Urinary peptide signature NDN99 non-invasively distinguishes nodular diabetic nephropathy, predicts kidney events.
Background
Diabetic kidney disease (DKD), particularly Diabetic Nephropathy (DN), poses significant challenges for risk stratification, often progressing to end-stage renal disease. Nodular Diabetic Nephropathy (NDN) represents the most specific histological feature of DN and is strongly associated with a poor prognosis. Current diagnosis of NDN relies on invasive kidney biopsy, a procedure with inherent risks and limitations. There is a critical need for non-invasive biomarkers to identify NDN early, enabling more precise risk stratification and timely intervention to improve patient outcomes.
Study Design
This case-control study analyzed urinary peptide profiles from 81 biopsy-proven Diabetic Nephropathy patients (43 NDN, 38 nNDN) using capillary electrophoresis-mass spectrometry. To enhance statistical power and generalizability, real patient data were combined with synthetic datasets generated via a Gaussian copula approach, then divided into discovery and validation cohorts. Differential peptides were identified using Wilcoxon testing with Benjamini-Hochberg correction, and a support vector machine model was developed to create the NDN99 classifier.
Results
In the discovery cohort, 207 urinary peptides were found to be significantly altered between NDN and nNDN groups. NDN was specifically characterized by a reduction in collagen fragments and an increase in fibrinogen, apolipoprotein A-IV, and α1-antitrypsin peptides. The developed NDN99 classifier demonstrated robust performance, discriminating NDN from nNDN with an AUC of 0.87 in the combined validation cohort and 0.79 in the real-patient validation cohort. Notably, the NDN99 classifier also predicted major adverse kidney events (MAKE; defined as death, kidney failure, or 40% eGFR loss), and significantly outperformed traditional histology in this predictive capacity.
NDN99 predicted MAKE with a hazard ratio (HR) of 2.9 (
p = 0.0006), compared to histology's HR = 1.9 (p = 0.04).
Key Findings
- 207 urinary peptides were significantly altered in nodular diabetic nephropathy (NDN).
- NDN showed reduced
collagen fragmentsand increasedfibrinogen,apolipoprotein A-IV, andα1-antitrypsinpeptides. - The NDN99 classifier discriminated NDN with an AUC of 0.87 (combined validation) and 0.79 (real-patient validation).
- NDN99 predicted major adverse kidney events (MAKE) with an HR = 2.9 (
p = 0.0006). - NDN99 outperformed histology in predicting MAKE (HR = 2.9 vs HR = 1.9).
Why It Matters
This research introduces a promising non-invasive diagnostic tool for Nodular Diabetic Nephropathy (NDN), potentially reducing the need for invasive kidney biopsies. For clinicians, the NDN99 urinary peptide signature could enable earlier and more accurate identification of NDN, facilitating personalized treatment strategies and improved risk stratification for patients with Diabetic Nephropathy. This could lead to earlier interventions to slow disease progression and prevent major adverse kidney events. While still in the research phase, this approach offers a pathway toward a more accessible and safer diagnostic protocol, complementing or even eventually replacing current invasive methods for subtype identification and prognosis.
diabetic-nephropathy
nodular-diabetic-nephropathy
kidney-disease
biomarkers
urinary-peptides
diagnosis