[Surrogate endpoints in clinical studies exemplified by HIV infection]

Surrogate endpoints are used extensively in clinical studies to evaluate the efficacy of therapy. The effect of a treatment on surrogate endpoints should predict efficacy with regard to clinical endpoints such as morbidity and death. Benefits and risks in the use of surrogate endpoints in clinical trials and daily practice are discussed using the example of HIV infection. CD4-lymphocyte counts and plasma HIV-RNA are powerful markers of disease progression in HIV infection. They are also increasingly used as surrogate endpoints to evaluate the efficacy of antiretroviral treatments. Surrogate endpoints should be biologically plausible and must be validated statistically. However, clinical endpoints are still of paramount importance in assessing the value of treatments for chronic diseases.