Fallout over FDA’s nod to Alzheimer’s drug highlights failures of accelerated system

An accelerated approval programme for US drugs that has been around for almost three decades is under fire for the criteria used by regulators to decide which therapies should be greenlighted, and for letting ineffective treatments linger on the market.

The Food and Drug Administration’s (FDA’s) accelerated process has been hailed for quickly addressing unmet medical needs with novel treatments. But critics say changes are needed to make it more transparent, to better measure efficacy and to quickly stop the sales of drugs that fail confirmatory trials.

Approval of the Alzheimer’s drug, Aduhelm, has revved up debate on the programme. Rather than being cleared based on its effectiveness, Aduhelm gained approval by showing it can reduce amyloid plaques in the brain, a physical biomarker, or surrogate, linked to the disease. Meanwhile, drugmaker Biogen has nine years to finish a trial on its efficacy.

“We’re still in a situation where we don’t know if Aduhelm actually helps patients,” said David Whitrap, vice-president for communications at the Institute for Clinical and Economic Review. “If approval would have been withheld and the company was just asked to run a third trial to provide further insight into whether the drug works, that trial surely could have been accomplished far sooner than 2030.”

Overall, there were 41 approvals for cancer through the accelerated programme last year, the most by far since the programme was initiated. Still, how well they work remains unclear with about half that number not yet reporting on their confirmatory trials.

While there have been many successes in the accelerated approval programme, there are problems that must be addressed, Whitrap said.

“We need more transparency,” he said “We need more consistency around evidentiary standards, we need incentives for companies to follow through and bring confirmatory trials to completion in a reasonable time frame.”

The approval of Aduhelm — which has been priced at $56,000 a year and is expected to be taken by millions of patients — could have a ripple effect for other potential Alzheimer’s drugs in the pipeline, Whitrap said, adding that it could make patients less willing to enrol in trials for other promising therapies.

Allison Parks, a Biogen spokesperson, said that the company is “working with urgency and putting resources and plans in place towards the goal of completing the confirmatory trial ahead of the nine-year time frame, with a focus on high-quality data”.

She referred to the accelerated programme as “a well-established regulatory pathway that can lead to tremendous innovation”.

Bishal Gyawali, an oncologist and associate professor at Queen’s University in Canada, is concerned that the use of disease surrogates for approvals could act as a disincentive for other drugmakers in the future to use survival as an endpoint.

Surrogates are a lab measurement or physical sign used in trials as a substitute for meaningful clinical result where it may take a long time to see an effect in treatment for a particular disease.

Follow-up trials

“If this trend continues,” Gyawali said, “then I think one big threat that I can see in the future is that none of these drugs will use overall survival as an endpoint.”

Approvals based on a disease surrogate are just one issue raising concern. Drugmakers also miss deadlines on the follow-up trials mandated by the programme, researchers say. And in some cases, the FDA can fail to take back the approval of a drug for years after a negative study finding.

Roche Holding’s Tecentriq, for instance, was granted accelerated approval in patients with metastatic urothelial carcinoma in May 2016. It failed its confirmatory trials the next year, but its approval for that purpose was not withdrawn until this year, a four-year delay.

Tecentriq continues to be approved for treatment of other cancer indications, said Nathalie Meetz, a Roche spokesperson.

A 2019 study found that only about 20% of 93 cancer drugs given accelerated approval showed improvements in overall survival, broadly considered the most important standard for whether a cancer treatment is effective.

Avastin, another drug made by Roche, is a case in point. It was given accelerated approval in 2009 on the basis of delaying tumour response for glioblastoma, one of the most common and aggressive forms of brain cancer. While confirmatory trials failed to show that the drug improved quality of life or survival, the FDA gave it full approval eight years later for confirming a surrogate measure called progression-free survival.

The FDA is aware of the concerns about the programme and has been looking more closely at how it works, said Chanapa Tantibanchachai, a spokesperson.

Advisory committee

During the early months of 2021, several pharma giants — including AstraZeneca and Merck — voluntarily withdrew cancer drug indications after failing to meet their postmarketing requirements. In all cases, the companies said the decisions were made in consultation with the FDA, as part of the agency’s industrywide review of accelerated approvals.

As part of its review, the FDA’s Oncologic Drugs Advisory Committee held a three-day meeting in April to vote on whether to keep six recent cancer immunotherapy approvals despite failing in their confirmatory studies.

The panel agreed to keep four of the six indications, though the final decision on whether to maintain the six indications will be made by FDA’s career officials in the Oncology Center of Excellence.

The FDA is also considering recommendations from a white paper issued by the Friends of Cancer Research in November 2020. The report urges the agency to consider ways in which the pathway could be made more nimble, making comparisons to similar programmes in Europe and Canada that require a yearly update of postmarket requirements to ensure commitments are met timely.

Confirmatory trials

The agency will continue to evaluate accelerated approval programmes and their confirmatory trials to determine if there is still patient benefit, the FDA’s Tantibanchachai said, and “may continue to periodically convene advisory committee meetings to discuss accelerated approvals”.

Lilly’s Lartruvo, a drug that aimed to treat advanced soft tissue sarcoma, is a good example of how the accelerated approval process should function, according to Queen’s University’s Gyawali.

Lartruvo was granted accelerated approval in 2016 but confirmatory trials provided no evidence that it prolonged patients’ lives and within three years it was off the market.

The company gained more than $500m in sales for the treatment, according to Tracy Henrikson, a Lilly spokesperson. “We don’t have a development cost estimate to provide,” she added, “but we can say that the development costs for Lartruvo exceeded the total revenues.”

Gyawali, meanwhile, called the $500m a small price to pay in the search for life-saving drugs. If the clinical trials for Lartruvo were not conducted in a timely manner, it could have cost the US billions, he said.

Surrogate use

And had they used a surrogate measure rather than overall survival as an endpoint, Gyawali added, we would probably never know that the drug did not work.

However, C Michael White, head of the department of pharmacy practice at the University of Connecticut, said the use of a surrogate in the Biogen approval could be helpful.

If Biogen’s confirmatory trial shows benefit, the agency is likely to be even more comfortable with the amyloid beta plaque surrogate endpoint. And if not, the FDA will be less willing to allow it going forward.

“The FDA has a list of surrogate endpoints they are comfortable with,” White said. “They expanded it to include amyloid beta plaques, but have shown they are reticent to add additional surrogates without stronger support for the linkages between the surrogate and the final health outcome.”

White said he is “not personally as confident as the FDA that this gamble will pan out with blocking amyloid beta, but it could. The net result would be earlier access to an effective therapy vs long delays and possibly giving up on the drug, and never doing the study to find out if it works.”

Bloomberg News. More stories like this are available on bloomberg.com