Even Low Levels of ApoB and LDL Linked With CAD, Mortality: UK Biobank

The “dose-response” relationships that apolipoprotein B (apoB) and LDL cholesterol have with coronary artery disease, all-cause mortality, and cardiovascular mortality do not appear to have lower thresholds at which those associations fall apart, according to results of a genetic analysis.

The positive associations with coronary events and mortality held across the natural distribution of genetically determined apoB and LDL cholesterol levels, a finding that further supports the concept of “lower is better” when it comes to reducing the risk of atherosclerotic cardiovascular disease (ASCVD), say investigators.

“The dose-response relationship appears to go all the way to the bottom end of the spectrum,” senior investigator Stephen Burgess, PhD (University of Cambridge, England), told TCTMD. “With triglycerides, there was less of a strong relationship, but still evidence for that kind of continuous relationship.”

The study, which was published this week in JAMA Network Open, was intended to explore the nature of the relationship between apoB, LDL cholesterol, and triglycerides and coronary artery disease and mortality, Burgess explained. While the randomized, controlled trials with statins have unequivocally shown that lowering LDL cholesterol reduces the risk of cardiovascular disease and mortality, later studies with some of the more potent lipid-lowering agents have shown that these agents, such as ezetimibe or PCSK9 inhibitors, reduce the risk of MACE but don’t reduce the risk of cardiovascular mortality.

“The evidence for the dose-response relationship with mortality is less clear,” said Burgess. “That’s something that these kind of natural experiments from genetics can tell us.”   

The study is based on 347,797 participants enrolled in the UK Biobank. As part of the study, the researchers identified 163 single nucleotide polymorphisms (SNPs) for apoB from the UK Biobank patients plus 313 SNPs for LDL cholesterol and 373 SNPs for triglycerides from patients enrolled in the Global Lipids Genetics Consortium. They created genetic risk scores for apoB, LDL cholesterol, and triglycerides.

In univariable mendelian randomization analyses, genetically determined apoB was associated with risk of coronary artery disease and mortality. Each 1-standard deviation (SD) increase in apoB was associated with a 65% higher risk of CAD and 11% and 36% higher risks of all-cause and cardiovascular mortality, respectively. The dose-response relationship appeared larger in males than in females. Genetically predicted apoB was not associated with cancer mortality, but there was a positive association with noncardiovascular/cancer mortality in men and older adults. Similar patterns were observed for LDL cholesterol.

Genetically determined triglyceride levels also were positively associated with CAD and mortality. Each 1-SD increase was associated with a 60% higher risk of CAD, an 8% higher risk of all-cause mortality, a 21% higher risk of cardiovascular mortality, and a 13% higher risk of noncardiovascular/cancer mortality.

On multivariable analysis, the positive associations with apoB, LDL cholesterol, and triglycerides with coronary events and mortality remained. However, after controlling for apoB, the association between triglyceride levels and mortality was not statistically significant.

The researchers also looked at the “shape” of the dose-response curve to understand if the relationship between these lipids and outcomes held even at very low levels. Here, the positive associations between apoB and CAD, all-cause, and cardiovascular mortality were seen across the entire spectrum of apoB levels. The same was true for LDL cholesterol. The association lessened with low levels of triglycerides but remained positive throughout the entire range.

In terms of why some of the more contemporary lipid-lowering studies haven’t demonstrated a reduction in mortality, which contradicts what would be expected based on this genetic analysis, Burgess indicated that it might come down to the relatively short follow-up in those trials and low number of deaths from cardiovascular causes.  

“Obviously, the trials are limited by scope and by timeframe, whereas with these genetic analyses we're looking at a gene that’s been assigned before birth,” he said. “We can [also] use quite large sample sizes.”

Burgess indicated that there is an ongoing debate about whether LDL cholesterol or apoB better predicts cardiovascular risk, but this analysis suggests that the epidemiological relationship between the two lipids and CAD and mortality looks very similar, “even if some of the biochemistry might be more complicated.”