Design issues in head and neck clinical trials: a statistician's perspective.

The purpose of this article is to present some of the challenges the trial statistician meets when designing a clinical trial of the head and neck cancer. In recent years, the field of head and neck cancer has been facing some exciting evolutions, such as the arrival of newly targeted therapies and findings of disease causality and prognosis. These evolutions are accompanied by challenges in trial methodology that continue even today, and will most likely grow in importance in the future. This article focuses essentially on the design of phase III trials and discusses three major topics: should the trial be designed for a broad or a targeted population? Is there a concern for lack of equipoise and if so, how will it affect the trial results? What are the key elements that need to be taken into consideration when choosing, defining, and measuring the primary endpoint?