Wednesday, October 27, 2010

USP Discussion on RMMs and recommendations for USP Chapter 1223

The final session of this year’s PDA Global Conference on Pharmaceutical Microbiology is focused on the direction that the USP will take with the existing informational chapter 1223, Validation of Alternative Microbiological Methods. Questions were provided to the meeting attendees and here is an excerpt of some of the comments and responses:

Question: Has USP 1223 Validation of Alternative Microbiological Methods been useful or an impediment for the selection, validation and implementation of RMMs?

Response 1: Yes, the chapter has been useful as providing guidance for RMM validation.

Response 2: Yes, but we would like to have more guidance on sensitivity and limit of detection strategies. For example, methods for developing very low inoculum levels, such as 1 cell.

Response 3: The use of statistics for each validation criteria should be more clearly defined and relevant.

Response 4: Much work has been done on limit of detection for sterility testing and these methods are appropriate for use and should be incorporated into the USP guidance.

Response 5: There needs to be a good balance between specific guidance, such as acceptance criteria, and background information into the reasons why the guidance is provided. The chapter should not be a white paper.

Response 6: We need a simple benchmark on how to validate RMMs. The food industry and AOAC have addressed this 20 years ago!

Question: Do you like to see examples?

Response 1: In the EU, regulators took the example in Ph. Eur. 5.1.6 as gospel and that caused many problems for companies validating RMMs. As a result, the next revision of 5.1.6 will not contain an example but will be published in PharmEuropa.

Response 2: I want to see real, practical and successful case studies of how RMMs have been validated, not theoretical examples.

Response 3: Recommendation is not to have an example but to put this information in Pharmacopeial Forum and/or reference other guidance documents that will provide examples, such as PDA Technical Report #33.

Response 4: I want to see guidance on controls based on technology platforms and applications (e.g., negative and positive controls). But don’t provide specific examples, but more suggestions on good controls that should be run.

Question: Should compendial guidance documents (JP, USP, and EP) on the validation of RMMs be harmonized?

Response: By a show of hands, many in the audience said yes.

Final summary statements from the USP based on these discussions:

We may be making this much harder than it should be. We need to go back and take a closer look at our recommendations in this chapter and provide a more “user-friendly” set of recommendations. Because there are many different processes and products that would utilize RMMs, it is really up to you, the users of RMMs, to define how to best validate these new systems. USP intends to provide better guidance on the use of alternative methods with input from stakeholders.

USP 1223 needs to be revised. There needs to be a clarification of sensitivity, limit of detection and limit of quantification, with specific validation criteria and not sole reliance on parallel testing. The use of CFUs is difficult because there is no good quantitative definition of what a CFU is. Statistical models used with validation criteria needs to be revisited. Microorganisms chosen during validation should be appropriate for the intended applications, process and product. We need to consider how to handle slow growing microorganisms, and address the use of appropriate controls.

Next, it appears that specific examples should not be included to avoid regulatory expectations that may not be appropriate. Instead, maybe we should reference PDA TR #33 and/or submit a Stimuli Article. Any reference to specific RMM technologies should not be included.

We should address the relevance of referee tests for short shelf-life products such as biologics.

We should work toward harmonizing all RMM guidance documents; however, this may be easier said than done. This is not a simple matter and could be time consuming. However, we will discuss how we can interact with other organizations to determine if our revisions can take into account other guidance documents.

Finally, we need to have more discussions with the stakeholders of RMMs more frequently than we have in the past, and we will discuss how we can seek out mechanisms to make this happen.

1 comment: