Yves Forestier / Getty

In 2016, Darren Baker and Jan van Deursen from the Mayo Clinic announced that they had discovered a new way to prolong the life of mice: They cleansed the rodents of retired cells.

Over time, the cells of complex organisms accrue damage in their DNA, which threatens to turn them into tumors. Some cells defuse this threat by entering a state called senescence: They don’t die, but they permanently stop growing and dividing. These retired cells accumulate as we get older, and despite their name, they’re not idle. They secrete molecules that trigger inflammation, and they’ve been implicated in some of the health problems of old age. By clearing them from mice, Baker and van Deursen slowed the aging process in many of the rodents’ organs, and in some cases extended their life.

Now the duo has shown that the same approach could benefit the brain by preventing degenerative diseases that afflict neurons. By removing senescent cells from a strain of mice that shows symptoms of Alzheimer’s, the team prevented one of the disease’s hallmarks: the buildup of a protein called tau, which forms toxic, tangled clusters inside neurons. Without these tangles, the aging mice didn’t lose neurons as they normally would, and their memories remained intact.

Baker cautions that he has no idea if the same approach would work in other strains of lab mice that imitate the symptoms of Alzheimer’s in different ways—and much less if it would work in people. Alzheimer’s researchers have an atrocious track record of translating promising findings from rodent studies into actual treatments. After decades of false starts and failed clinical trials, they are understandably wary.

Nevertheless, most of these previous trials involved drugs that were designed to get rid of a different protein that’s thought to lie at the heart of Alzheimer’s, amyloid-beta. The concept of going after senescent cells is new. “If these cells are found to play important roles in people with neurodegenerative disease, the implications for treatment could be quite significant,” says Li-Huei Tsai from MIT, who was not part of the new study.

That’s because there are several “senolytic” drugs that can eliminate senescent cells, and some of these have already been approved for treating cancers. “If they prove effective in preventing or slowing neurodegeneration, it would represent a truly major advance, especially in light of the continued failures of amyloid-based clinical trials,” Tsai says.

Despite those failures, or perhaps because of them, the National Institutes of Health has dramatically increased its spending on Alzheimer’s research, tripling its annual budget over the past three years to $ 1.9 billion. This financial explosion was partly meant to entice new researchers into tackling the puzzle of Alzheimer’s—and in Baker’s case, it certainly worked.

His team, including Tyler Bussian and Asef Aziz, worked with a strain of mice that accumulates tau tangles in its neurons by six months of age. By eight months, the mice’s neurons have started dying, their brain has started shrinking, and their memories have begun to falter. But Bussian and Aziz found that all of these problems are preceded by the buildup of senescent cells, and can be prevented by eliminating those cells at an early stage.

They did this in two ways: first, by genetically engineering the mice’s bodies to destroy their own senescent cells when they are fed a particular chemical; and second, by using a senolytic drug called “navitoclax” that kills those cells directly. Both approaches were successful at preventing tau tangles.

Navitoclax was originally developed as an anticancer drug, and it’s one of several tumor-fighting drugs that might also double as a senolytic. To kill tumors, such drugs must be delivered at high doses, so that no tumor cells escape. But when killing senescent cells, total annihilation isn’t necessary, and it might be possible to get away with far lower doses. A company called Unity Biotechnology is busy developing several senolytic drugs, and Baker is an inventor on some of its patent applications.

“Senescent cells have long been overlooked in research on neurodegenerative diseases,” he says. They’ve been detected in various conditions, but no one knew if they were a cause or a consequence. “We show that they can be an active part of the problem.”

To Baker’s surprise, it wasn’t the rodents’ neurons themselves that became senescent, but the glia—the surrounding brain cells that protect and support neurons. Baker suspects that when neurons start producing tau, they send distress signals to the glia. Those prolonged cries for help nudge the glia into a senescent state. And they, in turn, release chemicals that change the neurons, transforming their previously low levels of tau into toxic tangles. “It’s like throwing gasoline onto a match that keeps trying to light,” Baker says.

For the moment, that’s just a guess. A second team, led by Miranda Orr at the University of Texas Health Science Center at San Antonio, also recently linked senescent cells to tau tangles—but in its study, the neurons themselves showed signs of senescence.

These kinds of conflicting results are to be expected, and they reinforce how little is known about the complex interplay between senescent cells, tau, and brain diseases. Tara Tracy from the Buck Institute adds that tau comes in many guises that all contribute to brain problems. Baker’s team only looked at one kind—and in one particular strain of mice. Big questions remain, she says, but the team has taken some important initial steps.  

Baker notes that his team only cleared senescent cells in mice that have yet to show memory loss or signs of dying neurons. “What would happen if we did this in mice that already have those problems?” he asks. “Could we halt disease progression and revert them to a more youthful state? I’d be very surprised to find that, but we have to do the experiments and see.” That’s a crucial question, Tracy adds, given how difficult it is to diagnose people with Alzheimer’s disease before they show symptoms.

Baker and his team’s earlier work, in which they slowed the aging process by clearing senescent cells, “seems to be holding up in other labs,” says Norman Sharpless from the National Cancer Institute. “So I remain enthusiastic about the approach, although it is still early days.”

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.