Open questions on bioequivalence: some problems and some solutions.

This paper focuses on some specific situations where bioequivalence requires careful attention and tailored protocols in order to overcome intrinsic difficulties either marginally covered or fully neglected by operating guidelines. Some problems congregate with serious difficulties, namely high variability, very poorly absorbed drugs and endogenous substances with their own baseline. With endogenous substances, the dilemma faced is whether to subtract baseline from post-dose values in assessing bioequivalence. Either approach has intrinsic problems and is somewhat puzzling. In an attempt to resolve other existing problems, the most appropriate approach should be selected on a case-by-case basis, ensuring that the adopted procedure does not conflict with operating guidelines and scientific literature on the matter. Problematic cases include the management of trials with a predominant active metabolite, the absence of a reliable analytical bioassay, the availability of various strengths of the same drug on the market, a wide acceptability titre range, the management of studies on topical drugs that are devoid of systemic activity, the management of drugs that cannot be given for ethical reasons to healthy subjects or that may cause adverse events, especially when a steady state design is required. The parallel group study design appears to be more appropriate than the cross-over or the individual bioequivalence design in assessing drugs with a long half-life. Some pharmacokinetic and statistical analysis-related issues are also discussed such as the sequence/period interaction sometimes encountered in these trials, which, in the absence of the carry-over effect, does not bias the bioequivalence results and the need to process data with non-compartmental pharmacokinetic analysis. Copyright 1999 Academic Press.