|
Activity Number:
|
484
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308040 |
|
Title:
|
Evaluating and Using Surrogate Endpoints in Clinical Trials: A Bayesian Approach
|
|
Author(s):
|
Mary K. Cowles*+ and Qian Shi
|
|
Companies:
|
The University of Iowa and The University of Iowa
|
|
Address:
|
241 Schaeffer Hall, Iowa City, IA, 52242-1409,
|
|
Keywords:
|
accelerated failure time model ; censored data ; proportional hazards model ; Wishart distribution
|
|
Abstract:
|
Surrogate endpoints in clinical trials are biological markers or events that may be observed earlier than the clinical endpoints (such as death) that are actually of primary interest. We propose Bayesian models for use in settings in which the true endpoint of interest is time to a clinical event and there are one or more continuous-valued markers that may be useful surrogates, either alone or in combination. The time-to-event component of our models may be either parametric or semi-parametric. Our parametric models can produce posterior predictive distributions for the event times of individuals with censored data, given values of markers and other covariates.
|
