知镜 Cognispect · 医学文献与研究工作台

Urine peptides enable non-invasive differential diagnosis of nodular diabetic nephropathy.

Kondyli M, Siwy J, Jaimes Campos MA, Amann K, Beige J, Keller F, Mischak H, Frantzi M, Wendt R, Rossing P, Rupprecht H, Catanese L, Jankowski V.
Diabetes research and clinical practice · 2026-06-26 · PMID 42361854

文献摘要

Aims Risk stratification in diabetic kidney disease is challenging. Identifying nodular diabetic nephropathy (NDN), the most specific histological feature of diabetic nephropathy (DN), associated with poor prognosis, currently requires kidney biopsy. We aimed to identify urinary peptide biomarkers distinguishing NDN from other chronic kidney disease etiologies in diabetes (nNDN). Methods In this case-control study, urinary peptide profiles from biopsy-proven DN patients (43 NDN, 38 nNDN) were analyzed using capillary electrophoresis-mass spectrometry. Real patient data were combined with synthetic datasets generated using a Gaussian copula approach and divided into discovery and validation cohorts. Differential peptides were identified by Wilcoxon testing with Benjamini-Hochberg correction, and combined into a support vector machine model. Results In the discovery cohort, 207 peptides were significantly altered. NDN was characterized by reduced collagen fragments and increased fibrinogen, apolipoprotein A-IV, and α1-antitrypsin peptides. The 99-peptide classifier (NDN99) discriminated NDN from nNDN with an AUC of 0.87 (combined validation) and 0.79 (real-patient validation). NDN99 predicted major adverse kidney events (MAKE; death, kidney failure, or 40% eGFR loss), and outperformed histology (HR = 2.9, p = 0.0006 vs HR = 1.9, p = 0.04). Conclusions Urinary peptide signatures enable non-invasive discrimination of NDN and predict MAKE, potentially complementing kidney biopsy for subtype identification and risk stratification in DN.

在知镜精读原文来源DOI 原文

知镜提供医学文献检索、阅读笔记、人体图谱、统计绘图与研究复现工具。