Incidence of HIV-1 drug resistance among antiretroviral treatment-naive individuals starting modern therapy combinations

Background. Estimates of drug resistance incidence to modern first-line combination antiretroviral therapies against human immunodeficiency virus (HIV) type 1 are complicated by limited availability of genotypic drug resistance tests (GRTs) and uncertain timing of resistance emergence. Methods. Five first-line combinations were studied (all paired with lamivudine or emtricitabine): efavirenz (EFV) plus zidovudine (AZT) (n = 524); EFV plus tenofovir (TDF) (n = 615); lopinavir (LPV) plus AZT (n = 573); LPV plus TDF (n = 301); and ritonavir-boosted atazanavir (ATZ/r) plus TDF (n = 250). Virological treatment outcomes were classified into 3 risk strata for emergence of resistance, based on whether undetectable HIV RNA levels were maintained during therapy and, if not, whether viral loads were <500 copies/mL during treatment. Probabilities for presence of resistance mutations were estimated from GRTs (n 5 2876) according to risk stratum and therapy received at time of testing. On the basis of these data, events of resistance emergence were imputed for each individual and were assessed using survival analysis. Imputation was repeated 100 times, and results were summarized by median values (2.5th-97.5th percentile range). Results. Six years after treatment initiation, EFV plus AZT showed the highest cumulative resistance incidence (16%) of all regimens (500 copies/mL). Conclusions. The inclusion of TDF instead of AZT and ATZ/r was correlated with lower rates of resistance emergence, most likely because of improved tolerability and pharmacokinetics resulting from a once-daily dosage.