Impact of non-pharmacological interventions on serum metabolites and metabolic pathways of women living with HIV in reproductive age with mood disorders

To investigate the effects of cognitive behavioral therapy (CBT)-based multi-dimensional non-pharmacological intervention on serum metabolomic profiles and metabolic pathways among reproductive-age women living with HIV complicated by mood disorders, a total of 30 eligible participants were enrolled to receive a 3-month “psycho-behavior-social” integrated intervention. Seven participants withdrew during follow-up, leaving 23 subjects with paired pre- and post-intervention serum samples for final analysis. Baseline demographic, HIV-related and psychiatric indicators were compared between the 30 baseline enrollees and the 7 dropouts to assess whether the latter were representative of the initial cohort; no statistically significant intergroup differences were detected for all evaluated variables (all P > 0.05), indicating minimal attrition bias despite the loss to follow-up. Serum untargeted metabolomics based on UHPLC-QE-MS was performed to quantify metabolite abundances before and after intervention, and KEGG enrichment analysis was conducted on differential metabolites to screen mood disorder-related metabolic pathways responsive to intervention. A total of 5,517 metabolites were detected across all serum samples. Using the screening thresholds of variable importance in projection (VIP) > 1, 668 differential metabolites were identified, including 304 upregulated and 364 downregulated metabolites. After intervention, serum levels of mood-protective metabolites including hypoxanthine, uric acid, kynurenic acid, N-valine, tryptophan and proline were significantly elevated, whereas neurotoxic metabolites linked to aggravated mood dysfunction such as quinolinic acid, 5-hydroxylysine, alpha-tocotrienol and hypocholic acid were markedly decreased. KEGG pathway enrichment revealed 13 metabolic pathways closely correlated with mood disorders, among which tryptophan metabolism and purine metabolism exhibited the most prominent alterations. Observed metabolic shifts indicated that this integrated intervention remodeled tryptophan and purine metabolism, elevated circulating kynurenic acid and uric acid, and reduced quinolinic acid; these peripheral metabolite changes indirectly suggested potential modulation of neuroinflammation and oxidative stress, despite the absence of central neuroimmune biomarker measurements for mechanistic validation. Our preliminary results link the “psycho‑behavior‑social” non‑pharmacological intervention to improved circulating metabolite and mood‑related outcomes among reproductive‑age women living with HIV with mood disorders. Given limitations of small sample size and no control arm, future controlled investigations with larger cohorts and prolonged intervention‑follow‑up are needed to confirm these findings.