A Nonparametric Test for the Validation of Surrogate Endpoints

• Activity Number: 584
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Xiaopeng Miao*+ and Yong-Cheng Wang and Ashis Gangopadhyay
• Companies: Boston University School of Public Health and Biogen Idec and Boston University
• Keywords: Nonparametric test ; Kullback-Leibler distance ; Prentice's criteria ; Surrogate endpoints

Abstract

We present a novel nonparametric test to validate the surrogate endpoints based on multivariate density estimation and permutation test. This test is the first in literature to directly verify the Prentice statistical definition for surrogacy. The test does not impose distributional assumptions on the endpoints and it is robust to model misspecification. Our simulation study show that the proposed nonparametric test outperforms the practical test of the Prentice criterion in terms of both robustness of size and test power. The method is applied to the validation of MRI lesions as the surrogate endpoint for clinical relapses in multiple sclerosis trials.