Process validation is a legal requirement applied to active pharmaceutical ingredient manufacturers by governments and it is administered by their regulators. Process development chemists contribute to the documentation of process validation.
Philosophical Differences between Process Chemists and Regulators
Process chemists when faced with a synthesis problem ask themselves the question, “How might this be made to work?” We then seek the optimal operating parameters for a scheme that achieves the desired physical reality. Regulators look at what we have achieved and ask themselves, “How could this fail?” Instead of looking at the optimal conditions that have been found, they are looking at the limits for the failure of every critical parameter of the system we have optimized and are asking how this could upset the output quality of the system? As creators, we struggle to succeed at jigging the system to give the result we need. We struggle to understand the coded messages of our experiments to deduce what a practical process would be. The regulators analyze what we have created to see at what points it could break down. Since any process can be made to fail by deviating enough from its teaching, this thinking is alien and frustrating to us.
Philosophical Differences between Process Chemists and Regulators
Process chemists when faced with a synthesis problem ask themselves the question, “How might this be made to work?” We then seek the optimal operating parameters for a scheme that achieves the desired physical reality. Regulators look at what we have achieved and ask themselves, “How could this fail?” Instead of looking at the optimal conditions that have been found, they are looking at the limits for the failure of every critical parameter of the system we have optimized and are asking how this could upset the output quality of the system? As creators, we struggle to succeed at jigging the system to give the result we need. We struggle to understand the coded messages of our experiments to deduce what a practical process would be. The regulators analyze what we have created to see at what points it could break down. Since any process can be made to fail by deviating enough from its teaching, this thinking is alien and frustrating to us.
Drug Substance vs Drug Product
As Dr. Oljan Repic has pointed out in his book,[ Principles of Process Research and Chemical Development in the Pharmaceutical Industry, John Wiley & Sons, Inc. p. 179-194], a chemical process for making an active pharmaceutical ingredient (API) from commercial bulk chemicals is very, very different from making a final dosage form from its API. This is because, in the former, chemical steps result in the creation of new substances by first combining a great many undesired materials (substrates, substrate impurities, reagents, catalysts, solvents, etc) and then, after exposure to some reacting conditions, separating these and other undesirable materials (co-products, by-products, unanticipated and unidentified impurities) from the desired reaction product. In contrast, making a drug product from an API does not involve chemical reactions or purifications (other than drying). The drug product is some kind of mechanical mixture or micro-engineered assemblage of the inputs and so it is a simple combination of what is put into the process modulated by the manipulations. This is not to say that how an active ingredient is formulated into a drug product doesn’t affect the medicine’s effectiveness or even safety. It does. What I (and I think, Dr. Repic) are maintaining is that formulating is an inherently simpler change than synthesis and consequently failing to control the process of the former within the limits established is much more certain to cause the product to fail its specifications. Since there is no purification aspect in formulating, if it is garbage in, it will most certainly be garbage out. There is no possibility for purification in formulation so there is no capacity to rectify any serious deviations.
The synthesis of a pure chemical substance such as an API is nowhere near so dependent upon the starting material purity and the nitty-gritty of the process. In fact in the limit (100% purity), the properties of a pure chemical substance are completely independent both of its route of synthesis and the details of the most immediate transformation by which it is prepared. Even in the real world of actual APIs, the only evidence of the route of synthesis that remains in ‘pure’ product is the faint fingerprint of trace impurities that are still carried in it. Effective purification decreases this below any level of practical significance related to efficacy or safety and even, in many cases, below the level of analytical detection.
Even if it might come as a surprise to beginning formulators that how an API and the other excipients in a drug product recipe are combined can significantly alter the bioavailability of a tablet; it would not surprise even one tyro chemist that deviation from the details of a chemical process would cause the quality and yield of such product to suffer.
It is the Purification that Must Work
In a synthesis it is really only the final purification that needs to be effective to deliver a drug substance of the required purity, with the proviso of course that the analytical methods used to assess that purity need be validated. The process by which the final product is produced needs to be known essentially only so that the analytical methods can be tested for their capacity to detect and quantify the authentic likely/expected impurities.
This however is not the law. Regulators and politicians err on the side of caution and the validation rules they approve are more rigid than they need to be, because the rules derive historically from, and still closely resemble, those for formulated drug products. Process validation was originally conceived and implemented as a method to assure that each unit of a drug product could be guaranteed safe even when it could not be analyzed, either because the analysis would destroy the product or because the proper functioning of the product could not be analyzed. Drug security by validating the process was initially directed to drugs formulated for injection into the bloodstream. If these were contaminated by bacteria the patient would likely quickly die. Even if one took a proper statistically representative sample of the product there still could be some fatally toxic units of the product undiscovered. The solution was to control the process used to make the units so strictly that it would be extremely unlikely to make even one contaminated sample. This was the only course of action that could be effective because the fatal contamination came from the process (from the bacteria carried in the air) not from any starting materials.
Process validation also turned out to be valuable for assuring the efficacy of drug dosage forms where the delivery of the active ingredient from the dosage form ( ie tablet, capsule) into the bloodstream of the patient was challenging. In these cases, analysis of the dosage form was not showing reliably whether a particular lot of drug product could deliver the clinically required exposure to the medicine. Drug product prepared by a validated process was found much more likely to produce the desired clinical outcome.
Process validation differs significantly from chemical analysis in its ability to provide protection from different types of contamination problems. A drug substance is almost always a pure chemical compound. It is a single molecular entity and can be analyzed; precisely, to confirm its constitution, and sensitively, to show the presence of homogeneously dispersed contaminants. Drug substances are not mechanical mixtures like flakes of this and that stirred together as is often the case for tablets and capsules. Such mixtures are statistically homogeneous only at the scale of a dose but heterogeneous at the particle scale. This is what one finds in a mixture that is pressed into a pill. Process validation is better than analysis to guarantee that heterogeneous, most often particulate, contamination is not present. The reason for this is that chemical analysis of an active pharmaceutical ingredient (API) requires taking a representative sample from the entire batch to perform the analysis. It is quite likely that even with an approved statistically sound sampling procedure a heterogeneously contaminated batch would not get any contaminant into the sample being analyzed. Take for example the situation where glass particles from a chipped piece of equipment contaminated a batch. The dense glass would migrate to the bottom of the container and might escape the sampling. Thus it would not be detected. A validated process, however, could make sure that there was no conceivable condition where any glass could remain in a final product. The final filtration before the final crystallization of an API is part of a process for just such a validation purpose.
Process Validation: A Development Scientist’s Most Important Activity
Drug discovery chemists do not need to learn about process validation; process development chemists do. It is the latter’s work that is being examined. Nevertheless, scientists in general always like to know whatever they can about the synthetic routes they devise and discovery chemists may eventually transition into career development chemists. The additional experimentation that process validation requires does increase scientific knowledge about that process. It answers questions about the practical limits of those process conditions that most significantly impact final product quality. But aside from the scientific merits, it is important that pharmaceutical chemists, particularly process development chemists, understand process validation, since, because the pharmaceutical business is a highly government-regulated activity, any API they might make cannot be included in a medicine for sale unless it is made by a validated process. This makes validation the most economically significant test to which a scaled-up industrial process can be subjected and so all the chemist’s activities should be viewed from the perspective of its contribution towards success in validation.
Many and perhaps most companies that manufacture APIs exclude scientists below the managerial level from any contact with the government regulators who assess the quality and completeness of the validation process. Likewise, it is only senior technical people sign off on or assemble the validation report, which is the core document of the validation exercise. Yet exposing employees, who have very specialized tasks, to the all-encompassing validation program lets them see their precise role in it. Specifically, chemists, who are focused on process development, are likely to do much better in organizing their data, writing reports, etc. if they understand the larger purpose their data and conclusions will serve when process validation protocols are assembled and executed and the Development Report is folded into the Validation Report. Indeed it is highly recommended that general training in the concepts of process validation should be given to all individuals who are part of the development and manufacturing program not just to those who require such training as part of Good Manufacturing Practice (GMP).
The Corporate Validation Policy
Validation is so important to API-producing companies that very often it is heralded in a written Corporate Validation Policy. Where this document exists it is the overarching document pertaining to the validation methodology. As such it sets down general principles that employees should adhere to, rather than the nitty-gritty details; nevertheless, it is a good place to start in examining that most important activity called process validation.
A lifecycle policy approach to the validation concept is regarded by many experts as preferable or even essential. Validation activities begin with the inception of the development program and logically should not end until the product is retired. A corporate validation policy should provide validation expectations over the lifetime of the process. It should state the company’s expectations for both the organization’s scientific and documentation goals. Both of these classes of goals need to be kept in mind and adhered to from the first day of process development to the last. In other words, the Product Development Report (PDR) should be a ‘living’ progress report that is continuously updated through the product lifecycle. Many firms make the mistake of creating a static, concrete, dated, PDR paper which is archived once that first regulatory approval is obtained. Because improvements are being made throughout the lifetime of the process, these may require additional validation work and regulatory filings. These changes should set in motion an update of the PDR.
Since every organization has a somewhat different internal structure, any policy needs to allocate responsibilities for maintaining the product/process documentation including the mature product responsibilities relating to monitoring, trending, and change control.
The Validation Master Plan
The Validation Master Plan is the most senior planning document that pertains to a specific validation. It makes concrete and project specific what is set down in the Validation Policy only in a general way. It takes into account even the most recent changes in the organization’s structure, responsibilities, and reporting chains. The Validation Master Plan assigns actual project milestones to named persons and groups. It sets dates for meeting those milestones. It sets out what activities are to be performed when and it decides the science that will take place at small and at large scales. If the validation is for a new drug substance the Validation Master Plan needs to be revisited and modified as the clinical phases advance.
All aspects of the validation documentation are important. The regulators look not for what the company considers most important; what is scientifically most significant; what has been presented most clearly and completely or where the maximum inherent risk is, but what has been overlooked because that is from their perspective where that process deviation leading to dangerous medicine could most likely arise.
Some Aspects of Process Validation that are not the Development Chemists’ Responsibility
Validation must demonstrate that the process being examined is performed under good manufacturing practice (GMP). Mercifully there are many aspects of GMP that are not the responsibility of development chemists. Every organization that manufactures products for sale makes batches for clinical (human) tests or supports regulatory filings must have a separate and independent quality control unit to test, approve, and reject products; set and approve specifications for raw materials and products; and write and approve these and other written procedures. Another requirement of GMP is that all personnel associated with production must be trained in GMP and the company must be able to prove that training with the signatures of those trained. You may have been asked to submit your resume to a company representative responsible for meeting GMP standards. The company is required to show that its scientists too are adequately credentialed for their work.
The facilities and equipment to be included in a process need to be qualified before they can be used in an acceptable validation exercise. These qualifications are usually unfamiliar to practicing chemists. They are matters that are implicit in our everyday activities and are completely taken for granted by us.
Facility qualification relates to the suitability of the physical site and its infrastructure and personnel to carry out the process that is to be validated. Most of these matters are strictly the domain of architects, contractors, and engineers.
Drug substances in general have requirements for the facilities in which they are manufactured. The FDA will not allow someone to make a GMP drug substance in a bathtub in a garage! It insists upon the written assurance that the manufacturing facility has adequate space, heating, ventilation, plumbing, and maintenance.
The Development Report
The core document that development chemists do contribute to API validation is the Development Report. Process chemists recognize the Development Report as the history of the entire process development but it is at the same time a regulatory validation document of enormous importance. Validation would need to be done, if for nothing else, at least to convince regulators that the product is reproducibly safe, but API product safety is not the only goal of process development. Optimization aims to minimize cost, maximize throughput, achieve appropriate safety, reduce environmental burden, and deliver consistent, convenient operation of the steps. Consequently, some IPCs and some specifications are not critical in the regulatory sense of controlling final product quality. Deviations from these do not put the API quality at risk. Some of these aforementioned standard ranges alert batch sheet reviewers, when they are changing non-randomly, to potential creeping deviations (trends) that can have economic, safety, environmental, or simply consistency consequences, without there being quality ones. This means that an API could fail a noncritical IPC and still be safe and effective to use. Failing a specification or IPC that correlates with a critical process parameter (CPP) and in turn a critical quality attribute (CQA), on the other hand, would mark a failed batch.
A synthesis can in strict theory be traced back to the most elementary of chemical building blocks- even atoms themselves. What needs to be decided for validation is how many of the steps in the retrosynthetic direction need to be validated to ensure the required safety. This ‘depth’ of the validation study establishes what chemical transformations or treatments are included within the validation and what steps are accepted simply as the making of starting materials to be controlled by their material specifications independent of their particular synthesis.
Today, it is a good bet that most synthetic API processes are multi-step with at least one complex chemical transformation. Similarly, it is unlikely to be disputed that the final process step that creates an API, and the subsequently entailed purification, needs to be validated. Beyond this, there is no consensus answer as to when to apply process validation for intermediate steps in a multi-step process. Each process it seems must be evaluated as a separate case. So many factors are in play. This does not mean that a coherent policy or procedure cannot be established for a company describing how it should do that evaluation. In fact, putting such a decision tree in place can lead to better, faster, even less expensive, choices.
The process validation for an API is at least the sum of separate validations; one for each intermediate step that contains a critical parameter. Thus the determination of what steps are to be included in the full validation amounts to identifying what steps contain at least one critical parameter. This is much the same as asking what the starting materials are for the steps that will be validated because steps for making an API starting material do not need validation. Terminology can be the source of as much disagreement as the chemistry in these discussions so definitions are important. A ‘step’ leads either to an isolated or non-isolated intermediate. Other important determinations to get straight are: what is(are) the ultimate intermediate(s) or final intermediate(s), and what overall scheme of reactions and purifications will be required to produce the final API.
Determining what constitutes a starting material is debatable. According to Roger W. Koops, [Process Validation of Synthetic Chemical Processes for the Production of Active Pharmaceutical Ingredients (APIs), Journal of Validation Technology, Volume 8 Number 2 January 2002], “a starting material should be a readily available item that has a standard grade (or grades) associated with it, and has been well characterized. If produced under contract by a vendor, it should be produced by a known and established process, and the end product should be, again, capable of achieving a standard grade. Second, the vendor should be qualified (under a vendor qualification program)to ensure that a material meets consistent quality standards. A rudimentary quality agreement should be established to outline change notifications and quality requirements. What must be avoided are unshared, unannounced process changes that can change the impurity profile of the downstream drug. If a material is manufactured by a contracting party and the process was supplied by the innovator, the material is simply a third-party manufactured intermediate. The contractor now becomes an extension of the innovator, and the transferred process must be included in the full validation evaluation”.
Starting materials, generally speaking, are inputs to a synthesis whose atoms are partially incorporated into the drug substance. They do not need to have their own syntheses validated. Their quality is accepted based on their certificates of analysis. In contrast, intermediates are created directly or indirectly from starting materials. They arise downstream in the process with respect to some starting materials. Most intermediates arise in steps that need validation, but at least in principle, there can be exceptions. A step starting with an intermediate and resulting in another intermediate does not need to be validated if the interconversion process step has no critical parameters. That is, the step is so rugged that no matter how it is executed it does not affect the purity of the final drug product. This would be a rare case indeed. The reaction would need to be ultra-tolerant of reaction parameter deviations and/or the isolation would need to be exceptionally powerful at excluding impurities.
As the overall chemical development proceeds an impurity profile for the final product should emerge The validation policy should emphasize the importance of characterizing process impurities as early in the development process as possible. From this, each intermediate process step can be evaluated as to its influence on the purity of the end product. If an intermediate needs to be controlled because it could contribute to an adverse quality profile of the API, process validation should be applied to its related process step.
Each intermediate needs to be evaluated for how it contributes to the final API, in regard to the impurity profile and the specific process for that intermediate. Besides isolated intermediates, non-isolated intermediates should also be evaluated, since steps comprising these may be ones where even stricter controls may be required. For example, an intermediate may not be isolated due to structural instability when not in solution. In this case, the concentration and/or potency of the intermediate may be a critical attribute that needs to be controlled.
Either the corporate validation policy or the particular validation plan needs to provide guidance on whether the validation exercise at the plant scale should be designed so as to “stress” the accepted parameters by operating at the maximum or minimum accepted level (the edge of failure), or to aim for the set point target of each parameter. Some companies have attempted to matrix the parameter limits in an effort to test the extreme limits of process parameters.
This author holds the view that the laboratory development phase is the place to establish the boundaries of the operating ranges, and how combinations between extremes of parameters affect the final outcome of the process. Process validation, by definition, is used to demonstrate the limits of consistency of the process as designed and thus should aim for the set points of operating parameters. From a manufacturing standpoint, it is advantageous to operate based on the targeted parameter values for batch consistency.
Because the Process Development Report (PDR) is such a key element of validation documentation, it is important for the corporate validation policy to set out at what points interim reports need to be submitted during process development because interim reports make the compilation of the PDR easier to manage.
The Corporate Validation Policy may need to provide some guidance concerning choosing critical parameters since their number strongly affects the breadth and hence the expense of the experimentation.
Some organizations choose to call critical only parameters, that affect the impurity profile and the ability of a material to pass specifications. However, the ability of an output product to meet specifications is not the sole indicator that its process is running smoothly. There are sometimes parameters that, for example, affect yield, without affecting the final product’s properties. Nonetheless, variation of yield beyond normal experimental error points towards a process out of control. Whatever the inclusion/exclusion rules, the basis of the decision can be either a general policy prescription or something considered only on a case-by-case basis at the level of the Validation Master Plan (VMP).
Validation Protocols
The validation Master Plan amalgamates validation protocols. The hallmark of a process step that is ‘under control’ is that is predictable. The sequence of operations that constitute the step is unequivocally established in advance, the acceptable operating ranges of the critical parameters are set down in advance, and the critical specifications of the output material are set down in advance. So what sets a validation batch apart from any other pilot experiment is that there can be no recourse to any procedural or corrective activity that is not set out in advance in the protocol. The organization cannot redefine success in a process step being validated after the requisite batch is complete and so eliminate deviations between what was predicted and what transpired. The regulators are the referees who decide whether, when the organization executes the protocol, it follows its own procedures and meets its own standards.
Although some experts urge that the protocol should be drafted to be ‘as concise as possible’, protocols are normally known for their verbosity. A process validation protocol will describe in excruciating detail the ‘how to’ for the validation exercise. The protocol resembles a master record in many ways but provides a greater level of descriptive detail regarding the execution of the process, the monitoring of the steps containing critical parameters, and sampling instructions for the in-process checks, particularly critical in-process checks. In every case, either a procedure should be described in detail in the protocol or a reference should be provided in the protocol to an already established procedure for handling the task (ie a standard operating procedure). The instructions should not just be sufficiently unambiguous that they would direct a trained operator to choose the correct action. It should be such that a regulatory reviewer can only conceive a single proper performance of the instruction. Certain styles of drafting can condense the presentation without losing explicitness. Where applicable, data should be entered directly into the protocol. If an exercise is particularly complex, a series of protocols may be preferable. Protocols should have places in them for reviewers’ initials or signatures.
One way of looking at the components of a validation protocol is to see that they reproduce many of the parts of a summary report about a set of batches, but unlike the report, the protocol is written in advance (remember after developing a process you are supposed to be able to predict everything that is crucial to product safety in advance).
The protocol should include many sections that process chemists would likely consider report padding. There should be an introductory section defining the scope of the protocol relating back to the overall Validation Master Plan. The question of where this particular protocol fits into the overall validations should also be answered. The responsibilities for execution, testing, review, and approval for the overall validation exercise need to be reproduced here even if they also appear elsewhere. Also, for each particular protocol, the department and job responsibilities need to be restated and the entire protocol needs to be approved by the top technical officer and the top quality officer. The purpose of the sign-off is to make unequivocally clear that top management has seen and bought into the exact protocol exercise that is to be executed.
Because a description in words alone may be more difficult to follow than an illustration, a process flow diagram (PFD) for the particular step or portion of a process that is being validated should be provided annotated with starting materials, reagents, solvents, process conditions, transfers, by-product concentrates, waste streams, and purifications, etc. If there are any recycle loops or patches prepared in advance these need to be presented here in advance. The operations should be assigned an identification code so that further elaborations can be related unequivocally to the correct portion of the PFD. If seed crystals are used where they come from, what quantity they are what specifications they meet need to be set down. It is generally important and advantageous to identify critical process steps containing one or more critical process parameters on the PFD document. The idea is to give the operators no ‘wiggle room’ as to how the execution of the exercise is to proceed. This in turn gives regulators confidence that you have confidence in your process understanding and control.
A complete protocol should provide a bill of materials for every consumable item used during the process. Particular attention should be given to be sure that even materials that are not part of the chemical transformations also be there, such as filters, filter aids, packaging for chemicals before and after the main chemical transformation, seed crystals, etc. Some materials have both a name and an identification code. A reference to the specification or COA should be in the listing. If the PFD has the operations in the sequence labeled, this label should be here as a cross-reference.
All the equipment used in the process needs to be referenced. The actual documentation does not need to be reproduced in the report but the IQ, OQ, and PQ report numbers need to be given as well as the cleaning procedures and the cleaning validation reports and instructions. The ranges of parameters, particularly any critical parameters that will be used should be noted and highlighted. Of course, the required operating range needs to be within the validated range for each piece of equipment. If there are any special procedures that involve the equipment during the process these procedures must be detailed here or referenced accurately.
Facility documentation only needs to be referenced along with the DQ, IQ, OQ, and PQ materials. The only recapitulation of the facility validation would refer to the particular importance of any unusual facility capability for the protocol being validated. One example of such a special competence would be the inclusion in the facility of Class C clean rooms.
Critical Parameters
Only at this point do we get to elements that process chemists would consider the essence of scaled-up inventiveness. Specific page references should be made to the Development Report where the critical parameters are established and the data used to establish them is organized. Various ranges of the critical parameter need to be discussed. The master procedure range is the range that will be set out in the master batch sheet. Within that range will be the set point which is the numerical setting used on the regulating equipment. Under regular operating conditions, the set point can only be transiently met because of the equipment’s inherent limitations to overshoot and undershoot. The master procedure range of course cannot be less than the range set by the equipment’s inherent limitations. The next wider range is the verified range which is the range actually experienced in the plant. The Development Report may make predictions of a wider verifiable range from laboratory experiments. Using development data, a prediction will also be made of the process limits for the parameter. This is the edge of failure exceeding which the API is expected to suffer quality repercussions. Another limit that may be mentioned is the practical range of the equipment with respect to the critic parameter. For example, in a steam-heated reactor, the practical limit for external heating would be 150 C. For each critical parameter the controls that will be applied to maintain it need to be listed.
In-Process Tests
Every batch sheet comprises in-process tests that are used to assess the progress of the processing. A list is required of all the in-process tests within the protocol showing at what point the sample for testing is taken, what the test method will be, who will perform the test, and how the results will be reported. If the test is a go/no go test which is repeated until the go signal is obtained this should be explained. Achieving a particular result may be a critical parameter. If this is the case that needs to be made clear. For example, neither the reaction temperature nor the time may be critical but the outcome of an in-process test for reaction completion may be critical. Thus a lower temperature and longer time or a higher temperature and shorter time may each achieve the completeness required to pass a critical IPC.
Sampling Plans
Process development chemists know that sampling heterogeneous reaction mixtures does not give results representative of the mixture in its entirety. Even testing of the separated phases individually is fraught with difficulties. Solids that are to be sampled are rarely homogeneous. Even homogeneous solid mixtures can demix with mild shaking. As a consequence, if one is going to have a significant analytic result, the sampling must achieve a representative sample. Within the protocol, the plan should indicate the sampling location, the amount of sample to be taken, how it is to be collected, how it is to be stabilized if necessary, how stored, and how transported to the testing facility. Both operators and regulators need to know before the exercise exactly how the sampling is planned. Of course, at what point the test will be done must be indicated unequivocally. If there is a sampling deviation, it is helpful if it is reported in the batch sheet before any result for that sample is available.
Acceptance Criteria
All criteria that must be within predefined ranges must be listed in advance both critical parameters and IPC results and testing on the final product.
Deviations and Investigations
The most desirable outcome from executing a validation protocol is that all critical parameters are held within the operating range and the resultant intermediate has test measurements within the range of intermediates that have provided safe and effective IPA. That is what we aim for. What if some critical parameters are measured outside the master procedure range but not outside the verified range? What if some critical parameters have strayed outside the verified range but the intermediate seems unchanged in properties from the intermediate prepared without deviations? What if the migration out of bounds was brief or very small in magnitude? What is very important is that as much as possible the assessment of the impact of a possible deviation should be made before the validation is actually executed. Identifying a parameter as critical when it is not critical is unlikely after a thorough development but getting the limits of failure somewhat incorrect is rather likely. One does not want to abandon or fail a validation batch when in fact it will result in a pure safe final IPA. Falling outside of most critical parameter ranges is easily avoided but sometimes the desired operating range falls close to a limit of failure because the manufacturing needs to achieve some objective not related to final product safety or effectiveness but to minimizing cost, improving safety, or reducing waste. It is in these situations where preliminary considerations of less desirable outcomes can be invaluable.
Equipment and Facility Qualification
Equipment Design qualification (DQ) relates to equipment/machines. Design qualification is normally the responsibility of the equipment vendors. Successful Installation qualification (IQ) demonstrates that the equipment has been properly installed. Operational qualification (OQ) demonstrates that the equipment can achieve the operating parameters intended when exposed to typical substrates. An OQ does not need to specifically test the conditions of a particular process. That is the function of performance qualification (PQ). In general, it is best to have all facility and equipment qualifications completed prior to process validation (certainly all IQ/OQ should be completed). Performance Qualification (PQ) can be performed during process validation, but it is generally advisable to have completed this testing prior to process validation since it would add additional risk to the exercise. However, there may be some instances where it is useful to concurrently perform an equipment or facility PQ during process validation. In those cases, a separate PQ protocol should be drafted and referenced in the process validation protocol.
Analytical Validation
The validation of the required analytical methods that are used to follow the trends of the critical parameters during the process experimentation may be regarded as a sub-program to the overall validation project. Secure conclusions can only be derived from reliable data collected using validated methods, so methods validation needs to proceed hand-in-hand with the examination of the chemical transformation processes themselves.
Cleaning Validation
Equipment cleaning methods need to be included as part of the Validation Report. Cleaning validation is also sufficiently complex that it is better treated as a sub-discipline with its own report sub-section. Some API producers prefer to have standard cleaning procedures for their equipment which are always executed irrespective of the process that is being cleaned up. These methods are part of the SOPs for operations. Where these methods alone cannot be expected to satisfy the cleaning requirements additional cleaning may be set in place as part of the Master Batch Record following a particular exposure. Another organization might specify the cleaning treatment based on what the chemical step has exposed equipment to. That decision can be part of the Corporate Validation strategy. Cleaning methods can be developed in parallel with the chemical process development and cleaning validation can conveniently occur either prior to or concurrent with process validation. These decisions can also be policy decisions or part of a Validation Master Plan for a particular validation.
Cleaning of reactors in which any broadly insoluble, polymeric, or tarry by-product is formed can expend valuable reactor time and penalize throughput. When process chemists include such steps in a process going to scale up they should take a special responsibility to work with production personnel to develop efficient cleaning or at least pass along a warning of the difficulty and whatever they already know about overcoming it.
Conclusion
In conclusion, although process chemists need not become validation experts and although they may not be personally enmeshed in the validation exercise, some overall familiarity with the subject, from their own perspective and emphasizing their unique role, can enhance their work and their enjoyment of it.