· Genetic Therapy
At fifteen, a boy who had never walked without support took his first independent steps — not because of surgery or physical therapy, but because doctors silenced a single broken copy of one gene.
His treatment was designed for exactly one patient: him.
Researchers at UC San Diego and Rady Children's Institute for Genomic Medicine built individualized for two children with a severe, single-gene form of epilepsy, cutting seizures by up to 90% and helping one teenager walk independently for the first time — offering an early, real-world model for how personalized genetic medicine could scale beyond one-off miracles.
A colorful illustration of a brain with waves emanating from it, representing the SCN2A gene therapy study. Credit: UC San Diego Health Sciences.
A Teenager Takes His First Independent Steps
For most families, a teenager taking a few unassisted steps across a room wouldn't make the news. For one 15-year-old with a severe, genetically driven form of epilepsy, it marked something his medical team had spent years working toward: a treatment aimed not at his symptoms, but at the single genetic error causing them.
Researchers led by the University of California San Diego and Rady Children's Institute for Genomic Medicine designed an individualized gene-silencing therapy matched precisely to this teenager's specific mutation, and to that of a second, younger child with the same underlying condition. Published in Nature Medicine, the results describe two years of dramatically reduced seizures, meaningful developmental gains, and a possible blueprint for treating other diseases caused by a single faulty gene.
What Is SCN2A-Related Developmental Epileptic Encephalopathy?
SCN2A-related developmental epileptic encephalopathy, usually shortened to SCN2A-DEE, is a rare but severe childhood epilepsy syndrome. It's also one of the more common single-gene causes of , which is notable given how many neurodevelopmental conditions don't trace back to one identifiable genetic culprit at all.
Children with SCN2A-DEE typically experience seizures beginning early in life, alongside significant delays in movement, communication, and other developmental milestones. The severity and specific symptoms vary from child to child, but the underlying cause is remarkably precise: a single mutation in one gene.
One Broken Copy in a Sodium Channel
The gene at the center of this condition, SCN2A, provides instructions for building part of a sodium voltage-gated channel — a protein embedded in neurons that regulates how sodium ions move in and out of the cell. That movement is fundamental to how neurons fire electrical signals in the first place.
A mutation in SCN2A disrupts that regulation, making affected brain cells abnormally excitable. The downstream effects can include uncontrolled seizures, developmental delays, autism-related behaviors, movement difficulties, and gastrointestinal problems, depending on exactly how the mutation alters the channel's behavior.
Most SCN2A mutations arise de novo, meaning they appear spontaneously rather than being inherited from either parent. That randomness is part of what makes the condition so difficult to anticipate or screen for in advance.
Why Ordinary Anti-Seizure Drugs Fall Short
Standard anti-seizure medications are designed to dampen abnormal electrical activity broadly across the brain. For many forms of epilepsy, that's an effective strategy. For SCN2A-DEE, it's often only partially helpful.
The core problem is that conventional medications don't correct the actual genetic mechanism driving the disorder — they manage downstream symptoms without addressing the malfunctioning sodium channel itself. That's part of why families and researchers have been searching for a therapy that intervenes earlier in the chain of causation, at the level of the gene rather than the resulting electrical storm.
A Therapy Built for Exactly One Patient
Because SCN2A-DEE is rare, and because the specific mutation differs from patient to patient, the research team pursued something unusual in mainstream medicine: two separate n-of-1 clinical trials, each designed and evaluated for exactly one child.
The children were nine and 14 years old when their individual trials began. Each received a treatment engineered specifically around the exact SCN2A mutation they carried, rather than a one-size-fits-all drug meant to work across many different patients and mutation types.
Principal investigator Olivia Kim-McManus, MD, of UC San Diego School of Medicine and Rady Children's Institute for Genomic Medicine, described the underlying philosophy directly: "The therapy is deliberately designed to target the individual's genetic diagnosis." It's precision medicine taken about as far as precision medicine currently goes.
How an ASO Silences Just the Broken Copy
Every person carries two copies of most genes, one inherited from each parent. In SCN2A-DEE, only one of those two copies carries the disease-causing mutation; the other remains perfectly normal.
The researchers designed short synthetic strands of DNA called allele-selective antisense oligonucleotides, or ASOs, engineered to recognize a harmless stretch of genetic sequence sitting right next to each child's specific mutation. By targeting that neighboring sequence rather than the mutation itself, the ASO can selectively suppress activity from the mutated copy of the gene while leaving the healthy copy completely untouched, free to keep producing normally functioning protein.
Doctors delivered the ASOs directly into the spinal fluid while each child was under anesthesia, allowing the treatment to reach the central nervous system where the mutated gene was causing the most damage.
Two Years, Two Children, Measurable Change
Each child received a repeat dose every two to three months and effectively served as their own control, with researchers comparing seizure frequency, medication use, development, and behavior before and after treatment began.
After two years, the nine-year-old, who had previously experienced seizures almost daily, showed a 26% decrease in seizure frequency. The 14-year-old showed a far more dramatic 90% reduction, eventually reaching stretches of entirely seizure-free days. Both children were able to reduce or discontinue some of their anti-seizure medications, and both showed improvements in language, motor abilities, sensory processing, and adaptive behavior, alongside a reduction in autism-related behaviors.
The older patient's long-standing gastrointestinal problems also improved, reducing his need for related medication. Researchers reported no serious side effects, and routine lab testing, ECGs, and EEGs all remained stable throughout the study. Kim-McManus summarized the overall picture plainly: the changes appeared across nearly every measure the team tracked.
Walking Again — and Losing It Without the Next Dose
The most striking single result came at age 15, when the older patient walked independently for the first time. But the story didn't end there, and the detail that follows says as much about the therapy's nature as the achievement itself.
Because ASOs change how a gene is expressed rather than permanently rewriting the underlying DNA, their effect fades over time, which is why repeat dosing is necessary. During the trial, the researchers noticed the teenager beginning to lose some of his newly gained ability to walk independently as his next scheduled dose approached.
With FDA approval, the team shortened the interval between doses. According to Kim-McManus, he has been walking independently ever since. It's a vivid, almost literal demonstration of the treatment actually working: take it away, even briefly, and the specific ability it restored begins to recede.
From One-Off Miracle to Repeatable Model
Kim-McManus is careful to note that these treatments remain investigational, and two patients, however striking their results, don't constitute proof that this approach will work broadly. But the significance of this study isn't just what happened to these two children — it's that a rapid, systematic process now exists for going from an individual genetic diagnosis to a matched therapy.
That process, repeated for two separate children with two distinct mutations in the same gene, is what the researchers describe as a possible framework for extending highly personalized treatments beyond isolated cases toward larger groups of patients who share similar genetic profiles, even if no two patients' mutations are ever identical.
What This Could Mean Beyond Epilepsy
SCN2A-DEE is just one of many conditions caused by a single mutated gene. The same underlying strategy — designing an ASO to selectively silence a specific mutated allele while sparing the healthy copy — is, in principle, adaptable to other neurological and non-neurological diseases with a similar genetic architecture.
Kim-McManus described watching the field's reception shift over time: an idea that once drew skeptical, almost science-fiction looks is now being taken seriously enough to spread from academic labs into the pharmaceutical and biotech industry, where the infrastructure exists to eventually scale personalized approaches like this one to more patients.
Manohar Insight
What makes this study remarkable isn't just the clinical outcome; it's the economics of attention it represents. Two children, two custom-built therapies, years of individualized work by an international research team — on paper, that looks like the least scalable model of medicine imaginable.
Yet the actual achievement here is procedural, not just biological: the researchers demonstrated a repeatable pipeline for turning a specific genetic diagnosis into a matched treatment within a clinically meaningful timeframe. That's a very different claim than curing one rare disease. It's closer to proving that the "one patient, one therapy" model can work at all, which is the necessary first step before anyone can respond seriously to whether it can work at scale.
The detail I keep returning to is the dosing interval story — a teenager's walking ability quietly receding as a treatment wore off, then returning once the schedule was adjusted. It's a small, almost mundane logistical fix. But it's also a rare, unusually direct piece of evidence that a genetic therapy is doing exactly what it was designed to do, in real time, in front of the people delivering it.
Join the Signal
If personalized gene-silencing therapies like this one become more common, should healthcare systems prioritize funding "n-of-1" treatments for individual rare-disease patients, or wait until a therapy can be shown to help larger groups first? Share your thoughts in the comments.
Discussion