Background
Neurobehavioral disorders and gut dysbiosis continue to occur in treated PWH, but the relationship between the plasma microbiome and NBPs has not been investigated. Plasma microbiome analysis offers unique insights into systemic microbial translocation and immune responses. This study aimed to investigate the association between plasma microbiome composition and neurobehavioral phenotypes in PWH.
Methods
410 PWH who were taking antiretroviral therapy and had plasma HIV RNA ≤ 200 copies/mL underwent comprehensive neuromedical and neurobehavioral assessments and provided plasma for 16S rRNA amplicon sequencing. As previously described, 17 neurobehavioral features, including cognitive domains, depression subscales, and daily functioning, defined 4 NBPs using dimension reduction with a self-organizing maps (SOM) followed by k-means clustering methods: NBP1: healthy (n=187), NBP2: depression-dominant (n=94), NBP3: unhealthy in all features (n=55), and NBP4: cognitive impairment-dominant (n=74). Alpha and beta diversity were measured using diversity metrics. Differential species abundance was analyzed using Bayesian Inferential Regression for Differential Microbiome Analysis (BIRDMAn), adjusting for age and gender.
Results
Cohort characteristics: mean age 42.7 ± 8.9 years, 16.3% female, 44.9% Black, 59.3% AIDS, mean CD4+ T-cells 467.5/µL. Alpha diversity was lower in NBP2, than in either NBP4 or NBP1 (each p < 0.05). Beta diversity also differed between NBP2 and NBP4 using weighted (p = 0.03) and unweighted UniFrac metrics (p = 0.01). Genus Actinomycetospora and Mycobacterium, Flavobacteriaceae family, and species mucilaginosa were more abundant in people with NBP4 than NBP2, potentially contributing to the observed clinical differences (Figure 1).
Conclusions
These findings support a role for the plasma microbiome in modulating brain health in PWH. This may occur in part by modulating immune responses: Mycobacteria can induce proinflammatory immune responses and Flavobacteriaceae produce anti-inflammatory short-chain fatty acids. These results are provocative, but the cross-sectional design limits causal inference. Understanding these complex interactions support development of personalized, microbiome-based interventions to improve brain health in PWH.
Table or Figure
