Background
Antiretroviral treatment (ART) management during HIV low-level viremia (LLV) is still unclear. Our aim was to evaluate outcomes after LLV in people with HIV (PWH) according to ART switch.
Methods
Cohort study on PWH on ART with ≥1 episode of LLV since 2014. LLV defined as ≥2 consecutive viral loads (VLs) between 50 and 199 copies/mL ≥30 days apart, following a previous VL <50 copies/mL. In case of multiple LLV episodes, only the first was considered. In case of multiple ART switch during LLV, only the first was considered. Follow-up accrued from first VL 50-199 copies/mL [baseline (BL)] to VS (≥1 VL <50 copies), virological failure [VF (≥2 consecutive VLs ≥200 or ≥1 VL ≥1000 copies/mL)], second ART switch or last available VL measurement. Cox regression model applied to assess predictors of VS; ART switch considered as time-dependent variable.
Results
Among 6794 PWH in care, 451 (6.6%) developed LLV and were included in the analysis: at BL, 380 (84.3%) males assigned at birth, median age 52.6 (interquartile range 45.8-57.9) years, on ART since 14.2 (6.1-20.9) years, virosuppressed since 3.4 (1.3-8.2) years, CD4+/CD8+ 0.7 (0.4-1.0), 347 (76.9%) on a triple therapy, 296 (65.6%) on a high genetic barrier regimen (including bictegravir, dolutegravir, and/or a boosted PI). After a median follow-up of 0.8 (0.4-1.1) years, 366 (81.2%) individuals achieved VS [303/366 (82.8%) maintained stable VS for ≥2 VL determinations] and 16 (3.5%) had VF. ART was switched in 176 (39.0%) PWH after 0.3 (0.1-0.6) years [136/176 (77.3%) achieved VS (stable in 87.5% of them), 4/176 (2.3%) had VF]. Risk of VS was significantly higher after ART switch [unadjusted hazard ratio (uHR)=2.1, 95% confidence interval (95%CI)=1.7-2.6, p<0.001]. This result was confirmed in sensitivity analyses on PWH on high [uHR=2.3, 95%CI=1.7-3.0, p<0.001] or low [uHR=1.9, 95%CI=1.3-2.7, p=0.001] genetic barrier regimens. At multivariable analysis, a higher risk of VS was significantly associated with ART switch, pre-BL history of VF, and resistance to ≥2 drug classes (Table 1). According to RNA-based genotyping, resistance-associated mutations might have been selected during LLV in 7 (1.6%) PWH.
Conclusions
Low-level viremia during ART has a non-negligible prevalence in PWH and could be responsible for the emergence of resistance-associated mutations. Switching ART seems to be beneficial in achieving virological suppression, even in the case of high genetic barrier regimens.
Table or Figure
