SY16-5 Problems in clinical trials to evaluate de-escalation

Abstract

Recently, de-escalation of treatment has been evaluated in some clinical trials for breast cancer. If intensive treatments do not improve survival more than less intensive treatments, less intensive treatments may be better in terms of quality of life, toxicity, cost, and some burdens. To design a clinical trial of de-escalation of treatment, many factors should be considered such as definition of treatment (regimen), population selection (evaluation of biomarker), selection of endpoints (use of multiple endpoints), randomization, and analysis methods. For example, when de-escalation is evaluated in the non-inferiority setting, it may be difficult to conduct a randomized controlled trial because of a large sample size, and how to use multiple endpoints, such as overall survival, toxicity, quality of life, and cost, may also be difficult. In this talk, I discuss the use of historical data and multiple endpoints in clinical trials of de-escalation of treatment.