Abstract: SA-PO293
Improving the Histological Classification of Diabetic Nephropathy Based on Transformative Research in Diabetic Nephropathy (TRIDENT)
Session Information
- Diabetic Kidney Disease: Clinical Pathology, Diagnostic and Treatment Advances
October 26, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Diabetic Kidney Disease
- 702 Diabetic Kidney Disease: Clinical
Authors
- Mohandes, Samer, University of Pennsylvania, Philadelphia, Pennsylvania, United States
- Mottl, Amy K., The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States
- Scialla, Julia J., University of Virginia, Charlottesville, Virginia, United States
- Almaani, Salem, The Ohio State University Wexner Medical Center, Columbus, Ohio, United States
- Isakova, Tamara, Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States
- Bansal, Shweta, The University of Texas at San Antonio, San Antonio, Texas, United States
- Argyropoulos, Christos, University of New Mexico Health Sciences Center, Albuquerque, New Mexico, United States
- Coppock, Gaia M., University of Pennsylvania, Philadelphia, Pennsylvania, United States
- Jennette, J. Charles, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States
- Susztak, Katalin, University of Pennsylvania, Philadelphia, Pennsylvania, United States
- Palmer, Matthew, University of Pennsylvania, Philadelphia, Pennsylvania, United States
Background
Diabetic Nephropathy is the most common cause of kidney disease in the world. The diagnostic and prognostic utility of the current Renal Pathology Society Classification (RPS) system is not established. We aim to evaluate RPS system and compare it with our proposed Transformative Research in Diabetic Nephropathy (TRIDENT) histopathologic classification.
Methods
TRIDENT is a multicenter prospective observational study of subjects with diabetes mellitus undergoing a kidney biopsy. Unsupervised k-means clustering of 29 light microscopy features divided the cohort into 5 clusters. The association of this TRIDENT classification with kidney outcomes, defined as 40% change in eGFR or ESRD over 4 years of follow-up, was assessed by Cox proportional hazard models with Harrell’s C-statistic for predictive accuracy. Decision curve analysis was used to compare the potential clinical utility of the TRIDENT and RPS classifications across risk thresholds.
Results
The RPS class showed some association with kidney outcomes, but not between classes 3&4 (65% of the cohort). The TRIDENT classification (table 1) grouped the cohort into risk groups with better predictive accuracy of renal risk at 1 year (AUC 0.75 vs 0.66 ) and 2 years (AUC 0.8 vs 0.74) compared to RPS. Class 5, which includes epithelial features, had the most rapid progression. Addition of the TRIDENT classification to the 2 year kidney failure risk equation resulted in improved discrimination (C statistic 0.759-0.766, ROC AUC at 2 years 0.803-0.837). A higher net benefit in the TRIDENT classification compared to RPS was seen using a decision curve analysis.
Conclusion
We developed a new TRIDENT classification system that improves on currently used methods to capture the severity of DN and predicts kidney outcomes. This classification identifies a subgroup with significant epithelial changes and a more severe kidney outcome.
TRIDENT Classification System
| TRIDENT Class | Description |
| 1 | GBM thickness without mesangial expansion |
| 2 | Mesangial matrix expansion without KW nodules |
| 3 | Mesangial matrix expansion and KW nodules without mesangiolysis |
| 4 | Mesangiolysis without epithelial hyperplasia |
| 5 | Mesangiolysis with epithelial hyperplasia |
Funding
- Commercial Support – TRIDENT consortium