Validation Failure

3 validation batches were prepared for tablet product, the first two batches passed, but the third one was not conform with the
required specification during compression, but the reason was clear as a rework procedure was performed for this third batch, which means a significant quantity of rejected tablets from the first batch was grinded and added to the third batch
my question is: should i stay consider the first two batches as validation batches and just prepare an extra batch
or i have to prepare another three consecutive batches?

Dear Gretta,
Your “process” validation is not valid since you have used “two different processes”. As part of process validation, you need to manufacture at least three batches (as per your BMR) to prove its reproducibility. In this case, produce one more batch (number it as batch number 3) and prepare a validation report. This is nothing but violation of GMP but you don’t have second option. If you consider the new batch as batch number 4 and if you mention the deviation in your validation report, nobody will accept it and it will become a big headache to you in future.
To maintain the quality of your product, never mix the rejects of previous batch in next batch. Ask your R&D team to develop a new process for rework of rejects and validate it as a separate process. Also, apply frequent testing and stringent test methods to ensure the quality of reworked batch.


Thanks for youe help

There is no hard and fast rule that you have to validate 3 batches.

If you fail during validation always consider to opt for Root Cause Analysis at this stage. You may consider to include your R&D who will have primary authority on developing this technology.
You have to include your Process, Quality and Engineering.
They will check out if you have had missed Kep process control steps, Inprocess Quality control checks, Equipment sizing and working conditions such as Temp, Pressure, Shear, Stress, Material of constuction and other operating conditions.

Always include a quality group and engineering group in this whole validation study unless and otherwise you develop a copied product or generic after a patent expires.

Another most important thing to note is that when ever you follow a QbD process in your R&D the product will not go out of specifications (OOS) during validation or real time manufacturing.

When you have a proper QbD approach you need not validate for 3 batches.
Robustness and consitsancy will be repeatable.