Flickr / Allan Ajifo
In March 2015, researchers at the University of Queensland reported that repeated pulses of scanning ultrasound cleared amyloid plaques from the brains of mice bred to develop Alzheimer's-like disease, with plaques cleared in 75% of treated animals and better results on three memory tests. It was a striking result, but it was a mouse study, and early headlines claiming the treatment "fully restores memory" in people went well beyond the evidence. Ten years later the approach has reached its first small human trial, which found it safe but has not yet shown that it treats dementia.
Here is what the original study did and did not find, how the technique is meant to work, and where Alzheimer's treatment stands today.
Why Alzheimer's research matters so much
Alzheimer's disease is far more than forgetting names. It is a progressive brain disease that gradually erodes memory, language, judgement and independence, and it shortens life. The World Health Organization estimates that 57 million people were living with dementia worldwide in 2021, with nearly 10 million new cases each year, and that Alzheimer's accounts for 60 to 70% of them. Dementia is the seventh leading cause of death globally.
In the United States, the Alzheimer's Association estimates that more than 7 million people are living with Alzheimer's, a number projected to approach 13 million by 2050. Older reports often called it the sixth leading cause of death in the US; the association now lists it as the fifth leading cause among people aged 65 and over.
Plaques and tangles: what goes wrong in the brain
Two kinds of abnormal protein deposits are the hallmarks of Alzheimer's. Amyloid plaques form in the spaces between nerve cells, as fragments of a protein called beta-amyloid stick together into dense clumps. Neurofibrillary tangles form inside the cells, when a protein called tau, which normally helps hold the cell's internal transport tracks (microtubules) in shape, becomes abnormal and twists into insoluble fibres. With the tracks collapsed, nutrients and other essential materials can no longer move along the length of the neuron, and the cell eventually dies.
ElectronicProducts.com
For decades, a leading idea, the amyloid hypothesis, has held that clearing beta-amyloid early could slow or stop the disease. That is why a drug-free way to remove plaques attracted so much attention.
What the Queensland team did
Gerhard Leinenga and Jürgen Götz of the Queensland Brain Institute published their results in Science Translational Medicine on 11 March 2015. They used a mouse model that builds up amyloid deposits and treated the animals with repeated scanning ultrasound: focused sound waves swept across the whole brain, combined with an injection of tiny gas-filled microbubbles into the bloodstream. No drug or antibody was used.
The microbubbles vibrate in the ultrasound field and briefly loosen the blood-brain barrier, the tightly sealed lining of the brain's blood vessels that normally keeps most substances out. The barrier closes again within hours. During that window, the researchers found, microglia, the brain's resident immune and clean-up cells, became activated and took up amyloid into their internal waste compartments, called lysosomes.
What the study found, and what it didn't
- Plaques fell. Treated mice had a lower plaque burden than mice given a sham procedure, and plaques were cleared in 75% of the treated animals.
- Memory improved. Treated mice did better on three memory tasks: a Y-maze, recognising a new object, and learning to avoid a place where they had received a mild shock.
- No obvious damage. The authors reported no evident harm to brain tissue in the treated mice.
Two details of this study are often misreported. The animals were mice, not rats. And the 75% figure refers to the share of mice whose plaques were cleared, not to "full memory restoration"; the paper reports improved performance on memory tests, which is encouraging but not the same as a cure. Götz was enthusiastic at the time, saying in the university's press material that "the word 'breakthrough' is often misused, but in this case I think this really does fundamentally change our understanding" of treatment.
Mouse models also have limits. They usually carry human genes that drive amyloid build-up, but they do not reproduce the full disease, including the spread of tau and the extensive nerve cell loss seen in people. Many treatments that worked in mice have failed in human trials.
From mice to people: the 2025 safety trial
In 2015 the team said it hoped to test the method in larger animals such as sheep and to start human trials around 2017. Progress took longer. In December 2025 the University of Queensland reported the first human results: a pilot study of 12 people with Alzheimer's disease, carried out at Mater Hospital Brisbane and published in Brain Communications, tested different strengths of scanning ultrasound. Professor Peter Nestor described the treatment as fast, safe and well tolerated in all participants.
The trial was designed to check safety, not effectiveness, so it cannot show whether the treatment works. Some caregivers reported that participants seemed less agitated, and the team is planning a further trial focused on behavioural symptoms. Götz cautioned that "further trials in a larger group of participants will be needed" before any conclusions about dementia symptoms.
Other groups worldwide are also studying focused ultrasound, both to open the blood-brain barrier on its own and to help antibody drugs reach the brain more easily.
Where treatment stands today
In 2015, no medicine could change the course of Alzheimer's. That has changed. On 6 July 2023 the US Food and Drug Administration granted traditional approval to lecanemab (Leqembi), an anti-amyloid antibody that slowed clinical decline by 27% over 18 months compared with placebo in the Clarity AD trial. A second antibody, donanemab (Kisunla), was approved in the US in July 2024.
Both drugs are for people with mild cognitive impairment or mild dementia due to Alzheimer's, require regular infusions and scans, and carry warnings about brain swelling and small bleeds known as ARIA. They slow decline modestly rather than stopping or reversing it. Interest in drug-free or drug-assisting approaches such as ultrasound remains high for exactly that reason.
Myths and practical takeaways
Myth: ultrasound cures Alzheimer's. It does not, at least not yet. It cleared plaques and improved memory tests in mice and has passed a small human safety test.
Myth: nothing can be done. Medicines can now slow early disease for some people, and the WHO lists modifiable risk factors linked to dementia, including high blood pressure, diabetes, smoking, harmful drinking, physical inactivity, social isolation, hearing loss and poor sleep.
If you or someone close to you is noticing worrying memory changes, talk to a doctor. Many causes of memory problems are treatable, and an early assessment opens the door to support, planning and, where appropriate, clinical trials.
Sources: Leinenga and Götz, Science Translational Medicine (2015), PubMed abstract; University of Queensland News (December 2025): Human trial finds ultrasound treatment for dementia is safe; WHO: Dementia fact sheet; Alzheimer's Association: Facts and Figures; BioSpace: FDA grants traditional approval for Leqembi (6 July 2023); Wikipedia: Donanemab; story first reported by True Activist and ScienceAlert; images via Flickr (Allan Ajifo) and ElectronicProducts.com.
COMMENTS