Comment on: Long-term outcome of dolutegravir-containing regimens according to sex: data from the ICONA study.

Abstract

We read with interest the work by D’arminio Monforte et al.,1 evaluating the long-term risk of treatment failure of dolutegravir-based regimens. In their study, the authors found no difference in treatment failure by sex but observed a higher risk of discontinuation due to toxicity in women. Tolerability of dolutegravir, which has become one of the most used molecules as part of antiretroviral regimens, has been analysed by previous studies with various results.2,3 We conducted a retrospective analysis on a multicentre cohort of people living with HIV (PLWHIV) starting a dolutegravir-based regimen either as a first-line strategy or as a switch strategy. We evaluated time to dolutegravir discontinuation, as well as time to virological failure (VF). We also performed Cox regression analyses to find predictors of discontinuation or VF. We analysed data from 3960 PLWHIV: 2915 (73.6%) were males, with a median age of 50 years (IQR 41–56). In our cohort, 887 (22.4%) started dolutegravir as part of a first-line regimen; among the 77.6% of our population that started dolutegravir as treatment-experienced, median time of antiviral therapy (ARV) exposure was 10.2 years (IQR 3.2–18.9) and 1484 (48.3% of experienced PLWHIV) had at least one previous VF. Full population characteristics are shown in Table 1. We registered 1019 discontinuations during 10 129 patient-years of follow-up (PYFU), a rate of 10.0 per 100 PYFU. Reasons for dolutegravir discontinuation were: simplification to a 2-drug regimen/single tablet regimen (216, 21.2%), treatment intensification (163, 16.0%), CNS toxicity (136, 13.3%), gastrointestinal toxicity (58, 5.7%), hypersensitivity reaction (28, 2.7%), other toxicity (113, 11.1%), death (40, 3.9%), drug–drug interaction (21, 2.1%), concern of cardiovascular disease (10, 1.0%), pregnancy (10, 1.0%) and other/unknown (224, 22.0%). We found significant differences regarding reasons for dolutegravir discontinuation between naive and experienced PLWHIV (P < 0.001), with a higher rate of naive PLWHIV discontinuing dolutegravir due to simplification (31.0% versus 16.6%). We did not find significant differences in the rate of discontinuations due to toxicity between naive versus experienced PLWHIV, males versus females or between age groups. Overall, we observed estimated probabilities of maintaining dolutegravir of 73.7% at 144 weeks and 65.5% at 240 weeks. We found a significant difference in estimated probability of maintaining dolutegravir between treatment-naive and -experienced PLWHIV (51.0% versus 69.2%, respectively, at 240 weeks, log-rank <0.001). Among treatment-experienced individuals, a longer time of virological suppression was associated with a reduced probability of discontinuation [adjusted hazard ratio (aHR) 0.995, 95% CI 0.993–0.997, P < 0.001] in a multivariate analysis; meanwhile, among treatment-naive PLWHIV, the presence of an AIDS-defining event at diagnosis was associated with a higher risk of discontinuation (aHR 2.57, 95% CI 1.40–4.70, P = 0.002). Estimated probabilities of not discontinuing dolutegravir due to toxicity were 90.1% at 144 weeks and 89.3% at 240 weeks. Among treatment-naive individuals, the presence of an AIDS-defining event at diagnosis was associated with a higher risk of discontinuation due to toxicity (aHR 2.24, 95% CI 1.34–3.73, P = 0.002). During 9738.2 PYFU we registered 318 VFs, a rate of 3.3 VFs per 100 PYFU. Estimated probabilities of remaining free of VF during treatment with dolutegravir were 90.2% and 87.3% at 144 and 240 weeks, respectively. We did not observe differences in the rate of VF between naive versus experienced PLWHIV, males versus females or between age groups. At a multivariate analysis, among treatment-experienced PLWHIV, a previous VF was a predictor of failing during treatment with dolutegravir (aHR 3.47, 95% CI 2.54–4.74, P < 0.001), whereas a longer time since ARV initiation inversely correlated with VF (aHR 0.96, 95% CI 0.93–0.98, P = 0.003).