Changes in cerebral function parameters in persons with HIV with symptoms of insomnia switching from dolutegravir- to bictegravir-based antiretroviral therapy.

Sleep disturbances are frequently reported in persons with HIV and have been associated with the use of certain integrase strand transfer inhibitors (INSTIs), such as dolutegravir. This exploratory study assessed changes in cerebral function parameters in individuals with insomnia switching INSTIs. Individuals with an insomnia severity index (ISI) above 8 and virologically suppressed on a dolutegravir-containing ART regimen (DTG-ART) were randomised 1:1 to either continue DTG-ART or switch to bictegravir/emtricitabine/tenofovir alafenamide (BIC-ART) for 120 days. Cerebral function parameters were measured longitudinally at baseline (D0) and day 120 (D120) and included: (1) patient-reported outcomes (PROs) assessing sleep, quality of life (QoL) and symptoms related to ART, (2) resting-state functional cerebral MRI (fMRI), examining functional connectivity networks previously associated with DTG use or sleep and (3) plasma soluble inflammatory biomarkers associated with neuroinflammation or HIV disease progression (Neopterin, CXCL10 and IL-6). Functional connectivity analyses were performed using Seed-Based Correlations (SBC), and correlations between connectivity changes, PRO measures and biomarker concentrations determined. Of 19 individuals (12 DTG-ART, 7 BIC-ART), median age was 55 years (range 28-83), all were male and 17 of white ethnicity. Over 120 days, improvements in sleep and QoL in those randomised to BIC-ART vs. DTG-ART were observed. Median change in Insomnia Severity Index (ISI) score - 9 (-14 to -2) vs. -1 (-10 to -4), p = 0.030, Epworth Sleepiness Scale (ESS) -3.0 (-6 to -1) vs. 2 (-3 to 6), p = 0.007 and Short Form-36 Physical Function (SF36-PF) -5 (-40 to 5) vs. 0 (-5 to 15), p = 0.026) for BIC- vs. DTG- ART, respectively. BIC-ART was also associated with increased functional connectivity in the Default Mode and Salience Networks (both p < 0.05), which correlated with improvements in PRO measures (ESS and SF36-PF, both p < 0.05). No significant changes in soluble biomarkers were observed. Individuals with insomnia switching to BIC-ART had improvements in self-reported sleep, QoL and resting state fMRI networks associated with sleep, when compared to those continued on DTG-ART.