Myocardial DNA damage is responsible for the relationship between genotype and treatment response in patients with dilated cardiomyopathy

Abstract Background Dilated cardiomyopathy (DCM) is one of the leading causes of heart transplantation. Our group and others have previously reported that DNA damage accumulation in myocardial tissue and certain genetic variants such as those in LMNA are associated with treatment response and prognosis (ref 1-5). However, the potential association between myocardial DNA damage and genotype, and their contribution when combined to predicting treatment response, remains unclear. Methods We identified the deleterious variants in cardiomyopathy-related genes by analyzing the whole exome sequencing data from 89 patients with DCM who were recruited from two participating facilities. Immunostaining of γ-H2A.X, a DNA damage marker, was performed on endomyocardial biopsy specimens from the same patients, from which the percentage of nuclei positive for γ-H2A.X signals (%γ-H2A.X) was calculated. Left ventricular reverse remodeling (LVRR) was assessed by transthoracic echocardiography at 1 year after biopsy. We examined the associations among genotype, myocardial DNA damage, and the achievement of LVRR. Results Among the 89 patients with DCM, 11 (12.4%) carried pathogenic or likely pathogenic (P/LP) variants in sarcomere genes such as TTN, and 21 (23.6%) carried P/LP variants in non-sarcomere genes such as LMNA, while the rest had no P/LP variants in cardiomyopathy-related genes (Figure A). Patients with P/LP variants in sarcomere genes had a lower %γ-H2A.X [21.2 (10.7–29.0)% vs. 7.0 (4.3–12.4)%, P = 0.002] (Figure B), and a higher probability to achieve LVRR at 1-year (81.8% vs. 23.8%, P = 0.003) (Figure C) when compared to those with P/LP variants in non-sarcomere genes. The odds ratio for carriers of P/LP variants in sarcomere genes to achieve LVRR was 14.40 (95%CI 2.31–89.94) compared to that for carriers of P/LP variants in non-sarcomere genes. However, this effect was markedly attenuated after adjusting for the proportion of γ-H2A.X-positive nuclei [adjusted odds ratio 6.05 (95%CI 0.86–42.53)] (Table). Conclusions Among patients with DCM, carriers of P/LP variants in sarcomere genes showed less DNA damage and a favorable treatment response, when compared with carriers of P/LP variants in non-sarcomere genes. Our results suggest that myocardial DNA damage was partially responsible for the relationship between genotype and treatment response in patients with DCM.Figure, panels A-C