OBJECTIVE To evaluate the 24-week impact of saquinavir-enhancing antiretroviral therapy on viral replication in patients previously treated with nucleoside analogues with or without prior saquinavir hard-gel capsules (HGC). METHODS Patients were randomized in three groups to receive the following: Group 1-nelfinavir (750 mg tid), saquinavir soft-gel capsule (SGC) (800 mg tid), and stavudine (40 mg bid); Group II-ritonavir (400 mg bid), saquinavir-SGC (400 mg bid), and stavudine (40 mg bid); or Group III-delavirdine (400 mg tid), saquinavir-SGC (800 mg tid), and stavudine (40 mg bid). Viral loads, CD4 count, and safety were assessed over a 24-week period with an additional 6-month follow-up. RESULTS 73 patients received randomized therapy; 14 of whom were SQV naïve, with a median baseline viral load of 3.6 log(10) and a CD4 count of 370 cells/mm(3). By 6 months, the median decreases in plasma viral loads were 0.26, 0.71, and 0.29 log(10) copies/mL for groups I, II, and III, respectively. The median increases in CD4 counts, for groups I, II, and III, were 52, 40, and 69 cells/mm(3) at 6 months, respectively. Changes in viral load and CD4 counts at 6 months and 1 year were not significantly different between the treatment groups. More patients discontinued therapy in the ritonavir arm (35%) for drug intolerance or toxicity compared to either the nelfinavir or delavirdine arms (15% and 5%, respectively). In a multivariate analysis, baseline viral load, younger age, and baseline saquinavir resistance were significantly associated with detectable viral load at 24 weeks. CONCLUSIONS The use of antiretroviral agents that pharmacokinetically boost saquinavir levels has a modest benefit in saquinavir-experienced patients.