Statistical simulation study of new proposed uniformity requirement for bioequivalency studies.

The results of a statistical simulation study of the FDA-proposed 75/75 requirement for some bioequivalency studies are presented. The study used test drug products with the same true average bioavailability as the corresponding reference drug products but with various degrees of uniformity as measured by the coefficient of variation. The performance of the 75/75 requirement indicated that the probability of the test product passing was greater if both products had identical small coefficients of variation than if both had identical large coefficients of variation. Moreover, a test product compared to a reference product with equal variability had less probability of acceptance than an equally variable test product for another drug for which the reference product had less variability. Both results indicate that the 75/75 requirement should be withdrawn. An alternate uniformly most powerful, unbiased test, the Pitman-Morgan F test, is presented.