Background Integrase strand transfer inhibitors (InSTIs) are recommended for first-line treatment of HIV-infection. We identified risk factors, including baseline minor InSTI resistance mutations, for treatment failure of InSTI-based regimens. Methods We studied time to treatment failure and time to viral suppression among 1419 drug-naive patients in the Swiss HIV Cohort Study. We performed Cox regression models adjusted for demographic factors, baseline HIV RNA/CD4 cell counts, AIDS defining events and the type of InSTI. In 646 patients with a baseline genotypic resistance test of the integrase, we studied the impact of minor integrase resistance mutations. Results We observed 121 virological failures during 18’447 person-years of follow-up. A baseline viral load ≥100’000 cps/mL (multivariable Hazard Ratio (mHR): 2.2, 95% CI: 1.3-3.6) and an AIDS defining event (mHR: 1.8, 95% CI: 1.1-3.0) were associated with treatment failure. CD4 counts between 200-500 cells/µL (mHR: 0.5, 95% CI: 0.3-0.8) and >500 cells/µL (mHR: 0.4, 95% CI: 0.2-0.7) were protective. Median [IQR] time to viral suppression was 50 [29,107] days. Time to suppression was shorter in lower viral load strata (mHR: 0.7, 95% CI: 0.6-0.8) and in dolutegravir-based therapy (mHR: 1.2, 95% CI: 1.0-1.4). Minor resistance mutations were found at baseline in 104/646 (16%) patients with no effect on treatment outcome. Conclusion Among drug-naïve HIV-infected individuals treated with InSTI-based regimens, factors associated with treatment failure, in particular high viral load and low CD4 counts remain similar to older treatments. Minor InSTI resistance mutations had no impact in this large observational cohort.

No abstract