Guest Column | August 7, 2020

A Functional History Of Process Validation, Part 1 – A Weak Foundation

By Mark F. Witcher, Ph.D., biopharma operations subject matter expert

Using process validation (PV) is critical to building pharmaceutical processes that reliably produce high-quality products. Most of the pharmaceutical industry appears to undertake PV activities to satisfy regulatory requirements rather than as an integral part of efficiently and reliably developing effective manufacturing processes. Treating PV as a compliance exercise is a serious mistake because PV is one of two foundational paradigms that greatly increase the likelihood of quickly and efficiently developing fundamentally sound processes. Even worse, some companies treat PV and process characterization as a post-process development compliance requirement, adding more time and money to the product development effort. While PV is critical to getting things done by building the process, the second foundational paradigm is risk management to stay out of trouble by assuring the process has high reliability through minimizing the likelihood of failures.1

This article is the first of a two-part series that describes the origins of PV to explain the underlying concepts necessary to control the advanced bioprocess manufacturing technologies required to make the next generations of biopharmaceutical therapies.2 As the pharmaceutical industry expands from chemicals through biopharmaceuticals to cell and gene therapies, the importance of PV is becoming better appreciated as the key to successfully making high-quality products. The history of PV is discussed only to the extent necessary to evolve PV to meet these challenges.

History Of Process Validation

Initially, pharmaceuticals were manufactured and given to patients directly without controlling product quality, resulting in obvious adverse consequences. The problems encountered led to the creation of QC testing functions, regulatory agencies such as the FDA, clinical testing prior to approval, good manufacturing practices (GMPs), and other controls to prevent safety and efficacy disasters.3 As pharmaceuticals became more complex, it became apparent that simply testing the product was not sufficient and that a greater emphasis needed to be placed on the processes used to make the products. Even the comprehensive use of GMPs to control how the product was manufactured did not sufficiently control the quality of the product.

From the very beginning of biological products, it was readily apparent that many important quality attributes of the product were not directly measurable for release testing and that these unmeasurable attributes were critical to the product’s safety and efficacy. This understanding led to a concept that a biopharmaceutical product is significantly defined by its manufacturing process and thus the process needed to be both tightly controlled and validated.

In 1987, the FDA released the first process validation guidance.4 The fundamental approach was testing the process to assure it worked, along with periodic retesting of the manufacturing process to assure it was continuing to work. The fundamental flaw of the guidance was readily apparent in that biopharmaceutical processes are too complex to be defined and tested using a reasonable set of ranges for the multitude of parameters. In addition, some of us found out very quickly that periodic testing was a very poor way of assuring process performance. Any failures of the periodic tests, no matter how small, unleashed a flood of unanswerable questions with respect to when the problems began and what could be done to assure the quality of the product that had been manufactured. Some intrinsic method had to be found for the process to immediately signal if it might not be working properly.

In the mid 2000s, regulatory agencies developed ICH Q8 – Pharmaceutical Development in an effort to provide foundational concepts required for companies to develop high-quality manufacturing processes. The final 2009 guidance provided a number of definitions intended to provide the basis for very important concepts for building processes.5 The most important PV concepts begin with the following terms defined in ICH Q8 (R2):

  • Critical Process Parameter (CPP)
  • Critical Quality Attribute (CQA)
  • Design Space (DS)
  • Control Strategy (CS)
  • Quality by Design (QbD)
  • Real-Time Release Testing (RTRT)

The definitions of these terms turned out to be less than adequate for establishing the underlying concepts, causing a great deal of confusion within the industry. In an effort to be flexible, the broad definitions often created more chaos than enlightenment because they are subject to a wide range of interpretations based on experience, biases, and preferences. In addition, acknowledging that developing and manufacturing pharmaceuticals was primarily a risk management and control exercise, regulatory agencies in 2006 developed and issued ICH Q9 – Quality Risk Management.6 Unfortunately, ICH Q9 lists a number of risk management methods that do not work, primarily because they either do not include a risk’s uncertainty or treat uncertainty improperly by combining it with the risk’s severity.7

With PV in a state of confusion, the FDA issued its 2011 Process Validation, General Principles and Practices guidance.8 A revolutionary document, it describes validating a manufacturing process in three stages:

  • Stage 1 – Process Design: The commercial manufacturing process is defined during this stage, based on knowledge gained through development and scale-up activities.
  • Stage 2 – Process Qualification: During this stage, the process design is evaluated to determine if the process is capable of reproducible commercial manufacturing.
  • Stage 3 – Continued Process Verification: Ongoing assurance is gained during routine production that the process remains in a state of control.8

The 2011 PV guidance draws upon the definitions in ICH Q8, with the very unfortunate exception of QbD, which is left out altogether. With QbD having been so generally defined in ICH Q8, it had by 2009 fallen out of favor because no one could figure out what it really meant or how to use it effectively.9 With PV largely in a state of disarray because of numerous widely varying interpretations of its concepts and usage, it is appropriate to examine the current problems for which solutions need to be sought.

Problems With Current Process Validation

Most of the problems with the current status of PV are attributable to widely varying interpretations of the terms listed above. While the definitions are technically correct, the underlying concepts need to be better described and explained so that a level of common, industrywide understanding can be achieved that makes the terms usable for alignment and consensus.

The following describes the problems with the current FDA 2011 PV paradigm and the related ICH Q8 terminology so the underlying concepts can be evolved and explained to make them more effective.

  • Lack of process definition prerequisite The 2011 PV guidance gets off to a weak start by not precisely defining the goals and requirements that Stage 1 – Process Design must achieve.  ASTM E2500,10 arguably a good definition of good engineering practices, places a strong emphasis on starting any engineering activity with a good definition of what is to be accomplished.
  • Inadequate parameter definition The following ICH Q8 definition of CPP is an extremely important term:

Critical Process Parameter – A process parameter whose variability has an impact on a critical quality attribute and therefore should be monitored or controlled to ensure the process produces the desired quality.5

Although technically correct, it commingles both input and output parameters, leading to considerable confusion with systems where outputs of one unit operation become the inputs to the following unit operation.11 In addition, there are numerous other parameters that may not be “critical” but are nonetheless important for tracking and control to assure the process’ performance. A very clear definition of which parameters are inputs, and which are outputs, for each process step or unit operation is vital to defining control strategies.

  • Weak Design Space descriptionThe following ICH Q8 Design Space (DS) definition contains a simple single sentence followed by three interesting and potentially important policy statements, likely included to clarify the definition.

Design Space – The multidimensional combination and interaction of input variables, e.g., material attributes and process parameters that have been demonstrated to provide assurance of quality.

Working within the design space is not considered a change.

Movement out of the design space is considered to be a change and would normally initiate a regulatory post approval change process.

Design space is proposed by the applicant and is subject to regulatory assessment and approval.5

The DS definition suffers from the same problems as the incomplete CPP definition. In addition, material attribute is not defined, leaving it open to a variety of interpretations. Because the DS definition is central to describing the manufacturing process for operational as well as regulatory purposes, the DS concept must be more completely defined to make it user-friendly, especially for communicating and aligning the DS with regulatory agencies responsible for implementing the last three sentences of the definition.

  • Ineffective risk management methods While ICH Q9 provides valuable insights into the nature and necessity of managing risks, the methods described, particularly failure mode and effects analysis (FMEA), have proven to be ineffective.12 The primary defect of the methods described is that they either do not address or inappropriately deal with the uncertainty of a risk by combining it directly with severity.7
  • Weak control strategy development The term control strategy (CS) is defined in ICH Q1013 and the concept further expanded in ICH Q11.14 While the guidance talks about CSs being based on controlling risks, there is no functional mechanism provided for building a CS based on the results of a QRM exercise. The definition again provides a basic description of controlling manufacturing operations but provides little structure for building the various control systems for manufacturing unit operations necessary to establish and maintain product quality.15
  • Poor QbD implementation After the 2011 PV guidance was issued, QbD apparently was readopted by the regulatory agencies because the industry continued to use the catchy and attractive term. QbD is most often described and used in vague holistic terms, providing more confusion than guidance. QbD was originally a software development term used to describe modular, procedure-based design techniques for developing complex software. QbD has tremendous potential if the original concept of breaking complex problems into manageable pieces can be transferred to developing complex interconnected pharmaceutical processes.
  • Insufficiently designed qualification stage Testing complex biopharmaceutical processes has proven to be a challenge, especially managing the temporary enhanced testing required for qualifying the various process steps. The FDA’s 2011 PV guidance established process performance qualification (PPQ) lots as the primary mechanism for testing the manufacturing process. However, PPQ or conformance lots are more about testing the entire manufacturing enterprise’s ability to successfully operate the manufacturing process and the supporting facility and management systems than actually testing the process. To be more specific, testing a manufacturing process in a commercial facility is beyond risky for everyone involved. The entire purpose of PV’s first stage is to assure the process has a high degree of working at the point the PPQ lots are undertaken. Although classically three lots, the actual number of PPQ lots to qualify the manufacturing enterprise may vary widely depending on many factors. These factors include the extent of Stage 1 process validation, the design of the facility, the experience of the manufacturing organization, and the quality of the enterprises quality management systems.
  • Inadequate verification stage design – Proving to a reasonable certainty that a biopharmaceutical process is working in real time is far more complex than commonly understood by the industry. Statistical methods such as statistical process control (SPC) methods like the Western Electric Rules, originally designed for controlling mechanical processes, are extremely weak because the amount of data generated by a biopharma process is almost always insufficient to provide clear statistical signals in a timely fashion. Methods, frequently based on subjective human judgement, that stand a reasonable chance of providing immediate notification of possible problems in real time are required. To be effective, Stage 3 verification methods for monitoring important process parameters must be built into the process’ design by the scientists and engineers during Stage 1 - Design.
  • Inability to manage legacy products – The current PV paradigm has no clear path or tools for validating old processes used to manufacture legacy products. When an old process needs to be validated because of significant process changes or product quality problems, PV methods should provide an approach for better understanding and describing the process, developing improved control strategies, testing the revised process, and then establishing improved methods for releasing product.

When the above shortcomings are combined, a great deal of confusion is created because companies take widely varying approaches to each PV stage. The complexity and uncertainty add a great deal of communication overhead and ambiguity, both within the company developing the product and between the company and regulatory agencies responsible for understanding the process information on the path to approving the product.

Summary

Process validation must provide direction and guidance for building high-quality processes. The current PV approaches have a number of significant deficiencies leading to a great deal of confusion within the industry. This confusion has led to a wide variety of misalignments between companies and regulatory agencies, negatively impacting the review and approval of new products and their manufacturing processes. Part 2 of this series will describe methods to better define PV to significantly improve one of the biopharmaceutical industry’s most important tools for developing and presenting new products to regulatory agencies for approval.2

References:

  1. Witcher, M. “System Risk Structures: A New Framework For Avoiding Disaster By Managing Risks,” Pharmaceutical Online, July 13, 2020 https://www.pharmaceuticalonline.com/doc/system-risk-structures-a-new-framework-for-avoiding-disaster-by-managing-risks-0001
  2. Witcher, M. “A Functional History of Process Validation, Part 2 – The Key To A More Effective Future,” Pharmaceutical Online, https://www.pharmaceuticalonline.com/doc/a-functional-history-of-process-validation-part-the-key-to-a-more-effective-future-0001
  3. Immel, B. “A Brief History of GMPs for Pharmaceuticals,” Pharmaceutical Technology, July 2001.
  4. FDA (CBER/CDRH) Guidance on General Principles of Process Validation, May 1987
  5. FDA (CBER/CDER) Guidance for Industry, Q8(R2):  Pharmaceutical Development, Nov. 2009, ICH. Rev. 2.
  6. FDA (CDER/CBER) – Guidance for industry: Q9 quality risk management. June 2006. ICH.
  7. Witcher MF. Estimating the uncertainty of structured pharmaceutical development and manufacturing process execution risks using a prospective causal risk model (PCRM). BioProcess J, 2019; 18. https://doi.org/10.12665/J18OA.Witcher
  8. FDA (CDER/CBER/CVM) Guidance for Industry. Process Validation: General Principles and Practices, Jan. 2011, CGMP Rev. 1.  
  9. FDA Understanding Challenges to Quality by Design, December 18, 2009.
  10. ATSM E2500-07, Standard guide for specification, design, and verification of pharmaceutical and biopharmaceutical manufacturing systems and equipment. Aug. 2007.
  11. Witcher, M. F., Why Controlling CQAs Isn’t Good Enough For Gene & Cell Therapies, Cell and Gene Therapy, March 31, 2020, https://www.cellandgene.com/doc/why-controlling-cqas-isn-t-good-enough-for-gene-cell-therapies-0001
  12. Hubbard, D.W., The failure of risk management: Why it is broken and how to fix it, Wiley, 2009.
  13. FDA (CDER/CBER) ICH Q10 – Pharmaceutical Quality System, April 2009
  14. FDA (CDER/CBER) ICH Q11 – Development and Manufacture of Drug Substance, November 2012
  15. Witcher, M. “Developing Optimal Pharmaceutical Quality Control Strategies,” Pharmaceutical Online, June 14, 2019. https://www.pharmaceuticalonline.com/doc/developing-optimal-pharmaceutical-quality-control-strategies-0001

About the Author:

MarkMark F. Witcher, Ph.D., has over 35 years of experience in biopharmaceuticals. He currently consults with a few select companies. Previously, he worked for several engineering companies on feasibility and conceptual design studies for advanced biopharmaceutical manufacturing facilities. Witcher was an independent consultant in the biopharmaceutical industry for 15 years on operational issues related to: product and process development, strategic business development, clinical and commercial manufacturing, tech transfer, and facility design. He also taught courses on process validation for ISPE. He was previously the SVP of manufacturing operations for Covance Biotechnology Services, where he was responsible for the design, construction, start-up, and operation of their $50-million contract manufacturing facility. Prior to joining Covance, Witcher was VP of manufacturing at Amgen. You can reach him at witchermf@aol.com or on LinkedIn.