Design of Phase II Cancer Trials for Evaluation of Cytostatic/Cytotoxic Agents

Abstract

For experimental anticancer agents that may have both cytostatic and cytotoxic effects, assessment of response rates alone may not capture the full impact of the treatment. Oncologists are therefore interested in assessing both response and stable disease rates in early phase clinical trials of such therapies. We describe the design of a single-arm, Phase II clinical trial for the simultaneous evaluation of objective response and stable disease (lack of early tumor progression) rates using standard RECIST criteria. Demonstration of a sufficiently high rate for either of these endpoints will lead to rejection of the null hypothesis and a conclusion that the treatment warrants further study. A design is chosen that satisfies the desired type I error constraint and has sufficient statistical power at several selected points within the alternative hypothesis space using a restricted search algorithm. An early stopping rule for lack of efficacy is incorporated. The method is illustrated by the design of a Phase II clinical trial in head and neck cancer.

Key Words:

• Multinomial response
• Phase II clinical trial design
• Stable disease

Notes

CR = complete response, PR = partial response, SD = stable disease, NP = nonprogression, PD = progressive disease, NP = CR/PR/SD.