Abstract
Background: Elderly patients with COVID-19 are at increased risk of severe disease and mortality. Interactions between the plasma metabolome and the immune microenvironment may influence disease progression and outcomes; however, the associations of metabolic alterations with T-cell senescence, cytokine dysregulation, and clinical outcomes in this population remain poorly understood. Methods: Plasma metabolomic profiling was performed in elderly COVID-19 patients and healthy controls to identify differential metabolites and metabolic features associated with clinical outcomes. Machine learning approaches were applied to identify key metabolic features associated with 180-day mortality. Correlation analyses were performed to evaluate the associations of metabolic alterations with T-cell senescence and cytokine profiles. In vitro experiments were further performed to evaluate the effects of lysophosphatidylethanolamine (LPE) on T-cell senescence. Results: Plasma metabolomic profiles differed markedly between elderly COVID-19 patients and healthy controls. Distinct metabolic alterations were also observed between non-survivors and survivors, with phosphatidylethanolamine (PE) 36:3, glycochenodeoxycholic acid, and deoxycholic acid showing higher levels in non-survivors. Machine learning identified 24 key metabolites associated with 180-day mortality, primarily involving bile acid metabolism, fatty acid oxidation, and phospholipid metabolism. Dysregulated plasma phospholipid metabolism, particularly involving phosphatidylcholine (PC), lysophosphatidylcholine (LPC), phosphatidylethanolamine (PE), and lysophosphatidylethanolamine (LPE), was associated with T-cell senescence in elderly COVID-19 patients. Moreover, exogenous LPE 18:1 attenuated selected markers of T-cell senescence in vitro. Additionally, five cytokines—GRO-α, IL-6, IL-8, IP-10, and TSLP—showed higher levels in non-survivors. Metabolic alterations were associated with inflammatory cytokine profiles, particularly IP-10 and TSLP, while deoxycholic acid showed concurrent associations with IL-6, IP-10, and IL-8. Conclusions: Plasma metabolomic alterations were associated with T-cell senescence, inflammatory cytokine profiles, and clinical outcomes in elderly patients with COVID-19.