Kidney Week

Abstract: TH-PO862

Circulating Proteins Improve Prediction of Short-Term Kidney Disease Progression

Session Information

• CKD: Epidemiology, Risk Factors, Prevention - I
  November 03, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
  Abstract Time: 10:00 AM - 12:00 PM

Category: CKD (Non-Dialysis)

• 2201 CKD (Non-Dialysis): Epidemiology‚ Risk Factors‚ and Prevention

Authors

• Lopez-Silva, Carolina, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States

Carolina Lopez-Silva,

• Surapaneni, Aditya L., New York University Grossman School of Medicine, New York, New York, United States

Aditya L. Surapaneni,

• Coresh, Josef, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States

Josef Coresh,

• Chen, Teresa K., Johns Hopkins University School of Medicine, Baltimore, Maryland, United States

Teresa K. Chen,

• Grams, Morgan, New York University Grossman School of Medicine, New York, New York, United States

Morgan Grams,

Background

Chronic kidney disease (CKD) progression may be preventable with early medical intervention, but few tools reliably identify high-risk patients in early disease. The proteins KIM-1, TNFRSF1A, and TNFRSF1B have been proposed as promising early markers of CKD progression.

Methods

Using participants from ARIC visit 2 (mean age 57 years, 56% women, mean eGFR 98 mL/min/1.73m²), ARIC visit 5 (mean 76 years, 76% women, mean eGFR 69 mL/min/1.73m²) and AASK (mean age 55 years, 39% women, mean mGFR 46 mL/min/1.73m²) studies, we compared the performance of three clinical models: model 1 (age, sex, GFR, race and center/randomized group), model 2 (model 1 + albuminuria), model 3 (model 2 + diabetes, systolic BP, hypertension medication, smoking status) with those incorporating KIM-1, TNFRSF1A and TNFRSF1B. Individuals were stratified by diabetes status and eGFR level.

Results

The baseline clinical models had strong risk discrimination for 3-year 40% eGFR decline (C-statistic range for model 3: 0.73-0.89). Baseline levels of KIM-1, TNFRSF1A and TNFRSF1B improved risk prediction among 9,605 ARIC visit 2 participants (where albuminuria was not obtained). Among 2,935 ARIC visit 5 participants, the biomarkers enhanced risk prediction over model 1, except in patients without diabetes. Finally, among 557 AASK participants, the biomarkers added discrimination to model 1 but not to models 2 or 3 (Table 1).

Conclusion

Inclusion of KIM-1, TNFRSF1A and TNFRSF1B in clinical models resulted in small but significant improvements in risk prediction for short-term 40% eGFR decline in subgroups of patients at various levels of risk. Model improvement for 40% eGFR decline was more consistent among patients with diabetes and eGFR<60 ml/min/1.73 m², and before inclusion of albuminuria.

Funding

• Other NIH Support