Depression is a significant comorbidity in people living with HIV (PLWH), with prevalence estimates twice that of the general population. While the exact mechanisms are unclear, the association between depression and inflammation in the general population, coupled with the inflammatory nature of HIV infection, suggests a potential link. This study hypothesized that inflammatory biomarkers would mediate the relationship between HIV status and depressive symptoms. Depression is increasingly viewed not as a monolithic condition, but as encompassing multiple subtypes, with one subtype strongly linked to peripheral and central nervous system (CNS) inflammation. Shared symptoms like fatigue and anhedonia exist across depressive and inflammatory conditions. Elevated pro-inflammatory cytokines are found in depressed individuals, and anti-inflammatory treatments can alleviate depressive symptoms. In PLWH, the brain serves as a reservoir for HIV, leading to neuroinflammation even with viral suppression. Thus, HIV-induced inflammation may contribute to CNS pathology and depression. This study aimed to clarify the contribution of inflammatory biomarkers to the HIV-depression link using a well-characterized cohort.

Existing literature extensively documents the higher prevalence of depression among PLWH compared to the general population. Studies highlight the association between HIV infection and an increased likelihood of depression, potentially exacerbated by pre-existing depression. In the general population, research emphasizes the role of inflammation, both peripheral and central, in a subtype of depression characterized by sickness behaviors. Studies have shown elevated pro-inflammatory cytokines in depressed individuals, and the effectiveness of anti-inflammatory treatments in alleviating depressive symptoms support this link. However, the precise nature of the relationship between neuroinflammation and depressive symptoms, especially in the context of HIV, requires further investigation.

This study used data from the Comorbidities in Adults Living with HIV (COBRA) cohort, a prospective study of adults living with HIV in London and Amsterdam. Participants (aged 45+) were virally suppressed and on ART for at least 12 months. A demographically-similar HIV-negative control group was included. Depressive symptoms were assessed using the PHQ-9 questionnaire. Plasma and cerebrospinal fluid (CSF) samples were collected for biomarker analysis, including cytokines (IL-6, TNF-α), chemokines (MIG/CXCL9, IP-10/CXCL10, MCP-1/CCL2, IL-8/CXCL8, RANTES/CCL5, MIP-1α/CCL3), and other inflammatory markers (CRP, sCD14, sCD16, sCD163, neopterin, NFL, kynurenine:tryptophan ratio). Brain MR spectroscopy (MRS) was used to measure myo-inositol and choline in frontal white matter and prefrontal cortex. Logistic regression was used to analyze the association between HIV status and depressive symptoms (PHQ-9 score ≥ 10), adjusting for sociodemographic factors. Sequential adjustment for each biomarker was performed to assess mediation, with a >10% reduction in odds ratio (OR) indicating potential mediation. Linear regression was used in sensitivity analyses with PHQ-9 score as a continuous variable.

A total of 204 participants (125 HIV-positive, 79 HIV-negative) were included. HIV-positive participants had significantly higher odds of experiencing any depressive symptoms (OR 3.72, 95% CI 1.76–8.09). After adjusting for sociodemographic factors, plasma MIG (−15.0%) and TNF-α (−11.4%), and CSF MIP-1α (−21.0%) and IL-6 (−18.0%) showed potential mediation of the association between HIV and depressive symptoms. A >10% reduction in OR was observed for these biomarkers after adjustment. Other biomarkers, including neuroimaging markers, did not meet this criterion. Sensitivity analyses using PHQ-9 as a continuous outcome yielded consistent results. Moderate correlations were observed between plasma and CSF IL-6 concentrations, and between IL-6 and several chemokines.

This study provides evidence suggesting that peripheral and central inflammation, as indicated by specific biomarkers, may partially explain the increased risk of depressive symptoms in PLWH. The significant mediation effect of MIG, TNF-α, MIP-1α, and IL-6 highlights the potential role of inflammatory processes in the pathogenesis of HIV-associated depression. The finding regarding IL-6 aligns with prior research linking this cytokine to depression risk. The notable mediating role of CSF MIP-1α is a novel finding, suggesting it as a potentially valuable, yet understudied, biomarker for depression, particularly in PLWH. The lack of mediation by neuroimaging biomarkers warrants further investigation with more sensitive techniques. These results could inform the development of targeted interventions focusing on inflammation management in the context of mental health care for PLWH.

This study demonstrates that biomarkers of central and peripheral inflammation, specifically MIG, TNF-α, MIP-1α, and IL-6, are potential mediators of the association between HIV status and depressive symptoms. Future studies should validate these findings in larger, more diverse cohorts and explore the underlying mechanisms connecting inflammation, neuroendocrine processes, and depression in PLWH. Further research is needed to explore potential interaction effects among these biomarkers and their relationship with various types of adversity.

The cross-sectional design limits causal inference. The sample size, particularly for subgroups, might have limited statistical power. The exclusion of participants with severe depressive symptoms at baseline might affect generalizability. The cohort was primarily composed of White men who have sex with men, restricting the generalizability of the findings to other populations. Several relevant inflammatory markers were not measured, and some biomarker concentrations were near detection limits for many participants. Potential confounding effects from antiretroviral therapy side effects and selection bias related to study participation need consideration.