Abstract: PO2360
Urinary Peptidome Analysis to Predict the Risk of CKD Progression to Kidney Failure
Session Information
- Reassessing Race in Predicting Progression
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: CKD (Non-Dialysis)
- 2101 CKD (Non-Dialysis): Epidemiology, Risk Factors, and Prevention
Authors
- Massy, Ziad, Service de Néphrologie, Hôpital Universitaire Ambroise Paré, Assistance publique des Hôpitaux de Paris, Boulogne-Billancourt, France
- Lambert, Oriane, CESP, INSERM U1018, Equipe Epidémiologie Clinique, Université Paris-Saclay, UVSQ, Université Paris-Sud, Villejuif, France
- Metzger, Marie, CESP, INSERM U1018, Equipe Epidémiologie Clinique, Université Paris-Saclay, UVSQ, Université Paris-Sud, Villejuif, France
- Sedki, Mohammed, CESP, INSERM U1018, Pôle méthodologies et statistique, Université Paris-Saclay, UVSQ, Université Paris-Sud, Villejuif, France
- Glorieux, Griet Lrl, Department of Internal Medicine and Pediatrics, Nephrology Section, Ghent University Hospital, Ghent, Belgium
- Verbeke, Francis, Department of Internal Medicine and Pediatrics, Nephrology Section, Ghent University Hospital, Ghent, Belgium
- Mischak, Harald, Mosaiques Diagnostics GmbH, Hannover, Germany
- Siwy, Justyna, Mosaiques Diagnostics GmbH, Hannover, Germany
- Stengel, Benedicte, CESP, INSERM U1018, Equipe Epidémiologie Clinique, Université Paris-Saclay, UVSQ, Université Paris-Sud, Villejuif, France
- Schanstra, Joost, INSERM, U1297, Institut of Cardiovascular and Metabolic Disease, Toulouse, France
- Klein, Julie, INSERM, U1297, Institut of Cardiovascular and Metabolic Disease, Toulouse, France
Group or Team Name
- CKD-REIN Investigators
Background
Urinary peptidomics (UP) has emerged as one of the most attractive areas in the identification of biomarkers for characterizing CKD but its potential to predict the risk of CKD progression has not been fully investigated. The aims of the present study were to explore if a UP signature can improve the prediction of kidney failure (KF), compared to the risk factors (RF) included in the KF risk equation.
Methods
Within the Chronic Kidney Disease-Renal Epidemiology and Information Network prospective cohort of patients with CKD stage G3-G5 (N=3033), we conducted a case-cohort study of 892 patients, including 262 who progressed to KF as defined by the initiation of dialysis or preemptive kidney transplantation over 3-year follow-up. UP analysis was performed on samples collected at baseline using capillary electrophoresis coupled to mass spectrometry. Three logistic regression models with elastic-net penalty were developed with different sets of predictors: (1) peptides alone, (2) RF including age, sex, eGFR and urinary albumin to creatinine ratio, and (3) peptides and RF. We performed 50-repeated 2-fold cross-validation to choose the 3 optimal models and measure their performances (AUC, sensitivity and specificity). External independent validation was performed in a Belgian cohort of 270 patients with CKD Stages G3-G5 including 28 progressing to KF over 3-year follow-up.
Results
A signature of 174 peptides predicted KF risk in the first model (Figure). The independent validation of the UP signature in the Belgian cohort confirmed the prognostic potential of the peptide signature displaying a AUC of 0.928 [0.887-0.969] (sensitivity, 86% [71%-96%] and specificity, 81% [65%-94%]). RF alone also predicted KF risk with high precision, and the addition of peptides did not significantly improve this prediction (Figure).
Conclusion
We have identified a UP signature that predicts KF risk with high precision, but did not significantly ameliorate the prediction obtained by a combination of age, sex, eGFR and albuminuria.
Funding
- Government Support – Non-U.S.