THE MODERN BIOHEALTH, WITH PERSPECTIVE.
Clinician-led care in Florida

JOURNAL / HEART HEALTH & BIOMARKERS

Pelacarsen Cut Lp(a) 78%. Cardiovascular Events Did Not Fall.

In more than 8,300 people with established cardiovascular disease, pelacarsen dramatically lowered an inherited risk marker—but major cardiovascular events were virtually identical to placebo. The result challenges the simplest version of the Lp(a) hypothesis without making Lp(a) irrelevant.November 16, 2026 · 11 minute read
Editorial image accompanying Pelacarsen Cut Lp(a) 78%. Cardiovascular Events Did Not Fall.
Editorial illustration

Lipoprotein(a) has become one of preventive cardiology’s most closely watched biomarkers.

High levels are largely inherited. They are associated with heart attacks, strokes and calcific aortic-valve disease. Unlike LDL cholesterol, Lp(a) usually changes little with diet, exercise or statin therapy.

That made pelacarsen an unusually important test of a straightforward idea: if high Lp(a) contributes to cardiovascular disease, would substantially lowering it prevent cardiovascular events?

The first large outcomes trial has now delivered an uncomfortable answer.

In Lp(a)HORIZON, pelacarsen reduced Lp(a) by 78.4% after 12 months. But over a median follow-up of 4.7 years, major cardiovascular events occurred in 12.3% of participants assigned to pelacarsen and 12.1% assigned to placebo.

The hazard ratio was 1.02, with a 95% confidence interval from 0.90 to 1.15. There was no evidence of benefit and not even a favorable numerical trend.

The phase 3 results were presented November 7 at the American Heart Association Scientific Sessions and published simultaneously in The New England Journal of Medicine.

This was not a trial where the drug failed to move its target. It moved the target dramatically.

The outcome simply did not follow.

What happened

Lp(a)HORIZON enrolled adults ages 18 to 80 who had established atherosclerotic cardiovascular disease and an Lp(a) level of at least 70 mg/dL.

Established disease meant a previous heart attack, ischemic stroke or symptomatic peripheral artery disease. The qualifying event had to have occurred more than three months but less than 10 years before enrollment.

Participants entered a 30-to-90-day period during which cardiovascular risk-factor treatment could be adjusted to meet local secondary-prevention guidelines. They were then randomly assigned to receive either:

  • Pelacarsen, 80 mg by subcutaneous injection once monthly
  • A matching placebo

Pelacarsen is an antisense oligonucleotide designed to reduce hepatic production of apolipoprotein(a), a necessary component of the Lp(a) particle. It remains investigational; the trial was intended to determine whether its laboratory effect translated into fewer cardiovascular events.

A total of 8,350 people were randomized. The prespecified efficacy analysis included 8,322 participants: 4,157 assigned to pelacarsen and 4,165 to placebo. The AHA presentation materials rounded this as more than 8,300 participants; the NEJM publication provides the exact randomized and analysis populations.

The mean age was 59.7 years, and 27% were women. At baseline:

  • 76.6% had experienced a previous myocardial infarction.
  • 7% had experienced an ischemic stroke.
  • 10.3% had symptomatic peripheral artery disease.
  • 5.7% had a history of more than one qualifying event.
  • 25.8% to 26.9% had type 2 diabetes.
  • Approximately half had a history of premature cardiovascular disease.

The median starting Lp(a) was 235.8 nmol/L, or approximately 108 mg/dL—well above the trial’s minimum entry requirement.

These were already intensively treated patients

HORIZON was not a trial of people whose other cardiovascular risks were being ignored.

At baseline:

  • 77.6% were receiving high-intensity statin therapy.
  • 56.8% were using ezetimibe or a bile-acid sequestrant.
  • 11.1% were using a PCSK9 inhibitor.
  • 81% were taking aspirin.
  • Median LDL cholesterol was 64.6 mg/dL.
  • Median apoB was approximately 73 to 74 mg/dL.
  • Mean systolic blood pressure was approximately 125 mm Hg.

That treatment context is critical. Pelacarsen was being tested on top of aggressive secondary prevention—not as an alternative to lowering LDL or controlling blood pressure.

What the data actually show

The drug’s effect on Lp(a) was large and sustained.

After 12 months, the estimated Lp(a) level was:

  • 64.4 nmol/L with pelacarsen
  • 233.4 nmol/L with placebo

That represented a 78.4% reduction from baseline in the pelacarsen group and a 3.2% reduction in the placebo group.

Across the full trial, the time-averaged change from baseline was a 73.9% reduction with pelacarsen and an 8.3% reduction with placebo. The between-group difference in time-averaged percentage change was 65.6 percentage points.

But the primary clinical outcome did not improve.

OutcomePelacarsenPlacebo
Primary cardiovascular event511 of 4,157 (12.3%)503 of 4,165 (12.1%)
Absolute difference+0.2 percentage points—
Hazard ratio1.0295% CI 0.90–1.15
P value0.76—

The primary outcome combined cardiovascular death, nonfatal heart attack, nonfatal ischemic stroke and urgent coronary revascularization requiring hospitalization.

The result remained neutral in the prespecified group with even higher starting Lp(a)—at least 90 mg/dL:

  • 402 of 3,266 pelacarsen participants had an event: 12.3%.
  • 403 of 3,274 placebo participants had an event: 12.3%.
  • Hazard ratio: 1.00.
  • 95% confidence interval: 0.87 to 1.15.
  • P value: 0.99.

Narrower cardiovascular composites were also neutral. Death from any cause occurred in 5.7% of the pelacarsen group and 5.9% of the placebo group. None of the individual components produced a convincing signal of benefit.

The safety and tolerability tradeoff was real

Overall serious adverse events were reported in 37.3% of pelacarsen recipients and 41.1% of placebo recipients. But several treatment-related problems were substantially more common with pelacarsen:

  • Constitutional or flu-like symptoms: 31.9% versus 5.8%
  • Injection-site reactions: 25.6% versus 2.9%
  • Investigator-reported hypersensitivity: 17.1% versus 5.4%
  • Adverse events leading to permanent treatment discontinuation: 20.6% versus 7.5%

Nineteen pelacarsen recipients had reactions adjudicated as definite or probable anaphylaxis, and eight hypersensitivity reactions resulted in hospitalization. None were fatal.

Premature discontinuation of the assigned regimen occurred in 35.1% of pelacarsen participants and 18.3% of placebo participants.

That difference matters when interpreting a long-term injectable treatment. The trial still produced a large time-averaged Lp(a) reduction, but the tolerability profile could have reduced cumulative exposure for some participants.

Why this matters for the Lp(a) hypothesis

The evidence linking Lp(a) to cardiovascular risk did not begin with drug trials.

Large observational studies show that people with higher Lp(a) experience more cardiovascular disease. Genetic studies support a causal relationship: people born with genetic variants associated with higher lifelong Lp(a) exposure tend to have higher cardiovascular risk.

The 2026 ACC/AHA dyslipidemia guideline treats Lp(a) as a risk-enhancing factor. It estimates that 125 nmol/L, approximately 50 mg/dL, is associated with about 1.4 times the cardiovascular risk seen near the population median. Risk rises progressively at higher levels.

HORIZON does not erase that evidence. It does expose a gap between it and the therapeutic assumption that lowering Lp(a) later in life must reduce events within several years.

There are several possible explanations.

Lifelong exposure may not equal late intervention

Genetic studies approximate the effect of carrying a higher or lower Lp(a) level over decades. HORIZON lowered Lp(a) for a median of 4.7 years in people who already had cardiovascular disease.

Lowering a causal exposure after plaques have developed may not reproduce the benefit associated with having a lower exposure throughout life.

That is a plausible explanation, but it is not proof that a longer trial would have been positive. The event curves did not show a delayed favorable separation within the follow-up available.

Pelacarsen may not represent every way of lowering Lp(a)

Pelacarsen is an antisense oligonucleotide. Other investigational drugs use small-interfering RNA or block the assembly of Lp(a) particles by a different mechanism.

Some have produced larger or more durable reductions in early trials. They may also have different adherence and tolerability profiles.

That means HORIZON is definitive for this pelacarsen regimen in this population. It is not automatically definitive for every Lp(a)-lowering drug.

Very good background care may have narrowed the opportunity

Median LDL cholesterol was already approximately 65 mg/dL, mean systolic pressure was about 125 mm Hg and most participants were taking intensive preventive therapy.

When several major risks are tightly controlled, demonstrating additional benefit from modifying one residual risk factor becomes harder.

But that explanation should not be used too casually. More than 12% of participants still had a major cardiovascular event, leaving substantial residual risk to prevent.

The biomarker may not be a simple treatment target

Lp(a) may capture several related biological properties: an apoB-containing atherogenic particle, oxidized phospholipids, inflammatory activity and potentially prothrombotic effects.

A measured concentration may predict risk without every method of reducing that concentration producing the same biological effect.

The trial therefore challenges the simplest claim—“lower Lp(a), lower events”—more than it challenges the broader conclusion that high Lp(a) identifies risk.

What happens to the other Lp(a) drugs?

Several programs remain active, and HORIZON raises the evidentiary bar for all of them.

Olpasiran is a small-interfering RNA drug. The phase 3 OCEAN(a)-Outcomes trial enrolled 7,297 people with established cardiovascular disease and elevated Lp(a). The registry lists it as active but no longer recruiting, with primary completion estimated for March 2028.

Lepodisiran is another long-duration small-interfering RNA therapy. The phase 3 ACCLAIM-Lp(a) outcomes trial is enrolling approximately 17,300 participants with Lp(a) of at least 175 nmol/L. It includes people with established cardiovascular disease and selected people at high risk for a first event.

Muvalaplin is an oral small molecule that interferes with the binding of apolipoprotein(a) to apoB, reducing assembly of the Lp(a) particle. The phase 3 MOVE-Lp(a) outcomes trial is enrolling approximately 10,450 people and is expected to continue into 2031.

Zerlasiran is a small-interfering RNA therapy with completed phase 2 biomarker data. In a randomized trial of 178 participants with stable cardiovascular disease, three regimens produced placebo-adjusted time-averaged Lp(a) reductions of approximately 81% to 86% through 36 weeks. That study was designed to measure Lp(a), not cardiovascular outcomes; a phase 3 outcomes result has not been established.

These drugs differ in mechanism, dosing frequency, magnitude of reduction, patient population and trial design. One negative trial does not predetermine every result.

It does mean that lowering Lp(a) on a laboratory report can no longer be presented as a reliable proxy for preventing cardiovascular events. Each program now has to prove clinical benefit directly.

What this result does not mean

It does not mean Lp(a) is harmless

High Lp(a) remains associated with higher lifetime risk of atherosclerotic cardiovascular disease and calcific aortic-valve disease. The genetic and epidemiologic evidence remains substantial.

It does not mean testing is useless

An Lp(a) result can change a person’s estimated risk even when no medication specifically approved to lower Lp(a) exists.

The 2026 ACC/AHA guideline recommends measuring Lp(a) at least once in every adult for cardiovascular risk assessment. Because levels are predominantly genetically determined and usually remain relatively stable, repeated testing is generally unnecessary unless there is a specific clinical reason. Fasting is not required.

It does not prove that no Lp(a)-lowering treatment can work

HORIZON tested one drug, one dose, one mechanism and a secondary-prevention population over a defined period.

The result makes benefit less certain. It does not settle whether a different mechanism, greater reduction, earlier intervention or longer exposure could produce a different outcome.

It does not make modifiable risk factors less important

A high inherited risk marker is a reason to take controllable risks more seriously—not a reason to conclude that cardiovascular disease is predetermined.

What should a reader do now?

Get Lp(a) measured once

If you have never had it measured, ask whether Lp(a) can be added to your next lipid evaluation.

Results may be reported in either nmol/L or mg/dL. These units describe different properties, and there is no universally reliable fixed conversion between them. It is better to interpret the result in the units reported by the laboratory than to use a simple online conversion.

The 2026 guideline considers 125 nmol/L, or approximately 50 mg/dL, a risk-enhancing level. Risk is continuous, however; this is not a line separating “safe” from “dangerous.”

If Lp(a) is high, first-degree relatives may also benefit from being tested because the level is largely inherited.

Use high Lp(a) to sharpen the rest of the prevention plan

Until a targeted therapy proves that it prevents events, the practical response to high Lp(a) is to reduce the risks that can be changed.

That may include greater attention to:

  • LDL cholesterol
  • Non-HDL cholesterol
  • ApoB, particularly when LDL and particle burden may be discordant
  • Blood pressure
  • Smoking or nicotine exposure
  • Diabetes and glucose control
  • Physical activity and cardiorespiratory fitness
  • Dietary pattern
  • Kidney function
  • Body composition and waist circumference
  • Sleep and sleep apnea when clinically relevant

The 2026 ACC/AHA guideline specifically states that elevated Lp(a) should favor more intensive LDL-lowering and management of other cardiovascular risks. Statins generally do not lower Lp(a), but they reduce LDL-related risk—which may be especially valuable when inherited Lp(a) risk is also present.

Do not start aspirin or alter prescribed lipid therapy solely because of an Lp(a) result without discussing the balance of benefits and bleeding or medication risks with a clinician.

Questions to ask your clinician

  • Have I had Lp(a) measured, and what units did the laboratory use?
  • How does the result change my overall cardiovascular-risk estimate?
  • Is my LDL cholesterol appropriate for my risk level?
  • Would apoB or non-HDL cholesterol reveal residual particle-related risk not apparent from LDL alone?
  • Is my blood pressure controlled across home and office readings?
  • Do my personal or family history suggest testing first-degree relatives?
  • Would coronary artery calcium imaging or another risk-assessment tool meaningfully change management?
  • If I already have cardiovascular disease, are all evidence-based secondary-prevention therapies optimized?
  • Are there clinical trials for which I might qualify, and what is known versus still unproven about the investigational therapy?

The practical takeaway

Lp(a)HORIZON is the kind of result evidence-based medicine needs, even when it disappoints.

Pelacarsen did exactly what it was designed to do at the biomarker level. Lp(a) fell by nearly 80%. Yet cardiovascular events were essentially unchanged: 12.3% with pelacarsen and 12.1% with placebo.

The trial does not make Lp(a) meaningless. It makes the distinction between risk marker, causal factor and proven treatment target more important.

Measure Lp(a) once. Use a high result to understand inherited risk and intensify attention to LDL, apoB, blood pressure and other modifiable factors.

But do not confuse a dramatic change in a laboratory value with a demonstrated improvement in health. HORIZON is a reminder that outcomes—not biomarkers—are the final test.

Sources

  • Cho L, Nicholls SJ, Frndak S, et al. Pelacarsen and Cardiovascular Outcomes in Patients with High Lipoprotein(a). New England Journal of Medicine. Published November 7, 2026. DOI: 10.1056/NEJMoa2610159
  • American Heart Association. Lp(a) HORIZON trial: Lp(a) fell by 78%, but cardiovascular outcomes unexpectedly neutral. November 7, 2026. AHA presentation summary
  • American Heart Association. Lp(a)HORIZON data summary. Presentation PDF
  • ClinicalTrials.gov. Lp(a)HORIZON, NCT04023552. Trial record
  • Novartis. Lp(a)HORIZON Phase III topline results for pelacarsen. September 4, 2026. Company release
  • 2026 ACC/AHA multisociety guideline. Guideline on the Management of Dyslipidemia. Circulation
  • ClinicalTrials.gov. OCEAN(a)-Outcomes, NCT05581303. Trial record
  • Eli Lilly. ACCLAIM-Lp(a), NCT06292013. Trial record
  • Eli Lilly. MOVE-Lp(a), NCT07157774. Trial record
  • Nissen SE, et al. Zerlasiran—A Small-Interfering RNA Targeting Lipoprotein(a): A Phase 2 Randomized Clinical Trial. JAMA. Primary publication
Educational information—not medical care.

This article is not a diagnosis, prescription or substitute for care from a qualified clinician who knows your history.

YOUR NEXT STEP

Ready for a clinician-led plan?

Clinical review determines the appropriate next step. No treatment is guaranteed.

THE CARE PATH

  1. We order your labs

  2. A licensed clinician reviews results and recommends a plan

  3. Rx sent directly to you when appropriate

Care depends on eligibility and clinical judgment. No prescription is guaranteed. This pathway is not a recommendation for the treatment or test discussed in this guide.

Start your lab reviewExplore free educational tools