Being genetically predisposed to higher concentrations of high-density lipoprotein (HDL) cholesterol provides no protective benefits against MI, according to a meta-analysis published online May 17 in Lancet. The results put in question whether drugs designed to raise HDL levels will succeed in reducing the risk of heart attacks.
Numerous studies have shown a causal association between low-density lipoprotein cholesterol (LDL-C) and the risk of MI, and concluded that lowering LDL-C levels also reduces the risk of MI. Those studies include randomized clinical trials, observational studies and a type of observational study that uses an instrumental variable approach known as Mendelian randomization to minimize the potential of confounding. In this approach, the random assignment of a person’s genotype into a carrier or noncarrier group is considered equivalent to randomization in a clinical trial, potentially making causal inferences more robust.
The causal association for HDL has not been clearly established, wrote Sekar Kathiresan, MD, of the Center for Human Genetic Research and Cardiovascular Research Center at Massachusetts General Hospital in Boston, and colleagues. They termed the evidence from clinical trials and observational studies based on the Mendelian approach as “scarce and inconsistent.”
They pointed out that previous HDL studies using Mendelian randomization were hampered by restricted sample sizes, small numbers of single nucleotide polymorphisms (SNPs) and SNPs that affect other lipids as well as HDL. They added that their recent work with lipid-associated SNPs has allowed them to overcome these limitations.
In this meta-analysis, the researchers identified a SNP as affecting solely HDL-cholesterol (HDL-C) without changing other lipid or non-lipid cardiovascular risk factors. Carriers of this SNP, LIPG Asn396Ser, have an inherited increase in HDL-C. They hypothesized these carriers consequently should have a reduced risk of MI.
Kathiresan et al applied two different Mendelian randomizations to test for associations between HDL-C and reduced MI risks. In the first instrumental variable analysis, they used LIPG Asn396Ser as the instrument and tested it in 20 studies that included 20,913 MI cases and 95,407 controls. For the second part of the study, they used a genetic score based on 14 SNPs that had a small, exclusive effect of HDL-C as the instrument and tested them in 12,482 MI cases and 41,331 controls. They further tested the genetic score for 13 SNPs exclusively associated with LDL-C.
The results from the HDL-C analyses showed neither Asn396Ser nor the genetic score of the 14 combined SNPs reduced the risk of MI. The LDL-C results, on the other hand, reinforced previous research that associated LDL-C with MI risk. The researchers wrote that their results challenge the established thinking about HDL-C.
“First, these data question the concept that raising of plasma HDL cholesterol should uniformly translate into reductions in risk of myocardial infarction,” Kathiresan and colleagues wrote. “We raise the strong possibility that a specific means of raising of HDL cholesterol in human beings—namely, inhibition of endothelial lipase—will not reduce risk of myocardial infarction.”
They added that their findings also put in doubt the use of HDL-C as a surrogate measure for MI risk in trials testing interventions.
Based on their findings, they concluded that genetic mechanisms that raise HDL-C may not lower MI risk. “Hence, interventions (lifestyle or pharmacological) that raise plasma HDL cholesterol cannot be assumed ipso facto to lead to a corresponding benefit with respect to risk of myocardial infarction,” they concluded.
In an accompanying editorial Seamus C. Harrison of the Centre for Cardiovascular Genetics at University College London, and colleagues pointed to some possible limitations in the study. “The LIPG variant [Asn396Ser] used here had a fairly large effect on HDL cholesterol, but is rare in the population and so might not represent a strong instrument,” they pointed out. “We should note, however, that the case-control analysis of this variant was adequately powered to detect even a small effect, and the negative association of this variant can be regarded as definitive.”
The editorialists added that the SNPs used in the genetic score also had a small effect size, “which might not translate into a powerful instrument even when