The false hope of the new Alzheimer’s drugs The NHS was right to reject them

Now, we are told that “routine screening for over-50s [is] closer”. Research published this week in JAMA Neurology linked blood test results of a protein implicated in Alzheimer’s disease, tau, to a later changes on scans. The clamour for the “wonder drugs” will inevitably grow louder. But can we believe the hype?

Dementia has long existed across human history. And Alzheimer’s is the most common form. While advances in medicine have increased our understanding of the disease, and we know that age, genetics and lifestyle play a part in the progression of it, we still don’t know what triggers it. Aluminium exposure is one theory; infection and immune system malfunctions have also been posited, though are treated as unproven. But the most enduring one, and the one into which most funding flows, is the “Amyloid Hypothesis”.

Amyloid protein is found in all healthy people but in Alzheimer’s disease, the protein is found to have behaved abnormally: it clumps together forming plaques which cause “tangles” which disrupt communication between the brain cells. The cells malfunction, die and, over time, different areas of the brain shrink. The “clumps” cause the production of another protein, tau, which causes tangles and more damage; the two have been called the “trigger and bullet” in the disease. 

The two new “wonder drugs” both address the amyloid theory. Both are “monoclonal antibodies”, which target the protein in different ways. Donanemab’s antibodies recognise and remove amyloid when it is in plaque form, while lecanemab targets the amyloid as it forms fibres and tangles. But they don’t come cheap. Lecanemab is priced at £21,000 per patient annually with donanemab expected to be similar. And this, along with their efficacy, will be what Nice is looking at before it decides to recommend the drug — or not — later this year.

The new drugs certainly seem credible. Large trials have been conducted and published in high-profile journals, examining their impact on memory, brain scans and blood test results.

The Trailblazer study investigating donanemab was published in JAMA last year. It assessed over 1,700 adults, with an average age of 73, who had been diagnosed with early Alzheimer’s. Over 18 months, half of those studied received the active drug and half the placebo, all of them completing disease scores, indicating symptoms and cognitive ability as the trial progressed.

Eli Lilly, the drug developer, stated in a press release that “47% of participants on donanemab showed no decline on CDR-SB, a key measure of disease severity after 1 year, compared with 29% of participants on placebo.”  Moreover, “participants on donanemab had 40% less decline in ability to perform activities of daily living at 18 months”. Meanwhile, images of brain scans which seem to persuasively illustrate the disappearance of amyloid were widely published.

Given that dementia progresses over a period of time at differing speeds, the developers were keen to highlight their “rate of decline” measurement. The implication of their results was a significant slowing of the disease. But that by no means tells the whole story. Participants in the trial also had their cognitive function measured on the Alzheimer Disease Rating Scale. For those taking the placebo, their scores dropped by an average of 13 points. For those taking donanemab, their scores also dropped — but by an average of 10 points. So both groups found their cognitive function worsened, but with a difference of three points on a scale of 146. The question to ask must surely be: how much difference do those three points make.

As for the lecanemab trials, Biogen and Eisai announced “highly statistically significant results… representing a 27% slowing of decline”. These were conducted in a slightly different fashion. The 1,795 adults, ranging in age from 50-90, were assessed on an 18-point score in tests which looked at their memory, problem solving, hobbies and abilities in personal care. Half got the active drug, the other half, placebo.

After 18 months, both groups had experienced decline, as would be expected in people with dementia. Those taking lecanemab had scores which had declined by an average of 1.21 points. And those on the placebo had deteriorated more, by an average of 1.66 points over the course of the trial. That’s a difference of 0.45 points on an 18-point scale, described as “moderate” in the paper.

It is also important to note that the people who were recruited to the trials were only those with the “purest” forms of Alzheimer’s. For every 10 patients put forward, seven or eight were rejected. Those who were accepted had high amyloid levels but were relatively young and free of other diseases. To pick these patients, according to Dr Seb Walsh, a public-health doctor researching dementia at Cambridge University, is to misrepresent how the disease occurs among the vast majority of the population. Most people’s dementia is complex, occurring when they are in their eighties, and caused by several disease processes.

“By selecting amyloid-only patients,” he says, “they were giving the drug the best possible chance to show an effect, and yet even so they found an effect (after 18 months of fortnightly infusion treatment) that was so small it probably wouldn’t be noticeable to a doctor.” If the drug were given to a broader range of people diagnosed with Alzheimer’s disease, the already-small improvements might even disappear altogether.

And there’s another problem. The side effects.

Over a quarter of the participants in the lecanemab trial suffered drug infusion reactions, while there was brain swelling (which can be mild, such as flushing, or more severe, with changes in breathing and heart rate) in 13%. Others experienced headache, confusion and visual disturbance. Donanemab had similar side effects and three people also died in the trial — something researchers ascribed to the treatment.

So, while the press releases make a persuasive case for the effectiveness of the drugs, the overall difference is so small as to invite the question as to whether patients or families would even notice. What sort of improvement would these small differences represent in the real-world? And more important: are they worth the risk?

“My view,” says Professor Robert Howard, Professor of Old Age Psychiatry at University College London, “is that we’ve just about reached the point where we couldn’t have treatments that remove amyloid more aggressively because of the side effects of brain bleeds and swelling.”

It is this finding that is exercising — and dividing — scientists. For what does it mean for the new drugs if, as the trial showed, a big reduction in brain amyloid doesn’t translate into a hugely impressive reduction in or slowing down of symptoms.

One body of clinicians continues to hold to the hypothesis. Craig Ritchie, a Professor of the Psychiatry of Aging at the University of Edinburgh, worked on the donanemab trial and claims it has “a profound effect on the core pathology of Alzheimer’s disease namely cerebral amyloid”. He maintains that the slowing of the progression of the disease coupled with a reduction in amyloid build-up is not a coincidence. He believes that more time is needed to show the full impact of the reduction in amyloid on symptoms. “These drugs give us hope that we can do something, that these are the first drugs of an important new generation of treatments.” In order to push out this treatment, he has quit the NHS and set up Scottish Brain Sciences as an “independent research company” to run studies and give patients free access to the latest diagnostic tests and medications through trials funded by the drug industry. He, too, is funded by the industry.

These proponents of the new drugs also insist that monoclonal antibodies will show more impact if they are given before amyloid has damage to the brain. Therefore, they say, it needs to be given earlier — possibly even before someone has any symptoms of dementia.

That’s where the proposal that screening people — the so-called “Alzheimer’s blood test”  would come in. It would test people before they have any memory problems for their future dementia risk. But, offered as a “check up”, it would come with wide ethical issues. There would inevitably be false positives and negatives — causing needless anxiety or false reassurance. Nor would a result necessarily lead to people being offered treatments that could lower their chances of getting dementia later. But could it make sense in a research setting, as an opportunity to try these new treatments to see if they could make a difference if given earlier? 

Professor Howard is sceptical. The amyloid treatment, he says, “is a cul de sac, a dead end. And, worse, it means that the money and energy in drug development and trials isn’t being put into something that might work.” he says. “The argument that maybe the drugs need to be given earlier to make a difference is really just a wish or a hope. There is no data to suggest that it is true,” he says. “I worry that the “treat earlier” argument has become a way of saying ‘don’t blame the drugs for not working, blame the patients’.”

Various Alzheimer’s charities have heavily advocated for these drugs to be made available quickly. When lecanemab was approved in the US, a public letter hosted by the US Alzheimer’s Association in 2023 described it as “a foundational gamechanger”, saying: “No barrier can be allowed to stand between our patients and a treatment that has a reasonable risk-benefit ratio.” There may be a less philanthropic reason why they are so enthusiastic about the new treatments. A closer look at the signatories revealed that some had drug industry connections. And the charity itself received over $750,000 from Eisai and $430,000 from Biogen in 2023, who make lecanemab.

The inevitable result of these two amyloid-removing drugs being licensed means all drug companies will want to make and market their own versions, regardless of the criticisms. After all, it’s harder for regulators to say “no” when there are similar drugs on the market and the big charities are rooting for them.

But that hasn’t stopped the charities from investing £5 million in developing a blood test that will “revolutionise dementia diagnosis”. They say that currently only 2% of people get the detailed diagnosis they need and that significant investment is needed to ensure the NHS can identify people with Alzheimer’s disease far sooner. The Blood Biomarker Challenge, available within five years, will help do that. “If we can fix diagnosis, we open the door for a cure. It’s a bold ambition but, with someone developing dementia every three minutes in the UK, we must aim high,” they say.

But Dr Seb Walsh, a public health doctor researching dementia at the University of Cambridge, says the hype and hope is unkind: “For 20 years we have been promised wonder drugs within 5 years — but still we wait,” he says.

He also says it’s bad research practice to pre-empt your results before you do the research. “We don’t know how useful these tests will be, how well they’ll perform in the ‘typical’ NHS dementia patient (more complex than the people in existing research studies), that’s the whole point of doing the research.”

But the key question, and the one that Nice will be weighing up over the course of the year, is whether any of these new drugs represent the best use of public money. Given the small benefits, the risk of side effects, but the lack of other decent alternatives, the question will be how to weigh that up as independently as possible. Spending similar sums on social care might well provide greater benefits for patients and families than the drugs. Is it right that a drug — with minimal benefits — should be funded when the same money invested into social care might do more good for patients and families getting more practical help?

Obviously, a drug that worked would have massive, global market potential. But the hyperbole from doctors and researchers in this area — especially those with industry funding — is unhelpful. It means that the pressure to upsell the effectiveness of a drug is particularly intense. This, then, diverts money into look-alike drug development and drains it from continued research. Clearly the pharmaceutical industry and its relationship with the private and public sectors has a lot to answer for. Currently, Alzheimer’s has no cure and treatment has progressed little despite decades of research and billions of pounds. As our investigation shows, we should be cautious about the promise of these new “wonder” drugs which risk benefiting only the drugs companies.

“There is still much we don’t know, but we really need to know,” Dr Walsh says. “We are unlikely to ever, truly ‘cure’ dementia — this is more false hope.”

---