Bioequivalence and Bioavailability Forum

Hi Fabrice,

❝ ❝ Doesn’t matter because we are interested in PK (I) and safety (II).

❝

❝ OK but then how to justify in the dossier the extrapolation of some early phase outcomes, e.g. a food effect or an efficacy and/or safety exposure signal... should these key findings be evaluated/demonstrated again with the final formulation?

An example: The NDA of idelalisib. A lot of Phase I/II studies (women/men, Japanese/Caucasian, food effect, DDIs, renal/hepatic impairment) were performed in 6 (six!) to 12 subjects with early formulations.
There was also a study comparing formulations (n = 15, extremely underpowered for the x̃ CV of 29% in earlier studies). Some comparisons failed (e.g., 90% CI of C_max 106–153%). Funny:

2.5.2.2 What are the safety or efficacy issues, if any, for BE studies that fail to meet the 90% CI using equivalence limits of 80–125%?
None. The exposure for the different drug products is similar.

Quoting nobody: It’s originator, stupid!

Since idelalisib is in BCS II (where dissolution cannot predict in vivo performance), the FDA lamented about later manufacturing changes but accepted the package at the end of the day.