Investigation of Metabolic Dysregulation in Advanced-Stage Chronic Kidney Disease Using NMR-Based Urinary Metabolomics.

Abstract

Clinical assessment of kidney function is an integral part of routine medical care. More than 80% of clinical laboratories report an estimated glomerular filtration rate (eGFR) with endogenous markers such as serum creatinine. However, these markers are influenced by several non-GFR determinants. To address this limitation, we employed a ¹H nuclear magnetic resonance (NMR) based urinary metabolomics approach to investigate metabolic dysregulation in advanced-stage CKD (ASCKD). Urine samples from stage 4 CKD (S4, n = 22), stage 5 CKD (S5, n = 36), and healthy controls (HC, n = 22) were analyzed. Sixty-one metabolites were identified and quantified through 1D-¹H NMR. Statistical analyses showed clear separation in ASCKD and healthy controls (R² = 0.91, Q² = 0.79) and identified twenty-five significantly altered metabolites. Univariate receiver operating characteristic (ROC) analysis revealed a panel of fourteen metabolites distinguishing S4 from HC and twelve metabolites for distinguishing S5 from HC. Additionally, phenylalanine and mannitol demonstrated strong discriminatory performance (AUC > 0.80) in differentiating S4 from S5. Pathway enrichment and topology analyses revealed significant perturbations in glycolysis/gluconeogenesis, tricarboxylic acid (TCA) cycle, glyoxylate and dicarboxylate metabolism, pyruvate metabolism, and amino acid metabolism. These findings highlight significant disruption of energy and intermediary metabolism in ASCKD. Integration of urinary metabolomics with clinical markers may improve stratification and enable non-invasive monitoring of CKD severity.