Lp(a): The Cholesterol Test Most People Never Get, and the Drug Trial That Missed
The 2026 guideline says every adult should measure Lp(a) once. A large drug trial just lowered it without cutting events. What your result means, and what to do next.
Most of us read a cholesterol report the way we read terms of service: a quick scroll, a glance for anything flagged in red, done. LDL, HDL, triglycerides, a total. If the numbers sit near the line, the doctor says "keep an eye on it" and life resumes. What almost never appears on that page is a number that was set before you could walk, and that can matter more than any of the others.
It is called lipoprotein(a), Lp(a) for short. The 2026 ACC/AHA dyslipidemia guideline, published in March, recommends that every adult have it measured at least once. Then, this autumn, came news that seems to point the other way: a large drug trial that lowered Lp(a) and still did not prevent heart attacks and strokes. Both things are true at once, and the space between them is where the useful questions live.
What Lp(a) is, and why your panel skips it
Lp(a) is a particle that looks a lot like LDL, the familiar "bad" cholesterol carrier, with one extra protein tethered to it. Your genes decide how much of it you make. Diet, exercise and weight move it very little, and in adults the level stays roughly steady over the years. That is why a standard lipid panel leaves it out: it needs its own blood test, and until recently most doctors did not order it unless there was a reason.
Higher levels are linked in large population studies to more heart attacks, more strokes and a higher chance of the aortic valve stiffening over time. A commonly cited estimate is that around one person in five has an elevated level. Most of them have no idea, because nothing about it announces itself. No symptom, no change on the scale.
The first time I looked closely at a lipid panel, I realised I had been skimming the same four lines for years and trusting that the missing ones did not exist. A test that is not on the page feels like a test that does not matter. It is a quiet bias, and it is the reason a once-in-a-lifetime test needs a guideline to push it.
Who should prioritise the test
The guideline's position is that everyone should have one measurement as an adult. Because the number barely drifts, one result is usually enough, and you do not repeat it every year the way you might repeat LDL.
Some people have more reason to do it sooner:
- A parent, sibling or child who had a heart attack, stroke or stent at a young age.
- A heart attack or stroke yourself with no obvious cause, or at an age that surprised everyone.
- Familial hypercholesterolemia, or LDL that stays stubbornly high on treatment.
- A relative who already knows their Lp(a) is high. It runs in families, so the result is information for them too.
One practical snag: labs report Lp(a) in either nmol/L or mg/dL, and the two do not convert cleanly. Ask which unit your result is in before comparing it with any cutoff you read online, including the ones below.
What the pelacarsen trial showed
Pelacarsen is an antisense drug that tells the liver to make less of the protein that defines Lp(a). The Lp(a)HORIZON trial gave it, or a placebo, to more than 8,300 people who already had cardiovascular disease and elevated Lp(a). In early September, Novartis reported the top-line result: the drug lowered Lp(a) as expected, was reasonably safe, and did not significantly reduce the main endpoint, a combination of cardiovascular death, heart attack, stroke and urgent procedures to reopen a coronary artery.
The full data are due at the American Heart Association Scientific Sessions in Chicago this week, November 6 to 9. Until the details are public, anything about why is informed speculation. That includes what follows.
Why a lower number did not translate into fewer events
The cleanest reading is not "Lp(a) does not matter." The genetic evidence that high Lp(a) causes harm is strong, and a negative drug trial does not erase it. It means this drug, in this group, over this follow-up, did not show a benefit. There are a few honest explanations on the table.
Maybe the drop was not big enough. Genetic studies hint that a large absolute reduction may be needed before risk visibly falls. Lowering a number is not the same as lowering it far enough.
Maybe the follow-up was too short. Plaque builds over decades. A few years of lower Lp(a) may not undo that.
Maybe the people enrolled had little room left to improve. They all already had heart disease, and by early reports their LDL was well controlled. If the other risk was already pushed down, the remaining slice that Lp(a) contributes is harder to show.
That last point is a useful one for the rest of us. It suggests the stronger lever, today, is everything around Lp(a) rather than Lp(a) itself.
Your result, what next: a decision guide
The guideline treats an Lp(a) of about 125 nmol/L (roughly 50 mg/dL) or higher as a risk-enhancing factor. I checked that cutoff against several secondary summaries of the guideline rather than the full text, so confirm it with your clinician before treating it as a hard line. Here is how I would think about the three common situations.
Below about 125 nmol/L
Nothing extra to act on. You do not need to repeat the test, and your risk is shaped by the usual things: LDL, blood pressure, smoking, blood sugar, family history. Good news is allowed to be boring.
At or above about 125 nmol/L
This is where the result changes the conversation without changing your life overnight. In practice it means three things. The case for treating LDL becomes stronger, and the LDL target tends to get tighter. The other risk factors deserve real attention: blood pressure, glucose, weight, smoking, sleep, activity. And it can tip a borderline decision, for example whether to get a coronary calcium scan or to start a statin, toward doing it.
Very high, or you have already had a heart event
If your level is far above the cutoff, or you have had a heart attack, stroke or stent, ask for a referral to a cardiologist or lipid specialist. Family members may be worth testing. Some LDL-lowering medicines, such as PCSK9 inhibitors, also nudge Lp(a) down a modest amount, though that is not why they are prescribed. Lipoprotein apheresis, a dialysis-like filtering procedure, exists for selected patients with very high levels and established disease.
What not to do: buy a supplement sold as an Lp(a) lowerer. None has outcome evidence behind it. Niacin lowers the number a little, but trials did not show fewer heart events, and it brings side effects.
Which other Lp(a) drugs are still in the running
Pelacarsen is one of several. To my knowledge, other approaches still in late-stage testing include small interfering RNA drugs such as olpasiran and lepodisiran, which silence the same liver pathway more deeply, and an oral drug, muvalaplin, that works differently. Several of them lower Lp(a) by much larger amounts than pelacarsen did, which makes the "was it enough?" question the one to watch. Their outcomes results are still to come, so check ClinicalTrials.gov or a current cardiology source for where each stands.
Until then, nothing approved is proven to lower heart events by lowering Lp(a). The test is not useless for that reason. It tells you how hard to lean on everything else.
FAQ
Do I need to fast for an Lp(a) test?
Generally no. Lp(a) is not much affected by what you ate, so it can usually be drawn with other bloodwork at any time. Check with the lab, since some pair it with a fasting panel.
Can diet or exercise lower it?
Very little. Genetics sets most of the level, so lifestyle matters here mainly because it lowers every other risk. That is not a small thing when Lp(a) is high.
If mine is high, will I have a heart attack?
Not necessarily. A high level raises the odds, it does not decide the outcome. Many people with high Lp(a) never have an event, particularly when LDL, blood pressure and smoking are well handled.
Does the failed trial mean Lp(a) is a red herring?
No. It means one drug did not show a benefit in one population. The case for knowing your number rests on how it changes the rest of your plan today, not on a drug that is not yet available.
A test you take once and never think about again is a rare thing in medicine. It is worth the small errand, if only to stop wondering about the line that was never on the page.