The purpose of the in-process test and the specification is to monitor the process and to make
adjustments to the process. Take for example, a hypothetical case in which you might take a
sample to measure pH against an in-process pH specification and it doesn’t meet your in-process
pH specification. Do you have to stop the process at that point and try to determine why it failed
pH in the first place? Not necessarily, if it failed pH, it failed pH. It needs to be adjusted until the
proper pH is achieved. This is not reprocessing, it is processing to a specification. Since this is
not reprocessing, you would continue to monitor and adjust the pH as necessary in order to meet
the specification for the Intermediate or API.
Regarding reprocess a batch, it actually makes little scientific sense to say you need to limit the
number of times you can do it, at least from a product or material perspective. Generally, where
synthetic chemistry is involved, reprocessing improves material quality since you are doing
multiple purifications. For someone to try to scientifically object to reprocessing something more
than three times, or 10 times, or 100 times, makes no sense. Certainly for biological systems,
further manipulations could lead to product quality degradation and multiple reprocessing could
be undesirable.
The question becomes what knowledge is lacking that causes repeated reprocessing. Why this is
potentially important it is not immediately nor should it be a GMP issue. It is an important
economic issue.
