This year, gene-editing technology was customized to fix mutations in a single patient’s genes for the first time.

December 8, 2025, 8 AM ET

For a decade after its discovery, CRISPR gene editing was stuck on the cusp of transforming medicine. Then, in 2023, scientists started using it on sickle-cell disease, and Victoria Gray, a patient who lived with constant pain—like lightning inside her body, she has said—got the first-ever FDA-approved CRISPR gene-editing treatment. Her symptoms vanished; so did virtually everyone else’s in the clinical trial she was a part of.

This year, the technology has started to press beyond its next barrier. Most of the 8 million people globally who have sickle-cell disease share the same genetic mutation; treating rare disorders will require dealing with many different mutations, even within the same disease. And although rare diseases affect 30 million Americans in total, relatively few people are diagnosed with each one. Fyodor Urnov, the scientific director of UC Berkeley’s Innovative Genomics Institute (IGI), showed me a list of rare diseases and pointed to one carried by only 50 people. “Who’s going to work on a disease with 50 patients?” he asked. And even within one disorder, each person might need their own customized CRISPR treatment. Drug developers have little financial incentive to spend years and millions of dollars designing therapies that may need to be tailored to literally one person.

The technology is ready to treat at least some of these diseases, though. “There’s a whole toolbox now that can target arguably any part of the genome pretty precisely,” Krishanu Saha, a gene-editing researcher at the University of Wisconsin at Madison, told me. If researchers could build one CRISPR platform for a single disease, or even several similar ones, and tweak that template to suit each patient, they could target extremely rare disorders more quickly and economically. Maybe the first patient’s treatment for a disease takes $2 million and a year of development; by the third patient, the cost should be down to, say, $100,000 and a month of development, Urnov said, because you’ve already proved that the reused components are safe.

“We have been moving in the direction of thinking about CRISPR as a platform for some years,” Jennifer Doudna, the IGI’s founder, who shares the Nobel Prize for discovering CRISPR gene editing, told me. But, in her mind, 2025 was the first time many people understood its potential. A baby named KJ Muldoon is a big reason why.

In February, Muldoon became the first child to receive one of these customized CRISPR gene-editing treatments, tailored to fix his specific mutation. People born with his rare genetic disease, a type of urea-cycle disorder, have about a 50 percent chance of living past infancy. If they do, they live with extreme developmental delays and usually require a liver transplant. But when he was six months old, Muldoon got his bespoke treatment, and now he’s a healthy 1-year-old. His therapy was proof that custom gene-editing treatments can work and that they can be spun up relatively quickly, yet safely.

His treatment also gives scientists a chance to try the platform approach. The next child treated for a urea-cycle disorder should now be able to receive a CRISPR treatment from Muldoon’s template, tweaked to their unique DNA. CRISPR technology uses guide RNA, a molecular GPS of sorts, to send an editor protein to a particular address in someone’s DNA. Targeting a different mutation just means changing the address. Muldoon’s case put more momentum, too, behind personalized gene editing in general. The federal government recently announced two major programs that offer funding to scientists working on personalized treatments for rare diseases. The focus now, Doudna said, is figuring out how to make customized CRISPR “available to anyone who needs it.”

For years, one of the main roadblocks has been the U.S. drug-regulatory system. Its approval processes were designed for traditional drugs that help many people, not a bespoke treatment that helps one child in Philadelphia. The FDA has considered each treatment, even for the same disease, as a different drug. Biochemically, two therapies might be the equivalent of a pizza with pepperoni and another with artichokes. But under the FDA approval process, “you go back to square one. You recertify the oven. You recertify the person who throws the disk of dough. You confirm the cheese is still safe to eat,” Urnov, who was also part of the team that designed Muldoon’s treatment, said. The FDA has been trying to change that process over the past few years, and last month, two of its top officials, Marty Makary and Vinay Prasad, announced a new drug pathway that could speed up approvals for personalized rare-disease treatments. The framework was inspired in large part by the success of Muldoon’s therapy. (The FDA did not respond to a request for comment.)

The new pathway opens the door to the platform approach that scientists have hoped to take. If researchers could prove they’d successfully treated a small number of patients for one rare genetic disease, they could continue customizing treatments for other mutations, and potentially also for similar conditions. That streamlined process could finally attract for-profit players—the best shot at actually getting these customized therapies to patients en masse, Doudna said. “If we’re able to bundle trials together so that we’re able to treat multiple related diseases without starting from scratch, that could completely change the economics of treating rare disease,” she said.

The first clinical trials in this model will begin soon. Urnov and his colleagues plan to investigate a platform for rare immune disorders; Rebecca Ahrens-Nicklas and Kiran Musunuru, the geneticists who treated Muldoon at the Children’s Hospital of Pennsylvania, told me they are planning to start one this winter for children with various types of urea-cycle disorder. If all goes according to plan, another child should receive a treatment based on Muldoon’s in the near future.

Working this way does put more responsibility on scientists to test their therapies thoroughly, Ahrens-Nicklas said. Gene editing can go wrong: A treatment may accidentally alter the wrong part of a patient’s DNA, or the delivery mechanism could trigger a deadly immune reaction in their body. “If you have to treat fewer subjects in order to get that approval, you want to make sure that you’re really robustly measuring the safety on those few subjects” and communicating any risks to the wider gene-editing community, she said. But done well, these trials are a major step toward getting more custom treatments out to more people.

All of the researchers I spoke with emphasized that these are early days. Because of how the current gene-editing delivery mechanisms work, scientists are mostly limited to treating disorders in the blood and liver. And researchers are focused on single diseases, or groups of similar ones, for now. Their dream would be to have a CRISPR platform that could address many disparate disorders, but the current reality is that many, many families will still go without bespoke therapies. Muldoon’s treatment “took a team of people at both nonprofits and for-profit companies in multiple countries working at a scale I have never seen before,” Doudna said. And they changed his life. His parents weren’t sure if he’d ever be able to sit upright on his own, but recently, Muldoon took his first steps. The press has dubbed him a “miracle baby.” Now miracles like his need to become commonplace.

About the Author

Nancy Walecki is an associate editor at The Atlantic.

Explore More Topics