Large-Scale Proteomics Improve Prediction of Chronic Kidney Disease in People With Diabetes
Authors' conclusion
Adding the CKD protein risk score to a validated clinical risk model significantly improved the discrimination and reclassification of CKD risk in patients with diabetes.
Study context
- Population
- Patients with diabetes with proteomics and genetic information and no history of chronic kidney disease (CKD) at baseline, UK Biobank Pharma Proteomics Project1,047 test set
- Participants
- 2,094 patients
- Comparison
- CKD protein risk score vs CKD-PC clinical risk model and CKD polygenic risk score
- Outcome
- Incident CKD
- Follow-up
- Median 12.1 years
Methodological detailstraining and test sets, proteomics (nearly 3,000 plasma proteins), Training set 1,047 participants, 117 CKD events; test set 1,047 participants, 112 CKD events
- Training and test sets
- Proteomics (nearly 3,000 plasma proteins)
- Training set 1,047 participants, 117 CKD events; test set 1,047 participants, 112 CKD events
Performance
As reported in the abstract
Incident CKD prediction (test set)
Basic model
0.627
95% CI 0.578–0.675
Hazard ratio per SD increment (95% CI)
CKD protein risk scorevs Basic model
1.78(1.44–2.20)
Estimate (95% CI)
CKD protein risk scorevs Basic modelsignificant improvement
C-index increase 0.122(0.071–0.177)
CKD-PCvs Basic modelsignificant improvement
C-index increase 0.175(0.126–0.217)
CKD polygenic risk scorevs Basic modelno significant improvement
C-index increase 0.007(−0.016 to 0.025)
CKD-PC + CKD protein risk scorevs CKD-PCC-index from 0.802 to 0.825
C-index difference 0.023(0.006–0.044)
- 10-year continuous net reclassification, CKD-PC + protein score vs CKD-PC
- 0.199 (95% CI 0.059, 0.299)
- 10-year integrated discrimination index, CKD-PC + protein score vs CKD-PC
- 0.041 (95% CI 0.007, 0.083)
Original publication
Large-Scale Proteomics Improve Prediction of Chronic Kidney Disease in People With Diabetes. Diabetes Care. Published 23 July 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →