BACKGROUND: There are few data on the selection of resistance by ritonavir-boosted saquinavir (SQV/r), particularly in antiretroviral (ARV)-naive patients. OBJECTIVE: To assess the incidence of virological failure and evolution of resistance in ARV-naive individuals receiving SQV/r in the induction phase of the Staccato trial. METHODS: ARV-naive subjects (n = 272) received SQV/r 1,600/100 mg once daily with two nucleoside reverse transcriptase inhibitors (NRTIs) for at least 24 weeks. Patients were defined as having virological failure (VF) when there were two consecutive HIV-1 RNA measurements > 500 copies/ml after week 12. Viral genotypes (reverse transcriptase [RT] and protease [PRO]) were determined at baseline in all patients and as close as possible to the time of initial failure in patients experiencing VF. RESULTS: VF was observed in 9/272 patients receiving SQV/r 1,600/100 mg once daily with two NRTIs (3.3%) and occurred 19-48 weeks after treatment initiation. Eight of these patients were evaluable at failure. No major PRO mutations were detected, but 2/8 displayed single new minor PRO substitutions (M36I, L10I) at VF that were known or suspected not to have been present at baseline; both these substitutions exist as natural polymorphisms. A third patient displayed a single new RT mutation (M184I). CONCLUSIONS: SQV/r plus two NRTIs (1,600/100 mg once daily) is an effective initial treatment option for ARV-naive patients, resulting in a low rate of viral rebound (3.3%). Furthermore, no major protease mutations were detected following VF, suggesting that future treatment options are preserved.