In an ideal world, one clinical marker would explain all the variance in a disease system. In reality, however, this is rarely the case and HIV disease is no exception. This review considers several specific markers that have demonstrated some use in clinical trials and/or epidemiologic studies of antiretroviral therapy. CD4 lymphocyte count, HIV viral load and perhaps immune activation markers can be used to measure the activity of antiretroviral therapy. Some recent studies are presented and the results discussed. Sustained improvements in several markers (particularly HIV viral load and CD4 cell count) in combination appear to be the most predictive of clinical benefit. No current viral or immunologic markers adequately reflect toxicities attributable to antiretroviral therapy. Lamivudine/zidovudine combination therapy leads to sustained changes in several markers, and appears to be well tolerated. This may translate into significant clinical benefit but the duration of such benefit remains unknown.