Single-dose in-vivo base editing lowers LDL cholesterol by up to 62 per cent in first-in-human trial

A single intravenous infusion of an adenine base editor directed at the PCSK9 gene lowered LDL cholesterol by up to 62 per cent in adults at high cardiovascular risk. The result comes from a phase 1 trial published in the New England Journal of Medicine on 13 August 2026, whose first author is Scott B. Vafai of Verve Therapeutics in Boston. It is the first test in humans of the therapy, VERVE-102.

VERVE-102 combines a messenger RNA encoding an adenine base-editor protein with a guide RNA targeting PCSK9. The two are encapsulated in a lipid nanoparticle that incorporates N-acetylgalactosamine, which directs the particle to liver cells. The rationale is drawn from human genetics: people who naturally carry loss-of-function variants in PCSK9 have lower LDL cholesterol and fewer atherosclerotic cardiovascular events than people who do not.

The study was open-label and used a single-ascending-dose design. It is registered on ClinicalTrials.gov as NCT06164730 and enrolled adults with heterozygous familial hypercholesterolaemia or premature coronary artery disease. Thirty-five participants across six dose cohorts received one infusion at doses ranging from 0.3 to 1.0 mg of total RNA per kilogram of body weight, and each had at least 28 days of follow-up.

Reductions in blood PCSK9 protein were dose-dependent. Mean falls ranged from 51 per cent at 0.3 mg/kg to 88 per cent at 1.0 mg/kg. LDL cholesterol fell by 9 per cent at the lowest dose and by 62 per cent at the highest, an absolute reduction of 78 mg per decilitre at 1.0 mg/kg. The reductions appeared durable throughout follow-up, which was at least one year in 15 of the 35 participants.

No dose-limiting toxic effects were recorded. Mild-to-moderate infusion-related reactions and transient elevations in alanine aminotransferase were observed. Aspiration pneumonitis occurred in one participant who had gastro-oesophageal reflux disease.

The trial was funded by Verve Therapeutics, a wholly owned subsidiary of Eli Lilly. Its sites and academic investigators span the United Kingdom, Canada, Australia and New Zealand, among them Jörg Täubel of Richmond Pharmacology and St George's, University of London; Riyaz S. Patel of the Institute of Cardiovascular Science at University College London and Barts Health NHS Trust; Jaimini Cegla of Imperial College London; Handrean Soran and Bilal Bashir of the NIHR-Wellcome Trust Clinical Research Facility at Manchester University NHS Foundation Trust; Daniel Gaudet of the Université de Montréal and ECOGENE-21; Liam R. Brunham of the University of British Columbia; David E. Newby of the British Heart Foundation Centre for Research Excellence at the University of Edinburgh; Stephen J. Nicholls of the Victorian Heart Institute at Monash University; Jean-Claude Tardif of the Montreal Heart Institute; and Steve E. Humphries of the Institute of Cardiovascular Science at UCL.

The paper appears in the New England Journal of Medicine, volume 395, issue 7, pages 648 to 659, with the digital object identifier 10.1056/NEJMoa2601283 and the PubMed identifier 42187087.

The work sits alongside the CTX310 programme, which targets ANGPTL3 and whose durability data appeared in the same journal on 28 August 2026. Together the two mark the point at which one-and-done gene editing moves from ultra-rare disease into common cardiovascular risk.

Three questions remain open. The first is long-term safety in an otherwise healthy population. The second is the absence of any way to reverse an edit that spares the germ line but is permanent in the treated cells. The third is whether a cardiovascular outcomes trial can be designed for a therapy that is given once.