MISRYOUM Science News: Alzheimer’s has a way of making simple stories feel like they were built on sand. The basic plot—amyloid-β gets cut off from its precursor, clumps up in the brain, and that clumping drives dementia—was persuasive enough to steer decades of research.
Something about the idea caught on fast because it came with something scientists could point at. In 2006, a paper in Nature linked memory loss to a specific form of amyloid-β buildup outside of neurons. In a field where so much is murky, that kind of connection felt like a handle: if you can see it, maybe you can grab it.
The logic then became pretty straightforward: if aggregation is the problem, stop it—or remove the aggregates—and the disease should ease. So researchers leaned into what they do best: build targeted experiments. Since mice don’t naturally get Alzheimer’s, scientists had to engineer it. They inserted a mutated copy of the human APP gene into mice, creating a transgenic model that mimicked aspects of amyloid pathology. It’s a useful shortcut, though not everyone fully trusts the map.
In 1999, Elan Pharmaceuticals created a vaccine aimed at a particular part of amyloid-β. The company then showed that mice cleared plaques from their brains after treatment. And the result wasn’t limited to the kind of “prevention” scenario that makes headlines—Elan reported it worked in very young mice before plaques could form, but also in older mice where plaques were already present.
The vaccine strategy is fairly intuitive: prompt the body to produce antibodies against what the vaccine recognizes. Elan later moved to a related approach—anti-amyloid antibodies given directly. Those, too, helped clear plaques in the brains of the transgenic mice. For a while, you could almost feel the field relaxing—like, okay, this is it.
But the distance between animal results and human outcomes is brutal. Misryoum newsroom reported that Elan tried the vaccine in human patients with mild to moderate Alzheimer’s, and the trial had to be suspended after 360 patients developed brain inflammation. The vaccine didn’t go anywhere. Still, the amyloid idea didn’t vanish; other pharmaceutical companies and biotechs kept testing variations—different parts of the amyloid-β pathway, different ways to interfere.
Trial after trial failed to arrest or reverse the disease. Even when the science pushed forward, the outcomes often pushed back harder: targeting different parts of the amyloid-β pathway created side effects aplenty, some of them life-threatening or fatal. Amyloid-β stayed the preferred target anyway—because it was still the best-supported culprit on paper, even as the evidence in practice got messier. Then, in 2021, the Food and Drug Administration approved an antibody called aducanumab, made by Biogen.
It’s one of those moments where the room goes quiet—some people see permission to keep going, others see a warning label. And if you’ve ever sat through a lab update where someone reads off trial outcomes and the mood sinks, you know the feeling: not exactly surprise, more like a grim recognition of how hard this disease is. The amyloid theory isn’t “over,” exactly, but it certainly doesn’t look as clean as it once did—maybe it never did, not for long.
Project Hail Mary’s space science: how plausible?
Why social science needs honesty about uncertainty
Some GLP-1 drugs work better for people with a key gene variant