Whole-exome sequencing generated polygenic risk score and risk stratification in Taiwanese genetic cardiomyopathies: a nation-wide sudden arrhythmia death syndrome (SADS) Taiwan registry

Background Genetic cardiomyopathies (GCMs) are significant contributors to morbidity and mortality. While traditionally associated with rare pathogenic variants, emerging evidence suggests that common genetic variants also play a role in disease development.1 In addition, certain genetic mutations may lead to higher event rates.2 However, most studies have predominantly focused on Caucasian populations, with limited data available on Taiwanese individuals. Purpose This study aims to identify common and rare genetic variants associated with GCMs in Taiwanese patients and to develop a population-specific polygenic risk score (PRS) for disease susceptibility. Additionally, we aim to identify high-risk variants associated with sudden cardiac death (SCD). Methods Patients diagnosed with GCMs (n = 221) from the SADS-Taiwan registry (2015–2023) were included in this study. Whole-exome sequencing (WES) followed standard protocols and the GATK Best Practices workflow was used for variant calling.3 WES data of healthy controls (n = 1,484) were selected from the Taiwan Biobank. Variants with a minor allele frequency (MAF) >1% were identified to assess the association between common genetic variants and disease susceptibility. The dataset was divided into training and validation sets (8:2), and a pseudo-R-squared test was used to identify variants (R² = 0.1–0.4) for polygenic risk score (PRS) construction. To evaluate the effect of rare variants on SCD, we focused on 550 cardiomyopathy-related genes and investigated the associations with SCD using SKAT-O.4 Results Among 221 GCM patients (mean age 44 ± 22 years, 67.8% male), the phenotypes included HCM (n = 125), ARVC (n = 30), LVNC (n = 26), DCM (n = 26), AFD (n = 9), and ATTR-CM (n = 5). For common genetic variants and disease susceptibility, a pseudo-R-squared test identified 22 significant variants, including six located within the MFN1, KLHL2, SDHA, PCDHB3, PRKN, and AHNAK2 genes, which have previously been associated with cardiomyopathies. Polygenic risk score (PRS) construction based on these variants achieved an area under the curve (AUC) of 0.90 for GCM prediction (Figure 1A), with decision curve analysis confirming its clinical utility (Figure 1B). Regarding genetic risk of SCD, 23 patients (10.4%) developed SCD. High-risk SCD-related genes (VCL, FOXF1, SCN5A, TRNT1, CPT1A) were identified using the SKAT-O test (p < 0.02) (Figure 2A & 2B). Conclusion This study highlights the role of specific common genetic variants in predicting the development of GCM in Taiwanese individuals, consistent with findings from Caucasian studies. High-risk genes associated with sudden cardiac death (SCD) were also identified in Taiwanese. Our findings provide valuable insights into the genetic architecture of GCM susceptibility and lay the groundwork for future research on mechanisms and the enhancement of risk stratification.