Statistical methods to evaluate the effectiveness of cancer treatments in clinical trials

Overview

Abstract

Assessing the effectiveness of cancer treatments in clinical trials raises multiple methodological challenges that need to be properly addressed in order to produce a reliable estimate of treatment effects. The purpose of this seminar is to introduce the methodology to decide whether a new treatment should be preferred over the standard one in a cancer clinical trial. We first introduce the standard approach that compares the risk of death according to treatment received. Then, we discuss about the repeated measurements of tumors size as an alternative judgment criteria. Finally we conclude with recently introduced statistical methods that account for both the risk of death and the tumors measurements. These new methods can improve the statistical power for decision-making and reduce the delay in the availability of the new treatment when it is better than the standard treatment.

Brief Biography

Denis Rustand is a Post-Doctoral fellow in Statistics under the supervision of Professor Håvard Rue. He got his Ph.D. degree in Public Health Biostatistics at the University of Bordeaux, France and before that, he obtained his Master’s degree of Science in Statistics at the University of Southern Brittany, France. His research areas include Bayesian computational statistics, survival analysis and applications of statistics to medical research.