Creating Meaningful Protocols: Key Recommendations from FDA and EMA
Early in my career I had a Sr. VP tell me, quite seriously, after I presented CMC timelines for a given project that “the problem with timelines, Mike, is that people expect you to keep them”. I had reason to recall this Captain Obvious moment recently when discussing with colleagues how to set acceptance criteria for a Process Performance Qualification (PPQ) protocol. Protocol authors seem to be afraid to set meaningful acceptance criteria in their protocols for exactly the same reason. Not that I can blame them. Failing acceptance criteria in a PPQ protocol will generate headaches.
However, validation philosophy has evolved greatly over the last decade or so. There have certainly been new guidance documents issued on the subject, from the FDA in 2011, to the EMA in 2014 and the PDA in 2013. A thoughtful reading of these documents provides some insight into how to generate more meaningful, yet generally achievable, acceptance criteria.
The FDA guidance document states that the PPQ protocol should contain:
- “criteria and process performance indicators that allow for a science- and risk-based decision about the ability of the process to consistently produce quality products” and there should be
- “a higher level of sampling, additional testing, and greater scrutiny of process performance than would be typical …”.
The EMA guidance document advises that controls during PPQ:
- “are expected to go beyond the routine control system as described in S.2.2 and S.2.4”,
- validation of the upstream process should “focus on the confirmation of consistency of performance indicators and quality attributes” while,
- validation of the downstream process should “confirm the clearance capability”, and
- demonstrate that the process is “able to consistently generate the targeted quality of process intermediates and active substance”.
This language certainly suggests that much more is expected than simply following the batch record and meeting batch drug substance (BDS) specifications.
In my role at BPTC, I recently reviewed a PPQ protocol with acceptance criteria that stated “All batch record ranges and IPCs must be met” and “The BDS from all conformance lots must meet specifications”, but little else. I wondered whether implementation of this protocol would be a valueless exercise. I had to ask “Don’t all commercial lots need to follow the batch record and produce BDS that meets specifications?” “Of course”, was the answer. I followed up with “Then what’s the purpose of the PPQ protocol?” The response was “That’s the way we have always written the acceptance criteria”. Unfortunately, almost all the PPQ protocols I’ve seen over the years were written the same way.
Given the above observation, how then should we set the acceptance criteria? While we want to make the criteria meaningful, we certainly do not want to fail. That’s something that Sr. VP apparently did not understand: I crafted those CMC timelines with the expectation that they would be met. The same must hold true for the acceptance criteria in PPQ protocols. This is a topic for another blog. I’d be interested in your opinion about how to improve PPQ protocols.
Blog article by: Mike Glacken, Senior Consultant