In taking the unusual step of approving an old generic drug as a treatment for autism, the Food and Drug Administration stunned some experts by departing sharply from the agency’s typical standard for reviewing drugs.
The drug, leucovorin, has long been used to treat the toxic effects of chemotherapy, but it was endorsed as a therapy for some people with autism by President Trump and top U.S. health officials during a White House briefing on Monday.
The move flipped the standard process: Typically, a pharmaceutical company carefully studies a drug, often with input from the F.D.A. on the design of rigorous studies, and then files a formal application for approval.
But here, the agency said it reviewed medical research and made the approval decision to expand the drug’s use on its own.
“Mr. President, you told us to do what’s medically right — to go bold and not worry about the corporations and the lobbyists,” Dr. Marty Makary, the F.D.A. commissioner, said on Monday. “So that’s what we’re here doing today.”
Dr. Makary announced the sudden change during the president’s briefing on autism, which Mr. Trump repeatedly used to directly link Tylenol, the brand-name version of acetaminophen, to the disorder — a connection that is still unproven.
The F.D.A. was more cautious in a letter to doctors, saying evidence suggested a link between Tylenol and autism, though the agency added that the association remained an “area of scientific debate.” The health officials promised to fund more studies to explore a potential connection. Kenvue, the maker of Tylenol, has said the drug does not cause autism and is safe to use as directed.
Some outside experts said that the agency’s decision left many questions unanswered.
Dr. Aaron Kesselheim, a professor of medicine at Harvard Medical School, called the F.D.A.’s hasty endorsement of leucovorin “shocking.”
At the briefing, Dr. Makary promoted the drug as likely to help “hundreds of thousands” of children with autism. Yet the agency’s notice in the Federal Register indicated a narrower application.
The F.D.A. approved tablets of leucovorin specifically to treat “cerebral folate deficiency,” a condition characterized by low levels of folate, a B vitamin, in spinal fluid that is associated with delays in a child’s development and poor motor skills.
Before prescribing the tablets, doctors would first need to determine whether a patient suffered from the deficiency, Dr. Kesselheim said. How they would arrive at that diagnosis short of a risky, expensive spinal tap was not clear, he said.
The F.D.A.’s action also failed to clarify the appropriate dose of leucovorin for children and adults, which is typically determined by each drugmaker from clinical studies. Dr. Kesselheim also noted that the study cited by the F.D.A. showed only small improvements in autism symptoms among people with certain folate levels in their spinal fluid.
“This puts the cart about three miles in front of the horse,” Dr. Kesselheim said. “The reason we have F.D.A. is to try to help patients distinguish between things that do and don’t work.”
Holly Fernandez Lynch, an associate professor of law at the University of Pennsylvania, said the agency’s recommendation seemed like a “science communication disaster.” Few people would be likely to realize that a treatment touted as a salve for autism only applied to a subset of patients, she said, adding that the vagueness of the announcement seemed intentional.
“Like: ‘We solved it,’” she said. “That’s exactly what this administration wanted.”
The approval itself, including a description of a small study, also falls far below the agency’s standard for reviewing a treatment for a condition as common as autism, she said. Small studies are not unheard-of, but are typically done in the case of an extremely rare disease that is difficult to examine.
“I think this is a dangerous kind of subversion of what F.D.A. approval is supposed to be,” she said.
The F.D.A. asked GSK, the original company that applied for approval of leucovorin, to update the label on the product, which the company sold during the 1980s and 1990s. It is a procedural step that will enable several companies that make generic versions to adopt the label’s language. GSK said in a statement that it no longer sells the drug, but will update the label.
Daniel Aaron, an associate professor of law at the University of Utah who studies the F.D.A., said the agency’s decision on leucovorin could set a precedent for other companies that want to skip costly, careful and yearslong clinical trials before asking the agency for an approval.
“This approval could very much well come to haunt the F.D.A.,” he said.