Study's purpose: Only 1 in 5 chronic kidney graft failures can be attributed to human leukocyte antigen (HLA) mismatches. Unmeasured HLA variants, non-HLA mismatches, and gene polymorphisms affecting drug metabolism influence long-term graft survival. Susceptibility variants are not yet ready for incorporation into clinical practice because their interactions and cumulative risk remain unknown. This study aims to stratify the risk of chronic graft failure based on predisposing genetic modifiers in the D-R pairs.
Methods: We genotyped 114 D-R pairs (IRB-HSR210437) using Global Discovery Array with Enhanced Pharmacogenomic Content (GDA-PGx; Illumina) and generated genome-wide (GWAS) data. We performed a GWAS scan for graft failure using the Cox proportional hazards model and for post-transplant eGFR using the linear regression model. Next, using our in-house toolset SECRET (Screen and Evaluate Catalogued Risk Scores to Enhance Trait prediction), we leveraged the Polygenic Score Catalog to identify polygenic score (PGS) models that predict graft failure and post-transplant eGFR levels.  We calculated ~5,000 PGSs for each study subject and then evaluated the power of each PGS to predict graft failure with covariates being adjusted.  
Results: GWAS scan for the graft failure trait identified variant rs62231826 (P= 1.3×10-8), at which the minor allele has a hazard ratio (HR) of 9.3 (Figure 1). This minor allele was also negatively associated with post-transplant eGFR (P=0.002 and P=0.02 for the 1-year and 3-year eGFR). This variant was in the gene SLC6A6, which encodes the sodium-chloride-dependent taurine transporters (TAUT) and has been found in a previous study to predict graft function loss at the level of gene expression, protein abundance, and transporter function. SECRET identified PGS for systolic blood pressure in the donors as the most predictive for graft failure in recipients older than 60 (HR=1.4, P=0.004). The scan identified PGS for HbA1c in the recipient as the most predictive for eGFR decline 5 years after transplant (16% explained by this PGS).     
Conclusions: This proof-of-concept pilot study establishes a framework for risk stratification. In the future, fully powered studies may improve graft allocation and survival and provide faster access for those needing kidney transplantation.