This study evaluates the frequency and causes of dolutegravir (DTG) discontinuation along 5 years of follow-up, in both antiretroviral treatment (ART)-naive and experienced people living with HIV (PLWH). This is a prospective multi-center cohort study enrolling PLWH on DTG from July 2014 until November 2020. DTG-durability was investigated using the Kaplan-Meier survival curve. The Cox proportional-hazards model was used for estimating the hazard ratio (HR) of DTG discontinuation for any cause, and for adverse events (AEs). Nine hundred sixty-three PLWH were included, 25.3% were women and 28.0% were ART-naive. Discontinuations for any causes were 10.1 [95% confidence interval (95% CI) 8.9-11.5] per 100 person-years, similar in most regimens, with the apparent exception of tenofovir alafenamide/emtricitabine+DTG (p < 0.0001). In the multivariable Cox regression model, non-Caucasian ethnicity, age ≥50 years, and lower estimated glomerular filtration rate (eGFR) were associated with a higher probability of DTG interruption. The incidence rate of virological failure was 0.4 (95% CI 0.2-0.7) per 100 person-years, while the estimated discontinuation rate for AEs was 4.0 (3.2-4.9) per 100 person-years. Thirty-four DTG interruptions were due to grade ≥3 events (10 central nervous system, 6 hypersensitivity, 3 renal, 3 myalgia/asthenia, 3 abdominal pain, 2 gastrointestinal, and 7 other events). People with lower body mass index, age ≥50 years, and lower eGFR were at higher risk of AEs, while dual combinations were protective (HR 0.41 compared with abacavir/lamivudine/DTG, 95% CI 0.22-0.77). In this prospective observational study, we found high DTG durability and a low rate of virological failures. Dual therapies seemed protective toward AEs and might be considered, when feasible, a suitable option to minimize drug interactions and improve tolerability.
Keywords: HIV; adverse events; dolutegravir; durability; safety; toxicity; virolgical failure.