On 4 September, Novartis announced the topline result of Lp(a)HORIZON - the first large trial in history to test whether lowering Lp(a) with a drug actually prevents heart attacks and strokes.
8,323 people. Nearly a decade of work. The drug did exactly what it was designed to do on the blood test.
And the trial still missed.
If you've followed StrongHeart for a while, you'll know why I take this one personally. I started this company after a blood test showed my own heart risk was largely inherited - driven by Lp(a), a particle I'd never heard of until it showed up on my results. I've been waiting on this trial the way some people wait on the grand final.
So here's what happened, what it means, and - just as important - what it doesn't.
First: what is Lp(a)?
Lipoprotein(a) - say "L-P-little-a" - is a cholesterol-carrying particle made by your liver. Unlike ordinary LDL cholesterol, your level is set almost entirely by your genes. Diet and exercise barely move it.
Roughly 1 in 5 people carry elevated levels, and most have no idea, because it isn't on a standard cholesterol panel. High Lp(a) is consistently linked to earlier heart attacks, strokes and aortic valve disease.
That's why drug companies have spent a decade building drugs to switch it off.
The trial
Pelacarsen is a monthly injection that tells the liver to make less of the Lp(a) protein. In its phase 2 trial (Tsimikas et al., New England Journal of Medicine, 2020, 286 patients), the top dose cut Lp(a) levels by about 80%. Impressive.
Lp(a)HORIZON (NCT04023552) was the real test: 8,323 people with established cardiovascular disease and Lp(a) of at least 70 mg/dL (about 175 nmol/L in the units on Australian lab reports), randomised to pelacarsen or placebo on top of standard care. The endpoint that matters: cardiovascular death, heart attack, stroke, or urgent hospital treatment to reopen a blocked artery.
The topline result, announced 4 September 2026: Lp(a) came down substantially. Cardiovascular events didn't. Primary endpoint not met.
What that means — honestly
The uncomfortable lesson first: lowering a risk marker is not the same as lowering risk. A number on a blood test is a stand-in for the outcomes we actually care about, and this trial is a reminder that the stand-in and the real thing don't always move together.
But before anyone writes Lp(a)'s obituary, the caveats matter.
This is a topline announcement, not the full data. The detailed results - including the group with Lp(a) above 90 mg/dL — will be presented at an upcoming medical congress.
Everyone in the trial was already extremely well treated. Reports put average LDL around 1.7 mmol/L, which is lower than most heart patients ever reach. Showing extra benefit on top of that is a high bar.
And these were people who already had heart disease, treated for around five years. Whether lowering Lp(a) earlier in life, for longer, or in different people would help is still an open question - and other drugs are still being tested in large outcome trials: olpasiran (OCEAN(a)-Outcomes) and lepodisiran (ACCLAIM-Lp(a)), with results due over the next few years.
The genetics linking high Lp(a) to heart disease haven't changed. What changed is that the first attempt to fix it with a drug, in this group of patients, didn't work.
What to actually do
Knowing your Lp(a) is still worth it. It's one blood test, once in your life - ask your GP. If it's high, that's not a death sentence; it's a reason to take the controllable risks more seriously, earlier.
If yours is high (like mine), the playbook is the unglamorous one that already has outcome data behind it: work with your GP to get LDL and blood pressure to target. Do 150 minutes of moderate cardio a week, or 75 minutes of hard stuff, plus two strength sessions. Eat 25–40 g of fibre a day, with soluble fibre (oats, beans, lentils) doing the cholesterol work. Don't smoke.
And a straight word on supplements: no supplement lowers Lp(a) - including ours. Supplements can support your general heart health as part of the routine above, but they don't treat or prevent disease. If you ever see a product claiming to fix Lp(a), walk away.
A missed endpoint isn't a wasted trial. 8,323 people gave years of their lives to answer a question nobody had answered before, and that answer will make the next round of trials smarter. That's how the science is supposed to work - even on the weeks it breaks your heart a little.
Sources: Novartis topline announcement, Lp(a)HORIZON phase 3 trial, 4 September 2026 (NCT04023552, n=8,323); Tsimikas S et al., "Lipoprotein(a) Reduction in Persons with Cardiovascular Disease", NEJM 2020 (phase 2, n=286).
This article is general information, not medical advice. Talk to your doctor before changing any medication or treatment plan.