Prime Editing Rewrites DNA Inside a Living Patient for the First Time

On 5 October 2026, a patient with Wilson disease received an infusion that did something no medicine had done before: it sent a prime editing machine directly into a living person's body to rewrite a faulty gene where it sits. The company behind it, Prime Medicine of Cambridge, Massachusetts, announced the dosing that month, and Nature reported on 22 October that it was the first known use of prime editing inside a patient. No results have been published yet. What has happened so far is a first step, not a cure — but it is a first step into territory that gene-editing medicine has been approaching for years without ever entering.

Wilson disease is a rare inherited disorder in which the liver cannot clear excess copper from the body. The copper accumulates, damaging the liver and, in many patients, the brain. The cause is a faulty gene called ATP7B, which normally helps move copper out of liver cells. Existing treatment is lifelong management: drugs that bind copper or block its absorption, taken every day, with side effects and no fix for the underlying defect. The therapy now being tested, called PM577a, is designed as a single intravenous infusion that corrects one specific mutation in ATP7B, known as H1069Q — the most common Wilson-causing mutation in North America and Europe. It will not help every Wilson patient; the disease can arise from hundreds of different mutations, and Prime Medicine is already working on a version aimed at another common one, R778L.

Prime editing is a refined relative of CRISPR. Standard CRISPR-Cas9 cuts both strands of the DNA double helix at a target site and relies on the cell to repair the break, which can introduce unwanted changes. Prime editing instead nicks a single strand and uses an engineered guide RNA as a template to write in a corrected sequence, without the double-strand break. It has been used before to edit human cells in a laboratory dish, and Prime Medicine previously ran a trial in which blood stem cells were edited outside the body and then returned to patients with a rare immune disorder. This trial is different in the way that matters most: nothing is taken out. Lipid nanoparticles carry the editing machinery through the bloodstream to liver cells, and the repair is supposed to happen inside the patient.

That difference is also why the honest account of this moment is mostly a list of things we do not know. The trial is an open-label Phase 1/2 study, registered as NCT07748403, expected to enrol about 42 adults and adolescents who carry at least one copy of the H1069Q mutation and are stable on standard treatment. Safety and tolerability are the primary focus; the secondary measures include copper markers in blood and urine, liver copper measured by biopsy at 24 and 48 weeks, and — the number everyone will be watching — evidence of on-target editing in liver biopsy tissue. Prime Medicine says initial clinical data are expected in 2027. Until then, there is no editing percentage, no copper number and no safety table in public. One patient dosed is not a treatment, and even a successful edit may not reverse damage already done to the liver or nervous system.

Still, the reason this first dosing is worth more than a press release is what it tests. In vivo editing of the liver has been shown to be possible with cruder CRISPR tools, but prime editing is a larger, more intricate machine to deliver, and its precision is its whole selling point: if it cannot be delivered efficiently to enough liver cells, the elegance of the rewrite does not matter. If the 2027 data show measurable editing with an acceptable safety profile, the same delivery approach becomes a platform — a way to correct, in principle, a whole class of single-mutation liver diseases with one infusion each, rather than a lifetime of drugs. If the data show little editing or a safety problem, the field learns where the real barrier sits: delivery, dose, or the editing chemistry itself.

I find that second possibility as interesting as the first. A milestone like this is often reported as though the hard part were over, when in fact the hard part has only just become measurable. For years, prime editing's promise has rested on laboratory results; now it rests on a number that will either appear in a biopsy report or not. That is a better place for a technology to be — out of the realm of demonstrations and into the realm of evidence. The patient who received the first infusion will not know for months whether the rewrite took. Neither will the rest of us. The wait is the point: for the first time, there is something real to wait for.