Genetic Landscape of Kidney Failure in a Korean Transplant Cohort: Genome-Wide Association and Multi-Polygenic Risk Score Analyses.

Abstract

Kidney failure represents the final, irreversible stage of chronic kidney disease (CKD), yet its genetic architecture remains incompletely defined compared to CKD. While prior studies largely focused on kidney function traits, the genetic determinants of kidney failure itself and its etiologic subtypes are poorly understood. Therefore, this study aimed to identify genetic loci associated with kidney failure and its subtypes and to evaluate the predictive performance of multiple polygenic risk scores (PRSs) for stratifying kidney failure risk. We performed a large-scale genome-wide association study (GWAS) using a dataset of 2,355 kidney failure patients across three subtypes defined by primary disease, along with 152,131 controls. PRSs for type 2 diabetes, hypertension, estimated glomerular filtration rate (eGFR), and CKD were calculated, and a multi-PRS model was derived. The GWAS identified multiple kidney failure-associated loci, including HLA-DRB1 for all kidney failure and glomerulonephritis as the primary disease, and COL24A1 for hypertensive kidney failure at genome-wide significance (P<5x10-8), implicating distinct immune and hypertension-related pathways in kidney failure pathogenesis. PRS analysis revealed that genetically distinct components of type 2 diabetes, hypertension, CKD, and eGFR were significantly linked to kidney failure risk across subtypes (OR=1.1-2.5). A combined multi-PRS model demonstrated superior predictive performance (OR=1.5-2.5). Comparison with CKD cohorts, predominantly influenced by eGFR-related genetics, uncovered unique kidney failure-specific genetic signatures, highlighting kidney failure as a genetically heterogeneous and multifactorial disease beyond CKD. This study identified subtype-specific genetic loci and demonstrated that multi-PRS models improve kidney failure risk prediction beyond basic available clinical factors.