Is Novartis's Pelacarsen Exit a Real Blow to Lp(a) Drugs?

Novartis has decided to discontinue the pelacarsen atherosclerotic cardiovascular disease (ASCVD) development program, including ongoing ASCVD studies and the Managed Access Program. The decision follows the Lp(a)HORIZON trial, a phase 3 study of more than 8,300 adults with elevated lipoprotein(a).

The drug did exactly what it was designed to do at the biochemical level. Lp(a) fell by a mean of 78% over the trial. The clinical outcome did not follow. After a median follow-up of 4.7 years, 511 primary endpoint events occurred in the pelacarsen arm (12.3%) versus 503 in the placebo arm (12.1%), a hazard ratio of 1.02 (95% CI 0.90–1.15; P = 0.76).

That gap is not close. Novartis had already sunk tens of millions of dollars into development, and analysts had projected pelacarsen as a multi-billion-dollar product. Investing.com reported the company has largely abandoned it, which means the commercial case for the lead Lp(a) asset is gone.

Lipoprotein(a) is a cholesterol-carrying particle largely set by genetics. Roughly one in five people carries high levels, and it is an independent risk factor for heart attack, stroke, and aortic valve disease. Standard statins do little to move it, which left a clear gap in treatment.

That gap made Lp(a) one of the most sought-after targets in cardiovascular medicine. Antisense and siRNA drugs can knock the liver's production of the protein down sharply, and the early biochemistry looked convincing. The question was always whether lowering the number would translate into fewer events.

Pelacarsen was the first large test of that question, and the answer so far is no. Physicians now have a 78% reduction with no outcome benefit to point to.

The blow is real in three ways. Novartis loses its lead Lp(a) asset and a potential blockbuster. The first-generation class loses credibility on timing, since its outcome data arrived after years of anticipation. And physicians are left with a biomarker success and no clinical tool to show for it.

But the trial does not prove the target is wrong. Several design factors deserve scrutiny. Pelacarsen is an antisense oligonucleotide with a specific dosing and delivery profile, and its effect on Lp(a) may differ from other mechanisms. The trial enrolled patients with established disease, a population where many competing risk factors drive events, which can dilute a modest benefit.

Researchers quoted in coverage said the result raises questions that may force cardiologists to rethink how they interpret Lp(a) and which patients to treat. Competitors are the clearest test. Amgen's olpasiran and Eli Lilly's lepodisiran both lower Lp(a) by around 95% in earlier work, and both are in outcome trials.

Amgen's phase 3 results are expected in early 2028 and Lilly's in 2029. If either shows benefit, the verdict on Novartis will look like a timing and molecule problem. If both fail, the target itself comes under serious doubt. My position: this is a trial-design and endpoint risk for the class, and Novartis is the company that paid for learning it.

The target remains valid until a larger outcome trial says otherwise.

Three signals will settle this question: - Amgen's olpasiran phase 3 outcomes (expected early 2028): the most direct test of whether Lp(a) lowering prevents events. - Eli Lilly's lepodisiran outcomes (expected 2029): a second independent mechanism, which matters if olpasiran is ambiguous. - Subgroup and full-dataset analyses of Lp(a)HORIZON: any signal in patients with the highest baseline Lp(a) or lowest competing risk would reshape how the remaining trials are read. Until those readouts arrive, expect Novartis to keep its commercial exposure small and competitors to emphasise trial design differences in their own data.