Alzheimer’s drug is no miracle cure, says UK scientist
A drug to treat Alzheimer’s disease has finally won approval in the US. Why is the British scientist behind it not jumping for joy?
Almost four decades after a desperate woman sent John Hardy a desperate letter, and a decade after that woman began, as she always feared, to lose her mind, came a vindication of sorts. It arrived in a fact sheet for doctors produced by the US Food and Drug Administration providing information on the country’s newest drug. “Aduhelm,” it began, “is an amyloid beta-directed antibody indicated for the treatment of Alzheimer’s disease.”
With that mundane opening sentence an experimental drug became a clinical one. The US regulator had approved a drug to treat Alzheimer’s. If you accept that regulator’s controversial assessment (and we will come to that), it meant more than merely that Hardy was right – that the hypothesis he’d pioneered at the start of his career has been proved towards its end. It was, said Hilary Evans, head of Alzheimer’s Research UK, “a pivotal moment”. Yet Hardy, a neurogeneticist and professor at University College London, is not quite as jubilant as you might expect. “I wouldn’t put it as strong as vindicated,” he says. “I guess… I don’t feel as depressed as I had been.”
Aduhelm is the brand name for aducanumab, a medication that relies on – indeed, is the apotheosis of – Hardy’s work. For aducanumab to work it means, as he predicted, that Alzheimer’s is caused by a toxic protein called amyloid. And for the first time it means there is a possible treatment to clear that protein: to deal with the causes, rather than the symptoms, of this insidious disease. The drug’s existence is so much more than many dreamt possible just two years ago. It is also so much less than Hardy hoped for.
The story of amyloid and aducanumab is a story of science itself. It is a story of dogged research; occasional inspiration; years of thankless, potentially futile graft; hubris and disappointment; feuds; triumph and caveats. Lots of caveats.
It could begin in many places. It could begin withRoger Nitsch, the University of Zurich scientist who in the early Noughties identified a natural antibody that seemed to keep amyloid at bay. It could begin in 1905 with Alois Alzheimer, a German psychiatrist whose patient lived in Frankfurt’s Irrenschloss, the “Castle of the Insane”. She was confused. She forgot her name. She forgot what to call the food on her plate. “I am lost,” she said. When she died, Alzheimer cut her open and found clumped-up plaques of protein in her brain.
But there is a date in between, in the mid-1980s, when that desperate letter landed on Hardy’s desk, that is as good a place to start as any. Genetics was a new field then and Hardy had been inspired by research into Huntington’s disease that had found a single gene was the cause. He’d put an advert in an Alzheimer’s newsletter for people who thought the same was true in their family history of Alzheimer’s. Carol Jennings, from Nottingham, replied. “Dear Sir,” she began in careful handwriting. “I was very interested to read of your research in the Alzheimer’s Disease Society News and think my family could be of use.”
Her 63-year-old father had Alzheimer’s, she said, “as does his sister Audrey. His brother Arthur also may have the disease.” Hardy remembers reading it. “She said, ‘My doctor keeps telling me it doesn’t run in the family, but here’s my family tree, and it clearly does.’” Throughout her family were people who developed Alzheimer’s in their 50s, and she wanted answers. “She’s a very determined woman, a remarkable woman.”
This is the letter that launched a thousand studies. Other families with genetic Alzheimer’s would be found. Other researchers would follow the clues down the same path that Hardy took.
Most people don’t have genetic Alzheimer’s; not in the sense that a single gene causes the condition. But in those people like the Jennings family, he thought, might lie a clue. In their mutation could be a pointer to the disease in the population at large. These days, identifying the key gene from the family members might take a postdoctoral student a month. Back then it was very different. Without modern tools, identifying the genetic markers took a team of six researchers five years. “So, 30 person-years of work,” Hardy says.
They presented the results to the family, confirming what they suspected: that their situation was not normal. Each family member had a 50 per cent chance of getting the disease while young. They were, oddly, pleased. “They were just delighted that their fears, which had been dismissed, were proved.” Then began the job of working out what it meant.
A gene is a code for making a protein. And a protein is a building block for making a body. If you find out what a faulty gene is, you can find out what it codes for. If you find out what it codes for, you can identify the building block that has gone wrong. In the case of the Jennings family, that building block was called amyloid precursor protein (APP). As its name suggests, this protein is part of a pathway that leads to the amyloid protein. In those genetically predisposed to early onset Alzheimer’s, they make too much amyloid, or it is not as soluble as it should be. From here, the chain of events seemed obvious to Hardy.
When too much amyloid is deposited it causes tangles of a toxic protein called tau, which causes neuronal death. In those with genetic Alzheimer’s this happens early, because they make too much APP. In those with normal Alzheimer’s it happens later, because their brains are too slow at clearing amyloid. Either way, amyloid is the trigger.
In 1991, Hardy and his colleague David Allsop published what became known as the amyloid hypothesis. It would, they concluded, “facilitate rational design of drugs to intervene in this process”. A cure seemed just around the corner, and Alzheimer’s was a hot research topic. “The first thing we did was organise a meeting and invited all the other [researchers],” says Hardy.
But they all had their own ideas, and not all saw the amyloid hypothesis as so obvious. Could it be triggered, in fact, by infection? Was tau the crucial target? Many old people die with lots of amyloid but no dementia – might amyloid in fact be irrelevant? Could it even be nature’s way of protecting the brain? Could it be that identifying it as the cause of dementia was like identifying a scab as the cause of a stab wound? Could you really extrapolate from people with a very rare genetic disease to people with a common non-genetic disease? It didn’t help that our knowledge of the brain was – is – so poor. “Obviously [APP] has a function. We don’t know what that function is,” says Hardy. “That’s embarrassing.”
So the field of Alzheimer’s genetics fractured. Looking back, Hardy thinks, “the disagreement was good because we did different experiments, and although we were denigrating each other’s work, that’s how we led to progress”.
Progress, though, was slow. There were mutterings that amyloid was attracting all the funding but not providing the breakthroughs. In 2009, Hardy wrote an article in the Journal of Neurochemistry addressed to those who were claiming that his idea had corralled all the money to the detriment of the field. “Whether this chorus is like the dawn chorus, heralding a bright new era of Alzheimer’s research, or a malcontent’s chorus, merely whingeing that their grants go unfunded, is open to debate.” This is not the kind of language regular readers of the journal would have expected.
Still, consensus coalesced around amyloid – so much so that in 2015 Hardy received the $3 million Breakthrough Prize in life sciences. Soon after, the results of the first anti-amyloid drugs started coming through, and Hardy was not impressed. There was gantenerumab from Roche, solanezumab from Eli Lilly and bapineuzumab from Pfizer. Each ended in “umab”, meaning they used monoclonal antibodies: artificial proteins designed to attack amyloid. Each drug had made it as far as phase three trials. Each in turn failed. There was one more umab to come.
In the late 1990s, long before blockbuster failures had led some pharmaceutical companies to abandon Alzheimer’s research entirely, Roger Nitsch from the University of Zurich was taking a different approach. Instead of trying to find out what was going on in the brains of people who had Alzheimer’s, he looked at the brains of people who didn’t. He and his colleagues looked at lifestyle explanations – smoking, diet, exercise – but also at the genetics of the immune system. Here they found a “striking effect”: some made antibodies against the amyloid protein. “We basically had the blueprints,” he says. “We thought we just needed to decipher them and they will tell us how to treat the disease.”
They probed the brains of patients who had died of Alzheimer’s. “When you see these, you intuitively understand the hypothesis because actually it’s very simple,” Nitsch says. “There are these amyloid plaques in the brain. And then there are the neurons in the brain, all over the place. The closer they are to the plaque, the more messed up they are.” They tested hundreds of antibodies to see which bonded to the plaques in the brain sections. One of the ones that did is what became known, through Biogen and Nitsch’s company, Neurimmune, as aducanumab.
While other drugs sought to remove amyloid in the blood, aducanumab did something different: it ate away at the clumps themselves. The effect on the amyloid was dramatic. “Here we have John Hardy’s hypothesis at work in real patients,” says Nitsch. “Amyloid takes about 20 years to build up. Within 12 months it’s down to normal levels – where the patient wouldn’t even be diagnosed as an Alzheimer’s patient.” The effect on patient outcomes were less dramatic, though. Whatever the amyloid was doing, cognitively speaking they definitely still had Alzheimer’s. It was only by looking at a particular slice of the trial, containing patients at an early stage in the disease, that any cognitive improvement was found. Even then it was minimal, reducing decline by a quarter.
Nitsch hopes that ongoing research in those receiving the drug will reveal larger benefits when it is correctly targeted, but he makes no claims for it as a miracle cure. “It’s almost like you have to reduce that toxicity big time in order to get a positive effect on the neurons,” says Nitsch. It is in his view a proof of principle, of what could soon be possible. “This is not the end of Alzheimer’s disease, but it’s the beginning of the end.”
Hardy agrees. In fact, when the announcement was made he used just those words. “From a scientific point of view, once you know something turns the curve, then it’s easier to do it the second time,” he says. “The analogy I use is the Wright brothers. They just about got that plane off the ground, but within 20 years we would fly people commercially across the Channel. I think that this is a marginally effective drug, but I think it points the way and the next drug will be better still.”
Since the treatment was given accelerated approval in the US last month, to headlines around the world, three members of the regulator’s advisory committee have resigned, claiming that, at $56,000, it was too expensive, the statistics about its efficacy were too unclear, and that even if it did work its effects were minimal. David Knopman, a neurologist at the Mayo Clinic, told The Washington Post after stepping down that it had been a “sham process”. The FDA subsequently narrowed the drug’s label for use in those with early Alzheimer’s symptoms. Harvard Medical School Professor Aaron Kesselheim, another of the experts who stood down, told Bloomberg the label update was a welcome change but more needed to be done to counter what he calls the “excessively optimistic early messaging about this drug”.
In a sense, John Hardy thinks his critics had a point all along. He has slightly tweaked his idea of the amyloid hypothesis. Amyloid is important, but it’s not as simple as just removing it. Imagine that today, in the instant reading this, your Alzheimer’s begins. Amyloid starts to be deposited and it does so faster than immune cells called microglia can remove it. “You’re intellectually fine and nobody, including yourself, can tell there’s anything wrong. It’s rather like the build-up of cholesterol in your heart vessels. Everything seems fine, but the disease process has started.” This is when attacking amyloid could really have dramatic effects.
If we don’t, if we miss the signs, “then, in three or four years, the microglia will start to give up”. Amyloid will build up, and so will tau tangles. This is when beating amyloid is not enough. “Think of the amyloid as a match. The tangles are rather like a fire. And once it’s lit it starts to spread, independent of the initiating match.”
There is now consensus (or as close to it as there will ever be) that we need to get to amyloid earlier. We probably need to get to tau, too, and stimulate the immune cells that help carry it all away. It is possible the “failed” umabs of the past years may work in combination with aducanumab: it reduces amyloid build-up, they take it out of the blood. We need many therapies, not just one, and this shouldn’t be a surprise. “I’m a type 2 diabetic. I take insulin but like nearly all type 2 diabetics, I take statins and blood pressure medicines. Polypharmacy is where I go,” says Hardy. “And I think that’s where we’re going to end up in Alzheimer’s disease. We’re going to end up with an amyloid drug that we should give as early as we can, and then we’re going to be looking at other drugs.
“There’s a unity in the field now that we need all these targets. Of course, people always think there’s a unity of people who agree with them.”
For Carol Jennings, her future was always a coin toss. She lost. In her late 50s, the amyloid clumps began to twist like tangled vines. They squeezed her memory, her consciousness, her sense of self. They began to eat away at who she was. When she had approached Hardy first, “She said that she didn’t think it would help her, but she hoped it would help her kids,” says Hardy. “And in fact it didn’t help her.” She is still alive but “she’s now severely demented”. The last time he met Carol was four years ago, and he is pleased to say she remembered who he was.
Thirty-five years since she got in touch, with her own father in the early stages of Alzheimer’s, is there something that can at last help Carol’s kids as she had hoped? Perhaps. Can it cure them? No – not even close. Yet in the very existence of aducanumab lies a clue to the treatment that might. It is just conceivable we are on the cusp of something astonishing.
“If you’d have asked me how long it was going to take in 1991, I’d have said five years,” says Hardy. “And the trouble is, I’ve been saying five years ever since.”