When your brain swells during treatment for Alzheimer’s disease, that’s usually the point where clinicians pump the brakes. They slow the infusion, monitor you closely, wait for the puffiness to settle. It is, by any reasonable measure, a side effect you’d rather avoid. But a small study from Houston Methodist Research Institute now suggests that these swollen patches of brain may actually be the places where the treatment is hitting hardest, stripping away the amyloid plaques that are the hallmark of the disease.
The side effect in question is called ARIA-E, short for amyloid-related imaging abnormality with edema. It crops up in some patients receiving the newer generation of anti-amyloid antibody drugs like lecanemab and donanemab, which work by targeting and clearing sticky clumps of beta-amyloid protein from the brain. When it happens, plasma leaks from blood vessels into surrounding tissue, causing localised swelling visible on MRI scans.
ARIA-E has been a persistent headache for the Alzheimer’s treatment field. In clinical trials of these drugs, it shows up in a meaningful fraction of patients, particularly those carrying the APOE ε4 gene variant, which is itself a major risk factor for Alzheimer’s. Clinicians treat it cautiously, pausing or slowing drug infusions until the swelling resolves, which typically takes about two months. The worry, understandably, is brain damage.
Joseph Masdeu at Houston Methodist and his colleagues wondered whether something more interesting might be going on. Their study, published in the American Journal of Neuroradiology, looked at five patients who had developed moderate or severe ARIA-E while receiving lecanemab or donanemab. All five carried at least one copy of the APOE ε4 allele. Four were women. Their average age was roughly 74.
The team used PET brain scans to measure levels of beta-amyloid before and after the swelling had come and gone. They parcelled each patient’s brain into 40 regions, then compared the amyloid drop in regions that had swelled with those that hadn’t. In four of the five patients, the regions that had experienced ARIA-E showed a significantly larger decrease in amyloid signal. The brain, it seemed, wasn’t just swelling for nothing.
“This study shows that not all parts of the brain respond equally to anti-amyloid therapy,” Masdeu said. “That reframes ARIA-E from being only a side effect to a possible sign of strong local treatment activity.”
Why might swelling and amyloid clearance go hand in hand? The researchers proposed several mechanisms. One possibility is that the antibody drugs make local blood vessels more permeable, which lets more of the drug through to clear amyloid but also causes the fluid leakage that shows up as edema on scans. Another is that immune cells called microglia are gobbling up amyloid in those regions, a process known as phagocytosis, which has been documented in autopsy studies of patients treated with lecanemab. That engulfment could hide amyloid from PET tracers without necessarily removing it from the brain entirely. A third possibility involves the brain’s waste-clearance plumbing, the glymphatic system, getting gummed up by immune complexes deposited around blood vessels, leading to fluid pooling in the white matter.
It’s probably some combination of all three, and perhaps other mechanisms besides.
The study has obvious limitations, and Masdeu’s team is upfront about them. Five patients is tiny. All were APOE ε4 carriers, so the findings might not generalise to people with different genetic backgrounds. The one patient who didn’t show the pattern had only a single small region of ARIA-E and had been on treatment much longer before the swelling appeared, which may have blurred the regional differences. And without brain tissue to examine, the team can’t say definitively whether the amyloid is truly gone or merely hiding from the scanner.
Still, the consistency across four of five patients is striking. The team is now expanding its sample and collaborating with the LEADS consortium to validate the results in larger, more diverse groups. With nearly 7 million Americans currently living with Alzheimer’s and that number expected to double by 2060, any insight into how these new treatments actually work in the brain matters.
For clinicians and families navigating the uncertainties of anti-amyloid therapy, the finding offers a slightly different way to think about a worrying side effect. Brain swelling remains something to take seriously and manage carefully. But it might also be a signal that, in those particular patches of cortex at least, the drug is doing exactly what it was designed to do.
Study link: https://www.ajnr.org/content/early/2025/12/09/ajnr.A9111
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.