Precision medicine: myocardial fibrosis burden and genotype predict outcome in non-ischemic dilated cardiomyopathy (DCM)

Introduction Myocardial fibrosis occurs during pathological remodeling of the heart and can be associated with worse outcome in affected patients. Genetic background is also known to affect patients' survival. In this study we sought to estimate patients' overall fibrosis burden by combining independent modalities including left ventricular endomyocardial biopsy (LV-EMB), circulating biomarkers, and cMRI. We also aimed to use patients' genetics information to predict outcome. Furthermore, we investigated the correlation between cardiac fibrosis and genetic variations to detect novel susceptibility loci for fibrosis in DCM. Methods A total number of 542 DCM patients were included. Collagen volume fraction (CVF) was automatically estimated from biopsies. 13 circulating fibrosis biomarkers were measured using Human Magnetic Luminex Screening Assays, and the cMRIs were screened for presence of LGE. Whole exome sequencing (WES) was performed in 410 patients of the cohort using illumina HiSeq 2000. Common (MAF ≥0.05 in the study population OR gnomAD NFE AF ≥0.01) and non-common missense variants (MAF Results The median follow-up time was 43.2 months (2084 patient-years). Sixty-two patients reached the composite end point and 55 died. LV-EMB proved to be a safe procedure with a total complication rate of 2.3%. Machine learning based characterization of biopsies was highly accurate. Although the 3 different modalities did not significantly correlate with one another, the extent of CVF, levels of MMP-2, TIMP-1, OPN, and GDF-15, and presence of LGE were each significantly associated with worse outcome. Four possible susceptibility loci for cardiac fibrosis in DCM were introduced and underwent eQTL analyses. Rare missense variants in a list of 11 DCM-related genes showed to be associated with the 2 outcome measures or fibrosis burden. Conclusions LV-EMB, fibrosis biomarkers, and cMRI likely capture different aspects of a detrimental fibrosis process and may be combined to estimate patients' prognosis and monitor therapeutic success. Phenotype-genotype association studies help elucidate novel disease pathomechanisms and individualize patients' treatment. Funding Acknowledgement Type of funding source: Other. Main funding source(s): German Centre for Cardiovascular Research: DZHK