Plenary Talks

Dose Finding Under Model Uncertainty: the MCP-Mod Approach and its Extensions

Speaker: Dr. José Pinheiro, Janssen Research & Development

José Pinheiro José Pinheiro has a Ph.D. in Statistics from the University of Wisconsin–Madison, having worked at Bell Labs and Novartis Pharmaceuticals, prior to his current position as Global Head of Statistical Modeling & Methodology in the Statistics and Decision Sciences department at Janssen Research & Development. He has been involved in methodological development in various areas of statistics and drug development, including dose-finding, adaptive designs, and mixed-effects models. He is the current president of the International Biometric Society (IBS), a Fellow of the American Statistical Association, past-editor of Statistics in Biopharmaceutical Research, and past-president of the East North American Region (ENAR) of the IBS.

Abstract

Poor dose-regimen selection resulting from insufficient knowledge of the dose-response relationship remains one of the key challenges in clinical drug development, believed to be associated with the high attrition rate observed in confirmatory trials. Different methods have been proposed to improve on the conventional, often inefficient paradigm of pairwise testing of active doses versus placebo, among them MCP-Mod. This approach has the appealing feature of combining good aspects of hypothesis testing and modeling, implementing dose-response estimation and dose selection under model uncertainty. This talk will provide an overview of the MCP-Mod approach and some of its more recent extensions that increase the flexibility of the methodology and its range of applications. Examples from real and simulated clinical trials will be used to illustrate the use of MCP-Mod in practice, using its software implementation in the DoseFinding R package.


Lost in Translation

Speaker: Dr. Lee-Jen Wei, Harvard University

L.J. Wei L.J. Wei is a professor of Biostatistics at Harvard University. Before joining Harvard, he was a professor at University of Wisconsin, University of Michigan, and George Washington University. His main research interest is in the clinical trial methodology, especially in design, monitoring and analysis of studies. He has developed numerous novel statistical methods which are utilized in practice. He received the prestigious Wald Medal in 2009 from the American Statistical Association for his contribution to clinical trial methodology. He is a fellow of American Statistical Association and Institute of Mathematical Statistics. In 2014, to honor his mentorship, Harvard School of Public Health established a Wei-family scholarship to support students studying biostatistics. His recent research area is concentrated on translational statistics, the personalize medicine under the risk-benefit paradigm via biomarkers and revitalizing clinical trial methodology. He has more than 250 publications and served on numerous editorial and scientific advisory boards including data monitoring for governments and industry. L. J. Wei has extensive working experience in regulatory science for developing and evaluating new drugs/devices.

Abstract

Over the years, the process of designing, monitoring, and analyzing clinical studies for evaluating new treatments has gradually fallen into a fixed pattern. Clinical trialists have sometimes been slow to utilize new methodologies–perhaps to avoid potential delays in the review process for drug approval or manuscript submission. The underlying attitude toward innovation in drug development is in sharp contrast to that in other technologically-driven fields. Scientific investigation is an evolving process. What we have learned from previous studies about methodological shortcomings should help us better plan and analyze future trials. Unfortunately, use of inefficient or inappropriate procedures persists even when better alternatives are available. In this talk, we will explore various methodological issues and potential solutions to them. A goal of the clinical study is to obtain robust, clinically interpretable treatment effect estimate with respect to risk-benefit perspectives at the patient’s level via efficient and reliable quantitative procedures. We will discuss how to achieve this goal via various real trial examples.


© Department of Statistics, University of Connecticut