The founding idea at Verve Therapeutics arrived with an almost suspicious neatness. Some people are born with mutations that switch off a liver gene called PCSK9. They tend to carry less LDL cholesterol and suffer fewer heart attacks. Sekar Kathiresan, a cardiologist and human geneticist, had spent years studying these naturally protected people. His wager was direct: if medicine could reproduce their biology in adults, perhaps cardiovascular care could become less like brushing your teeth and more like installing a fire sprinkler.
That distinction matters. Statins work, as do injectable antibodies and RNA medicines that inhibit PCSK9. But they work only while people keep taking them. The chronic-care chain has many weak links: prescriptions, refills, copays, appointments, side effects, missed doses and years of motivation. Verve proposed an irreversible shortcut - one intravenous infusion carrying molecular machinery into liver cells, where it would alter a single DNA letter and disable PCSK9 for good.
A drug that edits its own instruction manual
VERVE-102, the company's lead program, is not a consumer product and is not approved. It is an experimental medicine for adults with heterozygous familial hypercholesterolemia or premature coronary artery disease. These are people whose risk is not theoretical: LDL can remain dangerously high despite standard therapy, and heart disease often arrives early.
The treatment has three essential pieces. Messenger RNA tells a liver cell to briefly manufacture an adenine base editor. A guide RNA supplies the genomic address. A lipid nanoparticle carries both through the bloodstream. The editor changes an A to a G at a splice site in PCSK9, disrupting the gene without making the double-strand DNA break associated with conventional CRISPR cutting. The machinery fades. The edit is meant to remain.
The molecular delivery run
This is where Verve differs from the familiar alternatives. A statin reduces cholesterol production. Repatha and Praluent are antibodies that block the PCSK9 protein. Leqvio uses small interfering RNA to suppress production and is dosed twice yearly after its starter schedule. All are reversible. Verve is trying to intervene upstream, once. That offers an adherence advantage but removes the comfort of simply stopping the drug if a late problem appears.
What failed first
Verve's first clinical version was VERVE-101. In 2023, early Heart-1 data did something important: they showed that in vivo base editing could reduce PCSK9 and LDL cholesterol in people. At a 0.45 mg/kg dose, the company reported LDL-C reductions as high as 73 percent. The biological thesis had crossed from slides into human blood tests.
Then the delivery problem announced itself. In April 2024, Verve paused enrollment after a participant in that dose cohort developed an asymptomatic Grade 3 rise in alanine aminotransferase, a liver enzyme, and Grade 3 thrombocytopenia, or low platelets. Both abnormalities were transient, but "transient" does not make them trivial when the proposed treatment permanently changes DNA. The company and its safety board stopped adding participants and prioritized VERVE-102.
That pivot was less a change of mind than execution of a contingency. Kathiresan later said Verve had explicitly built multiple candidates to reduce delivery risk. VERVE-102 kept the same base editor and guide RNA but replaced the ionizable lipid and added GalNAc, a sugar ligand that binds the ASGPR receptor abundant on liver cells. The first formulation depended on the LDL receptor for entry. The replacement had two doors.
There is a useful company-building lesson hiding in the biochemistry. Verve identified the least settled part of its stack early. Human genetics had validated the target. Base editing had a plausible mechanism. Getting enough cargo into the right tissue, with tolerable toxicity, was the bottleneck. So the company paid for redundancy before a crisis forced it. Many startups keep a backup pitch deck; Verve kept a backup nanoparticle.
The second swing, in 35 people
The first VERVE-102 participant was dosed in May 2024. Initial data from 14 people, released in April 2025, showed dose-dependent LDL lowering and no treatment-related serious adverse events at that cutoff. At 0.6 mg/kg, mean LDL-C fell 53 percent, with a maximum reduction of 69 percent in one participant. The FDA granted Fast Track designation.
A fuller Phase 1 report published in the New England Journal of Medicine in May 2026 covered 35 participants across six dose levels. At the highest dose, 1.0 mg/kg, mean PCSK9 protein fell 88 percent and mean LDL-C fell 62 percent. Fifteen participants had at least one year of follow-up, and reductions appeared durable. No dose-limiting toxic effects occurred. Researchers did observe mild-to-moderate infusion reactions and temporary ALT elevations. One participant developed aspiration pneumonitis that investigators assessed in the context of pre-existing reflux disease.
Those numbers are a signal, not a victory lap. Phase 1 studies are designed mainly to assess safety and dose. Thirty-five people cannot reveal rare harms. LDL reduction is a validated cardiovascular risk marker, but this trial did not show that VERVE-102 prevents heart attacks or strokes. Nor does one year establish lifetime durability. Larger randomized trials and longer observation must do that work.
What did the idea cost?
There is no announced treatment price because there is no approved treatment. The cost we can see is the company-building bill. Verve raised $58.5 million in its 2019 Series A, another $63 million in 2020, $94 million in a 2021 Series B, and $306.7 million gross in its IPO. Collaborations with Vertex and Lilly brought cash, equity investment, research reimbursement and the promise of milestones. In 2024 alone, Verve reported $204.3 million in research and development expense, $32.3 million in collaboration revenue and a $198.7 million net loss.
Lilly had been moving closer for years. It first partnered on the LPA program, then acquired rights previously held by Beam to opt into Verve's PCSK9 and ANGPTL3 programs. After reviewing the early Heart-2 results, Lilly agreed in June 2025 to buy the company. The acquisition closed the next month for $10.50 a share, roughly $1 billion in cash. A non-tradeable contingent value right can add $3 a share, about $300 million, if the first patient enters a U.S. Phase 3 VERVE-102 trial within ten years.
Who is this really for?
Today, the practical users are trial participants and clinical investigators. The intended first patients are adults at unusually high lifetime risk whose LDL remains uncontrolled. That is a sensible entry point: the medical need is severe enough to justify experimental permanence, and the benefit can be measured quickly in blood. If later studies establish safety and fewer cardiovascular events, the addressable population could widen to people with established atherosclerotic disease.
It may not work as a broad-market therapy if rare off-target edits or immune effects emerge, if LDL lowering fades as liver cells turn over, if manufacturing or administration remains too expensive, or if payers prefer cheap reversible drugs. It also cannot solve every cause of heart disease. Smoking, blood pressure, diabetes, inflammation and behavior do not vanish when PCSK9 does. A permanent medicine for a common risk factor will need a remarkably clean safety profile because many candidates can already be treated effectively with therapies that can be stopped.
The part worth copying
The same playbook will not travel everywhere. The liver is unusually accessible to lipid nanoparticles and has well-characterized surface receptors. A startup aiming for the brain, heart muscle or lungs cannot assume the same delivery trick will follow. Nor should founders confuse a biomarker win with a clinical-outcomes win. Verve's target comes with decades of genetic and pharmacologic support; a fashionable target with thin human evidence offers much less margin for error.
Verve now sits inside Lilly, a company with cardiometabolic scale, manufacturing depth and enough capital to run large outcomes trials. Its story is unresolved in the way the best biotech stories are unresolved: the mechanism has worked, the first formulation faltered, the replacement looks better, and the consequential tests lie ahead. If VERVE-102 succeeds, patients may one day receive cholesterol treatment with an expiration date measured in decades. If it fails, Verve will still have supplied a precise lesson. The edit can be elegant. Delivery, evidence and trust are the medicine.