Abstract Body

Background: In resource-limited countries, NNRTI-based regimen, as recommended by WHO, may result in emergence of more HIV drug resistance, because of a low genetic barrier.The purpose of the trial is to compare the efficacy and tolerance of LPV/r and NVP-based regimens and 2 WHO nucleoside backbones in naive HIV infected patients (p.). Methodology: Naive p. from 5 clinics in Lubumbashi (Congo-RDC) were randomized to receive LPV/r tablets versus NVP combined with TDF/FTC or ZDV/3TC. VL and CD4 were performed at baseline (BL) and every 24 weeks (W). The primary endpoint was the % of p. with therapeutic failure defined as clinical and virologic failures (VL>1000 c/ml) (missing data=failure), assessed at W48 and 96. We present here the results of 144 W of follow-up. Results: 425 Black African p. (72% female; median (md) age 38 years, md CD4 165/μL; md VL 5.2 log c/ml) were randomized (216 in LPV/r, 209 in NVP). BL characteristics were comparable. In the ITT analysis, previous results (abstract 88LB, CROI 2012) showed no difference between LPV/r and NVP treatment arms at W96 except a higher proportion of virologic failure (VF) in p. on NVP-based regimens. W144 ITT analysis showed a significant difference on endpoints between LPV/r (94/216) and NVP (111/209) (p=0.0479) and persistence of a significant difference in VF rate (20/216 vs 37/209 for LPV/r and NVP, respectively) (p= 0.015). BL genotypes showed NNRTI mutations (mt) in 3/31 NVP-failing p. and no PI mt in LPV/r-failing patients. At time of failure, NNRTI mt were seen in 23/26 NVP-failing p. and 0/13 primary PI mt in LPV/r failing patients. NRTI mt were seen in 19/26 p. in NVP arm (including K65R in 7 p. and M184V in 18 p.) vs 3/13 p. in LPV/r arm (M184V in 3 p.). Md CD4 change from BL was significant higher in LPV/r arm (251 cells/μL [interquartile range (IQR) 153;384]) compared with NVP arm (174 cells/μL [IQR 102-330]) (p= 0.0093). There was no difference in adherence > 95% between groups (73.6 vs 74.4 for LPV/r vs NVP). The discontinuation rate for toxicity was low: 2 p. (0.9%) in LPV/r and NVP groups. Hypersensitivity reactions were more frequent in the NVP (8/209) vs LPV/r (0/216) groups (p=0.0032). Diarrhea, nausea and vomiting were predominant with LPV/r (p=<0.0001). Md change from BL in estimated GFR was significant higher for NVP +20.5 ml/min [IQR 6.3;38.9] vs LPV/r +11.5[IQR -1;28.5] (p=0.0044) without significant difference between backbones. Compared to BL, total cholesterol, LDL, HDL cholesterol, triglycerides and haemoglobin increased over time, but without significant difference between treatment groups. Conclusions: In a resource-limited setting after 144 weeks of follow-up NNRTI-NRTI first-line regimen is associated with more virologic failure, more drug resistance mutations and a lower immunologic response than a PI-based regimen.