There are roughly 7,000 known rare genetic diseases. For the overwhelming majority of them, there is no cure — only management, symptom control, and hope that the science catches up before the disease wins.
For a baby boy named **KJ**, born in late 2024 with one of the rarest and most lethal of these conditions, the science did catch up. Just barely. And the story of how it happened is one of the most extraordinary events in the history of medicine.
**The Disease**
KJ was born with **CPS1 (carbamoyl-phosphate synthetase 1) deficiency** — a rare inherited disorder in which the liver lacks a critical enzyme involved in the urea cycle. The urea cycle is how the body disposes of ammonia, a toxic byproduct of protein metabolism. Without a functioning CPS1 enzyme, ammonia accumulates rapidly in the blood, crossing the blood-brain barrier and causing neurological damage, coma, and death.
CPS1 deficiency affects roughly 1 in 1.3 million births. Without a liver transplant — which carries its own enormous risks for an infant — the prognosis is devastating. Standard management involves protein restriction and medications, but these do not address the underlying genetic defect.
**The Decision**
In late 2024, physicians at the **Children's Hospital of Philadelphia (CHOP)** and Penn Medicine met with KJ's family and made an extraordinary proposal: they would attempt to build a gene therapy from scratch, specifically designed for KJ's exact mutation.
This had never been done before for CPS1 deficiency. And it had almost never been done at this speed, for any disease, for any patient.
**The Technology**
The team used a technique called **base editing** — a refinement of CRISPR gene editing pioneered by David Liu's laboratory at the Broad Institute of MIT and Harvard. Where conventional CRISPR cuts DNA (creating double-strand breaks that carry a risk of unintended edits), base editing makes **single-letter changes** — swapping one DNA base for another without cutting the double helix.
Think of it as the difference between crossing out a word and retyping the whole paragraph versus using correction fluid to fix a single letter.
The specific base editor used by the CHOP team was designed to correct KJ's particular mutation — a change in the CPS1 gene that had rendered his enzyme non-functional. The therapy was then packaged inside **lipid nanoparticles** (the same delivery technology that made the COVID-19 mRNA vaccines possible) and administered intravenously, allowing it to reach liver cells directly.
**The Timeline**
⚡ **Diagnosis:** shortly after birth ⚡ **Decision to pursue custom therapy:** within weeks of diagnosis ⚡ **Therapy design, synthesis, and testing:** ~6 months ⚡ **First administration:** May 2025 ⚡ **Result:** measurable improvement in KJ's ability to tolerate protein in his diet — a direct readout of improved urea cycle function
The treatment was described in detail in *The New England Journal of Medicine* in May 2025, alongside a companion Guardian report that brought KJ's story to global attention.
**What Makes This Different**
KJ's treatment is extraordinary for several reasons:
🧬 **It's the first personalised in-vivo base editing therapy ever delivered to a human being.** Previous gene therapies have used viral vectors, or targeted more common mutations shared across many patients. This was built for one person, for one mutation, delivered directly into living tissue.
🧬 **It was designed in six months.** Conventional drug development takes 10–15 years and costs billions of dollars. The team at CHOP, working with collaborators including Penn Medicine and the Broad Institute, compressed this to half a year — raising the possibility that this speed of personalised medicine could become a template.
🧬 **It worked.** KJ improved. He is not cured — base editing delivered by lipid nanoparticles is transient; his cells turn over and the edited versions dilute over time, requiring repeat doses. But the proof of concept — that a custom gene therapy can be designed and delivered to a living human infant in months — is now established fact.
**The Bigger Picture**
Dr. Kiran Musunuru, a cardiologist and gene editing researcher at Penn Medicine who helped lead the effort, told reporters: "Every child with a rare disease deserves a shot at a therapy. We hope this becomes a model."
The implications reach far beyond CPS1 deficiency. There are thousands of rare monogenic diseases — conditions caused by a single faulty gene — for which no treatment exists because the affected patient population is too small to make conventional drug development economically viable. Personalised base editing changes the equation. If the same approach can be applied to dozens, then hundreds of rare diseases, it would represent one of the largest expansions of treatable conditions in medical history.
KJ's story isn't the end of anything. It's the proof of concept that starts everything. 🧬
*Sources: The New England Journal of Medicine (May 2025) · The Guardian (May 15, 2025) · Children's Hospital of Philadelphia · Penn Medicine · Broad Institute · ScienceAlert · The Week*
