BACKGROUND: Resistance to antiretroviral agents remains a leading cause of treatment failure for patients infected with HIV-1. OBJECTIVE: To describe the efficacy and safety of tenofovir disoproxil fumarate (tenofovir DF) compared with placebo in patients with detectable viral replication despite current antiretroviral therapy. DESIGN: Randomized, double-blind, placebo-controlled study through 24 weeks. After 24 weeks, all patients received open-label tenofovir DF for the remainder of the 48-week study. SETTING: 75 North American, European, and Australian HIV clinics. PATIENTS: 552 HIV-1-infected adults who were receiving antiretroviral therapy and had stable HIV-1 RNA levels ranging from 400 to 10,000 copies/mL. MEASUREMENTS: Change in HIV-1 RNA level (time-weighted average from baseline through week 24); proportion of patients with grade 3 or 4 laboratory abnormalities and adverse events; and genotypic HIV-1 resistance testing in a separate substudy at baseline, week 24, and week 48. RESULTS: A statistically significant decrease in HIV-1 RNA level through week 24 (the primary end point) was observed in the tenofovir DF group versus the placebo group (-0.61 log10 copies/mL vs. -0.03 log10 copies/mL, respectively [P < 0.001]; difference, -0.58 log10 copies/mL [95% CI, -0.68 to -0.49 log10 copies/mL]). In a virologic substudy, 94% of 253 patients had plasma isolates expressing reverse transcriptase mutations associated with nucleoside resistance mutations at baseline. Through week 24, the incidence of clinical adverse events was similar between patients receiving placebo and those receiving tenofovir DF (14% vs. 13%). No evidence of tenofovir DF-related toxicity was seen through week 48. CONCLUSION: In treatment-experienced patients with suboptimal viral suppression, tenofovir DF significantly reduced HIV-1 RNA level and had a safety profile similar to that of placebo.