Neuropsychiatric adverse effects on dolutegravir

We read with interest the recent articles by Borghetti et al. [1] and Bonfanti et al.[2] that report and make an echo to recent concerns about the neuropsychiatric safety of dolutegravir in the real life setting. This prompted us to review data on dolutegravir use in our own real-life cohort of 2260 HIV-infected patients with a specific focus on neuropsychiatric adverse events (NP-AEs) leading to dolutegravir discontinuation, and on additional pharmacological data when available. We performed a retrospective analysis in patients who had initiated dolutegravir between 1 January 2014 and 30 November 2016, monitored in our infectious diseases unit in public hospitals of Marseille, Southeastern France. A total of 517 dolutegravir-based antiretroviral therapies were initiated during the observation period and 55 AEs (10.6%) led to their discontinuation, with 28 (51%) of these adverse events being NP-AEs. Patients’ characteristics and reasons for discontinuation are shown in Table 1. In almost all patients, symptoms occurred during the few months after dolutegravir initiation (median, 4 months), were not life-threatening, did not lead to hospitalization and disappeared quickly after dolutegravir discontinuation (median, 1 month). At initiation of dolutegravir, 94.5% of the patients were pretreated but 52% (29/55) were naïve of integrase strand transfer inhibitor. None of the reported deaths was considered to be drug related. Irritability and sleep disturbances were the most frequently observed NP-AEs. Pharmacological data were available for 12/55 patients with adverse events, including nine with NP-AEs. Median (range) dolutegravir trough concentration (Ctrough) was 1719 ng/ml (811–3730 ng/ml) for those on 50 mg once a day and a Ctrough of 5133 ng/ml was found for the only patient on dolutegravir 50 mg bid.Table 1: Patients characteristics on dolutegravir (n = 517).Our rate of dolutegravir discontinuation for NP-AEs (5.4%) is concordant with those reported in most previous real-life studies, which ranged between 3.4 and 6% [1,3,4]. In contrast, this rate was lower in the CISAI real-life study (1%) [2] and in clinical trials [5]. In addition, we did not observe in our larger cohort the association previously described between abacavir administration and NP-AEs [1,3,4]. In multivariate analysis, we found that women with NP-AEs were significantly older (median age, 51 years; range, 44–66) than women who discontinued dolutegravir because of non-NP-AEs (P = 0.03). In previous reports, age more than 60 years [2,3] and female sex [3] were significantly associated with NP-AEs occurrence. Noteworthy, in our study, among male patients who discontinued dolutegravir because of adverse events, BMI was significantly higher in those with NP-AEs than non-NP-AEs (24 versus 20; P = 0.03). Hoffmann et al.[3] suggested an interest in studying the association between BMI and NP-AEs on dolutegravir but to our knowledge we report here for the first time this association. Regarding pharmacological data, dolutegravir Ctrough was supratherapeutic and greater than the median value of 1340 ng/ml found by Yagura et al. as significantly higher in patients with than without NP-AEs (HIV Glasgow 2016; unpublished abstract), which suggested a relationship between dolutegravir exposure and toxicity. Taken together, previous findings show that NP-AEs emerge as a cause of dolutegravir discontinuation in real-life HIV cohorts in Europe. Other cohort studies at the national and international scale are needed to precise their incidence and putative mechanisms. This is particularly warranted as guidelines have moved to integrase strand transfer inhibitor-based combined antiretroviral therapies for initial treatments. Acknowledgements Conflicts of interest There are no conflicts of interest.