A question that has received a great deal of attention in evaluating treatments for HIV is that of finding a good surrogate marker for clinical progression. CD4+ lymphocyte count has been proposed as a potential marker for HIV trials because of its apparent correlation with clinical outcome. In order to evaluate the utility of CD4 as a surrogate, however, an understanding of the relationship of clinical outcome and individual evolution of CD4 over time is required. This talk will provide a non-technical introduction to these issues in the context of data from a double-blind placebo controlled trial of the HIV treatment zidovudine (ZDV or AZT) carried out by Burroughs-Wellcome in the early years of development of HIV therapies. The surrogate marker problem, including a formal perspective on what makes a "good" surrogate, will be discussed, and an approach to analysis that involves joint modeling of longitudinal CD4 counts and the primary endpoint, time-to-death, will be described.
Return to Biostatistics Working Group