Statistical Validation of Surrogate Endpoints: Problems and Proposals
- PDF / 537,253 Bytes
- 8 Pages / 504 x 719.759 pts Page_size
- 86 Downloads / 203 Views
Drug lnformution Journal. Vol. 34. pp. 447-454, Zoo0 F’rinted in the
Copyright Q ZOO0 Drug Information Association Inc.
USA.All rights reserved.
STATISTICAL VALIDATION OF SURROGATE ENDPOINTS: PROBLEMS AND PROPOSALS MARCBUYSE International Institute for Drug Development, Brussels, Belgium, and Limburgs Universitak Centrum, Biostatistics, Diepenbeek, Belgium
GEERTMOLENBERGHS, TOMASZ BURZYKOWSKI, DIDIERRENARD, AND HELENA GEYS Limburgs Universitair Centrum, Biostatistics, Diepenbeek, Belgium
Prentice proposed a definition of surrogate endpoints and operational criteria for their validation. Freedman supplemented these criteria with the proportion explained, which is supposed to represent the proportion of the treatment effect upon the true endpoint which is mediated by the surrogate endpoint. In this papel; we argue that the proportion explained should be replaced by other quantities, such as the relative effect linking the effects of treatment on both endpoints and an individual-level measure of agreement between both endpoints. The latter quantity carries over naturally when data are available on several randomized experiments, while the former can be extended to be a trial-level measure of agreement between the effects of treatment of both endpoints. This approach suggests a new definition of surrogacy, and allows one to predict the effect of treatment upon the true endpoint, given its observed effect upon the surrogate endpoint. Key Worak: Meta-analysis; Surrogate endpoint; Validation
INTRODUCTION
“final” endpoint (1). Secondary motivations for using surrogate endpoints are that they SURROGATE ENDPOINTS ARE referred may be easier and more convenient to meato as endpoints that can be used in lieu of sure, observed more frequently, less subject other endpoints in the evaluation of experito competing risks, and less affected by other mental treatments or other interventions. The treatments than the final endpoint of interest. primary motivation to use a surrogate endBefore an endpoint can be used as a surrogate point is that it can be measured earlier (somefor another endpoint in the evaluation of extimes much earlier) than the endpoint of inperimental treatments, it must be “validated,” terest, which is referred to as the “true” or a process that has caused a lot of controversies and has not been fully elucidated so far. In this paper, we review some of the issues involved in this process, we outline concepResented in part at the loth DIA Workshop “Statistical tual problems of previously proposed apChallenges in Drug Development,” March 15-16, proaches, and we suggest alternative ap1999, Hilton Head, South Carolina. proaches that may yield the kind of Reprint address: Marc Buyse. International Institute for Drug Development (iD2),430 Avenue Louise B14. information needed before a surrogate is deemed acceptable for future use. Technical B 1050, Brussels, Belgium. 447
Downloaded from dij.sagepub.com at NEW YORK UNIV LIBRARY on April 29, 2015
448
M. Buyse, G. Molenberghs, T. Burzowski, D. Renurd, and H.Geys
Data Loading...
