Abstract Body

Background: Poorer virologic response to NNRTI-containing regimens has been reported in children with prior sdNVP exposure. Despite WHO guidelines now recommending all children <5 years initiate LPV-containing ART, availability is limited and formulations challenging. We therefore compared VL response in children initiating ART aged <3 years in the ARROW trial according to previous sdNVP exposure. Methodology: In 2007-2008, 370 ART-naive children aged <3 years were enrolled from Uganda/Zimbabwe, and randomized to standard 3-drug NNRTI+3TC+ABC (Arm A) vs 4-drug NNRTI+3TC+ABC+ZDV induction, decreasing at 36 weeks to 3-drug NNRTI+3TC+ABC (Arm B) or to 3TC+ABC+ZDV (Arm C). VL was assayed using Abbott Taqman or Roche Amplicor, with lower detection limit 80 c/ml. Predictors of suppression <80 c/ml 48 and 144 weeks after ART initiation were identified using backwards elimination (p=0.1) from pre-ART CD4%, weight/height-for-age, CD4%, VL, sex, centre, ART randomisation, CD4 monitoring (yes/no), current ART as all syrups, and reported missed ART doses in the last 4 weeks, forcing sdNVP and age at ART initiation into the model. Results: Three children initiated ART with EFV and were excluded. In the remaining 367 initiating NVP-based ART aged <3 years, median age was 20 months (IQR 12-27).57(16%) children had received sdNVP. Children receiving sdNVP were younger (median 15 vs 20 months in those without sdNVP, p=0.003) and therefore had slightly higher CD4 counts (median 902 vs 724 cells/ul, p=0.02) and lower weight (median 7.8 vs 8.4 kg, p=0.05), but did not differ significantly in pre-ART CD4% (median 14%), VL (508314 c/ml), weight-for-age Z-score (-2.7) WHO stage (71% 3/4), sex (48% boys), centre/ country or initial ART (p>0.15).There was no evidence that suppression<80 c/ml was any worse with sdNVP vs no sdNVP exposure(Figure; wk48 adjusted (a)OR=1.84 [95% CI 0.83-4.05] p=0.13, wk144 aOR=1.39 [0.66-2.95] p=0.38). At week 48/144, suppression was lower in boys (aOR=0.45/0.56 p=0.004/0.03) and those with higher pre-ART VL (aOR=0.49/0.59 per log10 higher p=0.001/0.01).At week 48, it was also lower in those on syrups (aOR=0.50 p=0.04) or missing ART doses (aOR=0.34 p=0.07); and at week 144 in those on triple NRTI maintenance (aOR=0.58 p=0.04). There was no independent effect of age (p=0.42/0.14). Conclusions: In this cohort of African children <3 years of age starting ART with advanced HIV disease, there was no evidence of poorer VL response to NVP-containing ART in children exposed to sdNVP.