In One Donated Brain, Tau Stayed Quiet Where Amyloid Had Been Cleared

A case report in JAMA describes the autopsy of a single Alzheimer's trial participant: in brain regions where the anti-amyloid drug aducanumab had cleared plaques years earlier, tau tangles and tissue loss were markedly lower than in neighboring, less-cleared regions of the same brain.

Pathologists at the University of Pennsylvania who examined the donated brain of a man who had received the anti-amyloid antibody aducanumab for four and a half years found that regions where the drug had cleared most amyloid plaque carried far less tau pathology, the protein tangle most closely tied to memory loss, than nearby regions where amyloid persisted, according to a case report published in JAMA. The man, who carried a rare TREM2 gene variant linked to higher Alzheimer's risk and had mild cognitive impairment, received 30 doses of aducanumab over 4.5 years in the EMERGE and EMBARK trials and died four years after his last dose. At autopsy, amyloid clearance was patchy rather than complete: low-amyloid patches sat mainly at the crests of brain folds, while amyloid remained high in the depths of folds and in deep cortical layers, an uneven pattern the researchers say has not been described before.

Comparing the low-amyloid regions of this one brain to postmortem tissue from 14 untreated people matched by age or by the same TREM2 variant, the Penn team, led by co-senior authors Edward B. Lee and David Wolk, reported that the cleared regions had less tau buildup than is typical for Alzheimer's disease. Brain scans taken while the man was alive showed the same pattern: cortical thickness declined more slowly over time in the areas that would later prove low in amyloid at autopsy (beta estimate -0.50; 95 percent confidence interval -0.62 to -0.37). The finding matters because most evidence that anti-amyloid drugs help comes from cognitive scores and imaging in large trials; direct looks at human brain tissue years after treatment, showing what happened downstream to tau and tissue structure, are rare.

The researchers and outside readers of the case report note firm limits on what one brain can show. This is a single patient with a genetic variant that itself affects how the brain's immune cells respond to amyloid, so the pattern seen here may not generalize to people without that variant. Amyloid clearance in this brain was uneven, not complete, which is itself part of what is being described, not a controlled comparison of full clearance versus none. The comparison to 14 controls is observational: it shows that cleared regions and persistent-amyloid regions differed, not that clearing amyloid caused the lower tau seen later.

The authors say the pattern needs to be checked in more donated brains from people treated with amyloid-clearing drugs before anyone can say how reliably it holds.