Bioequivalence and Bioavailability Forum

Hi Mittyri,

❝ I've found very interesting paper which explains the vision of experts.

IMHO, rubbish.

❝ Do you remember your investigations regarding dosing groups? Looks like another factor starts the game in Russia!

Please, not opening another construction site (the paper is still on my TODO-list).

Let’s see what my my friend Alfredo* thinks about it (four authors of the Spanish agency and one of Health Canada):

The misunderstanding of the sex effect and the sex-by-formulation interaction
Other authors have suggested the need for recruiting male and females in bioequivalence studies simply because they exhibit different exposures when the reference product is administered. Although obvious, it seems necessary to highlight the fact that the existence of sex-related differences in pharmacokinetics of many drugs is not indicative of a sex-by-formulation interaction because if a drug exhibits different pharmacokinetic parameters between males and females, the same is expected for the test and for the reference product. A sex-by-formulation interaction in bioequivalence means that, hypothetically, a test product might be equivalent to the reference product in one sex group (e.g., males) and bioinequivalent in the other group (e.g., females); or bioinequivalent in both groups, but equivalent in the combined analysis (e.g., if the test product were 30% more bioavailable in males and 30% less bioavailable in females, a balanced study with respect to the sex of the recruited participants would be able to show equivalence).
      It is also important to clarify that even if the variability in females were larger than in males, which may occur in certain cases, this is not a reason to include women as suggested by some authors, because we are not interested in making the demonstration of bioequivalence more difficult, but in detecting differences if they exist. Where variability is larger, more subjects need to be recruited, which can be considered unethical since more subjects are exposed unnecessarily to the risks of the study. Importantly, once a large number of subjects is recruited, the estimation would be the same since the larger variability does not bias the study results (i.e., imprecision is not inaccuracy).

(my emphases)

---

• González-Rojano E, Abad-Santos F, Ochoa D, Román M, Marcotegui J, Álvarez C, Gordon J, García-Arieta A. Evaluation of sex-by-formulation interaction in bioequivalence studies of efavirenz tablets. Br J Clin Pharmacol. 2018;84:1729–37. doi:10.1111/bcp.13601.