Tumor response and survival end points in clinical trials: a clinician's perspective

Survival has become the gold standard for determining the relative effectiveness of chemotherapy regimens in phase III clinical trials and is measured generally as the median survival. Response could be a surrogate for survival in evaluating clinical trials for chemotherapy, but it has become a controversial measurement parameter because of the quantitative variability in measurement and the fact that differences in response rates are not commonly translated into differences in survival. However, if response is indeed a determinant of survival, the median survival (the point at which 50% of the patients are alive) will not be impacted because response rates (RR) for most advanced cancers are less than 50%. Nonetheless, a survival benefit has been identified for tumors in which response rates are 50% or less in 4 phase III randomized trials; 2 in breast cancer (a high response-rate tumor) also show significant differences in survival. These data would imply that factors other than tumor response could be operative in influencing survival. It is a categorical truth that "responders" live longer than "nonresponders," but the response and survival relationship is complex and both parameters are important in clinical trials.