Prospective Validation

Prospective validation should be conducted prior to the commercial distribution of an API produced by a new or substantially modified process. This validation approach should involve obtaining and evaluating documented processing and analytical control information for multiple batches manufactured, sampled, and tested according to a preestablished validation plan.

When prospectively validating a process, data from laboratory- and/or pilot-scale batches should identify critical quality attributes and specifications, critical steps, control ranges, and in-process tests. Scale-up batches could be used to generate data to confirm or refine earlier work, and production-scale batches should provide data showing consistency of the process.

The number of process runs should depend on the complexity of the process or the magnitude of the process change being considered. Although three consecutive, successful production lots should be used as a guide, there could be situations where additional process runs are warranted to prove consistency of the process (e.g., complex API processes or API processes with prolonged completion times). If a validation lot fails for reasons unrelated to process performance (e.g., power failure or equipment breakdown), that lot should be removed from the validation study and an additional validation run conducted.

Validated analytical methods capable of quantifying API quality attributes should be used during process validation. Resulting data should be evaluated and included in a validation report approved by the same organizational units that approved the protocol.