Lp(a): The Heart Risk Factor Science Still Can't Treat

There is a cholesterol number that diet barely moves, exercise doesn't fix, and statins mostly ignore. It is set largely by your genes, it raises the risk of heart attack and stroke, and roughly one in five people carries it at a high level. It is called lipoprotein(a), pronounced L-P-little-a. Until this fall, the great hope was that a new class of drugs could finally lower it, and that lowering it would save lives. The largest test of that idea just reported. The number fell. The heart attacks didn't.

The bet

Novartis's Lp(a)HORIZON trial enrolled 8,323 patients who already had cardiovascular disease and an Lp(a) of 70 mg/dL or above. They received pelacarsen, a once monthly antisense injection designed to cut Lp(a) production, or a placebo, on top of the standard care everyone was already getting: lipid lowering drugs, blood pressure treatment, the full guideline package. The primary endpoint was the one that counts: a composite of cardiovascular death, non fatal heart attack, non fatal stroke, and urgent procedures to reopen a blocked coronary artery.

The result

On September 4, 2026, Novartis announced the trial did not meet that endpoint. Lp(a) levels were lower on pelacarsen, as the drug was designed to do. The reduction did not translate into fewer events across the trial population. There is currently no approved treatment that specifically targets Lp(a), and after HORIZON, there still isn't one on the horizon with outcomes data behind it.

A biomarker is a promise. An endpoint is a fact. HORIZON delivered the promise and missed the fact.

Read the fine print, because there barely is any

Here is the part most coverage skipped: everything above comes from a company press release. The full results, the actual size of the Lp(a) reduction, the event rates in each group, the confidence intervals, the subgroup analyses, have not been published. The National Lipid Association said the obvious out loud on September 14: topline announcements limit interpretation. A missed endpoint does not prove the drug does nothing; it proves the trial did not demonstrate a benefit on its primary measure, in this population, over this timeframe. Those qualifiers are the story.

What HORIZON does change is the burden of proof. Rival programs from Amgen and Eli Lilly, built on different molecules chasing the same particle, now have to show events, not just prettier lab values. Investors understood this instantly. Readers of longevity content should too, because half the industry sells biomarker movement as if it were bought time.

The bottom line: if someone tries to sell you an Lp(a) test plus a plan to fix the number, ask them the HORIZON question. Show me the events. Until a therapy reduces heart attacks and strokes in a published trial, a lower Lp(a) is a hypothesis wearing a lab coat.