Discovery and analyses of pathogenic variants in explanted hearts from primary cardiomyopathy patients

Abstract

Abstract Background Dilated cardiomyopathy (DCM) and hypertrophic cardiomyopathy (HCM) are disorders of the myocardium that affect the structure and function of the heart. Purpose The primary aim of this study was to discover damaging genetic variants in myocardial tissue from patients with DCM or HCM, who underwent heart transplantation. Methods Whole exome sequencing was performed on myocardial tissue from 103 explanted hearts with diagnosis of cardiomyopathy; 80 DCM and 13 HCM. Sanger sequencing was performed to confirm the loss-of-function variants in genes known to be linked to cardiomyopathy. RNA sequencing was conducted to confirm copy number variation deletions detected in the cohort. Burden analysis was performed by comparing the frequency of variants found in the study cohort to the frequency in the population database gnomAD. Results Rare (minor allele frequency <1.0E-04) loss-of-function variants, deleterious missense variants, or copy number variation deletions, collectively described as damaging variants, were identified in cardiomyopathy genes in 42 of all 93 samples (45.2%). Damaging variants were identified in 37 of 80 DCM samples (46.3%) and 5 of 13 HCM samples (38.4%). The mean read depth for normal and variant allele were comparable. All the 28 loss-of-function variants in cardiomyopathy genes found in the cardiomyopathy cases were confirmed by Sanger sequencing. Two copy number variation deletions both in titin (TTN) were also detected and confirmed. Burden analyses showed that the genes TTN and lamin A/C (LMNA) had a higher frequency of loss-of function variants in the DCM cohort (17.5% and 3.75%, respectively) compared to the reference population with genome-wide significance (p=3.45E-22 and 4.34E-07, respectively). Furthermore, our analysis showed that deleterious missense variants in osteoclast-stimulating factor 1 (OSTF1), which previously has not been associated with cardiomyopathy, was highly enriched in the DCM cohort compared to the reference population (p=2.10E-06). Conclusions The frequency of damaging variants that are likely pathogenic (46.3%) is higher in DCM cases in this cohort compared to previous studies. These data indicate that patients with end-stage DCM are more likely to have a genetic cause for their disease. As read depth of variant and normal alleles were, these are likely germline and not mosaic variants, and can enable cascade testing in family members. Moreover, our study demonstrates that CNVs in TTN that alter the reading frame can cause DCM. Funding Acknowledgement Type of funding source: None