Abstract 4362480: East Asian-specific genome-wide association study and multi-ancestry meta-analysis reveal the genetic foundations for valvular heart disease and its subtypes

Background: The genetic architectures of valvular heart disease (VHD) and its subtypes have yet to be fully elucidated. Aims: We aimed to clarify the genetic architectures of VHD and its subtypes and to gain insights into VHD biology. Methods: We performed deep-phenotyping by analyzing detailed medical records and echocardiographic data from BioBank Japan. We classified VHD subtypes from stenosis to regurgitation and other abnormalities in the mitral, aortic, tricuspid, and pulmonary valves and the septum, followed by conducting East Asian-specific genome-wide association studies (GWASs). Then, we also performed GWASs in the UK Biobank, applying deep-phenotyping using ICD-10 codes. In addition, we utilized other available GWAS summary statistics, mainly from FinnGen and VA’s Million Veteran Program. Finally, we conducted cross-ancestry meta-analyses. Results: An East Asian-specific GWAS for VHD with 10,358 cases and 128,008 controls identified five genome-wide significant loci, while GWASs for VHD subtypes identified an additional 14 loci. Of the identified 19 loci, 14 had not been previously reported in GWASs for any valve diseases, whereas 12 exhibited very low allele frequencies (<0.001) in other ancestries. Moreover, cross-ancestry meta-analysis for VHD, which involves 93,584 cases and 1,619,230 controls, revealed 89 loci. Then, meta-analyses for VHD subtypes identified an additional 68 loci, totaling 157 loci, including 83 novel loci. Valve regurgitation and stenosis shared only a few loci, suggesting different etiologies. Interestingly, all regurgitation subtypes shared a locus prioritized for PITX2 with relatively high absolute betas (> 0.1), indicating a pivotal role of PITX2 . Pathway analysis revealed that genes associated with VHD were enriched for heart morphogenesis. Additionally, VHD exhibited a significant genetic correlation with atrial fibrillation (AF), congestive heart failure (CHF), and blood pressure (BP), while Mendelian randomization analysis indicated a causal role of BP in VHD. Furthermore, the polygenic risk score (PRS) for VHD stratified the risk of cardiovascular mortality in the other controls in BBJ. The combination of PRS for CHF or AF with that for VHD outperformed original PRS in the prediction of CHF or AF, suggesting VHD as a genetically independent factor. Conclusion: We conducted the first systematic GWASs for VHD and its subtypes, which elucidated the distinct genetic architectures of different VHD subtypes.