Statistical Considerations for Trials in Adjuvant Treatment of Colorectal Cancer

Abstract

Simple Summary Improvements in the treatment of patients with cancer largely depend on clinical trials comparing a new intervention with a previous standard of care. This is not different when it comes to improving adjuvant (postoperative) and neoadjuvant (preoperative) treatment for colorectal cancer. In clinical trials, there is usually a close collaboration between physicians and statisticians, and better and more reliable results can be obtained when such trials are designed, conducted and analyzed with attention to some important methodological issues. In this article, we provide a general review of useful methodological and statistical ideas that may help investigators and clinicians in their attempt to improve treatment results in colorectal cancer. The design of the best possible clinical trials of adjuvant interventions in colorectal cancer will entail the use of both time-tested and novel methods that allow efficient, reliable and patient-relevant therapeutic development. The ultimate goal of this endeavor is to safely and expeditiously bring to clinical practice novel interventions that impact patient lives. In this paper, we discuss statistical aspects and provide suggestions to optimize trial design, data collection, study implementation, and the use of predictive biomarkers and endpoints in phase 3 trials of systemic adjuvant therapy. We also discuss the issues of collaboration and patient centricity, expecting that several novel agents with activity in the (neo)adjuvant therapy of colon and rectal cancers will become available in the near future.