Bioethics Forum Essay
What a Bad Day Science Had
June 7, 2021, was a bad day for science. That was the day the U.S. Food and Drug Administration approved aducanumab for the treatment of Alzheimer’s disease, even though a committee of its own selected experts recommended strongly against approval. The committee found that studies failed to prove the medicine works, or that its benefits outweigh side effects, which include swelling and bleeding in the brain. In a break from precedent, the FDA did not accept its committee’s recommendation, but delayed its decision for months, culminating in an announcement of approval.
How, you might ask, could this happen? This is in a sense a failure of scientific integrity, not in the sense of bribery or overt cheating, but in the failure to honor the scientific method and to acknowledge when research fails. The FDA’s website states, “FDA has determined that there is substantial evidence that [aducanumab] reduces amyloid beta plaques in the brain and that the reduction in these plaques is reasonably likely to predict important benefits to patients.” The first part of the sentence is indisputably true – this drug does reduce the amyloid plaques long associated with Alzheimer’s disease. The second part of the sentence – that reducing amyloid is “likely” or “expected” to help patients — is repeated three times in FDA’s press release with slight variations. Despite this insistence, this claim is not about evidence but about hope, a hope that has been proved wrong repeatedly in multiple large studies, involving thousands of patients and hundreds of millions of dollars over 15 years.
The second part of the sentence is a statement of the amyloid hypothesis, which has been tested and retested and has always failed, including in the aducanumab trials, which were shut down in 2019 because they did not meet criteria for success. (The data were later resuscitated and submitted to re-evaluation that the FDA review committee and many other experts found highly dubious.) But many researchers have spent their careers shoring up the amyloid hypothesis, and they continue to insist it might work – though it never has, actually.
Science isn’t supposed to operate this way. Scientists develop a theory, they set up a test, maybe they tinker with the test if it’s inconclusive. But ultimately if they test and test and the theory fails, they are supposed to acknowledge as much and learn from it.
The FDA should have stood by the expertise of its highly qualified panel. It did not. Moreover, the FDA fails to admit the crucial fact that it rejected the recommendation of its advisory panel. An FDA statement disingenuously claims ”we followed our usual course of action” in deciding to approve, but notes that the approval process “garnered the attention of the press.” The agency’s statements omit that the press became interested when FDA radically deviated from standard practice by rejecting its committee’s recommendation.
The FDA’s decision starts U.S. taxpayers on the road to pay a vast sum for a drug that doesn’t work, based on a theory that has failed over nearly two decades of rigorous testing. How vast a sum? Biogen is demanding more than $50,000 each year per patient. The actual costs will be far higher, as the drug must be delivered by infusion which itself is quite expensive, and screening tests will add thousands of dollars for each person treated, and many more for those screened and found ineligible. Then, too, expensive MRI scans are indicated to monitor the ominous side effects of brain swelling and bleeding.
Even the drug’s supporters claim it can only help those with no symptoms or minimal ones, and that it cannot help those with advanced disease. Nonetheless, approval means all six million Americans who already have Alzheimer’s disease may ask for it. If Medicare follows FDA’s lead and ignores science, agreeing to pay for this drug for anyone with Alzheimer’s, the costs will be unprecedented. Try the math yourself. Take some percentage of those over 65 and with some degree of risk, for instance from family history, diabetes, heart disease, or a range of other common factors. Then add a portion of the millions who already have dementia. You will see numbers you’ll have to check carefully — it’s easy to reach $100 billion per year. Business analysts predict a bonanza of a mere $7 billion for Biogen, but that will depend on how Medicare and the large insurers reimburse its use.
If this were a drug that really changed outcomes, we’d have to work hard to find the money. But for a drug that doesn’t work, we will spend billions on a lie. We will raise false hopes for those suffering with or at risk for dementia. If costs are incompletely reimbursed, we will exacerbate the already severe disparities in access to dementia care, and some families will waste money crucially needed for necessities like safe housing. Putting funds toward research for a better medicine, or toward better nonmedical support for those struggling to live comfortable lives with dementia would be so much more useful.
Why does this matter? It matters not only because this is the wrong decision for aducanumab, but also because this is a terrible precedent for American health policy. The FDA faces extraordinary pressure from different groups – it always has and always will. Alzheimer’s disease is fatal and widespread, and there is no effective treatment. Progress has been agonizingly slow. Of course, patients want new drugs; we all do. And pharmaceutical companies are eager to make them, for a blockbuster drug creates unfathomable wealth. Advocacy groups like the Alzheimer’s Association, pharmaceutical companies, elected representatives, and a host of others can push for approval and may focus more on what would be desirable than on data. All these groups and their views have a role in our democracy.
Nonetheless, the FDA’s crucial, inviolable task is to determine whether an intervention is effective and safe. Only data from research can answer those questions. For the FDA to do its work, it must be trustworthy and must prove at every turn that it is so, especially today, when trust is ruptured across so many facets of our civic life. The FDA’s action yesterday wounds the beating heart of scientific integrity that should stand at its core. This is our real loss – it’s not just the error in approving aducanumab, nor even the unbelievable cost of that mistake. It is that FDA has lost the right to say that its crucial decisions, on which so much depend, are based on science, and not wishful thinking.
Tia Powell, MD, (@tiapowell) is Dr. Shoshana Trachtenberg Frackman Chair in Bioethics, Director of the Montefiore Einstein Center for Bioethics, and professor of epidemiology and psychiatry at Albert Einstein College of Medicine, and a Hastings Center fellow. Dr. Powell is the author of Dementia Reimagined: Building a Life of Joy and Dignity from Beginning to End.