--- title: "Fred Hutch's HSV gene therapy in 2026: mouse and guinea-pig results, no human trial, and the antivirals arriving first" kind: "Article" topic: "Drugs & public health" date: "2026-10-10" url: "https://canopticon.com/articles/fred-hutch-hsv-gene-therapy-in-2026" author: "Chris Telford" license: "CC BY 4.0" --- Every so often the news reports that herpes has been cured in a laboratory. The work behind those reports is real: a gene therapy that cuts the sleeping virus out of nerve cells and lets the body throw away the pieces. What the reports leave out is which animal it was cured in. ## What the therapy does to the virus Herpes survives by going dormant inside nerve cells, where antiviral drugs — which attack a virus while it copies itself — cannot touch it. The Fred Hutch approach sends an enzyme in after it. Martine Aubert, who runs the experiments, describes it plainly: We are using a meganuclease enzyme that cuts in two different places in the herpes virus's DNA. These cuts damage the virus so much that it can't repair itself. The enzyme is carried in by a harmless virus used as a delivery vehicle. Intended as a one-time treatment. ## In mice the virus nearly disappears The numbers that reach the news come from mice, and in mice they are close to total: ≥90% latent HSV reduction (orofacial), up to 97% (genital, dorsal root ganglia), with significant reduction in viral shedding frequency and amount. Less virus hiding, and less of it coming back out to be passed on. That is the figure that travels. ## Guinea pigs relapse like people, and the number falls Mice do not get recurrences the way people do. Guinea pigs do, which is why they are the better stand-in for a human being — and in guinea pigs the same therapy clears a fraction of what it clears in mice: In guinea pigs (natural recurrence model), latent HSV was reduced >30%, with fewer recurrences and 50% lower severity. The research does not bury this. It calls it the thing to watch: The much lower efficacy in guinea pigs (>30% latent reduction, versus 90–97% in mice) is the single most important caveat. Even inside the mice, the therapy is uneven. The nerve cluster behind cold sores is the one it reaches worst: Trigeminal-ganglia reductions (~41–48%) were notably lower than superior-cervical-ganglia reductions (~90–95%) — a delivery/completeness limitation. And every number above comes from an animal: All Fred Hutch efficacy figures are from animal models; none are human data. ## Nobody has yet asked to try it in a person There is a formal step between an animal result and a first human dose — an application to the regulator to begin a trial. It has not been made: The Fred Hutch/Jerome-lab HSV meganuclease program remains preclinical (pre-IND) as of mid-2026 — no IND has been filed and no human trial has begun. The date has been moving for years. The timeline has repeatedly slipped. Keith Jerome, who leads the lab, said in March last year that the team was still working through the preclinical studies needed prior to testing the strategy in human volunteers. Patient forums circulate a start date; the research says where those dates come from — projections of a trial beginning in the next year or two come from advocacy/secondary sources, not Fred Hutch directly — treat as unconfirmed. The distance the programme itself describes is longer than most people following it expect: a realistic IND-to-approval path is roughly 8–12+ years away. No large-pharma licensee has been announced. ## The drugs arriving first suppress, they do not remove While the cure waits, a different pipeline has been finishing trials in actual people. One long-acting antiviral, taken once a week, reported 94% reduction in HSV-2 shedding rate and 98% reduction in high viral load shedding rate, both statistically significant. Another, pritelivir, has already cleared a late-stage trial: met its primary endpoint of superiority (p=0.0047) in lesion healing in patients receiving treatment up to 28 days. Neither takes the virus out: These antivirals compete on convenience and suppression but do *not* eliminate latent virus. The third route, a vaccine, mostly closed: The vaccine field contracted sharply in late 2025: Moderna discontinued its mRNA-1608 therapeutic vaccine and GSK's candidate failed its Phase 2 efficacy endpoint. So the honest ranking for anyone who has the virus now is suppression first, by a wide margin, and the cure somewhere behind it. > **Objection: Ninety per cent is nearly a cure** > > For a virus that regrows from whatever survives, nearly is a different category from done. The research states the consequence: Because ~90% is not 100%, residual latent genomes could in principle reactivate; re-infection from a new exposure also remains possible. And the dormant virus is not in one place but spread across several separate nests of nerve cells, so a treatment has to empty all of them at once: Reaching all latent reservoirs simultaneously (trigeminal, dorsal-root, sacral, autonomic ganglia) is the central scientific challenge. ## What this explains It explains why the cure is announced every couple of years and never arrives: the figure in the headline is an animal figure, and the one step that would turn it into a human figure — a filing, a first dose — has not been taken. It explains why being first is slow: no AAV therapy has yet been approved for a latent/persistent viral infection, so there is no path already worn for this one to walk down. And it explains the strange economics underneath: the virus is carried by a large share of the world, which is both the reason to cure it and the reason it cannot be priced like a gene therapy for a rare disease — the Fred Hutch team explicitly emphasizes affordability and dose reduction. The cure is being built for a crowd, and that is slower than building one for a handful of patients.