The number of American children filling a leucovorin prescription rose roughly sevenfold over the course of 2025, and a sharp single-week jump coincided with a federal press conference promoting the drug for autism, according to an analysis published this month in The New England Journal of Medicine.
The research, led by Kao-Ping Chua of the University of Michigan, drew on the IQVIA Longitudinal Prescription Database, which captures about 93 percent of prescriptions filled at United States retail pharmacies. Across 2025, the database recorded 150,026 leucovorin prescriptions dispensed to 47,394 children.
For families, the finding lands in an uncomfortable place. Many parents who sought the drug did so after watching federal officials describe it as effective. In the months since, the evidence base supporting that description has not grown. It has contracted.
The Prescribing Curve Tracks the Announcement
The weekly number of children with a filled leucovorin prescription rose from 734 at the start of 2025 to 5,858 by the end of it, an increase of roughly 700 percent. Measured against the population, dispensing climbed from 1.0 per 100,000 children early in the year to 4.6 per 100,000 in the week before the September 2025 press conference.
During the week of the conference itself, the rate rose 46 percent, from 4.6 to 6.7 per 100,000 children. By the close of the study period, which ran through mid-December, it had reached 8.0 per 100,000. Dispensing to adults showed no comparable spike, which is the comparison that makes the pediatric pattern hard to explain by supply or coding changes.
The increases were concentrated among children ages 6 to 11 and in the southern United States. Michigan Medicine reported Chua as saying the findings "demonstrate the power of the federal government to influence medical practice," and that the increase is concerning because the effectiveness and long-term safety of leucovorin for children with autism is uncertain. The analysis also found dispensing was already rising before the press conference, a trend the authors suggest may have followed news coverage featuring individual patient stories.
A Narrow Approval and a Much Broader Use
Leucovorin is a prescription form of folinic acid that has been used for decades to counter the toxic effects of methotrexate in cancer and autoimmune treatment. The Department of Health and Human Services announced at the September 2025 event that it planned to update the drug's label to cover cerebral folate deficiency.
The approval that followed was considerably narrower than the announcement suggested. In March 2026, the Food and Drug Administration approved leucovorin for cerebral folate transport deficiency, an ultra-rare genetic neurological condition tied to a confirmed variant in the FOLR1 gene. The agency did not approve it for autism spectrum disorder.
That distinction is not academic. Cerebral folate transport deficiency affects a very small number of children and requires specific genetic confirmation. The prescribing surge documented in the new analysis extends far beyond any population plausibly carrying that diagnosis, which means most of it represents off-label use for autism itself.
The Supporting Research Got Smaller, Not Larger
The clearest development since the announcement runs in the opposite direction from the prescribing curve. The largest randomized trial supporting leucovorin's use in autism, a study of 77 autistic children originally published in 2024, was retracted by the European Journal of Pediatrics early this year after a post publication statistical review confirmed concerns about the data and could not replicate the reported results. It was one of only a handful of randomized clinical trials ever conducted on leucovorin in autistic people.
Zoe Gross, director of advocacy at the Autistic Self Advocacy Network, told CIDRAP News that "When the administration made that recommendation, the evidence wasn't strong enough," and said the retraction left the evidence weaker still.
The American Academy of Pediatrics maintains guidance for clinicians on leucovorin use in autism that families can review with their own pediatrician. The Child Neurology Society position statement reviews the three small randomized trials of leucovorin as a treatment for idiopathic autism and concludes that their limitations preclude firm conclusions about effectiveness.
A separate electronic health record analysis from the University of California, San Diego, published in JAMA Network Open earlier this year, examined more than 11.9 million outpatient encounters involving 838,801 children with autism and found prescribing rates rose more than 2,000 percent from a stable baseline. The two studies use different denominators, but both point in the same direction.
Questions Worth Bringing to a Pediatrician
Families already using leucovorin should not stop a prescribed medication on the basis of a news article. Abruptly discontinuing any prescription without clinical guidance can create its own problems, and a child taking leucovorin for a confirmed metabolic diagnosis is in a different situation from one taking it for autism generally.
The productive conversation is a specific one. Parents can ask which indication the prescription was written for, whether genetic or metabolic testing supported that indication, what improvement would look like and over what timeframe, and what the plan is if nothing changes. They can also ask what the drug costs out of pocket, since off-label prescriptions are frequently not covered.
Some researchers have raised the possibility that increased off-label prescribing contributed to leucovorin shortages that reduced access for cancer patients who depend on the drug. Families can ask whether a shortage affects their pharmacy and whether an alternative formulation is appropriate.
The research letter measured prescribing behavior, not clinical outcomes. It does not establish that leucovorin is harmful, and it does not establish that it is ineffective for every child. What it establishes is that a large behavioral shift occurred without a matching shift in evidence, and that families deserve to know which side of that gap their own prescription sits on.
Key Questions Answered
What did the new analysis find? Weekly counts of children filling a leucovorin prescription rose from 734 to 5,858 across 2025, roughly a 700 percent increase, with a 46 percent jump during the week of a September 2025 federal press conference.
Is leucovorin approved to treat autism? No. FDA approved it in March 2026 for cerebral folate transport deficiency in patients with a confirmed FOLR1 gene variant. It remains approved for methotrexate rescue. Use for autism generally is off label.
What happened to the research supporting it? The largest randomized trial supporting leucovorin in autism was retracted early this year after a statistical review could not replicate its results. The remaining evidence consists of small trials.
Should parents stop a current prescription? No one should stop a prescribed medication without speaking to the prescribing clinician. The appropriate step is a conversation about the specific indication and expected benefit.
Which children saw the biggest increases? Children ages 6 to 11, and children in the southern United States, according to the analysis. Dispensing to adults showed no comparable spike.
Does the study show leucovorin is unsafe? No. It measured prescribing patterns, not clinical outcomes. The researchers note that effectiveness and long term safety in this population remain uncertain.
What should families ask their pediatrician? Which diagnosis the prescription addresses, whether testing supported it, what measurable improvement to expect and when, the out of pocket cost, and what happens if there is no change.
