The 11th Summer Institute in Statistics for Clinical and Epidemiological Research

Module 3: Design, Conduct, and Analysis of Randomized Clinical Trials with Time-to-Event Primary Endpoints

Mon, July 8 to Fri, July 19
Instructor(s):
Scott Emerson
Schedule
  • Monday, July 8 – Pre-recorded module content available for registrants (approximately 14 hours of content)
  • Live sessions: Discussion of course content and case studies:
    • Tuesday, July 16, 10-11:30 a.m. PDT (1-2:30 p.m. EDT)
    • Wednesday, July 17, 10-11:30 a.m. PDT (1-2:30 p.m. EDT)
    • Friday, July 19, 10-11:30 a.m. PDT (1-2:30 p.m. EDT)
 
Description

There will be three 90-minute live sessions to discuss module material and case studies on July 16, 17, and 19.

The analysis of data measuring time to event is often complicated by incomplete observations: Some subjects have not yet had an event at the time of data analysis. A wide variety of statistical methods have been developed for this setting of “right censored data”, including parametric and semiparametric regression models, as well as a broad array of nonparametric methods. The common problems that arise in the clinical trial settings interact with the statistical behavior of the analysis methods in such a way as to warrant special attention.

In this module, topics will include

  • Review of statistical methods commonly used with time-to-event data, with particular emphasis on proportional hazards regression and restricted mean survival. Topics will include the requirements for noninformative censoring, statistical precision in both unadjusted and covariate-adjusted models, and time varying treatment effects (both crossing hazards and crossing survival curves).
  • General clinical trial design, including randomization schemes, sample size formulas,  group sequential sampling plans, and inferential methods appropriate for interim monitoring of both superiority and noninferiority studies. Particular consideration will be given to the impact of potentially time varying treatment effects on design decisions regarding accrual and length of follow-up.
  • Adaptive clinical trial design that transcends group sequential schemes, especially as it relates to sample size re-estimation that might adapt the number of subjects accrued or the time of follow-up, including in the setting of highly effective therapies. The potential impact of auxiliary data on adaptive trial design decisions will also be discussed.
  • Choice of estimands in the presence of intercurrent events that might affect the measurement of the most important clinical endpoints. Special emphasis will be placed on discussing regulatory guidances on missing data in clinical trials, including the International Conference on Harmonization (ICH) E9 (R1) addendum.