The Relationship of CD4 Counts over Time to Survival in Patients with AIDS: Is CD4 a Good Surrogate Marker?
Methods are developed to analyze the relationship of survival to time dependent covariates that are measured longitudinally with possible measurement error using a two stage approach. In the first stage, the longitudinal time dependent data are modelled using repeated measures random components models. In the second stage methods are developed for estimating the parameters in a Cox proportional hazards model when the time dependent data is of this form. These methods are applied to CD4 data from a randomized clinical trial of AIDS patients where half the patients received AZT and the other half received placebo. Although a strong corellation between CD4 counts and survival is demonstrated, we also show that CD4 may not serve as a useful surrogate marker for assessing treatments for this population of patients.
KeywordsHazard Function Conditional Expectation Partial Likelihood Time Dependent Covariates Good Surrogate Marker
Unable to display preview. Download preview PDF.
- Cox, D.R. (1972). Regression models and life tables, J. Roy. Statist. Soc Ser. B 34: 187–220.Google Scholar
- Lindstrom, M.J. and Bates, D.M. (1988) Newton-Raphson and EM algorithms for linear mixed-effects models for repeated-measures data, JASA 83: 1014–1022.Google Scholar
- Self, S.G. and Pawitan, Y. (1991). Modeling a marker of disease progression and onset of disease, Submitted for publication.Google Scholar
- Strawderman, R.L. and Tsiatis, A.A. (1991). Hazard estimation in the surrogate marker problem, Submitted to Biometrics.Google Scholar