Development phases and GMP requirements

 

  

GMP rules must be applied as soon as pharmaceutical products are manufactured for human use. The first production for human use usually takes place during development, when clinical samples are manufactured for initial tolerability studies on healthy volunteers (clinical phase I, see figure 16.B-5). A fixed routine procedure for manufacturing and control does not yet normally exist at this point. The law therefore prescribes that the GMP principles are applied "appropriate to the development stage of a drug product".

Before this time, i.e. while still in the research phase and the preclinical development phase, no GMP requirements apply. However, the distinction between research and development is defined very differently in different companies. This is an arbitrary interface with internally defined phase transition criteria that are not GMP-relevant. However, it is still advisable to perform some quality assurance activities already in the preclinical phase.

In the (preclinical) research phase, although there are no legal GMP-requirements, it is still useful to keep proper documentation (laboratory notebooks, log books, raw data) and to apply basic internal quality standards (SOPs, calibration, etc.).

In the preclinical development phase, there are also no legal GMP-requirements. However, in this phase, toxicological investigations are typically performed, which require compliance with GLP standards (see chapter 16.C.1 GLP - Good Laboratory Practice).

In this phase it is also advisable to comply with basic quality standards in order to maintain a reliable and reproducible data basis for compilation of the IND (Investigational New Drug Application) and IMPD (Investigation Medicinal Product Dossier) (see chapter 16.D.1 Prerequisites for the approval of clinical investigations) and the manufacture of clinical samples in the next phase (figure 16.B-2).

GMP standards must be complied with starting from the clinical development phase, because clinical samples for human use are being produced from this point onwards (figure 16.B-3).

The upscaling of the process and the transfer to production are performed in later development phases (e.g. in clinical phases II-III). In this phase, additional requirements apply as named in figure 16.B-4.

How the GMP regulations can be implemented in the individual phases of pharmaceutical development must be determined specifically in each company taking into consideration individual products and procedures.

Clinical studies are defined in four phases (see figure 16.B-5):

• In clinical phase I, the investigational drug product is tested in a relatively small number of healthy volunteers to test how, and in which dose range, an API can be tolerated. This phase also investigates how the medicinal substance acts on the human organism (pharmacodynamics), and conversely, how the human organism acts on the medicinal substance (pharmacokinetics), i.e. how the API is absorbed, distributed in the various organs, and how it is broken down (metabolised) and excreted.
• Clinical phase II is an exciting stage, since this is when the substance is tested in a small number of patients for the first time, to test whether the anticipated effect actually occurs, and within a dose range that is well tolerated.
• The studies of phase III deal with large-scale long-term studies on a large patient collective, often in several different clinics in several countries. As well as efficacy, these studies are also used to test long-term tolerability.
• After a new drug product has been approved, further studies are still required to test long-term tolerance, and for monitoring rare side effects (phase IV).

16.B.1 Formulation development

The aim of formulation development is to bring a new API into a galenic form that can be administered to a human or animal volunteer or patient for the first time. However, formulation development may also involve producing existing drug substances in a new form that is characterised by improved handling, improved stability, improved bioavailability, faster onset of effect, longer period of effectiveness, better compatibility, or similar.

In order to begin development of a new formulation, the formulation scientist needs the prerequisites shown in figure 16.B-6.

For the first formulation experiments, it can be useful to gain as much information as possible on the physico-chemical characteristics of the active pharmaceutical ingredient (and specifically on the batch of API used - as a preliminary specification), in order to select excipients systematically (rationale for the development report, see chapter 16.F Development report) on the one hand, while also enabling the detection of differences to subsequent batches of API, which may result from a changed or optimised synthesis route.

Examples of targeted excipient selection:

• Poor solubility of the API in aqueous media: Use of solubilisers, oily solutions, emulsions, micronised API etc., if a rapid onset of effect is required. If in contrast a depot effect is desired, for example, suspension in an aqueous medium can be beneficial.
• Oxidation sensitivity: Use of antioxidants, leak-tight packaging, inert gas
• Sensitivity to moisture: Use of excipients with low water content, anhydrous formulations, moisture-proof packaging
• High fine fraction of API (for solid dosage forms): Improvement of powder flow properties with suitable excipients and /or granulation or compacting

If the most important physico-chemical properties of the active pharmaceutical substance are known and the dosage form is determined, the relevant excipients are selected and their compatibility with the active pharmaceutical ingredient is analysed. The following are compared:

• Binary mixtures of API each with one excipient
• Mixtures of API with several excipients
• The influence of stabilising agents (antioxidants, complexing agents, preservatives, inert gas, etc.) on mixtures of API and excipient

To identify compatibility differences between these mixtures, low API fractions (for example 0.5-5%) are used and the mixtures are subjected to accelerated stability tests (see chapter 16.B.2 Analytical development and chapter 16.B.3 Manufacturing and testing of stability samples). From the results of these accelerated stability tests, it is then possible to identify suitable excipients and possibly suitable stabilising agents, and hence develop experimental formulations.

These pre-formulations are evaluated in terms of their technical feasibility, short-term stability, dissolution rate (in-vitro dissolution), maximum possible content of active substance etc., and are further optimised if necessary.

At the end of formulation development, a formulation is selected as test formulation to be used in early clinical investigations (phase I/II).

In order to demonstrate a targeted approach to formulation development (see chapter 16.F Development report), it is important to document accurately the results from this development phase listed in figure 16.B-7.

16.B.2 Analytical development

The typical tasks performed by the analytical control laboratory during pharmaceutical development are method development and validation:

• At the start of formulation development, a test method must already be available that is sufficiently sensitive to detect differences in stability in the first compatibility studies.
• A validated method for the determination of the content and purity of active substance in the clinical samples must be available at the time of submission of an IND or IMPD.
• Furthermore, concurrently to the relevant galenic activities, analytical methods must be provided, for example for quality control of new excipients and primary packaging materials, stability tests, in-process controls, and studies relating to process validation and cleaning validation activities.

Compatibility studies

In formulation development, the influence of individual excipients on the stability of the active substance is tested before studies on the stability of more complex formulations are performed (see chapter 16.B.1 Formulation development). In this test, the active substance/excipient mixtures are stored in suitable containers and subjected to stress conditions, such as:

• 4 weeks at 40 °C /75% r.h.
• 4 weeks at 50 °C
• 2 weeks cycle testing (e.g. daytime temperature 50 °C, night 5 °C)

The following should be used as references:

• The individual excipients alone stored under the above stress conditions,
• Fractions of all mixtures should also be stored at room temperature and in the refrigerator or deep-freeze.

The investigation of these compatibility samples not only provides information on the compatibility of excipients with the active substance and hence their feasibility, but also provides information on the suitability of the analysis method. It is easier to attribute problems in sample preparation or recovery to interactions with single excipients than in the analysis of more complex formulations.

Method validation

To ensure that the analytical data obtained during development is reproducible and reliable, the analytical methods used must be validated in the course of development. A stepwise approach is also practical and acceptable in this case:

• All laboratory equipment used for GMP purposes must be qualified; the test equipment used must be regularly calibrated in accordance with a written plan.
• For the release of clinical samples in phase I, all analytical methods including specifications must already be described in the IMPD (see chapter 16.D.1 Prerequisites for the approval of clinical investigations). The EMEA document Guideline on the Requirements to the chemical and pharmaceutical quality documentation concerning investigational medicinal products in clinical trials (CHMP/QWP/185401/2004, referred to in the following as the IMPD guide), requires, for phase I clinical trials, that "the suitability of the analytical methods used should be confirmed. The acceptance limits and the parameters (specificity, linearity, range, accuracy, precision, quantification, and detection limit, as appropriate) for performing validation of the analytical methods should be presented in a tabulated form." However, it does not mean that validation results or a validation report must already exist at this time. Nonetheless, it is strongly recommended that only validated test methods for content and purity are used for the release of phase I clinical samples, since these methods are essential for the reliability of the study.
  Additional test criteria can also be tested using non-product-specific standard methods. Although pharmacopoeia methods from individual monographs count as validated, they must still be tested for suitability for the specific application case (in particular in terms of sample preparation and recovery).
• For phase II studies, in contrast, a tabulated summary of the validation results obtained using ICH methods (described in ICH Q2A and Q2B) is explicitly expected - although a complete validation report is not yet required (see the IMPD Guide). A complete validation report must only be submitted to the authorities on request.
• As the development process progresses, depending on the level of detail of the specifications, the required breadth and depth of the validation of test procedures also increases. Due to limited availability of drug substance or intermediate, investigations into robustness and accuracy are often not possible in the early development phases. However, the reproducibility of measurement results must always be guaranteed, for example, using internal standards.
• During clinical phase III at the latest, the definitive test procedure for the commercial product must be defined. This contains all release-relevant test criteria, the test methods, and the corresponding specifications. At this point, each individual test method must be validated and documented as described in figure 16.B-9.

  Figure 16.B-9 Structure of a test method

  Structure of a test method

  Title of the test Principle (for example, photometric, microscopic, gravimetric determination) Reagents (if applicable, cross-reference to pharmacopoeia) Devices, equipment, and environmental conditions Sample preparation and preparation of standard solutions Method description, including system suitability test (SST) with requirements for: Selectivity Reproducibility Limit of quantification (LOQ) and limit of detection (LOD) Acceptance criteria Calculation formula for the test result Change history

Definition of (preliminary) specifications

Before initial clinical samples are manufactured and analysed, (preliminary) specifications must exist for the release-relevant test criteria of the medicinal product. GMP requires that these are determined in advance so that it cannot be subsequently (when test results are available) and arbitrarily decided whether or not a product batch is suitable for use in clinical studies. Testing of the final product is particularly important in development, since the manufacturing procedures are not yet standardised or properly validated. Legislators therefore place particular emphasis on specifications for

• the accuracy of the therapeutic single dose, such as homogeneity, uniformity of content,
• release of drug substances from the formulation: solubility, dissolution rate,
• estimation of stability and determination of the preliminary storage conditions in the preliminary packaging (see chapter 16.B.3 Manufacturing and testing of stability samples).

In addition to the release specification for the pharmaceutical product, concurrently with the development of formulations, manufacturing processes and analytical methods, the specifications listed in figure 16.B-10 must also be elaborated.

When compiling specifications, the following must be considered:

• The specifications should be appropriate, i.e. they must not permit too much variability, but must also not be too narrow.
• Any changes to specifications required during development must be documented and justified (Product Specification File, see chapter 16.E Documentation and recording of changes during development).
• A written procedure (SOP) must already exist for development that describes which measures to implement if a specification is not met. Problems in meeting specifications must not cause unjustified broadening of specifications, secondary releases, or similar.

Specification of impurities (including by-products, degradation products, residual solvents) in the active substance and in the medicinal product

The detection of impurities that occur normally or occasionally in a product, such as organic or inorganic impurities, residual solvents, by-products, degradation products, intermediate stages and reagents, is an important part of GMP for later control of the routine production process. In order to build up the required database ,impurity profiles must be created during development which record the following details:

• Identity (or at least a characteristic for identification) of each impurity that is present in a quantity of 0.1 % or more
• Identification of specific toxic impurities, even if less than 0.1 %
• Historical data on the quality and quantity of the impurities
• List of substances that occur in a product, even sporadically

The maximum content of impurities must be specified for the release of clinical supplies and in the stability plans.

Sampling

It is important to pay sufficient attention to the method of sampling, even during development. Correct sampling is of particular importance for the result of each test (and therefore sometimes for the further development), since no valid conclusions can be drawn about the total quantity on the basis of non-representative samples. A sample must be representative of the whole batch in terms of:

• Sampling point (samples should also be taken from positions at which it is known that problems may occur)
• Number of samples
• Quantity of samples (for example, is the quantity of samples different for determination of content compared to samples for content uniformity)

Difficulties in taking samples (incorrect sample size/number of samples, unsuitable packaging, non-representative sampling location, unsuitable sampling tools, demixing, intervals, etc. ...) should be identified in development to avoid problems and delays in later phases. Experiences from development are also helpful when creating the sampling plans (see chapter 14.A Sampling) for production.

16.B.3 Manufacturing and testing of stability samples

During the development of new dosage forms, stability tests must be performed with two different aims:

• Estimation of the stability of clinical supplies and storage conditions for the clinical samples. This data is important for the health and safety of study participants as well as for the reliability of the clinical studies.
  These stability tests must already start in the preclinical development phase, and often take place under accelerated or stress conditions. There are no specific GMP requirements for the manufacturing and testing of these stability samples, but tests should be fully documented.
• Collection of stability data in order to define the expiration dating period for the commercial product and prove the suitability of the packaging material under the specified storage conditions. If long-term stability tests are to be performed for submission purposes (registration stability), the respective product batches must be manufactured, packed, and tested in accordance with GMP.

Important results from stability testing are summarised in figure 16.B-11.

Depending on the goal of the stability tests, various approaches can be used:

• Short-term stability (4 weeks/3 months) at room temperature and if applicable, at a slightly increased temperature (for example 25 °C, 60 % r.h. and 30 °C, 60% r.h.): for example, as a prerequisite for initial clinical studies.
• Accelerated stability tests (1 week/3 months) at elevated temperatures/increased humidity (for example, 40 °C, 75 % r.h.) and if applicable, cycle testing at different temperatures (for example, daytime 30 °C, 60 % r.h., night 5 °C): for rough estimation of the expiration dating period and storage conditions
• Long-term stability (approx. 2-3 years, depending on the product) under different storage conditions, to justify the shelf life specified in the marketing application.

In all tests, reference samples should also be stored at refrigerator temperatures (5 °C) and in the deep freeze (-20 °C) .

The tests should also be performed in different containers (for example glass, PE, PP, PVC, Alu, HDPE, PVDC, to provide indications of particularly suitable primary packaging materials.

A detailed description of the tests to be performed, the methods, packaging (incl. number and size), the pull points, test criteria, different product batches to be tested, dosages, package sizes, and rationales for matrixing or bracketing, must be given in the stability plan. Depending on the galenic formulation and specific properties of the active substance, the test criteria listed in figure 16.B-12 may be relevant for stability investigations.

Stability tests for marketing authorisation purposes

In order to support the expiration dating period specified in the submission file for marketing authorisation, certain prerequisites must be fulfilled when performing long-term stability studies (see figure 16.B-13).

16.B.4 Packaging development

Packaging materials are classified into primary and secondary packaging materials. Primary packaging materials come into direct contact with the product and therefore have a significant influence on its quality. Examples of primary packaging materials are blister films, bottles, vials, ampoules, prefilled syringes, infusion bags, bottle packs, stoppers, pouches, tablet tubes, tubes, etc.

Secondary packaging materials include external covers such as folded boxes, as well as trays and labels. Secondary packaging materials are mainly used as mechanical protection and for information purposes (see chapter 13.A Packaging material).

Since primary packaging materials have a significant influence on product quality and product safety, it is important to identify suitable packaging materials by using stability tests concurrently with formulation development. For specific applications, it may even be necessary to develop new packaging principles. For certain dosage forms, the primary packaging material is not only used as protective packaging, but is also a necessary aid for application (medical device) such as metered dose inhalers, prefilled syringes, and powder inhalers. In these special cases, it must be identified individually, if special law governing medicinal products must be applied.

Important functions and resulting requirements of the primary packaging material are shown in figure 16.B-14.

A specification, i.e. a detailed description of the general appearance, material, material strength, printing, etc. must exist for each packaging material. Appropriate test methods must be developed for the individual quality criteria (see chapter 16.B.2 Analytical development).

Examples of tests are:

• Bottles, containers: Material, volume, dimensions
• Closures, stoppers: Material, dimensions, seal, dropper insert, tamper-evident packaging, etc.
• Film: Material, strength, no. of laminate layers
• Labels: Text, size, orientation

If primary packaging materials or specifications for packaging materials are changed during development, these changes must be documented (see chapter 16.E Documentation and recording of changes during development), and their possible influence on the stability of clinical samples or the reliability of long-term stability studies must be assessed. The following are examples of possible effects:

• Incompatibilities between the active substance or the formulation and the packaging material (for example, semi-solid formulation components soften plastic or rubber materials, drug substances cause brittleness, particles from suspensions or pastes lead to material abrasion on surfaces or (tube) closures, discoloration).
• Interactions such as separation of plastic components (leachables) or adsorption of drug ingredients (for example preservatives, proteins)
• Altered tightness, permeability to humidity or light
• Altered ratio of product volume to container surface or changed headspace above the product
• Different handling, for example of closures, droppers, pipettes
• Changes in removable volume (for example, residual volume that cannot be removed in ampoules, vials, tubes, spray bottles)

If changes are still made to the primary packaging in late development phases (phase III) the validation status of the packaging process must also be considered, since the changed packaging materials require different machine settings and may have different processing properties (for example, changes in the film thickness of a blister film or a sachet bag, changes to the dimensions of ampoules, vials, stoppers, material changes of film, stoppers, closures).

It must also be considered whether long-term stability results already obtained at this time for marketing application purposes must include additional new stability batches in the changed packaging material.

Further development and optimisation of packaging for marketed products.

Since the packaging material must be precisely specified in the submission file for marketing authorisation, packaging materials cannot simply be changed once the marketing authorisation has been granted. If a change is planned, it must first be proven that this change will have no influence on product quality. The equivalence of the new packaging must be proven by data that proves the safety, compatibility, appropriate protection of the formulation, and if applicable, the dosage accuracy/release behaviour (e.g. of aerosols).

Examples of changes to packaging materials after the marketing authorisation has been granted:

• Change from bottles to single-dose containers
• Change from HDPE to PVC or from glass to HDPE
• Different number of layers in laminated materials

Changes to packaging materials in approved products are taken very seriously by the authorities: These are classified as type-II changes (in Europe) or major change/supplemental approval (in the USA), and must therefore be reported to the authorities with the supporting data (evidence of suitability) before the change is made. The authorities then send an assessment within 90 days that states whether or not the change is permitted based on the data material provided.

16.B.5 Process development

The first pre-formulations on a laboratory scale are usually produced with a high degree of manual work. As soon as a preliminary formulation has been developed in this way, a suitable manufacturing process that meets the requirements for clinical samples for the initial phases can be designed. This process must take into account the factors listed in figure 16.B-15.

Drug substance attributes

Drug substances have a significant influence on the design of a manufacturing operation. For example, a high level of biological activity and/or toxicity leads to the selection of a procedure that can be executed in closed systems and/or using product-specific equipment. For employee protection, the equipment should also be CIP-capable (Cleaning in Place) (see chapter 4.I CIP (Cleaning in Place)). The same applies for drug substances that cannot be sterilised in their final container, but which are to be processed to form sterile products. Processes in which high shear strengths occur (compacting, tabletting, extrusion, certain types of powder mixture, screw conveyors, high throughput screens, etc.) can lead to problems in substances with low melting points or through the formation of eutectics. Sensitivity to moisture, for example, can be a reason for selecting compaction instead of a granulation process. For temperature-sensitive substances, care should be taken to ensure suitable drying conditions and avoid overheating during the process (for example, bulk temperature after tabletting). Unfavourable electrostatic properties of the active substance or low powder densities, for example, can make suction into containers and systems difficult or impossible.

Formulation properties

In addition to the drug substance attributes, the properties of the excipients, either alone or in combination (as an intermediate stage or bulk formulation), also have a significant influence on the selection of a manufacturing process. For example, a favourable flow behaviour of powders can enable direct compression, while an unfavourable particle size distribution and/or unfavourable flow characteristics may call for additional upstream steps, such as milling, sieving, compacting, or granulation. Agglomerations or demixing tendencies should be taken into account when selecting types of mixers and dryers, and when selecting conveyor systems. The viscosity of liquid or semi-solid systems influences the selection of filtration processes.

Available equipment

Not least, the equipment available is a decisive factor in the selection of a particular procedure. Wherever possible and practical, equipment should already be selected in the development phase that is comparable (in terms of geometry and processing principles) with the future production equipment, and hence offers scale-up capability.

If it is already known that the commercial product will be produced at a different manufacturing site (different plant, subcontractor, or in a different country), the equipment available at that site must be taken into account. Sometimes, for example, a high degree of automation may not be possible at foreign locations, and simpler although more personnel-intensive processes are preferred.

Finally, operational factors should also be taken into account in the selection of a process. For example, if the degree of utilisation of individual machines in production is already very high and further investment is not likely, alternative processing steps or process variants (for example, different mixing or drying methods) should be simultaneously developed and validated from the beginning in order to avoid unnecessary or unacceptable holding times for intermediate products.

Economic and ecological aspects

Economic factors (number of processing steps, process times, degree of automation, use of personnel, re-equipping time, machine down-time etc.) and ecological aspects (use of solvents, energy consumption, cleaning of exhaust air/drainage water, etc. can cause conflict in the selection of a manufacturing process. As a result, the definition of a manufacturing process becomes a matter of compromise between technical feasibility, environmental compatibility, GMP requirements, and cost-effectiveness.

Identification of critical processing steps

During process development and optimisation, it is important to perform an early, critical analysis of the process, in order to avoid delays in later phases and to lay the foundations for process validation. The determination and challenging of critical parameters therefore belongs to the early phases of process development, when technical approaches are defined, even if in relatively small measure and possibly not under GMP conditions. At this time, the following can be tested without great economic cost and without causing unnecessary time delays:

• Which process parameters are critical, i.e. which parameters have an influence on the quality of the product.
• At which threshold values product defects or significant drops in yield actually occur.
• Which process parameters are suitable for process control. (In development, process control is initially performed by intensive sampling and execution of in-process controls in short intervals).
• Whether alternative processing steps or the use of alternative equipment are possible.
• Whether processing steps can be exchanged in the sequence (examples: first heating, then evacuating, or first evacuating, then heating? Can a lubricant or parting agent already be mixed in before temporary storage of the tabletting mass, or not until immediately before tabletting?).
• Whether partial batches made on smaller-scale equipment for capacity reasons can be combined and further processed together following successful in-process controls.
• Whether the holding times of intermediate stages are critical for sensitive products, and in particular which packaging, storage, and if applicable, testing is necessary before further processing.

16.B.6 Cleaning verification and validation

In early development stages, cleaning validation usually does not yet exist. However, since at this stage only very little is known about the toxicity and efficacy (lowest effective dose) of new drug substances, and multi-use systems are usually used in development, it is particularly important to avoid carry-over of active pharmaceutical ingredients into subsequent products. Appropriate cleaning procedures must therefore be developed that are adapted to suit the manufacturing procedure, the equipment, and the properties of the substance (for example, the solubility of drug substance and excipients in various cleaning media). It is, of course, a prerequisite that the manufacturing equipment in development is constructed so that it can be easily and thoroughly cleaned (user requirements during design qualification). The effectiveness of these adapted cleaning methods must be confirmed, for example, by increased visual cleanliness checks. More recently, the term cleaning verification is sometimes used in this context, to indicate that although you are satisfied with the suitability of a cleaning method in early development phases, this is not yet a full cleaning validation. The term cleaning verification has not yet been defined and allocated requirements from a regulatory perspective. Therefore, if this term is used in development, it is important to formulate your own definition and document the procedure, scope (number of repetitions, for example, "one successful pass"), and criteria (for example, visually clean).

In development, particular general conditions apply, which are shown in figure 16.B-16.

During development, the handling of allergens, antibiotics, cytostatics, (peptide) hormones, biological preparations (for example of living micro-organisms), and blood products causes particular problems, because

• Even small traces of these substances may have fatal consequences for certain predisposed persons.
• Dedicated equipment is rarely available at the development stage.
• The protection of staff who process these substances must be guaranteed, although in development, closed systems, CIP facilities, and similar technologies to minimise contact with persons tend to be rare.

For these products, therefore, all measures must be considered that guarantee employee protection and product protection:

• Work in closed systems wherever possible.
• Plan manufacturing in a campaign style, i.e. use the required equipment/premises at least for a certain period of time exclusively for the problematic product.
• Switch to pilot/production equipment as soon as possible, if possible with CIP.
• Use very small, easy-to-clean laboratory instruments for as long as possible.
• Avoid cleaning processes wherever possible (employee protection), for example by using in-liner bags for transport, storage, or compounding containers, which are subsequently destroyed.
• If new investment is possible, in the design qualification it should be ensured in particular that closed systems are used, the fewest possible parts are contaminated with product, CIP cleaning is preferred or cleaning is avoided altogether (for example, in-liner technology with film lining for containers, tubes, etc.).

The measures listed in figure 16.B-17 can be used as a guide for the cleaning verification procedure in early development.

By using these measures, a large amount of basic data is already collected, which contributes to the product safety of the clinical samples, while also acting as a basis for cleaning validation in later development phases (phase II/III). From phase III onwards, it is necessary and practical to perform a cleaning validation (see figure 16.B-18), because in this phase, clinical samples are usually manufactured on a technical scale, that is, in comparatively large batches and using systems that are similar or identical to those used in production (see chapter 8 Cleaning Validation)

.

16.B.7 Process optimisation:
Basic principles for process validation

Poorly developed and insufficiently optimised processes are a serious deficiency frequently encountered in process validation on a production scale. In addition, there is often insufficient data material available to be used as a basis for determining critical processing steps and parameters, and defining the target variables and normal limits of process parameters. Therefore, it is often attempted during validation to compensate for steps that were missed in the development stage: terms such as "challenges", "worst-case" or "optimisation" are then used, although these are not actually part of the validation process, but instead are prerequisites for validation (see chapter 7.E.3 Timing of validation).

This is the aim of the optimisation phase, which should follow the period of formulation and process definition. Chronologically, this improvement phase runs parallel to the whole clinical development (phases I-III) and - depending on the definition - can also include the scale-up process (see figure 16.B-19).

In accordance with modern validation concepts, the main focus of this period is to learn the process capability, and hence the influencing factors and the process capability index of each individual part of the process. Once the influencing factors are known, the process can be optimised and statistical trust placed in the process as part of a permanent validation. This therefore requires permanent data recording, and not simply random data collation of three statistically insignificant "consistency batches". At the end of development, a proficient and validated manufacturing process should be in place.

Knowledge of the process capability is important because for many newly developed products, such as biotechnology products or blood products, tests on the finished product are not significant, which means that definitive statements on quality are only possible using process validation and in-process controls.

When "process validation" is mentioned in the context of pharmaceutical development, this does not refer to a three-fold reproduction of each development batch. Master validation plans are also not required for development. In accordance with PIC - Principles of Qualification and Validation (PR 1/99) and various FDA guidelines, no standard definition exists for the term "validation", and hence validation in development must be understood differently to validation during production (see figure 16.B-20).

This means that a principal task of development is to lay the foundations for later process validation (see figure 16.B-21 and figure 16.B-22).

It is useful to continually build on these principles, since:

• an early, critical analysis of the process avoids delays in later phases.
• stepwise validation during development makes an important contribution towards quality assessment in the release of the clinical samples.
• for clinical test formulations, requirements of the EU-GMP Guideline (article 10, section 4) state that "the manufacturing process as a whole" must be validated (for the validation of clinical samples, see chapter 16.D.2 Manufacturing of clinical samples and comparator drugs).

Equipment qualification

Since relatively little is known about new formulations and processes during development, it is important to continually monitor manufacturing processes and record the data in order to learn more about the process and to form a solid data basis for later validation work. This data recording is, however, only reliable if it is performed using calibrated measuring instruments and using qualified equipment and data interfaces, etc.

Supplier assessment and changes of supplier

If novel excipients or packaging materials are used in development, the qualification of suppliers must begin in sufficient time before handover to production (chapter 2 Procedure for assigning manufacturing contracts). The process of supplier qualification requires the activities listed in figure 16.B-23.

If supplier qualification cannot be performed during development, for example because purchase quantities are too small, it is necessary and acceptable to perform more intensive controls on incoming material instead.

Particularly for novel excipients or packaging material, it is desirable to look for possible alternatives and not merely to rely on one supplier. During optimisation, and also during the subsequent validation, starting materials of the various approved manufacturers must also be considered in order to test their equivalence in processing. Since it is unfortunately not yet possible to provide an exhaustive description of starting materials in the specification based on quality attributes and requirements, it is possible that surprises may occur during processing due to a change of supplier - despite starting materials that conform to specifications.

Recording changes to formulations, processes, and specifications

Changes to formulations, processes and specifications during development should not be made at will, but only if backed up by a scientific rationale. All changes to master formulae and processing instructions must therefore be made in accordance with a SOP and their effects on stability and bioequivalence must be evaluated. All changes must be approved and documented. The relevant current version of instructions and specifications must be included in the product specification file (chapter 16.E Documentation and recording of changes during development).

16.B.8 Upscaling to pilot plant and production scale

Scaling up. is an important step in process development. The greatest changes and surprises are to be expected in the transition from laboratory scale to pilot plant scale, whereas further enlargement to production scale usually only requires fine adjustment.

It is often true that processes which were easy to control in the laboratory using the machines available in the lab, first require considerable adaptation before they can be applied in pilot and production systems. This is not due to a lack of understanding of process development in development department. In order to avoid annoyance of mutual apportioning of blame, it is very important that the employees involved at this interface have the necessary insight into the working problems faced by their counterparts. In addition to altered operational prerequisites that normally govern the pilot or production environment, it is usually purely physical effects that lead to problems in scaling-up.

The following factors can play a role:

• Processing principle
• Surface-to-volume ratio
• Geometries
• High degree of automation
• Transport routes
• Shear strength: abrasion, agglomeration
• Phase separation, air separation
• Electrostatic effects
• Effects of heat
• Increased inertia (for example, prolonged heating and cool-down times)

The processing principle of development machines is often very different to pilot or production machines. Different types of mixers are a common example. Or, for example, a powder that is filled into capsules using a dosing system with vacuum tubes can cause great problems when transferred to the tamping pin dosing system. In some processes, even a change in (container) geometries can lead to unpredictable behaviour in processing. Dead spaces or altered shear strength can have a negative influence on the process. Even if so called "upscalable" development machines are actually available, merely the change in surface-to-volume ratio can cause unexpected effects in the larger machines, for example, slower heating or cooling, reduction of electrostatic effects, and so on. Larger systems are generally characterised by increased inertia, and not every product or process tolerates extended heating and cooling times or longer waits until the product is completely mixed.

Processing steps that cause (undesirable) heating of the (intermediate) product, such as compacting or tabletting, can suddenly become critical if the product can no longer cool down quickly enough due to high processing speeds, leading to rise in bulk temperature.

In general, in the production environment the aim is to achieve a higher degree of automation, for example automatic conveying, suction, pumping, automatic sampling, and automatic in-process controls. These automation steps greatly facilitate the processing work and contribute greatly to the reproducibility and safety of the process. However, unwanted effects can also occur with automation: During transport through pipes or tubes, shear forces act on the conveyed substance and, depending on the product, this can lead to abrasion, powder agglomeration, phase separation, or air separation. For example, if low-density powders are vacuumed automatically, it should be tested during process development whether the vacuum process is quantitative or whether the fine fraction remains caught in the filter - and hence causes a reduced content of active pharmaceutical ingredient in the final product.

These technical problems automatically identify critical processing steps and parameters. Ranges for the critical processing steps/parameters must be defined in the pilot phase, so that they can later be used as a basis for validating the production process. Of course, qualification of the pilot plants must be completed as a prerequisite.

In addition to the technical aspects of process optimisation, the pilot batches are also significantly important as

• clinical samples for long-term studies (phase III),
• long-term stability samples for marketing application purposes.

As a precondition the definitive formulation and the primary packaging foreseen for the market must be fixed, and the drugs must be produced in batch sizes of at least 1/10 of the planned production batch size.

In the subsequent handover of the manufacturing process to production, the following prerequisites must be fulfilled:

• Qualification of production facilities and equipment
• Environmental monitoring
• Cleaning validation of production facilities and equipment
• Compilation of definitive manufacturing description (processing instruction) with in-process controls and process parameters and sampling plans
• Establishment of specifications for the commercial product
• Manufacturing of consistency batches, i.e. validation of the normal operating range

16.B.9 Handover to other manufacturing sites

A change in manufacturing site, for example to a different, maybe newly-built production unit of the same company, to a subsidiary, or to a subcontractor, requires revalidation of manufacturing and cleaning process, since the results of the original validation are not transferable to the conditions in the changed environment. Generally there will not be the same systems or equipment available at the new manufacturing site - at best, equipment will be of similar design. In the case of relocation to a new production hall, the original system is dismantled and must be requalified and commissioned in the new location. In addition, different staff are usually responsible for manufacturing, sampling, and in-process control tasks at the new site. In some cases, even different raw material suppliers must be approved.

All these aspects explain why a process has to be revalidated (prospectively or concurrently). However, before validation takes place, the following basic prerequisites must be fulfilled at the new manufacturing site:

• Qualification of systems and equipment
• Calibration and maintenance according to written plans (SOPs)
• GMP-training of staff involved
• Environmental monitoring
• Effective change control system