Novartis has disclosed the detailed Phase 3 Lp(a)HORIZON results for pelacarsen, showing that the investigational antisense oligonucleotide reduced lipoprotein(a), or Lp(a), by about 78% but failed to reduce major cardiovascular events in more than 8,300 people with established cardiovascular disease. The findings were presented at the American Heart Association Scientific Sessions 2026 after Novartis had already announced in September that the pivotal trial missed its primary endpoint.
The detailed readout makes the failure more consequential than the earlier topline announcement alone. Participants were followed for a median of approximately 4.7 years, yet the large reduction in Lp(a) did not translate into a meaningful reduction in cardiovascular death, non-fatal myocardial infarction, non-fatal stroke or urgent coronary revascularisation requiring hospitalisation. The result does not establish that Lp(a) is clinically irrelevant, but it does show that selectively lowering the inherited risk factor with a dedicated therapy is not automatically sufficient to reduce cardiovascular events.
That distinction now hangs over an increasingly important cardiovascular development race involving Novartis, Amgen and Eli Lilly and Company. Amgen’s olpasiran and Eli Lilly and Company’s lepodisiran use different RNA-silencing approaches and have produced even deeper Lp(a) reductions in earlier studies, but neither programme has yet demonstrated that those biomarker effects translate into fewer heart attacks, strokes or cardiovascular deaths.
Why is the 78% Lp(a) reduction in Lp(a)HORIZON such an important negative result?
Lp(a)HORIZON was designed to answer a much harder question than whether pelacarsen could lower a laboratory measurement. The Phase 3 randomised, double-blind, placebo-controlled trial enrolled 8,323 adults with established cardiovascular disease and screening Lp(a) concentrations of at least 70 mg/dL, including people with previous myocardial infarction, ischaemic stroke or clinically significant symptomatic peripheral artery disease.
Participants received pelacarsen 80 mg by subcutaneous injection every four weeks or placebo while continuing guideline-directed cardiovascular therapy. The primary endpoint assessed time to cardiovascular death, non-fatal myocardial infarction, non-fatal stroke or urgent coronary revascularisation requiring hospitalisation. The trial was therefore built around major clinical events rather than biomarker improvement alone.
Pelacarsen clearly engaged its intended biological target. Lp(a) was reduced by about 78%, yet cardiovascular outcomes remained neutral over nearly five years of follow-up. Additional analyses examining different measures of response, including patients with greater Lp(a) reductions and analyses focused on participants who remained on treatment, did not reveal a convincing cardiovascular benefit.
That creates the central scientific problem. A therapy can demonstrate powerful pharmacological activity against a biomarker without demonstrating that changing that biomarker at that stage of disease alters patient outcomes. Lp(a)HORIZON therefore separates biological target engagement from clinical efficacy in unusually stark fashion.
Does pelacarsen’s failure mean the entire Lp(a) cardiovascular hypothesis is wrong?
The evidence does not support that conclusion. Elevated Lp(a) remains strongly associated with atherosclerotic cardiovascular disease and is largely genetically determined, making it an important source of residual cardiovascular risk. Cardiovascular researchers have treated the particle as a potentially causal target, while screening for elevated Lp(a) has gained increasing attention in risk assessment.
What Lp(a)HORIZON challenges is the simpler assumption that substantially lowering Lp(a) in people who already have advanced cardiovascular disease will necessarily reverse enough accumulated risk to produce fewer events within several years. Researchers now have to consider several possibilities, including whether treatment began too late, whether greater Lp(a) suppression is necessary, whether different patient populations would respond differently or whether characteristics specific to pelacarsen limited translation of biomarker reduction into clinical benefit.
The study population is particularly important. These were secondary-prevention patients who had already experienced major manifestations of cardiovascular disease and were being treated for other modifiable risk factors. By the time patients have experienced myocardial infarction, stroke or symptomatic peripheral arterial disease, years or decades of vascular injury may already have occurred.
The neutral result therefore cannot automatically be extrapolated to people with elevated Lp(a) who have not yet experienced a cardiovascular event. It also cannot establish that every pharmacological approach to Lp(a) lowering will fail, because different molecules may achieve different magnitudes, durations and patterns of suppression.
Why will Amgen and Eli Lilly now face a tougher test with olpasiran and lepodisiran?
Lp(a)HORIZON was the first major outcomes trial to force the emerging Lp(a) drug class beyond the relatively comfortable territory of biomarker reduction. That raises the evidentiary threshold for every programme following it.
Amgen’s olpasiran and Eli Lilly and Company’s lepodisiran are small interfering RNA therapies rather than antisense oligonucleotides. These programmes have produced Lp(a) reductions approaching or exceeding 90% in earlier clinical development, making them pharmacologically distinct from pelacarsen even though they ultimately target production of apolipoprotein(a).
Amgen’s OCEAN(a)-Outcomes programme is evaluating whether olpasiran reduces major cardiovascular events in patients with established atherosclerotic cardiovascular disease and markedly elevated Lp(a). The programme therefore offers an opportunity to test whether a different RNA-silencing mechanism and deeper suppression can succeed where pelacarsen did not.
Eli Lilly and Company’s ACCLAIM-Lp(a) programme is broader, enrolling approximately 17,000 adults with elevated Lp(a), including participants with established atherosclerotic cardiovascular disease as well as people at increased risk of experiencing a first cardiovascular event. That population design could become particularly important if timing of intervention proves to be one reason behind pelacarsen’s failure.
Neither programme can now rely on dramatic Lp(a) reduction as a persuasive proxy for cardiovascular benefit. Investors, clinicians and regulators will increasingly focus on hard clinical events, adherence, treatment discontinuation and whether benefits emerge in populations treated before extensive vascular disease is established.
What does the pelacarsen discontinuation rate reveal about the commercial challenge?
The newly disclosed safety and tolerability information adds another difficulty for pelacarsen. About 20.6% of patients reportedly discontinued treatment because of adverse effects compared with 7.5% receiving placebo. That difference is notable for a preventive cardiovascular therapy that could potentially need to be administered for years.
Earlier clinical experience had already identified injection-site reactions and other tolerability issues associated with pelacarsen. Separate studies have nevertheless shown that the therapy can produce strong Lp(a) reductions, including in patients undergoing lipoprotein apheresis, but those biomarker effects do not resolve the central outcomes question exposed by Lp(a)HORIZON.
The commercial significance is straightforward. A chronic preventive therapy must deliver enough clinical benefit to justify treatment burden, adverse effects and recurring administration. When the pivotal cardiovascular trial demonstrates no reduction in major events, a materially higher discontinuation rate becomes more difficult to overlook because there is no demonstrated cardiovascular benefit against which the treatment burden can be weighed.
This could become another point of differentiation for the next generation of Lp(a) therapies. Longer dosing intervals, deeper suppression and better tolerability may make siRNA approaches operationally more attractive, but those characteristics remain commercially secondary until cardiovascular outcomes benefit is demonstrated.
What does the Phase 3 failure mean for Novartis and Ionis Pharmaceuticals?
For Novartis, the result is disappointing but unlikely to threaten the broader investment case on its own. The Swiss pharmaceutical group remains supported by a diversified portfolio that includes major growth products across oncology, immunology, neuroscience and cardiovascular medicine. Pelacarsen was a potentially valuable pipeline asset, but it was not carrying the company’s valuation in the way a single late-stage programme might for a smaller biotechnology business.
The result matters more strategically because it weakens one of the most closely watched emerging cardiovascular opportunities. Novartis already has a significant presence in cholesterol management through Leqvio, and success with pelacarsen could have strengthened its position in residual cardiovascular risk beyond conventional LDL cholesterol lowering. That opportunity has now become substantially less certain.
Ionis Pharmaceuticals has greater direct exposure to the technology story because it discovered pelacarsen and licensed development and commercialisation rights to Novartis. The detailed data remove some residual hope that subgroup analyses or deeper examination might reveal an obvious path to cardiovascular-event reduction after the earlier topline miss.
The broader implication for antisense drug development is more nuanced. Pelacarsen successfully reduced its intended target, so the setback is not simply evidence of failed delivery or absent pharmacological activity. The more difficult question is whether the biological intervention occurred too late, was insufficiently potent for clinical effect or targeted a pathway that is harder to modify therapeutically than genetic and epidemiological evidence had suggested.
Could treating Lp(a) earlier prove more successful than treating established cardiovascular disease?
This question may become one of the most important consequences of Lp(a)HORIZON. Lp(a) exposure accumulates over a lifetime, while participants in the Novartis trial had already developed clinically significant cardiovascular disease before receiving pelacarsen.
If Lp(a) contributes to atherosclerosis over decades, reducing the particle only after extensive vascular injury has occurred could theoretically remove one source of damage without reversing existing disease. That explanation remains a hypothesis rather than a demonstrated reason for the negative trial, but it is now highly relevant to the design and interpretation of future studies.
The difference could help explain why trials that include patients without previous cardiovascular events deserve attention rather than being dismissed because pelacarsen failed. Earlier intervention could create a longer period during which reduced Lp(a) exposure might influence plaque development and downstream cardiovascular risk, although proving such an effect would require large populations and long follow-up.
Conversely, another neutral outcomes trial from a potent therapy in an appropriately selected population would put considerably greater pressure on the therapeutic hypothesis. The field therefore needs more than another impressive percentage reduction in a laboratory marker.
What will determine whether Lp(a) becomes a major cardiovascular drug market?
The pharmaceutical opportunity remains substantial because elevated Lp(a) is common, genetically determined and currently lacks a broadly approved targeted pharmacological therapy with demonstrated cardiovascular-event reduction. That combination has attracted major development investment and created expectations that Lp(a) could become another important preventive-cardiology treatment category alongside LDL cholesterol lowering.
Lp(a)HORIZON has now changed the commercial question. The race is no longer simply about which company can reduce Lp(a) the most or dose patients the least frequently. The decisive question is whether any approach can demonstrate that selective Lp(a) lowering prevents cardiovascular events.
Amgen and Eli Lilly and Company therefore carry unusually important next-stage evidence responsibilities. Their programmes can test whether greater suppression, different RNA technology, different dosing and broader patient populations can overcome the limitations exposed by pelacarsen.
For Novartis, the 78% biomarker reduction alongside a neutral clinical outcome is a reminder that cardiovascular drug development ultimately depends on outcomes rather than mechanism. For the wider Lp(a) field, the experiment is not over, but the burden of proof has become substantially higher.
