A Drug Cut Lp(a) by 78 Percent and Heart Attacks Did Not Budge
Lipoprotein(a), usually written Lp(a), has become the cholesterol test people are told they never had. It is largely set by genes, it tracks with heart attacks, and until recently there was nothing to do about it. The first large trial of a drug that lowers it has now reported, and the answer is not the one cardiology was hoping for.
- Lp(a)HORIZON, a phase 3 trial presented on November 7, 2026, gave more than 8,300 people with existing heart disease and high lipoprotein(a) either monthly pelacarsen or placebo for about 4.7 years.
- The drug did exactly what it was designed to do to the number: at one year, Lp(a) was 78.4 percent lower, against 3.2 percent on placebo. Major heart events did not change: 12.3 percent versus 12.1 percent, a hazard ratio of 1.02.
- Novartis has ended the program. The result does not settle the question for other Lp(a) drugs, but it is a strong warning against treating a lower number as a lower risk.
The trial

Lowering a risk marker is not the same as protecting the organ it predicts trouble for. German Hygiene Museum, Dresden.
DguendelCC BY 4.0
Illustrative. A museum model, not connected to the trial.
Lp(a)HORIZON enrolled people who had already had cardiovascular disease and whose Lp(a) was at least 70 mg/dL. Cleveland Clinic reports 8,322 people randomized; TCTMD reports 8,350 enrolled. They received pelacarsen, an injection that reduces the liver's production of Lp(a), at 80 mg once a month, or a placebo, on top of their usual care. Follow up lasted a median of about 4.7 years.
The main outcome was a composite of major cardiovascular events, the hard endpoints such as heart attack, stroke and cardiovascular death that a heart drug ultimately has to move.
The result
At one year, pelacarsen had lowered Lp(a) by 78.4 percent, against 3.2 percent on placebo. Over the whole trial, major cardiovascular events occurred in 12.3 percent of the pelacarsen group and 12.1 percent of the placebo group. The hazard ratio was 1.02, with a confidence interval of 0.90 to 1.15, which is a clean null: no benefit, and no sign of a hidden one.
The drug was also harder to stay on. Participants stopped it because of adverse events in 20.6 percent of cases, against 7.5 percent on placebo. Novartis, which funded the trial, has ended the program.
What this does and does not mean
High Lp(a) is still associated with more heart disease. What this trial shows is that lowering it this way, in people who already have heart disease, did not reduce events over five years. That separates two questions that are easy to blur: whether a marker predicts risk, and whether moving that marker changes the risk.
whether a marker predicts risk, and whether moving that marker changes the risk
Other drugs that lower Lp(a), including small interfering RNA drugs and an oral pill, are still in outcome trials, and this result does not decide their fate. It does raise the bar they will have to clear.
Honest caveat
The trial included only people who already had cardiovascular disease, so it says nothing direct about people with high Lp(a) and no history of heart disease. We read the Cleveland Clinic release and independent trade coverage rather than the full NEJM paper, and the two give slightly different enrollment totals. The trial was funded by Novartis.
What this means for you
If you have had your Lp(a) tested and it came back high, it is still useful information about your overall risk, and the measures with strong evidence, such as managing LDL cholesterol, blood pressure, smoking and activity, still apply.
What this trial changes is the expectation. A future Lp(a) drug will need its own outcome trial before anyone can say that lowering the number protects the heart, and no supplement has shown it can do this at all.
Primary sources
Common questions
Does this mean high Lp(a) does not matter?
No. High Lp(a) is still linked to more heart disease. The trial shows that lowering it with pelacarsen, in people who already had heart disease, did not reduce heart attacks, strokes or cardiovascular deaths over about five years.
How much did the drug lower Lp(a)?
At one year, by 78.4 percent, compared with 3.2 percent on placebo. Over about 4.7 years, major cardiovascular events were 12.3 percent on the drug and 12.1 percent on placebo.
Are other Lp(a) drugs dead too?
Not necessarily. Drugs that work by other mechanisms, including small interfering RNA drugs and an oral pill, are still in outcome trials. This result means each will have to prove it reduces events, not just the number.
Stay Curious.
Better health isn't found in a single article. It's built through curiosity, evidence, and small discoveries over time. Join thousands exploring longevity, gut health, recovery, nutrition, and the tools that truly make a difference.
