Bioequivalence and Bioavailability Forum

Dear Xipei and Srinivas,

❝ IMHO adding the ISR samples in the next batch of the unknown ones would not be considered as a good option.

Why ?

What is required is to re-analyse the samples in a separate run on a separate day. The usual practice is to analyse the ISR samples in dedicated runs, because there are quite a lot of samples to analyse. In this case it would make little sense to have a run with a calibration curve and 2 sets of QCs just for 3-4 samples. A pragmatic approach would indeed be to add them to the next routine run.

It would indeed be safer to start analysing the samples while they are in their demonstrated stability period, if the analyte is known to be unstable. Once you have obtained extended stability data you may be able to wait for more time and analyse more samples in dedicated runs if preferred. Still, it is good to have at least some data early in the trial: if something goes wrong it is always better to discover it in an early phase than at the end of the study, especially in a patient study with a long recruitment time.