Modern Approaches to Clinical Trials Using SAS: Classical, Adaptive, and Bayesian Methods

7.1 Introduction

A well-known problem of failed phase III programs is poor dose selection resulting from inappropriate knowledge of the dose-response relationship (efficacy and safety) at the end of the learning phase of drug development, i.e., phase II. Selection of a dose (or doses) to carry into confirmatory phase III trials is among the most difficult decisions in drug development. Although the exact numbers are unknown, it is believed that the high attrition rate plaguing the pharmaceutical industry in phase III studies are due, at least in part, to inadequate dose selection for confirming safety and efficacy in the intended patient population – doses that are too low to achieve adequate benefit, as well as doses that are too high and lead ...