The Drug Made for One

In January 2017, Timothy Yu, a neurologist and geneticist at Boston Children's Hospital, heard about a six-year-old girl in Colorado named Mila Makovec. Mila had been diagnosed with CLN7 Batten disease, a fatal neurodegenerative condition that destroys the nervous system. She was losing her vision, her speech, her ability to walk. She was having 15 to 30 seizures a day, each lasting up to two minutes.

Ten months later, Yu's team had identified her specific genetic mutation, designed a drug to correct it, manufactured the drug, received FDA approval for a single-person trial, and begun injecting it into her spinal fluid. The drug was named milasen, after her.

Milasen did not save Mila's life. Her disease was too advanced by the time treatment began. She died on February 11, 2021, at age 10. What milasen did was prove that a bespoke drug could be designed, built, and administered to one person with one mutation in one gene, on a timeline measured in months.

That proof became the foundation for every individualized therapy that followed.

The Mutation No One Had Seen Before

Batten disease is caused by mutations in several different genes. Mila's form, CLN7, is autosomal recessive, meaning both copies of the gene must be defective. Her doctors found a mutation in one copy. They could not find the second.

Yu's team sequenced Mila's genome more deeply and discovered the problem. A retrotransposon, a piece of mobile DNA, had spontaneously inserted itself into a noncoding region of her second CLN7 gene copy. The insertion disrupted RNA splicing, preventing the cell from producing functional CLN7 protein. This was not a known mutation. It was specific to Mila. No one else in any database had the same error.

That specificity was the obstacle and the opportunity. No existing drug targeted this mutation. No clinical trial was enrolling for it. No pharmaceutical company had an economic reason to develop a treatment for one person. The standard drug development pathway, designed for conditions affecting thousands or millions, had nothing to offer a child with a mutation shared by no one.

Building the Drug

Yu realized that an antisense oligonucleotide (ASO) could solve the problem. ASOs are short synthetic strands of nucleic acid designed to bind to a specific sequence of RNA and modify how the cell reads it. In Mila's case, the ASO would mask the retrotransposon insertion and restore normal splicing. The cell would read the gene correctly and produce CLN7 protein.

Yu's team designed candidate ASOs targeting Mila's specific mutation. They tested them in her skin cells grown in a laboratory dish. Three candidates corrected the faulty splicing. The team selected the most effective one, manufactured it under good manufacturing practices, conducted toxicology studies in rats, and prepared an application to the FDA for compassionate use under an emergency investigational new drug (IND) protocol.

The timeline from identification of the mutation to first injection: roughly 10 months. Traditional drug development for a rare disease takes 10 to 15 years.

What Happened

Mila began receiving milasen via intrathecal injection (directly into her spinal fluid) in January 2018. Over the course of treatment, her seizure frequency dropped. The seizures that did occur became shorter, often under one minute. Her neurological decline, which had been accelerating, appeared to slow.

The results, published in the New England Journal of Medicine in 2019, were cautious. The disease was not reversed. The damage already done to Mila's brain was irreversible. Milasen reduced seizures and may have slowed progression, but Mila's Batten disease continued. She lost remaining abilities over the following years.

Mila died three years after treatment began. Her mother, Julia Vitarello, founded Mila's Miracle Foundation to advance the field of individualized medicine that Mila's treatment had opened.

What Milasen Proved

Milasen demonstrated three things that the field of rare disease medicine had theorized but never executed.

First, a drug can be designed for a single person's unique mutation. The ASO chemistry is modular. The same platform (synthetic oligonucleotide, intrathecal delivery) can be adapted to target different mutations by changing the nucleotide sequence. Designing a new ASO does not require starting from zero. It requires changing the address.

Second, the development timeline can be compressed from years to months. Traditional drug development involves years of preclinical research, formal Phase I/II/III trials with statistical endpoints, and regulatory review designed for therapies that will be administered to thousands. For a child with a fatal disease and a unique mutation, that timeline is a death sentence. Milasen proved that the FDA's emergency IND pathway could be used to deliver an individualized therapy while the science was still young.

Third, each individualized therapy generates data that makes the next one faster. Yu's experience with milasen informed the design, manufacturing, toxicology, and regulatory approach for subsequent individualized ASO programs targeting other mutations in other genes. The first individualized ASO required building the infrastructure from scratch. The second required adapting it. The tenth will require adjusting it.

The Regulatory Bridge

Before milasen, individualized therapies existed in a regulatory vacuum. The FDA had no formal framework for evaluating drugs designed for one person. Expanded access (compassionate use) was available, but it was designed for acute emergencies, not systematic development programs. Right to Try, the 2018 law that allowed terminally ill people to access unapproved therapies, generated almost no useful data because it included no requirement for structured data collection.

Milasen helped build the case for the Plausible Mechanism Framework (PMF), which the FDA has since developed to evaluate therapies where the mechanism of action is well characterized even if the specific therapy has been tested in only a small number of people. The PMF allows master protocols under which multiple individualized therapies targeting different mutations in the same gene can be evaluated through a single regulatory application.

The logic is direct. If an ASO targeting Mutation A in Gene X works by exon skipping, and exon skipping is a validated mechanism with a known safety profile, then an ASO targeting Mutation B in the same gene via the same mechanism carries less regulatory uncertainty. The first treatment in a gene family bears the full burden. Each subsequent treatment in that family benefits from the data generated before it.

The Economics

Milasen's development cost has been estimated at roughly $2 million. The cost of manufacturing a custom ASO is measured in thousands of dollars for the molecule itself. The expensive components are GMP manufacturing, toxicology studies, and regulatory preparation.

By comparison, Zolgensma, the gene therapy for spinal muscular atrophy, costs $2.1 million per dose and was developed through a traditional pharmaceutical pipeline over many years. The per-treatment costs are converging. The development timelines are not: Zolgensma took more than a decade from concept to approval. Milasen took less than a year.

The cost curves matter because they determine how many conditions can realistically be addressed through individualized therapy. If each individualized ASO costs $100 million and takes 10 years, the approach is viable for a handful of conditions. If each costs $1 to 3 million and takes months, the approach is viable for hundreds.

What Mila Left Behind

Mila Makovec did not live to see the field her treatment created. She was six when her family began looking for help. She was seven when milasen was first injected. She was ten when she died.

Since milasen, multiple individualized ASOs have been developed for other children with other mutations in other genes. Terry Pirovolakis is building a gene therapy called Trophos for his son's ultra-rare neurological condition through public fundraising and academic collaboration. The FDA's PMF framework is formalizing the pathway milasen proved was possible.

Hundreds of families contacted Yu's lab after the milasen publication, asking whether the same approach could help their child. The answer, for many of them, is that the chemistry exists, the regulatory pathway is forming, and the question is whether the data infrastructure exists to make each treatment inform the next.

The drug was made for one. The infrastructure it requires is built for all of them.