The durability of virologic response to antiretroviral therapy is dependent on the potency, tolerability, and adherence level of the regimen. In a prospective, nonrandomized cohort study, we compared the treatment outcome of a nelfinavir-based highly active antiretroviral therapy (HAART) regimen with that of an indinavir-based regimen, over 1 year of routine clinical practice. Information was derived from 134 treatment-naïve HIV-1-infected patients initiated on triple therapy with either nelfinavir (n = 44) or indinavir (n = 90). The proportions of patients achieving a virological response were similar between treatment groups (>1 log(10) reduction in HIV RNA at 3 months in 95% of patients taking nelfinavir and 88% taking indinavir; HIV RNA <50 copies/mL after 1 year in 79% and 69% of patients, respectively). Predicting factors for 1-year virological suppression were initial virological response (p =.02) and adherence >90% (p =.0001). Over 90% adherence was achieved in 70% of patients taking nelfinavir compared with 41% of those taking indinavir (p =.01). The probability of remaining on the initial protease inhibitor (PI) after 12 months was 77% in the nelfinavir group and 66% in the indinavir group, with the median time to changing treatment being 519 days and 462 days, respectively. Gastric intolerance and nephritic colic were the most common reasons for changing therapy in the indinavir group. In the clinical setting, HAART based on initial nelfinavir and indinavir therapy was associated with similarly good virological and immunological suppression at 1 year, however, nelfinavir-based treatment was associated with a longer durability, probably due to a better adherence and tolerance pattern.