The gene-editing drug that doesn't cut the gene — a muscle disease is the first test

Biotech & Longevity 4 min 80 sources

A new kind of therapy changes what a gene does without altering the DNA itself. Early data in three patients with a muscular dystrophy is the first human sign it works. Plus China's first solid-tumor CAR-T, and a wave of rare-disease deals.

Key takeaways

A new therapy for a muscular dystrophy showed the first human sign you can treat disease by changing how a gene is read — without cutting the DNA at all.
China approved the world's first CAR-T therapy for a solid tumor, clearing a wall that had blocked the technique for years and shifting some of the field's centre of gravity east.
Billions in deals flowed toward rare and hard-to-treat diseases this week, while a rare bright spot for the developing world — a vaccine against a parasitic worm — quietly cleared its first trial.

For a decade, gene editing has meant the same thing: find the faulty letters in the DNA, cut them, and rewrite. This week brought the clearest clinical sign yet of a different idea — leave the code alone, and change how loudly it is read.

A drug that turns a gene down, not out

Epicrispr Biotechnologies said its therapy EPI-321 increased lean muscle in three patients with facioscapulohumeral muscular dystrophy, or FSHD — an inherited disorder where muscle weakness starts in the face and upper body and can spread, sometimes to a wheelchair [61]. Each patient had more muscle six months after a single IV infusion, ranging from about half a pound to 1.3 pounds [61]. The company calls it the first clinical evidence that a therapy may suppress the genetic driver of FSHD [61].

The mechanism is the news. Conventional gene editing, like CRISPR, cuts the DNA to correct or disable a gene. Epicrispr’s approach doesn’t cut anything. It uses a disabled version of the CRISPR protein — “dead” Cas9, or the smaller Cas12F — to carry a chemical tag to a target gene and switch it off, without changing a single letter of the code [9]. Think of the genome as sheet music and the epigenome as the volume and expression markings on top: the therapy edits the markings, not the notes. Biologist Fyodor Urnov of the University of California, Berkeley, calls it using an audio mixing board to change how the same music sounds [9].

Why it matters: in FSHD, the disease comes from a gene called DUX4 switching on in muscle cells where it should stay silent. You don’t need to delete DUX4 — you need it quiet again. And because the DNA is untouched, the change is, in principle, reversible, which a cut is not.

The caveats are large. This is three patients, six months, early data from a company release — not a completed trial [61]. Russell Butterfield, an investigator on the study, called the changes “encouraging” but stressed that more follow-up is needed [61]. Epigenetic marks are also powerful and delicate: Yann Joly, a bioethicist at McGill University, warns that these tags play a central role in development and reproduction, and the field should move carefully [9].

The field is filling in behind it

Epicrispr is not alone. nChroma, a Boston company, began dosing the first patients in January with an epigenetic silencer aimed at hepatitis B — a virus that chronically infects an estimated 240 million people and that existing drugs cannot fully clear, because bits of its genome hide inside human DNA [9]. Separately, researchers are testing whether the same silencing tools can clean up donor organs during machine perfusion — switching off risky genes in a kidney before transplant, so organs now discarded over quality or infection worries could be used [30]. Same core trick, three very different problems.

China clears the wall CAR-T couldn’t

In cell therapy, a genuine first: China’s drug regulator approved the world’s first CAR-T therapy for a solid tumor [3]. CAR-T means taking a patient’s own immune T cells, engineering them to recognise cancer, and putting them back — a technique with eight approvals so far, all for blood cancers [3]. Solid tumors, like those in the stomach or pancreas, had resisted it for years.

The therapy, satri-cel from Shanghai’s CARsgen, targets a protein called Claudin18.2 that sits on stomach-cancer cells but not on healthy stomach stem cells [3]. Carl June, who led the first CAR-T therapy at Penn, called it “the first large market approval for any CAR-T” and said it should keep the field’s momentum going [3]. The approval covers heavily pretreated gastric cancer; whether it works earlier, or in other solid tumors, is still to be tested [3]. That the breakthrough came from China, not Boston or San Francisco, is part of the story: the science was pioneered in the US, but the first solid-tumor approval was not [3].

Money moves toward rare and hard diseases

The deals this week traced the same direction — toward hard, specific targets. Ipsen agreed to pay up to €700 million for Memo Therapeutics and its rare-disease asset [4], and separately signed a $1.7 billion deal to acquire Kartos for a potential add-on therapy in myelofibrosis, a bone-marrow cancer [79]. Deerfield’s Boulevard Bio penned a $1.6 billion deal for a cancer antibody from China’s Metis TechBio [19]. And Moderna is moving its mRNA platform toward in-vivo CAR-T — making the engineered cells inside the body rather than in a lab, which could simplify manufacturing for autoimmune diseases if it works; the candidate is set to enter the clinic next year [12].

The quiet one: a vaccine for a worm

Easy to miss under the deal flow: an early-stage vaccine against schistosomiasis — a parasitic worm infection that affects hundreds of millions in the developing world and has no widely used vaccine — showed strong immune memory in its first clinical trials [46]. It is early, and immune memory is not the same as protection in the field. But neglected diseases rarely get this far, and this one just did.

02 · Lesson · why it matters

The difference between deleting a sentence and turning down the volume

The most powerful way to change a system is often not to rewrite it, but to change how loudly one part of it speaks — a lighter touch, and one you can take back.

Two ways to fix a broken instruction

Imagine a recipe with one bad line: “add two cups of salt.” There are two ways to fix it. You can take a pen and cross the line out — permanent, decisive, and if your hand slips you’ve scratched out the line below it too. Or you can write a note in the margin: “ignore this line.” The words are still there. Nothing is destroyed. And if you were wrong — if that salt turned out to matter — you can lift the note off and the line comes back.

For a decade, gene editing has meant the pen. CRISPR finds the faulty stretch of DNA and cuts it. This week, a small company showed the first human sign that the margin note works too. Its therapy didn’t touch the DNA of a muscular dystrophy patient. It changed a chemical marker sitting on top of a gene — turning the volume down on an instruction the body was reading too loudly, without editing a single letter.

The code and the markings on the code

Your DNA is the text. But the text alone doesn’t run your body. On top of it sits another layer — chemical tags that tell each cell which lines to read loudly, which to whisper, and which to keep silent. That layer is why a skin cell and a brain cell carry the exact same genome and behave nothing alike. Same book, different reading.

The disease here comes not from a broken gene but from a gene read in the wrong room — switched on in muscle cells where it should stay quiet. The old instinct would be to hunt that gene down and cut it out. The new therapy just re-applies the “stay quiet” marking. One biologist described it as using a mixing board rather than a red pen: the same music, played softer.

Why the lighter touch is the whole point

It’s tempting to hear “doesn’t cut the DNA” as a technical footnote. It isn’t. It changes what the tool can safely be used for.

A cut is a one-way door. Once the DNA is severed and rejoined, there’s no taking it back, and a cut in the wrong place can damage a neighbour you never meant to touch. A marking is a two-way door. If it lands wrong, in principle you can lift it. If the disease shifts, you can re-tune. The reversibility isn’t a nice-to-have bolted on — it’s the reason a whole class of problems suddenly looks approachable: a chronic virus that hides inside our own DNA, a donor organ that needs risky genes quieted just long enough for a transplant. When the change can be undone, you can afford to try.

The humility built into the method

Here the pattern reaches past biology. The move — change the setting, not the structure; adjust the volume, not the text — is one of the gentlest ways a person can act on a system they don’t fully understand. And that last part matters, because nobody fully understands this one.

The same scientists building these tools are the first to say the marking layer is potent and only partly mapped. It governs how bodies develop and reproduce; nudging it has effects that ripple further than the one gene in view. So the prize and the warning arrive together: a way to touch the machinery more lightly, delivered by people who admit they can’t yet see all the wires they’re touching.

That’s not a reason to stop. It’s a reason to notice which tool this is. The reversible nudge isn’t a weaker version of the permanent cut — it’s the humbler one, the one that leaves room to be wrong. Three patients, six months, a company’s own early numbers: this is a first glimpse, not a settled result, and it will be read straight in the trials to come. But the shift underneath is real. We’re learning, slowly, that the surest way to change a living system may be to touch it lightly enough that it can still be changed back — including by the people who walked in certain they already knew where the bad line was.

03 · Lab · your turn

Cut or Dial

Rehearse choosing between a permanent DNA cut and a reversible gene-dial, and feel why leaving room to be wrong beats a decisive edit on a system you don't fully understand.

04 · Hope · carry this

The most hopeful thing about this new kind of medicine is not how powerful it is, but how carefully it is being built — by people learning to change the body a little more gently, and to leave a way back for when they turn out to be wrong.