ABSTRACT
Most dose finding methods, including the CRM, are suitable for clinical settings that satisfy at least one of the following assumptions. First, there is a strictly increasing dose-toxicity relationship around the target dose. Second, the toxicity endpoint is an adequate surrogate for efficacy, which often takes much longer to observe. While these assumptions are reasonable for traditional cancer chemotherapy, they are debatable in other disease areas or treatment options. When neither assumption holds, the surrogacy perspective of the MTD is not an appropriate operational objective, and as a result the CRM is not a suitable design for dose finding. This chapter introduces alternative paradigms to the surrogacy perspective of the MTD. Section 13.2 presents the notion of maximum safe dose, and Section 13.3 outlines some basic ideas of bivariate dose finding in combined phase I/II trials.
