CRISPR in 2026: what gene editing can actually cure, and what is still hype
There is a real CRISPR cure on the market, a baby was saved by a medicine built for him alone, and its most advanced in-body trial just hit a serious safety scare. Gene editing in 2026 is genuinely miraculous and genuinely limited, often in the same breath. Here is the honest state of it.
Gene editing crossed from promise to medicine in 2026, but narrowly. There is exactly one approved CRISPR therapy, Casgevy for sickle cell and beta thalassaemia, and at roughly $2.2 million, punishing chemotherapy and months in hospital, only 64 people took it in all of 2025. One baby was saved by a treatment designed for his mutation alone. Meanwhile the frontier, editing genes inside the living body, hit a serious safety setback: a severe liver reaction that halted the most advanced trial and drew an FDA hold. The cures are real. So are the limits, the cost and the risk.
CRISPR won its inventors the Nobel Prize in Chemistry in 2020, and for a few years after that the story was all upside: a pair of "genetic scissors" that could, in principle, fix almost any disease written in our DNA. In 2026 the story is more interesting and more honest than that, because the technology has now met the real world of patients, prices and safety. It has produced a genuine cure, a genuine miracle, and a genuine tragedy, all inside about eighteen months. Holding all three in view at once is the only truthful way to describe where we actually are.
Is CRISPR actually curing people in 2026?
Yes, in one specific and remarkable case. Casgevy (exagamglogene autotemcel, or exa-cel), from Vertex and CRISPR Therapeutics, was the first CRISPR medicine approved anywhere, cleared in the US and UK in late 2023, and it is still the only approved CRISPR therapy in the world. It treats two inherited blood disorders, sickle cell disease and beta thalassaemia, and for the people it works for, it is not a management drug but a one-time functional cure.
The way it works is clever. Doctors take out a patient's own blood stem cells, use CRISPR to switch back on a gene that makes fetal haemoglobin (a form of the oxygen-carrying protein that our bodies normally shut off after birth), and infuse the edited cells back in. The corrected cells make healthy blood, and the crippling pain crises of sickle cell largely stop. In July 2026 the FDA widened the approval to children as young as two, making an estimated 5,500 more American children eligible. As a demonstration that you can edit a human genome to end a disease, it is settled. It works.
Why have so few people actually got the cure?
Because a cure you cannot reach is not much of a cure, and Casgevy is brutally hard to reach. The list price for the drug alone is about $2.2 million, and once you add the hospital stay and everything around it, analysts estimate the all-in cost approaches $3 million per patient. That is only the start of the barrier.
The bigger obstacle is the treatment itself. Before the edited cells go back in, the patient's existing bone marrow has to be wiped out with busulfan chemotherapy, a punishing conditioning regime that means weeks to months in hospital and, in most cases, permanent infertility. Add a tiny number of authorised treatment centres and you get the real-world result: Vertex reported just 64 patients infused worldwide in all of 2025, against roughly 100,000 Americans living with sickle cell disease alone. The science is finished. The delivery, the cost and the sheer ordeal of it are not, and that gap is the honest headline of gene editing in 2026.
The baby who got a medicine made just for him
The most moving thing CRISPR did this cycle was not a product at all. In 2025, a Philadelphia infant known as baby KJ (KJ Muldoon) was born with CPS1 deficiency, a rare urea-cycle disorder that lets ammonia build to toxic levels in the blood, often causing brain damage or death in the first months of life. There was no treatment for his specific mutation, so a team at Children's Hospital of Philadelphia and Penn built one from scratch, aimed at his DNA alone.
They used base editing, a more precise descendant of CRISPR, packaged in fat droplets and infused into his bloodstream to correct the faulty gene in his liver. He got his first dose at around seven months old, improved, and went home. The work was published in the New England Journal of Medicine in May 2025, and it is a real landmark: the first time a bespoke, one-patient gene-editing therapy was designed, built and given inside a matter of months. The caveat is the obvious one. A treatment engineered for a single child is a proof of concept, not a scalable medicine, and the regulatory path for "one patient, one drug" barely exists yet. It shows what is possible. It does not yet show what is repeatable.
Can you edit genes inside the body, not in a lab?
This is the leap the whole field is chasing, and it is exactly where 2026 got sobering. Casgevy edits cells in a dish outside the body. The far bigger prize is in-vivo editing: a single injection that edits your genes where they sit, no cell harvest, no chemotherapy, no months in hospital. That would turn gene editing from a heroic procedure into something closer to a normal treatment.
The furthest-along attempt was Intellia's nex-z (nexiguran ziclumeran), a one-time infusion that uses CRISPR to switch off a faulty gene in the liver in patients with transthyretin (ATTR) amyloidosis, a disease where a misfolded protein builds up and damages the heart and nerves. Early results were genuinely exciting. Then the programme hit a wall. A patient reportedly in his early eighties, dosed in a Phase 3 trial in late September 2025, developed a severe (Grade 4) liver injury, the kind of treatment-related toxicity that stops a trial cold. Intellia paused dosing on 27 October, and the FDA placed both Phase 3 trials on hold two days later.
Here the careful version matters, because it is easy to get wrong. The patient was hospitalised and died on 5 November, but the autopsy and the trial's lead investigator attributed his death to septic shock from a perforated ulcer, rather than to the gene-editing itself, so the alarming headline ("patient dies in CRISPR trial") is not quite the true story. What was real, and serious, was the liver toxicity itself, and it was enough to halt the most advanced in-vivo CRISPR programme in the world. The company notes that reactions of that grade showed up in under one percent of trial participants. The FDA let the nerve-disease trial resume in January 2026 and the heart-disease trial that March, both under tighter safety monitoring, and by August 2026 Intellia had restarted enrolling patients in the programme. The promise of editing genes inside the body is intact. The idea that it is routine or risk-free is not.
What are base editing and prime editing?
If plain CRISPR is a pair of scissors that cuts the DNA to make a change, the newer tools are scalpels. Base editing, developed in David Liu's lab at the Broad Institute, chemically swaps one DNA letter for another without cutting the double strand at all. Prime editing goes further, rewriting short stretches of code like a genetic find-and-replace. Both avoid the risky double-strand break that older CRISPR relies on, and both are now reaching patients.
The most important sign of where this is heading is VERVE-102, from Verve Therapeutics, now owned by Eli Lilly. It is an in-vivo base editor that switches off the PCSK9 gene in the liver to lower cholesterol, and in an early trial a single infusion cut LDL cholesterol by up to 62 percent, holding for well over a year. That matters because high cholesterol is not a rare disease, it is one of the most common conditions on Earth, and it hints at a future where a one-off edit replaces a lifetime of daily pills. Meanwhile Beam Therapeutics is running a base-editing therapy for sickle cell and hopes to file with the FDA by the end of 2026, and Prime Medicine is taking prime editing into the clinic for liver and lung disorders. This is the real trajectory: from editing cells in a lab for rare diseases, toward editing the body itself for common ones.
What about designer babies?
This is the fear that shadows every gene-editing story, and it is worth being precise. Everything above edits somatic cells, the ordinary cells of a living patient, so the changes are not passed to their children. Editing an embryo is a different thing entirely: those changes are heritable, rewriting the human line, and that is the bright ethical line the field has agreed not to cross.
One man crossed it. In 2018 the Chinese scientist He Jiankui used CRISPR on human embryos, producing the world's first gene-edited babies, twin girls. He was condemned worldwide, convicted of practising medicine illegally, and served three years in prison. Since his release in 2022 he has been attempting a comeback, floating proposals to edit embryos to lower the risk of Alzheimer's and reportedly courting cryptocurrency funding. It remains, rightly, off-limits: heritable human editing is banned or heavily restricted almost everywhere, the science is nowhere near safe enough, and the ethics are not close to resolved. The gene editing that is actually curing people in 2026 is emphatically not this.
The state of gene editing in 2026, at a glance
| Therapy | What it targets | How it edits | Status in 2026 |
|---|---|---|---|
| Casgevy (exa-cel) | Sickle cell, beta thalassaemia | Ex vivo, your own stem cells edited in a lab | Approved (the only one), now age 2+ |
| Baby KJ's therapy | One infant's CPS1 deficiency | In-vivo base editing, custom-built | One-off success, not yet a product |
| Intellia nex-z | ATTR amyloidosis | In vivo, one infusion edits the liver | Phase 3 hit an FDA hold on liver toxicity (nerve trial resumed 2026) |
| VERVE-102 (Lilly) | High cholesterol (PCSK9) | In-vivo base editing, one infusion | Early trial strong, Phase 2 next |
| Beam sickle cell | Sickle cell | Ex vivo base editing | In trials, FDA filing hoped by end 2026 |
| Prime Medicine | Alpha-1 antitrypsin, Wilson's | In-vivo prime editing | Entering the clinic |
So where does gene editing actually stand?
Somewhere far more remarkable, and far more fragile, than either the hype or the backlash suggests. We have one CRISPR cure that genuinely ends a cruel disease, yet almost no one can afford it or endure the chemotherapy it demands. We have a baby alive today because scientists wrote a medicine for his mutation alone, a genuine glimpse of personalised medicine that is not yet repeatable. We have an in-vivo frontier that could one day make gene editing as simple as a jab, and that just paid for that ambition with a life. And we have a heritable-editing line that should stay uncrossed, being tested by the one man who already crossed it.
The correct posture, then, is neither the breathless "we can cure anything now" nor the weary "it never lives up to the promise". Gene editing is real medicine as of 2026, doing things that were pure science fiction a decade ago. It is also expensive, arduous, still risky, and available to very few. Both of those are true at once, and anyone telling you only one of them is selling something. For more from the lab, see the Science section, and our looks at what a brain-computer interface can really do and quantum computing's real breakthrough.
For the first time, we can see whirlpools on the Sun's surface. They may explain why its atmosphere is so strangely hot
The world's largest solar telescope has taken the sharpest images ever of the Sun's surface and caught something long predicted but never seen there: tiny plasma whirlpools, only about 20 km across, the same wave-curl you get in ocean waves and clouds. They may be part of the answer to a decades-old mystery. Here is what was found, and what it does and does not prove.
JWST found a 'black hole star' in the early universe, and it might solve a real puzzle
The James Webb telescope spotted an object 660 million years after the Big Bang that looks like a giant star but is really a black hole wrapped in so much gas it disguises itself. If the interpretation holds, it helps explain one of astronomy's genuine mysteries: how monster black holes appeared so impossibly early. Here is the careful version.
A blood test can now help diagnose Alzheimer's. Here's what that really means
Diagnosing Alzheimer's used to require an expensive brain scan or a spinal tap. In 2026 a simple blood test crossed from the research lab into the clinic, sharply improving how accurately doctors can identify the disease. It is a genuine turning point, and also not a crystal ball. Here is what these tests do, and what they do not.