Re: A model to select chemotherapy regimens for phase III trials for extensive-stage small-cell lung cancer

Abstract

Chen et al.(1) recently argued that the results of phase II trials of new therapeutic regimens for extensive-stage small-cell lung cancer (SCLC) cold be used to predict the statistical power of subsequent phase III trials of the same regimens. Furthermore, they concluded that the median survival of patients on phase II trials was more useful than their response rate to predict the outcome of subsequent phase III trials. We believe, however, that their data call for more prudent conclusions. Phase II trials are typically performed with carefully selected patient populations and have often yielded impressive results that could not be reproduced in phase III trials(2). The results of phase II trials are imprecise because of a limited sample size and are unreliable because of the patient selection and the lack of a control group. The data presented by Chen et al. suggest that the results of phase II trials in SCLC poorly predict the outcomes of patients on phase III trials. The authors find no correlation between the median survival of patients on phase II trials and the median survival of patients on the phase III trials.