Web-Based Lectures

Title: Challenges in the Design and Analysis of Three-arm Non-Inferiority Trials
Presenter: Samiran Ghosh, Wayne State University (WSU) School of Medicine
Date and Time: Wednesday, July 22, 2020, 12:00 p.m. – 2:00 p.m. Eastern time
Sponsor: Mental Health Statistics Section

Registration Deadline: Monday, July 20, at 12:00 p.m. Eastern time

Randomized controlled trials (RCT's) are an indispensable source of information about efficacy of treatments in almost any disease area. However, design and analysis of effectiveness trial is much more complex than the efficacy trial. The effect of including an active comparator arm/s in a RCT is immense. This gives rise to superiority and non-inferiority trials. The non-inferiority (NI) RCT design plays a fundamental role in comparative effectiveness research (CER), which will be the primary focus of this talk. Traditional NI trials do not include any placebo arm; however, this comes at a cost of stringent assumptions, which must hold for making valid inference via such trial. At the recommendation of EMA a three-arm (including placebo) is often considered provided there is no severe ethical concern. These trials nevertheless require careful attention while constructing NI margin and the number of hypotheses that can be tested. Selection of appropriate outcome measure and proper functional of those outcomes also plays a vital role. In the past decade many statistical methods have been developed, though largely in the Frequentist setup. Albeit, the nature of outcome, primary end-point selection and temporal data collection points etc. dictates the design and sample size related issues. This talk will give an extensive guidance bringing examples from the state-of-the-art developments in this direction.

Additionally, a point to note that the availability of historical placebo-controlled trial is useful and if integrated in the current NI trial design, can provide better precision for CER. This may reduce sample size burden and improves statistical power significantly in current trial. Since NI trial involves active control which are often marketed drugs, the availability of the historical data is almost guaranteed. Bayesian paradigm provides a natural path to integrate historical as well as current trial data via sequential learning in the NI setup. In this talk, we will also discuss some of the recent Bayesian developments in three-arm NI trial. No prior knowledge of NI trial is necessary to understand most of the talk. The talk has significant contributions from my post-doctoral fellows, colleagues and FDA scientists at various point of time.

Dr. Ghosh is an Associate Professor with tenure at the Wayne State University (WSU) School of Medicine. He is also a director of Biostatistics at BERD core and co-director of CURES (NIEHS-P30) Biostatistics core at WSU. He obtained a PhD in Statistics from the University of Connecticut in 2006. Prior to joining WSU, he was an assistant professor biostatistics (in psychiatry) at the Weill Cornell Medical College. His primary area of research has two major directions namely developing novel statistical methods for, (i) Adaptive RCT design and (ii) High Dimensional modelling for Bioinformatics domain. He is interested in developing and supporting statistical design/analysis that deals with “within subject adaptation” (e.g. SMART and MOST type design) as well as more traditional “between subject adaptation” (e.g. Group Sequential and Bayesian). His research is funded by PCORI, NIH and other federal agencies at various point of time.

Member of the Mental Health Statistics Section: $0
ASA Member: $90
Nonmember: $110

Each registration is allowed one connection to the webinar. Multiple persons are encouraged to view each registered connection (for example, by projecting the webinar in a conference room).

Registration is closed.

Access Information
Registered persons will be sent an email the afternoon of Monday, July 20, with the access information to join the webinar and, if possible, the link to download and print a copy of the presentation slides.