(1997; 248 pages) [French]
Appendix 3. Technical aspects of bioequivalence statistics
The pharmacokinetic characteristics to be tested, the test procedure and the norms to be maintained should be specified beforehand in the protocol. A post hoc change in the methods specified for the statistical evaluation is acceptable only if adherence to the protocol would preclude a meaningful evaluation and if such a change in procedure has been fully justified.
Concentration-dependent data such as AUC and Cmax should be log transformed before statistical analysis in order to satisfy the fundamental assumption underlying analysis of variance that effects in the model act in an additive rather than a multiplicative manner.
Acceptance ranges for main characteristics
The 90% confidence interval for this measure of relative bioavailability should lie within a bioequivalence range of 80-125%. If the therapeutic range is particularly narrow, the acceptance range may need to be reduced. A larger acceptance range may be acceptable if clinically appropriate.
This measure of relative bioavailability is inherently more variable than, for example, the AUC-ratio, and a wider acceptance range may be appropriate. The range used should be justified, taking into account safety and efficacy considerations.
Statistical evaluation of tmax, makes sense only if there is a clinically relevant claim for rapid release or action, or signs of a relation to adverse effects. The non-parametric 90% confidence interval for this measure of relative bioavailability should lie within a clinically relevant range.