Aducanumab Amyloid Clearance Tracks With Less Tau Pathology in a TREM2 Carrier
A single clinicopathologic case is not a trial, but it can still answer a question that randomized endpoints leave open: what happens in brain tissue after amyloid is cleared? In this report from a memory center, a man in his 50s with mild cognitive impairment and the p.R47H TREM2 variant received aducanumab over 4.5 years and later underwent autopsy that allowed direct comparison between treated tissue and untreated controls. The finding was unusually granular. Regions with minimal residual amyloid were not uniform across the cortex, and the pattern of clearance lined up with less downstream tau burden and slower structural decline on imaging.
The biologic logic is straightforward even if the clinical proof remains narrow. Amyloid plaques are thought to act upstream of tau spread and neurodegeneration, and if a therapy meaningfully reduces amyloid in vulnerable cortical zones, the expectation is that downstream injury might ease as well. In this case, that expectation was visible in several modalities at once: pathology, positron emission tomography, and longitudinal magnetic resonance imaging.
Patchy Amyloid Removal at Autopsy
The patient received 30 doses of aducanumab, with a cumulative dose of 280 mg/kg, over 4.5 years. Four years after the final dose, he died, and autopsy revealed variable amyloid burden across brain regions rather than a simple global erase. Some areas showed very low amyloid levels, while adjacent areas retained the more typical distribution of amyloid in the deep cortical layers, with only low levels in the superficial layers.
The spatial pattern matters because amyloid plaques are not deposited evenly, and their clearance is unlikely to be uniform either. Gyral crests in the treated brain showed low amyloid levels that were preferentially located there when compared with untreated controls. In practical neuropathologic terms, this suggests that the antibody effect was not merely detectable on scan but also visible in the architecture of the tissue itself.
Less Tau Burden Followed the Amyloid Response
The strongest clinical relevance of this case lies in the coupling between amyloid reduction and tau pathology. Brain regions with lower amyloid after treatment were associated with less tau pathology at autopsy than corresponding regions in untreated controls. In Alzheimer disease, tau accumulation tracks more closely with cognitive and neurodegenerative injury than amyloid alone, so any signal linking amyloid clearance to reduced tau burden is biologically important.
A single sentence captures the central tension in the field: plaque removal is only useful if it changes the downstream disease process.
This case suggests it can, at least in a focal and regionally selective way. The data do not prove that amyloid removal reverses established tauopathy, and no one should read a single autopsy as a surrogate for population-wide benefit. Even so, the finding that the treated brain showed less tau in areas where amyloid had fallen is the kind of clinicopathologic correlation that amyloid-targeting therapy has long needed.
Imaging Mirrored the Tissue-Level Signal
The in vivo imaging findings reinforced the postmortem data. Amyloid and tau were assessed with positron emission tomography using standardized uptake value ratio, and cortical thickness was tracked longitudinally on magnetic resonance imaging. Treated regions with lower amyloid were linked to slower longitudinal atrophy on MRI, with β = -0.50, 95% CI, -0.62 to -0.37, and t = -7.96. The directional signal is clinically intuitive: if the cortical substrate is degenerating more slowly, then structural thinning should decelerate as well.
This imaging-pathology alignment gives the report its weight. A drug effect that appears only on amyloid PET but disappears at autopsy would be easy to dismiss as a biomarker artifact. Here, the pattern ran across modalities. Amyloid burden was low in selected cortical areas, tau pathology was reduced in those same regions, and cortical thickness declined more slowly where amyloid had been cleared.
What the TREM2 Variant Adds to the Interpretation
The participant carried the p.R47H TREM2 variant, a known risk factor for Alzheimer disease. That detail does not make the findings generalizable, but it does make them more interesting biologically. TREM2 is tied to microglial function, and microglia shape how the brain responds to amyloid deposition and clearance. A patient with altered TREM2-mediated immune signaling may not process amyloid in the same way as someone without the variant, and treatment response could therefore look different at the tissue level.
This is one reason clinicopathologic cases remain valuable. They let investigators examine how a therapeutic mechanism behaves in a brain with specific genetic vulnerability, rather than in an averaged cohort where region-specific effects can be blurred by heterogeneity. Here, the treated brain did not simply look “better” or “worse”; it showed a mosaic of amyloid clearance that could be read against tau and atrophy with unusual precision.
Why This Case Still Changes the Conversation
Amyloid therapies have often been judged by fluid or imaging biomarkers, then defended or criticized on the basis of cognitive outcomes that evolve slowly. This report moves the discussion back to tissue. If amyloid removal is uneven, if the deepest cortical layers retain more pathology than superficial layers, and if the regions with the least amyloid are also the regions with less tau and slower atrophy, then the therapy is interacting with disease biology in a meaningful way.
The findings also leave several hard questions untouched. A single patient cannot establish magnitude, durability, or class effect. The untreated controls help with context, but they do not replace a larger series that could test whether this pattern repeats across more brains, more genotypes, and more treatment durations. The report also cannot determine whether the observed slowing of cortical thinning translated into a clinically measurable cognitive difference over time.
Even so, the core message is concrete: in this patient, aducanumab was associated with patchy amyloid clearance at autopsy, less tau pathology in treated regions, and slower longitudinal atrophy on MRI. The next step is not speculation about whether amyloid matters; it is determining how often this tissue-level pattern appears, and in which patients, before amyloid clearance can be treated as a reliable proxy for downstream neuroprotection.