The two LPA polymorphisms studied in this study are some of the most important genetic markers for CAD. In this context, our main finding was that the LPA rs10455872 polymorphism is associated with coronary lesions in Brazilian patients submitted to coronory angiography. On the other hand, no association for the LPA rs3798220 was observed for any of the tested phenotypes.
Studies have reported higher Lp(a) concentration in sub-Saharan African descent and lower Lp(a) concentration in European descent [13, 15, 28, 29]. Regarding the ethnicity, Brazil has one of the most heterogeneous population of the world, composed by a mixture of different ethinic groups, mainly European descent, African descent and Amerindians. In our data, a stratified analysis by race supported a role for the rs10455872 polymorphism independent of ethnic group.
Corroboring with our study, Anderson et al. found that the rs10455872 polymorphism strongly predicted prevalent CAD (per allele OR = 1.43, 95% CI = 1.07-1.91) . Helgadottir et al. showed that patients with CAD carrying LPA risk alleles have increased susceptibility to atherosclerotic manifestations outside of the coronary tree and they are more likely to be diagnosed earlier with CAD than are CAD cases not carrying this variant . Other studies that analyzed the rs10455872 and rs3798220 polymorphisms together reported an increased risk of coronary disease and Lp(a) level that can be explained by these LPA polymorphisms [11, 32]. LPA rs10455872 is an intronic polymorphism associated with short KIV-2 repeat region (kringle IV type 2) which is associated with Lp(a) levels. In the present study, the frequency of the rs10455872 G variant allele was 6.4% for overall, but we observed higher allelic frequency in White compared with non-White groups. The association of the rs10455872 with CAD was significant even in the non-White group which had a small sample size. These data suggest that rs10455872 is also a strong genetic marker for CAD risk in ethnically mixed populations.
For the LPA rs3798220 polymorphism, we did not observe significant association with coronary lesions. Data from other studies also did not support a relationship between this LPA variant and CAD [33, 34]. Furthermore, Anderson et al., studying 1,400 participants with coronary angiography (more than 90% Whites), did not find an association signal between rs3798220 and CAD (OR = 1.47, 95% CI = 0.81-2.67, p = 0.20) . In contrast to our study, the rs3798220 has previously been reported to have an association with the Lp(a) level and the risk of coronary disease [35–37]. LPA rs3798220 results in an aminoacid substitution in the protease domain of LPA, but it can not provide stronger association than rs10455872 which might be representating a more complex group of genetic variants or repeat structures. A possible hypothesis for the lack of this association in the present study could be the lower value of linkage disequilibrium between rs3798220 and rs10455872 identified in the Brazilian patients compared with some studies with patients predominantly from European descent [11, 35–37]. Another hypothesis may be low statistical power, but less likely if the impact of rs3798220 was approximately equal to the impact of rs10455872. However, the exact reason is unclear and other genetic components differently expressed due to ethnicity might be important modulators.
The exact mechanism by which an increased Lp(a) level increases the CAD risk is not fully understood. Pathways modulated by Lp(a) may involve the LDL-cholesterol transport system, the inhibition of the expression of tissue factor, the inhibition of conversion of plasminogen to plasmin, the carriage of pro-inflammatory oxidized phospholipids, and an atherosclerotic stenotic mechanism [30, 38–44]. Some studies reported that the severity of coronary artery disease is associated with Lp(a) levels or LDL concentration [45–48]. Regarding to the Lp(a) level as a risk factor in different ethnic groups, Lp(a) has been associated with risk in European populations , but not unequivocally in African Americans [17, 18]. However, a recent study identified that the increased risk of CVD was at least as strong in African Americans as in White Americans . Another study investigated differential frequencies of LPA polymorphisms in non-Hispanic whites, non-Hispanic blacks, and Mexican Americans . Interestingly, 15 of the 19 polymorphisms tested were strongly associated with Lp(a) levels in at least one subpopulation, six in at least two subpopulations, and none in all three subpopulations. The lack of generalization of associations across ethnicities suggests that specific LPA variants may be contributing to the observed Lp(a) between-population variance. Authors also compared the allele frequencies in HapMap, and observed extremely high correlations (r ≥ 0.99) in allele frequencies between non-Hispanic whites and HapMap CEU (US individuals of northern and western European ancestry) and between non-Hispanic blacks and both HapMap YRI (Yoruba from West Africa) and ASW (individuals with African ancestry from the Southwest USA) .
There are some limitations in our study. First, we did not measure Lp(a) levels and we also did not genotype KIV-2 repeats to check their association with both the LPA polymorphisms and/or the CAD phenotype. Second, we did not assess ancestry through genetic markers; instead, we used a self-declared classification which is commonly applied in Brazil and correlates with genetic ancestry determination. In addition, in our stratified analysis by race, we observed significant association of the rs10455872 with CAD in the White and non-White patient groups. Third, our plaque burden data are derived from institutional records and represent real-life data, as opposed to core-lab derived hemodinamic data. Thus, and despite the greater external validity of our results, we were not able to determine inter- or intra-observed variability estimates. In addition, our choosen method for establishing atherosclerotic burden in the studied patients has relied upon the Gensini Score, which has been shown to highly correlate with this end-point. Other scores could also be used, although they are not as well fitted for quantifing plaque burden. One example is the Syntax score, an angiographic tool for grading the complexity of CAD and designed to better anticipate the risks of percutaneous or surgical revascularization. Finnaly, it is not possible to completely exclude the interaction of the covariates as other genetic markers, use of concomitant drugs, ethnicity, gender and age on our findings [51–54]. Nonetheless, our findings remained after multivariate analysis.