Serum kynurenine metabolites and cytokine levels: diagnostic and predictive implications in acute manic episodes of bipolar disorder

Abstract

In this study, we aimed to integrate serum cytokine and kynurenine metabolite levels to identify key biomarkers for diagnosing and predicting treatment outcomes in bipolar disorder during manic episodes (BDM). A total of 52 patients with BDM and 49 healthy controls (HCs) were recruited. Serum levels of cytokines and kynurenine metabolites were measured at baseline. Manic symptom severity was assessed using the Young Mania Rating Scale (YMRS) at baseline and after 8 weeks of treatment. Correlations between cytokines and kynurenine metabolites were analysed. Support Vector Machine (SVM) classifiers and Partial Least Squares (PLS) regression were employed to differentiate individuals with BDM from HCs and to predict treatment outcomes based on these profiles. Key findings included: (1) significant differences in kynurenine metabolites between individuals with BDM and HCs, whereas cytokine levels did not differ significantly; (2) weaker correlations between cytokines and kynurenine metabolites in individuals with BDM compared to HCs; (3) integrated models of kynurenine and cytokine systems achieved 89 % cross-validated accuracy in classifying individuals with BDM and HCs, outperforming models based on either system alone; and (4) interleukin (IL)-10, IL-8, kynurenine, IL-4, and tryptophan were key predictors of treatment outcomes, with IL-10 correlating significantly with 8-week treatment response. This study highlights the potential of integrating these systems to improve diagnostic accuracy and predict treatment outcomes in individuals with BDM, offering insights into novel therapeutic targets.