The huge spending bill that became law last December also included fixes to a problem that has dogged the Food and Drug Administration for years.

Drugmakers could get fast approval for certain drugs — with preliminary data — on the promise that they would do more research after the fact to make sure the drugs worked. But companies were often slow to follow through on these so-called accelerated approvals, leaving patients uncertain about their medicines.

The COVID-19 vaccines weren't granted this kind of approval. But the FDA has given accelerated approvals to more than 100 drugs aimed at treating cancer, HIV, preterm birth and other conditions over the last three decades.

Among other things, the new law gives the FDA more leverage to hold drug companies accountable for their promises about further research. Now, for instance, the agency has explicit authority to require drugmakers to have started confirmatory studies as a condition of accelerated approval.

"This essentially strengthens their hand," says Robin Feldman, a professor at the University of California College of the Law, San Francisco. "However, just because the FDA has the power doesn't mean it's going to use it."

The FDA wouldn't say whether it's doing anything differently on accelerated approvals, other than that it's working on a plan. "The FDA will be implementing these new authorities and will provide updates as appropriate," an agency spokesman said in a written statement.

Why accelerated approvals of medicines are controversial

Criticism of the program crescendoed following the controversial accelerated approval of Alzheimer's drug Aduhelm in 2021. Aduhelm's manufacturer initially set a price of $56,000 a year — even though the FDA approved it based on its ability to remove sticky amyloid plaques from the brain without proof that plaque removal could slow patients' loss of mental abilities. The drug flopped, with insurers refusing to cover it, and the drugmaker, Biogen, had to lower its price.

Dr. Aaron Kesselheim, a professor of medicine at Brigham and Women's Hospital and Harvard Medical School, resigned from an FDA advisory committee because of the Aduhelm approval.

His research has found flaws in the accelerated approvals program, including the fact that only 1 in 5 cancer drugs that were granted accelerated approval "demonstrated improvements in overall patient survival."

An NPR investigation last year found that many of the promised confirmatory studies are delayed, leaving doctors and patients in the dark for years about how well the medications really work. Specifically, NPR found that 42% of outstanding confirmatory drug studies required after accelerated approval either took more than a year to begin or hadn't started at all.

Meanwhile, drugmakers were more likely to repeatedly hike the price of drugs granted accelerated approval compared with drugs granted traditional approval.

In September, the Office of Inspector General for the Department of Health and Human Services issued a report about delinquent confirmatory studies for drugs granted accelerated approval. The report's findings were consistent with NPR's.

During NPR's investigation, a reporter asked the agency about confirmatory study timing. Dr. Jacqueline Corrigan-Curay, principal deputy center director in FDA's Center for Drug Evaluation and Research, said the agency would like to always have these confirmatory studies underway at the time of accelerated approval. But there's no regulatory requirement, she said.

"If you have the data and it meets accelerated approval and there's this unmet medical need, we need to go forward" with the approval, she said at the time.

However, according to former FDA principal deputy commissioner of food and drugs Dr. Rachel Sherman, the agency has always had the power to require that confirmatory studies be underway at the time of approval, but didn't often exercise it.

"I never lost an argument with a company," she says of her three decades with the agency. "Sometimes they convinced me they were right and I was wrong. But if I wanted something, I got it. Because at the end of the day, the FDA has – what's the right word? – a bully pulpit."

University of California's Feldman says the new law can come in handy if a drug company tries to push back on an FDA decision.

There have been times when the FDA has had powers but neglected to use them, says Feldman, pointing toward a recent study in JAMA about how 3,000 clinical trials hadn't reported results between 2017 and 2021, but the FDA only submitted notices of noncompliance to 58 of them.

The agency has also attempted to put its foot down, gotten sued by drugmakers and lost. That's what happened when it tried not to give rare disease drug perks to the makers of a chemotherapy drug called Bendeka. The drug's manufacturer sued, and the FDA lost.

The new legislation, at least, should offer clarity.

Congress passed a fix — but it almost didn't happen

Congress passed legislation to fix long-standing problems with accelerated approval in December, but it almost didn't happen.

Several bills were winding their way through Capitol Hill over the last year to reform accelerated approvals.

Changes were expected to get folded into renewal of the Prescription Drug User Fee Act in September, but Congress passed it without any riders. That worried Dr. Reshma Ramachandran, of Yale School of Medicine, who was one of the doctors who traveled to Washington, D.C., to talk to legislators about accelerated approval updates.

"We really wanted to make sure that they were going to hold to their commitment to revisiting those reforms as a part of the end-of-year spending package," she said, adding that the "stunning" public meetings around the withdrawal of the drug Makena (which was prescribed in hopes of preventing preterm births) seemed to push Congress to act.

Following confirmatory studies that didn't prove Makena's efficacy, the FDA's Center for Drug Evaluation and Research recommended pulling the drug from the market in October 2020. But because the drugmaker didn't voluntarily withdraw the drug, a hearing to discuss Makena's potential withdrawal was required. It took two years for the FDA's advisory committee to meet, and members voted in favor of removing it from the market. The FDA has not yet withdrawn the drug's approval, however.

"The FDA was experiencing trying to withdraw a drug which had failed — not just a confirmatory trial, but multiple other studies," says Ramachandran. "I think that really gave cause to members of Congress, but also the FDA, to really say, 'Hey, we need to have these common sense reforms.' "

FDA's powers are discretionary, not mandatory

Changes to the accelerated approval process ultimately passed in the year-end omnibus spending bill that was signed into law on Dec. 29. The new law includes several fixes, including one that allows the FDA to require companies to start confirmatory studies before the agency grants accelerated approval to the drug.

"We wanted that to be codified in legislation so that that wouldn't be an exception. That would just be the rule," says Ramachandran. "When that happens, that prevents significant delays for starting the trials, but also completing the trials."

Still, the law doesn't outright require that studies confirming efficacy of a medication be underway at the time of approval. It only says FDA leaders "may require" it.

"So generally these are discretionary rather than mandatory," says University of California's Feldman. "In other words, it allows the FDA to do things, but doesn't require the FDA to do things. So that's a big difference."

The law may make it easier for the agency to take a medicine off the market if studies find the drug is not effective.

"When a drug is given accelerated approval and then that confirmatory study fails, that's really an important finding that indicates that that drug does not have the same level of effectiveness that we expect of other approved drugs," says Harvard's Kesselheim. "And there is no reason for that drug to remain on the market at that point."

In addition, the law creates new transparency requirements and a council to get everyone at the FDA on the same page about how this kind of approval should be used.

So the rules may need to be tightened again in the future.

"The final language was not [as] ... airtight as it could be," Kesselheim said. "So I do think that there is wiggle room."