Smaller limbic structures are associated with greater immunosuppression in over 1000 HIV-infected adults across five continents: Findings from the ENIGMA-HIV Working Group

Abstract Background Human immunodeficiency virus type-1 (HIV) infection can be controlled with combination antiretroviral therapy (cART), but neurocognitive impairment remains common even in chronic and treated HIV-infected (HIV+) cohorts. Identifying the neuroanatomical pathways associated with infection has the potential to delineate novel neuropathological processes underlying persisting deficits, yet individual neuroimaging studies have yielded inconsistent findings. The ENIGMA-HIV Working Group was established to harmonize data from diverse studies to identify the common effects of HIV-infection on brain structure. Methods Data were pooled from 12 independent neuroHIV studies from Africa, Asia, Australia, Europe, and North America. Volume estimates for eight subcortical brain regions were extracted from T1-weighted MRI from 1,044 HIV+ adults (aged 22-81 years; 72.4% on cART; 70.3% male; 41.6% with detectable viral load (dVL)), to identify associations with plasma markers reflecting current immunosuppression (CD4+ T-cell count) or dVL. Follow-up analyses stratified data by cART status and sex. Bonferroni correction was used to determine statistical significance. Findings Lower current CD4+ count was associated with smaller hippocampal ( β = 20.3 mm 3 per 100 cells/mm 3 ; p = 0.0001) and thalamic volumes ( β = 29.3; p = 0.003); in the subset of participants not on cART, it was associated with smaller putamen volumes ( β = 65.1; p = 0.0009). On average, a dVL was associated with smaller hippocampal (Cohen’s d = 0.24; p = 0.0003) and amygdala volumes ( d = 0.18; p = 0.0058). Interpretation In HIV+ individuals across five continents, smaller limbic volumes were consistently associated with current plasma markers. As we assessed cohorts with different inclusion/exclusion criteria and demographic distributions, these deficits may represent a generalizable brain-signature of HIV infection in the cART era. Our findings support the importance of achieving viral suppression and immune restoration for maintaining brain health. Funding This work was supported, in part, by NIH grant U54 EB020403. Research in Context Evidence before this study HIV type-1 infection can be managed with antiretroviral therapy, however neurocognitive impairment persists even in treated HIV+ individuals. Given the challenges associated with standardized cognitive testing, there is a need to identify quantitative markers of central nervous system impairment. A number of neuroimaging studies have reported brain abnormalities in HIV-infected patients; however, prior studies investigating associations between CD4+ T-cell count or HIV viral load and subcortical brain volume report variable effect sizes and regional distributions of effects, limiting the generalizability of the conclusions drawn to date. We have conducted a literature search for reports in English language journals published until June 2019, using the following search terms: HIV AND subcortical AND neuroimaging AND brain AND viral load AND RNA AND CD4. After removing studies that were not applicable, there were 30 studies investigating CD4+ T-cell count and viral load associations with subcortical brain structure. Added value of the study The aim of the current study was to investigate structural brain associations with two biomarkers universally used to monitor immune function and treatment response, namely plasma RNA viral load and CD4+ T-cell counts. Prior analyses have been performed in smaller, heterogeneous cohorts, but by combining data across cohorts, we can identify consistent associations between brain volume and indicators of HIV infection across cohorts. The ENIGMA-HIV Working Group was established to identify common neurobiological signatures of the HIV-infected brain by harmonizing data analysis from HIV neuroimaging studies worldwide. The value of this dataset is that it is well-powered and representative of many HIV+ people living in the cART era. Implications of all the available evidence Our results provide robust evidence that despite demographic and clinical heterogeneity among HIV-infected individuals, brain abnormalities are consistently linked to HIV viral load and immunosuppression. This supports the importance of achieving viral suppression and immune system restoration in maintaining brain health in people living with HIV. The vulnerability of limbic regions, found in this study, extends beyond the classically implicated regions of the basal ganglia; this suggests that these regions remain an important target of cART era HIV research, especially given their heightened vulnerability to age-associated atrophy and neurodegeneration.