Number of vials to sample for parenteral assay

[domino1]

For those of you working on parenterals, how many vials do you sample for assay determination for release and stability (if the two are different) for early stage development compounds?

For example:

Single vial, duplicate preparations, single injection each prep.

I know I have seen a paper discussing this issue but can't find it. I don't believe that there is a regulation on it.

I just wanted to know what others out there are doing.

Thanks!

Domino

[KarenJ]

Domino,

It depends on what you want your data to show. If you are releasing a batch, you probably want to show an estimate of the vial-to-vial variability, especially in early phase development, so that you have an idea about how well the process is working. If you make multiple preps out of one vial, you are only showing the variability of your sample prep technique. Multiple injections from one prep only shows how well your instrument is working. For stability testing you probably want multiple vials also because you might see some variability in the sample degradation patterns. I would analyze three vials from each batch/lot.

KarenJ

[DR]

Assuming that:
the batch is a run of vials all filled from one tub,
some sort of inprocess assay shows units at either end of the run to be identical,
the assay method(s) are validated,

I would go with 1 prep of 1 vial for release and stability (although I would have reserve samples available for stability). If you have had any packaging related quality issues, you might want 3 units instead (stick with the single prep, single injection scheme).

[domino1]

We are doing in-process so that should verify we have a uniform batch.

I think the real issue is if we feel the need to "prove" that we have a uniform batch. For solid oral we do content uniformity but we do not for parenteral (which is probably a good assumption - but may not be the case for all). USP does not require it but USP is not a replacement for good science.

In early stage development you often do not have a lot of API so you do not have a lot of extra vials to do 3 at every timepoint.

There is lots to think about. Thanks for the feedback, I just want to know what others are doing.

Lisa.

[Alex Buske]

If properly prepared I would rate parenterals as homogenous. I would however avoid single preparations for analysis. If accidental errors occur during sample preparation only multiple prparations would reveal that.

Alex