Objectives: To examine the impact of transmitted drug resistance (TDR) on response to first-line regimens with integrase strand transfer inhibitors (INSTIs) or boosted protease inhibitors (bPIs).
Methods: From an Italian observational database (ARCA) we selected HIV-1-infected drug-naive patients starting two NRTIs and either an INSTI or a bPI, with an available pre-ART resistance genotype. The endpoint was virological failure (VF; plasma HIV-1 RNA >200 copies/mL after week 24). WHO surveillance drug resistance mutations and the Stanford algorithm were used to classify patients into three resistance categories: no TDR (A), TDR but fully-active ART prescribed (B), TDR and at least low-level resistance to one or more prescribed drug (C).
Results: We included 1365 patients with a median follow-up of 96 weeks (IQR 54-110): 1205 (88.3%) starting bPI and 160 (11.7%) INSTI. Prevalence of TDR was 6.1%, 12.5%, 2.6% and 0% for NRTI, NNRTI, bPI and INSTI, respectively. Cumulative Kaplan-Meier estimates for VF at 48 weeks were 11% (95% CI 10.1%-11.9%) for the bPI group and 7.7% (95% CI 5.4%-10%) for the INSTI group. In the INSTI group, cumulative estimates for VF at 48 weeks were 6% (95% CI 4%-8%) in resistance category A, 5% (95% CI 1%-10%) in B and 50% (95% CI 30%-70%) in C (P < 0.001). Resistance category C [versus A, adjusted hazard ratio (aHR) 12.6, 95% CI 3.2-49.8, P < 0.001] and nadir CD4 (+100 cells/mm3, aHR 0.6, 95% CI 0.4-0.9, P = 0.03) predicted VF. In the bPI group, VF rates were not influenced by baseline resistance.
Conclusions: Our data support the need for NRTI resistance genotyping in patients starting an INSTI-based first-line ART.