Posting: # 17150
In times where Bayesian techniques come more often I was wondering about BE trials and therefore my question: Has anyone of you evaluated a BE trial using Bayesian techniques? I guess in a confirmatory setting the answer of our regulators is clear: not acceptable. Or not? Ok, maybe as a secondary analysis. What was the feedback from the agency? What kind of prior did you use? (interesting would of course be a case which is not non-informative like in [1, Section 5.9])
Experiences or ideas are much appreciated.
 Patterson, Jones. Bioequivalence and Statistics in Clinical Pharmacology. Chapman & Hall/CRC, Boca Raton 2006.
Posting: # 17153
If current frequentist methods for studies where bayesian analysis may fit (two-stage) are full of controversies in the regulators point of view, I can't even imagine if you introduce bayesian methods to them in those cases
I'm not a frequentist phanatic. I believe both approaches are valid depending on the problems and study designs under consideration. Regarding bioequivalence however, I don't see how bayesian inference may add any value. The same is not true for therapeutic equivalence studies (biosimilars), where I believe bayesian methods would be very advantageous since you already "have" prior "information" regarding treatments.