Current treatments for human immunodeficiency virus (HIV) include both reverse transcriptase and protease inhibitors. Results from in vitro and clinical studies suggest that combination therapy can be more effective than single drugs in reducing viral burden. To evaluate compounds for combination therapy, stavudine (d4T), didanosine (ddI), or BMS-186,318, an HIV protease inhibitor, were combined with other clinically relevant compounds and tested in a T-cell line (CEM-SS) that was infected with HIV-RF or in peripheral blood mononuclear cells infected with a clinical HIV isolate. The combined drug effects were analyzed by the methods described by Chou and Talalay (Adv. Enzyme Regul. 22:27-55, 1984) as well as by Prichard et al. (Antimicrob. Agents Chemother. 37:540-545, 1993). The results showed that combining two nucleoside analogs (d4T-ddI, d4T-zidovudine [AZT], and d4T-zalcitabine [ddC]), two HIV protease inhibitors (BMS-186,318-saquinavir, BMS-186,318-SC-52151, and BMS-186,318-MK-639) or a reverse transcriptase and a protease inhibitor (BMS-186,318-d4T, BMS-186,318-ddI, BMS-186,318-AZT, d4T-saquinavir, d4T-MK-639, and ddI-MK-639) yielded additive to synergistic antiviral effects. In general, analysis of data by either method gave consistent results. In addition, combined antiviral treatments involving nucleoside analogs gave slightly different outcomes in the two cell types, presumably because of a difference in phosphorylation patterns. Importantly, no strong antagonism was observed with the drug combinations studied. These data should provide useful information for the design of clinical trials of combined chemotherapy.