In June 2021, the US Food and Drug Administration (FDA) approved the first Alzheimer’s drug in 18 years: aducanumab (also known by its brand name Aduhelm). At the time of writing, the drug is also under review in the EU, Japan and several other countries.
For the roughly 30 million people worldwide who live with Alzheimer’s, this is unprecedented news, and must seem like cause for optimism. Unlike existing drugs, which only feebly suppress cognitive symptoms, aducanumab attempts to get at the underlying cause of the disease, to stop and cure Alzheimer’s.
Having seen the disease up close and personal, I too long for a cure. Yet I’m not exactly overjoyed by the news of aducanumab. To borrow the words of the Harvard professor of medicine, Aaron Kesselheim, aducanumab was “probably the worst drug approval decision in recent US history”. Kesselheim was part of an FDA advisory committee that voted, almost unanimously, to not approve the treatment.
To understand the controversy, we need to start with the brain of an Alzheimer’s patient. It typically harbours two kinds of pathology: first, plaques, or deposits, that are scattered between nerve cells; second, fibres that form tangles inside the cells. The plaques consist of a protein fragment known as beta-amyloid. The tangles consist of a protein known as tau.
Scientists have long suspected that these two substances are the underlying cause of Alzheimer’s. Of the two, beta-amyloid has garnered far more interest, for some good reasons. In 1991, John Hardy’s team from St Mary’s hospital medical school in London discovered that genetic mutations affecting beta-amyloid cause early-onset Alzheimer’s, a rare inherited form of Alzheimer’s that exhibits the same brain changes as the more common, late-onset version of the disease. The discovery led Hardy to propose the amyloid hypothesis, which suggests that excessive beta-amyloid buildup is the cause of Alzheimer’s.
Since its birth, the amyloid hypothesis has been updated regularly to explain new findings, and has cemented its position as the dominant paradigm in the study of Alzheimer’s. The latest update places the blame not on the end state beta-amyloid plaques, but on intermediate deposits known as beta-amyloid oligomers.
The amyloid hypothesis is able to explain a range of phenomena related to Alzheimer’s, but it also struggles to answer important questions. Chief among them is why decades of clinical trials with anti-amyloid drugs have failed. From inhibitors that prevent the formation of beta-amyloid in the first place to vaccines that remove existing beta-amyloid, we have hit a brick wall in our efforts to stop cognitive decline. Until aducanumab – well, sort of.
Developed by Massachusetts-based Biogen, aducanumab is a vaccine comprising antibodies to beta-amyloid. It burst on to the scene in March 2015 when an early-stage clinical trial showed that it drastically reduced levels in the brain. More impressively, the cognitive and functional skills of those who took the drug didn’t decline as much as those who were given a placebo.
That trial was hailed a game-changer, a confirmation that if you reduce beta-amyloid by enough, you will see a clinical benefit. There is, however, a tiny problem. This supposed game-changing study only tested 165 people, and they were divided into smaller groups of 30 or so each. With these tiny sample sizes the cognitive findings carry little value beyond their usefulness as marketing hype. Not only that, but the trial didn’t even measure oligomers, the current most likely culprit. It merely measured plaques and assumed that oligomers also went down.
Undeterred, Biogen pushed forward with two large trials in autumn 2015 that were supposed to confirm the cognitive effect. But the plan went awry. In March 2019, an interim analysis showed that the trials were unlikely to succeed, and they were stopped in their tracks.
Then, a plot twist: seven months later Biogen declared that the interim analysis was, well, wrong, and that at its highest dose aducanumab did slow cognitive decline in one of the trials (in the other one, not enough participants received that increased dose). Biogen pushed ahead for FDA approval.
Ordinarily, one positive study isn’t enough to get you the green light. But the FDA put Biogen on the Accelerated Approval pathway, designed to speed up the process for potentially valuable therapies for serious diseases. As a result, less weight was placed on having definitive evidence of cognitive benefit: the fact that beta-amyloid plaque was reduced was deemed to be enough.
This is a problem, according to critics of the amyloid hypothesis. The connection between the two – plaque reduction, and cognitive improvement, that is – is far from certain. In fact, some believe beta-amyloid may not be the underlying cause of the disease at all.
Recall that other marker of Alzheimer’s, tau tangles. Tau proteins help to support axons, the long, thin part of a nerve cell that sends electrical signals to other nerve cells. If tau malfunctions, axons collapse, and brain signals will stall. That, say some, is the real cause of Alzheimer’s. Still others suggest that Alzheimer’s is a metabolic disease, in which the brain suffers from reduced glucose use and energy production. According to this view, Alzheimer’s can almost be thought of as a “type 3 diabetes” caused by impaired insulin signalling and reduced metabolism, which starves neurons.
For proponents of these theories, the continued dominance of the amyloid hypothesis represents a failure to think innovatively, and a frustrating block on meaningful progress. There is some truth to this, but equally, if there were an obvious alternative, it too would likely have borne fruit by now.
Yes, anti-amyloid drugs don’t have a good track record, but neither do tau or metabolic treatments. And each theory has its fair share of unanswered questions, of things that don’t stack up.
In any case the approval of aducanumab has made sure that for the foreseeable future the amyloid hypothesis isn’t going anywhere. After all, it has shown researchers and companies that pursuing anti-amyloid drugs can pay in the region of $56,000 (£40,000) per year per patient. Granted, if post-marketing studies show that aducanumab doesn’t actually work, approval can be revoked. But these studies are tricky to conduct: patients will not want to enroll in a trial and potentially receive a placebo when they can get an “approved” drug from their doctors.
More interesting than the aducanumab controversy are the several continuing trials designed to test anti-amyloid drugs on cognitively healthy people (for example, the A4 study). The researchers behind these are motivated by the idea that once cognitive symptoms appear, any reduction in beta-amyloid may be too little too late. In other words, prevention, not treatment, may hold the key to defeating Alzheimer’s, and those at risk may in the future be encouraged to take what might be thought of as “statins for the mind”. If the results of these studies, which are anticipated in one to four years, also turn out to be dismal, that may indeed be it for the amyloid hypothesis. But until then, there’s still a spark of hope. And as we labour to solve the riddle of this debilitating disease, close to an answer but with firm evidence of success still tantalisingly out of reach, hope is a precious commodity.