Bioequivalence and Bioavailability Forum

Hi Bebac user,

❝ The combination of Amlodipine and valsartan was published as two periods crossover but after checking the variability of Valsartan, it had more than 30%. Therefore, should I in those cases design this study as replicated or un-replicated?

There is no product-specific guidance of the EMA and the FDA recommends 2×2×2 crossovers (fasting and fed). Any replicated design can be assessed for (conventional, unscaled) average bioequivalence. However, if you aim at reference-scaling, I would be wary to base the sample size on the CV_w of a 2×2×2 design because the CV_wR is unknown (see this article).