Untargeted plasma metabolomics identifies metabolite signatures associated with suicidality status in chronic schizophrenia.

BACKGROUND AND HYPOTHESIS: Suicide contributes substantially to mortality in schizophrenia, yet metabolic signatures associated with suicidality remain poorly understood. We applied untargeted plasma metabolomics with multivariate modeling to identify plasma metabolic signatures associated with suicidality status and clinical features.

STUDY

DESIGN

We enrolled 146 frequency-matched psychiatric inpatients with chronic schizophrenia: 73 in the suicide-risk (SR) group, defined by current suicidal ideation or suicidal behavior/attempt history at admission, and 73 in the non-suicide-risk (NSR) group. Plasma samples underwent untargeted metabolomic profiling. Discriminating metabolites were identified using orthogonal partial least squares discriminant analysis (OPLS-DA) and Benjamini-Hochberg-adjusted Mann-Whitney U testing (variable importance in projection [VIP] >1.0 and adjusted p <0.05). Stepwise logistic regression with forced clinical covariates generated an exploratory metabolite-based classification model, and Spearman correlations examined associations with Positive and Negative Syndrome Scale (PANSS) and Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) scores.

STUDY

RESULTS

OPLS-DA identified 70 discriminating metabolites (R2Y = 0.938, Q2Y = 0.650), with nominal enrichment of lysine degradation, glutathione metabolism, phenylalanine/tyrosine/tryptophan biosynthesis, and sphingolipid metabolism. Stepwise regression retained six metabolites associated with SR status (area under the curve [AUC] = 0.774): Sphingosine, 3-OH-TML, Ile-Leu+Leu-Ile, TML, Taurine, and 12-HETE. Gamma-Glu-Thr showed the highest VIP (5.003) and broad clinical correlations.

CONCLUSIONS

Plasma metabolic differences involving sphingolipid, lysine/carnitine, neuroinflammatory, amino-acid, and glutathione-related pathways were associated with suicidality status in chronic schizophrenia. These exploratory cross-sectional findings require validation in larger longitudinal cohorts using dedicated suicidality assessments.