Kidney Week

Abstract: FR-PO1158

Integrating Bidirectional Mendelian Randomization into Proteomic Analysis to Prioritize Plasma Proteins for Regulating and Monitoring Kidney Function

Session Information

• CKD: Screening, Diagnosis, Serum and Urine Biomarkers, and Scoring Indices
  November 07, 2025 | Location: Exhibit Hall, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: CKD (Non-Dialysis)

• 2301 CKD (Non-Dialysis): Epidemiology, Risk Factors, and Prevention

Authors

• Wu, Ping-Hsun, Kaohsiung Medical University Chung Ho Memorial Hospital, Kaohsiung, Kaohsiung City, Taiwan

Ping-Hsun Wu, MD, MS, PhD

• Lin, Ya-Chi, China Medical University Hospital, Taichung, Taichung City, Taiwan

Ya-Chi Lin, PhD

• Chen, Hung-Lin, China Medical University Hospital, Taichung, Taichung City, Taiwan

Hung-Lin Chen, PhD

• Lin, Yi-Ting, Kaohsiung Medical University Chung Ho Memorial Hospital, Kaohsiung, Kaohsiung City, Taiwan

Yi-Ting Lin, MD, PhD

• Chang, David R., China Medical University Hospital, Taichung, Taichung City, Taiwan

David R. Chang, MD

• Kuo, Chin-Chi, China Medical University Hospital, Taichung, Taichung City, Taiwan

Chin-Chi Kuo, MD, PhD, MPH

Background

Chronic kidney disease (CKD) affects millions worldwide, yet our understanding of the molecular mechanisms remains incomplete. This study aimed to identify causal relationships between plasma proteins and kidney function, as approximated by estimated glomerular filtration rate (eGFR), using Mendelian randomization (MR) analysis.

Methods

We analyzed data from 46,451 participants in the UK Biobank with available proteomic profiles generated using the Olink Platform and corresponding eGFR measurements. Forward and reverse two-sample MR analyses, along with colocalization analyses, were performed to identify potential drug targets and to screen for novel biomarkers of kidney function, respectively. The ability to predict 5-year incident CKD by incorporating purely forwardly causal and reversely causal proteins, in addition to age and sex, was evaluated using Harrell’s C-index.

Results

Among participants (median age 58 years, 46.1% male), 1,282 (2.8%) had reduced kidney function (eGFR <60 mL/min/1.73m²). A total of 2,920 proteins were included in the final analysis, and 7 proteins were identified as having a purely forward causal effect on eGFR: Angiopoietin-2 (ANGPT2), Serum Amyloid P Component (APCS), Serine Protease 8 (prostasin, PRSS8), Cathepsin S (CTSS), Fc Receptor for IgA and IgM (FCAMR), Parathyroid Hormone (PTH), and Uromodulin (UMOD). Conversely, eGFR was found to causally affect the levels of 265 proteins, with no evidence of reverse causation. The Harrell’s C-index for predicting incident CKD was significantly improved by integrating top 10 biomarker candidate proteins compared to drug target proteins (0.798 vs. 0.723, p < 0.001).

Conclusion

Our study framework, which integrates proteomics with bidirectional Mendelian randomization analyses, helps prioritize key proteins for future research, both as potential drug targets and as biomarkers of kidney function. The prognostic value of these proteins also warrants further investigation.

Digital Object Identifier (DOI)

• doi: 10.1681/ASN.2025dddkwp6g