Abstract

Efavirenz (EFV)-based regimens are potent in treatment-naïve subjects across a wide range of pre-treatment HIV-1 RNA (VL) and CD4 cell counts: 3-year results from ACTG 5095 (A5095)

Ribaudo H.¹, Kuritzkes D.², Lalama C.¹, Schouten J.³, Schackman B.⁴, Gulick R.⁵, AIDS Clinical Trials Group

¹Harvard School of Public Health, Biostatistics, Boston, United States, ²Harvard Medical School, Section of Retroviral Therapeutics, Boston, United States, ³University of Washington, Seattle, United States, ⁴Weill Medical College of Cornell University, Division of Outcomes and Effectiveness Research, New York, United States, ⁵Weill Medical College of Cornell University, Division of International Medicine and Infectious Diseases, New York, United States

Background: Two nucleoside reverse transcriptase inhibitors (NRTI) plus EFV is a widely used initial treatment regimen for HIV-infected individuals but concern remains about virologic potency in subjects with high VL and/or low CD4. We compared VL and CD4 responses based on pre-treatment VL and CD4 in subjects receiving EFV with 2- or 3-NRTIs and assessed whether adding a third NRTI improved responses in any subgroup.

Methods: A5095 randomized 765 treatment-naïve subjects to 2- or 3-NRTIs+EFV with 3 years median follow-up. Pre-treatment VL and CD4 were categorized into pre-determined subgroups (<30,000, 30,000-99,999, 100,000-299,999, >=300,000 copies/ml and <50, 50-199, 200-349, 350-499, >=500 cells/mm³). VL and CD4 subgroup associations with risk of virologic failure (VF) (confirmed VL>200 copies/ml) were analyzed using Cox proportional hazards models; associations with CD4 increases from baseline at specific weeks used normal linear models. Analyses were intent-to-treat.

Results:The risk of VF was not significantly different among VL and CD4 subgroups and revealed no consistent trends (Figure). The risk of VF was not significantly different between subjects taking 2- vs. 3-NRTIs+EFV across VL (P=0.43) and CD4 subgroups (p=0.99). Similar CD4 increases over time were observed across all VL and CD4 subgroups (median=277 cells/mm³). At 48, 96, and 144 weeks, CD4 increases were not significantly different for 2- vs. 3-NRTIs+EFV across VL subgroups (P>0.55) and across CD4 subgroups at weeks 48 and 144 (P>0.12).

Conclusions: In this randomized study of 2- or 3-NRTIs+EFV pre-treatment VL and CD4 were not associated with treatment outcome over a median of 3 years demonstrating the potency of EFV-based regimens across a wide range of CD4 and VL. Adding a third NRTI did not enhance responses in any subgroup.

Back