Inflammation as a contributor to depression in people living with HIV

Abstract

People with HIV are at significantly higher risk for depression, but possible neurobiological mechanisms underlying this risk remain unclear. Inflammation may play a role in the pathogenesis of a subtype of depression, since some people with depression have elevated concentrations of biomarkers of inflammation, while anti-inflammatory medications can alleviate symptoms of depression. Given that people with HIV exhibit chronic inflammation, I hypothesised that inflammation may contribute to the higher risk for depressive symptoms in this community.

This thesis makes several theoretical contributions to this area of research. Using a narrative literature review, I showed that HIV, (neuro)inflammation, and depression are independently linked, but few studies have investigated the relationship between all three conditions. Then, using a scoping review of 33 studies, I showed that significant associations have previously been reported between key inflammatory cytokines (IL-6 and TNF-α) and depressive symptoms in people with HIV. The majority (70%) of these studies were carried out in the US, whereas the majority of people with HIV are located in eastern and southern Africa. Thus, there is an urgent need to diversify participant samples in this field. I further explored the need for authentic equity and inclusion in this field by demonstrating that research on HIV and co-morbidities, including mental health issues, is mostly carried out in the Global North with non-diverse participant samples. Finally, as my work is situated within the broader context of HIV research, I carried out a research priority setting workshop with young adults with HIV and their caregivers, the outcomes of which may inform future research design and ensure that community priorities are acknowledged and centred in this field.

To test the central hypothesis of my research, I first investigated whether inflammatory biomarkers in the brain, blood plasma, and cerebrospinal fluid mediated the association between HIV status and depressive symptoms. In a sample of N = 204 older adults in the UK and the Netherlands, I showed that participants with HIV were at greater risk for depressive symptoms (odds ratio [95% confidence interval] 3.27 [1.46, 8.09]), and that controlling for the effects of four biomarkers resulted in a reduction of at least 10% in this odds ratio: MIG (-15.0%) and TNF-α (-11.4%) in plasma and MIP1-α (-21.0%) and IL-6 (-18.0%) in cerebrospinal fluid. Thus, these cytokines and chemokines were found to be potential mediators of the association between HIV status and depressive symptoms in people with HIV.

I then investigated this relationship using inflammatory biomarkers measured in the brain (using magnetic resonance spectroscopy) and blood serum (using immunoassays) in a demographically distinct sample of N = 60 young people in South Africa. I observed that young people with HIV showed greater depressive symptom severity (B = 3.32, p = 0.022) and altered concentrations of certain inflammatory biomarkers (higher choline in the basal ganglia, higher MCP-1 and higher YKL-40 in blood serum, and lower IL-1β in blood serum) compared to young people without HIV. However, none of these biomarkers mediated a significant proportion of the association between HIV status and depressive symptoms in this sample (all β < |0.09|, pFDR > 0.05).

Finally, I explored the relationship between the severity of depressive symptoms and the intracellular diffusion of neurometabolites, measured using diffusion-weighted magnetic resonance spectroscopy, in N = 20 virally suppressed adults with HIV in the UK, n = 11 of whom were living with severe and persistent depressive symptoms (PHQ-9 score ≥ 15). Depressive symptom severity in these participants was significantly correlated with ADC of creatine (ρ = 0.48, p = 0.034), though this significance did not survive correction for multiple comparisons (pFDR = 0.135). Creatine is a neurometabolite involved in energy metabolism, whose expression is higher in glial cells and shown to be upregulated concurrently with neuroinflammation in animal models. Recent studies have also shown increased creatine diffusion in patients with multiple sclerosis and neuropsychiatric lupus, both conditions characterised by neuroinflammation. Findings from this study thus suggest that severe depressive symptoms in people with HIV may be linked to hypermetabolism and neuroinflammation.

This research offers early, but promising, evidence that inflammation may contribute significantly to depressive symptoms in people with HIV. Substantial further work is necessary to confirm these findings in well-powered and diverse cohorts, as well as to determine the direction and temporality of these relationships through longitudinal studies. Further characterising the relationship between inflammation and depression in people with HIV may pave the way for individualised anti-inflammatory interventions to reduce the burden of depression in this community.