Background: Primary analyses of TORO 1 and 2 established efficacy and safety and provided an assessment of treatment response among demographic, baseline (BL) characteristics, and viral sensitivity subgroups. Pooled analyses gives a more precise estimation of treatment benefit and summary of the safety of enfuvirtide (ENF).

Methods: Patients were randomized to ENF + optimized background (OB) antiretroviral (ARV) or OB alone. Change from BL in HIV-1 RNA was estimated by analysis of covariance with phenotypic sensitivity score as a covariate. Pooled analyses were pre-planned. Safety assessments included clinical and laboratory evaluations; the clinical evaluations included separate evaluation of injection site reactions (ISRs).

Results: Differences between the two arms in mean change from BL in HIV RNA between ENF+OB and OB groups was -0.85 log₁₀ and for CD4⁺ cell count was 37 cells/mm³, each significantly favoring ENF+OB (p < 0.0001). The ENF+OB regimen provided consistent additional benefit compared to OB alone across demographic, BL characteristics, and background ARV primary mutations, and by treatment factors such as use of other investigational ARVs and number of active drugs in the OB. The ENF+OB and the OB groups responded better when virus was sensitive to more ARVs, and the benefit of ENF+OB was maintained across the spectrum of viral sensitivity. ISRs were the most common (98%) adverse events (AEs) associated with ENF treatment. The majority was mild to moderate in intensity with only 9.4% of patients (pts) having severe pain requiring analgesics or limiting usual activities. There was no evidence of an increasing severity of ISRs over time. The most common AEs were diarrhea (ENF+OB, 27%; OB, 34%), nausea (ENF+OB, 20%; OB, 24%), and fatigue (ENF+OB, 16%; OB, 17%). The most common AEs with higher numerical incidence on the ENF+OB group were headache (ENF+OB, 11.9%; OB, 11.1%), insomnia (ENF+OB, 11.3%; OB, 8.7%), and peripheral neuropathy (ENF+OB 8.9%; OB 6.3%). Four (4; 0.6%) pts had an AE consistent with a systemic hypersensitivity reaction to ENF. Eosinophilia was more common on ENF+OB (10.1%) compared to OB (2.4%), and there were no other consistent differences in laboratory abnormalities between the 2 arms.

Conclusions: Combined analyses from TORO 1 and 2 confirmed and clarified the benefit of ENF+OB versus OB alone. Across the spectrum of characteristics in the multi-resistant, treatment-experienced population, ENF+OB provided consistent efficacy and safety.