Methods and Applications of Statistics in Clinical Trials, Volume 1: Concepts, Principles, Trials, and Designs

Chapter 6

Application of New Designs in Phase I Trials

Elizabeth Garrett-Mayer

6.1 Introduction

Phase I dose finding studies have become increasingly heterogeneous in regards to the types of agents studied, the types of toxicities encountered, and degree of variability in outcomes across patients. For these reasons and others, when designing a phase I study it is especially important to select a design that can accommodate the specific nature of the investigational agent. However, most current phase I studies use the standard 3 + 3 design, which has been demonstrated to have poor operating characteristics, allows little flexibility, and cannot properly accommodate many of current dose-finding objectives. Novel phase I trials have been introduced over the past 20 years that have superior properties and are much more flexible than the 3 + 3 design, yet they have not been fully embraced in practice. These newer designs are in general not cumbersome and provide a more accurate and precise estimate of the appropriate dose to pursue in phase II and later studies.

6.2 Objectives of a Phase I Trial

It is important to be reminded of the goals of the phase I trial when selecting a design. Phase I studies are generally dose-finding studies, where the goal of the trial is to identify the optimal dose to take forward for further testing (usually for efficacy testing). For ethical reasons, patients cannot be randomized to doses: one cannot treat patients at high doses until lower ones have been ...