|
Activity Number:
|
599
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304018 |
|
Title:
|
An Assessment of Several Trial-Level Surrogacy Measures in the Meta-Analytical Framework of Evaluating Surrogate Endpoints in Clinical Trials
|
|
Author(s):
|
Qian Shi*+ and Daniel Sargent
|
|
Companies:
|
Mayo Clinic and Mayo Clinic
|
|
Address:
|
200 First Street SW, Rochester, MN, 55905,
|
|
Keywords:
|
trial-level surrogacy ; meta-analysis ; clinical trial ; survival analysis
|
|
Abstract:
|
Various meta-analytical approaches have been applied to evaluate putative surrogate endpoints (S) of primary endpoints (T) in multiple diseases. Naive and systematic trial-level surrogacy (TLS) measures were assessed and compared using simulated multi-trial data sets with time-to-event endpoints. Naive TLS assess the association between treatment effects (TE) on S and T, including correlation coefficients and R-square from weighted linear regression. Systematic TLS included trial-level R-square proposed by Burzykowski (2001) which incorporates the measurement errors of TE estimates and Surrogate Threshold Effect (Burzykowski 2006). Factors examined included number of trials, subject-level association between S and T and range of magnitudes of TE. Naive TLS tends to be over-optimistic; and the precision of estimating systematic TLS depends on number of trials available.
|
