HIV-1 low-level viraemia predicts virological failure in first-line and second-line ART-experienced individuals in India: A retrospective longitudinal study

Objective: To study the prevalence of low-level viraemia (LLV) and its association with virological failure (VF). Methods: We conducted a retrospective analysis of 3498 participants at YRG CARE, Chennai, India (2013-2018) on antiretroviral therapy (ART) for >= 6 months with two or more plasma viral load (pVL) measurements. Results were stratified for those with pVL <1000 copies/mL: fully suppressed (FS) (pVL <40), low-LLV (pVL 40-199), mid-LLV (pVL 200-399), and high-LLV (pVL 400-999). The study assessed the association with VF (pVL >1000 copies/mL) using Cox proportional hazard model. Results: Among 3498 participants, 2965 (84.8%) were FS and 533 (15.2%) were LLV. During the follow-up, 348 (10%) experienced VF, with 222 (6.3%) experienced after LLV (42% of LLV) and 126 (3.6%) experienced after FS (4.3% of FS). When compared with FS, those with LLV had a greater risk of VF [adjusted hazard ratio (aHR) = 12.7; 95% confidence interval (CI): 10.2-15.9]. First-line participants had a higher VF incidence (aHR = 15.8, 95% CI: 11.4-21.9) than second-line participants (aHR = 5.6, 95% CI: 4.1-7.7). Those with high-LLV had the highest VF risk (aHR = 22.856, 95% CI: 15.204-34.359 vs. aHR = 8.186, 95% CI: 5.564-12.043, for first-line vs. second-line participants, respectively), followed by those with mid-LLV (aHR = 13.375, 95% CI: 8.327-21.483 vs. aHR = 6.261, 95% CI: 4.044-9.695) and low-LLV (aHR = 12.976, 95% CI: 7.974-21.118 vs. aHR = 4.158, 95% CI: 2.826-6.119). Conclusions: The prevalence of LLV was intermediate in our study population. There was a higher risk of VF among individuals with LLV, and this risk increased with the increasing levels of LLV. Close monitoring of individuals experiencing LLV could help in the early identification of VF.