Bioequivalence and Bioavailability Forum

Good evening John!

❝ I meant modified release excluding DR (Delayed Release). Just a hypothetical question that's all. I have experience with DR and we had to show that Tlag was not significantly different as part of the BE criteria. We used Wilcoxon sign rank test.

Yep. Luv nonparametrics. What do you mean by “we had to show that Tlag was not significantly different …”? In the US? Surprise.

❝ What about modified release? As far as I know from a US regulatory point of view this is not considered an issue since BE is AUCs and Cmax only.

AFAIK yes. Maybe it’s better to talk about controlled release (CR); IMHO MR includes both CR and DR (my simple mind would say: CR = everything which ≠ IR). With CR we might cross the border of flip-flop PK (k_a ≤ k_e). AUC_∞ might be problematic, since we are extrapolating absorption – which does not continue forever. Nobody knows what happens after the last sampling time point with a CR formulation. From a PK point of view I can’t imagine a situation where AUC_t, AUC_∞, and C_max are equivalent and t_max is not (i.e, a large difference like in your OP).