ABSTRACT
Phase II trials aim to gather information on the clinical activity of a new agent in order to determine whether the agent should proceed to phase III trials for further development. Numerous phase II trial designs are used in oncology (Simon 1989; Rubinstein et al. 2005, 2009; Green et al. 2012). Increasingly phase II clinical trials of new cancer therapies are incorporating biomarkers that are thought to be potentially informative for the likelihood that a patient will beneˆt from the therapy under investigation (i.e., predictive biomarkers) (McShane et al. 2009). The strength of preliminary evidence for potential predictive biomarkers may vary widely from study to study, and this is generally re¡ected in the choice of phase II design. At the time a phase II trial is initiated, there might be reasonable conˆdence in what biomarker is likely to be predictive, but there could be uncertainty about the most appropriate
CONTENTS
5.1 Introduction .................................................................................................. 71 5.2 Biomarker Measurement ............................................................................72 5.3 Nonrandomized Single-Arm Trials with a Single Biomarker
and Single Experimental Agent ................................................................. 73 5.4 Randomized Two-Arm Trials with a Single Biomarker and a
Single Experimental Agent ......................................................................... 76 5.5 Nonrandomized Multiarm Trials with Multiple Biomarkers
and Multiple Experimental Agents ........................................................... 78 5.6 Randomized Multiarm Trials with Multiple Biomarkers
and Multiple Experimental Agents ........................................................... 82 5.7 Randomized Phase II/III Trials with Biomarkers ...................................84 5.8 Summary and Generalizations ..................................................................86 References ............................................................................................................... 87
form of the biomarker (e.g., gene mutation, gene ampliˆcation, gene expression, or protein expression), and the assay might still be research grade. It may be acceptable to use such a biomarker in a phase II setting with the understanding that the phase II trial provides an opportunity to establish activity of the therapy and to reˆne the biomarker or discover new biomarkers that might ultimately be helpful for guiding use of the therapy in clinical practice.
