Cortical White and Grey Matter Volume Differences Associated with Plasma Cytokine and Chemokine Levels in PLWH in Cape Town

HIV infection is accompanied by production of pro-inflammatory cytokines, which are regarded as critical in neuronal damage, leading to brain dysfunction. To develop diagnostic tools and therapeutic interventions, we need to measure CNS response to immune activation, hence the need to identify specific cytokine biomarkers that are associated with brain damage in HIV infection. This cross-sectional retrospective study applied Magnetic Resonance Imaging (MRI) for brain volumetric measurements and high-throughput Luminex-based immunoassays to quantify plasma cytokine and chemokine concentrations. We then used generalized linear models and Partial Least Square Regression models to evaluate the association between brain volume and plasma cytokines in predominantly treatment-naïve participants with HIV. After adjusting for clinical and demographic variables, we observed that higher MCP-1 (<i>p</i> = 0.013) and RANTES (<i>p</i> = 0.002) remained significantly associated with lower cortical white matter volume, whereas the anti-inflammatory cytokine IL-9 (<i>p</i> = 0.025) and the growth factors PDGFBB (<i>p</i> = 0.012) and VEGF (<i>p</i> = 0.001) were associated with higher cortical white matter volume. Only IL-6 (<i>p</i> = 0.010) was significantly associated with lower subcortical grey matter volume. Higher concentrations of five pro-inflammatory cytokines, IL-6 (<i>p</i> = 0.0001), IL-8 (<i>p</i> = 0.018), GCSF (<i>p</i> = 0.004), MCP-1 (<i>p</i> = 0.004), and RANTES (<i>p</i> = 0.015), were associated with lower total grey matter volume. Associations of pro-inflammatory cytokines with lower brain volume could imply a link to mechanisms of HIV-associated brain damage, which may lead to neurocognitive impairment. Therefore, the use of highly sensitive neuroimaging and high-throughput immunoassays in HIV-associated brain disorders has potential applications in clinical assessments and therapeutic monitoring.