Induction therapy with protease-inhibitors modifies the effect of nevirapine resistance on virologic response to nevirapine-based HAART in children

Background: Nevirapine resistance after failed prophylaxis to prevent mother-to-child human immunodeficiency virus (HIV) transmission can compromise subsequent nevirapine-based highly active antiretroviral therapy (HAART). Methods: Nevirapine-exposed children who achieved virologic suppression with lopinavir/ritonavir-based induction HAART before switch to nevirapine-based HAART or who continued the lopinavir/ritonavir regimen were studied. Nevirapine-resistant HIV was quantified (≥1% frequency) in plasma before therapy and archived in peripheral blood mononuclear cells after induction HAART with ultradeep pyrosequencing. The primary endpoint was virologic failure (confirmed viremia ≥1000 copies/mL by 52 weeks) on nevirapine-based HAART, and Receiver operating characteristic analysis identified threshold levels of resistance associated with failure. Results: Nevirapine resistance mutations were detected in plasma at a median frequency of 25.6% in 41 (33%) of 124 children starting HAART at median 9 months of age. After a median nine months of induction HAART, nevirapine-resistant HIV remained archived in cells in 59 (61%) of 96 children (median 13.6% of cells). The threshold frequency of nevirapine resistance in plasma most predictive of virologic failure on nevirapine-based HAART was 25%. Children maintaining resistance before therapy at or above this threshold frequency had a 3.5 fold higher risk of failure (95% confidence interval, 1.1-10.8) than children without detectable plasma resistance. In contrast, virologic failure was not independently associated with age, resistance in plasma below 25% frequencies, or archived in cells. Conclusions: Virologic suppression with lopinavir/ritonavir-based HAART in nevirapine-exposed children raises the threshold level of resistance at which reuse of nevirapine-based therapy is compromised. Standard genotyping may allow identification of children likely to benefit from an induction-switch approach. © The Author 2011.