Phase III trials of targeted anticancer therapies: redesigning the concept.

Randomized phase III trials provide the gold-standard evidence for the approval of new drugs: an experimental treatment is compared with the current standard of care to identify clinically relevant differences in a predefined endpoint. However, there are several problems relating to the current role of phase III trials in drug development including the limited clinical benefit observed for some approved agents, the necessity for large trials to detect these differences, the inability of such trials to identify rare but important toxicities, and high cost. The design of phase III trials evaluating drug combinations, and those including biomarkers, presents additional challenges. Here, we review these problems and suggest that phase III trials with adaptive designs in selected prescreened populations could reduce these limitations.