Therapy of HIV infection with protease inhibitors (PIs) may be associated with improvements in CD4 T-cell number via a mechanism that is independent of effects on plasma viral load (VL). To compare CD4 responses of patients who receive reverse transcriptase inhibitor (RTI) therapies with or without a PI, matched for viral exposure. Patient data were analyzed from two prospective randomized trials of antiviral therapy with or without nelfinavir. Total viral exposure over 24 weeks was estimated by viral area under the curve (AUC), which reflects baseline viral load, slope of virologic decay, viral nadir, and duration of suppression. Patients were stratified into quartiles on the basis of viral AUC, and CD4 T-cell responses were evaluated between PI-containing and RTI-only treatment groups within each quartile. In both trials, patients receiving nelfinavir had greater CD4 T-cell increases than patients receiving RTI alone. Analysis of variance modeling revealed increased CD4 T-cell responses in PI-treated groups at all time points after the second week. These differences were significant (p < .05) at weeks 12, 24, 28, 32, 36, 40, and 48 in one study, and weeks 1, 2, 4, 6, 8, 12, 16, 20, 24, 28, 32, 36, and 44 in the other. Within quartiles matched for viral AUC, absolute CD4 T-cell change from baseline was greater in the PI-treated patients at 84% (101/120) of time points analyzed. Nelfinavir-containing therapy is associated with enhanced increases in CD4 T-cell number compared to RTI therapy alone with equivalent antiviral effect. These data suggest that PIs influence CD4 T-cell number through a nonvirologic effect.