Get Social With Us

The evolution of process validation

When FDA published its draft guidance Process Validation: General Principles and Practices in November 2008, the response from industry was so strong that the agency extended its deadline for comments well beyond the normal two months. Almost two years later, the guidance remains in draft form despite FDA’s assurance in June of this year that issuing the final guidance is a priority for 2010. The comments posted on the regulations.gov website regarding the draft guidance were not unexpected, such as highlighting concerns over the need to harmonize the recommendations with ICH Guidelines, clarifying language in the Guidance, and a desire not to abandon the traditional reliance on three consecutive conformance batches as an essential component of process validation. Others expressed concern over the potential for confusion in the use of the term “qualification” and a lack of guidance on the manufacture of investigational products.

 

In our view, addressing these concerns and comments should not present much of a problem to those responsible for finalizing the Guidance. It appears that the agency has decided that, as with other guidance documents, leaving the document in draft form may be more appropriate at this point than issuing a final version. FDA has a history of leaving guidance documents in draft form for a long time as this gives them (and industry) the flexibility to either do exactly as the guidance suggests or to do something slightly different. As written, the draft guidance is fairly well aligned with other (ICH and FDA) guidances, risk-based approaches to validation, QbD, etc., so that it is unlikely FDA will change their view much in the near future. A draft guidance gives companies more flexibility in how they interpret its contents and they can’t be faulted for not following it to the letter.

While industry is waiting for the guidance to be finalized, our advice is to concentrate on assuring that early development (process design) work yields data that can be used to support decisions on acceptable process parameter ranges. Once the process steps have been defined, follow this with a risk evaluation of each step to identify those process parameters which must be controlled in order to maintain product quality. When these critical process parameters have been identified, a series of DoE-based runs using a well-qualified scaled-down process model can be used to evaluate the acceptable ranges for these parameters and their effect on critical quality attributes. With sufficient data to enable meaningful statistical analysis of the results, followed by confirmation at full scale to ensure the parameter control ranges produce product that meets specifications, a complete validation package can be assembled. The draft Guidance also emphasizes the need to perform additional in-process testing during these runs, again to ensure process and product consistency. Thus, the number of consistency lots necessary for process validation will depend on the extent to which the processes can be shown to be in control by these analyses. Following marketing approval and commercialization, statistical analysis of in-process control data and final product test results should continue, to a greater extent than in the past, to ensure that the manufacturing process is maintained in a state of control.

 

Blog article by: Howard Levine, Alex Kanarek, James Blackwell, and Patti Seymour