Cochrane Review that concluded treatments that remove amyloid from the brain don’t work reaps backlash over flawed methodology

A Cochrane Review of amyloid‐beta-targeting monoclonal antibodies in people with mild cognitive impairment or mild dementia due to Alzheimer’s disease has concluded that the impact of the drugs on cognitive function and dementia severity at 18 months is “trivial,” and any improvement in functional ability is “small at best.”

It went further, saying that successfully removing amyloid from the brain “does not seem to be associated with clinically meaningful effects” and future research on disease-modifying treatments should focus on other mechanisms of action.

But the backlash has been swift, with experts around the world slamming the review for fundamentally flawed methodology.

Questionable decision to pool all seven drugs into a single class-level analysis

The review included 17 studies of seven different drugs. Just two of the studies were for the two monoclonal antibodies that are currently approved by the TGA and FDA; one for lecanemab and one for donanemab.

In other words, the researchers analysed two trials of drugs that showed robust benefit on the outcomes measures alongside 15 trials for drugs that never made it to market, explains Steve Macfarlane, head of clinical services at Dementia Support Australia and an Associate Professor of Old Age Psychiatry at Monash University.

“I really don’t think it’s a fair comparison to look at 15 failed trials and two trials that have clearly been successful and have led to approval of the drugs by the TGA and the FDA, and say that on average, these drugs don’t work,” Associate Professor Macfarlane says.

Professor Chris Rowe, a neurologist, Alzheimer’s researcher and director of the Australian Dementia Network, agrees.

Analysing the drugs that we know don’t work together with the ones that do, “is just totally illogical,” he says.

Drugs included in the review had different mechanisms and target different areas of amyloid – and the trials used different designs and outcome measures.

“They’re not the same drugs. Some of them had the wrong dose, and it was only from learning from the failures that we got to the ones that worked in the appropriate dosage,” Professor Rowe notes.

“Only about 0.1% of what you inject, or less than that, actually makes it into the brain. So early on people using far too low doses.”

“And they are all targeting different areas of amyloid. Some of those targets, while they look good in a test tube, clearly didn’t work in the human brain,” Professor Rowe adds.

Associate Professor Macfarlane says the stage of disease also has a big impact on results, and earlier trials often involved patients who were too far advanced in the disease.

“For patients with established disease, it’s really asking a lot more of a drug to show benefit because much more damage has been done.”

“What we’ve seen in particular with the lecanemab and donanemab trials is that they’ve been moving the intervention back in the process to people with very early Alzheimer’s disease and indeed mild cognitive impairment that’s due to Alzheimer’s pathology.”

An important starting point

Clinical trials suggest lecanemab and donanemab may slow the progression of cognitive decline by around 25-35%, with patients who start at an earlier stage of illness getting the most benefit. While the initial trials were 18-months, we now have data suggesting the benefits continue out to four years. Still, while the medicines “buy time,” they don’t stop or reverse decline – and they do have risks.

“These drugs aren’t perfect. They’re just a first start. They don’t make you better. They don’t improve your symptoms. They just buy you a bit of time, like the early days of chemotherapy,” Professor Rowe says.

“This is just the start of the journey to finding really effective treatments for Alzheimer’s disease.”

Associate professor Macfarlane says the forest plots in the full review show unambiguous benefit of lecanemab and donanemab, where the other drugs have failed.

“Sure, the effect size is small. It’s not a dramatic, world-changing benefit that’s seen, but it’s a clear and obvious and undeniable benefit,” he says.

Moving beyond amyloid?

Professor Paresh Malhotra, head of the division of neurology at Imperial College London, has also weighed in on the Cochrane Review.

“I agree with the authors that other mechanisms of action beyond targeting amyloid for treatment of Alzheimer’s should be explored,” he says.

“However, the findings to-date do not justify ‘throwing the baby out with the bathwater’ and dismissing all the results of well-conducted individual studies.”

Currently multiple clinical trials are also looking at drugs that target another protein called pTau, Associate Professor Macfarlane says.

“I suspect that in 10 years’ time we’ll be seeing combination therapies of both anti-amyloid and anti-pTau targeting drugs used in combination in the same patient.”

Key takeaways

If patients want to discuss the risks and benefits of anti-amyloid therapies it’s crucial to limit the discussion to the two that are approved: donanemab and lecanemab, Associate Professor Macfarlane says.

“The results of failed trials on different drugs are not relevant to a decision a patient needs to make when they’re sitting in front of their GP,” he says.

For more information

Nonino et al. Cochrane Database Syst Rev. 2026;4:CD016297.