An 8,323-patient trial found that pelacarsen lowered lipoprotein(a) but did not significantly reduce cardiovascular death, heart attacks, strokes or urgent coronary procedures, Novartis reported Friday. The Phase III result is the first definitive setback for a drug built specifically to neutralize this common inherited risk factor, and it leaves a field crowded with promising laboratory numbers waiting for proof that patients actually fare better.

The announcement supplies only the topline conclusion. Novartis did not disclose the treatment effect, confidence interval, event counts or results in the group with the highest starting levels. Those details matter because a missed statistical endpoint can cover a range of outcomes, from no detectable benefit to a smaller benefit than the trial was designed to prove. Full findings are due at a medical meeting, the company said in its release.

Still, the result is consequential. Elevated Lp(a), pronounced “L-P-little-a,” is widely regarded as a causal contributor to atherosclerotic cardiovascular disease. It affects roughly one person in five, is mostly inherited and changes little with diet or exercise. Pelacarsen’s failure does not erase that evidence, but it does show that lowering a plausible biomarker with one medicine, at one dose and disease stage, is not the same as preventing clinical events.

Why the Biomarker Looked So Compelling

Lp(a) resembles an LDL particle with an additional protein attached. That structure can carry cholesterol into artery walls and may also promote inflammation and clotting. Large genetic and epidemiologic studies have repeatedly tied higher lifelong exposure to heart attack, ischemic stroke, peripheral artery disease and calcific aortic-valve disease. An AHA statement estimates that genetics accounts for about 70% to more than 90% of the variation between people.

The risk is continuous rather than divided neatly into safe and dangerous categories. The American Heart Association says levels at or above 125 nanomoles per liter, roughly 50 milligrams per deciliter, may raise cardiovascular risk, while levels around 250 nanomoles per liter may double it. Its current guidance also stresses that a person’s absolute risk depends on age, blood pressure, smoking, diabetes and other factors, not Lp(a) alone.

Pelacarsen was designed to intervene close to the source. It is an antisense oligonucleotide that binds the genetic message used by liver cells to make apolipoprotein(a), reducing production of the particle. In a smaller randomized Phase II study, different dosing regimens produced dose-dependent reductions of roughly 35% to 80% at six months. Those striking biomarker results established biological activity, but that study was never large or long enough to determine whether fewer patients would have heart attacks or strokes.

What Lp(a)HORIZON Actually Tested

Lp(a)HORIZON enrolled people who already had cardiovascular disease and an Lp(a) level of at least 70 milligrams per deciliter. Participants had a prior heart attack, ischemic stroke or symptomatic peripheral artery disease. They were randomly assigned to receive an 80-milligram injection of pelacarsen or placebo every month, on top of standard preventive care. The published design called for follow-up until 993 independently adjudicated primary events had accumulated.

The primary measure combined four outcomes: cardiovascular death, nonfatal heart attack, nonfatal stroke and an urgent coronary revascularization requiring hospitalization. Investigators planned controlled statistical tests in all participants and in a prespecified group whose Lp(a) was at least 90 milligrams per deciliter. According to the company, pelacarsen lowered Lp(a) as intended, yet the trial did not demonstrate a significant reduction in that composite compared with placebo.

This is a stronger test than observing whether a blood value moves. Randomization balances measured and unmeasured differences between groups, placebo control limits expectation effects, and adjudication applies common definitions to suspected events. The public registry lists a study that began in 2019 and spanned hundreds of sites. Its long duration and event-driven design were intended to answer the question drug developers had previously left open.

Why Lower Levels May Not Have Lowered Events

Several explanations remain possible, and the topline disclosure cannot distinguish among them. Pelacarsen may not have lowered Lp(a) far enough, early enough or for long enough to reverse risk built over decades. The molecule could affect the measured concentration without changing every harmful property of the particle. Alternatively, any incremental benefit may have been too small to separate from chance in patients already receiving intensive cholesterol, blood-pressure and antithrombotic treatment.

Cardiovascular trials also confront lower event rates as routine care improves. That is good for participants but makes added benefit harder to prove. Lp(a)HORIZON focused on secondary prevention, when arterial disease was already established; a therapy might behave differently before plaque and valve damage accumulate. These are hypotheses, not explanations established by the result. The event curves, adherence, achieved levels, background treatment and component endpoints will be necessary to judge them.

The outcome also sharpens a familiar lesson in medicine: causal biology does not guarantee that every intervention on the pathway will work. LDL cholesterol is supported by multiple successful drug classes and outcome trials, but other apparently persuasive cardiovascular targets have failed when tested prospectively. As Reuters reported, specialists now need to know whether the negative finding reflects pelacarsen, the depth of reduction, trial design or the broader therapeutic hypothesis.

Rival Outcome Trials Now Carry More Weight

The field does not end with pelacarsen. Amgen’s olpasiran and Eli Lilly’s lepodisiran use small interfering RNA, a related but distinct method of silencing liver production of apolipoprotein(a). Earlier studies reported deeper and more durable Lp(a) reductions than those seen with pelacarsen, but they too primarily measured the biomarker. Cross-trial comparisons are unreliable because doses, populations, assays and follow-up differ.

Amgen’s Phase III OCEAN(a)-Outcomes trial has enrolled about 7,300 people with established atherosclerotic disease and Lp(a) of at least 200 nanomoles per liter. Its primary objective is to determine whether olpasiran reduces coronary death, heart attack or urgent coronary revascularization. Its narrower entry threshold and different composite could reveal whether selecting people with very high concentrations, or achieving greater suppression, changes the clinical result.

Lilly’s ACCLAIM-Lp(a) study is even larger, with a planned enrollment of about 17,300 adults who have cardiovascular disease or are at high risk for a first event. The trial is testing lepodisiran against placebo and is expected to continue for years. Until those outcome data arrive, dramatic percentage reductions should be treated as evidence that the drugs engage their target—not evidence that they prevent disability or death.

What Patients and Clinicians Can Conclude

Friday’s result does not make an Lp(a) measurement meaningless. It can identify inherited risk that is not visible on a standard lipid panel, inform the intensity of conventional prevention and prompt testing discussions within families. The 2026 U.S. dyslipidemia guideline recommends that adults have Lp(a) measured at least once, according to the AHA’s project. A test is a risk-assessment tool; it is not a promise that a targeted medicine exists.

There is currently no approved drug whose indication is specifically to lower Lp(a) and thereby prevent cardiovascular events. People with high levels therefore still rely on proven measures that reduce overall risk: controlling LDL cholesterol and blood pressure, treating diabetes, avoiding tobacco, exercising as medically appropriate and taking prescribed preventive medicines consistently. The pelacarsen news is not a reason to stop therapy or alter care without a clinician.

The next decisive evidence will come in two stages. First, complete Lp(a)HORIZON data must show how close the trial came, whether individual outcomes moved in the same direction, how the highest-risk subgroup fared and whether safety remained acceptable. Then rival trials must establish whether a different molecule or more profound suppression changes events. Until then, the most accurate conclusion is narrow but important: pelacarsen lowered the inherited marker, and this large trial did not prove that doing so improved cardiovascular outcomes.