| |
Brain Affects by Integrase Inhibitors & ARTs - Altered neuropsychological performance and reduced brain volumetrics in people living with HIV on integrase strand transfer inhibitors
|
| |
| |
Download the PDF here
I will mention there is no consideration in the analysis of diabetes & if that might have affected the outcome since other studies have found an increase in diabetes among PLWH on dolutegravir. Jules
2019
Jane A. O'Hallorana, Sarah A. Cooleyb, Jeremy F. Strainb, Anna Boerwinkleb, Robert Paulc, Rachel M. Prestia and Beau M. Ancesb,d
Here we present data indicating a modest worsening in the learning and memory domain of PLWH on INSTI-based regimens. We also observed significant lower brain volumes, especially within the frontal, brain stem and cerebellar regions. Frontotemporal connections are involved in multiple cognitive functions including learning and memory, and are affected in PLWH [49].
in the Discussion section: Although not definitive, direct neurotoxicity has been reported in vitro with ART. cART has been shown to lead to oxidative stress and neuronal damage and may account for some of the observed reductions in brain volumes [35]. A recent autopsy study demonstrated higher likelihood of neuronal phospho-tau lesions in the putamen of patients on darunavir, whereas ritonavir use was associated with marked microgliosis in the putamen, both suggestive of cerebral degenerative changes. However, there was insufficient INSTI use to allow for their assessment [36]. Significantly, a recent study in primary rat neuroglial cultures demonstrated neurotoxicity of ETG but not RAL or DTG [37]. Clinically, concerns have been raised regarding increased prevalence of neural-tube defects (NTD) in infants born to mothers treated with DTG at the time of conception. In a surveillance study that is currently ongoing in Botswana, the prevalence of NTD in those who received DTG-based therapy was 0.94% compared with 0.12% in those in non-DTG-based regimens. Neurotoxicity concerns have resulted in cART guideline changes for women of childbearing age. However, limited data exist on differences in brain volumes according to type of cART in PLWH [38].
Interindividual variabilities in CSF penetration of antiretrovirals may contribute to the relative risk of neuropsychiatric side effects. Higher CSF concentrations of certain medications can produce neurotoxicity, whereas lower concentrations may increase viral replication and inflammation in the central nervous system [39,40]. Neuropsychiatric symptoms in PLWH have been associated with interindividual variability in CSF concentration of RAL [41]. EFV use was associated with improvement in neuropsychological testing, likely due to viral suppression; however, higher blood levels of EFV corresponded with worse neuropsychological performance [42]. The expression of neuropsychiatric symptoms may be due to complex interactions among HIV, cART and the host.
Abstract
Objectives:
Neuropsychiatric symptoms have been reported in people living with HIV (PLWH) on integrase strand transfer inhibitors (INSTIs) in postmarketing analysis. Limited data exist regarding brain integrity (function and structure) in PLWH prescribed INSTIs compared with other HIV treatment regimens.
Design:
A cross-sectional analysis of PLWH on combined antiretroviral therapy aged more than 18 years at a single institution.
Methods:
Neuropsychological tests were administered to calculate domain deficit scores in learning/memory, executive function and motor/psychomotor domains. Cortical and subcortical volumes from MRI were obtained using the FreeSurfer software suite (v5.3).
Results:
Of 202 participants, median age 55 (48, 60) years old, 49% were on INSTI-based combined antiretroviral therapy. PLWH on INSTIs were similar to individuals on non-INSTIs in terms of age, sex, race, education years, smoking history, depression scores, psychiatric medication use, presence of hepatitis C infection, history of substance use, HIV infection duration and recent or nadir CD4+ T-cell count. Participants in the INSTI group performed worse than non-INSTI users in the verbal learning and memory domain [1.5 (interquartile range 0, 2.5) versus 1 (0, 2); P = 0.016]. The INSTI and non-INSTI groups were similar for other cognitive domains. Frontal, brain stem and cerebellar volumes were reduced in INSTI compared with non-INSTI users (all P = <0.05).
The proportion of PLWH with abnormal learning/memory domain scores (primarily within learning and retention) was significantly higher in the INSTI group compared with the non-INSTI group [73 (74%) versus 61 (59%), P = 0.029]. There was no difference in executive function or motor/psychomotor DDS. Participants in the INSTI group had worse TDS than those in the non-INSTI group [0.75 (0.33, 1.25) versus 0.5 (0.25, 1); P = 0.049] that primarily reflects deficits in the learning/memory domain score. The proportion of PLWH with abnormal TDS was also significantly higher in the INSTI group compared with the non-INSTI group [70 (71%) versus 57 (55%), P = 0.024].
Conclusion:
We demonstrated modest differences in learning/memory performance and smaller brain volumes in PLWH on INSTI-based regimens compared with non-INSTI users. Prospective studies are needed to define mechanisms and the clinical significance of reduced brain integrity in PLWH on INSTIs.
Introduction
HIV-associated neurocognitive disorder (HAND) continues to occur in people living with HIV (PLWH), even in the era of combined antiretroviral therapy (cART) [1–3]. The prevalence of HAND is less common among virally suppressed PLWH when compared with demographically similar individuals with detectable viral load [4]. Similarly, studies report lower frequencies of progressive neuropsychological decline [5,6] and/or structural brain changes in virally suppressed PLWH [7]. However, concerns remain about iatrogenic effects of cART on brain integrity (function and structure), which may contribute to HAND. The nonnuclease reverse transcriptase inhibitors (NNRTI), especially efavirenz (EFV), have been linked to incident neuropsychiatric adverse events and worsening cognition [8,9]. NNRTIs have been replaced as first line therapy by integrase strand transfer inhibitors (INSTIs).
Although INSTIs were well tolerated in clinical trials, concerns have emerged regarding the risk of neuropsychiatric symptoms. Several postmarketing studies have reported insomnia and depressive symptoms among PLWH who initiated dolutegravir (DTG)-based [10–12] or raltegravir (RAL)-based regimens [13–15]. However, studies of the effects of INSTIs on brain function and structure are lacking. To address this gap in knowledge, we examined neuropsychological performance and brain volumetrics in a cohort of PLWH on INSTI-based regimens compared with a well matched group of PLWH receiving non-INSTI-based regimens.
Discussion
Antiretroviral drugs have previously been associated with neuropsychiatric symptoms including dizziness, sleep disturbance, abnormal dreams, anxiety and depression. Despite this, limited data exist in relation to the effect of specific antiretroviral drugs on neuropsychological function [34]. The primary aim of this study was to determine if differences existed in neuropsychological function and brain structure in PLWH on INSTI-based cART compared with a well matched group on non-INSTI-based cART. Here, we demonstrated worse learning/memory performance and smaller regional brain volumes in PLWH on INSTI-based regimens. This could be due to multiple possible causes including direct neurotoxicity of the INSTI cART regimens themselves, variation in cerebrospinal fluid (CSF) concentrations of different cART agents, the effect of a recent switch of regimen to an INSTI-based regimen or different drug interactions between INSTI versus protease inhibitor or NNRTI-based cART.
Although not definitive, direct neurotoxicity has been reported in vitro with ART. cART has been shown to lead to oxidative stress and neuronal damage and may account for some of the observed reductions in brain volumes [35]. A recent autopsy study demonstrated higher likelihood of neuronal phospho-tau lesions in the putamen of patients on darunavir, whereas ritonavir use was associated with marked microgliosis in the putamen, both suggestive of cerebral degenerative changes. However, there was insufficient INSTI use to allow for their assessment [36]. Significantly, a recent study in primary rat neuroglial cultures demonstrated neurotoxicity of ETG but not RAL or DTG [37]. Clinically, concerns have been raised regarding increased prevalence of neural-tube defects (NTD) in infants born to mothers treated with DTG at the time of conception. In a surveillance study that is currently ongoing in Botswana, the prevalence of NTD in those who received DTG-based therapy was 0.94% compared with 0.12% in those in non-DTG-based regimens. Neurotoxicity concerns have resulted in cART guideline changes for women of childbearing age. However, limited data exist on differences in brain volumes according to type of cART in PLWH [38].
Interindividual variabilities in CSF penetration of antiretrovirals may contribute to the relative risk of neuropsychiatric side effects. Higher CSF concentrations of certain medications can produce neurotoxicity, whereas lower concentrations may increase viral replication and inflammation in the central nervous system [39,40]. Neuropsychiatric symptoms in PLWH have been associated with interindividual variability in CSF concentration of RAL [41]. EFV use was associated with improvement in neuropsychological testing, likely due to viral suppression; however, higher blood levels of EFV corresponded with worse neuropsychological performance [42]. The expression of neuropsychiatric symptoms may be due to complex interactions among HIV, cART and the host.
Previous studies have demonstrated neuropsychological dysfunction in individuals with both sleep disturbances and depression [43,44]. High prevalence of sleep disturbances have been reported in PLWH and have been associated with a myriad of factors including depression and specific antiretroviral drugs [45]. However, we did not observe differences in subjective sleep measures between the groups. Of note in a recent study that demonstrated increased DTG peak concentration in older (>60 years) PLWH, there was also no differences in sleep quality noted at 3 or 6 months [46].
Empirical data on INSTIs and neuropsychological function are limited. A small prospective study reported decline in neuropsychological function after 24 weeks of RAL in virally suppressed PLWH at least 60 years old [47]. Of note, a recent pilot study included comprehensive assessment of cerebral function parameters including cognitive function, CSF parameters and MRI imaging in 20 patients who switched from a RAL-based regimen to a DTG-based regimen and did not demonstrate changes in any parameters [48]. Here we present data indicating a modest worsening in the learning and memory domain of PLWH on INSTI-based regimens. We also observed significant lower brain volumes, especially within the frontal, brain stem and cerebellar regions. Frontotemporal connections are involved in multiple cognitive functions including learning and memory, and are affected in PLWH [49].
The current study has several limitations. Although the groups are well matched for baseline demographics as well as HIV clinical variables, the cross-sectional design means that it is subject to the inherent biases associated with this study design and does not allow examination of causal pathways. As is common in the management of PLWH for several years, the majority of study participants had previous cART treatment but data on the reason for switching regimens were not available. Although many study participants were likely switched to regimens with better safety and tolerability profiles as part of routine standard of care, we cannot rule out potential channelling bias introduced by these switches. To account for this, the use of previous cART was included in our regression model. To address this further, we examined the cohort of participants who were on their first cART regimen. Although the cohort was small, the learning/memory DDS was also worse in the subgroup on INSTIs in this analysis. This study is also prone to survival bias as some participants may have discontinued non-INSTI drugs prior to the time of analysis because of neuropsychiatric adverse events. Also as the non-INSTI group had been taking their current regimen much longer than the INSTI group, the non-INSTI group may be selected for participants who tolerate NNRTI or protease inhibitor therapy. Finally, data on the participant's functional or menopausal status were not available [48].
In conclusion, in this large cohort of PLWH, the use of INSTI-based regimens were associated with modestly reduced neuropsychological performance in the learning and memory domain as well as lower brain volumes when compared with their counterparts on non-INSTI-based regimens. These findings emphasize the importance of clinical monitoring of PLWH being switched to INSTI-based regimens. Longitudinal studies with preplanned hypotheses are needed to confirm the observed findings and explore potential underlying mechanisms.
|
|
| |
| |
|
|
|
