Kidney Week

Abstract: FR-PO0239

Urinary Proteomic Signatures and Prognostic Markers in the Progression of Diabetic Kidney Disease

Session Information

• CKD: Mechanisms of Injury and Fibrosis - 2
  October 23, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: CKD (Non-Dialysis)

• 2203 CKD (Non-Dialysis): Mechanisms

Authors

• Wang, Hao, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, China

Hao Wang, DrPH

• Sun, Wei, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, China

Wei Sun, PhD

• Li, Mingxi, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, China

Mingxi Li, PhD

• Liu, Yuhao, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, China

Yuhao Liu, MD, PhD

Background

Diabetic kidney disease (DKD) is a leading cause of end-stage renal disease (ESRD). Currently, robust non-invasive biomarkers for the early detection and prediction of DKD progression are lacking. This study aimed to characterize the dynamic changes in the urinary proteome during DKD progression and to identify novel protein biomarkers for early diagnosis and the prediction of renal function decline.

Methods

This study was conducted in two phases. Initially, a cross-sectional proteomic analysis was performed on urine samples from 72 participants: 23 with type 2 diabetes mellitus (DM), 27 with early DKD (24-h urine protein [24hUP] < 0.5 g or albumin-to-creatinine ratio [ACR] < 300 mg/g, estimated glomerular filtration rate [eGFR] > 80 mL/min/1.73 m²), and 22 with advanced DKD (24hUP > 3 g, eGFR < 45 mL/min/1.73 m²). Urinary proteins were enriched using nanomagnetic beads and analyzed via data-independent acquisition (DIA) mass spectrometry. Differentially expressed proteins (DEPs) were identified, and diagnostic models were constructed using LASSO regression. Subsequently, a longitudinal cohort of 31 patients with DKD was assessed over a 36-month follow-up period to evaluate the predictive value of candidate proteins for eGFR decline, utilizing both univariate and multivariate linear regression analyses.

Results

A total of 582 DEPs were identified between the early DKD and DM groups, and 272 between the advanced and early DKD groups. LASSO-based diagnostic models demonstrated excellent performance, achieving an area under the curve (AUC) of 0.934 for early DKD detection (incorporating BLMH, ABHD14A, CFD, HEXB, CFHR1, TAGLN, CPN1, TPP1, SAA4, PFN2) and an AUC of 0.919 for advanced DKD identification (incorporating C6, CHGA, IGFBP1, AGT, MYOC, CFD). In the longitudinal cohort, after adjusting for age and baseline eGFR, baseline urinary levels of HEXB (beta = -0.770,P < 0.001) and CHGA (beta = -0.279, P = 0.046) were identified as independent predictors of accelerated renal function decline, demonstrating a significant inverse association with eGFR at 36 months.

Conclusion

This study delineates the dynamic urinary proteomic landscape characterizing the progression of DKD from the early to advanced stages. The established diagnostic and prognostic models hold promise for personalized risk stratification, non-invasive monitoring, and early intervention in patients with DKD.