Skip to Main Content

A generic drug used widely to treat swelling associated with hypertension and heart failure showed hints in early research that it may also prevent the devastating brain damage of Alzheimer’s disease, a surprising twist that suggests scientists have a lot more to learn about the root cause of the neurodegenerative condition.

The findings, reported Monday in Nature Aging, show how the drug, bumetanide, reversed signs of Alzheimer’s in mice, as well as in human brain cells in lab dishes. The new study also detailed real-world data mined from millions of patients’ electronic health records showing that people over the age of 65 who regularly took bumetanide were 35% to 75% less likely to be diagnosed with Alzheimer’s.


Since efforts to develop drugs for the disease have been riddled with failures and controversy, the results suggest a therapeutic approach that radically departs from the one that has dominated for three decades.

Any new therapies are likely still years away — the study was in mice and human cells, which are often poor predictors of what will work in patients. And the researchers can’t yet explain why the drug would have the purported effects on Alzheimer’s. But the fact that bumetanide has already been approved by the Food and Drug Administration should speed up clinical testing, which the study authors are now pursuing.

“Developing new drug targets for Alzheimer’s disease takes a lot of time and money, so we wanted to find a faster way to move drugs to patients,” said Yadong Huang, a neurobiologist at Gladstone Institutes and a co-author of the new study.


In 2017, he launched a new center to apply a computational approach for repurposing FDA-approved drugs for new uses. And from day one, his team set their sights on Alzheimer’s disease. “We didn’t want to be biased toward any hypotheses about the mechanisms of Alzheimer’s disease, so we took a step back and instead looked at how the whole gene expression profile altered with disease progression.”

One of the hallmarks of Alzheimer’s disease is the buildup of mangled beta-amyloid proteins into plaques. Many scientists believe those plaques trigger neuron death, leading to the disease’s characteristic cognitive declines. Drug development has been driven by the idea that clearing out the plaques should reverse or at least slow the disease’s progression. But a litany of failed clinical trials has made the idea harder and harder to defend. Then, this summer, the FDA approved Aduhelm, Biogen’s amyloid-eliminating drug, despite minimal impact on patients’ cognition. (A STAT investigation found that the drug was resuscitated from clinical failure in part by a secret Biogen persuasion campaign dubbed “Project Onyx,” the details of which are now being investigated by the Office of Inspector General in the Department of Health and Human Services.)

The new study does not overturn the controversial “amyloid hypothesis.” But it did uncover a host of previously uncatalogued biological irregularities that show up inside the brains of Alzheimer’s patients.

One of the biggest genetic risk factors for Alzheimer’s disease is having one or more copies of the gene APOE4. About 25% of people possess one copy — which raises the risk of developing Alzheimer’s by three- to four-fold. About 2% carry both copies, raising the risk 12 to 14 times compared with those who don’t carry the APOE4 variant. Huang’s group looked at what was going on inside the brains of Alzheimer’s patients carrying one or more copies of APOE4 and found nearly 2,000 other genes whose expression had been altered compared to people without the disease. Many of these genes link back to pathways that have nothing to do with beta-amyloid metabolism, including those involved with circadian rhythms, morphine addiction, and GABA — a neurotransmitter that keeps neurons in check, preventing them from firing too often.

Huang’s team then scanned a database of 1,300 drugs to find ones that could flip those altered genes back to a healthy state. Among the top five hits was bumetanide, a powerful diuretic first approved by the FDA in 1983.

They tested its effect first on a line of mice engineered to have two copies of the human APOE4 gene, causing them to develop memory problems and other cognitive deficits around 15 months of age — the equivalent of 60 in human years. Bumetanide treatments significantly boosted how well these mice performed on various cognitive and spatial memory tests.

They then repeated the experiment in another line of mice, who in addition to APOE4 genes also carried the human gene for APP — a protein that when broken down becomes beta-amyloid. These mice typically develop plaques at 6 months of age. But bumetanide treatments delivered at 10 months shrank these plaques and restored healthier brain function. Taken together, the results suggest that bumetanide is improving symptoms without specifically targeting plaques.

“There are many cellular and molecular changes in Alzheimer’s disease patients besides plaques, but we usually don’t talk about them,” said Huang. “These results suggest that in order to treat Alzheimer’s we should probably not target only one or two but multiple genes and multiple pathways involved in the disease.”

The findings raise hopes that aiming not so narrowly at beta-amyloid, and instead at the full cascade of molecular changes, might be more effective. But this hypothesis-free approach can also be a bit unsettling. Sure, it looks like the drug might work for Alzheimer’s disease. But why would it work? No one knows.

“The mechanism of the drug is well-known, but what the authors haven’t addressed is how that mechanism is related to what they think might happen if they were to give this drug to Alzheimer’s patients,” said Shilpa Kadam, a neurologist at Johns Hopkins University who studies developmental disabilities. She has been closely following a trend of physicians prescribing bumetanide off-label to people with autism, epilepsy, and other brain disorders. But in these instances of drug repurposing, the mechanism for why the drug might provide a benefit is well-understood. Bumetanide reduces water retention by blocking proteins that shuttle salts across cell membranes. Those proteins are also found in neurons, and if those neurons are overexcited — as occurs in conditions like epilepsy — blocking them can restore balance.

However, bumetanide can also be tricky to use because it can leave patients dehydrated and with their electrolytes off-kilter, said Jeffrey Cummings, director of the Chambers-Grundy Center for Transformative Neuroscience at the University of Nevada Las Vegas, who was not involved in the study.

“This drug’s relationship to Alzheimer’s disease is not quite proven and its side effect profile is undesirable in older people,” said Cummings. Which is one of the reasons why he thinks rushing this drug into clinical trials might not be prudent. The other is lapsed intellectual property, which will make the generic harder to profit from, and thus less appealing to biotech companies.

“I would see this much more as pointing us toward a repertoire of pathways that have not been adequately investigated,” said Cummings. “And expanding the targets that might be beneficial in human cognition strikes me as a really good outcome.”

Huang told STAT his team is now working with a number of medical research centers to launch a clinical trial for Alzheimer’s patients who carry at least one copy of APOE4. He hopes it is a first step in overturning another idea that has long shaped Alzheimer’s drug development; that it is just one disease treatable with one magic bullet.

“These patients may have different underlying cellular mechanisms that lead to their neurodegeneration,” he said. “If that’s true, then you can imagine the disease can be divided into subgroups that require different treatments. More and more people are accepting this concept, but it’s definitely still an emerging idea.”

Correction: A previous version of this story incorrectly described the year bumetanide was first approved by the FDA.


Create a display name to comment

This name will appear with your comment

There was an error saving your display name. Please check and try again.