Abstract
The problem of concern in early clinical trial is to find an appropriate dose of the new drug under development for further investigation. If the anti-cancer drug is, for example, chemotherapy for solid tumor, then both the toxicity probability and efficacy probability are increasing with dose. Hence, the primary goal of trial designs is to estimate the maximum tolerated dose (MTD) under a prespecified targeted toxicity probability (TTP). However, when the single anti-cancer drug is the molecularly target agent (MTA), the efficacy probability may be plateaued after a high dose. Therefore, the optimal biologic dose (OBD) that gives the highest efficacy yet preserves the tolerable toxicity is of major interest. In this talk, phase I trial designs for finding the MTD and phase I/II trial designs for exploring the OBD are discussed based on the credible intervals for toxicity probability and/or efficacy probability. Since the dose-escalation procedure and the estimation of the MTD or OBD are guided by the lower and upper bounds of the credible intervals, the trial designs under study are denoted as CLUB designs.