Abstract: TH-PO450
The Polygenic Burden of Rare Variants Predicts Onset of CKD in the UK Biobank
Session Information
- Genetic Diseases: Glomerulopathies - I
November 02, 2023 | Location: Exhibit Hall, Pennsylvania Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Genetic Diseases of the Kidneys
- 1202 Genetic Diseases of the Kidneys: Non-Cystic
Authors
- Lali, Ricky, McMaster University, Hamilton, Ontario, Canada
- Vlasschaert, Caitlyn, Queen's University, Kingston, Ontario, Canada
- Lanktree, Matthew B., McMaster University, Hamilton, Ontario, Canada
Background
Genetic contributors to chronic kidney disease (CKD) have been explored through monogenic mechanisms by rare mutations and polygenic mechanisms through the aggregate impact of many small-effect common variants. In this work, we test whether a polygenic burden of rare variants across numerous genes contributes to CKD by constructing a rare variant polygenic risk score (rvPRS).
Methods
We first conducted a discovery exome-wide association study (ExWAS) of rare protein-truncating variants with a calculated severe CKD phenotype based on eGFR below 30 ml/min/1.73m2 using a discovery set of 834 cases and 147,855 British European controls in the UK Biobank (UKB). After excluding known Mendelian CKD genes, an rvPRS was constructed of 124 nominally significant (P<0.05) risk genes associated with severe CKD. As such, the effect conferred through rvPRS124 would not be driven through underlying monogenic mechanisms. We then tested the predictive power of rvPRS124 using a validation set consisting of 688 independent CKD events in the UKB.
Results
In the validation set, rvPRS124 conferred a 21% increase in hazard for incident CKD onset with one rare protein-truncating allele (HR=1.21; 95%CI, 1.01-1.46; P=0.045) after adjusting for age, sex, the first 5 principal components of ancestry, and pertinent clinical risk factors including obesity, myocardial infarction, and smoking. Individuals with 2 or more rvPRS124 alleles (N=1,352) had a 10-fold increase in hazard for CKD onset compared to individuals with no rvPRS124 variants (HR=10.0, 95% CI, 2.5-39.4; P=1.3 x 10-3). No single gene in rvPRS124 replicated an association with CKD after adjusting for multiple hypothesis testing (P<4x10-4; 0.05/124), which emphasizes the importance of rare variant polygenic mechanisms underlying CKD. Lastly, through 1000 permutations of random gene sets, we show that the association of rvPRS124 with CKD was specific to the selected genes used to construct the score and not solely due to the gene count (P<0.05).
Conclusion
Using the UKB, we demonstrate a cumulative impact of rare protein-truncating variants in genes not known to have monogenic effects on CKD. An omnigenic score, incorporating established clinical risk factors and both Monogenic and common variant polygenic effects, should also include the polygenic burden of rare variants in non-Mendelian CKD-related genes.