Statistical issues in HIV non-inferiority trials

One of the most common study designs currently used to evaluate new treatments for patients infected with the human immunodeficiency virus (HIV) is the non-inferiority (NI) clinical trial. While the goal in a conventional randomized superiority trial is to demonstrate that the new therapy is more efficacious than the control, the objective in an NI trial is to establish that the new treatment is not worse by more than a prespecified margin than the comparator, which is usually a standard therapy. This goal is of interest when the new treatment offers benefits such as improved safety, increased tolerability, lower cost, or greater convenience that make it a desirable alternative even if it is not necessarily more efficacious than the standard. An NI trial is also conducted to evaluate an experimental therapy when the use of a placebo is unethical due to the availability of existing effective regimens. In this case, the efficacy of the new drug is demonstrated by showing that it is non-inferior to an approved treatment.