Abstract
BACKGROUND: In a randomized trial comparing nevirapine (NVP)-based versus lopinavir/ritonavir (LPV/r)-based antiretroviral therapy (ART) in HIV-infected children [primary endpoint discontinuation of study treatment for any reason or virologic failure by week 24] aged 2 months to 3 years, we assessed whether clinical, virologic, immunologic and safety outcomes varied by prior single-dose NVP exposure (PrNVP) for prevention of mother-to-child HIV transmission and other covariates. METHODS: Efficacy was assessed by time to ART discontinuation or virologic failure, virologic failure/death and death; safety by time to ART discontinuation because of a protocol-defined toxicity and first ≥ grade 3 adverse event; immunology and growth by changes in CD4%, weight/height World Health Organization z-scores from entry to week 48. Cox proportional hazards and linear regression models were used to test whether treatment differences depended on PrNVP exposure and other covariates. RESULTS: Over a median follow up of 48 (PrNVP) and 72 (no PrNVP) weeks, there was no evidence of differential treatment effects by PrNVP exposure or any other covariates. LPV/r-based ART was superior to NVP-based ART for efficacy and safety outcomes; however, those on NVP had larger improvements in CD4%, weight and height z-scores. Lower pretreatment CD4% and higher HIV-1 RNA levels were associated with reduced efficacy, lower pretreatment CD4% with shorter time to ART discontinuation because of a protocol-defined toxicity, and no PrNVP with shorter time to first grade ≥ 3 adverse event. CONCLUSIONS: Differences between LPV/r and NVP ART in efficacy, safety, immunologic and growth outcomes did not depend on PrNVP exposure, prior breast-feeding, sex, HIV-1 subtype, age, pretreatment CD4%, HIV-1 RNA or World Health Organization disease stage. This finding should be considered when selecting an ART regimen for young children.