BACKGROUND: Bepridil is an antiarrhythmic drug used to treat paroxysmal atrial fibrillation; however, its use is limited by corrected QT interval (QTc) prolongation and the risk of ventricular arrhythmias. This study aimed to identify genetic variants associated with bepridil-associated QTc prolongation after catheter ablation for paroxysmal atrial fibrillation. METHODS: A total of 523 patients who initiated bepridil therapy after catheter ablation at Hiroshima University between November 2013 and March 2023 were enrolled. Of these, 445 patients were included in a genome-wide association study, and 78 were used as an independent replication cohort. QTc was measured 1 month after treatment initiation, with QTc prolongation defined as ≥450 ms in men and ≥460 ms in women. Logistic regression analyses were conducted using additive, dominant, and recessive genetic models. Variants showing suggestive associations in the genome-wide association study (P<1.00×10-6) were evaluated in the replication cohort, followed by random-effects meta-analyses, with genome-wide significance defined as P<5.00×10-8. Linear regression analyses were conducted for QTc prolongation. RESULTS: After quality control, 443 patients were analyzed in the genome-wide association study. Of these, 122 (28%) exhibited QTc prolongation. Under the recessive model, rs12622919, an intronic variant in the FSHR (follicle-stimulating hormone receptor) gene, showed a suggestive association. In the random-effects meta-analysis combining the discovery and replication cohorts, rs12622919 reached genome-wide significance. Exploratory sex-stratified analyses suggested a potential association in women (P=0.026), although no statistically significant genotype-by-sex interaction was observed. Inclusion of rs12622919 improved discrimination for QTc prolongation (area under the curve, 0.77 [95% CI, 0.70-0.84]; Welch's t test, P<0.001), with the improvement remaining significant by DeLong's test (area under the curve, 0.78 [95% CI, 0.74-0.84]; P=0.0078). CONCLUSION: The FSHR variant rs12622919 is associated with bepridil-associated QTc prolongation and may serve as a genetic biomarker for individual risk assessment. Because the findings are based on meta-analysis, independent validation is required.