OBJECTIVES: No large, randomized head-to-head comparison of atazanavir/ritonavir (ATV/r) and darunavir /ritonavir (DRV/r) for first-line treatment of HIV-1 is currently available. This study compares the efficacy of ATV/r and DRV/r at 48 weeks using a matching-adjusted indirect comparison. METHODS: Two similarly designed randomized trials were identified. Individual patient-level data were available for the CASTLE trial comparing ATV/r (n=430) vs. lopinavir/r (LPV/r) (n=438), each in combination with tenofovir/emtricitabine (TDF/FTC); published summary data were used from the ARTEMIS trial comparing DRV/r (n=343) vs. LPV/r (n=346), each in combination with TDF/FTC. To adjust for cross-trial differences, CASTLE patients were re-weighted to match summary baseline characteristics in ARTEMIS. The primary endpoint was virologic response (HIV-1 RNA<50 copies/mL) at 48 weeks assessed using time to loss of virologic response (TLOVR), and was compared between balanced ATV/r and DRV/r trial populations after matching. As a negative control, outcomes in the two LPV/r arms were compared. RESULTS: Data from all patients in the two trials were included. Before matching, baseline characteristics differed significantly between CASTLE and ARTEMIS. CASTLE had a higher proportion of patients with HIV-1 RNA>100,000 copies/mL, a lower proportion with CDC class C, and lower median CD4 cell count. Without matching, a naïve comparison showed ATV/r-treated patients had a significantly lower virologic response rate at week 48 than DRV/r-treated patients (78% vs. 84%, p=0.040). After matching, mean baseline characteristics were exactly balanced between CASTLE and ARTEMIS, and virologic response to LPV/r was comparable (77% vs. 78%, p=0.811). There was no significant difference between ATV/r and DRV/r in virologic response rate (80% vs. 84%, p=0.138). CONCLUSIONS: After adjusting for cross-trial differences in baseline characteristics, the analysis suggests that ATV/r and DRV/r, each in combination with TDF/FTC, are equally efficacious. A randomized head-to-head trial will provide the gold standard of the comparative efficacy between the two treatments.