Kidney Week

Abstract: FR-OR04

Genome-Wide Association Study of Hospitalized AKI

Session Information

• AKI Prediction and Assessment: Traditional and Novel Tools
  November 03, 2023 | Location: Room 118, Pennsylvania Convention Center
  Abstract Time: 04:57 PM - 05:06 PM

Category: Acute Kidney Injury

• 101 AKI: Epidemiology, Risk Factors, and Prevention

Authors

• Siew, Edward D., VA Tennessee Valley Healthcare System, Nashville, Tennessee, United States

Edward D. Siew, MD, MS, MSc

• Hellwege, Jacklyn N., Vanderbilt University Medical Center, Nashville, Tennessee, United States

Jacklyn N. Hellwege,

• Hung, Adriana, VA Tennessee Valley Healthcare System, Nashville, Tennessee, United States

Adriana Hung,

• Birkelo, Bethany, Vanderbilt University Medical Center, Nashville, Tennessee, United States

Bethany Birkelo,

• Parr, Sharidan, Nebraska Western Iowa Veterans Affairs, Omaha, Nebraska, United States

Sharidan Parr,

• Robinson-Cohen, Cassianne, Vanderbilt University Medical Center, Nashville, Tennessee, United States

Cassianne Robinson-Cohen,

• Susztak, Katalin, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania, United States

Katalin Susztak,

• Greevy, Robert, Vanderbilt University Medical Center, Nashville, Tennessee, United States

Robert Greevy,

• Vincz, Andrew J., Vanderbilt University Medical Center, Nashville, Tennessee, United States

Andrew J. Vincz,

• Matheny, Michael Edwin, VA Tennessee Valley Healthcare System, Nashville, Tennessee, United States

Michael Edwin Matheny,

• Velez edwards, Digna R., Vanderbilt University Medical Center, Nashville, Tennessee, United States

Digna R. Velez edwards,

Background

AKI commonly complicates hospitalization and has a potential genetic basis for susceptibility.

Methods

We conducted a genome-wide association study of AKI in the Million Veteran Program(MVP) and Vanderbilt University's DNA biobank(BioVU). Among patients hospitalized during 2002-2019, AKI cases were defined as KDIGO Stage ≥2, or Stage 1 AKI for ≥2 days. Non-AKI controls were identified among hospitalized patients without historical evidence of AKI. Single variant logistic regression analyses were performed on imputed genetic variants, adjusting for sex, age, and the top 10 principal components of ancestry. Results were meta-analyzed using fixed effects analyses. A second meta-analysis was performed with summary statistics from UK Biobank, FinnGen, and BioBank Japan. Bioinformatic analyses were performed to provide functional information for significant variants.

Results

There were 58,891 patients with AKI(53,079 MVP,5,812 BioVU) and 139,098 non-AKI controls(127,627 in MVP,11,471 BioVU). Three novel loci reached genome-wide significance (Fig): rs11642015 near FTO(odds ratio[OR] 1.07(95% confidence interval [CI], 1.05-1.09), p=8.2x10-17, EAF[effect allele frequency]=0.36); rs20654180 near MPPED2(OR 1.05(95% CI 1.03-1.06), p=2.9x10-8, EAF=0.68]); and rs8042910 near INTS14(OR 1.06(95%CI:1.04-1.08), p=3.9x10-8, EAF=0.20). Effects were consistent across groups and replicated in diagnosis code-based datasets. Within MVP, FTO remained significant after adjusting for BMI. Genome-wide meta-analysis of Biobank Japan, FinnGen, and UK Biobank (N=67,051 cases, 788,014 controls) identified one additional significant locus: rs76704066 in NALCN [EAF = 1.9%; OR 0.82 (95%CI:0.77-0.88)]. FTO and MPPED2 remained significant (p= 1.0x10-17 and 9.6x10-9, respectively), while INTS14 did not (p= 8.1x10-8).

Conclusion

Three loci were associated with AKI including one (MPPED2) previously associated with CKD. FTO may contribute to the risk of AKI independently of BMI. Further work to determine how these genes may contribute to AKI is ongoing.

Funding

• Veterans Affairs Support