Abstract 3561: Genetic Variants Associated with Acute Myocardial Infarction Risk in Five Ethnic Groups: The INTERHEART genetics study
Background: The INTERHEART case-control study showed 9 modifiable risk factors (tobacco, diabetes, hypertension, dyslipidemia, abdominal obesity, physical inactivity, psychosocial stress, low fruit and vegetable intake, and no alcohol consumption) account for >90% of the population attributable risk for MI globally. We tested the association of SNPs within selected candidate genes with MI, and MI risk factors in 8,034 individuals from 5 ethnic groups.
Design and Methods: Individuals of South Asian, Arab, Iranian, Nepalese and European origin were genotyped for 1,536 HapMap SNPs capturing common variation in 103 candidate genes using an Illumina panel. 1,442 SNPs were polymorphic and in HWE in all 5 ethnic groups. We employed a staged approach in which 1,344 SNPs were first assessed versus at least one of 9 MI risk factors with which they were hypothesized to be associated. SNPs that were significantly associated with a risk factor (after adjusting for age, sex and ethnic group, and multiple testing using the Bonferroni correction) were passed to Stage 2 and then tested in an additive model versus MI. Cases and controls were matched by ethnic group, sex and age. An empirical P value for significance was determined using permutation testing in Stage 2.
Results: Fifteen SNPs passed Stage 1 testing. Twelve SNPs were associated with Apo B/A ratio, 2 SNPs were associated with alcohol use, and 1 SNP was associated with fruit and vegetable consumption. Three SNPs from the two genes Apo E (rs# 7412) [MAF=5.7%; odds ratio = 0.78 (95% CI: 0.70 – 0.89) P value=0.0004] and LDLR (rs# 1433099 [MAF=29.9%, OR=0.90; (95% CI: 0.84 – 0.96); P=0.002; rs# 6511720; MAF =9.5%, OR =0.86; (95% CI: 0.77– 0.95); P=0.004] were confirmed to be directly related to MI. After adjustment for Apo B/A levels only one SNP (rs# rs1433099) from the LDLR gene remained independently associated with MI [OR=0.91, 95% CI: 0.85– 0.97; P= 0.0065].
Conclusions: Three SNPs from lipid related loci were associated with MI in 5 ethnic groups. The association of two of these SNPs with MI were explained by their influence on Apo B/A levels. These data suggest that the greatest gains in understanding genetic associations of MI may come from understanding how genetic variants are associated with its key intermediate phenotypes.