Key points are not available for this paper at this time.
Background Diabetic kidney disease (DKD), a major microvascular complication of diabetes mellitus, is the leading cause of chronic kidney disease and end-stage kidney disease worldwide. However, current clinical markers are limited in sensitivity for early detection and prognosis. Methods We recruited 96 patients with biopsy-confirmed diabetic nephropathy based on type 2 diabetes (T2DN), 76 patients with type 2 diabetes (T2D), and 79 healthy controls (HC). Untargeted urinary metabolomics was performed, followed by pathway and network analyses. Prognostic metabolites were identified using Cox regression and Kaplan–Meier survival analyses adjusted for confounders. Associations with renal histopathology were evaluated using partial Spearman correlation. For key prognostic metabolites, targeted quantification was performed in urine and plasma for validation. Results Among 148 detected metabolites, 101 significantly differed across groups, spanning multiple metabolic pathways. Amino acid metabolism, particularly branched-chain amino acid pathways, showed progressive changes along the HC–T2D–T2DN continuum. Urinary isoleucine emerged as the top prognostic metabolite (2.18 1.18, 4.03, P = 0.013), stratifying renal survival in T2DN (log-rank P 0.001) and correlating with four renal histological scores, which together capture the extent of renal structural damage and disease progression. Targeted quantification confirmed these findings and demonstrated that urinary, but not plasma, isoleucine was prognostically relevant. Conclusions Urinary isoleucine is a potential non-invasive biomarker of DKD progression, reflecting tubular injury and underlying metabolic dysregulation. These findings support urinary metabolomics as a tool for prognostic risk stratification, with the potential to enable future precision clinical management in DKD.
Li et al. (Mon,) studied this question.