Dr. Aaron Kesselheim had been on an advisory committee for the US Food and Drug Administration for a half-dozen years, but he had never been to a meeting like this one.
The FDA establishes advisory committees to assist the federal agency with one of its most important duties: deciding whether to approve the distribution of new drugs. The stakes of these decisions are enormous. Based on the outcome of the FDA’s deliberations, patients may gain access to lifesaving medicines, and manufacturers may reap billions in profits. Kesselheim, a professor at Harvard Medical School, was one of the members of this committee because of his expertise on pharmaceuticals that address diseases of the brain, including Alzheimer’s, the irreversible, progressive brain disorder that destroys memory and thinking skills, and eventually causes death. Alzheimer’s is the sixth leading cause of death in the United States.
The public meeting, conducted as an all-day video call on November 6, 2020, concerned the application for aducanumab, a drug that would be marketed under the name Aduhelm by the company Biogen, which is based in Cambridge, Massachusetts. Aduhelm, if effective, would address one of the most pressing needs in modern medicine: to slow the symptoms of Alzheimer’s disease. Research into Alzheimer’s treatments has long been an exercise in frustration, with no new drugs approved since 2003. The field of Alzheimer’s research had proved so difficult that it was dubbed by some with the macabre nickname of the “Valley of Death.”
In light of this record, and the need to improve it, Kesselheim was looking forward to examining the prospects for Biogen’s new drug. “The great thing about advisory committees is that they are independent, and they don’t have a stake in the outcome,” he said. “We were just an independent group providing their opinion.”
As in all advisory committee meetings, this one included several representatives of the FDA and of the applying pharmaceutical company. Dr. Billy Dunn, the director of neuroscience at the FDA, spoke at length on the call, and Samantha Budd Haeberlein, a senior vice president of Biogen, led the representatives for the company. What was unusual about this meeting, though, according to Kesselheim, was the apparent relationship between the FDA and company representatives.
“There was a strange dynamic, compared to the other advisory committee meetings I’ve attended,” Kesselheim said. “Usually there’s some distance between the FDA and the company, but on this one the company and the FDA were fully in line with each other in support of the drug.”
Dunn in particular was outspoken in his support for the effectiveness of the drug, calling some of the evidence in its favor a “home run,” while opponents of approval described the same evidence as inconclusive or worse. Dunn noted further that the FDA “has determined that it is appropriate to exercise the broadest flexibility in applying the statutory standards for these conditions.”
From a regulatory perspective, drugs can work in two ways: They can work to treat disease or they can address a surrogate measure for the disease. Take heart attacks, for example. A drug can directly reduce the chances of heart attack or it can address a surrogate for heart attack – like cholesterol.
At the November meeting, according to Kesselheim, the issue was treatment of Alzheimer’s itself, not any surrogate. “The discussion at the committee related to the clinical benefits of the drug” – that is, whether it slowed the cognitive decline of Alzheimer’s patients, he said.
Most of the meeting was devoted to analysis of Biogen’s own studies of the effectiveness of Aduhelm. The results were not promising in terms of clinical outcomes. Indeed, two of Biogen’s major studies of people taking Aduhelm had been shut down before they were completed because they showed no significant benefit to patients. (Aduhelm, which is injected intravenously, also caused significant side effects, including brain swelling in about a third of patients. Brain swelling, at a minimum, can produce painful headaches as well as more serious problems, including, in rare cases, death.)
In light of this, the advisory committee voted, with one member voting uncertain but no one dissenting, to recommend that the FDA reject the drug.
Seven months later, in June, the FDA gave Biogen final approval to treat patients with Aduhelm in all stages of Alzheimer’s disease. Reflecting the confidence in its judgment, the agency even acted on an accelerated basis.
Kesselheim and two other members of the advisory committee promptly resigned in protest. In his letter of resignation, Kesselheim called the Aduhelm process “probably the worst drug approval decision in recent US history.”
Of course, an advisory committee’s recommendations are just that – advisory. The FDA is not required to follow them. Indeed, one study found that between 2008 and 2015 the agency rejected advisory committee recommendations 22% of the time. But Kesselheim, along with the other experts CNN interviewed, could not recall a single occasion when the FDA had rejected the unanimous conclusion of an advisory committee – which is what happened here.
The FDA’s decision to approve Aduhelm has already had significant consequences. In the first place, the drug may give hope to the roughly 6 million Alzheimer’s sufferers in the United States and their families.
“The approval has already renewed investment activity in Alzheimer’s disease research and development, and we are optimistic that other innovative treatments will soon join Aduhelm,” said Allison Parks, a spokeswoman for Biogen. “We are proud of the work our dedicated team has done to develop Aduhelm, and of the potential it brings to Alzheimer’s patients.”
The financial implications are huge as well. Aduhelm will be extremely expensive, as an intravenous treatment that will cost about $56,000 per year, plus several thousand more in related expenses. Given the number of Alzheimer’s patients in the United States, the cost for widespread use of this treatment could quickly run into the many billions of dollars – much of which would go to Biogen and its partner in the project, Eisai, a Japanese pharmaceutical company. If Aduhelm is approved as a treatment by Medicare, which is currently weighing the issue, the taxpayers will pay much of the tab.
At its core, though, the controversy about Aduhelm raises a fundamental question: Did the FDA approve a drug that doesn’t help people – and if so, why?
‘What’s going on here?’
“Alzheimer’s is really like nothing else in chronic disease because it is so devastating to the family and caregivers as well as the people who have it,” said Harry Johns, the president of the Alzheimer’s Association, the nation’s largest advocacy group for victims of the disease. “Six million people have it now, and by mid-century that’s going to double because of the aging of the population. If you figure at least two caregivers per afflicted person, that’s an enormous toll on the country.”
Alzheimer’s first robs victims of their memory, and then takes their lives. Despite all research efforts, Alzheimer’s remains both incurable and fatal.
The dire course of the disease, and its prevalence, have led to a rush of investment in potential treatments. Sixteen years ago, the federal government spent $450 million a year on Alzheimer’s research; today the annual total is $3.1 billion. Private investment, based on the promise of millions of potential customers, has soared as well. But progress has been glacial to nonexistent.
This prior record of futility appears to have played some role in the FDA’s decision to approve Aduhelm. As Courtney Rhodes, an FDA spokeswoman, said, “Given the unmet needs for patients with Alzheimer’s disease – a serious, progressive and ultimately fatal disease – the agency chose to use the accelerated approval pathway to allow earlier access to patients while we continue to acquire data on the drug’s benefit.” The FDA will continue to collect data on the effectiveness of Aduhelm for the next nine years.
But other parts of the FDA’s explanation for its decision surprised and angered advisory committee members.
Aduhelm is designed to address a surrogate end point for Alzheimer’s disease, just like cholesterol medication addresses a surrogate end point for heart disease. Advanced medical scans reveal that many Alzheimer’s patients have what are called amyloid plaques – toxic nests of proteins – in their brains. As the FDA said in its original statement on Aduhelm, “While the specific causes of Alzheimer’s disease are not fully known, it is characterized by changes in the brain—including amyloid plaques and neurofibrillary, or tau, tangles—that result in loss of neurons and their connections.” Aduhelm is designed to remove some of the plaques, and thus, the theory goes, slow the progress of symptoms of Alzheimer’s disease.
“For the last 20 or 30 years, there has been a tremendous emphasis in the research on what’s known as the amyloid cascade hypothesis – the idea that amyloid isn’t just a symptom of the disease but that it’s a cause of the disease,” said Dr. Caleb Alexander, an epidemiologist at Johns Hopkins who has served on the same FDA advisory committee as Kesselheim for eight years. “There has been enormous investment in various therapies to reduce amyloid and some have even succeeded, but none of them have produced the corresponding gains in cognition. In other words, reducing amyloid has not meant that anyone is actually getting better.”
Still, when the FDA approved Aduhelm in June, it did so because the drug reduced plaque. As the agency said in its announcement of the approval, three studies involving more than 3,000 patients in total showed that “patients receiving the treatment had significant dose-and time-dependent reduction of amyloid beta plaque, while patients in the control arm of the studies had no reduction of amyloid beta plaque.” The logic was straightforward for a surrogate medication: Less plaque meant less disease meant fewer symptoms. As Rhodes, the FDA spokeswoman, said, the FDA “concluded that it is reasonably likely that this reduction in amyloid plaque will result in meaningful clinical benefit to patients.”
But that reasoning immediately came under attack from the members of the advisory committee.
“The reduction of plaque is a surrogate measure, a lab measurement which stands in for a clinical end point, which is how a patient feels, function or survives,” Kesselheim said. “Amyloid plaque is a protein deposit on the brain that you can observe on a PET scan. Some Alzheimer’s patients have it, and some don’t have it. It’s associated with Alzheimer’s, but it’s not a perfect association. We don’t know what it means to reduce the amyloid plaque. It’s not 100% clear. In contrast, your LDL cholesterol level is a clear surrogate measure. If you change your LDL, then you reduce the number of heart attacks and strokes.”
As Dr. David Knopman, a neurologist at the Mayo Clinic, wrote in his email of resignation from the advisory committee, “Biomarker justification for approval in the absence of consistent clinical benefit after 18 months of treatment is indefensible.”
But committee members were surprised by more than the FDA’s decision to use what they regard as a flawed surrogate measure to approve the drug. They were especially offended because the issue of amyloid plaque hadn’t really been raised at the committee meeting in November.
“The whole focus of the advisory committee meeting was whether the medication affected cognitive function – which it didn’t improve – but not whether it affected some surrogate for cognitive function,” said Kesselheim.
Worse yet, according to the critics, the FDA gave Aduhelm accelerated approval in June, another possibility that had not been raised before the committee. (In notable contrast, the FDA did not act in an accelerated fashion when it finally approved the Pfizer/BionTech vaccine for Covid-19 in August.)
The FDA remains adamant that both the process and the result of the review of Aduhelm were correct.
“We appreciate the perspective of the members of the advisory committee and value their input,” the FDA’s Rhodes said, “The advisory committee’s view was that there was i