Unlocking Lp(a): Baseline Levels Matter, but So Too Does Absolute Reduction

New FOURIER data shed some light on this little understood risk factor, while a mendelian analysis underscores why only big drops yield statistical benefit.

Unlocking Lp(a): Baseline Levels Matter, but So Too Does Absolute Reduction

LISBON, Portugal—New data from the FOURIER trial show that patients with higher baseline lipoprotein(a) levels may be more likely to derive benefit from PCSK9 inhibition.

Michelle O’Donoghue, MD (Brigham and Women’s Hospital, Boston, MA), who presented the subanalysis during a late-breaking clinical study session here yesterday at the 2018 European Atherosclerosis Society meeting, told TCTMD that “people are very interested to try and get a better handle on which patients may derive enhanced benefit from a drug like evolocumab.”

Lp(a), a marker of risk that’s becoming more well-known but is still incompletely understood, is “also a risk factor,” according to O’Donoghue. “It is sort of an appealing concept to think that the baseline concentration may identify people who derive enhanced benefit.”

As previously reported in the main FOURIER findings, evolocumab (Repatha; Amgen) reduced Lp(a) concentrations in patients by 26.9% from baseline through 48 weeks. The new data show that when patients are stratified into quartiles by baseline Lp(a) level, those who start out with the highest levels seem to see the largest reductions in this marker with treatment.

Median Shifts in Lp(a) With Evolocumab by Quartile of Baseline Level

 

Quartile 1

Quartile 2

Quartile 3

Quartile 4

Absolute Change, nmol/L

-1

-9

-24

-36

Percent Change

-9.1%

-41.9%

-32.5%

-16.0%

 

Importantly, those with higher baseline Lp(a) concentrations also seemed to derive the most protection against adverse events from evolocumab. In patients with baseline Lp(a) levels above the median, the rate of the primary endpoint (CV death, MI, or stroke) was substantially smaller with evolocumab compared with placebo (8.17% vs 10.97%; HR 0.76; 95% CI 0.66-0.86), with a number needed to treat of only 36. By contrast, for patients with baseline Lp(a) levels at or below the median at baseline, the difference in the rate of the primary endpoint between groups was attenuated at 3 years, although still lower with evolocumab than with placebo (7.48% vs 8.74%; HR 0.85; 95% CI 0.73-0.97). Here, the number needed to treat was 79.

Additionally, patients who achieved the lowest levels of both Lp(a) and LDL cholesterol saw a greater reduction in primary endpoint event risk beyond week 12 compared with those who finished with the highest Lp(a) and LDL cholesterol levels (6.57% vs 9.43%; P < 0.001).

Defining a Meaningful Benefit

Referring to a presentation from the first day of the meeting given by Brian Ference, MD (University of Cambridge, England)—a mendelian randomization study that showed an absolute reduction of 100 mg/dL in Lp(a) would be required in order to have about a 20% relative risk reduction in clinical events—O’Donoghue said: “It may just be that the existing therapies that we have to lower Lp(a) concentration don't lower it sufficiently enough to be translating into meaningful clinical benefit.”

The next question, she continued, “is do the Lp(a)-lowering properties of evolocumab translate greater benefit than might be expected from LDL-lowering alone? There we can't be quite as sure.”

For his part, Ference told TCTMD that “without question, there's compelling evidence that Lp(a) is causally related to the risk of cardiovascular disease.” Now, what is “really pertinent,” he added, “is how much do we have to lower Lp(a) to reduce clinical events in a short-term trial?”

The effect of Lp(a) on the risk of cardiovascular events is clinically proportional to the absolute reduction in Lp(a), not the percent reduction. Brian Ference

O’Donoghue’s findings showed “no quantitative evidence to suggest that there was an additional benefit from the Lp(a) lowering itself” compared with the benefit due to LDL-cholesterol lowering, Ference said. Specifically, he explained, even a 16% lowering of Lp(a) in the highest quartile of patients only translates to a 13-14 mg/dL absolute reduction.

“Based on the genetic evidence, we wouldn't expect that to have very much effect at all—maybe as much as a 2-4% reduction in additional risk, which really would be too small to capture within a short-term randomized trial,” he commented.

Ference said his study shows “for the first time that the effect of Lp(a) on the risk of cardiovascular events is clinically proportional to the absolute reduction in Lp(a), not the percent reduction. And so one cannot say that a 30% Lp(a) reduction or a 50% reduction should be beneficial.” Even a 90% reduction may not be “useful because it will depend entirely on what somebody's baseline Lp(a) level is,” he added. “That's particularly important for Lp(a) because it has such an asymmetrical distribution that the Lp(a) value can vary by as much as 1,000-fold in the same population.”

The genetic findings should inform the design of future pivotal cardiovascular outcomes trials of new drugs in development to specifically reduce Lp(a), like antisense oligonucleotides and RNA inhibitors, Ference said. Researchers need to “make sure we don't incorrectly conclude that Lp(a) doesn't reduce the risk of events because we didn't reduce Lp(a) enough in the trials to translate into a clinically meaningful effect,” he cautioned, suggesting that the median Lp(a) level of all patients enrolled in these trials should be 100 mg/dL. With that, Ference said, “one can reasonably expect a 15-20% reduction [in clinical events] in a short-term trial, which would be good.”

A Call for ‘Worldwide Screening’

Assuming promising results are seen with new drugs in development that specifically target Lp(a), O’Donoghue said, “it makes sense to think about how can we identify people who are going to benefit the most,” although cost will be an obvious concern. However, she noted, “we know there's a lot of people who have low LDL-cholesterol concentrations who go on to have recurring cardiovascular events, and right now we throw up our hands a little bit and say, ‘Well, we don't have really much more to offer you.’”

If Lp(a) became an additional target, “that would be appealing,” she added.

Session co-chair Eran Leitersdorf, MD (Hadassah Medical Center, Jerusalem, Israel), told TCTMD that these findings give “the whole field a boost in importance to move forward.”

We know there's a lot of people who have low LDL-cholesterol concentrations who go on to have recurring cardiovascular events, and right now we throw up our hands a little bit and say, ‘Well, we don't have really much more to offer you.’ Michelle O’Donoghue

As for where Lp(a) testing stands for now, Ference said, “if somebody has early aggressive atherosclerosis and they don't have other risk factors, it's important to check for Lp(a).” Another finding from his genetic study was that an additional 100 mg/dL of Lp(a) has the same effect as 1 mmol of LDL cholesterol. “So if you have 200 mg or higher of Lp(a), you have the equivalent effect of 2 mmol higher of LDL and a fourfold increase in lifetime risk of cardiovascular disease,” he explained, adding that this is equivalent to what is seen with heterozygous familial hypercholesterolemia.

Because so much of what leads to high Lp(a) levels is known to be inherited, this could represent “a new inherited cardiovascular lipid disorder that has the same lifetime risk as heterozygous familial hypercholesterolemia, but is probably twofold more prevalent,” he continued.

O’Donoghue said she has seen an uptick in patients asking to have their Lp(a) tested, partially spurred on by coverage in the media, most notably a recent feature in the New York Times.

These data should ultimately “stimulate the conversation for people to begin contemplating worldwide screening for elevated Lp(a) levels to detect people who are at really extreme risk,” Ference said.

Sources
  • O’Donoghue M. Lipoprotein(a), PCSK9 inhibition and cardiovascular risk: insights from the FOURIER trial. Presented at: EAS 2018. May 7, 2018. Lisbon, Portugal.

  • Ference BA. LP(a) variants, risk of coronary disease, and estimated clinical benefit of lipoprotein(a) lowering therapies: a mendelian randomized analysis. Presented at: EAS 2018. May 6, 2018. Lisbon, Portugal.

Disclosures
  • O’Donoghue reports receiving grant support from Amgen.
  • Ference reports receiving research grants from Merck, Novartis, Amgen, Esperion Therapeutics, and Ionis Pharmaceuticals and fees associated with consulting or serving on advisory boards from Merck, Amgen, Pfizer, Regeneron, Sanofi, Ionis Pharmaceuticals, dalCOR, The Medicines Co, CiVi Pharma, KrKa Phamaceuticals, Medtronic, Celera, Quest Diagnostics, American College of Cardiology, and the European Atherosclerosis Society.

We Recommend

Comments