Heart Drugs That Cut Risk Markers Sharply Still Failed to Prevent Heart Attacks, Stunning Cardiologists
Two experimental heart drugs did exactly what they were designed to do in the blood — and still failed to help a single patient avoid a heart attack or stroke. The full results, presented on Saturday at the American Heart Association meeting in Chicago and published the same day in the New England Journal of Medicine, have left cardiologists searching for explanations.
“Everybody in the world, me included, thought that this was going to work,” said Dr. Steven Nissen of the Cleveland Clinic, who presented the results of the Novartis trial on Saturday.
The cholesterol drug that cut Lp(a) by 78%
The first disappointment was pelacarsen, an experimental Novartis drug developed with Ionis Pharmaceuticals. It targets lipoprotein(a), known as Lp(a) — a cholesterol-carrying particle, largely determined by genetics, that is estimated to affect roughly one in five people worldwide and has been repeatedly tied to heart attacks and strokes.
In the Lp(a)HORIZON trial, more than 8,300 patients with established cardiovascular disease received monthly injections of pelacarsen or a placebo and were followed for a median of 4.7 years. The drug worked on its target: it cut Lp(a) levels by about 78%.
But the clinical results were virtually identical in the two groups. Major cardiovascular events — cardiovascular death, non-fatal heart attack, non-fatal stroke or urgent hospitalisation to unblock a clogged artery — occurred in 12.3% of patients on pelacarsen and 12.1% of those on placebo, a hazard ratio of 1.02. A separate analysis in NEJM Evidence, looking at patients whose Lp(a) fell the most, still showed no benefit.
Side effects were also a problem. Some 20.6% of patients stopped taking pelacarsen because of adverse effects, versus 7.5% of those on placebo.
Novartis had already announced in September that the trial missed its main goal, without giving numbers. The company is now largely abandoning the drug for atherosclerotic cardiovascular disease, although one trial will continue in patients with calcific aortic valve stenosis.
The trial “is completely neutral,” Nissen said. “It is not harmful. It is not helpful. It is just stone cold neutral.”
Related Articles
The anti-inflammatory drug that also missed
The second failure, presented at the same meeting, followed the same pattern. Novo Nordisk’s ziltivekimab is designed to neutralise interleukin-6, a key driver of inflammation in high-risk heart patients. In the ZEUS trial of about 6,300 patients with heart disease, advanced kidney disease and systemic inflammation, the drug cut the inflammation markers hs-CRP by 64% and interleukin-6 by 76%.
Yet it did not reduce deaths, heart attacks or strokes compared with placebo: the primary outcome occurred in 561 patients on the drug and 565 on placebo, effectively no difference.
Novo Nordisk had announced in July that the study failed, and in September it ended two other late-stage ziltivekimab trials in heart failure patients, on the advice of an independent data monitoring committee.
Why the failures matter
Together, the two trials challenge a central assumption of modern cardiology: that fixing a biological risk marker will automatically prevent disease. Both drugs hit their targets decisively. Neither changed outcomes. Experts say the results do not clear Lp(a) or inflammation of their role in heart disease, but they do show that lowering a marker is not the same as lowering risk — and that the biology may be more complicated than the theories suggested.
The setbacks also leave patients with no approved treatment that specifically lowers Lp(a). Hope now rests on other drugs still in testing, including Amgen’s olpasiran, with Phase III results expected in early 2028, and Eli Lilly’s lepodisiran, expected in 2029. Novartis has another experimental Lp(a)-lowering drug, DII235, in development.
For researchers, the message from Chicago is sobering but not final. As Nissen cautioned, the failure of one drug does not doom the entire approach — but it does mean the next Lp(a) therapy will have to prove itself in patients, not just in blood tests.
[…] increase patient and investor concern, but the rarity of NAION and substantial weight, glycemic, cardiovascular, renal, and sleep apnea benefits should preserve a favorable benefit-risk profile and limit changes […]