Associations between reported atrial fibrillation burden and cardiac substructure dosimetric variables in patients receiving thoracic radiotherapy for malignancies involving the thorax: a systematic review and meta-analysis

Introduction: Thoracic radiotherapy is integral to multimodal treatment for malignancies involving the thorax, but the literature on the burden of reported atrial fibrillation (AF) after thoracic radiotherapy and its relationship with cardiac substructure dose exposure has not been systematically synthesized. Methods: We searched PubMed, Embase, the Cochrane Library, and Web of Science through May 17, 2026, for studies that reported AF incidence or comparative/dosimetric associations among adults receiving thoracic radiotherapy for malignancies. Pooled incidence was estimated using logit-transformed random-effects proportion meta-analysis. Because cardiac structures, dose metrics, cutoffs, statistical models, and AF definitions differed across studies, dosimetric evidence was summarized using structured tables and evidence plots. Results: Ten studies were included. Eight studies contributed nine cohort-level estimates to the incidence synthesis, including 3,192 patients, of whom 293 developed study-defined reported AF. The pooled incidence across all eligible cohorts was 9.1% (95% confidence interval (CI), 6.3%-12.9%; I²=90.0%; prediction interval, 2.5%-28.5%). After the exclusion of two esophageal cancer cohorts with substantial perioperative confounding, the main analysis included seven cohorts with a total of 2,766 patients, among whom 217 developed reported AF; the pooled incidence was 7.6% (95% CI, 5.6%-10.3%; I²=80.3%; prediction interval, 2.9%-18.8%). Nine reports described associations between cardiac or cardiac substructure radiation doses and AF risk. Adjusted findings suggested candidate signals involving the pulmonary veins, atria, sinoatrial node, and selected whole-heart dose metrics, although the findings were heterogeneous and unsuitable for quantitative pooling. Conclusion: Study-defined reported AF after thoracic radiotherapy represents a clinically relevant burden. Current evidence suggests candidate dosimetric regions involving the pulmonary veins, atria, and sinoatrial node, with selected whole-heart metrics showing signals. However, the evidence remains insufficient to support causal inference or clinical dose constraints. Prospective multicenter studies using standardized AF definitions, cardiac substructure delineation, and individual-patient data are needed to validate these signals.