‘Stronger’ Evidence That High Lp(a) Concentrations Cause CVD

High molar concentrations of lipoprotein(a)— not the biomarker’s apolipoprotein(a) size—are responsible for increasing cardiovascular disease risk, confirms a new mendelian randomization study out of Iceland. Further, despite concerns over very low Lp(a) levels potentially causing type 2 diabetes, the results show that a 20% reduction in Lp(a) in people with the highest levels will safely bring them down to the median without conferring this risk.

“Further studies are needed to verify the generalizability of these observations to other populations and to develop risk assessment and therapeutic strategies based on Lp(a) molar concentration rather than apo(a) isoform or sequence variants,” lead author Daniel F. Gudbjartsson, PhD (University of Iceland, Reykjavik), and colleagues write.

The findings have implications for the drugs under development that target high Lp(a) levels.

“The clinical relevance is that for people . . . with very marked elevations of Lp(a), it looks like that the antisense [oligonucleotides] would be beneficial, and even if you knocked the levels down 80% or 90%, you wouldn't actually get into the low levels where you might have some increased risk for diabetes,” Christie Ballantyne, MD (Baylor College of Medicine, Houston, TX), who was not involved in the study, told TCTMD.

For the case-control study, published in the December 17, 2019, issue of the Journal of the American College of Cardiology, the researchers included the genetic information of 143,087 Icelanders, including 17,715 with CAD and 8,734 with type 2 diabetes.

Lp(a) was measured for 12,137 adults and the molar concentrations for the rest of the population were imputed through genetic analysis. Gudbjartsson and colleagues also genetically imputed kringle IV type 2 (KIV-2) repeats, which determine apo(a) size, as well as a splice variant in the LPA gene associated with small apo(a) but low Lp(a) molar concentration to distinguish the relationship between Lp(a) and cardiovascular risk.

They found that Lp(a) molar concentration was associated in a dose-dependent fashion with CAD risk among the 2,930 cases and 8,913 controls who had their Lp(a) directly measured (OR 1.15 per 50 nM; P < 0.0001) and among 14,785 cases and 116,826 controls with genetically imputed Lp(a) concentrations (OR 1.15 per 50 nM; P < 0.0001). Lp(a) molar concentration was additionally linked with peripheral artery disease, aortic valve stenosis, heart failure, and life span. Also, the association between Lp(a) and CAD could be fully explained by the molar concentration of the biomarker with no residual association of apo(a) size.

The 10% of participants with very low Lp(a) molar concentrations (< 3.5nM) did have a greater risk of developing type 2 diabetes (OR 1.44; P < 0.0001), but this risk was independent of molar concentration in those with Lp(a) levels above the median. Those who were homozygous carriers of loss-of-function mutations (cumulative allele frequency of 6.2%) had little or no Lp(a) and an increased risk of diabetes (OR 1.45; P = 0.022) as did those with otherwise very low genetically imputed Lp(a) molar concentrations (OR 1.16; P = 0.0012), “which demonstrated a causal link between very low Lp(a) levels and type 2 diabetes risk,” the authors write.

‘Highly Encouraging’

“The data is highly encouraging,” Ballantyne said, noting that it has been previously established that the concentration of Lp(a) is what really matters.

“Mendelian randomization is increasingly being used to infer causality of established and emerging [atherosclerotic cardiovascular disease] risk factors, to inform clinical trial design, and to set expectations with regard to the outcomes of cardiovascular outcomes trials (CVOTs),” writes Benoit Arsenault, PhD (Université Laval, Québec, Canada), in an accompanying editorial.

“In this regard, the findings of Gudbjartsson et al are of particular importance because, if replicated in other studies with more diverse populations, they could help eliminate one of the hurdles of using Lp(a) as a therapeutic target for residual cardiovascular risk in patients with high Lp(a) levels,” he explains. “However, it must be emphasized that most genetic studies that are used to inform clinical trial design are performed in Caucasians, whereas CVOTs are currently performed in dozens of countries with study participants of different ethnic backgrounds.”

At this point, “the case for absolute Lp(a) concentrations as a causal and ‘druggable’ risk factor for residual cardiovascular risk (and potentially all-cause mortality) is stronger than ever,” Arsenault says. But this hypothesis will “ultimately need to be tested in long-term CVOTs,” of which several are ongoing.

Ballantyne agreed. “We'll have to wait and see what the clinical trial [says about the efficacy] and then what the risk of the therapy is,” he said. “Genetics cannot tell you anything about side effects of the drug. It can tell you that it's a reasonable target, and this paper suggests it is a reasonable target.”

Arsenault points out that, as of now “Lp(a) is unfortunately not routinely measured in most healthcare facilities around the world.” One challenge may be identifying people whose Lp(a) levels are high enough to show an effect in Lp(a)-lowering CVOTs, he observes.

He recommends at least a one-time measurement of Lp(a) in all adults with a family history of cardiovascular disease, as suggested in recent guidelines.

“In the poststatin and postgenomic era, finding much needed therapeutic targets for residual cardiovascular risk can be compared to a gold-digging expedition,” Arsenault concludes. “Like a map to the location of the gold, several genome-wide association studies and mendelian randomization studies are consistently pointing us in the direction of Lp(a). It is time to coordinate our efforts to dig where the map told us, to see once and for all, if we will find the golden target of residual cardiovascular risk that we are hoping for and to give hope to high-risk patients with elevated Lp(a) levels.”