In a highly anticipated announcement on Monday (22 SEP 2025), U.S. President Donald Trump outlined his administration’s approach to what he called the “crisis of autism.” He devoted much of the news conference to exhorting parents to “spread out” vaccinations over several months or years and avoid taking any acetaminophen during pregnancy or giving it to young children — advice scientists say is not justified by evidence. However, federal officials also said the Food and Drug Administration (FDA) would approve a drug called leucovorin, a form of folate (vitamin B9), as “the first FDA-recognized treatment pathway for autism” — an unusual move for a pill only tested in a few small studies.
Those studies do hint at possible benefits for children with certain autism-like conditions. However, the drug approval struck many as premature. “I don’t think we’re at the normal stage where we’d get FDA approval,” says Rebecca Schmidt, an Epidemiologist at the University of California Davis who has studied whether giving folate during pregnancy can lower autism risk in children. As part of the announcement, the administration also revealed the recipients of $50 million in funding for projects seeking to understand factors that influence autism prevalence and outcomes. Further studies of leucovorin were not on the list.
Leucovorin is commonly given to cancer patients to help counteract side effects of certain chemotherapies that interfere with folate metabolism. It has also been tested in a rare genetic condition called cerebral folate deficiency (CFD), estimated to occur in one in 1 million children. CFD occurs when the brain does not get enough folate even though overall levels in the blood are adequate. Children with the condition initially develop normally; however, starting between 5 months and 2 years they begin to regress and develop symptoms that can resemble autism, including language delays and lack of motor control.
Some scientists have identified antibodies they suspect might cause some cases of CFD by interfering with folate absorption in the brain. No definitive studies have compared rates of this autoimmune form of CFD in autistic versus neurotypical children. However, some small studies suggest treatment with leucovorin may lead to improvements in verbal communication in autistic children, and some have suggested particular benefits for the subset who have these antibodies. However, none of the studies of leucovorin in autism had more than a few dozen subjects, and none was longer than 6 months.
Several also had methodological problems, such as incorrect adjustment of scores on tests of language skills by age group, says Lawrence Scahill, a Professor of Pediatrics at Emory University who specializes in the design and conduct of clinical trials in children with autism. “It could be possible to have a subset of autism cases that have this folic acid deficiency, but I would be skeptical of any one proposed explanation or treatment,” says Kristen Lyall, an Epidemiologist at Drexel University who will lead one of the 13 studies the administration announced Monday, studying the interactions of diet, genetic background, and environmental exposures on the risk of autism.
Despite the limited evidence, FDA Commissioner Martin Makary announced that his agency had already filed a Federal Register notice to change the label “on an exciting treatment called prescription leucovorin so that it can be available to children with autism.” In an accompanying fact sheet, the administration clarified that the drug would be approved to treat CFD, not autism in general, and it acknowledged “additional studies are needed” to tell whether leucovorin could be helpful to people with autoimmune CFD. Makary did not mention that caveat in the news conference, claiming “hundreds of thousands of kids, in my opinion, will benefit.”
Scahill says before doctors turn to leucovorin, scientists need to perform larger studies of its effects that focus on specific age groups and use appropriate tests, which “would be challenging” to conduct and likely take several years to complete. Even if the drug works, he adds, it is not clear how long a child would need to be on leucovorin before they might expect to see changes and whether children who do see improvements would need to continue taking the drug indefinitely.
Also missing are large-scale studies on leucovorin’s safety and the optimal dose, Schmidt says. Autism is a complex condition, she adds, and “there’s not going to be a cure-all.”
Many researchers, advocates, and autistic people object to the view of autism as a disease to be treated or cured at all, arguing instead for supports and services that improve the well-being of autistic people.
The administration will attempt to collect some of the missing data on leucovorin as it becomes more widely used, Mehmet Oz, head of the Centers for Medicare & Medicaid Services, said at the news conference. More than half of U.S. children are covered by federally funded health care programs including Medicare and the Children’s Health Insurance Program, he noted, and the administration will ask states to collect data on children who receive the drug under those programs. Oz noted that scientists and clinicians had told administration leaders it would take “at least 5 years” to collect the data usually needed for FDA approval; however, “parents are unwilling to wait 5 years. The president is unwilling to wait 5 years for these results.”
However, Scahill says the approval opens the doors to “an experiment in the wild” rather than methodical testing of the drug. Pediatricians “have nothing to go on” when it comes to prescribing leucovorin, he says. The agency’s decision, he adds, “is so remarkable to me — and, dare I say, irresponsible.”
REFERENCE: Science; 23 SEP 2025; Gretchen Vogel and Phie Jacobs