A BAYESIAN STRATEGY FOR SCREENING CANCER TREATMENTS PRIOR TO PHASE-IICLINICAL-EVALUATION

We address the problem of selecting a treatment for phase II evaluatio n when several candidate treatments emerging from phase I testing are available. A pre-phase II Bayesian selection design which randomizes p atients among treatments is proposed. The patient group in the trial h as prognosis more favourable than that of phase I patients but less fa vourable than the target group of the subsequent phase II trial. The p atient response rate distribution in each treatment arm is continually updated during the trial for comparison with early termination cutoff s, and the best final treatment must satisfy a minimal posterior effic acy criterion. The primary aim is to replace the usual informal treatm ent selection process with a fair comparison formally based on a combi nation of prior opinion and clinical data.