FDA's Draft Guidance for Process Validation: Can It Be Applied Universally?

Published on: 
Pharmaceutical Technology, Pharmaceutical Technology-05-01-2009, Volume 2009 Supplement, Issue 2

The author describes various manufacturing processes and evaluates whether the guidance can be applied to each of them.

In November 2008, the US Food and Drug Administration issued Draft Guidance for Industry—Process Validation: General Principles and Practices (1). The guidance outlines regulatory expectations for process validation following a "life-cycle concept" that describes a "cradle-to-grave" approach for validating pharmaceutical processes (2). The life-cycle approach builds upon the results of experimental activities during development to define operating parameters and product specifications that are used in initial and ongoing process qualification. The life-cycle concept provides a robust means for the development, manufacture, and control of pharmaceutical products.

The guidance document covers validation largely at a conceptual level and avoids narrow precepts and specific examples. This approach is appropriate because the document addresses the subject from active pharmaceutical ingredient (API) production (by either chemical synthesis or biological processes) through drug-product production for all pharmaceutical dosage forms. The intended breadth of coverage embraces a myriad of unit operations in the preparation and manufacture of these products. Unstated is whether the draft guidance is intended to be applied to supportive processes that are not an inherent part of the formulation process. Among the support processes are cleaning, inspection, sterilization, and aseptic processing. Each of these processes can be an essential part of pharmaceutical manufacture that requires validation.

Basics of the draft guidance

The draft guidance recommends a defined and structured approach for process-validation activities within an organization. During the design and development stage, experiments should define the relationship between the independent process parameters and the dependent product attributes. These studies should be conducted in a predefined manner, and the results should be documented for later reference. The goal of process development is the attainment of knowledge regarding the process–product relationships to support later commercial production. The greater the knowledge accumulated at this early period, the more assurance the firm will have that it can successfully launch and maintain the process at a commercial scale. From a compliance perspective, this approach makes excellent sense; the knowledge gained during preclinical and clinical-stage experimentation provides a way to link clinical data obtained at the smaller scales with data from the later production process. A well-developed process is one for which the critical process parameters have been identified and control ranges for each have been established (3). The development experiments should evaluate the interaction between the independent and dependent variables until the results of the process are predictable and routinely acceptable. Multifactorial experiments can assess the relationships between the variables and build knowledge about the process's limitations. The experiments performed at a smaller scale establish the acceptable ranges for the various independent process parameters. Thus, when the process is operated under the appropriate conditions, operators have substantially greater confidence that the desired quality attributes of the product will be realized. The acceptability of the end result is ascertained using samples of the completed materials.

Initial and ongoing qualification of production processes are the means for establishing and confirming the experimental experience at significantly larger scales of operation. Knowledge gleaned from the development simplifies later activities. Production processes are always operated within the defined operating ranges because there is no reason to experiment with conditions at the extreme ends of the ranges on this larger scale. Challenges during these stages are primarily in the number of tests performed on the produced materials. Qualification lots are customarily sampled at a substantially higher rate than are routine production lots, and testing of these expanded samples is the challenge of the commercial process. The guidance recommends using appropriate statistical tools in the full-scale qualification efforts to provide the desired confidence in process and product acceptability and thus attain the desired validated state. The expectations for statistical evidence in process validation are well founded; sampling batches at the modest levels associated with pharmacopeial tests provide little, if any, proof of end-product quality. Although those tests may be legally binding, they have only limited value. Industry has largely ignored the levels of "real quality" needed to support its claims for patient welfare (4).

The draft guidance has a broad scope from API production through to finished dosage forms and therefore must be applicable in various situations. The validation methodology outlined provides substantial flexibility to the industry, avoiding details that might fit well in one instance but be wholly inappropriate in another. Objectively speaking, the guidance's lack of definitions can be viewed both positively and negatively. Firms with extensive knowledge of their processes will appreciate that FDA believes that they should be abe to validate their own processes without having to accommodate arbitrary precepts. On the other hand, firms lacking the requisite process knowledge would likely have preferred substantially more detail with regard to regulatory expectations for process validation. The draft describes FDA's process-validation expectations from a "what to" rather than a "how to" perspective. That approach is wholly consistent with the 21 CFR Part 211 current good manufacturing practices (CGMPs), where firms are given great latitude with respect to the methods used to realize the compliance requirements (5).

The document properly places process equipment in a secondary role. Equipment qualification is certainly expected, but only as a means to support the process and not as the focus of the effort or an end in itself. Providing clarity in this regard is an important step forward and is consistent with FDA's risk-based compliance objective, the recent American Society for Testing and Materials effort at refocusing equipment qualification activities, and industry publications that cite the need for focus on critical end-product quality concerns (6–8).

In summary, the usefulness of the draft guidance for validating pharmaceutical production processes and products is unquestionable. The life-cycle model will result in development and validation exercises that provide relevant and meaningful information. The link between the process parameters that influence the critical quality attributes will serve the industry well. The use of statistical methods will add a rigor to the validation efforts that has been sorely lacking.

Applying the guidance outside the process or product focus

Perhaps the only question that need be asked is how easily the draft guidance can be applied to processes and systems that are less clearly related to end-product quality attributes, or where the development model described might be substantially different. The most important of these systems and processes commonly used in the pharmaceutical industry include:

  • Critical process utilities (e.g., water systems and compressed gases)

  • Classified and controlled environments (e.g., ISO 5–8 rooms and cold boxes)

  • Computerized systems (e.g., process control and manufacturing resource planning)

  • Inspection of product attributes (e.g., visible particle and labeling)
Advertisement

  • Cleaning and preparation procedures (e.g., product-contact parts and surfaces)

  • Sterilization processes (e.g., steam, dry heat, and radiation)

  • Aseptic processing (e.g., filling and compounding)

  • Manual procedures (e.g., gowning and sanitization).

The ease with which the guidance can be applied varies substantially with the particular process, and each general category is best considered individually.

Critical process utilities. These systems are largely mechanical and include multiple components assembled to deliver an essential process utility throughout the facility. The equipment components of these systems are easily addressed in a manner essentially identical to that defined for production equipment. The major difference is that the utility equipment requires substantially less development to place it into operational service. The processes to prepare the liquids or gases are largely unchanging, and thus the concepts such as design of experiments (DOE) and multivariate analysis are of limited value. The process utility is performance qualified using a sampling plan that may have statistical elements, though they are not necessary. These systems are customarily subjected to routine monitoring on a frequent basis; many water systems are monitored daily. These systems include stages of initial and ongoing performance qualification, but the attributes of greatest concern, microbial identity and population, are poor fits with most statistical tools. Thus the guidance is not readily applicable to these systems.

Classified and controlled environments. Similar in some ways to the utility systems described above, these systems provide a condition rather than a specific material at a well-defined state. The equipment-qualification side of these systems fits the guidance model reasonably well. Because the systems perform in essentially the same manner at all locations and for nearly all applications, development in the sense required in the guidance does not apply at all. A design phase certainly exists, but its elements are not at all like those related to formulation or synthesis processes.

The simplest of these systems (e.g., cold boxes, incubators, and similar items) require little performance qualification, and the use of statistics is hardly warranted. At the other extreme, aseptic processing areas, the evaluation of successful performance is subject to numerous external influences, so these systems can't be considered performance qualified at all, at least not in an independent manner. Acceptable results in an aseptic processing environment result more from diligence in manual routine decontamination, housekeeping practices, operator gowning, and the like. It would be inappropriate to suggest that once the environment is properly qualified, the activities performed inside it would always be successful. Success simply can't be guaranteed, at least not in any manned environment. The application of the guidance to classified environments without adjustment is not feasible.

Computerized systems. Computerized systems are well established in pharmaceutical manufacturing. The validation of these systems within the GMP environment was outlined by a Pharmaceutical Manufacturers Association committee many years ago (9). The original document described a life-cycle approach that was openly adapted from the software-development life cycle. The similarity between the computerized system and draft process validation guidance life cycles is primarily in their cradle-to-grave treatment of the system being validated.

The similarity ends there. The methods for software and hardware development for computerized systems are substantially different from those recommended for process validation. The software and hardware elements of computerized systems roughly parallel process methods and process equipment in the production environment. The equipment components have some degree of similarity, but software development bears little resemblance to the development of a pharmaceutical process on a small scale. The performance qualification of computerized systems in the pharmaceutical industry should be considered a background activity and has neither a requirement for repetitive confirmation nor a statistical component. The real proof of a computerized system in the pharmaceutical industry is in the end-product, be it a completed product, test result, or data field that resulted from the operation of the system. The draft guidance is an extremely poor fit for computerized systems because the testing and documentation required is so different from that associated with process and product situations.

In-process inspection of product attributes. The inspection of pharmaceutical products is an important part of the overall process for their preparation. The purpose of the inspection is to identify and remove nonconforming units before further processing or final release. Inspection processes are quite varied. They include:

  • Inspection of filled parenteral containers for visible particulates

  • Inspection of uncoated and coated tablets

  • Lot-number and expiry-date inspection for printed materials.

The design and development of inspectional systems can be a hybrid of process development and equipment design. The use of DOE practices can be effective in establishing the inspection parameters for automated systems. When the inspection is performed by qualified personnel, a certain amount of that evaluation is possible, but because the operator's inherent diligence and visual acuity are perhaps of greater importance than the controllable inspection parameters, statistical inferences are less useful. Initial and ongoing qualification of these systems is reconfirmation of the optimized inspection parameters in commercial settings with real-world defects. The life-cycle model for process validation appears to fit inspectional systems reasonably well.

Equipment cleaning and component preparation. Containers and closures for parenterals require cleaning before use to remove trace particles and endotoxins. In addition, some of these items may be treated with silicone-based lubricants to facilitate assembly and use. Nondisposable equipment in contact with the product must be cleaned before use in the next lot and subsequent procedures. Cleaning and preparation processes are production processes in which the objective is to provide items or surfaces that are free of contamination. From that perspective, the design and development for cleaning and preparation procedures are well suited to the draft guidance. The initial and ongoing validation of these processes in the manner outlined for products and processes is possible as well, though the expectations defined in the guidance do seem somewhat excessive for these background processes.

Sterilization processes. All parenteral products include several sterilization processes for the formulation, equipment, and supportive elements. Depending upon the particular sterilization process, the number of parameters to be evaluated can be substantial, and the DOE or multivariate experimental designs mentioned in the design stage of the guidance may be appropriate. Included in most sterilization-cycle development efforts are activities such as component or container mapping, load mapping, and confirmation of material quality and stability postexposure. These activities are well established at many firms in a formal manner, and the application of more scientific approaches would be beneficial where such knowledge is not well structured. The draft guidance can be applied to the initial and ongoing qualification phases, because it would largely align with the statistical approaches for evaluating products exposed to the sterilization process.

The difficulty in applying the guidance to sterilization stems from the biological destruction aspects of these processes. Although statistics are commonly used to calculate sterility assurance levels and the probability of a nonsterile unit, these values are focused on biological-indicator results and usually do not consider the bioburden to a significant extent. The bioburden resistance and population in the commercial-scale process are of critical importance, but development knowledge cannot be translated to a larger scale easily. Sterilizing filtration, which is widely used, represents an even greater challenge because information about the bioburden is essential, and little can be done from a design or development perspective to support it more fully in a commercial setting.

The rigors expected in development and the use of a life cycle from the guidance certainly fit sterilization, but it seems inappropriate to force statistics onto the actual qualification of the sterilization processes themselves, especially the biological destruction components of those processes. Most sterilization processes are validated using worst-case conditions and highly resistant biological indicators, which represent extreme challenges for the process. Does the imposition of statistical expectations add any real value?

Aseptic processing. Successful aseptic processing relies on various elements. Facility design, equipment design, numerous sterilization procedures, environmental decontamination, aseptic assembly, and aseptic technique are some of the more prominent components (10). Success with aseptic processing relies on attention to detail in these and other areas. Unlike formulation processes, where the influence of the independent processing parameters can be assessed by reviewing the dependent quality attributes, aseptic processing lacks any meaningful correlation potential.

On a basic level, there are simply too many independent—and interrelated—variables in aseptic processing, and the most meaningful one of all lacks metrics of any type. Personnel are an integral part of virtually all current aseptic processing systems. Their effect on the most important aspect of all, sterility, is also the greatest, and creates the greatest opportunity for failure. Aseptic processing performed by human operators is devoid of any measurable variable that could be used to predict the outcome.

Another concern regarding the adaptation of the draft guidance to aseptic processing is perhaps even greater. In 2008, FDA proposed a change to the CGMP regulations that required aseptic processing to be validated. To many in industry, this expectation could not reasonably be fulfilled with the available tools. The author provided the following comment to FDA:

Aseptic-processing simulations cannot validate an aseptic process. The results obtained demonstrate the capability of the facility, equipment and operational controls to provide a minimal microbial contamination rate in a single event. They cannot be utilized to predict the outcome of a similar process performed at a different time, and thus cannot 'validate' the aseptic process. Successful aseptic processing incorporates a myriad of necessary controls; however these controls, alone or in concert, cannot be relied upon to support the absence of microbial contamination as is routinely accomplished in sterilization validation.

There is sufficient support within the CGMP regulations [implied in 211.113(b)] and guidance documents for aseptic processing (explicit in the 1994 submission guide and 1986–2004 aseptic processing guidance) to support the periodic execution of process simulations. Considering these as validation overstates their utility, and implying that firms having successfully passed a process simulation have attained a 'validated' condition is inappropriate. Sterilization and depyrogenation processes are validated such that the routine controls utilized with them are adequate to support the efficacy of the process. Aseptic processing, because it relies on controls of limited sensitivity, poor robustness, and a substantially greater involvement of personnel with their attendant variability, is inadequately supported by the currently available in-process controls. To consider it 'validated' overstates our imperfect ability to measure and control it. In addition, it could be interpreted as license to relax the controls that are necessary for its success. Aseptic processing is perhaps the most difficult of all processes to control in this industry; suggestions that it can be 'validated' imply a level of control not yet attainable. (11)

Others made similar observations, and the general thrust of the comments on the proposed change was that the presently available tools to control aseptic processing do not constitute adequate confirmation to be considered validation. The arguments against the proposed revision notwithstanding, FDA made the change to the regulation. The arguments presented above were deemed inadequate, though no clear rationale or supportive scientific evidence to the contrary was offered. In the context of this larger document, the earlier objections to FDA's CGMP revision and its relevance to this draft guidance have perhaps been made clearer, and the specific difficulties more evident.

FDA indicated that media fills, environmental monitoring, routine processing controls, and satisfactory sterility tests constituted a validation in principle if not in fact (12). That position is inconsistent with the concepts presented in the draft validation guidance, which requires substantially more robust evidence of the link between input and results. The difference may seem on one level to be semantic, yet a gulf nevertheless exists between the industry's and regulators' positions with respect to validation of aseptic processing. Regardless of how one views this conundrum, one point should cause little or no confusion: the statistical component of the guidance really doesn't work with respect to linking any process parameters directly to performance.

Manual processes. Operators' skills and proficiency play a major role in the outcome of a substantial number of important pharmaceutical processes such as manual sanitization of equipment and environments and aseptic gowning. In addition, the operator's abilities may sometimes negate other controls that are present. Processes that rely heavily on operator proficiency may not be considered adequately validated, regardless of the outcome. These processes include manual cleaning, wet granulation, sugar coating, and manual inspection. It is hard to conceive that any of these could attain the level of control associated with validation. These processes are likely best considered as verified, given their heavy dependence on the operator.

A final perspective

A substantial amount of clarification, and perhaps broader interpretation, is needed to reconcile the draft guidance with the elements of pharmaceutical validation that do not directly result in the preparation of an intermediate material, drug substance, or drug product. In some situations, the draft guidance certainly can be used without change, but a degree of caution is necessary. Recommendations for a life-cycle approach are certainly acceptable; many pharmaceutical firms had already instituted comparable programs. It seems that the less the particular process results in or influences a material that can be analyzed or is subject to operator variability, the less useful the statistical elements of the guidance are. Without parameters and attributes that are readily quantifiable, using statistics is certainly inappropriate. The design and development experimental evidence works quite well in some instances but appears to be a force fit in other applications, largely according to whether the process outcome is numeric. Successfully linking design and development and initial and ongoing qualification for these processes also seems to depend on the extent to which process success is readily quantifiable.

The qualification and validation model provided in the draft guidance appears fully applicable to validation of products and processes but only partially applicable in other areas. Its usefulness for sterilization and aseptic processing, two of the more important pharmaceutical processes, is highly questionable. An extensive effort to provide a common level of expectations between FDA and industry (as well as within FDA and industry) is urgently needed to clarify how implementation is to be addressed. For sterile products, where FDA theory and industry practice are most divergent, an update of FDA's 1994 Guidance Submission for Sterile Products might be the best way to accommodate the desired approach with current methods for validation (13).

Conclusion

The draft guidance about process validation is refreshingly simple and supports good science, yet it is demanding with respect to the level of effort required to properly validate a pharmaceutical production process. The variety of approaches that the pharmaceutical industry uses for process validation extends from the merely cosmetic, providing little if any real support to product quality, to overblown efforts that have nearly crippled firms with their complexity and restrictive approach. The draft guidance is not perfect, and improvement and clarification are certainly necessary, especially regarding the guidance's application to sterile products. The guidance requires a restructuring of validation programs to tie development science more closely to commercial-scale manufacturing. It clarifies FDA expectations for validation in a more coherent manner than previous documents. The expectation for life-cycle treatment with heavy statistical emphasis mandates that sound science be applied to validation more clearly than ever before.

For applications in the validation of processes and products that result from them, the author commends those who prepared the guidance for a job well done. It is essential, however, that FDA and industry proceed with special caution in the areas reviewed above because blind adherence to the concepts of the draft guidance will likely lead many astray.

James Agalloco is the president of Agalloco and Associates, PO Box 899, Belle Mead, NJ 08502, tel. 908.874.7558, jagalloco@aol.com

References

1. FDA, Draft Guidance for Industry—Process Validation: General Principles and Practices (Rockville, MD, Nov. 2008).

2. J. Agalloco, "The Validation Life Cycle," J. Parenter. Sci. Technol. 47 (3), 142–147 (1993).

3. K. Chapman, "The PAR Approach to Process Validation," Pharm. Technol. 8 (12), 24–36 (1984).

4. R.E. Madsen, "Real Compliance and How to Achieve It," PDA J. Pharm. Sci. Technol. 55 (2), 59–64 (2001).

5. Code of Federal Regulations, Title 21, Food and Drugs (General Services Administration, Washington, DC, September 2008), Part 211, pp. 51919–51933.

6. FDA, "Pharmaceutical CGMPs for the 21st Century—A Risk-Based Approach," Final Report (Rockville, MD, Sept. 2004).

7. ASTM, "E 2500-07 Standard Guide for Specification, Design, and Verification of Pharmaceutical and Biopharmaceutical Manufacturing Systems and Equipment," (ASTM, West Conshohocken, PA, 2007).

8. J. Agalloco, "Compliance Risk Management: Using a Top Down Validation Approach," Pharm. Technol. 32 (7), 70–78 (2008).

9. J. Harris et al., "Validation Concepts for Computer Systems Used in the Manufacture of Drug Products," in Proceedings: Concepts and Principles for the Validation of Computer Systems in the Manufacture and Control of Drug Products (Pharmaceutical Manufacturers Association, Chicago, 1986).

10. PDA, "Process Simulation Testing for Aseptically Filled Products, PDA Technical Report #22," PDA J. Pharm. Sci. Technol. 50 (6), supplement (1996).

11. J. Agalloco, Comments to FDA, submitted Feb. 5, 2008, Docket No. 2007N-0280.

12. Federal Register, 73 (174), pp. 51919–51933, Sept 8, 2008.

13. FDA, Guidance for Industry for the Submission Documentation for Sterilization Process Validation in Applications for Human and Veterinary Drug Products (Rockville, MD, Nov. 1994).