address. Ideally, dialogue with the FDA PAT
team should begin early in the preparation phase.
4.1.6.2 Assessment
The PAT guidance clearly states that industrial implementations should be risk
based. Soon after the PAT team and objective have been identifi ed, the project
should commence with a formal risk assessment. The risk assessment should be
focused on identifying and characterizing the failure modes which present risks to
product quality; the outcome of the risk assessment will provide a means prioritizing
the allocation of PAT resources and a baseline for review of the effect of PAT in
mitigating risks to quality.
4.1.6.3 Analyze
The “ analyze ” phase of the project consists of the activities which are typically
associated with PAC, including identifi cation and assessment of potential sensor
technologies, method development, qualifi cation, and validation. In addition,
designed experiments (DOE) or data - mining exercises may be performed to
PAT IMPLEMENTATION PROCESS 345
346 REGULATORY AND INDUSTRIAL PERSPECTIVES
generate process understanding or to support PAT goals. Plans for the IT infrastructure,
sampling protocols, and development of controls should also be considered.
4.1.6.4 Control
The implementation of controls begins as each new analytical method or technology
is deployed. Controls may be as simple as automated termination of a unit operation
upon reaching an endpoint. With greater process understanding, more complex
controls can be deployed, including feedback (e.g., control of punch force during
tablet compaction, control of temperature or airfl ow during fl uid bed processing)
or feedforward controls (e.g., adjustment of process parameters based on incoming
raw - material quality). The development and implementation of controls should also
consider operating procedures for adverse situation management and should initiate
a reassessment of risk to determine the suitability of controls.
4.1.6.5 Release Philosophy
For PAT projects including implementation of RTR or some modifi cation of a preexisting
release mechanism for an approved process, additional method development
and validation procedures will be required. The real - time release decision will
typically be determined by a process model, which can be a mathematical equation
or algorithm within the control system; furthermore, the IT system must accurately
convey the release decision and supporting data to downstream operations (i.e.,
warehouse, logistics), upstream operations (i.e., production scheduling, accounting),
or the facility information repository. The interwoven IT and scientifi c components
require an integrated systems approach to development, validation, deployment,
and operation. Finally, implementation of PAT systems enables redefi nition of
product quality acceptance criteria for release; the task of identifying robust release
criteria suitable for large sample sizes, for example, continues to merit examination
[43] .
4.1.6.6 Optimization
The optimization phase of the project provides an opportunity to assess the performance
of the PAT system relative to the goals of the project as well as the level of
latent risk in the system. Ideally, with the PAT s2ystem in place, the level of process
understanding will be improving as more data are collected for every batch. The
added insight into the operation may yield new opportunities for improving quality
or effi ciency or for solving similar problems with another product. The key to success
in the optimization stage is realizing that it is only the beginning of continuous
improvement.
4.1.7 PERSPECTIVES ON THE IMPACT OF PAT
PAT and the twenty - fi rst - century cGMPs initiative have clearly made an impact
within the pharmaceutical and associated industries. Signifi cant sums of capital
are now fl owing in new directions to meet the challenges and opportunities pre
sented by the changes. Some people within the industry, however, question
whether there will be much of a long - term impact, citing the litany of new eras
in the industry (and their careers) that turned out to be more of the same. With
just a bit of observation, though, it is not hard to see that it really is different
this time.
The modern pharmaceutical manufacturing industry fi nds itself in a diffi cult situation
that perhaps few anticipated just 10 or 15 years ago. The rate of new blockbuster
drug approvals has continued to wane, while new drug therapies become
inexorably more expensive to discover and develop. Despite the fact that the market
for drug sales has never been larger, drug company profi t margins are shrinking
while consumers, feeling that pharmaceutical company profi ts are unjust, have
reached new lows in their opinion of the industry. A recent survey by the Kaiser
Family Foundation placed pharmaceutical companies just above oil and tobacco
companies, and right below health management organizations (HMOs), in terms of
public opinion [67] . Entities of signifi cant magnitude in both the public and private
sector are increasingly applying pressure to capture an even greater portion of the
industry ’ s compensation. Indeed, there is no shortage of industrial and fi nancial
publications which have chronicled the pharmaceutical industry ’ s troubles [34, 45,
68, 69] .
The pharmaceutical industry is fortunate, perhaps, to follow (rather than lead)
most other major industries in adopting truly automated controls, process analytics,
quality management, and lean manufacturing. The performance of pharmaceutical
companies relative to the benchmarks for world - class manufacturers provides a
roadmap for improvement. If the pharmaceutical industry, as a whole, were able to
at least approach the benchmarks for world - class manufacturing performance (by
implementing PAT), the savings returned to consumers and shareholders would be
immense (Figure 15 ). Finally, the returns on investment in PAT are not limited to
major producers. Estimates based on recent benchmarks suggest that, by successfully
transforming operations through the deployment of PAT and lean, a typical
small or mid - sized pharmaceutical manufacturer could improve operating margins
by up to 600 basis points [44] .
Forces which are out of the industry ’ s control are providing more reasons
than ever before to seek effi ciency in pharmaceutical manufacturing, and the
FDA is doing its part to clear the way. While the pharmaceutical industry has
likely been unjustly cast as a culprit behind America ’ s fi scal crisis in health care, the
industry has ample opportunity to change for the benefi t of patients as well as
investors.
ACKNOWLEDGMENTS
The author would like to thank the following reviewers for their input, which was
essential to the quality of this manuscript: James K. Drennen, III, Ph.D., Director,
Duquesne University Center for Pharmaceutical Technology, Senior Consultant,
Strategic Process Control Technologies; Robbe C. Lyon, Ph.D., Deputy Director,
Division of Product Quality Research FDA/CDER; D. Christopher Watts, Ph.D.,
Team Leader, Standards & Technology, FDA/CDER/OPS; and Tom Knight, Founder
& Chief Strategy Offi cer, Invistics Corp.
ACKNOWLEDGMENTS 347
348 REGULATORY AND INDUSTRIAL PERSPECTIVES
FIGURE 15 Potential fi nancial returns from deployment of PAT and lean. The curves are
calculated based on the aggregate COGS and inventories reported in the 2005 annual reports
of the top 16 branded and generic pharmaceutical manufacturers (according to market capitalization).
It is important to keep in mind that working capital savings are a one - time - only
benefi t, while cost of quality and inventory fi nancing and overhead savings represent on - going
returns on investment. Furthermore, while the curves may overestimate savings because of
innacuracies in benchmark data or the limits on the opportunities for PAT implementation,
they do not account for numerous other potential pathways for returns from PAT such as
capacity increase, labor productivity enhancement, reduction of QC expense, or decreased
time to market.
Cost of Quality Saving ($ Billions)
10
9
8
7
6
5
4
3
2
1
0
0.5 0.6 0.7 0.8 0.9 1
Process Capability (Cpk)
2 4 6 8 10 12 14
Total Inventory Turn Rate (Turns/Year)
35
30
25
20
15
10
5
0
Supply Chain Optimization Savings ($ Billions)
Working Capital Savings
Financing & Overhead Savings
REFERENCES
1. U.S. Department of Health and Human Services ( 2004 ), Pharmaceutical CGMPs for
the 21st century — A risk - based approach, Final report, Food and Drug Administration,
Rockville, MD.
2. U.S. Department of Health and Human Services ( 2004 ), Guidance for industry: PAT — A
framework for innovative pharmaceutical development, manufacturing, and quality
assurance, Food and Drug Administration, Rockville, MD.
3. Clevett , K. J. ( 1986 ), Process Analyzer Technology , Wiley , New York .
4. Illman , D. L. ( 1986 ), CPAC: An industry — university cooperative research center for
process analytical chemistry , TrAC Trends Anal. Chem. , 5 , 164 .
5. Callis , J. B. , Illman , D. L. , and Kowalski , B. R. ( 1987 ), Process analytical chemistry , Anal.
Chem. , 59 , 624A .
6. Balboni , M. L. ( 2003 ), Process analytical technology: concepts and principles, Pharm.
Technol .
7. Koch , M. V. ( 2006 ), Optimizing the impact of developments in micro - instrumentation on
process analytical technology: a consortium approach , Anal. Bioanal. Chem. , 384 , 1049 .
8. K u ppers , S. , and Haider , M. ( 2003 ), Process analytical chemistry — future trends in industry
, Anal. Bioanal. Chem. , 376 , 313 .
9. K u ppers , S. , and Haider , M. ( 2006 ), Process analytical chemistry , Anal. Bioanal. Chem. ,
384 , 1034 .
10. Hinz , D. C. ( 2006 ), Process analytical technologies in the pharmaceutical industry: the
FDA ’ s PAT initiative , Anal. Bioanal. Chem. , 384 , 1036 .
11. Beebe , K. R. , et al ., ( 1993 ), Process analytical chemistry , Anal. Chem. , 65 , 199R .
12. Blaser , W. W. , et al., (1995), Process analytical chemistry , Anal. Chem. , 67 , 47R .
13. Workman , J. J. , et al ., ( 1999 ), Process analytical chemistry , Anal. Chem. , 71 , 121R .
14. Workman , J. J. , et al ., ( 2001 ), Process analytical chemistry , Anal. Chem. , 73 , 2705 .
15. Workman , J. J. , Koch , M. V. , and Veltkamp , D. J. ( 2003 ), Process analytical chemistry , Anal.
Chem. , 75 , 2859 .
16. Workman , J. J. , Koch , M. V. , and Veltkamp , D. J. ( 2005 ), Process analytical chemistry , Anal.
Chem. , 77 , 3789 .
17. Gluckman , P. , Roome , D. R. , Deming , W. E. , and Delavigne , K. ( 1993 ), Everyday Heroes
of the Quality Movement: From Taylor to Deming, the Journey to Higher Productivity ,
2nd ed., Dorset house , New York .
18. Shewhart , W. A. ( 1980 ), Economic Control of Quality of Manufactured Product , ASQC
Quality Press , Milwaukee, WI .
19. Shewhart , W. A. ( 1986 ), in Deming W. E. ed., Statistical Method from the Viewpoint of
Quality Control , Dover , New York .
20. Scherkenbach , W. W. ( 1991 ), The Deming Route to Quality and Productivity: Road Maps
and Roadblocks , ASQC Quality Press , Milwaukee, WI .
21. Juran , J. M. ( 2003 ), Architect of Quality: The Autobiography of Dr. Joseph M. Juran ,
McGraw - Hill , New York .
22. Bhote , K. R. ( 2002 ), What is Six Sigma? The Ultimate Six Sigma . Vol 1. New York :
AMACOM ; 9 – 14 .
23. Galvin , R.W. ( 2002 ), Forward . The Ultimate Six Sigma . Vol 1. New York : AMACOM ;
xxi – xxii .
24. Marash , S. A. , Berman , P. , and Flynn , M. ( 2004 ), Fusion Management: Harnessing the
Power of Six Sigma, Lean, ISO 9001:2000, Malcom Baldridge, TQM and Other Quality
Breakthroughs of the Past Century , QSU Pub., Fairfax, VA .
REFERENCES 349
350 REGULATORY AND INDUSTRIAL PERSPECTIVES
25. Ealey , L.A. ( 1988 ), Quality by design. Taguchi Methods and U.S. industry . Dearborn, MI :
ASI Press .
26. Taguchi , G. , Chowdhury , S. , and Wu , Y. ( 2005 ), Taguchi ’ s quality engineering handbook .
Hoboken, NJ : John Wiley & Sons .
27. e - Handbook of Statistical Methods. NIST/SEMATECH . Available at: http://www.itl.nist.
gov/div898/handbook/ .
28. Macher , J. , and Nickerson , J. ( 2006 ), Pharmaceutical manufacturing research project: Final
benchmarking report, Georgetown University, McDonough School of Business and
Washington University Olin School of Business, Washington DC.
29. Levinson, W. A. (2002), Henry Ford ’ s Lean Vision , Productivity Press , New York .
30. Leiper , K. ( 2006 ), paper presented at the The Heidelberg PAT Conference, Heidelberg,
Germany.
31. Lewis , N. A. ( 2006 ), A tracking tool for lean solid - dose manufacturing , Pharm. Technol.
30 , 94 – 108 .
32. Roumeliotis , G. ( 2006 ), Lean proves mean in drug manufacturing , available: in -
PharmaTechnologist.com .
33. Gerecke , G. , and Knight , T. , Improving performance and reducing cycle time using fl ow
path management: A case study , Pharm. Eng. 21 .
34. Abboud , L. , and Hensley , S. ( 2003 ), Factory shift: New prescription for drug makers:
Update the plants, Wall Street J. , Sept. 3, p. 1.
35. Cooley , R. E. , and Egan , J. C. ( 2004 ), The impact of process analytical technology (PAT)
on pharmaceutical manufacturing , Am. Pharm. Rev. , 7 , 62 – 68 .
36. Cogdill, R. P. (2006), in Brittain, H. G. , Ed., Spectroscopy of Pharmaceutical Solids , Taylor
& Francis Group , New York , pp. 313 – 412 .
37. Cogdill , R. P. , et al ., ( 2005 ), Process analytical technology case study, Part II: Development
and validation of quantitative for tablet API content and hardness , AAPS Pharm. Sci.
Tech. , 6 , Article 38.
38. Cogdill , R. P. , et al ., ( 2005 ), Process analytical technology case study, Part I: Feasibility
studies for quantitative NIR method development , AAPS Pharm. Sci. Tech. , 6 , Article
37.
39. Cogdill , R. P. , Anderson , C. A. , and Drennen , J. K. ( 2005 ), Process analytical technology
case study, Part III: Calibration monitoring and transfer , AAPS Pharm. Sci. Tech. , 6 ,
Article 39.
40. Lorber , A. ( 1986 ), Error propagation and fi gures of merit for quantifi cation by solving
matrix equations , Anal. Chem. , 58 , 1167 .
41. Braga , J. W. B. , and Poppi , R. J. ( 2004 ), Figures of merit for the determination of the
polymorphic purity of carbamazepine by infrared spectroscopy and multivariate calibration
, J. Pharm. Sci. , 93 , 2124 .
42. Lunney , P. , and Drennen , J. K. I. ( 2005 ), A prevention based strategy for quality control
using PAT , NIR News , 16 , 7 .
43. Sandell , D. , Diener , M. , Vukovinsky , K. , Hofer , J. , and Pazdan , J. ( 2006 ), Development of
a content uniformity test suitable for large sample sizes , Drug. Info. J. , 40 , 337 .
44. Cogdill , R. P. , Knight , T. P. , Anderson , C. A. , and Drennen , J. K. ( 2007 ), The fi nancial
returns on investments in process analytical technologies and lean manufacturing: Benchmarks
and case study , J. Pharm. Innov. , 2(1–2) , 38 – 50 .
45. du Pre Gauntt , J. ( 2005 ), Quality manufacturing: A blockbuster opportunity for pharmaceuticals,
Economist Intelligence Unit.
46. Steyer , R. ( 2005 ), Merck names clark CEO, available: TheStreet.com , accessed May 5,
2005.
47. Watts , D. C. ( 2006 ), PAT — An FDA paper presented at the The Heidelberg PAT Conference,
Heidelberg, Germany.
48. Hussain , A. S. ( 2001 ), Emerging science issues in pharmaceutical manufacturing: Process
analytical technologies , paper presented at the Science Board Presentations to FDA,
Rockville, MD.
49. Chisholm , R. S. ( 2001 ), TQMS, statistically based in-process control with real time quality
assurance, the AstraZeneca total quality management strategy , paper presented at the
Science Board Presentations to FDA, Rockville, MD.
50. Raju , G. K. ( 2001 ), Pharmaceutical manufacturing: New technology opportunities , paper
presented at the Science Board Presentations to FDA, Rockville, MD.
51. Scherzer , R. H. ( 2002 ), Quality by design: A challange to the pharma industry , paper
presented at the Science Board Presentations to FDA, Rockville, MD.
52. U.S. Department of Health and Human Services (2004), Risk-based method for prioritizing
cGMP inspections of pharmaceutical manufacturing sites — A pilot risk ranking
model, Food and Drug Administration, Rockville, MD.
53. U.S. Department of Health and Human Services ( 2006 ), SMG 2020 — FDA quality
system framework for internal activities, Food and Drug Administration, Rockville,
MD.
54. U.S. Department of Health and Human Services ( 2006 ), Guidance for industry: Quality
systems approach to pharmaceutical cGMP regulations, Food and Drug Administration,
Rockville, MD.
55. U.S. Department of Health and Human Services ( 2004 ), Guidance for industry: Sterile
drug products produced by aseptic processing — Current good manufacturing practice,
Food and Drug Administration, Rockville, MD.
56. U.S. Department of Health and Human Services ( 2003 ), Guidance for industry: Comparability
protocols — Chemistry, manufacturing, and controls information, draft guidance,
Food and Drug Administration, Rockville, MD.
57. U.S. Food and Drug Administration (FDA) ( 2004 , Mar.), Sec. 490.100 Process validation
requirements for drug products and active Pharmaceutical ingredients subject to pre -
market approval (CPG 7132c.08), FDA, Rockville, MD.
58. International Conference on Harmonization (ICH) ( 2004 ), Q8: Pharmaceutical development,
International Conference on Harmonization of Technical Requirements for Registration
of Pharmaceuticals for Human Use, ICH, Geneva.
59. International Conference on Harmonization (ICH) ( 2004 ), M4: Organisation of the
common technical document for the registration of pharmaceuticals for human use,
International Conference on Harmonization of Technical Requirements for Registration
of Pharmaceuticals for Human Use, ICH, Geneva.
60. U.S. Department of Health and Human Services ( 2006 ), Guidance for industry: Providing
regulatory submissions in electronic format — Human pharmaceutical product applications
and related submissions using the eCTD specifi cations, Food and Drug Administration,
Rockville, MD.
61. Yu , L. X. ( 2006 ), Implementation of quality - by - design: Question - based review , paper
presented at the Drug Information Association 42nd Annual Meeting, Philadelphia,
PA.
62. International Conference on Harmonization (ICH) ( 2005 ), Q9: Quality risk management,
International Conference on Harmonization of Technical Requirements for Registration
of Pharmaceuticals for Human Use, ICH, Geneva.
63. International Conference on Harmonization (ICH) ( 2005 ), Q10: Pharmaceutical quality
systems, fi nal concept paper, International Conference on Harmonization of Technical
Requirements for Registration of Pharmaceuticals for Human Use, ICH, Geneva.
REFERENCES 351
352 REGULATORY AND INDUSTRIAL PERSPECTIVES
64. Hussain , A. S. ( 2006 ), Quality by design and bioequivalence/bioavailability assessment ,
paper presented at the The Heidelberg PAT Conference 2006, Heidelberg, Germany.
65. Lyon , R. C. , and Hammond , S. ( 2006 ), Process monitoring of pilot - scale pharmaceutical
blends by near - infrared chemical imaging and spectroscopy , paper presented at the
Eastern Analytical Symposium, Somerset, NJ.
66. D ’ Sa , A. ( 2005 ), Process analytical technology (PAT): regulatory process, review and
inspection , paper presented at the 19th International Forum on Process Analytical Technology
—IFPAC 2005, Arlington, VA.
67. Views on prescription drugs and the pharmaceutical industry, The Kaiser Family Foundation,
2005 .
68. Arlington , S. , Barnett , S. , Hughes , S. , Palo , J. , and Shu , E. ( 2002 ), Pharma 2010: The threshold
of innovation, IBM Business Consulting Services, Somers, NY.
69. Arlington , S. , et al ., ( 2005 ), The metamorphosis of manufacturing, IBM Business Consulting
Services, Somers, NY.
353
4.2
PROCESS ANALYTICAL TECHNOLOGY
Michel Ulmschneider and Yves Roggo
F. Hoffmann-La Roche Ltd, Basel, Switzerland
Contents
4.2.1 Basic Concepts and Impact
4.2.1.1 Defi nition
4.2.1.2 What Motivated PAT?
4.2.1.3 Root - Cause Analysis and Process Control
4.2.1.4 When to Introduce PAT
4.2.1.5 PAT Enhances Process Understanding
4.2.1.6 Changing Current Practice Using PAT
4.2.1.7 Promoting Physical Pharmacy and Pharmaceutical Sciences
4.2.1.8 Data Mining
4.2.1.9 Data Warehousing
4.2.1.10 Data - Mining Methods for Pharmaceutical Processes
4.2.1.11 Data - Mining Practice
4.2.1.12 Comments about Data Mining
4.2.1.13 PAT Methods
4.2.1.14 Conclusion
4.2.2 Vibrational Spectroscopy
4.2.2.1 Introduction
4.2.2.2 IR Spectroscopy Theory
4.2.2.3 Mechanical Model of IR Vibration
4.2.2.4 Quantum Mechanical Model
4.2.2.5 Anharmonicity
4.2.2.6 Structure Elucidation Using MIRS
4.2.2.7 Extending Use of MIRS
4.2.2.8 Raman Spectroscopy
4.2.2.9 Introducing NIRS
4.2.2.10 Benefi ts of NIRS
4.2.2.11 Introducing MIR/NIR Chemical Imaging
4.2.2.12 Design of MIR Instruments
4.2.2.13 Conclusion
Pharmaceutical Manufacturing Handbook: Regulations and Quality, edited by Shayne Cox Gad
Copyright © 2008 John Wiley & Sons, Inc.
354 PROCESS ANALYTICAL TECHNOLOGY
4.2.3 Chemometrics
4.2.3.1 Introduction
4.2.3.2 From Univariate to Multivariate Regression
4.2.3.3 Sample Quality and Data Error
4.2.3.4 Mathematical Preprocessing of Spectroscopic Data
4.2.3.5 Preprocessing NIR Data
4.2.3.6 Mathematical Pretreatment and Transformation
4.2.3.7 Principal - Component Analysis
4.2.3.8 PCA Practice for NIRS
4.2.3.9 Pattern Recognition
4.2.3.10 SIMCA Classifi cation
4.2.3.11 Regression
4.2.3.12 Multiple Linear Regression
4.2.3.13 PCR and PLS Regression
4.2.3.14 Regression Practice in NIRS
4.2.3.15 Some Pitfalls
4.2.3.16 Example Analytical Applications of NIRS
4.2.3.17 Conclusion
Bibliography
4.2.1 BASIC CONCEPTS AND IMPACT
4.2.1.1 Defi nition
Process analytical technology (PAT) is one of the objectives contained in the Initiative
for Pharmaceutical cGMPs for the 21st Century published by the Food and Drug
Administration (FDA). In a few words and according to the FDA ’ s guideline, PAT
can be defi ned as a system for designing, analyzing, and controlling pharmaceutical
manufacturing through the measurement of critical quality and performance parameters.
The measurements performed on raw and in - process materials or process
parameters are intended to enhance fi nal product quality.
Process analytical technology encourages technological innovation, specifi cally
the adoption of new analytical techniques by the pharmaceutical industry designed
to improve the understanding and control of manufacturing processes. Both the
FDA and industry experts expect benefi ts over conventional manufacturing practices:
higher fi nal product quality, increased production effi ciency, decreased operating
costs, better process capacity, and fewer rejects. Correspondingly, fundamental
changes are also expected within the regulatory framework. The future of pharmaceutical
production will require innovative technological approaches and more
science - based processes. PAT will boost collaboration between research and development
(R & D) and manufacturing departments inside companies and increase
overall effi ciency. Approvals and inspections will increasingly focus on scientifi c and
engineering principles. As a result, regulators will set higher expectations for new
products from the outset.
4.2.1.2 What Motivated PAT ?
Preliminary discussions of PAT concepts between the FDA and certain pharmaceutical
companies already active in this fi eld date back to the late 1990s. In September
BASIC CONCEPTS AND IMPACT 355
2004 the FDA released a document for the industry entitled “ PAT Guidance for
Industry: A Framework for Innovative Pharmaceutical Development , Manufacturing,
and Quality Assurance. ” PAT is clearly anchored in FDA corporate culture.
Pharmaceutical companies are facing growing demands for increased productivity
and reduced manufacturing costs. They also have to meet the evolving need for
higher quality standards and higher drug expectations. At the same time the quest
for new active substances remains a signifi cant issue. Reducing the attrition rate
among selected candidates will bring more new medicines onto the market. In terms
of drug marketing, the goal is to improve formulations so as to offer patients innovative
and more effi cient solutions, and thus achieve commercial success or breakthrough.
By prioritizing science - based design and introducing novel or improved
process techniques, backed by the generation of increased critical data throughout
a drug ’ s life cycle, the aim of the emerging PAT strategy is to direct the drug industry
toward these essential goals.
Because they have been used for many years, a variety of existing experimental
methods and manufacturing processes are considered well established. They are
trusted to generate few errors and make only modest contributions to process variation.
Due to their longevity, they continue to be widely used in recent drug developments.
Improvements in existing technologies are always possible and are constantly
being made. However, this makes it diffi cult to consider or identify potential technological
alternatives without critical review or a voluntary management decision
to replace well - established techniques. The FDA noticed that nearly all recent drug
developments lacked the possibility of enhancing and extending process capabilities
toward newer or alternative technologies. More specifi cally, the FDA wanted to
encourage drug manufacturers to achieve more innovation and improve risk management
when releasing new medicines on the market.
4.2.1.3 Root - Cause Analysis and Process Control
When a quality problem arises in present - day production, it is increasingly diffi cult
to identify the root cause. Thorough understanding of process and product performance
often comes up against knowledge barriers, whether due to the escalating
documentation burden, lack of time, or loss of expertise. The goal of PAT is to
enhance process control and understanding so that procedures can be performed
differently and more effi ciently. The PAT initiative facilitates and encourages the
introduction of innovative approaches. It makes it possible to consider shifting from
validation to continuous verifi cation. The next step is effective real - time release with
continuous processing as an alternative to the conventional batch - after - batch production
scheme.
4.2.1.4 When to Introduce PAT
Building quality into a pharmaceutical product has to be considered from the very
beginning of the product ’ s life. Essential preconditions are the equal involvement
of — and seamless communication between — R & D and manufacturing. One purpose
of PAT is to provide a motivating framework to bring quality into a product from
the outset. It is thus essential for it to be involved in the R & D phase. If product
quality requirements are understood and implemented from the beginning,
356 PROCESS ANALYTICAL TECHNOLOGY
root - cause analysis of quality or process failure after scale - up to commercial manufacturing
will be much easier. This is why PAT could play an even more important
role in the design and analysis of manufacturing processes, enabling performance
control to be based on timely measurement of well - described critical processing
data.
Data processing needs should also be considered in the context of overall process
analysis strategy to meet emerging requirements for the speed and volume of data
collection. Real - time analysis supported by knowledge management requires collecting
and gathering all production batch information, for example, by data warehousing.
Thus, a PAT data management strategy based on online process analysis
or data mining can be set up long before generating large sets of measurement data.
Historical data analysis should aim to cover method development, method validation,
and ongoing performance monitoring, as well as routine results for a given
manufacturing process.
4.2.1.5 PAT Enhances Process Understanding
Process analytical technology can greatly enhance process understanding. In fact,
introducing PAT can act as a key driver to better process knowledge. The expected
steps in implementing the PAT approach are the collection of online, in - line, and
at - line data (Figure 1 ) on critical attributes, extraction of information, and analysis
of process status data, ending with closure of the loop by dynamic process control.
Innovating during development, applying cutting - edge techniques, and process
modeling whenever possible, all contribute to a more fundamental exploration of
the science behind the process. It is important to realize that PAT is not only the
straightforward introduction of additional analytical techniques into a process but
also the development of methods to predict future behavior according to given settings
of the critical parameters. That means being able to predict fi nal product
quality. For example, while implementing the process, it is important to explore all
sources of component variation as well as their effect on the fi nished product in
order to select which quality parameters (i.e., attributes) have to be measured for
optimal and realistic process control.
Science, engineering, and control technologies can provide a very high level of
process understanding and control capability. A process is well understood when all
FIGURE 1 In - line, online, and at - line process measurements.
Spectrometer Spectrometer Spectrometer
Reactor
Inline Online Atline
Sampling
Process flow
BASIC CONCEPTS AND IMPACT 357
critical sources of variability are identifi ed and explained. The process should be
robust enough to manage this variability. It is also expected that critical quality
attributes can be accurately and reliably predicted in an adequate design space
when other unexpected variables are encountered (e.g., change of raw material
supplier).
4.2.1.6 Changing Current Practice Using PAT
An approach integrating R & D and manufacturing will enhance process understanding
and make acceptable risk management possible. By establishing transferable
process models, it will be possible to develop and implement adequate measurement
technologies that match process needs rather than vice versa. More effi cient and
cost - effective technology transfers will facilitate process knowledge, continuous
process verifi cation, and compliance, thereby enhancing fi nal product quality. Better
process understanding makes it possible to operate by continuous process verifi cation
instead of three - batch validation. Measurement technique selection and integration
occur very early. Accumulated pertinent knowledge is readily available
through data - mining techniques to confi rm or control processing. A series of dynamic
closed control/compliance loops at the process steps identifi ed as critical will increase
confi dence in fi nal product quality. In addition knowledge accumulated over time
will provide a basis for immediate and rapid intervention in the event of deviation
or failure.
A typical illustration of a PAT approach to quality improvement is the use of
near - infrared spectroscopy (NIRS) to qualify excipients and active principles just
before they enter the production process, for example, in dispensing. As discussed
in the next part, near - infrared (NIR) spectra are informative about product structure
and overall quality. Because with substances such as excipients the quality
range was investigated at some time in the past and fi xed into a calibration, NIR
measurement can provide simultaneous nondestructive confi rmation of the predominant
physical and chemical parameters. This is an effective method of reducing
uncertainties about possible causes of failure or poor quality during production.
Each time a given excipient fails its quality requirements at the moment of use,
immediate action can be taken. Control is possible before the risk of failure is
increased. Such an approach is complementary to container - wise identifi cation of
materials on delivery to a warehouse.
4.2.1.7 Promoting Physical Pharmacy and Pharmaceutical Sciences
Process analytical technology supposes a more science - based approach to pharmaceutical
processes. As a matter of fact, it underlines the observed weakness in formal
knowledge of the physical phenomena behind pharmaceutical processes. The physics
is less well understood than the chemistry. Conventional physics has moved increasingly
into the fi eld of activity of engineers and technologists. Formal approaches are
lacking. As a consequence, much highly valuable knowledge of physical phenomena
is dispersed across various disciplines. Expertise in physics is often purely technological
rather than being formalized and integrated into a specifi c discipline.
Just as the boundaries of physics and chemistry once merged to create
physical chemistry, there is an opportunity now for assembling complementary
358 PROCESS ANALYTICAL TECHNOLOGY
scientifi c knowledge from various disciplines. It is a major challenge to improve
understanding through in - depth investigation of the physical phenomena behind
pharmaceutical processes. This objective motivates the enforcement of physical
pharmacy to improve process understanding through a grounding in theoretical
physics.
One major issue is the science and technology of solid particles and powders:
characterization, size and shape analysis, processing understanding, and so forth.
Others include particle formation and fl uid – particle separation, mixture stability,
and understanding and simulating the dynamics of powder mixtures. For example,
the compaction state of powders and mixtures may change rapidly depending on
storage time and conditions. Time to use is not always under control and unexpected
changes may occur. Stirring a mixture of two free - fl owing powders of different size
may result in segregation rather than improved mixture quality. The fl ow properties
of powders depend not only on intrinsic characteristics of the different materials,
such as particle size distribution, particle shape, and surface properties, but also on
external conditions, such as humidity or compaction status. Further areas of interest
include liquid drops, emulsions and colloids, bubbles, and polymers, as well as
surface properties, surface analysis, interfacial and electrostatic phenomena, surface
reactivity, wet chemistry properties, and solubility.
4.2.1.8 Data Mining
Complex processes generate large volumes of data over time. As ever - increasing
volumes are collected and stored, the gap between buried information and usable
accessible knowledge can quickly expand if care is not taken. Data mining extracts
new knowledge out of accumulated observations and thus provides a basis for decision
making and action. How to turn understanding of buried knowledge to best
use? How to extract operational feedback from preexisting, but latent, dormant
empirical knowledge? Such questions precede any data - mining project.
As a multidisciplinary technique, data mining sits at the interface between statistics,
mathematics, and computer science. It is a collection of methods for detecting
regularities and patterns and for extracting knowledge from massive databases
using conventional and advanced analytical tools. Another approach to data mining
is to view it as the multivariate modeling of a real environment on the basis of
multidimensional and accumulated historical data. Thus, data mining is similar to
explorative data analysis. It is driven by the data itself. However, it must be considered
as different from conventional statistics due to the huge volume of processed
data, far above the megabyte scale. Beyond this critical database dimension, most
conventional statistical packages exceed their operational limit. Data mining can
also be performed without the help of professional statisticians. It runs according
to semiautomatic procedures, which makes it widely attractive and more likely to
be used in an industrial environment.
Such situations are characteristic of pharmaceutical processes which accumulate
a variety of historical data without consideration of pertinence. Accumulation is
systematic and exhaustive. However, cross - links between data sources or types may
not be established, leading to irrelevant and undetected redundancies. Reliability
of the collected data is not clearly established over time and variations may not be
detected.
BASIC CONCEPTS AND IMPACT 359
4.2.1.9 Data Warehousing
The 1990s saw the development of data warehouses. An ideal data warehouse is a
collection of historical data varying with time, organized by topic, aggregated in a
unique database, and stored in a way that facilitates decision making (Figure 2 ).
Three main functions are required to manage data warehouses. First, the data must
be collected or else accessed by an alternative method, for example, as preexisting
databases or fi les. Second, the data warehouse requires management and control
tools. Only then can the third function operate, namely data analysis for the purpose
of decision making and new knowledge. Dedicated information management tools
mediate all external, operational, and historical data to the warehouse. Decisional
information management components are used to extract and visualize the data
warehouse information. Online analysis processing (OLAP) consists of the real -
time analysis and visualization of the historical data. Data mining involves the
extraction of rules and models constructed from the collected data.
Online analysis processing mainly comprises the interactive exploration of multidimensional
data sets, or data cubes, which are manipulated by operations from
matrix algebra, for example, slice - and - dice, roll - up, and drill - down. Computing performance
is related to data warehouse size and also data quality, for example,
missing data, unsharpness, and redundancy. The multidimensionality issue is critical
for extracting pertinent information and selecting the results to be stored and
visualized.
The data - mining tools now incorporated in much commercial software are a set
of techniques and algorithms for exploring large databases in order to extract
semantic links pertinent to event explanation and new knowledge acquisition. The
more general goal of data mining is to extract rules and models for understanding
connections and assisting decision making. There are numerous fi elds of application:
risk analysis, manufacturing trends, raw material management, maintenance, process
validation, development, quality control, and so forth. The idea behind data mining
consists in introducing or proposing rules associated with likelihood coeffi cients
established from a large set of existing (i.e., historical) data. The techniques used
FIGURE 2 Schematic structure of a data warehouse.
Data
warehouse
External data sources Internal data sources Operational data sources
Data-mining
modeling of rules
prediction
OLAP
Data analysis
visualization
Transformation or
pretreatments
360 PROCESS ANALYTICAL TECHNOLOGY
are drawn from the fi elds of artifi cial intelligence and numerical and statistical data
analysis, for example, functional modeling, learning machines, neuronal networks,
Bayesian networks, support vector machines, modeling of associations, and explanatory
rules, classifi cations, and segmentations. Their computing complexity derives
from the dramatic up scaling from database to data warehouse level (from megabase
to petabase, i.e., 10 6 . 10 15 ).
4.2.1.10 Data - Mining Methods for Pharmaceutical Processes
The data warehouse is a central repository of data accumulated over time from
various origins: quality control, quality assurance, production, development, and the
like. The accumulated data represent a potential gold mine, conferring competitive
advantage by facilitating understanding of pharmaceutical process and optimizing
it in the light of buried empirical knowledge.
Data mining is used to extract previously unexploited data and knowledge. Its
potential for acquiring knowledge and generating explanatory rules can overcome
the loss of data or underused accumulated data. There are two ways of proceeding.
The fi rst is proactive or directed, for example, hypothesis testing. Particular groupings
or features are suspected, and verifi cation or confi rmation of identity is sought.
The second is reactive or undirected, consisting of simple data exploration. Groupings
are unknown, properties undetected or latent, and patterns unidentifi ed. Alternative
terms for these approaches are supervised and unsupervised learning,
respectively. Top - down and bottom - up approaches complement one another. For
example, the confi rmatory tools of supervised learning can be used to verify and
certify the quality of the discoveries obtained using the exploratory approach.
What can be obtained using data - mining tools? Here is a short list of achievable
goals:
• Data characterization to extract or determine descriptors or indicators, for
example, by generalizing, summarizing, or grouping
• Establishment of associative and explanatory rules
• Classifi cation (supervised learning) of items or objects in classes according to
a given probability
• Clustering of data items (unsupervised learning) in classes, after establishing
class limits inductively from existing data sets
• Detection of similarities in time series
• Pattern recognition
Data from external and internal sources is integrated, aggregated, or associated in
time series. Data items may contain errors or the data may be missing, unsharp,
redundant, or contradictory. A language with operators and variables is required to
establish models. Validity levels also have to be defi ned using suitable optimization
and validation criteria. In addition, a search method is required to extract the data
from the data warehouse and prepare it for analysis.
Data mining can, therefore, be considered as a three - step operation. Prior to any
analysis, the collected data is preprocessed to integrate the warehouse, and some
verifi cation is performed to maintain the data level: for example, integration,
BASIC CONCEPTS AND IMPACT 361
aggregation, or grouping of data from different internal and external sources. The
data is then selected and data mining performed applying the appropriate algorithms
or models. Results are visualized and interpreted for experts in the fi eld.
4.2.1.11 Data - Mining Practice
Data mining is part of an action process known as a business intelligence chain
(Figure 3 ). Data mining is a fl exible solution to the recurrent problem of how to
derive knowledge from data. The source for data mining is the existence of a large
but buried data set. The corresponding data analysis is an intellectual method that
applies only if integrated into the current operational process. Hypothesis testing,
knowledge acquisition, and the generation of explanatory rules are directed by
active collaboration between different process actors. Data mining is teamwork that
requires expertise in various areas, such as information technology (IT), database
management, and data analysis. However, the methods are available in commercial
packages and may not require the expertise of traditional statisticians. It is the
computer which is responsible for discovering patterns or identifying rules or features.
In summary, data mining is a logical loop involving the following steps:
• Business understanding
• Precise setting of the data - mining project, for example:
Defi nition of realistic objectives
Field of treated data
Inventory of available or usable data
• Data preparation
Extraction from internal or external sources
Verifi cation and correction
Pretreatment
• Warehouse construction
• Modeling, for example:
Description and visualization
Affi nity grouping
Rules of association, explanatory rules
Clustering
FIGURE 3 Place of data mining in the decision chain.
Data Information Knowledge Decision Action
Business intelligence chain
Data mining
0
0
0
0
0
0
0
0
0
0
362 PROCESS ANALYTICAL TECHNOLOGY
Classifi cation
Estimation
Prediction
• Evaluation and comparison of models
• Documentation and presentation of results
• Deployment for action
• Back to business understanding.
4.2.1.12 Comments about Data Mining
Data mining provides an explanatory analysis from a confi rmatory analysis. It is
tempting to extract maximal value from available resources such as any kind of
accumulated data. But maximal effi ciency requires critical insight into the expertise
actually buried in data collections or warehouses. The goal of data exploration is to
access the buried data to acquire the knowledge that will make explanation, prediction,
or estimation possible. That is why data mining requires team effort from data
specialists, users, information technologists, and specialists in the relevant fi eld (in
this case, pharmaceutical process). It also requires senior management support
throughout the organization. Mining is a matter of good practice according to established
rules but also a challenge for innovative mathematical techniques. Not all
patterns or rules found by data mining are interesting, although the results should
remain logical and actionable by experts in the relevant fi eld. Because the
algorithms involved tend to be complex and the data volume is huge, software
implementation together with the level of information technology are major
considerations.
Data mining is driven by the accumulated data but always directed at solving a
process, business, or research problem. The results are designed to make it easier to
reach a diagnosis or make a decision. They are only likely to be useful in context:
that is, they are not simply numbers and graphics but an aid to insight for experts
in the relevant fi eld. Also, no single mining technique is equally applicable. A range
of different methods or algorithms should be considered, as no one particular technique
will work equally well or outperform all other techniques on all problems.
Nor will the value of an analytical technique exceed that of the data upon which it
is based.
4.2.1.13 PAT Methods
Almost any existing analytical method can serve the objectives of PAT. Many online
applications already exist. With newer techniques, like NIR imaging or matrix -
assisted laser desorption/ionization time - of - fl ight (MALDI - TOF) mass spectrometry,
there are technological problems about performing online or inline analytics.
Implementing a given analytical technique close to or during process does not
always provide better process understanding. Attributes which are not informative
should not be measured at all and are not worth the burden of complex process
implementation.
Use of the various techniques listed in Table 1 depends on process requirements.
The validity of a given technique or analytical application is challenged by every
0
0
0
TABLE 1 Analytical Methods for PAT
Method Description Online Application Chemical Identifi cation
Pharmaceutical Application
Examples
Infrared,
near - infrared,
and
Raman spectroscopy
Vibrational spectroscopy
(discussed in this chapter)
.
.
Reaction monitoring
Polymorphism
Content determination
Process monitoring (drying,
granulation, blending)
Hyperspectral imaging
Vibrational spectroscopy
coupled with a spatial
analysis (cf. chemical
imaging chapter)
.
Chemical compound distributions
Counterfeit detection
UV – Vis spectroscopy Photoelectron spectroscopy
.
.
Color measurement
Dissolution testing
Cleaning validation (ppm - level
detection)
Terahertz spectroscopy Far - infrared spectroscopy;
3D imaging
.
Polymorphism
Coating integrity and thickness
API distribution possible
Laser - induced breakdown
spectroscopy
Plasma generated by a laser
pulse and detection of the
emitted light (destruction
of sample)
.
Drug development
Process troubleshooting
Laser diffraction Interaction of a laser beam
with particles and
detection of the scattered
light
.
Particle size determination
Effusivity Combines thermal
conductivity, density, and
heat capacity
.
Mixing,
blending,
granulation
monitoring
Acoustic methods
Active or passive
.
Solid,
semisolid,
and high viscose
sample
High shear granulation
monitoring Crystallization
monitoring
363
364 PROCESS ANALYTICAL TECHNOLOGY
technological advance or new analytical technique. Innovation continuously drives
optimization of overall process performance.
4.2.1.14 Conclusion
Process Analytical Technology can be viewed as a constellation placing greater or
less emphasis on a given activity depending on the current problem or situation
(Figure 4 ). There is no written rule or straightforward path to progress through PAT.
Experience and expertise are necessary, together with a good knowledge of the
pharmaceutical environment. Once a pharmaceutical company has decided
to implement PAT, continuous management support for the development and
maintenance of PAT - related activities is critical. It is a strategic and necessary step
for the future success of PAT to encourage, stimulate, and initiate scientifi c collaboration
and interaction as well as the relevant education and training. Better understanding
and control of chemical and pharmaceutical processes are greatly needed,
as well as the development of advanced measurement tools and data analysis
methods.
A summary of PAT benefi ts follows:
• Immediate action if quality is not met
• Better process control and understanding
• Less uncontrolled variation and less production waste
• Better and more stable products
• Data collection and improved historical knowledge
Process analytical technology continuously improves product quality, extends the
acquired knowledge base for new projects, and shortens time to market.
FIGURE 4 PAT constellation (DoE, design of experiments).
DoE
Risk
analysis
Analytical
methods
PAT
Chemometrics
Data
mining
Physical
pharmacy
Pharmaceutical
sciences
Sensor
technology
Pharmaceutical
technology
4.2.2 VIBRATIONAL SPECTROSCOPY
4.2.2.1 Introduction
Modern infrared (IR) spectroscopy is a versatile tool applied to the qualitative and
quantitative determination of molecular species of all types. Its applications fall into
three categories based on the spectral regions considered. Mid - IR (MIR) is by
far the most widely used, with absorption, refl ection, and emission spectra being
employed for both qualitative and quantitative analysis. The NIR region is particularly
used for routine quantitative determinations in complex samples, which is of
interest in agriculture, food and feed, and, more recently, pharmaceutical industries.
Determinations are usually based on diffuse refl ectance measurements of untreated
solid or liquid samples or, in some cases, on transmittance studies. Far - IR (FIR)
is used primarily for absorption measurements of inorganic and metal - organic
samples.
Within the electromagnetic spectrum (Figure 5 ), the IR region ranges from 12,800
to 10 cm . 1 or from 0.78 to 1000 . m. The IR domain is conveniently subdivided into
NIR, MIR, and FIR, respectively, with the following limits:
Near 0.78 – 2.5 4000 – 12,800
Mid 2.5 – 50 200 – 4000
Far 50 – 500 20 – 200
Methods and applications differ with the IR subregion considered. Academia and
analytical chemists commonly consider MIR as the default region of interest.
Current MIR instruments are completely different from traditional grating spectrophotometer
technology. The generalization of Fourier transform (FT) – based spectrometers
in the early 1980s lowered instrument prices and increased the number
and types of MIR applications, in particular thanks to the use of interferometers in
improving signal - to - noise ratios and detection limits. IR applications were originally
limited to qualitative organic analysis. Almost from the outset, absorption MIR
became a well - established application for structure elucidation. Organic chemists
were trained in the visual and direct interpretation of MIR spectra. Nowadays mid -
IR spectroscopy (MIRS) tends to be more viewed as a useful tool for the quantitative
analysis of complex samples by absorption and emission spectrometry, which
may require calibration and data pretreatment.
Near - IR measurements can be performed similarly to those using dedicated
ultraviolet (UV) or visible spectrophotometers. Historically, the most important
FIGURE 5 Limits and designation of the spectroscopic domains.
cm-1 3.3 20 200 4 000 12 500 25 000 10 5
Far Middle Near
Microwaves IR VIS UV
mm 3 000 500 50 2.5 0.8 0.4 0.1
VIBRATIONAL SPECTROSCOPY 365
366 PROCESS ANALYTICAL TECHNOLOGY
application was quantitative analysis in the food and feed industries. Only more
recently have the chemical and pharmaceutical industries shown increasing interest
in the NIR range. The major reason for the delay is in the type of information
delivered. All observed bands result from overtones or combinations of overtones
originating in the fundamental MIR region of the spectrum. Because the measurement
method is nondestructive, samples are measured with little or no specifi c
preparation. NIR spectra contain chemical and physical information on the sample.
Direct interpretation is limited, if not impossible, meaning that multivariate data
processing is routinely required to extract the relevant information. This led most
analytical chemists to ignore the potential of NIR. Until the early 1990s, NIR spectrophotometers
tended to be the dispersive type based on diffraction gratings. Subsequent
technological advance has brought FT and diode array instruments. Filter
instruments remain used for ultrarapid measurement of material composition in the
food and feed industries.
Being at the edge of the IR region, FIR is believed to have less industrial potential.
This is partly due to unresolved experimental and technological diffi culties. FIR
may provide relevant information, but at the cost of disproportionate effort. Routine
use in the pharmaceutical environment is not anticipated in the near future, and for
this reason we shall not discuss FIR further.
The most recent developments in IR/NIR technology include imaging large sample
surfaces, nondestructive analysis of solids by attenuated total refl ectance (ATR), and
photoacoustic measurement. Instrument performance continues to increase, with particular
respect to reliability and modularity. Spectrometer downsizing, speed of measurement,
and mobility no longer represent critical challenges. However, what has
really expanded the scope of MIR applications, and use of the full NIR region, has
been the constant increase in computing power. The fi eld of application of IR spectroscopy
is moving toward the quantitative analysis of complex samples in various
measurement modes. These types of samples are characteristic of the pharmaceutical
industry. Noninvasive spectral sampling using light probes is at last making in situ
analytics attractive, for example, for performing online real - time measurements.
Infrared microscopy was introduced in the early 1980s. Two microscopes, an
ordinary optical microscope and an FT IR instrument with refl ection optics, were
combined. The optical microscope is used to visually locate the spot of interest. The
spot is then irradiated with the IR or NIR beam. There are numerous applications
for noninvasive measurement, including of contaminants, particles, imperfections,
and for fi ber identifi cation. Chemical imaging systems (CIS) are a refi nement of the
technique. Spectra are collected from adjacent areas (pixels) on a larger surface. In
practice, an imaging breakthrough became possible after moving away from pixel -
after - pixel scanning. CIS fl exibility and speed of acquisition improved with the
introduction of new detectors, for example, focal plane array (FPA) detectors. Multiple
IR/NIR spectra (up to many thousand) are scanned in a single step on the
sample surface. With image analysis algorithms and fast computers, current NIR/IR
imaging techniques hold fresh promise for resolving quality problems.
4.2.2.2 IR Spectroscopy Theory
In a typical IR absorption spectrum of an organic substance (Figure 6 ), the ordinate
is transmittance and the abscissa is the wavenumber. A linear wavenumber scale is
preferred because of the linear relationship between wavenumber and energy and
frequency. The frequency of an absorbed radiation is the molecular vibrational frequency
actually responsible for the observed absorption.
Infrared absorption, emission, or refl ection for molecular species can be explained
by assuming transitions from one rotational or vibrational energy state to another.
IR radiation is not energetic enough to produce electronic transitions similar to
those resulting from UV, visible (Vis), or X - ray radiation. Absorption of IR radiation
is limited to molecular species with small energy differences between various
vibrational and rotational states. In order to absorb IR radiation, a molecule must
undergo a net change in dipole moment as a consequence of its vibrational or rotational
motion. Under these circumstances an alternating electrical fi eld interacts
with the molecule and causes changes in the amplitude of one of its motions. The
dipole moment is determined by the magnitude of the charge difference and the
distance between the two charge centers. In addition, regular fl uctuation in dipole
moment occurs, and a fi eld is established which interacts with the electrical fi eld
associated with the incident radiation. If the radiation frequency exactly matches a
natural vibrational frequency of the molecule, a transfer of energy takes place that
changes the amplitude of molecular vibrations and absorption of radiation results.
Similarly, the rotation of asymmetric molecules around their centers of mass results
in periodic dipole fl uctuations which interact with radiation. Homonuclear species
are not concerned and such compounds cannot absorb in the IR.
The amount of energy required to cause a change in energy level is approximately
equivalent to radiation of 100 cm . 1 or less. The relative positions of atoms in a molecule
fl uctuate continuously, and multiple types of vibrations and rotations about
the bonds in the molecule are possible. Exact analysis of all movements becomes
FIGURE 6 Typical example of an infrared absorption spectrum.
55
60
65
70
75
80
85
90
95
%T
500 1000 1500 2000 2500 3000 3500
Wavelength ( cm–1)
VIBRATIONAL SPECTROSCOPY 367
368 PROCESS ANALYTICAL TECHNOLOGY
impossible for molecules comprising several atoms. Not only do larger molecules
have more vibrating possibilities, but intercenter interactions occur that must be
taken into account. Vibrations may be of the stretching and bending variety. Stretching
vibration involves a continuous change in interatomic distance along the axis of
the bond between the atoms. Bending vibration is characterized by a change in the
angle between two bonds and comes in four types: scissoring, rocking, wagging, and
twisting. All vibration types may be possible in a molecule containing more than
two atoms. In addition, vibration interaction or coupling may occur if the vibrations
involve bonds to a single central atom with a change in the characteristic of the
vibrations concerned.
4.2.2.3 Mechanical Model of IR Vibration
Infrared spectra result from light absorption by organic molecules. The easiest way
to describe vibrational spectroscopy from a theoretical perspective is to consider
the isolated vibrations of a mechanical model called the harmonic oscillator. Atomic
stretching vibration behavior can be approximated by a mechanical model consisting
of two masses, m 1 and m 2 , connected by an ideal spring. Displacement of one
such mass along the spring axis results in harmonic motion. Many fundamental frequencies
may be calculated by assuming that band energies arise from the vibration
of the ideal diatomic harmonic oscillator (Figure 7 ), obeying Hooke ’ s law, that is,
. . = 1
2
k
u
where . is the vibrational frequency, k the classical force constant, and
u mm m m = + ( ) 1 2 1 2 , the reduced mass of the two atoms.
The model provides a good description of true diatomic molecules and is not far
from the average value of two atoms stretching within a polyatomic molecule. The
corresponding potential - energy curve is the typical parabola illustrated in Figure 8 .
This approximation gives the average vibration frequency of the bond. For example,
the reduced masses for C — H, O — H, and N — H are 0.85, 0.89, and 0.87. These
fi gures are similar, so the frequencies would be quite similar too. However, the
electron - withdrawing and - donating properties of neighbors within molecules act
FIGURE 7 Ideal diatomic harmonic oscillator.
x
x2
0
requilibrium
r
x
m1 m2
x 1
2 1
on the observed band strength, length, and frequency. An average value is of little
use in structural determinations and these differences cause a real spectrum to
develop. The force constant k is a measure of the stiffness of the chemical bond and
is the equivalent of the force constant of the spring in the harmonic model. The k
values vary widely and cause energy differences which can both be calculated and
utilized in spectral interpretation. It has been possible to evaluate some force constants
for various types of chemical bonds by IR spectroscopy. Generally, k has been
found to range between 3 . 10 2 N/m and 8 . 10 2 N/m for most single bonds (average:
5 . 10 2 N/m). Double and triple bonds are found to have k values two and three
times this average, respectively. In practice, these average experimental values can
be used to estimate the wavenumbers of fundamental absorption peaks, that is,
peaks of the transition from the ground state to the fi rst excited state, for a variety
of bond types.
Classical mechanics does not apply to the atomic scale and does not take the
quantized nature of molecular vibration energies into account. Thus, in contrast to
ordinary mechanics where vibrators can assume any potential energy, quantum
mechanical vibrators can only take on certain discrete energies. Transitions in vibrational
energy levels can be brought about by radiation absorption, provided the
energy of the radiation exactly matches the difference in energy levels between the
vibrational quantum states and provided also that the vibration causes a fl uctuation
in dipole.
4.2.2.4 Quantum Mechanical Model
Unlike the classical spring model for molecular vibrations, there are not an infi nite
number of energy levels. Instead of a continuum of energies, there are discrete
energy levels described by quantum theory. The time - independent Schr o dinger
equation is solved using the vibrational Hamiltonian for a diatomic molecule. Values
for the ground state ( . = 0) and succeeding excited states can be calculated by
solving the equation (Figure 8 ). Absorption of a photon of the correct energy can
cause the molecule to change between vibrational energy levels. At room temperature
only the ground state has a signifi cant population, and so transitions due to
absorption at these temperatures occur from the ground state. Transitions between
ground state to energy level 1 give the fundamental absorption if this leads to a
FIGURE 8 Energy diagram of the ideal diatomic oscillator.
Potential energy
V=0
V=1
V=2
V=3
Interatomic distance
Energy
level
VIBRATIONAL SPECTROSCOPY 369
370 PROCESS ANALYTICAL TECHNOLOGY
change in molecular dipole moment. Transitions between ground state and energy
level 2 or above give overtones. Transitions between multiple states can occur and
give rise to combination bands.
A simplifi ed version of the energy levels may be written for the energy levels of
a diatomic molecule:
E
h k
u . .
.
. = + ( ) = 1
2 2
0, 1, 2, . . .
in which Hooke ’ s law terms can be seen. Rewritten using the quantum term
hV h k u =( ) 2. , the equation reduces to
E hV . . . = + ( ) = 1
2
0, 1, 2, . . .
In the case of polyatomic molecules, the energy levels become quite numerous.
Ideally, one can treat such a molecule as a series of diatomic, independent, harmonic
oscillators and the above equation can be generalized:
E hV
i
N
i i . . . . . . . 1 2
1
3 6
1
1
2
0 , , , . . . , , , . . . , 1, 2, 3, . . . 3 23 ( )= + ( ) =
=
.
.
Any transition of an energy state from 0 to 1 in any one of the vibrational states
( . 1 , . 2 , . 3 , . ) is fundamental and allowed by selection rules. Where the transition
is from the ground state to . i = 2, 3, and so on and all others are zero, it is known
as the fi rst overtone, the second overtone, and so on. Transitions from the ground
state to a state for which . i = 1 and . j = 1 simultaneously are combinations. Other
combinations, such as . i = 1, . j = 1, . k = 1, or . i = 2, . j = 1, and so forth are also
possible. Typically, NIR spectra will contain these overtones and combinations
derived from the fundamental vibrations which appear in the MIR. Overtones and
combinations are not allowed, but appear as weak bands due to anharmonicity or
Fermi resonance. As a rule, overtones occur at one - half and one - third of the fundamental
absorption wavelength or 2 and 3 times the frequency. The majority of
overtone peaks arise from the R — H stretching and bending modes because the
dipole moment is high: O — H, C — H, S — H, and N — H are strong NIR absorbers
and form most NIR bands. Since most absorption is repeated in the NIR range, this
region is likely to be used to identify a molecule, as with MIR. As a consequence,
IR bands are traditionally used to identify functional groups which have characteristic
frequencies. NIR spectra are more overlapping, and, although bands can be
identifi ed, they cannot be placed in relation to the rest of the molecule. NIR spectra
are, therefore, mainly used to confi rm the identity of a material, as for true
identifi cation.
As given from the quantum mechanics equations, the energy for transition from
energy levels 1 to 2 or 2 to 3 should be identical to that for transition from 0 to 1.
Furthermore, quantum theory states that the only transitions that can take place
are those for which, according to vibrational quantum theory, the vibrational
quantum number changes by unity. This is the so - called selection rule.
So far we have illustrated the classic and quantum mechanical treatment of the
harmonic oscillator. The potential energy of a vibrator changes periodically as
the distance between the masses fl uctuates. In terms of qualitative considerations,
however, this description of molecular vibration appears imperfect. For example, as
two atoms approach one another, Coulombic repulsion between the two nuclei adds
to the bond force; thus, potential energy can be expected to increase more rapidly
than predicted by harmonic approximation. At the other extreme of oscillation, a
decrease in restoring force, and thus potential energy, occurs as interatomic distance
approaches that at which the bonds dissociate.
In theory, the wave equations of quantum mechanics can be used to derive near -
correct potential - energy curves for molecular vibrations. Unfortunately, the mathematical
complexity of these equations precludes quantitative application to all but
the very simplest of systems. Qualitatively, the curves must take the anharmonic
form. Such curves depart from harmonic behavior by varying degrees, depending
on the nature of the bond and the atom involved. However, the harmonic and
anharmonic curves are almost identical at low potential energies, which accounts
for the success of the approximate methods described.
Anharmonicity leads to deviations of two kinds. At higher quantum numbers, . E
becomes smaller, and the selection rule is not rigorously followed; as a result, transitions
of . ± 2 or ± 3 are observed. Such transformations are responsible for the
appearance of overtone lines at frequencies approximately two or three times that
of the fundamental line; the intensity of overtone absorption is frequently low, and
the peaks may not be observed. Vibrational spectra are further complicated by the
fact that two different vibrations in a molecule can interact to give absorption peaks
with frequencies that are approximately the sums or differences of their fundamental
frequencies. Again, the intensities of combination and difference peaks are generally
low.
It is ordinarily possible to deduce the number and kinds of vibrations in simple
diatomic and triatomic molecules and determine whether these vibrations contain
several types of atoms as well as bonds; for these molecules, the multitude of possible
vibrations gives rise to IR spectra that are diffi cult, if not impossible, to analyze.
The number of possible vibrations in a polyatomic molecule can be calculated as
follows. Three coordinates are needed to locate a point in space; fi xing N points
requires 3 N coordinates. Each coordinate corresponds to one degree of freedom for
one of the atoms in a polyatomic molecule; for this reason, a molecule containing
N atoms is said to have 3 N degrees of freedom. A molecule features three types of
motion. First, the motion of the entire molecule through space; second, the rotational
motion of the entire molecule around its center of gravity; and, third, the
vibrations of each of its atoms relative to the other atoms. Since all atoms in the
molecule move in concert through space, defi nition of translational motion requires
three of the 3 N degrees of freedom. Another 3 degrees of freedom are needed to
describe the rotation of the molecule as a whole. The remaining 3 N . 6 degrees of
freedom involve interatomic motion and hence represent the number of possible
vibrations within the molecule. In a linear molecule 2 degrees of freedom suffi ce to
describe rotational motion. Thus, the number of vibrations for a linear molecule is
3 N . 5. Each of the 3 N . 6 or 3 N . 5 vibrations is a normal mode. For each normal
mode of vibration there is a potential energy relationship. In addition, to the extent
that a vibration approximates harmonic behavior, the differences between the
VIBRATIONAL SPECTROSCOPY 371
372 PROCESS ANALYTICAL TECHNOLOGY
energy levels of given vibrations are the same; that is, a single absorption should
appear for each vibration in which there is a change in dipole.
However, fewer experimental peaks may be observed than would be expected
from the theoretical number of normal modes. Fewer peaks can be found when the
symmetry of the molecules is such that no change in dipole results from a particular
vibration. The energies of two or more vibrations can be identical or nearly identical.
In some cases absorption intensity is too low to be detected by ordinary means. It
may also happen that the vibrational energy is in a wavelength region which is
beyond the range of the instrument.
Conversely, more peaks may be found than expected from the number of normal
modes. This is the typical situation that concerns the NIR domain. Overtone peaks
at two or three times the frequency of a fundamental peak, or addition combination
bands at approximately the sum or difference of two fundamental frequencies, are
sometimes encountered. The energy of a vibration and thus the wavelength of its
absorption peak may be infl uenced by, or coupled with, other vibrators in the molecule.
A number of factors infl uence the extent of such coupling. Vibration coupling
is a common phenomenon. As a result, the position of an absorption peak corresponding
to a given organic functional group cannot always be specifi ed exactly.
While interaction effects may lead to uncertainties in the identifi cation of functional
groups contained in a compound, it is this very effect that provides the unique features
of an IR absorption spectrum that are so important for the positive identifi cation
of a specifi c compound.
4.2.2.5 Anharmonicity
The ideal harmonic oscillator is a somewhat limited model. As the oscillating masses
get very close, real compression forces — which are neglected in calculations — fi ght
against the bulk of the spring. As the spring stretches, it eventually reaches a point
where it loses its shape and fails to return to its original coil. This ideal case is shown
in Figure 9 . The barriers at either end of the cycle are approached in a smooth and
orderly fashion. Likewise, in molecules, the respective electron clouds of the two
bound atoms limit approach by the nuclei during the compression step, creating an
energy barrier. At extension of the stretch, the bond eventually breaks when the
vibrational energy level reaches the dissociation energy. The barrier at smaller dis-
FIGURE 9 Energy diagram of the anharmonic diatomic oscillator.
Interatomic distance
Potential energy
V=0
V=1
V=2
V=3
Energy
level
tances increases at a rapid rate, while the barrier at the far end of the stretch slowly
approaches zero (Figure 9 ). The shape of the potential energy curve is typical of an
anharmonic oscillator.
Energy levels in the anharmonic oscillator are not equal, although they become
slightly closer as energy increases. This phenomenon can be seen in the following
equation:
E hW W X e e e . . . = + ( ) . + ( ) + 1
2
1
2
2
higher terms
where W Ku e e = ( )
1
2
1 2 . is the vibrational frequency, W e X e the anharmonicity
constant, K e the anharmonicity force constant, and u the reduced mass of the two
atoms. In practice, anharmonicity is between 1 and 5%. Thus, the fi rst overtone of
a fundamental vibration set, for example, at 3500 nm would be
. = + .[ ] ( )
3500
2
3500 0 01 . , 0.02, . . .
Depending on structural or steric conditions, the number may range from 1785 to
1925 nm for this example. However, it would generally appear at 3500/2, plus a relatively
small shift to a longer wavelength. As forbidden transitions, the overtones are
between 10 and 1000 times weaker than the fundamental bands. Thus, a band arising
from bending or rotating atoms would have to be in its third or fourth overtone to
be seen in the NIR region of the spectrum. For example, a fundamental carbonyl
stretching vibration at 1750 cm . 1 or 5714 nm would have a fi rst overtone at approximately
3000 nm, a weaker second overtone at 2100 nm, and a third very weak overtone
at 1650 nm. The fourth overtone, at about 1370 nm, would be so weak as to be
useless. These fi gures are based on an illustrative 5% anharmonicity constant.
The detailed examination of the spectra of simple molecules is a direct source to
determine the characteristic NIR frequencies for selected vibration modes. For
qualitative and quantitative analyses there is the requirement to interpret as much
as possible the NIR spectrum. Although interpretation of spectra in a manner
analogous to MIR is not conceivable, attempts exist to defi ne and categorize
observed NIR frequencies. Examples of reported frequencies for aliphatic hydrocarbons
are given in the following list:
8547 cm . 1 C — H second overtone in CH —— CH
8474 cm . 1 C — H group in cis olefi ns
7700 – 9000 cm . 1 C — H second overtone
8696 cm . 1 Second overtone of CH 2 antisymmetric stretching
8285 cm . 1 Second overtone of CH 2 symmetric stretching
1080 – 1140 cm . 1 Second overtone olefi n
7692, 8237, 8576 cm . 1 C — H stretching second overtone in CH 2
4.2.2.6 Structure Elucidation Using MIRS
Mid - IR absorption and refl ectance spectroscopy is typically used for determining
the structure of organic and biochemical species. When used in conjunction with
VIBRATIONAL SPECTROSCOPY 373
374 PROCESS ANALYTICAL TECHNOLOGY
other analytical methods, such as mass spectroscopy, nuclear magnetic resonance,
and elemental analysis, IR spectroscopy usually achieves positive species identifi cation.
Spectra are obtained after sample preparation, usually involving dilution of
the analyte. Sample handling is the diffi cult and time - consuming part of the analysis.
Organic samples exhibit numerous IR absorption peaks used for qualitative structure
confi rmation. First, presumptive functional groups are identifi ed by examining
their frequency region from about 3600 to 1200 cm . 1 . As mentioned earlier, the frequency
at which an organic functional group absorbs radiation can be approximated
from the atomic masses and bond forces between them. These group frequencies
are not totally invariant because of interactions with other vibrations. However, such
interaction effects are small, and a range of frequencies can be assigned within which
it is highly probable that the absorption peak for a given functional group will be
found. Group frequencies are listed in correlation charts, which serve as a starting
point in the identifi cation process.
Second, the spectrum of the unknown is compared with the spectra of reference
compounds featuring all the functional groups found in the fi rst step. The fi ngerprint
region from 1200 . 1 to 600 cm . 1 is extremely useful because small differences in
structure and constitution produce signifi cant changes in the appearance and distribution
of absorption peaks in this region. Most single bonds give rise to absorption
bands at these frequencies. Because their energies are about the same, strong interaction
occurs between neighboring bonds. The absorption bands are thus composites
of these various interactions and depend upon the overall skeletal structure of
the molecule. Exact interpretation in this region is seldom possible because of spectral
complexity. On the other hand, it is this complexity that leads to uniqueness and
the consequent usefulness of the region in fi nal identifi cation. A close match between
two spectra in the fi ngerprint region constitutes almost conclusive compound
identifi cation.
In employing group frequencies it is essential that the entire spectrum rather
than a small isolated portion be considered and interrelated. Correlation charts
serve only as a guide for further and more careful study. Catalogs of IR spectra that
assist in qualitative identifi cation by providing comparison and reference spectra
for a large number of pure compounds are commercially available on electronic
media. Optimized search systems for identifying compounds from IR spectral databases
and algorithms for the matching step produce rapid and reliable potential
hits.
4.2.2.7 Extending Use of MIRS
Organic and inorganic molecular species (except homonuclear molecules) absorb
in the IR region. IR spectroscopy has the potential to determine the identity of an
unusually large number of substances. Moreover, the uniqueness of a MIR spectrum
confers a degree of specifi city which is matched or exceeded by relatively few other
analytical methods. This specifi city has found particular applications for the development
of quantitative IR absorption methods. However, these differ from quantitative
UV/Vis techniques in their greater spectral complexity, narrower absorption
bands, and the technical limitations of IR instruments. Quantitative determinations
obtained from IR spectra are usually inferior in quality and robustness to those
obtained with UV/Vis and NIR spectroscopy. In addition, univariate or linear cali
bration curves require meticulous attention to numerous details. One cause of
failure is the frequent nonadherence to Beer ’ s law due to the inherent complexity
of IR spectra, featuring overlapping absorption peaks or disturbance by stray radiation.
Analytical uncertainties cannot be reduced to a level which is comparable to
other methods, despite considerable effort or care.
Diffuse - refl ectance MIRS has found a number of applications for dealing with
hard - to - handle solid samples, such as polymer fi lms, fi bers, or solid dosage forms.
Refl ectance MIR spectra are not identical to the corresponding absorption spectra,
but suffi ciently close in general appearance to provide the same level of information.
Refl ectance spectra can be used for both qualitative and quantitative analysis. Basically,
refl ection of radiation may be of four types: specular, diffuse, internal, and
attenuated total.
Specular refl ection is encountered when the refl ecting medium is a smooth polished
surface. The angle of refl ection is identical to the incident angle of the radiation
beam. If the surface is IR absorbent, the relative intensity of refl ection is less
for wavelengths that are absorbed than for wavelengths that are not. Thus, the plot
of refl ectance R , defi ned as the fraction of refl ected incident radiant energy versus
the wavelength (or wavenumber) appears similar to a transmission spectrum for the
sample.
Diffuse - refl ectance spectra are obtained directly from powder samples after a
minimum of preparation. In addition to the time saved, measurement is nondestructive,
leaving the sample intact for further analysis. The widespread use of diffuse
refl ectance was only possible with the introduction of the FT technique. Refl ected
radiation from powders is too low to be measured at medium resolutions or inadequate
signal - to - noise ratios. Diffuse refl ectance (Figure 10 ) occurs when a beam of
radiation strikes the surface of a fi nely divided powder. With this type of sample,
specular refl ection occurs at each plane surface. However, since there are many of
these surfaces and they are randomly oriented, radiation is refl ected in all directions.
The intensity of the refl ected radiation is independent of the viewing angle. If peak
locations are identical in refl ectance and transmittance spectra, relative peak heights
differ considerably. For example, minor transmittance peaks generally appear larger
in refl ectance spectra.
Internal - refl ection spectroscopy is used to obtain IR spectra of hard - to - handle
or hard - to - prepare samples such as solids with limited solubility, fi lms, pastes, adhesives,
and powders. Refl ection occurs when a beam of radiation passes from a denser
to a less dense medium. The fraction of incident beam which is refl ected increases
as the angle of incidence becomes larger. Beyond a certain critical angle, refl ection
is complete. During the refl ection process the beam penetrates a small distance into
FIGURE 10 Diffuse refl ectance, transfl ectance, and transmittance measurements.
Diffuse
reflectance
Transflectance
Diffuse
transmittance
VIBRATIONAL SPECTROSCOPY 375
376 PROCESS ANALYTICAL TECHNOLOGY
the less dense medium before refl ection occurs. The depth of penetration varies
from a fraction of a wavelength up to several wavelengths and depends on the
wavelength of incident radiation, the refraction indices of the two materials, and the
angle of incident beam with respect to the interface.
Attenuated total refl ection (ATR) is the most common refl ectance measurement
modality. ATR spectra cannot be compared to absorption spectra. While the same
peaks are observed, their relative intensities differ considerably. The absorbances
depend on the angle of incidence, not on sample thickness, since the radiation penetrates
only a few micrometers into the sample. The major advantage of ATR spectroscopy
is ease of use with a wide variety of solid samples. The spectra are readily
obtainable with a minimum of preparation: Samples are simply pressed against the
dense ATR crystal. Plastics, rubbers, packaging materials, pastes, powders, solids, and
dosage forms such as tablets can all be handled directly in a similar way.
4.2.2.8 Raman Spectroscopy
When radiation passes through a transparent medium, a fraction of the beam scatters
in all directions. A small fraction of the scattered radiation differs from the
incident beam, showing shifts in wavelength determined by the chemical structure
of the molecules in the medium. The same types of quantized vibrational changes
associated with IR absorption occur, and the difference in wavelengths between
incident and scattered radiations corresponds to wavelengths in the MIR. The
Raman scattering spectrum and IR absorption spectrum for a given species are very
similar. Figure 11 illustrates a typical Raman spectrum. IR is generally the method
of choice, but in some cases Raman spectroscopy offers more information about
certain types of organic compounds. For example, it is sensitive to conformational
FIGURE 11 Example plot of two Raman spectra (two polymorphic forms of an
excipient).
12,000
14,000
16,000
18,000
20,000
22,000
24,000
26,000
28,000
30,000
32,000
34,000
36,000
38,000
400 600 800 1000 1200 1400 1600
and environmental information. Peak overlap in compound mixtures is less likely,
and quantitative determinations are easier. In particular, accurate quantitative
determination can be performed on very small samples. Despite these advantages,
Raman spectroscopy has not yet been exploited due to the rather high cost of the
instruments.
There are differences between the kinds of groups that absorb in the IR and
those that are Raman active. Parts of Raman and IR spectra are complementary,
each being associated with a different set of vibrational modes with |