Left ventricular mass (LVM) is an important clinical phenotype, whose assessment can predict adverse cardiovascular events and premature death in all genders, races, and ages. Increase in LVM defines left ventricular hypertrophy (LVH) with the thickening of the left ventricle of the heart. In community-based cohorts, the presence of left ventricular hypertrophy (LVH) and increased LVM predict the development of coronary heart disease, congestive heart failure, stroke, and cardiovascular disease. Thus this trait serves not only as measures of cardiac structure, but also as intermediate phenotype for clinical cardiovascular disease outcome. Several studies have indicated that LVM is influenced by genetic factors. Genome wide linkage and association studies in diverse population have been performed to identify genes influencing LVM, but much of the heritability remains unexplained, the identity of the underlying gene pathways and functional variants remain unknown, and the promise of genetically based risk prediction remains unfulfilled. The aim of the study was to investigate the association of common genetic variants with left ventricular mass using a genome wide approach in a large cohort of never treated mild-to-moderate essential hypertensive subjects. From the linear analysis, we selected 85 single nucleotide polymorphisms (SNPs), with a suggestive p-value of association with LVM (≤ 10-5). In particular, some SNPs lying in genes previously described as having a role in the pathogenesis of cardiac hypertrophy, such as ROCK1, IGF1, CACNA1D, FGFR1, TRAF5, SOX5, and KSR2. Each of them might play a putative role in determining the LVM phenotype as well as other pathophysiological pathways directly or indirectly linked to cardiac pathophysiology. To assess the risk alleles associated to the most interesting findings in relation to the phenotype studied, we performed a case-control analysis by dividing our sample in two subsets according to LVM values. Most of the SNPs associated with LVM in linear regression presented a significant association, showing that the carriers of the risk alleles have an odds ratio higher than 1 to have increased LVM, i.e. to be cases respect to controls. Nevertheless as for most of the complex traits, the observed odds ratios are modest (except for those biased by the absence of homozygous risk genotypes), so their relevance for a clinical use is uncertain. Thus, we defined a weighted genetic risk score using the effect size of the risk allele (beta value of the linear regression analysis) to account for the strength of the genetic association with each allele. The possibility to combine more variants in a global genetic risk score could be interesting and could add relevance to the results. In conclusion, our GWAS allowed us to pinpoint genes whose role in heart function and/or cardiac hypertrophy has been demonstrated in previously publications by different authors. Moreover, we highlighted the usefulness of an aggregate measure of risk of LVH to discriminate high-risk subjects. However, the results must be interpreted within the context of some potential limitations and perspectives. No SNPs reached a Bonferroni’s significance level probably due to a limited sample size. However, the phenotypic homogeneity of our cohort and the absence of previous antihypertensive treatment are prerequisites for the identification of true genetic effects. A replication in independent cohorts is needed VII to further confirm the findings; however an independent cohort with similar criteria was not available for replication. Moreover, it often happens, as in our study, the significant SNPs map in non-coding regions, making it difficult to explain their causative role. These limitations should not reduce the relevance of the genes identified and confirmed by previously published papers. Future perspectives of this study should be the replication of the GWAS findings in independent cohorts and the assessment in independent samples of the prediction ability of wGRS to correctly classify true positives and true negatives according to their genetic background.
A genome wide approach to identify genetic variants associated with left ventricular mass / M. Chittani ; tutore: C. Barlassina. - Milano : Università degli studi di Milano. Università degli Studi di Milano, 2016 Jul 01. ((28. ciclo, Anno Accademico 2016.
|Titolo:||A genome wide approach to identify genetic variants associated with left ventricular mass|
|Data di pubblicazione:||1-lug-2016|
|Settore Scientifico Disciplinare:||Settore MED/04 - Patologia Generale|
|Citazione:||A genome wide approach to identify genetic variants associated with left ventricular mass / M. Chittani ; tutore: C. Barlassina. - Milano : Università degli studi di Milano. Università degli Studi di Milano, 2016 Jul 01. ((28. ciclo, Anno Accademico 2016.|
|Digital Object Identifier (DOI):||http://dx.doi.org/10.13130/m-chittani_phd2016-07-01|
|Appare nelle tipologie:||Tesi di dottorato|