Abacavir (1592U89), a nucleoside reverse transcriptase inhibitor with in vitro activity against human immunodeficiency virus type-1 (HIV-1), has been evaluated for efficacy and safety in combination regimens with other nucleoside analogs, including zidovudine (ZDV) and lamivudine (3TC). To evaluate the potential pharmacokinetic interactions between these agents, 15 HIV-1-infected adults with a median CD4⁺ cell count of 347 cells/mm³ (range, 238 to 570 cells/mm³) were enrolled in a randomized, seven-period crossover study. The pharmacokinetics and safety of single doses of abacavir (600 mg), ZDV (300 mg), and 3TC (150 mg) were evaluated when each drug was given alone or when any two or three drugs were given concurrently. The concentrations of all drugs in plasma and the concentrations of ZDV and its 5'-glucuronide metabolite, GZDV, in urine were measured for up to 24 h postdosing, and pharmacokinetic parameter values were calculated by noncompartmental methods. The maximum drug concentration (C_max), the area under the concentration-time curve from time zero to infinity (AUC_0-), time to C_max (T_max), and apparent elimination half-life (t_1/2) of abacavir in plasma were unaffected by coadministration with ZDV and/or 3TC. Coadministration of abacavir with ZDV (with or without 3TC) decreased the mean C_max of ZDV by approximately 20% (from 1.5 to 1.2 µg/ml), delayed the median T_max for ZDV by 0.5 h, increased the mean AUC_0- for GZDV by up to 40% (from 11.8 to 16.5 µg · h/ml), and delayed the median T_max for GZDV by approximately 0.5 h. Coadministration of abacavir with 3TC (with or without ZDV) decreased the mean AUC_0- for 3TC by approximately 15% (from 5.1 to 4.3 µg · h/ml), decreased the mean C_max by approximately 35% (from 1.4 to 0.9 µg/ml), and delayed the median T_max by approximately 1 h. While these changes were statistically significant, they are similar to the effect of food intake (for ZDV) or affect an inactive metabolite (for GZDV) or are relatively minor (for 3TC) and are therefore not considered to be clinically significant. No significant differences were found in the urinary recoveries of ZDV or GZDV when ZDV was coadministered with abacavir. There was no pharmacokinetic interaction between ZDV and 3TC. Mild to moderate headache, nausea, lymphadenopathy, hematuria, musculoskeletal chest pain, neck stiffness, and fever were the most common adverse events reported by those who received abacavir. Coadministration of ZDV or 3TC with abacavir did not alter this adverse event profile. The three-drug regimen was primarily associated with gastrointestinal events. In conclusion, no clinically significant pharmacokinetic interactions occurred between abacavir, ZDV, and 3TC in HIV-1-infected adults. Coadministration of abacavir with ZDV or 3TC produced mild changes in the absorption and possibly the urinary excretion characteristics of ZDV-GZDV and 3TC that were not considered to be clinically significant. Coadministration of abacavir with ZDV and/or 3TC was generally well tolerated and did not produce unexpected adverse events.