Objective: To examine the effect of adding nevirapine (NVP) and/or hydroxyurea (HU) to a triple nucleoside analogue reverse transcriptase inhibitor (NRTI) regimen in terms of efficacy and tolerability.
Methods: : HIV-1-infected, treatment-naive adults were randomized, using a factorial design, to add NVP and/or HU to the triple NRTI backbone of zidovudine plus lamivudine plus abacavir. Primary endpoint was treatment failure, defined as having plasma HIV RNA levels > 50 copies/ml after week 24, or discontinuation of randomized treatment. Follow-up was 72 weeks.
Results: For the 229 subjects, median plasma HIV-1 RNA was 4.61 log10 copies/ml and median CD4 cell count was 269 x 10 cells/l. NVP users reached plasma HIV-1 RNA < 50 copies/ml more rapidly than subjects using no NVP (log-rank test; P = 0.011). In the as-treated analysis, 21.6% of subjects using NVP versus 48.8% using no NVP reached the primary endpoint (P = 0.013). In the intent-to-treat analysis, 83.3% of subjects using HU versus 73.0% using no HU experienced treatment failure (P = 0.060), while no difference was observed in the as-treated analysis (34.5 versus 36.7%). Differences in the intent-to-treat analysis were accounted for by toxicity: 52.6% of subjects using HU experienced toxicity leading to discontinuation of randomized treatment versus 28.7% of subjects using no HU.
Conclusion: The use of NVP in addition to a triple NRTI regimen improved both short- and long-term antiretroviral efficacy. The use of HU significantly contributed to treatment failure because of toxicity.