A heart drug that cut a key cholesterol marker by more than three-quarters failed to prevent a single extra heart attack or stroke, full results published on Saturday show. Pelacarsen, Novartis's treatment for people with high lipoprotein(a), lowered the particle known as Lp(a) by 78.4 percent, yet patients taking it suffered major cardiovascular events at exactly the same rate as those given a dummy pill. The findings were presented at the American Heart Association's Scientific Sessions in Chicago and published in the New England Journal of Medicine.

Key Points

• Pelacarsen cut lipoprotein(a) by 78.4 percent at 12 months: median levels fell from 236 nmol/L to 64.4 nmol/L, against 233.4 nmol/L on placebo.
• It did not reduce major cardiovascular events: 12.3 percent of treated patients had a heart attack, stroke, cardiovascular death or urgent revascularisation, against 12.1 percent on placebo.
• The hazard ratio was 1.02, meaning risk was effectively identical, and no subgroup benefited, including patients with the highest Lp(a).
• Novartis is halting development of pelacarsen for atherosclerotic cardiovascular disease; a smaller study in calcific aortic valve stenosis continues.
• The result casts doubt on rival Lp(a) programmes from Amgen and Eli Lilly.

What Happened

The full results of the phase 3 Lp(a)HORIZON trial were presented on Saturday, November 7, 2026, in Chicago, and published the same day in the New England Journal of Medicine. Novartis had warned in September that the trial had failed, but withheld the numbers.

It had not come close. Over a median follow-up of 4.7 years, 511 of 4,157 patients on pelacarsen, or 12.3 percent, suffered a major cardiovascular event, against 503 of 4,165 on placebo, or 12.1 percent. The hazard ratio of 1.02, with a 95 percent confidence interval of 0.90 to 1.15, means the drug's effect was statistically indistinguishable from no effect.

The trial enrolled 8,323 to 8,350 patients with established cardiovascular disease and Lp(a) of at least 70 mg/dL, at 797 sites in 42 countries; 82 percent had already had a heart attack. Patients received a monthly 80 mg injection of pelacarsen, an antisense drug developed with Ionis Pharmaceuticals, or placebo. Median Lp(a) fell from 236 nmol/L to 64.4 nmol/L after 12 months on the drug, against 233.4 nmol/L on placebo, yet that biochemical success brought no clinical benefit.

Background and Context

Lipoprotein(a), or Lp(a), is a cholesterol-like particle in the blood. Unlike LDL, the bad cholesterol that statins lower so effectively, Lp(a) levels are set almost entirely by genetics and barely respond to diet, exercise or existing medicines. About one in five people has elevated Lp(a), and high levels are linked to a greater risk of heart attacks, strokes and aortic valve disease.

That combination, a clear risk factor with no approved treatment, made Lp(a) a prized target in preventive cardiology. Pelacarsen, an antisense drug that blocks the liver's production of the protein carrying Lp(a), had cut levels by about 72 percent in phase 2 studies. Novartis bet that lowering the marker would lower the risk, and the Lp(a)HORIZON trial was designed to prove it.

Pelacarsen and the Lp(a) Question

The most striking feature is how complete the failure was. Pelacarsen did not merely miss its main target; it failed to show even a numerical advantage over placebo, with the event rate in the treated group slightly higher, not lower. Nor did any subgroup offer hope: among patients with baseline Lp(a) of at least 90 mg/dL, the hazard ratio was 1.00, a perfect tie.

That does not necessarily settle the Lp(a) question. Amgen's olpasiran and Eli Lilly's lepodisiran lower Lp(a) more deeply and enrolled patients with higher starting levels, at least 200 nmol/L and 175 nmol/L, against about 149 nmol/L in the pelacarsen trial. Their trials are still running, but the burden of proof on them has grown sharply.

Reactions and Quotes

Fierce Biotech, reporting the data on Saturday, described the outcome as "a total absence of efficacy", noting that the failure could intensify questions about the upcoming readouts from Amgen and Eli Lilly. Analysts had hoped rival drugs might succeed where pelacarsen failed, but the complete lack of benefit, even in the highest-Lp(a) patients, took the shine off that argument.

BMO Capital Markets analysts had said in a November 3 note that a minor separation on Novartis's primary endpoint could help the rival programmes, and that any signal in the high-Lp(a) subgroup would likely be positive for olpasiran and lepodisiran. Pelacarsen delivered neither. Novartis has said little beyond the September announcement, while keeping a mid-stage study in calcific aortic valve stenosis running.

Why It Matters

The result matters first for the millions of people with high Lp(a). For years, patients told their cholesterol was otherwise well controlled but their Lp(a) was high have waited for a drug aimed specifically at the problem. Pelacarsen was the closest to reaching them, and its failure means that wait continues, with no approved Lp(a)-lowering treatment on the horizon.

It also matters for how medicine understands risk. Lp(a) is one of the strongest inherited risk factors for heart disease, and genetic studies have long suggested it directly causes harm. The pelacarsen result does not overturn that evidence, but it shows that lowering a risk marker is not the same as lowering risk in patients. Novartis has written off a major programme, and Amgen and Eli Lilly must now show their drugs can succeed where pelacarsen did not.

What Happens Next

Novartis will wind down the pelacarsen programme for atherosclerotic cardiovascular disease, while the calcific aortic valve stenosis study continues. Full details, including safety data, are now in the New England Journal of Medicine for independent scrutiny.

Attention now turns to the rival trials. Amgen's olpasiran and Eli Lilly's lepodisiran are both in large outcomes studies, with results expected in the coming years. If either shows a benefit, the field will argue that deeper Lp(a) lowering, or a different patient population, was the key. If they fail too, researchers will have to rethink whether Lp(a) can be treated at all. For patients, the advice is unchanged: control LDL cholesterol, blood pressure and smoking, and ask a doctor about Lp(a) testing.

Frequently Asked Questions

What is pelacarsen? An antisense drug from Novartis and Ionis Pharmaceuticals, given as a monthly injection, that blocks the liver from producing the protein carrying lipoprotein(a), cutting Lp(a) by about 78 percent.

Why did pelacarsen fail if it lowered Lp(a) so much? Cutting Lp(a) by 78 percent did not reduce heart attacks, strokes or cardiovascular deaths: 12.3 percent on the drug against 12.1 percent on placebo, a hazard ratio of 1.02, meaning no benefit at all.

What happens to the rival Lp(a) drugs? Amgen's olpasiran and Eli Lilly's lepodisiran lower Lp(a) more deeply and enrolled patients with higher starting levels. Their trials are still running, but the pelacarsen failure raises the bar for them considerably.

Should people with high Lp(a) do anything differently? No new treatment is available, so doctors advise controlling other risk factors, including LDL cholesterol, blood pressure and smoking, and discussing Lp(a) testing with a doctor.

Read more: FDA investigates Zolgensma tumour risk; StatStrip glucose meter recall.