Bioequivalence and Bioavailability Forum

Hi CECIF,

❝ […] I found a cross-over study comparing pharmacokinetics of biperiden immediate release against extended release.

❝ Is it reasonable to assume the CVwithin of biperiden to be equal to the one on the study that I found?

Difficult to say. Can you give the reference?

❝ I mean, I think it will be smaller or equal at most, therefore if I assume it is the same for the sake of sample size calculation, I could be overestimating sample size, but I would not be sacrificing any statistical power, which is not statistically incorrect, right?

Maybe. Maybe not. Even if sampling times were properly chosen in the study you found, the variability of C_max of the MR product might be lower than the one of the IR product – simply because MR profiles are “flatter”. What we get in a 2×2×2 crossover study is the pooled variability (from the unknown s_wR and s_wT). Given that, it might well be that the CV_w of IR is higher than the one you extracted from the literature.
I suggest to perform a pilot study, a Two-Stage Design, or – if you want to go directly to the pivotal study – allow for a safety margin.