Stage 2 Process Validation: Regulatory Expectations and Approaches to Determine and Justify the Number of PPQ Batches

Kuklos/shutterstock.com

The biopharmaceutical sector has been growing rapidly in the pharmaceutical industry. A key driver for its growth is an increasing demand for biotechnology-derived drugs to prevent, treat, and/or completely eradicate a number of chronic diseases. The global biopharmaceutical market, which was valued at $162 billion in 2014, is predicted to grow at a compound annual growth rate of 9.4% from 2014-2020 to reach $278 billion (1). Manufacturing of biopharmaceutical products, however, is a highly specialized and complex process. One regulatory requirement is to validate the process (2, 3), which is expected to ensure that the process consistently deliver quality products. 

FDA’s 2011 guidance for industry (2), which recommends a lifecycle approach to process validation, includes three stages (Stage 1-Process Design, Stage 2-Process Performance Qualification [PPQ], and Stage 3-Continued Process Verification) over the lifecycle of the process and product. In Stage 2, the manufacturing process, as designed and developed in Stage 1, is evaluated to determine if it consistently produces products that have high quality, safety, and efficacy via manufacturing of conformance lots/PPQ batches. This guidance, however, does not recommend a specific number of PPQ batches required nor specific methods to determine the number of batches. Biopharmaceutical manufacturers are expected to determine and rationalize the number of PPQ batches based on a reasonable approach (2, 3). In this article, the authors have sought to summarize the current regulatory expectations regarding the number of PPQ batches and the authors’ critical perspective on science- and risk-based and statistical methods used to determine and justify the number of PPQ batches. Conceptual evolution of process validation is also summarized.

Conceptual evolution of process validation

The process validation concept was first introduced by Byers and Loftus in the 1970s (4, 5). This concept became a specific guideline when it was adopted and included in FDA’s first guidance for industry in 1987 (4-6). The historical concept was conceived to represent consistency and reproducibility in the process conditions that were used for the manufacture of a drug product without necessarily defining, understanding, or controling critical aspects of the process or critical quality attributes (CQAs). One important aspect (i.e., inherent variability associated with the manufacturing process), however, remained unidentified. 

Recent guidelines introduced by FDA and the European Medicines Agency (EMA) represent a paradigm shift in the process validation approach (2, 3). These guidance documents incorporate the lifecycle concept into the validation process. The lifecycle-based approach involves a sound science- and risk-based approach to identify the critical process parameters (CPPs) and failure modes, establish robust control strategy, and meet the defined CQAs (2, 3, 7). Both agencies now consider the validation as an exercise that is meant not only to achieve process understanding and consistency, but also to demonstrate ongoing verification, which ensures the process remains within its validated design space and consistently produces product that meets all specifications (2, 3). The differences between the FDA’s historical and lifecycle-based approaches are provided in Table I.

Table I: Key differences between FDA's historical and lifecycle-based approaches to process validation.

The paradigm shift in FDA’s approach to process validation acknowledges the need for improvement in the manufacturing processes. It has now enabled the manufacturers to better understand the process and implement effective controls to consistently produce quality products over the lifecycle of the process and product. This concept is also in alignment with the QbD approach (8). The FDA and EMA guidances require adherence to current good manufacturing practice regulations, aligning with the International Council for Harmonization Q8 (R2), Q9, Q10, and Q11 (9–12) guidelines. Unlike the FDA guidance, the EMA guidance does not divide the process validation into three stages, but allows for a hybrid approach that combines the new process validation guidance with the traditional approach (3). In case of the FDA guidance, the new guideline supersedes the historical approach (2). 

Current regulatory expectations regarding the number of PPQ batches

Based on the most recent guidelines for process validation introduced by different regulatory agencies and/or talks given by the regulatory agency, the expectations with regard to the number of PPQ batches and rationale are summarized in Table II (2–3, 13–17).  

Table II: Current regulatory expectations regarding the number of process performance qualification (PPQ) batches and justification.

Potential approaches to determine and justify the number of PPQ batches

The FDA and EMA guidances promote the “lifecycle” approach to process validation that includes scientifically sound process design practices, robust process qualification, and continued process verification (2, 3). Although the FDA guidance is less prescriptive regarding the number of batches, there are clear recommendations to use a science- and risk-based approach and objective measures such as statistical methods wherever feasible and meaningful to achieve adequate assurance (2). 

Science- and risk-based approach. The FDA and EMA guidances emphasize the documented evidence that answers some of the key questions, such as: Is there sufficient knowledge and manufacturing experience to demonstrate the design space and robustness of the manufacturing process? Does the manufacturing process consistently produce a product that meets pre-determined specifications? How variable is the manufacturing process? Is the impact of sources of variability on CQAs well understood? What is the level of risk associated with the manufacturing process? Is the manufacturing process in a state of control? 

The following is proposed as a potential science and risk-based approach to determine and justify the number of PPQ batches. Based on the product knowledge, process understanding, and large-scale/clinical manufacturing experience (Figure 1), the overall residual risk level associated with the manufacturing is determined. The overall process risk (low, medium, or high) can be determined on the basis of a reasonable approach (e.g., on the basis of process impact score, risk priority number, or qualitative assessment based on overall product, process, and manufacturing knowledge, process variability, and control strategy). The overall process risk may then be translated into the number of PPQ batches (Table III).

Figure 1: Schematic diagram showing the knowledge used and steps undertaken to determine the number of process performance qualification (PPQ) batches via scienceand risk-based approach. (Courtesy of the Authors)

Table III: Overall process risk, minimum number of process performance qualification (PPQ) batches and authors' perspectives.

Statistically-based approaches. The FDA’s process validation guidance section on Stage 2: Process Qualification includes recommendations and/or expectations related to statistical measures. The section states that FDA “strongly recommends firms employ objective measures (e.g., statistical metrics) wherever feasible and meaningful to achieve adequate assurance (2).” This recommendation makes explicit that FDA gives significant importance to statistical measures in lifecycle-based process validation.  

The manufacturers are expected to provide documented evidence regarding the capability of the manufacturing process that it will consistently produce a finished drug product that meets pre-determined specifications. The evidence can be established through a systematic PPQ plan that involves science, manufacturing experience, risk management, and statistical tools. Statistical tools are an integral part of a comprehensive lifecycle approach to process validation. Hence, it is the responsibility of the manufacturers to ensure that the process is adequately qualified and to demonstrate adequate statistical assurance.  

After the introduction of the lifecycle approach, the question of number of batches required for PPQ campaign does not have a clear answer. The pharmaceutical industry has embraced this challenge and continuously making efforts to identify appropriate strategies. For example, the International Society of Pharmaceutical Engineering (ISPE) has published two discussion papers that describe different statistically-based approaches (Table IV) to determine and justify the number of PPQ batches (18, 19). The ISPE papers are part of frameworks to translate the overall process risk into appropriate number of PPQ batches and gain statistical confidence. Although these approaches do not represent a consensus position within the industry, the two papers have provided different options for manufacturers and stimulated discussions for further improvements. 

Table IV: Summary of statistically-based approaches (described in International Society for Pharmaceutical Engineering [ISPE] chapters) used to determine and justify the number of process performance qualification (PPQ) batches (19, 20)[Click to Enlarge].

The selection of a suitable statistical method is influenced by a number of factors including manufacturer’s need, simplicity, benefits and limitations associated with the method, and flexibility to incorporate prior knowledge (e.g., process knowledge, manufacturing experience, control strategy, and overall process risk) in the calculation. Each method involves some assumptions and has its own benefits and limitations. The authors have performed a critical review of the ISPE methods and summarized in Table IV. The determination and justification of the number of PPQ batches are evolving topics and there is no industry standard yet. Hence, the manufacturers can modify these methods as appropriate for their need. 

After the introduction of the lifecycle approach, the question of number of batches required for PPQ campaign does not have a clear answer.

Conclusion

Validation of biopharmaceutical manufacturing processes via a lifecycle approach is an expectation from numerous regulatory agencies. The lifecycle-based approach represents both improvisations and challenges in the validation process. Stage 2-Process Qualification in the lifecycle approach is a critical phase, wherein the process is qualified by manufacturing a certain number of conformance lots. There are no clear recommendations in the guidances released by different regulatory agencies regarding the number of PPQ batches or specific methods to determine and justify the number of batches. Different types of statistically-based methods described in the ISPE discussion papers involve some assumptions, and offer both benefits and limitations/challenges. Regardless of the method employed, the critical aspect is to rationalize the number of batches based on product and process knowledge, large-scale/clinical manufacturing experience, control strategy, overall process risk, and statistical assurance/confidence. Efforts to improve the practicality of the methods are warranted.   

Acknowledgement

The authors thank Dr. Myrna Monck (Manager), Dr. George Crotts (Manager), and Dr. Dany Doucet (Senior Scientific Investigator), Department of Biopharmaceutical Product Sciences, GlaxoSmithKline, King of Prussia, for their feedback.

Disclaimer

This article reflects the views of the authors and should not be construed to represent any regulatory agency’s views or policies.

References

1. Persistence Market Research, “Global Market Study on Biopharmaceuticals: Asia to Witness Highest Growth by 2020,” Press Release, Jul 27, 2015.
2. FDA, Guidance for Industry, Process Validation: General Principles and Practices (2011).
3. EMA, Guideline on Process Validation for Finished Products-Information and Data to Be Provided in Regulatory Submissions. EMA/CHMP/CVMP/QWP/ BWP/70278/2012-Rev1,Corr.1.
4. T. E. Byers, PDA J. Pharm. Sci. Technol. 32(1), pp. 22-25 (1974).
5. B. T. Loftus, PDA J. Pharm. Sci. Technol. 32(6), pp. 268-272 (1978).
6. FDA, Guideline on General Principles of Process Validation (1987).
7. R. Sharnez, BioPharm International 28 (1), pp. 32-34, 45 (2015).
8. A. S. Rathore, H. Winkle, Nat Biotechnol. 27 (1), pp. 26-34 (2009).
9. ICH, Q8(R2) Pharmaceutical Development (ICH, August 2009), www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q8_R1/Step4/Q8_R2_Guideline.pdf
10. ICH, Q9 Quality Risk Management (ICH, Nov. 9, 2005).
11. ICH, Q10 Pharmaceutical Quality System (ICH, June 4, 2008).
12. ICH, Q11 Development and Manufacturing of Drug Substances (Chemical Entities and Biotechnological/Biological Entities) (ICH, May 1, 2012).
13. European Medicines Agency, Guideline On Process Validation For The Manufacture Of Biotechnology-Derived Active Substances And Data To Be Provided In The Regulatory Submission, EMA/CHMP/BWP/187338/2014.
14. Health Canada, Validation Guidelines for Pharmaceutical Dosage Forms (GUI-0029), 2009.
15. 2014 CMC Strategy Forum Japan Wrap-Up Summary Notes, Presentation [Korea Ministry of Food and Drug Safety].
16. K. Oyama, PhD, “PMDA Perspective: Regulatory Updates on Process Validation Standard,” Presentation at CMC Strategy Form Japan 2014, Tokyo, Japan, Dec. 8-9, 2014.
17. ASEAN, ASEAN Guideline on Submission of Manufacturing Process Validation Data for Drug Registration, 2012.
18. ISPE, Discussion Paper: Topic 1-Stage 2 Process Validation: Determining and Justifying the Number of Process Performance Qualification Batches (2014).
19. ISPE, Discussion Paper: Determining the Number of Process Performance Qualification Batches Using Statistical Tools-Supplement to Prior Discussion Paper (2016). 

Article Details

Pharmaceutical Technology
Vol. 41, No. 6
June 2017
Pages: 56-61

Citation

When referring to this article, please cite it as K. G. Desai et al., ""Stage 2 Process Validation: Regulatory Expectations and Approaches to Determine and Justify the Number of PPQ Batches," Pharmaceutical Technology 41 (6) 2017.