Genetics Said These Heart Drugs Should Work. The Patients Disagreed.

The Genetic Promise That Didn't Deliver

What makes these failures sting is the quality of evidence behind both targets.

Lp(a) is one of the most genetically validated cardiovascular risk factors in medicine. Through a technique called Mendelian randomization (essentially using people's inherited DNA differences as a natural experiment), researchers have shown that lifelong elevations in Lp(a) cause heart disease. One large analysis found that every 10 mg/dL increase in genetically predicted Lp(a) was associated with roughly a 5.8% higher risk of coronary heart disease. The causal chain looked airtight.

The IL-6 inflammation pathway had its own strong pedigree. Back in 2017, the landmark CANTOS trial proved that targeting inflammation (via a different molecule, IL-1β) could reduce cardiovascular events. Ziltivekimab was supposed to be the next step: a more precise anti-inflammatory that lowered hsCRP even more dramatically than canakinumab did in CANTOS. In the Phase 2 RESCUE trial, the suppression of hsCRP was nearly twice what canakinumab achieved.

More biomarker suppression, better outcomes, right? Wrong.

Why Genetics Can Point You to the Door but Can't Push You Through

So what went wrong? The honest answer is: we're still figuring that out. Full data presentations are expected later in 2026. But the broad lesson is already crystallizing.

Genetic evidence tells you what happens over a lifetime. A drug trial lasts a few years. Mendelian randomization estimates reflect decades of inherited exposure. A pill or injection given to a 65-year-old for three to five years is a fundamentally different intervention. Think of it like this: growing up bilingual makes you fluent, but a weekend Duolingo binge probably won't get you there.

There are other possible explanations, too. Maybe pelacarsen needed to lower Lp(a) by more, or patients needed treatment earlier in life, or the trial enrolled the wrong population. Maybe IL-6 inhibition helps in some cardiovascular contexts but not in patients who already have advanced disease and kidney problems. These aren't excuses; they're genuine scientific questions that the topline data can't resolve.

The uncomfortable truth is that genetic validation roughly doubles the success rate of drug development programs, according to published analyses. That sounds great until you remember that doubling a low number still gives you a low number. Genetics is a filter, not a crystal ball.

What This Means for the Rest of the Class

Pelacarsen's failure sends shockwaves through the Lp(a) field. Amgen's olpasiran and Eli Lilly's lepodisiran are both in large cardiovascular outcomes trials of their own. They work through different mechanisms and may achieve different magnitudes of Lp(a) lowering, so the story isn't necessarily over. But investor sentiment has taken a hit, and the burden of proof just got heavier.

For anti-inflammatory cardiovascular drugs, the picture is murkier. Canakinumab proved the concept in CANTOS but was never commercially pursued for heart disease (Novartis shelved it, partly due to a modest effect size and infection risks). Ziltivekimab was supposed to be the refined sequel. Instead, it showed that suppressing inflammation more aggressively doesn't automatically translate to fewer cardiac events.

The Bigger Lesson Pharma Can't Ignore

Cardiovascular medicine has a history of humbling assumptions. We once thought raising HDL (the "good cholesterol") would prevent heart attacks; multiple drugs that raised HDL flopped spectacularly. We thought certain diabetes drugs would help the heart; some did, some harmed it. The pattern is consistent: moving a biomarker is not the same as moving outcomes.

Genetics-guided development is still the best tool the industry has for picking targets. Programs with human genetic support genuinely do fail less often. But "less often" is not "never," and the pelacarsen and ziltivekimab results are a reminder that biology is not destiny, at least not on a clinical trial's timeline.

The road ahead for cardiovascular drug development isn't closed. It just requires more humility about what genetic evidence can and cannot promise. As one of medicine's oldest rules goes: the patient is the final arbiter. And this week, the patients spoke clearly.

Merck Just Paid $400M for a Drug That Doesn't Exist Yet

Merck just dropped $400 million on a Chinese biotech's preclinical cancer drug targeting a protein scientists called "undruggable" for 40 years. The KRAS G12D race is on, and Merck's gamble says everything about where Big Pharma is shopping for its next blockbuster.