Failed cancer immunotherapies cost Medicare hundreds of millions. How should FDA revamp accelerated approvals?

A recent FDA review of past conditional go-aheads for cancer immunotherapies has led to several withdrawals. Health policy and economics researchers see the campaign as another opportunity to push for a revamp of the agency’s accelerated approval pathway.

Across six cancer indications for four medicines through the FDA’s accelerated approval program which have since been pulled, Medicare had spent $224 million between 2017 and 2019, a new study published in JAMA Internal Medicine shows.

Describing the expenditure as “waste,” the team of researchers from Harvard University and London School of Economics and Political Science suggests the FDA should “enforce timely completion of confirmatory trials and accelerated withdrawal” if a drug fails to confirm its benefit.

The six indications span small cell lung cancer, bladder cancer, stomach cancer and liver cancer and belong to Merck & Co.’s Keytruda, Bristol Myers Squibb’s Opdivo, Roche’s Tecentriq and AstraZeneca’s Imfinzi. The $224 million didn’t calculate Tecentriq’s newly withdrawn use in triple-negative breast cancer. That indication cost the Medicare system about $31 million in accumulative spending, Mahnum Shahzad, the study’s corresponding author, told Fierce Pharma. It also didn’t consider how Keytruda’s full approval in previously untreated bladder cancer was for a smaller patient population than its original conditional nod.

They were all part of 10 accelerated approvals that have attracted FDA scrutiny after failures in confirmatory trials; the larger group cost Medicare $569 million during the three years, the researchers found.

RELATED: Inspector general review of FDA's quick approval process won't take the accelerated pathway

Since the start of the accelerated approval program in 1992, the FDA has mainly used the pathway to approve cancer drugs based on biomarkers, such as tumor response rate, which are considered “reasonably likely” to predict clinical benefit, namely, extending patients’ lives. As a condition for the early green light, drugmakers are required to prove their meds’ efficacy in confirmatory trials.

But the program has repeatedly come under fire for its lack of accuracy and follow-on oversight. A 2019 JAMA Internal Medicine study led by Harvard researcher Aaron Kesselheim, M.D., found that only 19 of 93 cancer drug indications the FDA waved through under the pathway between late 1992 to mid-2017 had positive overall survival data from confirmatory trials.

Earlier this year, the controversial approval of Biogen’s Aduhelm for Alzheimer’s disease, based on the drug’s ability to decrease amyloid plaques in the brain, sparked wide debate over the FDA’s use of a questionable surrogate marker.

Critics have also picked on the duration of time the FDA has allowed accelerated approvals to stay in place without clear efficacy data or even after confirmatory trial failures. For example, for Keytruda’s now withdrawn third-line stomach cancer indication, a confirmatory trial for second-line treatment failed in late 2017. Another trial for newly diagnosed patients bombed in April 2019. But the FDA didn’t take any action until an advisory committee meeting this April, which led to Merck’s voluntary withdrawal in July.

RELATED: FDA's ongoing review of accelerated cancer approvals targets 2 drugs from Secura and Aurobindo

While the accelerated approval criticisms aren’t exactly new, the issue has seemed to gain steam this year. Amid the Aduhelm fallout, The HHS Office of the Inspector General in August launched a review of the FDA’s accelerated approval pathway, with a report expected in 2023.

In another JAMA Internal Medicine study in July, Kesselheim and collaborators returned to suggest that the FDA should incorporate better consensus in the use of biomarkers for accelerated approvals. In a point that has gained support lately, the researchers argue that the FDA should finalize confirmatory trial protocols as a condition of accelerated approval. A negative confirmatory trial readout would automatically trigger a withdrawal, the team said.

The team also suggested the Centers for Medicare & Medicaid Services and the Department of Veteran Affairs take steps to control costs of drugs with uncertain evidence. Limiting how much a drug may be reimbursed is a tactic that powerful drug cost watchdog, the Institute for Clinical and Economic Review, has recently proposed in its own list of recommendations (PDF) to improve the FDA’s accelerated approval program. 

Last week, two researchers at the Brookings Institution and the University of Pennsylvania also proposed a four-part project titled “Pay for Drugs That Work” for Medicare. The plan includes a four-year cap for drugmakers to provide confirmatory trial results before Medicare Part B stops reimbursement.

Earlier this year, citing “the risk of significant adverse drug events,” plus a “lack of evidence of a positive impact on cognition,” The Department of Veteran Affairs declined to add Aduhelm to its formulary.

Editor's Note: The story has been updated with Medicare expenditure on Tecentriq's triple-negative breast cancer indication.