RECURRENT GENOMIC COPY NUMBER VARIANTS IMPLICATE NEW CANDIDATE GENES FOR EARLY ONSET BICUSPID AORTIC VALVE DISEASE

Background Bicuspid Aortic Valve (BAV), the most common adult congenital heart defect, is a major cause of aortic insufficiency or stenosis requiring valve replacement and thoracic aortic aneurysms predisposing to acute aortic dissections. The spectrum of BAV ranges from early onset valve and aortic complications to sporadic late onset disease. We hypothesize that genomic copy number variants (CNVs) modify the penetrance and severity of early onset BAV (EBAV). Methods We performed genome-wide microarray analysis of 140 EBAV probands who were enrolled in the UTHealth Bicuspid Aortic Valve Research Registry. CNVs were detected in Illumina Global Screening Array genotypes using PennCNV, Nexus and CNVPartition algorithms. The EBAV cohort (22% female, mean age 18) was compared to 15,414 controls without known cardiac disease from the Database of Genotypes and Phenotypes. Only CNV calls that contained at least six consecutive variants, were identified by at least two algorithms and intersect with known genes were included. For comparison, 140 older late presenting BAV patients (LBAV group) and 58 BAV probands with early onset thoracic aortic disease (ETAD group) were assayed using identical methods. CNV burden and associations were tested using PLINK (v1.07). CNVs were validated using quantitative PCR. Results We identified 7 recurrent CNV regions that are absent or extremely rare (frequency <0.1%) in controls and 1 polymorphic duplication that is enriched in EBAV cases. Six of these regions are shared by EBAV and LBAV or ETAD cases and involve candidate genes that interact with each other during heart development. Four CNVs involve genes that cause BAV when mutated. The largest and most prevalent of the recurrent CNVs are at 2q37.3 (1 duplication, 8 deletions), 22q11.21 (3 duplications, 2 deletions) and Xp22.33 (8 duplications). The rate of large (>200 Kb) genic deletions and the percentage of individuals with rare genic CNVs (27%) are both increased in the EBAV cohort in comparison to the LBAV (9%) or control (10%) cohorts (empiric P < 1×10-5). Conclusion We identified rare CNVs in one quarter of EBAV patients, implicating alterations of candidate genes in these loci in the pathogenesis of BAV