Abstract 844: No Clear Association Between Candidate Gene Variants and Outcomes in 3239 Patients with Chronic Heart Failure: Results from the CHARM Program
Background A number of gene variants have been reported to be related to outcome in patients with chronic heart failure (HF), although none have been well validated in large populations.
Methods and results DNA was analyzed from 3239 patients enrolled in the Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) Program with chronic HF and with preserved or reduced left ventricular ejection fraction (EF). Outcomes were 773 deaths and 1105 cardiovascular (CV) deaths or HF hospitalizations. We chose 14 candidate genes based on prior association studies, biologic plausibility, and/or adrenergic or renin-angiotensin system involvement, including angiotensin converting enzyme (ACE), angiotensinogen (AGT), angiotensin type 1 receptor (AGTR1), and adrenergic receptor genes (ADBR1, ADBR2, ADRA2C). A total of 165 tagging SNPs were selected across the genes (including 5kb in 5′ and 3′ flanking regions) from HapMap validated Caucasian SNPs with frequency >5% using Haploview tagger algorithm, including amino acid changing SNPs (frequency >3%). Analyses were stratified by race, and adjusted for clinical predictors using a validated model and by EF, in European origin (n=2932) and Blacks (n=213) separately. Variants in adenosine deaminase and in adrenergic receptor genes were not significant even in crude analyses. The strongest and most consistent signals across unadjusted and adjusted analyses were in AGT and AGTR1, primarily for death. For the most significant allelic tests in European unadjusted analysis in respective gene, the effects were OR=1.18 (1.04, 1.33), p=0.0096 for rs2478523 (AGT) and OR=0.80 (0.69, 0.93), p=0.004 for rs6801836 (AGTR1). However after adjusting for multiple testing, for the four analyses (CV death or HF hosp, death; allele, genotype), no individual SNP was significant at p<0.05.
Conclusions In this large study of chronic HF, we found no significant associations of common genetic variants in candidate genes and outcomes after adjustment for multiple testing (although validation is needed to exclude false negatives). These results raise doubts about findings from previous smaller studies and underscore the need for large studies and for validation before findings can be considered reliable.