New Genetic Variants Associated with Acquired Dilated Cardiomyopathy. Towards a New Predisposing Polygenic Panel

Background: Heart failure with reduced left ventricular ejection fraction (HFrEF) is one of the cardiovascular pathologies with the highest mortality.The existence of a genetic predisposition for the development of this pathology in the presence of a noxa is still uncertain. Objectives:The aim of this study was to identify associations between genetic variants in polymorphic alleles, single nucleotide polymorphisms (SNPs) with the presence of rEF and intraventricular conduction disorders (IVCD) in a model of acquired cardiomyopathy such as chagasic cardiomyopathy.Methods: A study model including patients seropositive for Chagas disease with a history of infection of more than 20 years, one group with rEF (35%) and another group with preserved LVEF (pEF, 50%).A blood collection was performed and processed to obtain deoxyribonucleic acid (DNA), which was sent to an international genotyping laboratory.A panel of gene SNPs was prepared using the international TheGenomeAggregationDatabase (GnomAD), choosing SNPs with a frequency in the population of 10% to 40%.The SNPs chosen are related to genes responsible for the process of ventricular contraction and relaxation (TTN, BAG3, MTSS1), myocardial metabolism (PPARGC1A, SIRT1, AKT1 mTOR, AMPK), adrenergic receptor beta 1 (ADRB1), cholinergic receptor muscarinic 2 (CHRM2), angiotensin II type 1 receptor(AGTR1B), and atrial natriuretic peptide (NPPA) Using various artificial intelligence models for supervised learning (Logistic Regression, Support Vector Machines, Artificial Neural Networks, NaiveBayes, Classification Trees, and RandomForest) 68 SNPs were evaluated as predictors of the presence of two phenotypes: rEF, and IVCD.The performance of the models for the prediction of rEF was evaluated by cross-validation technique, with the F1 metric as a measure of accuracy to select the best model.Results: One hundred and eighty-two patients were included, with a median age of 62 years, 39.6 % men.Thirty-one percent had rEF and 53% had some IVCD.The model with the best performance was logistic regression (F1=0.85),so it was used to express the association by Odds Ratio and its 95% CI.It was observed that the variants rs2076300, rs61772962 and rs7071853 were independent predictors of rEF, whereas the only predictor for IVCD was the SNP rs72840788.The SNP rs61772962 corresponds to the PRKAA2 gene, which encodes the alpha-2 catalytic subunit of 5'-AMP-activated protein kinase.On the other hand, rs7071853 and rs72840788 correspond to the gene encoding the BAG3 protein (chaperone regulator 3).Finally, the SNP rs2076300 corresponds to the DSP gene encoding a desmoplakin, a protein of the desmosome binding plate.Conclusions: In this derivation model, in a set of patients with positive serology for Chagas disease, 3 SNPs predictive of rEF and one SNP predictive of IVCD were identified.The reproducibility of these results should be confirmed in a validation model with a larger sample of individuals.Furthermore, this finding could be useful in dilated cardiomyopathies of other etiologies.