In the current drug development paradigm, regulatory approval requires agent efficacy and safety that is demonstrated in rigorous randomized controlled clinical trials. Given the length of time and intense resource commitment that are required for phase III cancer prevention clinical trials, it is imperative that only agents that are highly likely to meet these standards be selected for late-phase clinical development. The major challenge for the field of cancer prevention (and cancer treatment) is to identify the best agents to move forward into definitive phase III testing. Preliminary indications of activity against carcinogenesis are obtained from in vitro and animal in vivo experiments (including mechanistic analyses), epidemiologic case-control and cohort studies, and early-phase clinical trials or secondary end points from clinical trials done for other indications (1). Early-phase clinical trials have the potential to be most informative because they actually test the specific agent at the requisite dose in human beings and assess a surrogate end point that should predict the phase III end point of cancer development. Identification of suitable end points for early-phase clinical trials, however, has been problematic.