Oceanside teen with SCN2A epilepsy took his first unassisted steps after a drug made for his mutation

"He carries a more complex variant that requires an allele-selective approach because he has both seizures and autism," said Dr. Olivia Kim-McManus, the UC San Diego neurologist who treated Dalby at Rady Children's Hospital ^[5]. His particular variant, and the mix of symptoms it produces, set the terms for the drug. His mother, Kelly Del Real, said he was "the first human ever to be treated with this molecule" ^[15].

The denominator here is one patient, with no comparison group. In place of a control, a single case can offer a long, stable baseline. Dalby was diagnosed when he was almost five ^[1] and had his first dose at 14 ^[7], roughly nine years apart ^[16]. Over his life he tried 12 or 13 anti-seizure medications ^[3]. "And nothing really helped very much," Del Real said ^[3].

Against that record, the timing matters. "He started trying to take steps unassisted. So the first time in his entire life," Del Real said ^[8]. A change that arrives within months of a first dose, after years in which a dozen drugs did little, is harder to put down to ordinary development than a gradual improvement would be. Kim-McManus described the clinical impression. "It's hard to deny what we were seeing at the beginning," she said. "So that was very exciting and hopeful." ^[12]

The effect sizes need more care. The step count and the 90 percent seizure reduction come from his mother's account to KPBS ^[9]. The 50 to 100 seizures a day she described date from his early years ^[2], so the report does not specify the baseline that the 90 percent is measured against. The Nature Medicine paper documents his treatment and progress ^[11], and its measurements are the figures to set beside hers. A teenager also changes a lot between 14 and 17 for reasons unrelated to any drug. Dalby is now 17 ^[10], about three years past his first dose ^[17].

Scale is the open problem. A drug designed specifically for one mutation ^[6] does not automatically carry over to a different one. Kim-McManus said the question of whether doctors can build treatments for rare mutations with no existing therapy is still being worked out ^[13]. KPBS reported that bringing such medicines to more patients will require research and funding ^[14].

I think the case shows that a molecule can be designed for a variant no therapy addressed, given to the patient for years, and documented in a journal ^[4][7][11]. The thing it doesn't tell you is how often the next molecule, built for the next child's variant, will work as well.