Genome-wide association meta-analysis of 5,864 individuals highlights 25 new loci implicated in the pathogenesis of cancer therapy-related cardiac dysfunction

Abstract

Abstract Background Cancer therapy-related cardiac dysfunction, which commonly includes varying degrees of asymptomatic ventricular dysfunction and/or heart failure, is the main adverse effect of anticancer therapy. The mechanism by which it develops is still quite uncertain. In recent years several genetic association studies have identified common genetic variants associated with cardiotoxicity in genes related to DNA damage, drug transportation and metabolism, antioxidant mechanism, sarcomere dysfunction and impaired iron metabolism, but the evidence is still limited. Rare variants in familial cardiomyopathies genes have also been identified. Purpose This uncertainly requires larger studies to identify genetic variants allowing to predict the risk of cardiotoxicity. In this sense, we carry out a genome-wide association meta-analysis to identify genetic variants associated with cancer therapy-related cardiac dysfunction. The development of a risk predictor model combining clinical and genetic factors would make it possible to anticipate the adverse effects of this therapy. Methods Genome wide meta-analysis was performed with a total of 594 patients from the CARDIOTOX registry and 3545 healthy individuals from the National DNA Bank, together with three previously published independent cohorts with 1446, 227 and 52 individuals receiving cancer therapy. Patients were classified with different cardiotoxicity degrees based on the CARDIOTOX registry and exhaustive clinical information was obtained from medical records. The study was ethically approved by local institutional review boards and all participants provided written informed consent. DNA samples were genotyped and imputed to perform a GWAS meta-analysis for case-control (n=5864) and extreme phenotypes (n=5317) looking for genetic variants that predispose to developing cancer therapy-related cardiac dysfunction. A functional characterization of the significant loci was performed by colocalization with eQTLs. Results The case-control analysis showed 17 loci that achieved suggestive significance (< P=1 x10-5) (Figure 1). The rs6765460, an intergenic variant between TPRG1 and TP63 genes, had the strongest relationship with the development of cardiotoxicity (P=1.89x10-7). The extreme phenotypes analysis showed additional 8 loci that achieved suggestive significance. The rs12904266, an intron variant of SLCO3A1, had the strongest relationship with severe cardiotoxicity (P=3.20E-07). eQTLs colocalization of the significant loci revealed association with genes involved in the mitochondria, inflammation, immune system, and fibrosis. Conclusions Our GWAS meta-analysis identified 25 novel genetic variants that may be associated with cancer therapy-related cardiac dysfunction. Genotyping of these markers could be considered in a genetic risk score which, together with the already known clinical risk factors, would improve the cardiotoxicity prediction in cardio-oncology units.Figure 1.Miami plot with P values