Baby KJ's bespoke CRISPR therapy worked. The company built to replicate it collapsed nine days after the FDA's new pathway was announced — a sign the science is ahead of any business that could deliver it at scale.
image from grok
The FDA released draft guidance for its "Plausible Mechanism Pathway" on February 23, 2026, creating a framework for individualized genetic medicines targeting ultra-rare diseases—but EveryONE Medicines, the company built specifically to commercialize such therapies, shut down nine days later. The pathway addresses regulatory evidence requirements but explicitly does not reduce manufacturing burden, as it still mandates cGMP-compliant facilities for therapies affecting fewer than 20 patients. The Baby KJ case in February 2025 demonstrated that custom CRISPR base-editing can work (the infant is now over a year old and walking), but that success relied on an emergency IND and roughly 100 scientists working voluntarily for six months—an unsustainable model rather than a replicable business.
EveryONE Medicines shut down on March 4, 2026. Nine days earlier, the FDA had released draft guidance on a new pathway for individualized genetic medicines — the same pathway the company was built to use.
Julia Vitarello founded EveryONE Medicines after her daughter Mila became the first person in the world to receive a personalized genetic medicine. The company was supposed to be the commercial vehicle for exactly what the FDA's "Plausible Mechanism Pathway" was supposed to make scalable: bespoke CRISPR therapies for children with ultra-rare genetic diseases.
The Plausible Mechanism Pathway was announced by HHS and the FDA in November 2025 and was supposed to solve the economics problem. The original Baby KJ case had proven the science could work: in February 2025, an infant in Philadelphia born with CPS1 deficiency, a urea cycle disorder that prevents the liver from breaking down ammonia, received a custom CRISPR base-editing therapy under an emergency investigational new drug application. A team including Kiran Musunuru, a cardiologist and geneticist at the University of Pennsylvania, and Rebecca Ahrens-Nicklas, a physician-scientist at Children's Hospital of Philadelphia, assembled roughly 100 scientists and developed the therapy in six months. Baby KJ went home. He is now over a year old and walking.
The pathway announcement was accompanied by a press event featuring HHS Secretary Robert F. Kennedy Jr., who said a disease with 100 distinct mutations in the same gene would no longer require 100 clinical trials under the new framework.
The draft guidance, issued February 23, 2026, told a more complicated story. The framework applies to individualized genetic medicines for diseases affecting fewer than 20 patients and requires a manufacturing facility meeting federal standards. That means cGMP compliance — the same manufacturing standards required for any approved drug. The FDA's own language states that the most resource-intensive aspect of an individualized genetic medicine is the manufacturing process. The pathway does not reduce that burden. It reorganizes the evidence package.
"The types of individualized genetic therapies that we are trying to develop simply do not fit in a traditional model of drug development," Ahrens-Nicklas told CEN/ACS.
Baby KJ's therapy worked because an emergency IND let the team skip many of the usual steps. It was a sprint assembled specifically for one patient — a remarkable feat of voluntary scientific labor, not a business model. Scaling that to a repeatable process for additional patients with urea cycle disorders requires something structurally different. It requires a manufacturing operation that can produce a different construct for each patient, with quality controls sufficient for an FDA approval, at a cost a research institution can sustain.
A platform approach could get there: Musunuru and Ahrens-Nicklas are planning a clinical trial in 2026 that can enroll patients with any of seven different urea cycle disorders, caused by variants in any of seven different genes, under a single IND. Switching out one small part of the therapy that acts like a GPS of the genome could allow it to correct many patient-specific variants — the editing machinery stays the same; only the guide RNA changes. The research team intends to submit the primary IND for the UCD platform in 2026. The approach is scientifically coherent. The cost trajectory is theoretically favorable: if the initial treatment costs $2 million and requires a year of development, by the third patient, a new treatment could cost $100,000 and be created in a month. Ahrens-Nicklas said her team had received thousands of requests from desperate parents seeking help.
The problem is that no company currently exists to do it.
EveryONE Medicines was supposed to be that company. The company began winding down the same week the Plausible Mechanism Pathway draft guidance was released, in large part because of investor skepticism that there was a financially sound way forward for individualized medicines, according to reporting by CEN/ACS. Julia Vitarello, who co-founded the company, said at the time: "If investors do not believe in it, there are no companies in this space. The companies will not last". HHS disputed any causal link between the FDA guidance and EveryONE's shutdown.
The FDA's own track record adds texture. The agency has recently rejected several drug candidates for rare diseases, including a gene therapy for Huntington's disease from uniQure and a gene therapy for mucopolysaccharidosis type II from Regenxbio — both diseases with few or no treatment options, despite the new pathway's existence. The plausible mechanism framework does not establish a new regulatory approval pathway; it operates within existing approval frameworks. Ropes & Gray and McDermott Will both published analyses noting that cGMP compliance is still mandatory under the pathway.
Baby KJ is now a thriving toddler. His mother, Nicole Muldoon, described him at one year as walking, talking, and full of unbridled joy. He is able to handle more dietary protein and requires less nitrogen-scavenging medication, with better control of ammonia levels during ordinary childhood illnesses. No serious side effects have been observed.
The science works. The regulatory intent is real. The manufacturing wall has not moved.
Story entered the newsroom
Research completed — 0 sources registered. Baby KJ treated Feb 2025 (not 2024). CPS1 deficiency — urea cycle disorder, 1 in 800K-1.2M newborns. The FDA Plausible Mechanism Pathway (draft guidan
Draft (582 words)
Reporter revised draft (844 words)
Reporter revised draft based on fact-check feedback
Reporter revised draft based on fact-check feedback
Approved for publication
Headline selected: FDA Finally Opens Path for Custom Gene-Editing Drugs. The Company Built for It Already Died.
Published (835 words)
Get the best frontier systems analysis delivered weekly. No spam, no fluff.
Biotech & Life Sciences · 20h 4m ago · 4 min read
Biotech & Life Sciences · 1d ago · 4 min read
