GENOME-WIDE ASSOCIATION STUDY OF MORTALITY BENEFIT FROM BETA-BLOCKERS IN PATIENTS WITH HEART FAILURE WITH REDUCED EJECTION FRACTION
Shugg T, Li J, She R, Gui H, Sabbah HN, Williams LK, Luzum JA, and Lanfear DE. GENOME-WIDE ASSOCIATION STUDY OF MORTALITY BENEFIT FROM BETA-BLOCKERS IN PATIENTS WITH HEART FAILURE WITH REDUCED EJECTION FRACTION. J Am Coll Cardiol 2019; 73(9):1005.
J Am Coll Cardiol
Background: Beta-blockers (BB) reduce the risk of mortality in most patients with heart failure with reduced ejection fraction (HFrEF), but not all patients benefit. Genome-wide association studies (GWAS) discovered variants influencing BB benefit in patients with hypertension, but a GWAS evaluating BB benefit in HFrEF patients has not been performed. Therefore the objective of this study was to use GWAS to discover variants influencing BB mortality benefit in HFrEF patients. Methods: HFrEF patients (n = 1,060) were enrolled in a single institution registry from 2007-15 and genotyped with the Axiom® Biobank array. The Michigan Imputation Server imputed ~7 million variants with 1000 Genomes as the reference panel. Variants with minor allele frequency < 1% or deviating from Hardy Weinberg equilibrium were excluded. BB exposure was quantified from pharmacy claims based on dose and adherence. The primary endpoint was time to all-cause mortality. BB exposure by variant interactions were tested using Cox regression in R, adjusted for clinical risk factors, BB propensity, and the first two principal components. Additive genetic models were used and statistical significance was defined as p < 5 × 10-8. HaploReg v4.1 identified putative mechanisms of significant variants. Results: The registry is 35% female, 51% self-reported African-Americans, and n = 993 had survival data and passed genotyping quality control. At baseline, age = 68 ± 12 yrs (mean ± sd), left ventricular ejection fraction = 35 ± 11%, and 77% were treated with BB. Follow-up was 2.4 ± 1.7 yrs with 175 deaths (18%). One variant on chromosome 19 was statistically significant: rs10416900; p = 1.2 × 10-8. This variant is 7.5 kb 5' of MKNK2 and not in linkage disequilibrium (LD) with other variants. It alters DNAse in 5 tissue types and 4 regulatory motifs. Two variants on chromosome 3 were nearly significant: rs6441824 & rs4682720; p < 5.9 × 10-8; 59 kb 5' of TOPAZ1. These variants and many others in LD alter several regulatory mechanisms, including histone marks and DNAse. Conclusion: GWAS discovered novel candidate variants for BB mortality benefit in HFrEF patients with putative regulatory mechanisms. These data require replication in other HFrEF cohorts.
9 Suppl 1