BACKGROUND: Early virological response to antiretroviral therapy is predictive of long-term treatment outcome in therapy-naive patients. In treatment-experienced patients, such correlations are less well defined, because initial responses may be less pronounced and transient because of accumulated cross-resistance to prior therapies. Our objectives were to explore how the virological and immunological status of treatment-experienced patients at an early time point (week 12) during enfuvirtide-based therapy predicted their responses at weeks 24, 48, and 96 in the T-20 versus Optimized Regimen Only (TORO) trials. METHODS: Post hoc, modified, on-treatment and intent-to-treat analyses were performed to determine whether the relationship between virological and immunological outcomes at weeks 24, 48, and 96 were predicted by the patients' week-12 responses to therapy. RESULTS: Using a modified on-treatment analysis for patients who, by week 12, achieved a decrease in their HIV-1 RNA load of > or =1 log10 copies/mL, 39.2% (95% CI, 33.6%-44.8%) and 59.5% (95% CI, 53.8%-65.1%) achieved a viral load of <50 copies/mL or <400 copies/mL at week 96, respectively, compared with 1.3% (95% CI, 0%-3.8%) and 2.6% (95% CI, 0%-6.1%) of patients, respectively, who did not achieve an early virological response. Using the same modified on-treatment analysis method for patients who, at week 12, achieved a CD4 cell count increase of > or =50 cells/mm3, 87.2% (95% CI, 82.6-91.8) maintained or improved this response through week 96, compared with 56.6% (95% CI, 47.5-65.8) of patients who did not achieve this early categorical immunological response. CONCLUSION: Enfuvirtide-based treatment regimens are associated with a rapid and durable response. Week-12 virological and immunological responses to treatment with enfuvirtide are predictive of subsequent outcomes in triple-class treatment-experienced patients.