Identification of Metabolic Pathway Dysregulation in Diabetic Retinopathy: A Propensity Score-Matched Metabolomic Study

Abstract Background Diabetic retinopathy (DR) is a major diabetes-related disease linked to metabolism. However, scientifically assessment of serum metabolic alterations in DR is scarce. We aimed to investigate the changes in metabolic coregulation from type 2 diabetic patients (T2DM) to DR and identify corresponding metabolite predictors via a widely targeted metabolomics approach.Methods In this case-control study, we tested 613 serum metabolites in 69 pairs of T2DM with DR (case) and propensity score-matched T2DM without DR (control) utilizing the ultra-performance liquid chromatography-electrospray ionization-tandem mass spectrometry system. The discrimination capability of differentially expressed metabolites (DEMs) in DR identification was also evaluated using a least absolute shrinkage and selection operator (LASSO) regression-based linear support vector machine (SVM) classifier. Metabolic pathway dysregulation in DR were comprehensively investigated by metabolic pathway analysis, chemical similarity enrichment analysis and MetaMapp approaches.Results A total of 89 DEMs were identified after paired univariate analysis and partial least squares discriminant analysis. The linear-SVM model based on LASSO regression selected DEMs had an excellent discrimination with an area under the ROC curve (AUC) as 0.99 (95% confidence interval: 0.95, 1.00). The biosynthesis of polyunsaturated fatty acids (PUFAs), thiamine metabolism, amino acids (mainly glycine, serine and threonine metabolism), hydroxyeicosatetraenoic acid (HETE), disaccharides, indoles and nucleotides were significantly enriched in DR. Conclusions This study systematically demonstrates that distinct metabolic alterations are linked to DR initiation. n-3 PUFAs, trehalose and vitamin B1 play an important role in inhibiting DR progression.