Design Issues in Noninferiority/Equivalence Trials *

Abstract

When designing a noninferiority/equivalence trial, the sponsor intends to show efficacy by demonstrating that a new treatment is as good as or not worse than a known effective treatment by a small predefined margin. To confirm noninferiority/equivalence of the new treatment to an active control, “sensitivity-to-drug-effects” and “assay sensitivity,” as defined in the International Conference on Harmonization (ICH) E10 Guideline (1, 2) must be supported. Otherwise, a finding of mere nonsignificant difference between treatments in the traditional setting of significance testing leaves the question unanswered: Would the trial have concluded noninferiority while the new treatment was, in fact, inferior?

This paper first reviews the choice of control and the crucial issues of sensitivity-to-drug-effects and assay sensitivity. Then, it discusses the choice of the noninferiority/equivalence margin and the forms of the null and alternative hypotheses and confidence intervals. Finally, it addresses the inherent difficulties and some useful design alternatives to the noninferiority/equivalence trials.