Treatment & Procedure
Lipoprotein(a)
Biohacking
Lipoprotein(a) is a transport particle in the blood whose level is genetically determined and barely changes over a lifetime. That high levels drive the risk of heart attack is as cleanly established as for hardly any other risk factor. On September 4, 2026, the first large study that was meant to lower this level missed its primary endpoint.
In short
Lipoprotein(a) is an LDL-like particle with an additional protein attached, apolipoprotein(a). About one in five people has an elevated level and usually does not know it, because lipoprotein(a) is not measured in the standard cholesterol check. That a high level increases the risk of heart attack is considered as good as causally established through Mendelian randomization. That lowering it reduces this risk again, by contrast, is not established: Lipoprotein(a)-HORIZON, with 8,323 patients, clearly lowered the level and nonetheless missed its primary endpoint on September 4, 2026. Getting it measured remains worthwhile — a drug that demonstrably prevents events by lowering it does not exist.
What the value is
Imagine an ordinary LDL particle, what is colloquially called bad cholesterol. In lipoprotein(a), a long, twisted protein is additionally attached to it: apolipoprotein(a). This appendage makes the particle stickier and apparently more pro-inflammatory.
Lipoprotein(a) is therefore not a sub-item of cholesterol, but a risk factor in its own right. It does not appear in the standard lipid profile; anyone who wants to know it has to request it.
Why it has a special role
The level is inherited, is practically fixed at birth and barely changes over a lifetime. Diet has hardly any influence, and neither does exercise. Statins do not lower lipoprotein(a); they even raise it slightly.
That makes it the cardiovascular risk factor that so far cannot be changed. The stable level, however, has a practical side: a single measurement in a lifetime is enough. Anyone with early heart attacks in the family has a good reason to get it done.
What genetics shows
The link with heart attack is not only established, it is unusually cleanly established. This is thanks to a method called Mendelian randomization. At conception, nature rolls the dice on which gene variants someone gets. Some people have genetically low lipoprotein(a) from birth, others high, and this allocation is random — independent of lifestyle, income, everything.
This is essentially a randomized trial that nature has conducted over an entire lifetime. The analyses show clearly: people with genetically high lipoprotein(a) have heart attacks more often. Not mere correlation, but as close to causality as you can get without a drug.
Pelacarsen and the HORIZON study
Pelacarsen was developed out of this expectation, by Novartis together with Ionis. It is an antisense molecule: it does not act on the finished particle, but prevents apolipoprotein(a) from being made in the liver in the first place. In the preliminary studies it lowered lipoprotein(a) by up to 80 percent.
The endpoint study is called Lipoprotein(a)-HORIZON. 8,323 patients, all with established cardiovascular disease and elevated lipoprotein(a) of 70 milligrams per deciliter or more; in addition, a group with levels of 90 or more had been defined in advance. What was measured was what matters: cardiovascular death, heart attack, stroke and urgent bypass or stent procedures with hospitalization. Placebo-controlled.
On September 4, 2026, the result came: primary endpoint missed. Lipoprotein(a) went down, that worked — it just did not translate into fewer events. Novartis’s head of development said so himself: lower levels seen, but no evidence that this translates into lower cardiovascular risk. More candid than such announcements usually are.
Three explanations, none of them proven
The first is the most interesting, because it is 8 years older than the result. In 2018, Stephen Burgess and Brian Ference calculated how much lowering would be needed at all: for the same risk reduction as lowering LDL by almost 39 milligrams per deciliter, lipoprotein(a) would have to fall by a good 101 milligrams per deciliter. In absolute terms, not as a percentage. Anyone starting at 70 or 90 cannot lower by these 101 at all; even with an 80 percent reduction you end up at perhaps 60 or 70. According to this calculation, the expected effect was small from the outset — possibly too small to be seen in 5 years in 8,000 patients.
That does not mean the failure was foreseeable. Finding an explanation in hindsight is always easy. The calculation was on the table, it was discussed, and this design was chosen anyway — for understandable reasons: people with very high levels are rare, and a study needs participants.
The second explanation: all participants were already ill and being treated, with statins, blood pressure drugs, blood thinners. If the risk is already pushed down, the contribution of a further drug becomes smaller and harder to detect. The third: perhaps it is the lifelong high lipoprotein(a) that does the damage. Anyone who has had elevated levels for 60 years carries vascular changes that 5 years of medication can no longer reverse. Genetics measures an entire life, a study a few years.
What is still underway
The idea is not finished with this. 3 further drugs are in large studies: olpasiran from Amgen with results around 2028, lepodisiran from Eli Lilly around 2029, and muvalaplin, also from Lilly — interesting because it is a pill; its study runs even longer. The first two work with siRNA instead of antisense and in some cases lower more strongly.
The practical yield of the failed study lies here: anyone working on this now looks very closely at the baseline levels they include. HORIZON has shown where the bar might lie.
What is well supported
Two things hold up. First, the link: that high lipoprotein(a) contributes to causing heart attack risk is as well established through Mendelian randomization as is possible without a drug trial. Second, the drug: pelacarsen did what it was built for and clearly lowered the level. Plus an unspectacular third point: the measurement is simple, the level stable, and one determination in a lifetime is enough.
What the studies show
Lipoprotein(a)-HORIZON — pelacarsen
Placebo-controlled endpoint study with 8,323 patients, all with established cardiovascular disease and lipoprotein(a) of 70 milligrams per deciliter or more, plus a subgroup of 90 or more. Endpoint: cardiovascular death, heart attack, stroke, urgent bypass or stent procedures with hospitalization. Result of September 4, 2026: level lowered, primary endpoint missed. Full data only at a scientific congress.
Burgess and Ference 2018
The calculation that, in retrospect, defines the bar: for the same risk reduction as lowering LDL by almost 39 milligrams per deciliter, an absolute reduction of a good 101 milligrams per deciliter would be needed — mathematically unachievable with inclusion thresholds of 70 and 90.
The three ongoing programs
Olpasiran (Amgen) and lepodisiran (Eli Lilly) work with siRNA, with results around 2028 and around 2029; muvalaplin (Lilly) is a pill with a longer study duration. None of these programs has so far shown a benefit on events.
Where the data stop
There is currently no drug that has been proven to prevent heart attacks by lowering lipoprotein(a). None. Anyone selling something else is selling a hope. Genetics has shown that lipoprotein(a) contributes to the damage; it has not shown that you get rid of it again by lowering it later. These are two different statements, and they are constantly lumped together.
The explanations for the failure also remain explanations: too small an absolute reduction, patients already well treated, vascular damage accumulated over a lifetime — each plausible, none proven. In addition, the full HORIZON data have not yet been published. Until they are available, the manufacturer’s announcement is all there is.
Status, approval and legal
Pelacarsen is not approved, and neither are olpasiran, lepodisiran and muvalaplin — all four are investigational substances. No approved drug for lowering lipoprotein(a) exists. What is available is the measurement: a laboratory test that has to be requested separately and, depending on the reason, paid for yourself. The topic has no doping relevance.
Safety
The announcement of September 4, 2026, contains no information on the tolerability of pelacarsen; without the full publication nothing reliable can be said about it, neither good nor bad. The blood draw itself is routine. The real risk lies elsewhere: a high level is no reason to resign yourself, but the argument for being more consistent with the modifiable factors. The interpretation of an elevated level belongs in a conversation with a doctor, not in self-interpretation of a lab report.
BK-Score Well studied – effect not confirmed
| Human evidence | 8 | |
|---|---|---|
| Mechanism | 7 | |
| Safety data | 5 | |
| Hype gap | 5 | |
| Track record of use | 6 |
Evidence 8, because with Lipoprotein(a)-HORIZON there is a large, placebo-controlled study in 8,323 patients over 5 years with hard endpoints – it simply turned out negative and has not yet been fully published. Mechanism 7, because the target structure is confirmed in humans: pelacarsen lowered lipoprotein(a) by up to 80 percent in the preliminary studies, and Mendelian randomization quantifies the lifelong contribution of the particle – but the chain from lowering to event has just broken in humans. Safety 5, because although 5 years of controlled data were collected, the announcement of September 4, 2026, says nothing about it; there are no reliable safety statements on pelacarsen. Hype 5, because the clean genetic finding is regularly turned into a promise of therapy, while the manufacturer’s own announcement was worded unusually candidly. Use 6, because the measurement has been available for years as a regular laboratory value, whereas the lowering drugs are available exclusively in studies. The negative direction refers to lowering as a treatment, not to the level as a risk marker: there, the HORIZON data speak against the hoped-for benefit.
The score rates the state of knowledge, not the substance. “Safety data 9” means well studied – not harmless.
Subjective assessment by Biohacking Kompakt based on published scoring rules – not a scientific rating and not a medical recommendation. Rules and all ratings (German)
Frequently asked questions about lipoprotein(a)
How often do I need to have lipoprotein(a) measured?
Once in a lifetime is enough. The level is genetically determined, is practically fixed at birth and barely changes over the years. It is not included in the standard cholesterol check and therefore has to be requested explicitly.
Can I lower my level through diet or exercise?
Practically not. Diet has hardly any influence, and neither does exercise. Statins do not lower lipoprotein(a) either; they even raise it slightly. That is why this risk factor is considered the one that so far cannot be changed.
How many people are affected?
Roughly speaking, about one in five people has an elevated level. Most do not know it, because the level is rarely measured. Anyone with early heart attacks in the family has a good reason to know it.
Why did the HORIZON study miss its primary endpoint?
No one knows for sure. Three explanations are being discussed: the absolute reduction was probably too small with baseline levels of 70 milligrams per deciliter or more, the participants were already heavily treated because of their existing disease, and vascular damage that has developed over decades may no longer be reversible in 5 years. All three are plausible, none is proven.
Is the idea finished with this?
No. 3 further drugs are in large studies: olpasiran with results around 2028, lepodisiran around 2029 and muvalaplin as a pill with a longer study duration. The first two in some cases lower more strongly. Whether one of them shows what pelacarsen did not show is open.
What should I do if my level is high?
That belongs in a conversation with a doctor and not in self-interpretation. The established consequence is to be more consistent with everything else: LDL, blood pressure, smoking, exercise. A risk factor you cannot change is the argument for taking the modifiable ones more seriously, not the excuse to give up on them.
Related
- Often mentioned togetherCutting down on alcohol
- Often mentioned togetherZone 2 & VO2max training
- Mentioned togetherMuscle as an organ (grip strength & myokines)
- Mentioned togetherSleep & sleep hygiene
- Mentioned togetherBeetroot / nitrate
- Mentioned togetherMicronutrient concentrates (LaVita & Co.)
- QuestionLongevity: Where do I start?
Sources
- Lipoprotein(a)-HORIZON (pelacarsen, Novartis and Ionis), announcement of the missed primary endpoint of September 4, 2026
- Lipoprotein(a)-HORIZON, registry entry (NCT04023552, 8,323 patients, Novartis)
- Burgess, Ference and colleagues, JAMA Cardiology 2018 — Mendelian randomization: a 101.5 mg/dl lipoprotein(a) reduction corresponds to a 38.67 mg/dl LDL reduction
- Kamstrup et al., JAMA 2009 — Mendelian randomization on genetically elevated lipoprotein(a) and heart attack risk
- Olpasiran (Amgen), ongoing endpoint study OCEAN(a) Outcomes (NCT05581303), results around 2028
- Lepodisiran (Eli Lilly), ongoing endpoint study ACCLAIM-Lp(a) (NCT06292013), results around 2029
- Muvalaplin (Eli Lilly), oral investigational substance, ongoing phase 3 endpoint study (NCT07157774)
Open in the database – with search, filters and comparison (German app)
Information only, not medical advice and not a usage or dosage recommendation. Prescription-only and unapproved substances belong in the hands of a physician. Last updated: 2026-09-19.