2026 ECNP Congress

PS02-2043

Divergent immunometabolic signatures across adult males and females with major depressive disorder revealed by multi-omics integration

Siyuan Li — T. Xu, D. Zeng, H. Xu, X. Zhao, Y. Shi, L. Zhang, L. Zhang, Y. Zheng, J. Shi, S. He, H. Li · Shanghai Mental Health Center- Shanghai Jiao Tong University School of Medicine, Department of Psychiatry, Shanghai, China

2026 ECNP Congress

Abstract Background

Major depressive disorder (MDD) shows a pronounced sex disparity in prevalence, with women having a 1.6-fold higher risk than men [1]. Accumulating evidence suggests that sex shapes the biological architecture of depression, including immune activation, metabolic regulation, and stress responsiveness [2,3]. However, most biomarker studies in MDD have not distinguished shared from sex-specific alterations, potentially obscuring disease heterogeneity. Integrating high-dimensional immune-cell phenotyping with cytokine and metabolomic profiling may therefore provide a more comprehensive approach to resolving sex-specific immunometabolic signatures and informing biologically grounded stratification in MDD.

Methods

We profiled plasma targeted multi-omics data, including mass cytometry by time-of-flight (CyTOF), cytokines, and metabolomics, in 83 patients and 49 healthy controls using a sex-stratified analytical approach. We applied differential, pathway enrichment, rank-rank hypergeometric overlap (RRHO), and cross-omics integration based on Data Integration Analysis for Biomarker Discovery using Latent cOmponents (DIABLO) analyses to characterise peripheral immunometabolic alterations in MDD. As a secondary analysis, machine-learning models were trained and evaluated using leave-one-out cross-validation (LOOCV), bootstrap optimism correction, decision curve analysis (DCA), and SHapley Additive exPlanations (SHAP). All statistical analyses were conducted in R.

Results

Differential analyses revealed opposite sex-specific directional patterns across omics layers, characterised by predominant metabolite upregulation in women (148 upregulated vs 9 downregulated features; men: 5 vs 5) and predominant cytokine downregulation in men (116 downregulated vs 15 upregulated features; women: 23 vs 23). KEGG analysis revealed substantial cross-sex overlap in cytokine-related pathways (45 shared, 35 male-specific and 2 female-specific pathways [adjusted P < 0.05, count > 2]), but highly sex-specific pathway enrichment in metabolites. Lipid metabolic pathways were significantly altered only in women and accounted for most female-specific metabolic enrichment. Shared immune pathways were predominantly related to pro-inflammatory signalling networks, including cytokine-cytokine receptor interaction, IL-17, JAK-STAT, TNF, NF-κB, and chemokine signalling, with stronger enrichment in men. RRHO2 analyses indicated that metabolic dysregulation is strongly sex-dependent, whereas immune pathway perturbations—especially those involving immune-cell populations—are more broadly conserved across sexes. DIABLO analyses identified four cross-omics axes: a Treg–spermidine–CXCL14 axis and a hypoxanthine/lactate–cytokine–non-Treg immune-cell axis in women, and a Treg–HexCer–cytokine axis and a monocyte–amino acid–ULBP-1 axis in men. Sex-stratified models showed high diagnostic performance with compact feature panels (AUC = 0.983 with 6 features in men; AUC = 0.899 with 5 features in women). TREM1+ monocytes were the only shared predictive feature in the final diagnostic models for both sexes, with the highest importance observed in men.

Conclusions

These findings support a sex-stratified framework for understanding peripheral immunometabolic heterogeneity in MDD. Methodologically, this study establishes an integrative framework that combines plasma CyTOF, cytokine, and metabolomic profiling in the same individuals with both network-based and predictive analyses. Biologically, it suggests that MDD is not organised around a single shared peripheral immunometabolic architecture, but rather around sex-divergent programmes, with stronger metabolic specificity in women and more pronounced cytokine-centred immune perturbation in men. Translationally, TREM1+ monocytes emerged as a sex-conserved candidate predictor, while compact sex-specific feature panels showed strong exploratory diagnostic performance.

Keywords

Major depressive disorder; sex differences; multi-omics; immunometabolism; machine learning; metabolomics

References

  1. Li S, Zhang X, Cai Y, Zheng L, Pang H, Lou L: Sex difference in incidence of major depressive disorder: an analysis from the Global Burden of Disease Study 2019. Ann Gen Psychiatry 2023, 22:53.
  2. Wang Z, Xiang Y, Dang R, Wang P, Du X, Xie P: Sex-specific differences in peripheral blood metabolites and biological functions in major depressive disorder: A systematic review and meta-analysis. Neurosci Biobehav Rev 2025, 170:106052.
  3. Labaka A, Goñi-Balentziaga O, Lebeña A, Pérez-Tejada J: Biological Sex Differences in Depression: A Systematic Review. Biol Res Nurs 2018, 20:383-392.