Metabolomic and transcriptomic features of the cardiometabolic disease spectrum in people living with HIV

Abstract Background: Human immunodeficiency virus (HIV) infection is associated with significant metabolic disruptions, and understanding how these disruptions contribute to cardiovascular disease (CVD) is essential for improving patient management and therapeutic strategies. By integrating metabolomics and transcriptomics, we aimed to elucidate the cardiometabolic disease spectrum of people living with HIV (PLWH) and to identify biomarkers and pathways associated with disease progression in this population. Methods: We conducted a cross-sectional observational study between October 2022 and June 2023 at the Third People’s Hospital of Shenzhen, China. Participants were categorized into one of four groups: HIV+ without metabolic abnormalities (healthy control [HC]), HIV + with untreated metabolic (UM) abnormalities (HIV + UM), HIV + with treated metabolic (TM) abnormalities (HIV + TM), and HIV + with CVD (HIV + CVD). Key metabolic and transcriptomic features were identified through a comparative analysis using untargeted metabolomics and RNA sequencing, followed by statistical and pathway enrichment analyses. Results: This study included 114 PLWH: HC group ( n = 30), HIV + UM group ( n = 19), HIV + TM group ( n = 46), and HIV + CVD group ( n = 19). Through comparative analyses of untargeted metabolomic and transcriptomic data across these four groups, we identified 580 dysmetabolic features (DMFs) based on significant changes between the HC group and the HIV + UM/TM groups, with no statistically significant differences in the HIV + UM/TM vs. HIV + CVD comparisons. Notably, 90.52% of these DMFs (525/580) exhibited reduced abundance in the HIV + CVD group. In contrast, 241 CVD-specific features were identified based on significant differences between the HIV + CVD group and the HIV + UM/TM groups. These features, primarily found in the HIV + CVD group, included metabolites and genes strongly associated with chronic inflammation and endothelial dysfunction—key processes in the pathogenesis of CVD among PLWH. Features that indicated a transition from metabolic abnormalities to CVD were termed escalation features (ESCFs). ESCFs included early biomarkers of lipid dysregulation, such as phosphatidylcholine (O-22:1/20:4), and upregulated genes, such as CAMP , both of which showed progressive changes from the HC group through the HIV + UM/TM groups to the HIV + CVD group. Features that indicated a reversal in expression trends—such as those increasing from the HC group to the HIV + UM/TM groups but decreasing in the HIV + CVD group, or vice versa—were termed de-escalation features (DSCFs). DSCFs included lipid species and genes involved in lipid metabolism and mitochondrial function. The integration of metabolomic and transcriptomic data revealed disruptions in pathways, such as glycerophospholipid and arachidonic acid metabolism, with concurrent upregulation of genes involved in cholesterol biosynthesis and the immune response. Receiver operating characteristic curve analysis based on the integration of metabolomic and transcriptomic data markedly improved CVD prediction among PLWH, with an area under the curve of 0.965, sensitivity of 84.2%, and specificity of 96.9%. Conclusions: This study identified key metabolic and transcriptomic alterations associated with CVD among PLWH. These alterations, primarily driven by immune activation, inflammation, and metabolic dysregulation, reflect unique molecular characteristics of CVD in PLWH. Our findings underscore the importance of early detection and tailored interventions to manage CVD risk in this population, providing insights into potential biomarkers for disease progression.