Abstract 564: Genetic Basis of Thoracic Aortic Aneurysms in A Large Unselected Clinical Population

Objectives: Thoracic aortic aneurysms (TAAs) are associated with protein coding variants in genes associated with Loeys-Dietz Syndrome (LDS), vascular Ehlers-Danlos Syndrome (VEDS) and ACTA2 . The purpose of this study was to estimate the prevalence of these variants, along with potential novel variants, in a large unselected clinical population. Methods: We analyzed exome sequence data from 174,452 participants. Rare variants (frequency <0.001) were annotated by functional consequence and classification as pathogenic or likely pathogenic (P/LP) in the ClinVar database. TAA diagnoses were determined by ICD-10 codes. Results: 2,670 (1.5%) individuals had a TAA diagnosis; mean age was 66.1 years, compared to 56.7 for the study population (p=0.014). 69% of TAA cases were male and 71% were smokers. There were 20 carriers of ClinVar P/LP variants in LDS genes (8 variants in 10 carriers for SMAD3 , 3 variants in 3 carriers for TGFB2 , 2 variants in 2 carriers for TGFBR1 , 3 variants in 5 carriers for TGFBR2); 5 had a diagnosis of TAA (2 for SMAD 3, 1 each for the others; ORs =16-65); the age of diagnosis was significantly lower than for non-carriers (p<0.0001). 11 P/LP carriers without a TAA diagnosis were <55 years old and could be at risk for TAA. There was 1 TAA case in 21 COL3A1 P/LP carriers and 0 cases in 10 COL1A1 P/LP carriers. We also identified LDS and VEDS variants that are classified as uncertain significance that could be reclassified as P/LP based on their association with early TAA (Table 1). ACTA2 P/LP variants were also strongly associated with TAA (OR=26; p<0.001). The prevalence of P/LP variants in these genes is 1/3,300 (1/1,350 if potential novel pathogenic variants are included). Conclusions: Our results provide estimates of the prevalence and penetrance of variants associated with TAA in a large unselected clinical population, reveal potential novel pathogenic variants, and support the use of genetic data to identify individuals with high disease risk.