thebench.bio #1 Ecosystem for the Bio-Tech Industry
All Process Validation questions

Question 16 of 20 Intermediate Troubleshooting

A PPQ batch fails an acceptance criterion. What happens next?

Short answer

Record it as a deviation under the protocol, investigate the cause, and decide whether the failure undermines the process design or is an isolated, assignable event. You cannot drop the batch and run another to reach three passes. The report must address it, and PPQ may need to be repeated after correction.

Why

The FDA's 2011 guidance says data should not be excluded from PPQ without a documented, science-based justification, and that the PPQ report must summarize and discuss all nonconformances, including deviations and aberrant results, before concluding whether the process is in a state of control. Annex 15 requires validation protocols to define acceptance criteria based on development data. A failed criterion is evidence about the process, and that is exactly what PPQ is designed to find.

The investigation asks whether the failure has an assignable cause unrelated to the process design, such as a mislabeled sample or a failed instrument with confirmed lab error, or whether it reflects real process variability. If blend uniformity fails because the mixing time in the design space is marginal for a new excipient lot, the process design is in question. Then the right move is to correct the process, which may mean returning to Stage 1 work, and repeat PPQ.

If the failure is a true lab error that is documented and confirmed, the protocol's provisions for invalid data apply, and the report explains it. Either way, the batch's own disposition is a separate decision. The common failure is to call the batch an outlier, make an extra batch and report three passing batches, which inspectors read as testing into compliance.

How to work it through

  1. Raise a deviation against the protocol and quarantine the batch.
  2. Investigate using the OOS or deviation procedure to find whether there is an assignable cause.
  3. Decide whether the cause affects the process design or is isolated and documented.
  4. If the design is affected, correct the process through change control and plan repeat PPQ.
  5. Address every failure and exclusion explicitly in the PPQ report.
  6. Decide the batch's disposition separately on its own evidence.

An example

The second of three PPQ batches fails the protocol criterion for stratified content uniformity at the end of compression, with RSD 6.1 percent against a limit of 5.0 percent. Investigation shows segregation in the hopper as it empties, a mechanism that applies to every batch. Batches 1 and 3 passed only because they were run with fuller hoppers.

The team adds a minimum hopper level as a CPP through change control, demonstrates it at pilot scale, and runs three new PPQ batches. The report explains the original failure and why the new design addresses it.

Where the answer stops holding

A failure with no clear cause leaves uncertainty that more batches may or may not resolve. Repeat PPQ delays supply, which creates pressure to explain failures away. Some failures relate to acceptance criteria set too tightly, and revising criteria after the fact must be justified scientifically, not by the result.

A strong answer covers

  • Deviation under the protocol, never silent exclusion
  • Assignable cause versus process design weakness
  • Correct and repeat PPQ when design is affected
  • Every failure discussed in the report
  • Batch disposition decided separately

Common mistakes

  • Running an extra batch and reporting only the three that passed.
  • Calling a failure an outlier without a documented assignable cause.
  • Loosening acceptance criteria after seeing the failing result.