Lecanemab and Vascular-Amyloid Deposition in Brains of People With Down Syndrome

This study investigates the binding properties of lecanemab in the brains of people with Down syndrome, in anticipation of their inclusion in clinical trials or access to antiamyloid immunotherapies.

Key Points

Question

What types of Alzheimer disease neuropathology does lecanemab recognize in aged brains of people with Down syndrome?

Findings

Lecanemab bound to amyloid plaques in all 15 Down syndrome cases studied in patients older than 43 years. Notably, lecanemab also extensively labeled cerebral amyloid angiopathy in Down syndrome.

Meaning

These results show that significant staining of vascular amyloid lesions by lecanemab in DS brain tissue underscores the need for careful consideration of cerebral amyloid angiopathy–associated amyloid-related imaging abnormalities in Down syndrome–Alzheimer disease clinical trials.

Abstract

Importance

Anti-β-amyloid immunotherapy using lecanemab is becoming increasingly available to patients with Alzheimer disease (AD). Individuals with Down syndrome (DS) develop AD neuropathology by age 40 years, representing a significant cohort of genetically determined AD.

Objective

To investigate the binding properties of lecanemab in the brains of people with DS, in anticipation of their inclusion in clinical trials or access to antiamyloid immunotherapies.

Design, Setting, Participants

The study included cases of postmortem brain tissue analysis from 15 individuals with DS aged 43 to 68 years that were acquired from Alzheimer Disease research centers at the University of California, Irvine and the University of Kentucky from 2008 to 2021. Data were analyzed from August 2023 through May 2024.

Exposure

The binding properties of lecanemab were assessed in brain tissue.

Main Outcome

The primary outcome was the extent of lecanemab binding to amyloid plaques and brain blood vessels.

Results

Tissue from 15 people (8 were female [53%]) with DS ranging in age from 43 to 68 (mean, 56.6) years were included in the study. Lecanemab-labeled amyloid plaques appeared in all 15 DS cases studied, indicating potential target engagement. However, extensive binding of lecanemab to brain blood vessels in DS was observed, raising significant safety concerns. These findings underscore the necessity for clinical trials of lecanemab in people with DS to evaluate both safety and efficacy, particularly in individuals older than 43 years.

Conclusions and Relevance

These findings suggest significant binding of lecanemab to cerebral amyloid angiopathy in DS. Lecanemab should be rigorously tested in clinical trials for AD in the DS population to determine its safety and efficacy, especially in those older than 43 years.

Introduction

Individuals with Down syndrome (DS) develop significant Alzheimer disease (AD) pathology by 40 years of age.¹ The prevalence of dementia in this vulnerable cohort also increases with age with a median onset at 53 to 54 years.² DS is a genetic cause of AD and limits life span in people with DS, highlighting the need for interventions.² However, people with DS have been excluded³ or underrepresented⁴ in previous and ongoing clinical trials using amyloid-targeting immunotherapies. Despite a lower prevalence of systemic vascular risk factors, people with DS show high levels of cerebral amyloid angiopathy (CAA) as compared with individuals affected by late-onset AD (LOAD).⁵ As in LOAD, CAA in DS is associated with microhemorrhages,⁶ though this is a relatively late manifestation of CAA pathophysiology. In the context of antiamyloid immunotherapies, the presence of CAA-related changes has become a potential concern, as CAA appears to be a risk factor for amyloid-related imaging abnormalities (ARIA) during antiamyloid antibody treatment.⁷ For people with DS or autosomal dominant AD, the possibility that treatment using antiamyloid antibodies may compromise vascular integrity highlights an important concern for prognosis and screening prior to treatment. The purpose of the current study was to characterize lecanemab binding in the brains of people with DS to inform planning for DS clinical trials and clinical use of antiamyloid antibodies in DS-related cases of AD.

Methods

Human Samples

This was a cross-sectional case study of brain tissue acquired between 2008 and 2021. Formalin-fixed, paraffin-embedded sections (5 μm) from the dorsolateral prefrontal cortex, occipital cortex, and hippocampus of 15 cases of DS with AD pathology ranging in age from 43 to 68 years (mean age = 56.6; 8 patients were female [53%]; postmortem interval mean 8.17 hours) from the University of California Irvine Alzheimer Disease Research Center (n = 7) and University of Kentucky Alzheimer Disease Research Center (n = 8) were examined. Demographic characteristics of the cases are detailed in the eTable in Supplement 1. The study did not meet the definition of human participants research as it involved deidentified human tissue samples and was exempt from institutional review board review. Race and ethnicity were reported by the family of the donor, which was categorized based on the National Institutes of Health Standards for the Classification of Federal Data on Race and Ethnicity. Recording race and ethnicity was required by the National Institute of Health for this study.

Antibodies

Using publicly available sequences of the lecanemab and ponezumab, biosimilar human immunoglobulin G (IgGs) were sourced from Proteogenix and Med Chem Express, respectively. Antitrinitrophenol human IgG was generously provided by Megan Batson from Sanofi, while recombinant antihuman amyloid-β antibody (21F12) mouse IgG was a gift from Elan/Johnson. Purified anti-β-amyloid, x-40 recombinant (QA18A67) rabbit IgG was procured from Biolegend and the Fab of Goat antihuman from Jackson ImmunoResearch Laboratories. Biotinylated secondary antibodies were purchased from Vector Laboratories and Alexa 488 plus and Alexa 594 plus secondary antibodies were acquired from ThermoFisher. The Alexa 647 secondary antibody was obtained from Jackson ImmunoResearch Laboratories.

Immunostaining of Human Brain Sections

The inherent challenge in using human IgG for staining human tissue lies in the presence of abundant endogenous human IgG, leading to a significant background signal. This elevated background complicates the differentiation between true- and false-positive signals, as illustrated in the representative immunohistochemical result of the single previous study employing lecanemab.⁸

To surmount this technical challenge, we devised a blocking method using an antihuman IgG (heavy and light chain) fragment antigen-binding region (Fab region) to mask endogenous human IgG in the human brain tissue, as outlined in Figure 1A. Paraffin sections underwent rehydration and antigen retrieval through treatment with 88% formic acid for 15 minutes. Endogenous peroxidases were blocked by treating with 0.3% H₂O₂ in PBS containing 0.03% Triton X-100 (PBST), and nonspecific binding was prevented by incubation in 5% skim milk in PBST. Subsequently, sections were incubated with goat antihuman Fab to mask endogenous human IgG. To validate this method, we used a negative control antibody (antitrinitrophenol) alongside lecanemab on adjacent paraffin sections of DS brain tissue. Following the blocking steps, sections were incubated with primary antibodies at 4 °C overnight, followed by incubation with biotinylated secondary antibodies for 1 hour. For visualization, sections were treated with avidin/biotin-horseradish peroxidase complex (Vector) and then with diaminobenzidine substrate (TCI), with intensification by Nickel Ammonium Sulfate. Photomicrographs were captured using a CX33 microscope (Olympus) equipped with an Mlchrome 5 Pro camera (Tucsen). Brightness, contrast, and threshold were adjusted using ImageJ 2.0 (National Institutes of Health). In the case of triple immunofluorescence labeling, before primary antibody incubation, sections underwent rehydration, formic acid treatment, and blocking with 5% skim milk and antihuman Fab. After all blocking steps, sections were incubated with primary antibodies at 4 °C overnight and then with Alexa-labeled secondary antibodies for 1 hour. Photomicrographs were captured using a Leica DMi8M inverted microscope.

Figure 1. Lecanemab Labels Extensive Amyloid Deposits in Brains of People With Down Syndrome (DS).

A, Schematic of immunohistochemical protocol for using human antibody on human brain tissue. B, Low magnification of DS brain tissue labeled with lecanemab along with negative control antibody red numbers highlight different structures labeled with lecanemab but not negative control antibody (1, amyloid plaques; 2, parenchymal CAA; 3, meningeal CAA; scale bar = 500 μm. C, High magnification of different amyloid deposits recognized by lecanemab; red numbers indicated different type of deposits (1, coarse-grained; 2, cotton wool; 3, classic cored structures; 4, blood vessel amyloid angiopathy in the cortex; 5, capillary amyloid angiopathy in the cortex; 6, diffuse parenchymal Aβ deposits near blood vessels; 7, frontal cortex); scale bar = 50 μm. ABC-DAB indicates Avidin-biotin complex method using 3, 3'-diaminobenzidine; IgG, immunoglobulin G.

Results

Lecanemab Binds to Amyloid Deposits in Brains of People With Down Syndrome

As depicted in Figure 1B, lecanemab distinctly labeled significant amounts of parenchymal amyloid plaques, while the negative control antitrinitrophenol antibody exhibited minimal staining, confirming a strong reduction in background human IgG staining. Specifically, lecanemab-labeled amyloid plaques, parenchymal CAA, and meningeal CAA, whereas the antitrinitrophenol antibody failed to label any of these structures.

Lecanemab Labels Vascular Amyloid Deposition in Brains of People With Down Syndrome

Under higher magnification (Figure 1C), lecanemab bound to a diverse subset of amyloid plaques, including coarse-grained, cotton wool, and classic cored structures. Lecanemab also labeled cerebrovascular amyloid deposition in CAA, including blood vessel and capillary-amyloid angiopathy in the cortex. Indeed, lecanemab exhibited a pronounced affinity for vascular amyloid deposits and diffuse parenchymal Aβ deposits near blood vessels in DS. As shown in Figure 1C from frontal cortex, plaques surrounded by amyloid-fused capillaries in the same field were observed, further highlighting the intensive labeling of vascular amyloid by lecanemab. In 15 of 15 cases that included the brains of people with DS and AD older than 43 years (eTable in Supplement 1), lecanemab consistently labeled leptomeningeal and cortical amyloid angiopathy, including blood vessel and capillary-amyloid angiopathy in dorsolateral prefrontal cortex (Figure 2A; eFigure 1 in Supplement 1). Only case 1 (45-year-old female) demonstrated fewer labeled amyloid plaques, but still showed meningeal vascular CAA labeling. Confirmatory staining with another human IgG2 antibody recognizing Aβ40, ponezumab,⁹ produced similar results (eFigure 2 in Supplement 1), supporting the robustness of these findings.

Figure 2. Lecanemab Labels Extensive Vascular Amyloid Deposits in Brains of People With Down Syndrome (DS) Among Different Brain Regions.

Low magnification of representative DS brain tissues triple labeled with anti-Aβ 40, anti- Aβ 42, and lecanemab; scale bar = 200 μm (A). High magnification of representative DS brain tissues triple labeled with anti-Aβ 40, anti- Aβ 42, and lecanemab; scale bar = 500 μm (B through D). ApoE indicates apolipoprotein E; DSAD, Down Syndrome pateints with Alzheimer Disease; F, female; M, male; NA, not applicable.

The study team further tested whether lecanemab could label plaques and CAA in other brain regions, including regions in a different vascular territory. Occipital cortex and hippocampus samples were stained with lecanemab using the same protocol. In 4 cases examined (ages 43 to 57 years), labeling of amyloid plaques in 3 cases and extensive amyloid vascular deposition in 1 case was observed (Figure 2B and C).

Lecanemab Labels Aβ40-Positive Amyloid Deposits in Brains of People With Down Syndrome

To further characterize the types of Aβ within plaques and CAA identified by lecanemab, the study team examined the association with Aβ40 or Aβ42 positivity. Double immunofluorescence with lecanemab, 21F12 (anti-Aβ42 antibody), and QA18A67 (anti-Aβ40 antibody) within the same DS brain sections revealed that anti-Aβ40 (green) and lecanemab (red) labeled vascular amyloid deposits. In contrast, lecanemab labeled only a fraction of amyloid plaques positive for anti-Aβ42 (magenta), which weakly stained vascular structures (Figure 3A). The clear difference of anti-Aβ42 and lecanemab binding was striking and consistent with a previous report.⁸ This observation was consistent for 12 cases tested. At higher magnification, anti-Aβ40 and lecanemab were notably positive for diffuse parenchymal Aβ deposits around blood vessels (Figure 3B, white arrowhead), whereas anti-Aβ42 exhibited weak staining of such structures. In Figure 3C, anti-Aβ40 and lecanemab showed a preference for labeling the periphery of the plaque (white arrowhead), while anti-Aβ42 predominantly stained the core. Furthermore, anti-Aβ40 and lecanemab exhibited strong colabeling of parenchymal CAA (Figure 3D; white arrowhead), whereas anti-Aβ42 showed minimal staining of vascular amyloid deposits. Collectively, these findings suggest that lecanemab has an affinity for Aβ40-positive structures, including extensive CAA and a subset of amyloid plaques in brains with DS.

Figure 3. Lecanemab Preferably Labels Aβ 40–Positive Amyloid Deposits in Brains of People With Down Syndrome (DS) .

Low magnification of representative DS brain tissue triple labeled with anti-Aβ 40, anti-Aβ 42, and lecanemab, scale bar = 200 μm (A). High magnification of representative DS brain tissue triple labeled with anti-Aβ 40, anti-Aβ 42, and lecanemab; scale bar = 50 μm.

Discussion

There is increasing interest in providing antiamyloid immunotherapies, such as the US Food and Drug Administration–approved lecanemab, for the treatment of AD-related cognitive impairment in people with DS. Given the potential for ARIA-hemorrhage (ARIA-H) and hemorrhagic complications common to antiamyloid antibody treatments, it is critical to examine the extent to which lecanemab may bind to vascular amyloid deposits in DS brains, as people with DS have significantly higher levels of and more severe CAA than LOAD.^5,10 The pathophysiology of ARIA detected by magnetic resonance imaging remains uncertain, including ARIA-edema and ARIA-H.¹⁰ ARIA-edema reflects vasogenic edema and ARIA-H involves microhemorrhages, macrohemorrhages, and hemosiderosis. Thus, ARIA-H is linked to CAA and occurs in patients with AD in response to amyloid-modifying therapies.

In this study, we characterized lecanemab immunoreactivity in the brains from a cohort of aging adults with DS, some as young as 43 years. Lecanemab identified a broad spectrum of vascular amyloid-deposition, encompassing leptomeningeal and cerebral amyloid angiopathy, including blood vessel and capillary amyloid angiopathy, in all 15 cases. Using ponezumab, QA18A67, and 21F12, we observed that lecanemab recognized only a subset of Aβ42-positive parenchymal amyloid plaques but overlapped more strongly with Aβ40-positive vascular amyloid deposits, indicating CAA. This finding is particularly noteworthy considering that the high affinity of therapeutic antiamyloid antibodies for vascular amyloid deposition may underlie the etiology of ARIA. In a previous study, lecanemab was also characterized immunohistochemically in the brains of 4 cases (58 to 70 years of age) with DS and in LOAD cases.⁸ The current study builds and expands on the previous work by including 15 cases with DS from 43 years and older, includes more brain regions, and people who carried the E4 allele, to highlight the consistent observation of plaque and extensive blood vessel binding. This observation heightens concerns about the safety and suitability of lecanemab for use in individuals with DS aged 40 years and older, emphasizing the need for careful evaluation in clinical trials. Indeed, CAA in DS is also associated with neuroinflammation, as reported in a recent case study further emphasizing the need for the assessment of safety.¹¹

The findings in this study help to inform future clinical trials using Aβ immunotherapies in people with DS.¹² On a positive note, preliminary results of the ACI-24.060 pilot study,¹³ which involves an active vaccine, for individuals with DS showed immunogenicity with anti-Aβ IgG antibodies and an increase of Aβ 1-40 and Aβ 1-42 plasma among participants who received the higher-dose active treatment. During this 96-week trial, treatment-emergent adverse events were mostly mild with no evidence of observed ARIA in participants aged 25 to 41 with a mean (SD) age 32.6 (4.4) years.

Taken together, the results of the present study suggest that pathological evidence of CAA is readily observable in the brains of people with DS past the age of 40 and that lecanemab extensively binds CAA-related vascular amyloid deposits in the DS brain. This underscores the need for vigilance regarding hemorrhagic complications in patients with DS who are treated with lecanemab, as there may be an enhanced risk of ARIA in this population. As in LOAD, neuroimaging to identify microhemorrhages suggestive of CAA in patients with DS will be critical, both as a screening measure and to monitor for potential adverse ARIA outcomes.^14,15 However, given that the emergence of microhemorrhages visible on susceptibility-weighted magnetic resonance imaging is a relatively late event in the CAA disease course, some people with DS may have a high burden of vascular amyloid deposition despite the absence of magnetic resonance imaging-visible hemorrhagic lesions.

Limitations

This study was limited in terms of sample size and the diversity of our cohort. In a follow-up study, as cases become available, we will expand to include younger participants to determine the age at which blood vessel labeling is first identified by lecanemab.

Conclusions

The results here also suggest that younger people with DS with lower or absent CAA may have less risk of vascular complications from anti-amyloid immunotherapy and that careful selection is important in the implementation of these new treatments for this underserved population. We are investigating whether a similar binding profile of lecanemab applies to brains of patients with LOAD as a follow-up to this study. Combined with our current research, this will further elucidate the general nature of lecanemab binding to brain amyloid deposition.

Supplement 1.

eTable. Case demographics

eFigure 1. Lecanemab labels extensive amyloid deposits in DS brains

eFigure 2. Similar to lecanemab, ponezumab labels extensive amyloid deposits in DS brains

Supplement 2.

Data sharing statement