Details from Novartis’ failed cardiovascular trial offer ‘little silver lining’ for rivals

After Novartis and Ionis Pharmaceuticals reported that pelacarsen failed to improve cardiovascular outcomes in a late-stage test, analysts had held out hope that the details would still leave some room for other therapies that target Lp(a). But now, with the fine print revealed, industry watchers see a tough path ahead for the space.

“HORIZON’s full data offer little silver lining,” Truist Securities analysts said in a Sunday evening note.

While the antisense oligonucleotide greatly lowered levels of lipoprotein(a)—aka, the bad cholesterol—HORIZON found no benefit in terms of major adverse cardiovascular events (MACE), according to the company’s presentation at the 2026 American Heart Association Scientific Sessions on Saturday and a simultaneous publication in The New England Journal of Medicine (NEJM). The trial featured 8,300 adults with cardiovascular disease who were followed for seven years.

Investors had been focused on the full HORIZON dataset for possible trends pointing toward a benefit in cardiovascular outcomes “that could have read-through to ongoing trials with siRNA and oral molecules that have greater Lp(a) lowering capacity,” William Blair analysts wrote in a Monday morning note, but “HORIZON cut many ways remains negative.”

Among patients with high Lp(a), a MACE event—which includes nonfatal heart attack, stroke, an urgent revascularization procedure to open clogged heart arteries or death from any cardiovascular cause—occurred in 380 participants (9.1%) treated with pelacarsen compared to 434 participants (10.4%) on placebo, according to a subgroup analysis published in NEJM Evidence.

William Blair pointed to the high rate of discontinuations among participants taking pelacarsen as “one major caveat.” In HORIZON, 1,462 patients (35.1%) given 80-mg pelacarsen every four weeks prematurely discontinued treatment, the analyst said, vs. 761 (18.3%) given placebo on this timeframe.

But even with the discontinuations out of the equation, the drug still failed. For patients treated with uninterrupted pelacarsen, 167 or 11% had a MACE event vs. 204 or 10.1% in the placebo group, according to the subgroup analysis.

“This discontinuation rate is higher than we expected heading into the Lp(a)HORIZON readout and may have censored data impacting the primary endpoint analysis,” William Blair said.

Novartis and Ionis revealed in September that the trial failed to meet the primary endpoint of lowering the risk of MACE vs. placebo but did not provide details at the time.

The revelation shook Novartis’ stock amid several other clinical stumbles, with the market damage spreading to Lp(a)-targeting peers Amgen and Eli Lilly. It also raised questions about the value of targeting Lp(a) in cardiovascular disease (CVD). Amgen’s olpasiran and Lilly’s lepodisiran are both siRNA therapies that lower Lp(a) expression.

“We do see meaningful risk to a potential future in Lp(a)-driven CVD trials in light of results from Lp(a) HORIZON,” William Blair said at the time.

Upon seeing the detailed results, Leerink Partners lowered the probability of success for olpasiran from 50% to 20%. “[W]e are not lowering odds to 0% because it has a different structure and mechanism,” they wrote in a note to investors on Saturday, adding that Amgen’s drug “drives deeper Lp(a) lowering.”

William Blair had previously modeled sales of $710 million by 2030 for olpasiran but has now removed these estimates from its updated model. “[W]e still struggle to find confidence in olpasiran based on the totality of data presented,” the firm wrote on Monday.

Even when looking at the secondary prevention setting—in patients with established CVD—“there was no apparent difference in total events between participants randomly assigned to pelacarsen or placebo,” the NEJM Evidence authors wrote.

William Blair spoke specifically to this point, writing, “With no demonstration of efficacy in these high Lp(a) reduction subgroups . . . we believe it will be challenging for future studies to demonstrate benefit in secondary prevention in an otherwise well-controlled population.”

For now, the Lp(a) field will have to wait to learn more about whether this approach is viable. The Phase 3 OCEAN(a) study of olpasiran is set to read out topline data in early 2028, while Lilly is anticipating initial data from lepodisiran’s Phase 3 ACCLAIM-Lp(a) trial in March 2029.

For Novartis, the public reminder of pelacarsen’s failure couldn’t come at a worse time. Last week, the company confirmed nine reported cases of tumors in patients treated with its spinal muscular atrophy treatment Zolgensma, including two cases where that have been linked by evidence to the gene therapy. The FDA on Friday announced an investigation into the risk of tumor formation associated with the therapy.

And these headlines follow a summer of cascading challenges for Novartis. In September, the antibody-oligonucleotide conjugate del-desiran failed in a Phase 3 trial to significantly improve hand function in patients with myotonic dystrophy type 1, calling into question the company’s $12 billion buyout of Avidity Biosciences last year.

Then in late August, news broke that Novartis had paused a clutch of Phase 2 trials for the CAR T cell therapy rapcabtagene autoleucel after three patient deaths.