Some statistical considerations on the FDA draft guidance for individual bioequivalence

Abstract

In December of 1997, the FDA proposed a draft guidance for future in vivo bioequivalence studies. The guidance suggested specific criteria for new drug sponsors to show individual bioequivalence (IBE). The criteria use a mixed-scaling aggregate strategy. It has been generally accepted that, under some particular situations, the proposed criteria would result in a relaxation of the current bioequivalence standard set by the average bioequivalence (ABE) criterion. Here we study the magnitude of this relaxation under three scenarios: when the conditions for an ABE investigation are met; when the drugs are highly variable, and when the experiments are poorly conducted. The magnitude of relaxation we report here may be surprisingly large to many. For example, when a drug is highly variable (with the intrasubject coefficient of variation reaching 40 per cent), the allowable limit for the ratio of the formulation means could reach 55-180 per cent in an IBE investigation. In comparison, the usual allowable limit in an ABE investigation is 80-125 per cent. Our investigation raises doubts on whether the implied standard of the new proposed IBE criteria would adequately ensure switchability in highly variable drugs.