Methods and Applications of Statistics in Clinical Trials, Volume 1: Concepts, Principles, Trials, and Designs

Chapter 49

Noninferiority Trial

H. M. James Hung, Sue-J. Wang, and Robert O’Neill

49.1 Introduction

Traditionally, the effect of an experimental treatment in treating a disease is mostly evaluated in a placebo-controlled clinical trial with the objective of demonstrating that the experimental treatment is more effective than the placebo with or without the standard-of-care medical treatment on the background. This clinical objective drives the formulation of a statistical hypothesis for testing. The trial is designed to provide statistical evidence to reject the null hypothesis that there is no treatment difference in the therapeutic effect and consequently to accept the intended alternative hypothesis that the experimental treatment is more effective than the placebo. This is often referred to as superiority testing. When a placebo can no longer be used ethically in the trial (e.g., with a life-saving drug on the market), a treatment regimen that has been proven to be effective and safe can be selected as a control, often called active or positive control, for an experimental treatment to be compared against [1]. In an active-controlled trial, demonstration of superiority over the active control certainly establishes the effectiveness of the experimental treatment. Alternatively, proving the efficacy of the experimental treatment can be based on the so-called noninferiority testing, by which the experimental treatment will be shown to be not much inferior to the active control. ...