Adaptive designs for binary treatment responses in phase III clinical trials: controversies and progress.

In phase III clinical trials the problem is to compare two treatments or therapies for their effects. The randomized 50:50 allocation to the two competing treatments is popular in most of the cases. In contrast there is a class of adaptive designs available in literature for this purpose. The object is to treat a larger number of patients by the eventual better treatment. A few such adaptive designs were conducted out in practical applications and consequently a lot of controversy arose. In spite of that there is steady theoretical progress in this area over the past three decades. The present paper reviews some of the real-life adaptive clinical trials and discusses some logistics of such trials.