Neuropathological study of the effects of aducanumab anti-Aβ immunotherapy on patients with Alzheimer’s disease Authors: Lindsay A. Welikovitch, Derek H. Oakley, Rachel E. Bennett, Alberto Serrano-Pozo, Haoyang Zhu, Nancy E. Ruiz-Uribe, Theodore J. Zwang, Lori B. Chibnik, Teresa Gomez-Isla, Matthew P. Frosch, Donald S. Marks, Stephen Salloway, Charles Bernick, David Greeley, Caitlin S. Latimer, Amber Nolan, R. Matthew Hutchison, Carrie E. Rubel, Thierry Bussiere, Edward D. Plowey, C. Dirk Keene, Bradley T. Hyman Published: 2026 ABSTRACT Fluid and imaging biomarker data show that anti-amyloid (Aβ) antibodies promote the clearance of Aβ from the brains of patients with Alzheimer’s disease (AD). We examined postmortem brain tissue of individuals who participated in aducanumab clinical trials and investigated the drug’s effects on Aβ pathology and other AD phenotypes. The medial temporal lobe of six aducanumab clinical trial participants—who had extensive exposure to aducanumab and happened to die between 7 weeks and 5 years after their last antibody infusion—was compared with that of nine untreated AD patients matched for age, APOE genotype, and Braak neurofibrillary tangle stage, to determine how aducanumab impacts AD pathobiology. Patients treated with aducanumab displayed a robust reduction in Aβ burden. As observed in previous studies, Aβ was associated with non-arterial microvessels in aducanumab-treated patients, suggesting a redistribution of Aβ within the neuropil. Neuritic phospho-tau decreased in parallel with fewer Aβ plaques, but the density of PHF-1+ and AT8+ neurofibrillary tangles remained unchanged relative to the average untreated AD donor. Measures of microglial and astroglial reactivity were also comparable to those in untreated AD controls. These findings confirm aducanumab’s potent ability to target and remove brain Aβ. On average, Aβ plaques increased in proportion to the length of time after the last dose, consistent with the idea that plaques gradually redeposit post-treatment. The selective reduction of neuritic, but not neurofibrillary tangle phospho-tau implies that Aβ-targeted antibodies such as aducanumab alleviate plaque-associated dystrophy but may not address established tangles. This study describes the long-term outcomes of anti-Aβ immunotherapy in AD.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00401-026-03037-y. Introduction Aducanumab was the first disease-modifying monoclonal antibody approved for the treatment of Alzheimer’s disease (AD). Its accelerated approval by the U.S. Food and Drug Administration (FDA) under the trade name Aduhelm marked a pivotal shift in the field, spurring broader efforts to develop new biologics targeting amyloid-β (Aβ), intracellular pathological protein aggregates such as tau, and cell receptors implicated in neurodegeneration (Reviewed in [34],reviewed in [13]). Although aducanumab is no longer commercially available, its impact on AD drug development remains significant, with additional anti-Aβ drugs subsequently gaining regulatory approval worldwide [32, 36]. Early clinical trials offered promising evidence that aducanumab could meaningfully alter the AD course. In the Phase 1b PRIME study, the drug produced a dose-dependent reduction in brain Aβ, as measured by positron emission tomography (PET), with the strongest effects at the 10 mg/kg dose [12]. These biomarker changes were accompanied by indications of slowed cognitive decline on the Clinical Dementia Rating-Sum of Boxes (CDR-SB) and Mini-Mental State Examination (MMSE) [31]. These results led to two identical Phase 3 trials, ENGAGE and EMERGE, evaluating aducanumab’s efficacy in individuals with prodromal or early AD and confirmed Aβ pathology. Although both trials were initially halted after interim futility analyses suggested they would not meet their primary endpoints, subsequent analyses with extended follow-up data revealed that EMERGE, but not ENGAGE, achieved its primary endpoint at the highest dose, showing a statistically significant delay in cognitive deterioration relative to placebo [9, 22]. Despite the termination of the aducanumab program in 2024, continued study of these clinical trial cohorts provides an important opportunity to examine how anti-Aβ antibodies act in the human brain. Neuropathological assessments can link in-trial biomarker trajectories to underlying cellular and molecular changes, identify correlates of therapeutic response, and inform the design and clinical application of drugs with similar mechanisms of action. However, neuropathological data from treated individuals remain limited: to date, only eight autopsy cases have been reported, and just five of these include case-control analyses [7, 28, 38]. Here, we present a comprehensive neuropathological analysis of six additional aducanumab clinical trial participants who consented to autopsy and brain donation for research. These cases offer insights into the mechanisms and long-term consequences of aducanumab treatment, as well as the brain’s response following treatment discontinuation. Materials and methods Brain donors Neuropathological changes in the medial temporal lobe (MTL) and temporal neocortex (A20) of aducanumab-treated patients (n = 6) were compared with those of AD patients who had never received anti-Aβ medications (n = 9). Aducanumab brain specimens were obtained from the Massachusetts Alzheimer’s Disease Research Center (MADRC) Neuropathology Core Brain Bank (n = 3) and the University of Washington BioRepository and Integrated Neuropathology (BRaIN) Laboratory and Precision Neuropathology Core (n = 3). Untreated AD cases were selected from the MADRC Brain Bank to be matched by age, APOE genotype, and Braak neurofibrillary tangle stage (Tables 1, 2). All study participants or their next-of-kin provided informed consent for brain donation at their respective institutions. Tissue was formalin-fixed and paraffin-embedded (FFPE). Table 1Baseline characteristics of study subjectsUntreated AD (n = 9)Aducanumab AD (n = 6)Age, years (mean ± SD)76.3 ± 9.275.0 ± 8.5Sex, n (%) female6 (66.7)1 (16.7)APOE4 carriers, n (%)5 (62.5)3 (50.0)APOE4 alleles, n (%)5 (31.3)4 (33.3)Post-mortem interval, hours (mean ± SD)17.3 (12.9)35.3 (27.9)AD Alzheimer’s diseaseTable 2Age, sex, APOE genotype, and postmortem neuropathological evaluation of study subjectsGroup(Subject label)Age at death (years)SexAPOEAβ Thal phaseBraak NFT stageCERAD neuritic plaque scoreLATE-NC stageCAA scoreLewy body pathologyOther neuropathological diagnosesAdu 182M3/42VI113–Cerebrovascular diseaseAging-related tau astrogliopathyAdu 268M3/34VI322––Adu 387F3/32V111––Adu 478M4/45VI323–Mild arteriosclerosisAdu 567M3/45VI301Amygdala predominantModerate arteriosclerosisAdu 668M3/32VI302Limbic/transitionalModerate arteriosclerosisAD 162M3/35VI301––AD 275F3/35VI201–ArteriosclerosisAD 376F3/45VI302–ArteriosclerosisAD 482M3/45VI303–Small vessel arteriosclerosisAcute hypoxiaAD 565Fn/a5VI301–ArteriosclerosisAD 685F3/45VI213––AD 771F3/44VI300––AD 890M3/35VI221Amygdala predominant–AD 981F3/45VI222–Adu Aducanumab, AD Alzheimer’s disease, M male, F female, Aβ amyloid-β, NFT neurofibrillary tangles, LATE-NC limbic-predominant age-related TDP-43 encephalopathy neuropathologic change, CAA cerebral amyloid angiopathy, n/a = not available; − = absent Table 1Baseline characteristics of study subjectsUntreated AD (n = 9)Aducanumab AD (n = 6)Age, years (mean ± SD)76.3 ± 9.275.0 ± 8.5Sex, n (%) female6 (66.7)1 (16.7)APOE4 carriers, n (%)5 (62.5)3 (50.0)APOE4 alleles, n (%)5 (31.3)4 (33.3)Post-mortem interval, hours (mean ± SD)17.3 (12.9)35.3 (27.9)AD Alzheimer’s disease Baseline characteristics of study subjects AD Alzheimer’s disease Table 2Age, sex, APOE genotype, and postmortem neuropathological evaluation of study subjectsGroup(Subject label)Age at death (years)SexAPOEAβ Thal phaseBraak NFT stageCERAD neuritic plaque scoreLATE-NC stageCAA scoreLewy body pathologyOther neuropathological diagnosesAdu 182M3/42VI113–Cerebrovascular diseaseAging-related tau astrogliopathyAdu 268M3/34VI322––Adu 387F3/32V111––Adu 478M4/45VI323–Mild arteriosclerosisAdu 567M3/45VI301Amygdala predominantModerate arteriosclerosisAdu 668M3/32VI302Limbic/transitionalModerate arteriosclerosisAD 162M3/35VI301––AD 275F3/35VI201–ArteriosclerosisAD 376F3/45VI302–ArteriosclerosisAD 482M3/45VI303–Small vessel arteriosclerosisAcute hypoxiaAD 565Fn/a5VI301–ArteriosclerosisAD 685F3/45VI213––AD 771F3/44VI300––AD 890M3/35VI221Amygdala predominant–AD 981F3/45VI222–Adu Aducanumab, AD Alzheimer’s disease, M male, F female, Aβ amyloid-β, NFT neurofibrillary tangles, LATE-NC limbic-predominant age-related TDP-43 encephalopathy neuropathologic change, CAA cerebral amyloid angiopathy, n/a = not available; − = absent Age, sex, APOE genotype, and postmortem neuropathological evaluation of study subjects Cerebrovascular disease Aging-related tau astrogliopathy Small vessel arteriosclerosis Acute hypoxia Adu Aducanumab, AD Alzheimer’s disease, M male, F female, Aβ amyloid-β, NFT neurofibrillary tangles, LATE-NC limbic-predominant age-related TDP-43 encephalopathy neuropathologic change, CAA cerebral amyloid angiopathy, n/a = not available; − = absent APOE genotyping APOE genotyping of donors from the MADRC Brain Bank was performed via restricted fragment length polymorphism or commercially available Taqman assays (ThermoFisher Scientific, United States, cat#4,351,379, assay IDs: C___3084793_20 for rs429358 and C____904973_10 for rs7412) at Massachusetts General Hospital or was received from the National Centralized Repository for Alzheimer’s Disease and Related Dementias (NCRAD). APOE genotype was unavailable for one untreated AD donor. Immunohistochemistry and Prussian blue staining Histological analyses were performed using tissue from the right hemisphere of patients Adu 1, 3, and 4, and the left hemisphere of patients Adu 2, 5, and 6. FFPE sections were deparaffinized in xylene and rehydrated through a graded ethanol series. Heat-mediated antigen retrieval was performed in citrate buffer (pH6.0; 10 mM citric acid, 0.05% Tween-20) at 95 °C for 20 min in a microwave oven, followed by a 15 min cooling period at 4 °C. Tissue was rinsed in water and blocked in 5% bovine serum albumin (BSA) in Tris buffered saline (TBS) with 0.25% Tween-20 (TBS-T) for 1 h at room temperature. Sections were incubated overnight at 4 °C with primary antibodies (Supplemental Table 1), rinsed in large volumes of TBS, and incubated with fluorophore-conjugated secondary antibodies or streptavidin for 1 h at room temperature. Nuclei were stained with 4′6-diamidino-2-phenylindole (DAPI) for 5 min. Tissue autofluorescence was reduced using TrueBlack Autofluorescence Quencher (1x; Biotium, United States) before coverslipping with Fluoromount G mounting medium (SouthernBiotech, United States). Thioflavin S (ThioS) staining was performed after immunolabeling. Slides were rinsed in phosphate buffered saline (PBS) for 2 min, incubated in 0.05% ThioS solution in 50% ethanol for 8 min in the dark, transferred to 80% ethanol for approximately 15 s, and washed in water for 2 min. For immunolabeling of synaptic puncta, ethanol pre-treatment was omitted and the processed tissue was mounted onto charged glass slides and coverslipped with ProLong Glass Antifade mounting medium (Cat. No. P36980; Invitrogen) using a 170 ± 5 μm No. 1.5H high-precision coverglass for super-resolution imaging. Iron stain solution was prepared using equal parts potassium ferrocyanide solution (Abcam, United Kingdom) and 2% hydrochloric acid. Sections were stained for 3 min, rinsed, counterstained with nuclear fast red (Abcam) for 5 min, dehydrated, and coverslipped with Eukitt Quick-hardening mounting medium (Millipore-Sigma, United States). Image acquisition Whole-section widefield fluorescent images were acquired at 20 × magnification using an Olympus VS120 Virtual Slide Scanner (Japan). Brightfield images were acquired at 40 × using a Hamamatsu NanoZoomer (Japan). Acquisition settings were held constant for all images being directly compared by quantitative analysis. Pre- and post-synaptic terminals immunolabeled with anti-Bassoon and -Homer1, respectively, were captured in CA1 using the Zeiss Elyra 7 super-resolution microscope with Lattice SIM2. Z-stacks, 2.27 μm thick, were imaged at a step size of 0.126 μm using a 63 × oil objective. All images were processed using SIM2 ‘Low contrast’ reconstruction settings. Image analysis Brain regions were manually segmented, and fluorescence thresholding was performed using QuPath [2]. SMI312+ axonal dystrophies within 10 μm of Aβ+ThioS+ plaques were manually counted and normalized by plaque size. Because dense-cored ThioS+ plaques are concentrated in the hippocampus, unlike the presubiculum and entorhinal cortex, which typically harbor more diffuse ThioS− deposits that are difficult to reliably delineate, this analysis was restricted to hippocampal subregions. Aβ+ and ThioS+ deposits overlapping at least partially with GLUT1+ endothelial cells were classified as vascular. All vascular Aβ+, ThioS+, and hemosiderin deposits were manually identified, outlined, and measured; vessel diameters were also manually recorded. Microglial recruitment towards Aβ plaques and astrocyte coverage of blood vessels were assessed by delineating Aβ plaques and GLUT1+ blood vessels using fluorescence thresholding and expanding these masks radially by 10 µm. Neurofibrillary tangle density was quantified in the hippocampus using Ilastik’s pixel and object classifier workflow [3]. The pixel classifier was trained using a digital paintbrush to distinguish positive signal from background, producing a probability map. This map was used in the object classifier, which was trained to differentiate neurofibrillary tangles from dystrophic neurites based on morphological features. Classifiers were then applied to all images. The total number of Bassoon and Homer1 puncta was determined by thresholding using Imaris Software (Oxford Instruments) and normalized per millimeter cubed of tissue. Statistical analysis All statistical tests were performed in GraphPad Prism 10 (Dotmatics, United States). Nonparametric pairwise comparisons were conducted using the Mann–Whitney U test. Correlations were evaluated with the Spearman’s rank test. Box-and-whisker plots show Tukey distributions; scatterplots show individual data points. Data are presented in box-and-whisker graphs using the Tukey method. Correlations were carried out using the Spearman’s rank test. Data are presented in point graphs. Outliers were identified using Grubb’s test (p < 0.05) and excluded. Sample size (n) represents the number of subjects analyzed in each analysis and is represented in each figure. Significance thresholds were p < 0.05, with ns = p > 0.05; * = p < 0.05; ** = p ≤ 0.01; *** = p ≤ 0.001. Results Subject characteristics and clinical trial data Tables 1 and 2 summarize the demographic and neuropathological characteristics of six aducanumab clinical trial participants and nine untreated AD patients. Groups were matched for age, APOE genotype, and Braak neurofibrillary tangle stage. CAA severity was graded according to the scheme established by Love et al. [21], where 0 = absent, 1 = mild, 2 = moderate, and 3 = severe. All clinical trial subjects received the maximum dose of 10 mg/kg. Baseline Aβ PET Standardized Uptake Value ratio (SUVr) and Centiloid values, baseline MMSE and CDR-SB scores, trial participation, total number of infusions, cumulative dose, time between last dose and death, and clinical trajectories are provided in Table 3. Data from EMERGE and ENGAGE phase III trials show that 48% and 31% of patients treated with 10 mg/kg aducanumab, respectively, reached a PET composite SUVr of ≤ 1.10 after 78 weeks of treatment—approximately 18 infusions—a proposed threshold distinguishing Aβ-negative from Aβ-positive patients [9]. At the group level, patients enrolled in EMERGE and ENGAGE demonstrated a 71% and 59% in PET Aβ burden compared with baseline, respectively. All six immunotherapy patients studied here received between 19 and 76 monthly infusions, a treatment duration previously shown to produce substantial reductions in Aβ burden [9, 31]. The interval between final infusion and death ranged from 7.86 weeks to 4.97 years, enabling direct histological examination of both the consequences of long-term treatment and the sequelae following treatment discontinuation. Asymptomatic amyloid-related imaging abnormalities-hemosiderin (ARIA-H) and -edema (ARIA-E) were reported in four individuals during active treatment (Table 4). Cause of death for each participant was unrelated to treatment. Table 3Baseline Aβ PET and clinical trial participation and data of study subjectsSubject labelBaseline Aβ PET SUVrBaseline Aβ CentiloidsBaseline MMSE scoreBaseline CDR-SB scorePRIMEEMERGEENGAGEEMBARKTotal number of infusionsCumulative dose (mg/kg)Time since last dose (years)Time between baseline and most-recent cognitive test (years)Change in MMSE score from baselineChange in CDR-SB score from baselineAdu 11.2452.45301.5XX766540.154.27− 2.0 + 1.5Adu 21.4192.21253.5X191504.793.81− 9.0 + 6.0Adu 31.45100.85264XX463661.002.44− 3.0 + 3.0Adu 41.5982.76253X211564.973.81− 7.0 + 2.5Adu 51.49109.33243.5XX191102.512.04− 7.00Adu 61.73165.64283.5XX484003.123.81− 2.0 + 2.5Adu Aducanumab, Aβ amyloid-β, PET positron emission tomography, SUVr Standardized Uptake Value ratio, MMSE Mini-Mental State Exam, CDR-SB Clinical Dementia Rating Scale-Sum of BoxesTable 4Amyloid-related imaging abnormalities (ARIA) in aducanumab clinical trial subjectsGroup(Subject label)APOEStudyTrial phaseType of adverse eventsNumber of ARIA eventsAffected brain regionsTime between most-recent adverse event and autopsy (years)Adu 13/4PRIMELTEARIA-H1n/a5.15Adu 23/3ENGAGELTEARIA-H/E2Right parietal5.28Adu 33/3––none––Adu 44/4EMERGELTEARIA-H/E2Right frontalLeft occipitalLeft parietalRight parietalLeft temporal, non-hippocampalRight temporal, non-hippocampal4.79Adu 53/4EMBARKCoreARIA-H/E1Right frontalLeft occipitalRight occipital2.22Adu 63/3––none––Adu Aducanumab, LTE long-term extension, ARIA-H ARIA-hemosiderin, ARIA-E ARIA-edema, n/a not available Table 3Baseline Aβ PET and clinical trial participation and data of study subjectsSubject labelBaseline Aβ PET SUVrBaseline Aβ CentiloidsBaseline MMSE scoreBaseline CDR-SB scorePRIMEEMERGEENGAGEEMBARKTotal number of infusionsCumulative dose (mg/kg)Time since last dose (years)Time between baseline and most-recent cognitive test (years)Change in MMSE score from baselineChange in CDR-SB score from baselineAdu 11.2452.45301.5XX766540.154.27− 2.0 + 1.5Adu 21.4192.21253.5X191504.793.81− 9.0 + 6.0Adu 31.45100.85264XX463661.002.44− 3.0 + 3.0Adu 41.5982.76253X211564.973.81− 7.0 + 2.5Adu 51.49109.33243.5XX191102.512.04− 7.00Adu 61.73165.64283.5XX484003.123.81− 2.0 + 2.5Adu Aducanumab, Aβ amyloid-β, PET positron emission tomography, SUVr Standardized Uptake Value ratio, MMSE Mini-Mental State Exam, CDR-SB Clinical Dementia Rating Scale-Sum of Boxes Baseline Aβ PET and clinical trial participation and data of study subjects Adu Aducanumab, Aβ amyloid-β, PET positron emission tomography, SUVr Standardized Uptake Value ratio, MMSE Mini-Mental State Exam, CDR-SB Clinical Dementia Rating Scale-Sum of Boxes Table 4Amyloid-related imaging abnormalities (ARIA) in aducanumab clinical trial subjectsGroup(Subject label)APOEStudyTrial phaseType of adverse eventsNumber of ARIA eventsAffected brain regionsTime between most-recent adverse event and autopsy (years)Adu 13/4PRIMELTEARIA-H1n/a5.15Adu 23/3ENGAGELTEARIA-H/E2Right parietal5.28Adu 33/3––none––Adu 44/4EMERGELTEARIA-H/E2Right frontalLeft occipitalLeft parietalRight parietalLeft temporal, non-hippocampalRight temporal, non-hippocampal4.79Adu 53/4EMBARKCoreARIA-H/E1Right frontalLeft occipitalRight occipital2.22Adu 63/3––none––Adu Aducanumab, LTE long-term extension, ARIA-H ARIA-hemosiderin, ARIA-E ARIA-edema, n/a not available Amyloid-related imaging abnormalities (ARIA) in aducanumab clinical trial subjects Right frontal Left occipital Left parietal Right parietal Left temporal, non-hippocampal Right temporal, non-hippocampal Right frontal Left occipital Right occipital Adu Aducanumab, LTE long-term extension, ARIA-H ARIA-hemosiderin, ARIA-E ARIA-edema, n/a not available Aducanumab-treated patients exhibit lower Aβ burden, with a positive correlation between withdrawal duration and Aβ load Chronic aducanumab administration improves Aβ fluid and imaging biomarkers within 18 months of treatment initiation [9, 31]. To evaluate these effects histologically, we performed Aβ immunostaining in the MTL of six aducanumab clinical trial participants (Fig. 1a; Supplemental Figs. 1, 2). Consistent with prior PET studies, we observed a pronounced reduction in Aβ-immunopositive material in this brain region, with all treated subjects showing markedly lower Aβ levels than the untreated AD group average (Fig. 1b).Fig. 1Low-magnification micrographs illustrate a spectrum of Aβ pathology in the MTL revealed by Aβ immunolabeling (red). Images show an untreated AD case representing the “natural” AD neuropathological trajectory in the absence of anti-Aβ therapy (left), an aducanumab-treated subject with modest Aβ plaque deposition (~ 50% of the average untreated burden; Aducanumab 6, center), and an aducanumab-treated subject with almost no detectable Aβ plaques (~ 1% of the average untreated burden; Aducanumab 1, right). Micrographs were generated from fluorescent widefield images using an inverse background setting (scale bar = 1 mm). b, c Quantification of insoluble Aβ in the MTL revealed a significant reduction in total Aβ deposits in immunotherapy patients, as detected using antibodies D54D2 (p = 0.002) and 4G8 (p < 0.001). d Aβ area fraction, measured using D54D2 immunolabeling, was strongly associated with time since last aducanumab dose (r = 0.829; p = 0.03), but not with cumulative exposure (r = − 0.371; p = 0.25) Low-magnification micrographs illustrate a spectrum of Aβ pathology in the MTL revealed by Aβ immunolabeling (red). Images show an untreated AD case representing the “natural” AD neuropathological trajectory in the absence of anti-Aβ therapy (left), an aducanumab-treated subject with modest Aβ plaque deposition (~ 50% of the average untreated burden; Aducanumab 6, center), and an aducanumab-treated subject with almost no detectable Aβ plaques (~ 1% of the average untreated burden; Aducanumab 1, right). Micrographs were generated from fluorescent widefield images using an inverse background setting (scale bar = 1 mm). b, c Quantification of insoluble Aβ in the MTL revealed a significant reduction in total Aβ deposits in immunotherapy patients, as detected using antibodies D54D2 (p = 0.002) and 4G8 (p < 0.001). d Aβ area fraction, measured using D54D2 immunolabeling, was strongly associated with time since last aducanumab dose (r = 0.829; p = 0.03), but not with cumulative exposure (r = − 0.371; p = 0.25) The Aβ antibody used in this initial analysis (clone D54D2) recognizes an N-terminal epitope overlapping with aducanumab’s binding site, which could potentially interfere with immunohistochemical labeling. To rule out this possibility, we applied two fluorophore-conjugated anti-human IgG antibodies to detect residual aducanumab bound to Aβ plaques. No IgG+ plaques were identified in any brain donor, including one patient who received their final dose just 7 weeks before death, indicating rapid antibody washout (Supplemental Fig. 3). Findings were confirmed using antibody clone 4G8, which recognizes the mid-segment (aa17–24) of the Aβ peptide (Fig. 1c), as well as antibodies targeting the C-termini of Aβ40 and Aβ42 (Supplemental Fig. 4a)—outside aducanumab’s binding domain—all of which confirmed lower Aβ deposition. To assess changes in Aβ plaque structure and associated neuritic pathological changes, we quantified Aβ+ThioS+ dense-core plaques and SMI312+ dystrophic neurites. Although overall Aβ burden was considerably lower in treated individuals, the proportion of ThioS+ Aβ plaques and the number of plaque-associated dystrophic neurites were comparable between groups, suggesting that Aβ plaques may retain similar biochemical and neuropathological characteristics following treatment (Supplemental Fig. 4b). To examine dose- and time-dependent responses to aducanumab, we correlated postmortem Aβ levels with total drug received and time since last administration (Fig. 1d). No clear association was found between cumulative dose and Aβ burden (Supplemental Table 2). By contrast, a statistically significant positive correlation emerged between withdrawal duration and Aβ burden, indicating that longer treatment gaps are associated with greater Aβ load (Fig. 1d; Supplemental Table 2). Notably, even individuals with treatment gaps of approximately 4–5 years had considerably less Aβ than the average untreated AD donor. This effect was not attributable to brain-antibody retention, as all plaques were IgG− (Supplemental Fig. 3). We also confirmed these findings in Brodmann area 20 (BA20), which likewise exhibited reduced Aβ deposition in immunotherapy patients compared with untreated AD controls (Supplemental Fig. 4c). Baseline, pre-treatment Aβ Centiloid values did not correlate with postmortem Aβ levels in either BA20 or the MTL, indicating that Aβ burden at treatment initiation, as measured by PET, was likely reduced by therapeutic intervention (Supplemental Fig. 4d). Aβ plaques are broadly categorized as diffuse or fibrillary/dense core based on ultrastructural and tinctorial characteristics. Given that aducanumab binds both oligomeric and fibrillar Aβ [14, 31], we asked whether specific plaque subtypes are more effectively impacted by immunotherapy. We quantified Aβ in the presubiculum, where diffuse ‘lake-like’ deposits predominate, and in the hippocampus proper—hereafter referred to as ‘hippocampus’—where cored and diffuse dense plaques accumulate (Supplemental Fig. 5a) [35]. Relative to untreated AD donors, immunotherapy patients exhibited 87% and 50% lower Aβ in the presubiculum and hippocampus, respectively (Supplemental Fig. 5b). A similar regional disparity was observed using the 4G8 antibody (Supplemental Fig. 5c). Neuritic tau is reduced in association with Aβ burden, but neurofibrillary tangles remain Antemortem biomarker studies show that Aβ removal by aducanumab corresponds with reductions in cerebrospinal fluid (CSF) phosphorylated tau (pTau) levels and improvements in tau PET [9]. To investigate whether analogous neuropathological changes occur in aducanumab-treated patients, we quantified several forms of pathological tau, including somatic tangles, neuropil threads, and plaque-associated neuritic dystrophies (Fig. 2a, b).Fig. 2Immunolabeling of pTau217 (green), PHF-1 (magenta), and AT8 (cyan) revealed abundant neurofibrillary tangles, neuropil threads, and Aβ plaque-associated neuritic dystrophies in untreated AD subjects, as well as in immunotherapy patients with Aβ plaque pathology (yellow) (scale bars = 100 µm in A; 200 µm in B). C Overall pTau217 burden did not differ between treatment groups (p = 0.12), whereas PHF-1 and AT8 area fractions were significantly lower following aducanumab treatment (PHF-1: p = 0.01; AT8: p = 0.03). D In contrast, the numbers of PHF-1+ and AT8+ neurofibrillary tangles were comparable between untreated and treated AD subjects (PHF-1: p = 0.34; AT8: p = 0.17), indicating that decreased total tau burden primarily reflects changes in plaque-associated neuritic pathology. E Super resolution imaging of Bassoon+ and Homer1+ pre- and post-synaptic terminals respectively in region CA1 of untreated and aducanumab-treated patients revealed no change in synapse abundance (Bassoon: p = 0.26; Homer1: p = 0.50) (scale bar = 50 µm) Immunolabeling of pTau217 (green), PHF-1 (magenta), and AT8 (cyan) revealed abundant neurofibrillary tangles, neuropil threads, and Aβ plaque-associated neuritic dystrophies in untreated AD subjects, as well as in immunotherapy patients with Aβ plaque pathology (yellow) (scale bars = 100 µm in A; 200 µm in B). C Overall pTau217 burden did not differ between treatment groups (p = 0.12), whereas PHF-1 and AT8 area fractions were significantly lower following aducanumab treatment (PHF-1: p = 0.01; AT8: p = 0.03). D In contrast, the numbers of PHF-1+ and AT8+ neurofibrillary tangles were comparable between untreated and treated AD subjects (PHF-1: p = 0.34; AT8: p = 0.17), indicating that decreased total tau burden primarily reflects changes in plaque-associated neuritic pathology. E Super resolution imaging of Bassoon+ and Homer1+ pre- and post-synaptic terminals respectively in region CA1 of untreated and aducanumab-treated patients revealed no change in synapse abundance (Bassoon: p = 0.26; Homer1: p = 0.50) (scale bar = 50 µm) Total pTau-217 did not differ between groups (Fig. 2c). However, total PHF-1 and AT8 percent area were markedly lower in treated individuals relative to untreated AD controls despite equivalent Braak stages (Fig. 2c). We also confirmed these findings in BA20, which likewise exhibited reduced AT8 percent area in immunotherapy patients compared with untreated AD controls (Supplemental Fig. 6). We considered whether resolution of dystrophic neurites and/or neurofibrillary tangles could account for this change. Tangle counts in the hippocampus were similar between groups (Fig. 2D), and no clusters of pTau+ dystrophic neurites were detected outside of clear association with an Aβ plaque core (Fig. 2a, b). Comparable results were obtained using a non-phospho-directed tau antibody (Supplemental Table 1). Post hoc power analysis indicated that, with the current sample sizes, we would have had 90% power to detect a 75% change in tangle number, and 60% power to detect a 50% change, assuming a one-sided test at α = 0.05. The observed group difference was only ~ 15%, suggesting that a biologically meaningful effect was unlikely to have been missed due to insufficient power. Together, these findings indicate that mature neurofibrillary tangles are largely stable and may be resistant to Aβ-directed therapy, whereas plaque-associated neuritic tau and neuropil threads were reduced in parallel with lower parenchymal Aβ burden. We also applied super-resolution microscopy to quantify pre- and post-synaptic boutons in region CA1 (Fig. 2e). Bassoon+ and Homer1+ pre- and post-synaptic densities were comparable between aducanumab-treated and untreated AD patients. Aducanumab treatment is associated with increased Aβ on cortical microvessels Active immunization with AN1792 has been associated with a shift in Aβ accumulation from plaques to microvessels [6, 29]. To evaluate whether this also occurs with passive immunization, we quantified vessel-associated Aβ in available tissue samples. Leptomeningeal vessels were excluded from analyses to mitigate potential artifacts from tissue sectioning. The incidence of Aβ+ and ThioS+ deposits colocalizing with GLUT1+ endothelial cells was almost threefold greater in aducanumab-treated individuals compared with untreated AD patients (Fig. 3a, b). While the vast majority of Aβ (mean 93.5%) was concentrated in parenchymal plaques in untreated AD donors, up to 75% of Aβ was vessel-associated in immunotherapy patients (Fig. 3c). Vessel-type analysis showed an almost fivefold increase in the average number of Aβ+ capillaries (≤ 10 μm diameter) in clinical trial subjects, with no corresponding change in larger vessels (Fig. 3d). Capillary Aβ was also detected with the 4G8 antibody (Supplemental Fig. 7a), but, notably, not with Aβ1-40- or Aβ1-42-directed antibodies. The proportion of Aβ40/42 deposited on larger vessels was similar between treatment groups (Supplemental Fig. 3e).Fig. 3b GLUT1+ blood vessels (cyan) exhibited substantially more Aβ+ and ThioS+ accumulation (yellow) in the MTL of aducanumab-treated vs. untreated AD patients (Aβ: p = 0.01; ThioS: p = 0.03). Arrows denote sites of vascular Aβ deposition (scale bar = 50 µm). c Despite an overall decrease in parenchymal Aβ following immunotherapy, the percentage of total Aβ localized to blood vessels was nearly fourfold higher in treated participants (p = 0.02). d This shift was driven by an increase in the number of affected capillaries (p = 0.04), whereas the frequency of affected vessels larger than 10 μm did not differ between groups (p = 0.33). e The ratio of Aβ40/42 deposited on larger vessels was similar between treatment groups b GLUT1+ blood vessels (cyan) exhibited substantially more Aβ+ and ThioS+ accumulation (yellow) in the MTL of aducanumab-treated vs. untreated AD patients (Aβ: p = 0.01; ThioS: p = 0.03). Arrows denote sites of vascular Aβ deposition (scale bar = 50 µm). c Despite an overall decrease in parenchymal Aβ following immunotherapy, the percentage of total Aβ localized to blood vessels was nearly fourfold higher in treated participants (p = 0.02). d This shift was driven by an increase in the number of affected capillaries (p = 0.04), whereas the frequency of affected vessels larger than 10 μm did not differ between groups (p = 0.33). e The ratio of Aβ40/42 deposited on larger vessels was similar between treatment groups Although aducanumab-treated individuals showed considerably lower overall cerebral Aβ burden, a disproportionate fraction of Aβ pathology was localized to the brain vasculature following immunotherapy. Importantly, this change in distribution appears to be a lasting and/or progressive phenomenon, as most subjects had discontinued treatment years before autopsy. One subject whose last dose was only 7 weeks prior to death had a similar pattern of loss of Aβ40/42 on capillaries. A trend toward a significant positive correlation between time since last dose and vascular Aβ was observed (Supplemental Fig. 7a), though this fell just below the threshold for significance, suggesting that redistribution of Aβ toward cerebral blood vessels may occur during treatment and persist beyond the treatment period. Whole-slide assessment of the MTL revealed a pronounced negative spatial correlation between GLUT1+-Aβ and areas of persistent parenchymal Aβ plaques: vascular Aβ was most prevalent in regions where plaque burden was minimal, as observed in the four cases with detectable plaque pathology (Supplemental Figs. 8–11). No such relationship was found in untreated AD donors (Supplemental Figs. 12–14). Local extravasation of albumin, fibrinogen, and human IgG in the MTL lobe showed no treatment-dependent alterations. Prussian blue staining for hemosiderin deposits likewise revealed no treatment-related differences in microhemorrhage frequency (Supplemental Fig. 7c), consistent with the absence of ARIA detected by MRI in the MTL of any analyzed subject during the study period (Table 4). Microglia and astrocytes do not show overt reactivity after immunotherapy Anti-Aβ antibodies with full effector function activate microglial Fc receptors to trigger phagocytosis and cytokine production. In a recent study using chimeric murine aducanumab in a transgenic mouse model harboring both Aβ and tau neuropathologies, microglia remained in proximity to plaques at least one week after drug washout [39], accompanied by pronounced astrocytic GFAP expression in both plaque-adjacent and more distal cortical regions. To determine whether analogous cellular responses persist in patients following aducanumab treatment and discontinuation, we quantified IBA1+ microglia and GFAP+ astrocytes in the MTL. Despite reduced Aβ burden in aducanumab-treated cases, there was only a nonsignificant trend towards increased IBA1 coverage (Fig. 4a, b). Though only sparse plaques were visible in some immunotherapy patients (Supplemental Fig. 15), microglial association with plaques was comparable to that observed in the untreated AD group (Fig. 4c). A modest trend towards expanded GFAP expression was similarly observed, but did not reach statistical significance (Fig. 4d, e). These results may reflect variability in treatment duration, withdrawal interval, and limited sample size. Given the increase in vascular Aβ+ and ThioS+ deposits, we also assessed perivascular astrogliosis by quantifying vessel-associated GFAP expression [18]. No treatment-dependent differences were detected (Fig. 4f).Fig. 4b Total microglial IBA1+ area fraction (green) showed no significant treatment-related differences (p = 0.06). C IBA1 coverage of Aβ plaques (magenta), reflecting plaque-associated microglial recruitment, was also comparable between treatment groups, potentially due to drug washout and prolonged intervals between last dose and death in most patients (p = 0.38). D, E Total astrocytic GFAP+ area fraction (cyan) was largely unchanged in aducanumab-treated cases (p = 0.09). F Vascular astrogliosis was not observed (p = 0.19) (scale bar = 100 µm) b Total microglial IBA1+ area fraction (green) showed no significant treatment-related differences (p = 0.06). C IBA1 coverage of Aβ plaques (magenta), reflecting plaque-associated microglial recruitment, was also comparable between treatment groups, potentially due to drug washout and prolonged intervals between last dose and death in most patients (p = 0.38). D, E Total astrocytic GFAP+ area fraction (cyan) was largely unchanged in aducanumab-treated cases (p = 0.09). F Vascular astrogliosis was not observed (p = 0.19) (scale bar = 100 µm) Discussion This postmortem neuropathological assessment of individuals who received aducanumab offers insight into the drug’s enduring impact on the AD brain. Our histopathological findings corroborate and extend CSF- and imaging-based studies demonstrating that aducanumab is associated with substantially lower Aβ burden, even in regions typically enriched with dense-core plaques, such as the hippocampus. High-dose exposure was associated with significantly lower Aβ levels relative to untreated AD subjects, regardless of the interval between the last infusion and death. In several cases, Aβ deposits were nearly absent, supporting the drug’s potential to achieve profound Aβ plaque reduction under chronic dosing regimens. The strong positive correlation between time since last dose and Aβ burden allowed us to model a hypothetical rate of Aβ plaque re-emergence following treatment withdrawal. Assuming that all study subjects may have been near Aβ-negative at the end of trial participation—an assumption that cannot be confirmed given the absence of serial PET imaging—regression of postmortem Aβ burden against time since last dose yields an estimated rate of increase of 0.267% Aβ coverage per year (95% CI 0.116–0.418, p = 0.006). Using the untreated AD group’s average percent area fraction as a putative baseline (2.71% area), we estimate that Aβ plaque burden in aducanumab-treated patients would approach levels comparable to untreated AD donors after approximately 10.2 years—assuming a linear trajectory, and with the caveat that this calculation is based on a sample of six individuals. These timelines align with estimates derived from Aβ PET studies of lecanemab clinical trial participants [24] (16–20 years) and carry important implications for designing intermittent or maintenance dosing strategies aimed at suppressing Aβ pathology while minimizing long-term treatment risks. Aducanumab’s affinity for both soluble and fibrillar Aβ likely enables broad depletion of multipole plaque subtypes [14, 31]. However, regional analyses uncovered differences in efficacy. This disparity could reflect either reduced epitope accessibility in compact Aβ plaques or faster deposition of dense deposits after treatment ends. Regression analyses favored the latter explanation: we observed a steeper association between withdrawal duration and Aβ load in the hippocampus (0.216% Aβ/year) compared to that in the presubiculum (0.059% Aβ/year,Supplemental Fig. 5d), consistent with known regional differences in Aβ accumulation rates, and the known early accumulation of hippocampal plaques in young individuals with Down syndrome [19, 23]. These findings suggest that plaque structure may influence both therapeutic response and re-aggregation dynamics following treatment cessation. Notably, the temporal and regional pattern of Aβ burden across withdrawal timepoints is reminiscent of the stereotypical staging described by Thal et al. [35]. Chronic anti-Aβ immunotherapy was also associated with less PHF-1+ and AT8+ tau, predominantly in dystrophic neurites and neuropil threads. Given that neuritic dystrophies arise partly from Aβ-driven toxicity, it is reasonable to assume that their reduction may occur downstream of lower parenchymal Aβ burden. Neurofibrillary tangles, by contrast, remained unchanged, consistent with the view that, once formed, tangles are largely irreversible and unresponsive to anti-Aβ interventions alone. This selective reduction of pathological tau in neural processes but not cell bodies parallels observations in individuals treated with aducanumab [28], lecanemab [11], and AN1792 [4, 5, 26, 27, 30], as well as PET and CSF data showing stabilization or modest declines in tau pathology during aducanumab therapy [9]. Animal studies similarly demonstrate resolution of neuritic dystrophies following Aβ clearance [10, 39]. Collectively, these findings suggest that somatic and neuritic tau differ in their stability and degradation pathways, and that Aβ-targeted therapies may be most effective for preventing or reversing plaque-associated neuritic tau, while additional strategies will likely be required to address mature tangles. We speculate that the reduction of neuritic dystrophies may contribute, in part, to the clinical benefits observed with anti-Aβ antibodies [8, 9, 15, 20]. Regional and global plaque clearance were associated with elevated vascular Aβ, mirroring observations from the AN1792 active immunization trial [6, 25, 29]. In our cohort, vascular Aβ accumulation was evident even in patients whose treatment had ended years before death, and trended positively with longer withdrawal periods. Although no evidence of generalized blood-brain-barrier dysfunction or inflammatory vascular gliosis was found in the MTL, these data raise the possibility that antibody-mediated dispersion of Aβ toward the vasculature lays the groundwork for ARIA in vulnerable brain regions, particularly in patients with predisposing genetic or vascular risk factors. The spatial relationship between parenchymal Aβ plaque removal and vascular Aβ accumulation—documented in AN1792-treated patients [6, 25, 27], reviewed in [4, 5] and in case reports of ARIA colocalizing with regions of reduced Aβ-PET signal following bapineuzumab and gantenerumab treatment [16, 33]—further supports this interpretation. These observations provide in situ evidence linking antibody-mediated clearance of parenchymal plaques with microvascular Aβ accumulation, at least in this patient subset. Future work should evaluate whether specific dosing regimens or combination therapies could mitigate this effect while preserving Aβ clearance efficacy. Microglia and astrocytes are key mediators of antibody-induced Aβ clearance [1, 37] and have been implicated in ARIA pathogenesis. Preclinical models show that aducanumab activates microglia via Fc receptor signaling, leading to transient inflammatory responses [17, 39]. In our postmortem analysis—conducted in the absence of acute inflammation from microhemorrhage or other brain injury—we found no evidence of sustained gliosis in the MTL, even in individuals treated at high doses or who died shortly after their last infusion. Similarly, vessel-associated gliosis was not elevated despite prominent vascular Aβ deposition. These findings, consistent with another recently reported series of aducanumab-treated cases [7], suggest that immunotherapy-induced glial activation does not necessarily lead to persistent neuroinflammation. This may reflect resolution of transient inflammatory responses after treatment ends or fundamental differences between rodent and human immune signaling. The absence of long-term gliosis supports the potential tolerability of anti-Aβ biologics as dosing intervals and durations are extended in clinical practice. A comprehensive neuropathological study was recently published describing five patients who participated in the ENGAGE and EMBARK clinical trials [7]. The principal findings relevant to our work were that Aβ levels in cortical layer I of the middle frontal cortex, primary visual cortex, occipital gyrus, and parahippocampal gyrus were significantly lower in aducanumab-treated patients compared with untreated AD controls. Beyond cortical layer I, however, no differences in total Aβ burden were detected across the regions examined. Likewise, no differences in tau pathology, vascular Aβ deposition, or inflammatory markers were reported, with the exception of increased CD68 immunoreactivity in the middle frontal cortex. We believe the contrasts between those findings and the present results are largely attributable to key baseline and treatment-related differences between the cohorts. First, whereas APOE genotype was evenly distributed in our cohort (three APOE4 carriers and three non-carriers), all participants in Boon et al. were APOE4-positive. Second, two of the five individuals examined by Boon et al. carried pathogenic PSEN1 mutations, while none of the participants in our study showed clinical or neuropathological evidence of familial AD. These differences in APOE genotype and the relative proportion of sporadic versus familial AD may have contributed to a more rapidly progressive disease course in the Boon et al. cohort, consistent with their younger average age at death (66 years vs. 75 years in the present study). Although the interval between last dose and death was similar across studies, the most substantial difference appears to be the cumulative drug exposure. In our cohort, all participants were titrated to the highest dose and received between 19 and 76 infusions in total. In Boon et al., only three of the five patients received at least six infusions at 10 mg/kg; one received primarily 3 mg/kg, and the remaining patient received only three infusions and never exceeded 3 mg/kg. Accordingly, the average cumulative aducanumab exposure in the Boon et al. cohort (121.0 mg/kg) was less than half that of the present study (303.7 mg/kg). Taken together, these differences suggest that substantially lower drug exposure and shorter treatment duration in Boon et al. may have limited the extent of Aβ clearance, while genetic factors associated with more aggressive Aβ accumulation may have further attenuated treatment response. The restricted pattern of Aβ reduction they observed—primarily localized to layer I of the frontal cortex—may therefore represent partial Aβ plaque removal insufficient to produce downstream effects on tau pathology or vascular Aβ. The present cohort likely reflects a more complete therapeutic effect of prolonged high-dose aducanumab exposure, consistent with what fluid and imaging biomarkers indicate. Limitations This study has several important limitations. First, the sample size is small and likely underpowered to detect subtle effects. Patient heterogeneity, including differences in trial cohort, treatment duration, and time since last dose, also complicates between-subject comparisons. Although tissue from aducanumab-treated participants was obtained from two sites (MADRC and the University of Washington), untreated AD controls were sourced exclusively from the MADRC; however, all histological processing and image analysis were performed at a single site (MGH), which reduces technical variability. The absence of post-treatment imaging, and placebo-treated clinical trial controls limit our ability to directly assess baseline pathology and longitudinal treatment responses. In addition, our analyses were focused on the MTL and BA20 and may not reflect treatment effects in other brain regions. Future studies incorporating larger cohorts, broader anatomical sampling, and participants spanning a range of treatment timelines will be essential for characterizing the long-term consequences of anti-Aβ biologics and for optimizing treatment strategies. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 19 KB)Supplementary file2 (DOCX 17 KB)Supplementary file3 (DOCX 23,261 KB) Supplementary file1 (DOCX 19 KB) Supplementary file2 (DOCX 17 KB) Supplementary file3 (DOCX 23,261 KB)