The Changing Face of Clinical Trials: Challenges in the Design and Interpretation of Noninferiority Trials

New England Journal of Medicine
October 5, 2017  Vol. 377 No. 14

Review Articles
The Changing Face of Clinical Trials: Challenges in the Design and Interpretation of Noninferiority Trials
Mauri and R.B. D’Agostino, Sr.,
[Initial text]
Noninferiority clinical trials have become a major tool for the evaluation of drugs, devices, biologics, and other medical treatments. Treatment with placebo or with a no-treatment control in a study is not ethical when an effective treatment has already been established. Effective medical treatments exist for many medical conditions and are the relevant bar to be surpassed by a new treatment. Although some new treatments offer greater efficacy, others may promise greater safety or convenience, or less expense, while providing similar efficacy. The concept of a good substitute was the original rationale for the design of noninferiority trials (i.e., to evaluate a new treatment for efficacy similar to that of an established treatment). Recently, noninferiority trial methods have also been applied in evaluating whether an effective treatment is safe enough. The number of randomized trials assessing noninferiority increased by a factor of 6 in a decade — in 2005, just under 100 trials were listed in MEDLINE under the general rubric of “noninferiority,” whereas in 2015, there were almost 600 such trials. These trials span multiple medical and surgical disciplines and diverse treatment strategies.
In this article, we provide a framework for considering the features, including pitfalls, of noninferiority studies. We use cardiovascular treatment trials as examples, although noninferiority trials can be conducted in many fields. These trials include studies designed for regulatory approval of new therapies and trials designed to compare established treatments. In addition, we consider the application of noninferiority concepts and design to emerging areas of clinical investigation. The term “placebo” is used to denote either a true placebo or a no-treatment control in situations in which a true placebo is not available…