|
Activity Number:
|
243
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #309737 |
|
Title:
|
Evaluation of a Binary Response Variable as a Surrogate for Survival: A Gynecologic Oncology Group Study
|
|
Author(s):
|
Virginia Filiaci*+ and Randolph L. Carter
|
|
Companies:
|
GOG Statistical and Data Center and University at Buffalo-SUNY
|
|
Address:
|
Elm and Carlton Streets, Buffalo, NY, 14263,
|
|
Keywords:
|
Surrogate endpoint ; survival ; indirect effect ; explained variance ; clinical trial
|
|
Abstract:
|
Identification of valid surrogates for survival that lead to a decrease in study times or the resources necessary to complete clinical trials is important. Methods were developed to evaluate a binary response variable as a surrogate for survival using data from a single trial. Two criteria are suggested for perfect surrogacy: 1) the total effect of treatment on survival must be comprised solely of its indirect effect through the surrogate; and 2) survival must be perfectly predicted by the surrogate and treatment. Two statistics are proposed in terms of these criteria. The first is the proportion of the total treatment effect on mean censored survival time truncated at T0 that is mediated through the surrogate. A measure defined by Korn and Simon is adopted as the second. These methods will be applied to data from a randomized trial of 263 women with metastatic endometrial cancer.
|
