CardiologyPublished: September 2026·Updated: September 2026·10 min read

Lowering Lp(a): So Close, Yet So Far Away

Lowering Lp(a) has been one of the most anticipated goals in cardiology. Lipoprotein(a) is a powerful, genetically fixed predictor of heart attack and stroke, and after decades with no way to touch it, we finally have drugs that cut it by around 80%. On 4 September 2026, Novartis announced that the first large outcomes trial of one of them, pelacarsen, in 8,323 patients with existing heart disease — the Lp(a) HORIZON trial — did not reduce cardiovascular events, despite lowering Lp(a) exactly as intended. It is a real disappointment. It does not mean Lp(a) is unimportant. It means that measuring it and successfully lowering it are, for now, two very different things.

Dr. Peter Chang

Triple Board-Certified Cardiologist & Vascular Specialist

What the Lp(a) HORIZON Trial Found

The Lp(a) HORIZON trial randomised 8,323 people who already had cardiovascular disease and a raised lipoprotein(a) — 70 mg/dL or higher — to a monthly injection of pelacarsen 80 mg or a placebo, on top of good standard treatment including statins. Pelacarsen is an antisense drug that switches off production of the apolipoprotein(a) protein, and it did its job: Lp(a) fell by roughly 80%.

The primary endpoint was a composite of cardiovascular death, non-fatal heart attack, non-fatal stroke, and urgent hospitalised coronary procedures. After several years of follow-up, there was no significant reduction in that endpoint compared with placebo. Novartis released only the topline result on 4 September 2026; the full data, including the higher-Lp(a) subgroup and the actual hazard ratio, are expected at the American Heart Association meeting in November, and cardiologists in Singapore are watching for it closely. But the headline is clear: lowering Lp(a) with this drug did not prevent events in this population.

What Lp(a) Is and Why It Predicts Risk So Well

Lipoprotein(a), said ‘LP-little-a’, is an LDL-like cholesterol particle with an extra protein, apolipoprotein(a), wrapped around it. Your level is set almost entirely by the genes you inherit, is largely fixed from childhood, and barely moves with diet, exercise or statins. Around one in five people worldwide carries a high level, and it is not included in a standard cholesterol panel in Singapore or anywhere else, so most never know.

As a predictor it is excellent. Large population studies and genetic (Mendelian randomisation) analyses consistently link lifelong high Lp(a) to more coronary disease, more stroke, and more calcific aortic valve disease, independent of LDL cholesterol. That strong, causal-looking signal is exactly why a drug that lowers it was expected to work — and why the HORIZON result is such a surprise.

So Close: We Can Finally Lower It

For most of my career, a high Lp(a) was a number we could measure and then do nothing about. That has changed fast. A group of RNA-targeted drugs now lowers lipoprotein(a) dramatically: pelacarsen, an antisense oligonucleotide, by around 80%, and the small interfering RNA agents olpasiran, lepodisiran and zerlasiran by 80 to 95%, some with only two or three injections a year.

The tolerability in trials has been reassuring, dominated by injection-site reactions rather than anything serious. In other words, the pharmacology is solved. We can take a lifelong genetic risk factor and, chemically, make most of it disappear. The only question left was the one that matters: does doing so actually protect the heart? HORIZON was built to answer it for pelacarsen, and the first answer is no.

So Far Away: Why a Great Predictor May Not Be a Good Target

This is the hard lesson cardiology keeps relearning: a number that predicts risk is not guaranteed to reduce risk when you lower it with a drug. The classic example is HDL cholesterol. Low HDL is one of the most reliable predictors of heart attack we have, yet every large trial that raised it — with niacin in AIM-HIGH and HPS2-THRIVE, and with CETP inhibitors in dal-OUTCOMES and REVEAL — failed to translate that into fewer events.

Lp(a) may be turning out the same way, at least in this setting. The genetics show that being born with lifelong low Lp(a) is protective. Whether lowering Lp(a) for a few years in your sixties, after decades of exposure and on top of good LDL control, does anything measurable is a completely separate question — and HORIZON suggests the honest answer might be very little.

The Explanations Still on the Table

One negative trial does not close the book. Several explanations are still live, and the full data may favour some over others:

Too late: lifelong exposure is what drives the risk, and a few years of lowering in older patients with established disease may simply come too late to matter
Not enough headroom: in a trial population already well treated with statins, the extra risk attributable to Lp(a) that is left to remove over four or five years may be small
Wrong population: the benefit, if any, might show up in younger people or in primary prevention, not in secondary prevention
Magnitude and threshold: the higher-Lp(a) subgroup (90 mg/dL and above) has not been reported yet, and could look different
Other drugs, other trials: olpasiran (OCEAN(a)-Outcomes) and lepodisiran (ACCLAIM-Lp(a)) are still running, and will give an independent read within the next couple of years

What This Means If You Have a High Lp(a) in Singapore

Practically, not much changes today, because there was never an approved drug to lower lipoprotein(a) for the purpose of preventing events — and after HORIZON, there may not be one soon. A high Lp(a) is still worth knowing about, because it tells us your genetic risk runs higher than your LDL number alone suggests, and that should make us treat everything modifiable harder.

In our clinic on Orchard Road, that means pushing LDL cholesterol to a lower target (often below 1.4 mmol/L, sometimes lower), controlling blood pressure tightly, keeping well away from tobacco, treating diabetes properly, and considering low-dose aspirin where the rest of the risk profile justifies it. Lipoprotein apheresis, a dialysis-like procedure, exists for rare extreme familial cases but is not a practical option here. Niacin lowers Lp(a) by around 20% but does not improve outcomes and is poorly tolerated, so we do not use it for this.

When to Get Your Lp(a) Checked

Because the level is fixed for life, a single test at any age is enough. In Singapore it is worth adding to a lipid panel once if any of the following apply to you: heart disease or stroke at a young age (men under 55, women under 65), a parent or sibling with early cardiovascular disease, a coronary calcium score that is unexpectedly high for your risk factors, familial hypercholesterolaemia, calcific aortic valve disease before your seventies, or heart events that keep happening despite a well-controlled LDL.

The test is a simple blood draw, usually an add-on of around SGD 30 to SGD 80 to a standard lipid profile, and no referral is needed to see a cardiologist privately. Bring any previous lipid results. Knowing the number will not give you a drug to fix it, but it will sharpen every other decision we make about your heart.

Frequently Asked Questions

Common Questions About Lowering Lp(a)

Can you lower Lp(a)?

Yes. A new class of injectable RNA drugs, including pelacarsen, olpasiran and lepodisiran, lowers lipoprotein(a) by roughly 80 to 95%. The problem is that the first large outcomes trial, Lp(a) HORIZON, showed pelacarsen did not reduce heart attacks or strokes despite that reduction. No Lp(a)-lowering drug is currently approved for preventing cardiovascular events.

Does lowering Lp(a) reduce heart attack risk?

On the current evidence, not clearly. The Lp(a) HORIZON trial lowered Lp(a) by about 80% for several years in people with established heart disease and found no significant reduction in cardiovascular events. Genetic data suggest lifelong low Lp(a) is protective, but lowering it later in life may be a different matter. Other trials are still running.

What is a normal Lp(a) level?

Most laboratories treat below about 30 mg/dL (roughly 75 nmol/L) as low risk, and 50 mg/dL (about 125 nmol/L) or above as elevated. Risk rises continuously with the level. The Lp(a) HORIZON trial enrolled people at 70 mg/dL and above. Because it is genetically set, one measurement in your lifetime is enough.

Should I get my Lp(a) tested in Singapore?

It is worth checking once if you have had heart disease young, have a strong family history of early cardiovascular disease, have an unexplained high coronary calcium score, or keep having events despite a good LDL. It is a simple blood test, usually an add-on of around SGD 30 to SGD 80 to a lipid panel.

Is there a drug to lower Lp(a) available in Singapore?

No. There is no approved therapy anywhere for lowering lipoprotein(a) to prevent cardiovascular events, and the disappointing Lp(a) HORIZON result makes near-term approval of pelacarsen for that purpose unlikely. For now, a high Lp(a) is managed by treating every other risk factor more aggressively.

Do statins or niacin lower Lp(a)?

Statins do not lower Lp(a) and may raise it slightly. Niacin lowers it by around 20% but has not been shown to improve outcomes and is poorly tolerated. PCSK9 inhibitors lower Lp(a) by about 20 to 25%, but they are prescribed for LDL lowering, not for their modest effect on Lp(a).