Inflammation Plays Into the ASCVD Risk Associated With Lp(a): ACCELERATE

The risk of major adverse cardiovascular events resulting from elevated levels of lipoprotein(a) is mediated by the degree of residual systemic inflammation, suggests a new analysis from the ACCELERATE randomized trial.

In patients at high risk for atherosclerotic cardiovascular disease (ASCVD), increased levels of Lp(a) were significantly associated with a higher risk of cardiovascular death, MI, and stroke among those with elevated high-sensitivity C-reactive protein (hsCRP) concentrations. In contrast, for patients without evidence of systemic inflammation, it didn’t matter how high Lp(a) levels were at baseline or on treatment—there was no increased risk of cardiovascular events in patients with elevated Lp(a) compared with those with low levels.

Rishi Puri, MBBS, PhD (Cleveland Clinic, OH), who led the secondary analysis of ACCELERATE, published online July 8, 2020, in JAMA Cardiology, said the study suggests there is a link between Lp(a) and inflammation, at least in this high-risk population.

“Across a range of analyses, high-sensitivity CRP levels regulated the Lp(a)-associated cardiovascular risk,” he told TCTMD. “In other words, if your CRP was less than 2.0 mg/L, irrespective of whether you fell into the highest quintile of on-treatment Lp(a), which were very high levels, your relative risk for a cardiac event in the context of being on the best available preventive therapies, was not significantly higher than if you were in the first quintile of Lp(a), or effectively had a normal level of Lp(a).”

The interdependence of these two known mediators of ASCVD risk—Lp(a) and hsCRP—might help identify patients who will benefit from novel Lp(a)-lowering therapies should these drugs be approved by regulators. “You can potentially start to bring these patients in to have these therapies at lower Lp(a) levels under the proviso that their high-sensitivity CRP level is greater than 2.0 mg/dL,” said Puri. “If CRP is less than 2.0 mg/L, you may not derive as much of a benefit—the bang for the buck—with these Lp(a)-lowering therapies.”

Guillaume Paré, MD (McMaster University, Hamilton, Canada), who wasn’t involved in the research, said the concept that Lp(a) is intertwined with systemic inflammation has been suggested in previous studies, but praised the ACCELERATE investigators for clearly showing a synergy between the two markers of ASCVD risk.

“It’s a great discovery, but we’ll need a little bit more before we can truly understand what’s happening here,” he said. “I’d also be curious to see if the finding extends to a community sample instead of a very-high-risk population.”

He raised the possibility of the “index of event bias,” which can occur in studies that include patients who are highly selected for the occurrence of an event. In certain settings, for example, studies have shown that obesity may be protective against secondary events in patients with heart disease, which may be a statistical artifact rather than a true finding. In ACCELERATE, investigators enrolled patients with high-risk vascular disease, such as those with ACS within the previous 30 to 365 days, cerebrovascular disease, PAD, or type 2 diabetes with CAD.

“Biologically, the findings are very interesting,” said Paré, referring to the new analysis, “but how did the patient selection influence the results?”

If CRP is less than 2.0 mg/L, you may not derive as much of a benefit—the bang for the buck—with these Lp(a)-lowering therapies. Rishi Puri

To TCTMD, Puri called the new findings provocative and hypothesis-generating, but also stressed the need for further study, including in lower-risk patient populations. While Lp(a) is a particularly hot target for ASCVD risk reduction, no clinical trials have shown that reducing it lowers the risk of clinical events. The first large-scale randomized clinical trial testing a Lp(a)-lowering antisense oligonucleotide just launched in late 2019 and that trial, known as HORIZON and led by Steven Nissen, MD (Cleveland Clinic), and colleagues, won’t be completed for at least 4 years.

Findings First Observed in SATURN

Halted for futility in 2015, the ACCELERATE trial was part of a group of large-scale randomized trials that failed to identify a clinical benefit in patients treated with cholesteryl ester transfer protein (CETP) inhibitors. The study included 12,092 patients with high-risk vascular disease who received evacetrapib (Eli Lilly) or placebo in addition to optimal medical therapy, including maximally tolerated statins. The primary endpoint was the first occurrence of all-cause mortality, MI, stroke, revascularization, or hospitalization for unstable angina.

To TCTMD, Puri said the present analysis stemmed from findings that emerged in the SATURN trial looking at plaque progression/regression in patients treated with high-intensity rosuvastatin or atorvastatin. In SATURN, Lp(a) levels were not associated with coronary plaque regression or progression, and although the trial was not designed to assess clinical endpoints, it found only those with high Lp(a) and high hsCRP levels had a higher risk of major adverse cardiovascular events.    

In this ACCELERATE analysis, they focused on a stricter endpoint—cardiovascular death, MI, or stroke—and found that when patients were stratified by hsCRP levels, there was an association with elevated Lp(a) levels and MACE among those with high hsCRP concentrations. For example, those with Lp(a) levels in the top quintile had a 70% higher risk of MACE compared with those with the lowest levels. Similarly, those in the third and fourth quintiles had 42% and 50% higher risks of MACE, respectively, compared with those in the bottom quintile. This association was not observed in patients with hsCRP levels less than 2.0 mg/L. In a sensitivity analysis that focused only on placebo-treated patients and eliminated any potential confounding effects of evacetrapib on outcomes, the relationship between MACE and high Lp(a) in those with elevated hsCRP was even stronger.

To TCTMD, Puri said the cardiology community was “spooked” somewhat by a recent analysis from Brian Ference, MD (University of Cambridge, England), suggesting that doctors would need to “take a sledgehammer” to Lp(a) to achieve any clinical benefit. By Ference’s estimate, a reduction in the range of 80 to 100 mg/dL would be needed to achieve a therapeutic effect akin to a 40-mg/dL reduction in LDL cholesterol. Using data from the UK Biobank, only patients in the 95th to 99th percentile would have sufficiently high enough levels of Lp(a) to warrant treatment, according to Ference.

Intervening in Those Mostly Likely to Benefit

Puri said his team’s findings might mean that physicians could intervene at lower levels of Lp(a) in patients with evidence of systemic inflammation, although that remains to be proven. It could also enhance uptake of the drugs, which may be quite expensive, if studies show that those with high Lp(a) and high hsCRP derive the largest benefit from treatment. 

“We’ve seen some of the economic and uptake challenges of the PCSK9 inhibitors, which we know work,” said Puri, noting that the funding, reimbursement, and cost-effectiveness of those drugs have been under scrutiny since their approval. “The same might hold true for Lp(a) molecules. There might be a marginal benefit overall, but if we can find groups of patients who benefit more at potentially lower Lp(a) levels, then we can certainly target these groups more effectively. It’s not only a clinical improvement, but it’s also a cost-effective improvement as well.”

We’ll need a little bit more before we can truly understand what’s happening here. Guillaume Paré    

Paré pointed out that in the overall analysis not stratified by hsCRP levels, patients with Lp(a) levels ≥ 22.9 nmol/L (the median) did not have an increased risk of MACE events compared with those with Lp(a) levels less than the median. In 2017, a study published in the Lancet Diabetes and Endocrinology failed to show an association between Lp(a) concentrations or genetic variants for LPA and all-cause or cardiovascular mortality. That paper, said Paré, surprised a lot of proponents of Lp(a) as a treatment target for reducing residual ASCVD risk but lines up with the present analysis.

“Overall, there was no association with Lp(a), but the association does emerge once they stratified by levels of inflammation,” he said. That analysis raises the possibility, as noted by the researchers, that Lp(a)-lowering therapies might best be used in patients with elevated Lp(a) and high CRP, particularly if the still-investigational drugs turn out to be quite expensive. “It gives us a strong rationale that we should also measure inflammation in the upcoming phase III trials of Lp(a) inhibitors to really address the question.”

Paré believes Lp(a) is a significant risk factor for ASCVD, noting that in his clinic he encounters patients with elevated Lp(a) concentrations—but no other cardiovascular risk factors—and an extensive family history of heart disease. “I’m very optimistic that Lp(a)-lowering drugs will benefit patients,” he said. “The key will be to target the right patients for these drugs, and inflammation might very well be part of the equation.”