Abstract: FR-PO0534
Plasma Proteogenomics Identifies Glucose-Independent Residual Risk Drivers in Diabetic Kidney Disease
Session Information
- CKM: Clinical - Trials, Epidemiology, and Biomarkers
October 23, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Cardiovascular-Kidney-Metabolic Health
- 602 Cardiovascular-Kidney-Metabolic Health: Clinical
Authors
- Chen, Lei, West China Hospital of Sichuan University, Chengdu, Sichuan, China
- Tang, Wanxin, West China Hospital of Sichuan University, Chengdu, Sichuan, China
Background
Diabetic kidney disease (DKD) can progress despite adequate glycemic control, suggesting residual kidney risk beyond glucose toxicity. We aimed to identify plasma proteins associated with glucose-independent DKD risk and prioritize biologically plausible drivers using proteomics and genetics.
Methods
We analyzed UK Biobank Olink proteomics data from 2,890 participants with type 2 diabetes (T2D), 4,016 healthy controls, and 2,923 plasma proteins. Among T2D participants, albuminuric DKD was defined by urine albumin-to-creatinine ratio (UACR) ≥30 mg/g and classified as microalbuminuric or macroalbuminuric DKD; T2D controls had UACR <30 mg/g and preserved estimated glomerular filtration rate (eGFR). Multivariable regression identified DKD-associated proteins. Metabolic decoupling prioritized kidney-linked proteins weakly coupled to HbA1c and metabolic sensor proteins. Prioritized proteins were evaluated by genetic prioritization, kidney single-cell data, DKD severity trends, decision-curve analysis, prospective outcomes, and PheWAS safety.
Results
Among 127 DKD-associated proteins, 63 showed concordant associations with lower eGFR and higher UACR. Metabolic decoupling retained 22 kidney-linked proteins, from which BTN2A1, TNFRSF9, TNFRSF11A, and EFEMP1 emerged as four glucose-independent residual-risk candidates. These proteins increased stepwise from T2D controls to microalbuminuric and macroalbuminuric DKD and mapped to immune activation and matrix-remodeling pathways. BTN2A1 showed the strongest colocalization support (PP4 = 98.8%). In participants with HbA1c <7%, the four-protein signature discriminated residual-risk DKD (AUC = 0.70) and was associated with CKD progression, cardiovascular disease, and all-cause mortality.
Conclusion
Plasma proteogenomics prioritized four glucose-independent DKD risk proteins linking immune activation and matrix remodeling. These findings support risk stratification beyond HbA1c and highlight non-glycemic pathways for kidney protection.