Burden Analysis Reveals Genetic Variants in PKD1 and SMAD2 Associated with Intracranial Aneurysm Prevalence

Abstract

Objectives We explored whether genes previously linked to intracranial aneurysms (IAs) are associated with IA prevalence in a larger population and whether there are additional variants within the same genes that exert similar effects.

Background Several Mendelian risk variants for IAs have been identified in family studies. Additionally, several rare monogenic disorders exhibit an inherent predisposition for IA, suggesting a link between the genes implicated in these disorders and the manifestation of IA.

Methods We assessed the collective effect of genetic variants within putative IA-associated genes, including those identified in family studies and known causal genes for monogenic disorders. We tested the association between genic burden with IA prevalence using data from the UK Biobank (N=1,646 IA cases, 925 aneurysmal subarachnoid hemorrhage (ASAH) cases, 391,385 controls). Separate analyses stratified by variant allele frequency and predicted impact were performed. We utilized a: Burden test, Sequence Kernel Association Test (SKAT), and variant-level aggregated Cauchy association test to identify association under various assumptions. Odds ratios with confidence intervals were calculated based on Firth logistic regression.

Results Statistically significant associations with IA prevalence were identified for ultrarare variants with moderate or high impact in PKD1 (p = 1.33×10-5 [SKAT], odds ratio 1.36 [95% confidence interval 1.00-1.79]) and SMAD2 (p = 6.79×10-5 [SKAT], odds ratio 5.51 [95% confidence interval 2.07-11.6]), and for ultrarare synonymous variants in TGFBR2 (p = 7.04×10-5 [SKAT], odds ratio 4.93 [95% confidence interval 1.64-11.1]).

Conclusions In conclusion, our study confirmed associations between IA and ultrarare damaging variants in two genes linked to medical conditions with a known predisposition for IA: PKD1, associated with polycystic kidney disorder, and SMAD2 associated with Loeys-Dietz syndrome, even in absence of a diagnosis of these disorders. Our study failed to identify an association with IA for any of the genes previously identified in family studies.