What Level of Lp(a) Indicates Increased CVD Risk?

Lipoprotein a — Lp(a) — is a new focus of interest as an independent cardiovascular (CV) risk factor, with several new drugs that lower this lipoprotein now on the horizon.

Given that Lp(a) levels are largely genetically determined, those with elevated levels are exposed to increased CV risk throughout their lifetimes, so identifying these individuals and lowering their Lp(a) levels will be of the utmost importance. But what threshold levels of Lp(a) should be used to indicate a significantly increased CV risk, and do these thresholds differ for various risk groups?

Findings from a new US registry study are helping to answer these questions. The registry, which represents the largest cohort of Lp(a) thresholds that have been studied in the United States so far, suggests the threshold related to increased risk for CV events may be different for primary and secondary prevention populations.

“In individuals who already have known cardiovascular disease, we found the threshold where they had a higher risk of cardiovascular events was about the 70th percentile, which roughly correlated to 112 nm/L; then the risk starts to plateau,” senior author Ron Blankstein, MD, told theheart.org | Medscape Cardiology.

“In the population who have not yet had a cardiovascular event, there appears to be a more graded relationship between Lp(a) and risk of cardiovascular events, and the true high risk is not seen until Lp(a) levels are at the 90th percentile — that is, about 216 nm/L,” he noted.

Blankstein, who is a preventive cardiologist at Brigham and Women’s Hospital and professor of medicine and radiology at Harvard Medical School, Boston, Massachusetts, believes there is still a lack of knowledge about the importance of Lp(a).

“We may be nearing a time when we have therapies for lowering Lp(a), and we are starting to think about how we should use Lp(a) levels to tell us which patients are at increased risk, and beyond that, which patients to include in future trials or to treat with these medicines,” he said. “It is important that the medical community and the public understand what this lipoprotein is, how it increases cardiovascular risk, and who is affected. Our data will help with that.”

These latest findings on Lp(a) levels and CV risk are reported in a paper published online on February 26 in the Journal of the American College of Cardiology.

The registry study included 16,419 individuals from two medical centers in Boston, Massachusetts, who had their Lp(a) level measured between 2000 and 2019 as part of routine care. Of these individuals, 62% already had CV disease (secondary prevention patients), and 38% did not have baseline CV disease (primary prevention population).

CV events (CV death, myocardial infarction, stroke, or revascularization) were recorded over a median follow-up of 12 years.

“In terms of the Lp(a) threshold of where risk really starts to increase, we found quite different results for the secondary prevention and primary prevention populations,” Blankstein said.

Results showed that among the 10,181 patients with baseline CV disease, those in the 71st-90th percentile group for Lp(a) levels had a 21% increased hazard of CV events compared with the reference group — those with Lp(a) levels below the 50th percentile, with an adjusted hazard ratio (HR) of 1.21 (P < .001). This was similar to that of individuals in the 91st-100th group (adjusted HR, 1.26; P < .001).

Among the 6238 individuals without established CV disease, there was a continuously higher hazard of CV events with increasing Lp(a), and individuals in the 91st-100th Lp(a) percentile group had the highest relative risk with an adjusted HR of 1.93 (P < .001).

In terms of absolute risk, the secondary prevention population is at a much higher risk for events regardless of whether they have raised Lp(a), Blankstein pointed out.

In individuals who do not have prior CV disease, those at 90th percentile of Lp(a) or above had an absolute annual event rate of 2.2% vs 1.1% for those with Lp(a) below the 50th percentile.

But in patients with prior CV disease, even the reference group of Lp(a) < 50th percentile had an annual event rate of 4.1%. With Lp(a) levels above the 70th percentile, this increased to 5.3%.

“However, in the primary prevention population, the relative risk of a cardiovascular event imposed because of a high level of Lp(a) — above the 90th percentile — appears to be greater than in the secondary prevention population,” Blankstein noted.

Implications for Future Trials

He noted that the Lp(a) level where risk is significantly increased in the secondary prevention population (above 112 nm/L) is quite a bit lower than the levels used to include patients in the two ongoing phase 3 trials of Lp(a) lowering medications: 175 nm/L in the HORIZON trial and 200 nm/L in the OCEAN(a)-OUTCOMES trial.

“So, these observations suggest that if those trials find a benefit of lowering Lp(a) in the population studied, there will be many more individuals with lower levels of Lp(a) who may also benefit from having their levels lowered.”

Blankstein acknowledged that these first phase 3 trials needed to select patients with very high Lp(a) levels at baseline to try and prove the Lp(a) hypothesis — that lowering Lp(a) will lower events.

“In that way, I think the thresholds chosen are appropriate. But our results suggest that if those trials do show a benefit of lowering Lp(a), then future trials in secondary prevention may involve a lower threshold.”

With regard to future trials of Lp(a) lowering in the primary prevention population, Blankstein said these results suggest that it will be important to pick a relatively high threshold (above the 90th percentile or 200 nm/L) for enrollment.

“The primary prevention population is much larger than the secondary prevention population — there are a lot more people out there living with high Lp(a) who have never had an event. Ultimately, our goal should be to prevent events before they happen.

If we can show that Lp(a)-lowering drugs reduce cardiovascular events in a primary prevention population, then our data would suggest that about 10% of the population could be eligible and could benefit from such treatment.”

Support for Universal Screening

Given that Lp(a) levels are genetically determined, Blankstein believes it is reasonable that everyone should have their levels checked at least once.

“Lp(a) has been under the radar. Most people will not be aware of it, and even physicians often don’t think it’s important to test as there isn’t a specific therapy to lower it yet. But I think it’s important to know if you have raised levels,” he said. “It can help us do better risk stratification, and hopefully in the not-too-distant future there will be therapies to lower it.”

In an editorial accompanying the publication in JACC, Nathan D. Wong, MD, from the Heart Disease Prevention Program at the University of California, Irvine, California, said these new data extend the understanding of Lp(a) and CV disease risk thresholds and add support for universal Lp(a) screening.

“The failure to screen and identify those with Lp(a)-associated risks represents a missed opportunity to address this risk, not only with our existing repertoire of treatments but hopefully in the future with the development of promising therapies targeting Lp(a),” he concluded.

This study was funded, in part, by Amgen. Blankstein has received research support and consulting fees from Amgen Inc. and Novartis.