Kidney Week

Abstract: TH-PO080

Urine Proteomic Analysis Identifies Interferon Gamma Downstream Chemokine CXCL-9 as a Biomarker for Diagnosis of Acute Interstitial Nephritis

Session Information

• AKI: Biomarkers, Risk Factors, Treatments, Outcomes
  November 03, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
  Abstract Time: 10:00 AM - 12:00 PM

Category: Acute Kidney Injury

• 101 AKI: Epidemiology‚ Risk Factors‚ and Prevention

Authors

• Moledina, Dennis G., Yale School of Medicine, New Haven, Connecticut, United States

Dennis G. Moledina,

• Obeid, Wassim, Johns Hopkins University, Baltimore, Maryland, United States

Wassim Obeid,

• Sise, Meghan E., Massachusetts General Hospital, Boston, Massachusetts, United States

Meghan E. Sise,

• Rosales, Ivy A., Massachusetts General Hospital, Boston, Massachusetts, United States

Ivy A. Rosales,

• Smith, Rex Neal, Massachusetts General Hospital, Boston, Massachusetts, United States

Rex Neal Smith,

• Colvin, Robert B., Massachusetts General Hospital, Boston, Massachusetts, United States

Robert B. Colvin,

• Kuperman, Michael Benjamin, Arkana Laboratories, Little Rock, Arkansas, United States

Michael Benjamin Kuperman,

• Kashgarian, Michael, Yale School of Medicine, New Haven, Connecticut, United States

Michael Kashgarian,

• Moeckel, Gilbert W., Yale School of Medicine, New Haven, Connecticut, United States

Gilbert W. Moeckel,

• Wilson, Francis Perry, Yale School of Medicine, New Haven, Connecticut, United States

Francis Perry Wilson,

• Parikh, Chirag R., Johns Hopkins University, Baltimore, Maryland, United States

Chirag R. Parikh,

Background

Targeted analyses have identified TNF-a and IL-9 as diagnostic biomarkers of AIN. However, unbiased analysis may reveal more accurate biomarkers for AIN diagnosis.

Methods

In a prospectively enrolled cohort of participants who underwent a kidney biopsy for evaluation of acute kidney disease with adjudicated histological diagnosis, we performed urine proteomics via Olink assay. We validated proteomics findings by developing a sandwich immunoassay in urine, and by examining gene expression in kidney tissue using Nanostring.

Results

We observed that 32 (17%) of 184 proteins were significantly different between AIN cases (n=31) and non-AIN controls (n=57). Of these differentially expressed proteins, CXCL-9 had the highest strength of association (Figure). Pathway analysis showed that activation of interferon-γ, the key upstream regulator of CXCL9, explained observed changes in the urine proteome. Using a sandwich immunoassay, we showed that CXCL-9 was 6-fold higher in AIN cases (n=32) than non-AIN controls (n=186). Participants in the top quartile of CXCL-9 had higher odds of AIN than those in the bottom quartile after controlling for a diagnostic model for AIN (adj. OR, 5.9 (95% CI, 1.8, 20)) with an AUC of 0.82 (0.74, 0.89). Finally, CXCL-9 expression was higher in kidney tissue from patients with drug-induced AIN as compared to those with diabetes, acute tubular injury, or healthy kidney donors.

Conclusion

Using urine proteomics, we identified CXCL-9 as a novel biomarker for AIN diagnosis. We validated the association of CXCL-9 using immunoassay and kidney tissue expression data.

Funding

• NIDDK Support