Lipids play a critical role in the progression of CAD. This central role has prompted investigators to examine polymorphisms in genes encoding lipoprotein and enzymes involved in lipid metabolism. Apolipoprotein B, the protein component of low-density lipoprotein (LDL), was a logical initial target of interest. In a small case-control study, the X1 allele of the RFLP defined by the endonuclease XbaI was associated with an increased risk of MI, although it was not associated with differences in LDL levels (96). Unfortunately, subsequent studies failed to validate the association with clinical events (97,98). However, they did demonstrate that LDL levels were 8 mg/dL higher in homozygotes for the variant allele (98).
Apolipoprotein E (apoE) is a component of very low-density lipoproteins and, via binding to receptors on the liver, plays a role in clearing cholesterol from the blood. Three iso-
forms of apoE are known, apoE2, apoE3, and apoE4, coded by the APOE £2, e3, and e4 alleles, respectively. The e4 allele has been associated with higher circulating levels of LDL and an increased risk of CHD in both men (OR: 1.53; p = 0.04) and women (OR: 1.99;p = 0.05) (99). In a meta-analysis of nine studies, using the common e3 allele as the reference, the e2 allele was not associated with an increased risk of CHD (OR: 0.98; 95% CI: 0.85-1.14) but the e4 allele was (OR: 1.26; 95% CI: 1.13-1.41) (100). Initial reports suggested that this risk was dependent on an individual's smoking status (101), but this was not confirmed in a large study (102). In the Scandinavian Simvastatin Survival Study (4S), carriers ofthe e4 allele were at a significantly increased risk of dying (RR: 1.8; 95% CI: 1.1-3.1). Simvastatin treatment appeared to be twice as efficacious in the high-risk £4 carriers (mortality risk reduction to 0.33) than in noncarriers (mortality risk reduction to 0.66), although the formal interaction test was nonsignificant (p = 0.23) (103).
Cholesteryl ester transfer protein (CETP) catalyzes the transfer of cholesteryl esters from high-density lipoprotein (HDL) to LDL, where they can then be taken up by the liver. Whether CETP is pro- or antiatherogenic remains a matter of debate. LDL particles with increased cholesteryl ester content are atherogenic, but the reverse transport of cholesterol from the periphery to the liver is antiatherogenic. Initial studies of the TaqIB RFLP in CETP demonstrated that B1 homozygotes had HDL levels 5 mg/dL higher than did B2 homozygotes, that B1 carriers had increased progression of coronary atherosclerosis, and that pravastatin therapy was more effective in slowing the progression of atherosclerosis in B1 homozygotes (104,105). The association between the TaqIB RFLP and HDL levels was validated in studies from the Framingham Offspring Study and the Veterans Affairs High-Density Lipoprotein Cholesterol Intervention Trial (VA-HIT) (106,107). In both of these studies the B1 variant was associated with an increased risk of CHD, although these associations were no longer significant after adjusting for other cardiac risk factors. Subsequently, the TaqIB RFLP was studied in the Cholesterol and Recurrent Events (CARE) trial (108), in which it was associated with HDL levels but not the risk of cardiovascular end points (109). Two other polymorphisms in CETP have been studied: Ile405Val and Ala373Pro. In a nested case-control study from the Copenhagen Heart Study, the 405Val variant was associated with higher levels of HDL in women and an increased risk of ischemic heart disease in women not treated with hormone replacement therapy (HRT) (110). Also in the Copenhagen Heart Study, the 373Pro variant was associated with lower levels of HDL in both men and women and a reduced risk of ischemic heart disease in women not treated with HRT (111). The possible interaction with HRT may reflect the fact that HRT alters HDL levels and may override any genetic effects.
Examining 148 SNPs in 10 candidate genes related to lipid metabolism, Chasman et al. (112) found two tightly linked SNPs (r2=0.90) in the HMG-CoA reductase gene (HMGCR) that were associated with changes in lipid levels with pravastatin therapy. Individuals heterozygous for the variant (6.7% of the study cohort) had a 22% smaller reduction in total cholesterol and a 19% smaller reduction in LDL. These differences remained significant after correction for multiple testing. Neither SNP was associated with baseline lipid level, itself a predictor for the change seen with statin therapy. Both SNPs are located in introns and are far away from known splicing borders. Thus, these SNPs may be in tight linkage disequilibrium with the true causal variant.
As part of a large genotyping project, we have recently found an Ala210Pro polymorphism in the ADAMTS-1 gene (which encodes a protein with metalloproteinase, disintegrin, and thrombospondin domains) that was significantly associated with both clinical outcomes and the efficacy of statin therapy (113). In 2421 Caucasian males from the CARE trial (108), the rate of death or MI was 9.5, 12.3, and 17.4% in individuals treated with placebo with 0, 1, or 2 copies of the 210Pro allele, respectively. By contrast, in those treated with pravastatin, the rates were 9.5, 9.2, and 4.8%. This translated into a significant treatment interaction between pravastatin therapy and genotype (^interaction = 0.049). These findings were then validated in the West ofScotland Coronary Prevention Study (WOSCOPS) study (114). Figure 2 shows the ORs for death or MI with pravastatin therapy stratified by genotype in the two trials. When the data from the two trials were pooled, the ORs for the reduction in death or MI with pravastatin were 0.77, 0.72, and 0.23 in individuals with 0, 1, or 2 copies of the 210Pro allele, respectively (^interaction = 0.008).
Was this article helpful?