When Timothy Yu developed milasen, a custom drug for a young girl named Mila with Batten disease, he ignited a spark in the field of personalized medicine. Milasen was the first medicine specifically designed for a single person, and it was developed in just about a year.
In response to milasen, nonprofit organizations have emerged calling for the development of personalized therapies for the estimated 400 million people living with rare diseases worldwide. These medicines, colloquially called “n of 1” therapies, are often made to treat debilitating genetic conditions that are too rare to garner interest from pharmaceutical companies.
Now, in a study published Wednesday in Nature, Yu, an attending physician at Boston Children’s Hospital in the division of genetic and genomics, and his colleagues have developed another personalized medicine, called atipeksen, for a young child with a genetic disorder called A-T, or ataxia-telangiectasia.
A-T is caused by a mutation in the genetic code of a key enzyme involved in DNA damage repair. The mutation prevents proper processing of the enzyme, rendering it inactive. Ultimately, this inactivity causes severe neurodegeneration and reduces a person’s life span to 25 years on average.
Like milasen, the development of atipeksen was sparked by another passionate parent to two sons with A-T.
“I was at a small conference … presenting our proposal, at that point, to try to develop a drug for [Mila],” said Yu. “This is an audience of scientists, physicians, and pharmaceutical executives — and there was a gentleman in the back of the room who kept on asking these pesky pointed questions, especially directed at the pharmaceutical industry, about why they weren’t moving faster for particular orphan or rare diseases.”
The “gentleman” was Brad Margus, the founder and president of the A-T Children’s Project (ATCP), which is dedicated to finding a cure for A-T through fundraising, recruiting scientists and clinicians, and compiling patient data.
While his own two sons had missed the window to receive A-T treatments, Margus was still adamant on finding treatments for other A-T families. “I just thought it was obscene, that just because [A-T] was rare, that first-rate science wasn’t being done on it,” he said.
Margus had taken a lot of factors into his own hands through ATCP. At the time of meeting Yu, he had already compiled a database containing the sequenced genomes of 235 children with A-T. This was exactly the data that Yu needed to answer his next big question on the broader applications of individualized antisense oligonucleotide (ASO) therapies. “We knew that this field deserved a more systematic assessment of what the true opportunity was” for antisense oligonucleotides, said Yu, and “that A-T provided that opportunity to actually address it in a non-anecdotal way.”
ASOs are specialized pieces of synthetic DNA that bind to a specific area of an individual’s faulty genetic code, effectively patching it so the cell can no longer recognize the faulty mutation. They could potentially be used to patch a mutation in the DNA damage repair enzyme responsible for A-T, restoring its proper processing and function.
With the patient data, Yu and his colleagues were able to perform systematic analyses to not only identify all the genetic mutations that contributed to A-T in this population, but also to assess which mutations would be amenable to treatment with ASOs.
They concluded that 15% of children within this cohort had favorable mutations. From this 15%, Yu and his team selected one mutation that had a high likelihood of responding favorably to ASOs, then chose to develop a therapy for a young girl.
“I had to manage the [A-T] community — all the families — and explain how we would pick the first kid,” said Margus. “We made it really clear that it would be based on whichever kid had a mutation that was very tractable with an ASO approach,” he said. “If we could just treat one kid successfully, it’d be huge.”
Yu’s team began testing ASOs on the girl’s cells to see if they could restore function of the faulty DNA damage repair enzyme. Within a year, they had developed a therapy ready for administration. The child has now been receiving her bespoke medicine for over three years, starting treatment at just age 2.
“It struck us that this case was kissings cousins with Mila, a very similar situation with an ASO-amenable mutation,” said Yu. However, “this case was one that we had identified … at a very, very young age, unlike Mila.”
Milasen had shown promising initial results, but the disease was already so advanced that Mila died at age 10, three years after starting her treatment. Yu hopes that starting therapeutic intervention on the girl with A-T at such a young age may have a more significant impact on staving off disease progression.
“It was part of our clinical thinking that the chance to intervene in these diseases early would be really important,” said Yu. “Mila is a good example of that.”
The study only reports the tolerability of the treatment, so the clinical outcome of the young girl with A-T is still unknown. However, Yu said he’s “been really pleased that the treatment’s been very well tolerated, and she seems to be doing very well.”
To Toshifumi Yokota, a professor of medical genetics at the University of Alberta who was not involved in the study, Yu’s results provide hope for the expanded utility of ASO treatments, saying that “the framework is possibly applicable to many other genetic diseases.”
Specifically, Yokota believes this approach can be used on genetic diseases that arise from short mutations. However, they say that genetic mutations that cause shortened versions of a protein, or cause a shift in the genetic code that results in a new protein sequence altogether, will likely be more resistant to ASOs.
ASOs for some genetic diseases have seen commercial success, like Spinraza, which treats spinal muscular atrophy, and Amondys 45, which treats Duchenne muscular dystrophy. The drugs can treat a large percentage of patients since there are only a handful of genetic mutations that cause these genetic diseases. In contrast, Yu and his team identified 469 potentially disease-causing mutations in 235 kids with A-T.
While Yu’s advancement is promising, Margus emphasized how important it is to remember that the vast majority of children with A-T are not viable candidates for ASO therapy. Both Margus and Yu hope that some of this remaining population will be responsive to other genetic technologies, like nucleotide base editors and siRNAs.
According to Yu, however, “if we’re going to use these [genetic] tools to their ultimate potential, we’ll have to actually learn to use them much more nimbly than we are currently.” This will require significant “attitude shifts” from people in regulatory agencies, academia, and the pharmaceutical industry alike, said Yu.
Julia Vitarello, a co-founder of the N=1 Collaborative and Mila’s mother, agreed that regulatory processes for personalized medicines need to change. “With [the FDA’s] current model, they can’t possibly receive an IND for every single child. That’s not feasible,” she said. An IND or investigational new drug application must be filed with the Food and Drug Administration to start any human clinical trials.
Encouragingly, Margus said that so far, the FDA has been receptive to loosening these regulations at some point. “They just want to see data, and that’s fair,” he said. “If they see that we’ve done, say, 12 ASOs — changing up the sequence — but the toxicity has been exactly the same, then maybe they’ll start to lower the bar, which makes it faster and cheaper for us.”
Vitarello said she is actively exploring reimbursement plans for businesses that create custom medicines through her company, EveryONE Medicines. Still other companies like Quantile Health are working to reimburse insurers, which could help democratize access to custom medicines. Currently, however, financing is still a big hurdle for scalability.
Another issue, according to Vitarello, is the lack of infrastructure to allow the scaling process to happen. The N=1 Collaborative is working to facilitate a productive model between academia and industry.
“We’re finally getting companies involved,” she said. With industry starting to step in, Vitarello believes a viable company model is on the horizon.
Yu, Margus, and Vitarello all commented on the fact that personalized medicine is not a new concept. “The good news is that there are precedents,” said Yu, referencing CAR-T therapies that are individually customized for patients with certain types of cancer.
According to Vitarello, the hesitancy around “n-of -1” treatments is a matter of perception, and questions why society has “decided it’s worth taking a risk and spending a lot of money … when it comes to end-stage cancer,” but not for rare genetic diseases.
Margus agreed that personalized medicine is misrepresented. “I think it’s the way it’s framed,” he said. “If we say that we’re going to do drug development for each kid, it sounds terrible.” With enough case studies, however, Margus envisions a “plug and play” framework for individualized medicines, in which kids are taken through sequencing screens for ASO qualification and into a routine pipeline for ASO development.
Perhaps the biggest misrepresentation of all, said Vitarello, is the name “n of 1” itself. “It implies that these are one-offs. And it gives the sense that there’s just not a lot of them,” she said. In light of Yu’s discovery that up to 15% of children with A-T stand to benefit from ASO therapies, Vitarello is hopeful that a similar statistic will apply to other rare genetic diseases.
For Vitarello, “it doesn’t actually matter what the number is. It matters that you’re going after the underlying genetic cause,” she said. “Even if there’s one, or there happens to be 15 more, you’re still considering each patient genetically, as one.”