Pediatrician Kristin Sohl has lost count of how many times parents of children with autism have asked her for a prescription for leucovorin — the drug thrust into the spotlight after President Donald Trump touted it at a White House event this fall.
Since September, despite the rising queries, Sohl has typically told her patients no.
Early clinical trials of the drug showed hints of promise in boosting communication and cognition for some children with autism. But the studies have been small, often just a few dozen participants. Normally, approval by the Food and Drug Administration comes only after years of large-scale testing. But Trump’s pledge to fast-track the drug in September, bypassing that process, has left many doctors on the front lines divided.
“It leaves me as a practicing physician with a lot of unanswered questions,” said Sohl, a professor of pediatrics at the University of Missouri School of Medicine, who has been working in the field of autism for over 20 years.
As interest in the drug surges, Facebook groups devoted to it are swelling in membership, message boards are inundated with questions, and Google searches are climbing. Physicians, who typically rely on evidence-based guidelines and clear treatment algorithms, are finding that with leucovorin they must — lacking robust scientific data — improvise. Some are cautiously moving forward with prescribing the drug, but many are still holding off.
At Children’s National Hospital in Washington, D.C., neurodevelopmental pediatrician Sinan Turnacioglu said the hospital convened a meeting of various departments — including those specializing in autism, developmental pediatrics, genetics and psychiatry, as well as primary care doctors — to come up with a systemwide policy. Their conclusion: that they would like to see more robust research before prescribing it.
Peter Crino, chair of neurology at the University of Maryland School of Medicine and who runs a clinic for neurodivergent adults, likewise said he believes the medication is not ready for prime time.
“People are asking me a lot about it, but I do not prescribe it. Gosh I hope there will be something to the drug and it will help people in the future, but the data is simply not there yet,” he said.
Limited evidence
Each conversation Sohl has with families unfolds differently, she said, shaped by a child’s history, a parent’s worry, a flicker of hope. But the script she follows is steady: she lays out what research has shown — and what it doesn’t — about the treatment, then asks what the family hopes the drug might change.
In a field with no cure and few therapies, she uses that same framework to guide discussions about the other latest supposed breakthroughs drifting across social media — broccoli extracts, CBD oil, stem cell therapy, and more. The goal isn’t to dismiss any ideas outright but to ground them in evidence, or show the lack of it, before families decide what to do next.
For leucovorin, Sohl’s main message is that “we’re not on solid science yet.” However, there are “potential suggestions of benefit.”
Leucovorin or folinic acid has a long history of use in the context of cancer for about 50 years. It’s been shown to protect healthy cells from the toxic effects of one particular chemotherapy drug and to enhance the effectiveness of another one. Side effects were very minimal but in cancer patients have included nausea and fatigue.
For some children with autism, the immune system may produce antibodies that block the body’s ability to move folate — a vitamin essential for cell growth and DNA production — into the brain. Leucovorin, a prescription form of folate, offers a potential workaround. It crosses the blood-brain barrier by a different route, delivering the nutrient where it’s needed.
The U.S. clinical trial that got Trump’s attention is being conducted by Richard E. Frye, a pediatric neurologist who was formerly an associate professor at Arizona Children’s Hospital in Phoenix. Its design was considered the gold standard — a randomized, double-blind placebo-controlled trial — but it only had 48 children, ages 5 to 12, in it. In the trial — published in 2018 in the journal Molecular Psychiatry — the drug was well-tolerated and the parents and doctors reported improvements in communication and behavior.
Frye said in an interview that leucovorin did not work on all of his patients. But it did work for many and that children with no verbal utterances began showing meaningful word approximations, for example, and that those with phrase speech began forming full sentences.
There have been four subsequent trials in other countries, and all four of them also reported significant improvements and no serious harm. But they were also very small. A study in France with 19 patients was published in 2020, in Iran with 55 patients in 2021, in India with 40 patients in 2024, and in China with 80 patients in 2025.
The Trump administration latched onto promising research and promoted efforts to expand access to leucovorin for autism, despite the lack of large-scale clinical trials.
But since then, doctors have been proceeding cautiously. At least two influential medical societies have come out with their own interim recommendations. The American Academy of Pediatrics and the Society for Developmental & Behavioral Pediatrics both do not recommend the routine use of leucovorin for children with autism. But the AAP left an opening for doctors to prescribe it, stating that pediatric care providers “are encouraged to engage in shared decision-making with families who inquire about or request leucovorin, providing clear information about current evidence and potential risks.”
Crino said that many medical research papers — including those on leucovorin — are written in ways families can understand, and he encourages patients and their families to read the primary studies themselves. He often reviews the papers with them, he added, pointing out the limitations of the research. In the 2024 study, which was published in the European Journal of Pediatrics, for example, the authors reported that many children showed improvements in speech, but none went from nonspeaking to speaking, and the study offered no evidence about whether those changes affected daily life.
“There is a lot going on in scientific research that is getting twisted,” he said.
Turnacioglu said that some of his patients receive leucovorin from other providers. In those cases, he focuses on monitoring their progress by first establishing a baseline assessment of language and adaptive skills and then repeating the same evaluations periodically to track any changes.
In those cases, he focuses on monitoring their progress by first establishing a baseline assessment of language and adaptive skills and then repeating these evaluations periodically to track any changes.
He said the growing interest in leucovorin reflects a broader shift toward more personalized autism treatments, fueled by recent research that supports what clinicians have long observed: autism is not a uniform condition that exists along one continuous spectrum, but rather a collection of distinct conditions that have been grouped under a single label. As a result, different people may require different treatments.
“We don’t yet have enough information to use those findings to guide leucovorin treatment,” Turnacioglu said. “But it’s the kind of direction I’m excited about — figuring out which patients are going to respond to particular treatments.”
An exception
Sohl is part of a team of pediatricians who helped draft the AAP guidelines.
The patients that have approached her are all ages and across the spectrum, including adults and individuals with strong verbal skills. For months, she’d explained her reasons for holding back on leucovorin, and most families accepted them.
Then, in October, a patient sat across from her and she began to wonder if this might be an exception.
He was a teen boy she describes as minimally speaking, whom Sohl had been treating for 10 years. She was impressed by his knowledge of the research on leucovorin, his deep and realistic understanding of the potential risks and benefits, and his eagerness to document any changes both quantitatively and in narrative form. Sohl will be meeting with her patient each month to go over any changes.
“I have low expectations, his mom has low expectations, he has low expectations. But we all agreed it was worth a therapeutic trial,” Sohl said.
With the recent national attention, information about leucovorin has been spreading online far faster than through the slow, methodical channels of medical research, where studies and peer-reviewed papers can take years to emerge. She learns from the parents and patients who are often the first to encounter new ideas circulating in their communities and online.
Sohl tells families that while the drug has shown very little in the realm of side effects, this is in the context of adults with cancer, not children with autism. She said she emphasizes that she does not think this is a dangerous medicine, but there has not been enough research.
“I think it’s my duty as a doctor to say that I don’t know,” Sohl said, “and I want you to know I don’t know.”
