Abstract
Background
ALZ-101 is a therapeutic vaccine comprised of a stabilised oligomeric derivative of amyloid-β42 (Aβ42), denoted Aβ42CC, that elicits antibodies targeting a low-abundance, neurotoxic Aβ oligomer species. This first-in-human, single-center Phase 1b trial evaluated the safety, tolerability, and immunogenicity of ALZ-101, with exploratory biomarker and clinical outcomes, in participants with early Alzheimer’s disease (AD).
Methods
Thirty‑two participants with biomarker‑confirmed early AD (mean age ∼69 years, ∼30% female, Mini‑Mental State Examination ∼24, Alzheimer’s Disease Composite Score [ADCOMS] ∼0.55, Clinical Dementia Rating–Sum of Boxes ∼3.0, ∼70% APOE ε4 carriers) were enrolled in a randomised, double‑blind, placebo‑controlled main study (Part A1) and a semi‑blinded extension (Part B). In Part A1, participants received intramuscular ALZ‑101 125 µg (n = 10) or 250 µg (n = 10) or placebo (n = 6) at Weeks 0, 4, 8, and 16. After ≥ 14 weeks of follow‑up, eligible participants (n = 23) entered Part B and received two or four additional 250 µg doses of ALZ‑101 with partial blinding of original allocation. An open‑label arm (Part A2) explored ALZ‑101 400 µg (n = 6) on the same schedule as Part A1. Primary endpoints were safety and tolerability. Secondary and exploratory endpoints included Aβ‑specific antibody responses, cerebrospinal fluid (CSF) and plasma biomarkers, and cognitive/functional scales.
Results
ALZ-101 was not associated with any clear safety signals, and no serious adverse drug reactions were observed. Sixteen asymptomatic ARIA‑H events, including two superficial siderosis events, occurred in nine participants (two initially on placebo, six on 125 µg, one on 250 µg), and one asymptomatic ARIA‑E occurred in the extension phase in a participant originally randomised to placebo; no ARIA was observed in the 400 µg arm. ALZ‑101 induced robust Aβ‑specific IgG responses in nearly all vaccinated participants, with classical primary/booster kinetics and high responder rates at all dose levels. No statistically significant between‑group differences were detected for pre‑specified CSF biomarkers or cognitive and functional outcomes versus placebo in the randomised phase. Over longer follow‑up in Part B, participants initially randomised to ALZ‑101 showed numerically smaller increases in CSF neurofilament light chain (NfL) and tau and a flatter ADCOMS trajectory than those initially on placebo, but these differences were not statistically significant. For CSF tau, attenuation of the increase seemed to be present already in Part A1, whereas for the more slowly changing NfL, potential attenuation emerged only over the longer Part B.
Conclusion
In this small Phase 1 study in early AD, ALZ‑101 was not associated with clear safety signals and elicited strong, durable Aβ‑specific antibody responses. Exploratory biomarker and ADCOMS findings are compatible with a possible disease‑modifying effect, but formal evidence of clinical efficacy is lacking, and all non‑primary analyses should be interpreted with caution. These results support further evaluation of ALZ‑101 in larger, adequately powered trials.
Registered at clinicaltrials.gov: NCT05328115, 14 April 2022.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s13195-026-02082-9.
Keywords: Early Alzheimer’s disease, Vaccine, Phase 1, Aβ42 oligomer
Introduction
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder and the leading cause of dementia worldwide [1]. The pathophysiology of the disease is marked by gradual accumulation of protein aggregates in the brain, including extracellular plaques mainly composed of insoluble amyloid-β (Aβ), soluble Aβ oligomers, and intracellular tau tangles [2]. This aggregation process evolves over decades, ultimately disrupting brain function and triggering neurodegeneration by mechanisms that remain largely unknown.
Aβ accumulation is widely regarded as the central causal factor in AD pathophysiology. However, the precise causal relationship between Aβ and tau remains debated, and substantial experimental and clinical evidence highlights a synergistic relationship between Aβ and tau pathologies in AD [3]. To date, therapies targeting tau pathology have not shown robust, consistent cognitive benefit in late‑stage trials. Several monoclonal antibodies directed against the N‑terminus of extracellular tau, e.g., gosuranemab [4], tilavonemab [5], and semorinemab [6], have demonstrated robust target engagement and reductions in cerebrospinal fluid (CSF) tau but no meaningful improvement on cognitive or functional measures in symptomatic AD. Similarly, agents preferentially directed against misfolded, aggregated tau species thought to mediate seeding and spread have not been more efficacious: the conformational monoclonal antibody zagotenemab [7] failed to slow clinical disease progression versus placebo in a Phase 2 trial in early symptomatic AD despite evidence of target engagement, and the active tau vaccine AADvac1 showed favourable effects on neurodegeneration biomarkers but did not demonstrate clear improvement in cognition or function [8, 9]. Several other tau targeting therapies are in early stages of development, including the intrathecally administered tau‑targeting antisense oligonucleotide BIIB080 (MAPT Rx) which has shown marked reductions in CSF total and phosphorylated tau, with numerically slower declines in exploratory clinical measures in early AD individuals, pending confirmation in Phase 2 [10, 11].
In contrast, novel monoclonal antibodies designed to clear Aβ plaques have demonstrated the ability to reduce the rates of cognitive and functional decline [12–14]. Their effect sizes, however, are modest [15–17], and these drugs carry risks of adverse effects, including amyloid-related imaging abnormalities (ARIA). These are brain lesions detected by magnetic resonance imaging (MRI), including edema (ARIA-E) or haemorrhage (ARIA-H), that commonly occur as potentially serious side effects during anti-Aβ antibody treatments targeting amyloid plaques in the brain [18]. ARIA thus represent a major risk–benefit constraint for this group of therapies, particularly in the context of apolipoprotein E (APOE) ε4. Of note, pooled analysis of gantenerumab Phase 3 studies shows a strong genotype dependence for new ARIA‑H: any new ARIA‑H in 17.3% of non‑ε4-carriers, 19.9% of ε4 heterozygotes, and 41.1% of ε4 homozygotes [19]. This genotype dependence is now reflected in the product information and regulatory guidance of anti-Aβ antibody drugs, where APOE ε4 homozygotes are identified as a subgroup with substantially elevated ARIA risk, leading to boxed warnings for plaque-reducing monoclonal antibody treatments Leqembi® (lecanemab) and Kisunla® (donanemab) and, in some jurisdictions, recommendations against their use in APOE ε4 homozygous patients, or only with heightened monitoring.
Although pathological Aβ deposition is central to AD, converging evidence indicates that soluble oligomeric Aβ assemblies are the principal mediators of synaptic dysfunction and neurotoxicity, rather than fibrillar Aβ plaques. Synthetic Aβ oligomers can disrupt long‑term potentiation, form membrane ion channels, perturb Ca²⁺ homeostasis and mitochondrial function, induce oxidative stress, and promote downstream tau pathology, as summarised in several comprehensive reviews [20, 21]. Brain‑derived Aβ assemblies isolated from AD tissue are even more potent on a molar basis than synthetic preparations [22, 23], and can drive synaptic failure, network hyperexcitability, and tau misfolding in experimental systems [24–27]. Importantly, only a small subset of these brain‑derived oligomers appears to be highly toxic, while most oligomeric and plaque‑bound Aβ species are less bioactive, highlighting the need for therapies that preferentially target disease‑relevant oligomer conformations [27, 28].
Many plaque‑reducing monoclonal antibodies — including mAb158/BAN2401/lecanemab [29], aducanumab [30] and 19.3/ACU193/sabirnetug [31] — bind linear epitopes in the flexible N‑terminus of Aβ, a region that structural studies show remains solvent‑exposed and structurally similar across monomers, oligomers, and fibrils [32, 33]. Under assay conditions that normalise epitope valency, such antibodies indeed show comparable apparent affinity for monomeric, oligomeric, and fibrillar Aβ, suggesting that their apparent ‘oligomer selectivity’ largely reflects avidity for multivalent aggregates rather than preferential recognition of oligomer species [28]. This may limit their ability to discriminate between pathogenic Aβ oligomers and more inert Aβ oligomers and plaques and motivates complementary strategies that more directly target conformational oligomer epitopes.
Given these limitations of N‑terminally directed immunotherapies for engaging brain‑derived toxic oligomers, alternative strategies that more directly target oligomeric Aβ merit consideration. One such example is the small‑molecule aggregation inhibitor valiltramiprosate: in a Phase 3 trial in APOE ε4/ε4 homozygotes with early AD, valiltramiprosate did not significantly benefit the overall population but did decelerate brain atrophy (hippocampal ~ 18%, cortical thickness ~ 20%), and in a prespecified mild cognitive impairment (MCI) subgroup with a Mini-Mental State Examination (MMSE) score > 26 showed more pronounced slowing of brain atrophy (hippocampal ~ 26%, cortical thickness ~ 35%), together with a nominally significant effect on the Alzheimer’s Disease Assessment Scale – Cognitive Subscale 13 (ADAS‑Cog13) [34]. These findings provide exploratory clinical support for the oligomer‑targeting concept — suggesting that modulation of Aβ oligomer formation can reduce brain atrophy — while also indicating that the current compound is not sufficiently efficacious to deliver robust clinical efficacy across the broader early‑AD population, even within APOE ε4/ε4 homozygotes.
Active vaccination against Aβ is a promising modality in AD therapeutics as it stimulates the immune system to produce sustained, specific antibodies that may target pathological Aβ assemblies or slow their accumulation over decades, thereby providing long-lasting protection with minimal treatment burden. Most therapeutic vaccine development projects targeting Aβ have, however, failed. One early candidate, AN1792, was discontinued in early clinical development due to an unsafe Th1-driven cellular immune response [35, 36]. Several later vaccine candidates have safely elicited humoral, antibody-mediated Th2 responses mainly targeting the N-terminus of Aβ (recently reviewed in ref. [37]). None have proven efficacious on cognitive outcomes, possibly due to insufficient specificity for the toxic Aβ species. Developing active vaccines that effectively and selectively target neurotoxic Aβ oligomers remains difficult, as non-stabilised oligomers and even fibrils tend to induce immune responses predominantly targeted against the immunodominant N-terminus of Aβ, a structure widely accessible in most Aβ species [38, 39].
Interestingly, in comparison with monoclonal antibody therapies targeting amyloid plaques, active Aβ vaccines evaluated to date (e.g., CAD106 [40], ACC‑001 [41], UB‑311 [42]) have shown much lower incidence of ARIA‑E and generally low, often placebo‑like rates of ARIA‑H, although small subject numbers and limited experience, particularly in APOE ε4 homozygotes, preclude definitive safety conclusions. Thus, while active immunization may offer a more favourable ARIA profile than currently approved plaque‑removing monoclonal antibodies, especially in genetically high‑risk patients, careful MRI surveillance and genotype‑informed risk assessment remain essential throughout clinical development.
ALZ‑101 is an active immunotherapy developed for the treatment of early AD. It uses oligomers of Aβ42CC as immunogen - a protein‑engineered Aβ42 variant with an intramolecular disulfide lock in the central region that prevents fibrillisation and stabilises soluble oligomers with an antiparallel β‑sheet architecture distinct from fibrillar Aβ [43]. The earliest β structured Aβ42CC oligomers detectable during aggregation are on the order of ~ 100 kDa by size‑exclusion chromatography with multi‑angle light scattering, and larger oligomers are formed at higher peptide concentrations and prolonged incubation times, eventually resembling what is typically referred to as protofibrils [28, 43, 44]. The oligomeric antigens in ALZ-101 used for study ALZ-C-001 had an average size of 744 kDa (SD 99 kDa; n = 7 independent measurements; Sandberg, A., data on file).
When formulated with aluminium hydroxide, these Aβ42CC oligomers elicit a polyclonal IgG response in animals that is strongly dominated by oligomer‑specific antibodies: pre‑absorption on Aβ42 fibrils largely abolishes fibril binding while preserving binding to Aβ42CC oligomers (see Supplementary Fig. 1 for an example from immunised rabbits). The unique loss of N-terminal immunodominance in ALZ-101 may result from conformational restriction of residues 16–33, the main T-cell epitope of the Aβ peptide [45], which we hypothesise to alter T-cell recognition and redirect T-cell help toward B cells recognizing non-N-terminal epitopes. It may also reflect altered antigen processing and presentation by antigen-presenting cells.
The monoclonal antibody ALZ‑201, derived from ALZ‑101‑immunised mice, recognises the same conformational Aβ42CC oligomer epitope but not monomeric or fibrillar Aβ42 and can specifically remove a low‑abundance, neurotoxic oligomeric species from AD brain tissue ex vivo, supporting the disease relevance of the targeted oligomer [28]. ALZ-201 binds all Aβ42CC oligomers ≥ 100 kDa with equal affinity, indicating that affinity is structure-dependent rather than size-dependent, and this epitope is not detectable on human plaques by standard immunohistochemistry [28]. A tissue cross-reactivity screen of human ALZ-101–derived polyclonal serum against AD brain tissue has not been performed, however, so potential cross-reactivity with conformational epitopes on plaques or vascular amyloid therefore remains unassessed.
The current clinical study ALZ-C-001 (NCT05328115) was conducted to assess the safety, tolerability and immunogenicity of ALZ-101 in participants with early AD, and is the first clinical trial to evaluate a therapeutic vaccine designed to specifically target neurotoxic Aβ42 oligomers in AD.
Methods
This study was conducted in accordance with international and national regulations governing biomedical research involving human subjects, including the Declaration of Helsinki, Good Clinical Practice, and all applicable Finnish national regulations. Prior to initiation, the study protocol and informed consent materials received approval from the National Committee on Medical Research Ethics (TUKIJA). All participants provided written informed consent before any study procedures were undertaken.
Study design
This was a first-in-human, Phase 1, randomised, double-blind, placebo-controlled, parallel group study of ALZ-101 in participants diagnosed with mild AD or MCI due to AD according to the National Institute on Aging-Alzheimer’s Association criteria [46], collectively referred to as early AD patients. The study was conducted at a single contract research organisation in Finland, operating two clinical units located in Turku and Helsinki. The study consisted of a main study (including a placebo-controlled Part A1 and an open-label Part A2) and a semi-blinded extension phase for participants who completed the placebo-controlled study (denoted Part B). The study design is outlined in Fig. 1.
Fig. 1.
ALZ-C-001 study design. In Part A1, the length of the follow-up period after the last administration varied from 14 to 52 weeks, depending on whether participants proceeded to Part B. In Part A2, all participants had a follow-up period lasting 4 weeks after their last administration. Participants randomised to placebo in Part A1 received active doses at Weeks 4 and 8. All other participants received placebo injections at Weeks 4 and 8, in order to maintain blinding until completion of Part A1
ALZ-101 and placebo were given as intramuscular injections and were formulated with aluminium hydroxide (Croda, Denmark) as a wet gel suspension. Part A1 included four doses of ALZ-101 at 125 µg (n = 10) or 250 µg (n = 10), or matching placebo (n = 6), administered at Weeks 0, 4, 8, and 16 in a randomised, double-blind manner. Part A2 was an open-label cohort where all participants (n = 6) received four doses of 400 µg ALZ-101 on the same schedule. In the extension phase, participants initially randomised to placebo in Part A1 received active 250 µg doses at Weeks 0, 4, 8, and 16 (n = 5), while those initially on active treatment in Part A1 received active 250 µg doses at Weeks 0 and 16, and placebo at Weeks 4 and 8 (n = 18) to maintain blinding of the original treatment allocation in Part A1. This design created a semi-blinded scenario where treatment allocation remained blinded for Part A1 doses, but participants and investigators were aware that active treatment was administered to all participants in the extension phase.
MRI scans
Brain MRI was performed at two sites. At site 1 (Turku University Hospital, Turku, Finland), brain MRI was acquired on a 3T GE Signa Premier scanner (GE Healthcare, Wisconsin, USA) using a protocol including sagittal 3D T2‑fluid‑attenuated inversion recovery (FLAIR), axial proton density/T2, axial diffusion‑weighted imaging (DWI), sagittal 3D T1‑weighted, and axial T2*‑weighted sequences. At site 2 (Docrates Cancer Center, Helsinki, Finland), brain MRI was acquired on a 3T Siemens Magnetom Vida scanner (Siemens Healthineers, Erlangen, Germany) using a protocol including sagittal 3D FLAIR, axial proton density/T2, axial DWI, sagittal 3D T1‑weighted, and axial T2*‑weighted sequences. For each participant, the same scanner platform and harmonised protocol was used across visits. MRI was performed at screening/baseline, at prespecified on‑treatment time points (Weeks 3, 7, 15, 30/42; Part B: Weeks 2, 6, 14, 42; Part A2: screening, Weeks 3, 7, 15, 20), and whenever clinically indicated in the event of new or worsening neurological symptoms, reflecting current guidance for ARIA surveillance under anti‑amyloid therapies [47]. See Supplementary Material for full sequence details.
All MRIs were transferred pseudonymised to a central MRI laboratory (Turku University Hospital) for blinded assessment. Screening MRIs were compared against prospectively defined exclusion criteria (e.g. >4 microhaemorrhages, superficial siderosis, prior macrohaemorrhage > 1 cm, significant infarcts or extensive white‑matter disease). ARIA‑E was defined as new or worsening FLAIR hyperintensity compatible with vasogenic edema and/or sulcal effusions and graded as mild, moderate or severe by maximal extent; ARIA‑H included new or increased microhaemorrhages (≤ 1 cm), macrohaemorrhages (> 1 cm) and superficial siderosis, with lesion counts and locations recorded and changes versus the prior scan classified as new, increasing, stable or resolving. Local radiology reports focused on safety (ARIA and meningoencephalitis); in case of disagreement, the central read was specified as primary for study reporting. MRI findings informed both eligibility and on‑treatment safety decisions. For any ARIA finding, further dosing and the intensity of MRI follow‑up (typically every 4 ± 1 weeks until ARIA‑E resolution or ARIA‑H stability) were decided by the Investigator according to the study protocol.
Participant selection
Men and women aged 50 to 83 years (inclusive) were eligible for inclusion if they had MCI due to AD or mild AD according to the National Institute on Aging-Alzheimer’s Association criteria [46]. Eligible participants were required to have a Clinical Dementia Rating (CDR) [48] global score of 0.5 or 1, a CDR memory box score of at least 0.5, and a MMSE [49] score of at least 20 at screening. CSF analysis had to show a biomarker profile consistent with increased amyloid plaque load and AD pathology as indicated by the Aβ42/40 ratio. Participants receiving acetylcholinesterase inhibitors (AChEIs), memantine, or both, were required to have been on a stable dosage for at least eight weeks prior to the first administration of the study drug. All participants were also required to have an identified, reliable, and knowledgeable study partner willing and able to support the participant throughout the study and provide follow-up information.
Participants were excluded if they had any contraindications to MRI of the brain or were unable to undergo brain MRI throughout the study. Screening MRI scans (3T) revealing clinically significant pathological lesions suggestive of a dementia diagnosis other than early AD led to exclusion. Additional exclusion criteria included a modified Hachinski Ischemic Score greater than 4 at screening; a history of cerebrovascular events, including transient ischemic attack or stroke, within 12 months prior to screening; seizures occurring within the past two years; or the presence of any psychiatric diagnosis or symptoms (such as hallucinations, major depression, delusions, schizophrenia, or bipolar disorder) that could interfere with study participation. Detailed inclusion and exclusion criteria are provided in Supplementary Material.
Baseline demographics of the ALZ-C-001 study population are provided in Table 1.
Table 1.
Baseline characteristics (safety analysis set)
| Placebo (n = 6) | ALZ-101 125 µg (n = 10) | ALZ-101 250 µg (n = 10) | ALZ-101 400 µg (n = 6) | |
|---|---|---|---|---|
| Age, yearsa | 67.7 ± 5.2 | 72.1 ± 6.0 | 66.8 ± 7.4 | 66.2 ± 5.2 |
| Age >64 years, n (%) | 5 (83.3) | 9 (90.0) | 7 (70.0) | 3 (50.0) |
| Female, n (%) | 5 (83.3) | 4 (40.0) | 2 (20.0) | 1 (16.7) |
| Not Hispanic or Latino, n (%) | 6 (100.0) | 10 (100.0) | 10 (100.0) | 6 (100.0) |
| Body weight, kga | 67.7 ± 9.3 | 73.2 ± 8.0 | 72.3 ± 12.2 | 76.1 ± 19.1 |
| Body mass index, kg/m2 a | 25.8 ± 3.4 | 25.5 ± 4.4 | 24.3 ± 3.1 | 24.0 ± 6.0 |
| MMSE scorea | 24.5 ± 2.8 | 24.1 ± 3.3 | 23.3 ± 3.2 | 25.0 ± 3.1 |
| Clinical state, n (%) | ||||
| MCI due to AD | 1 (17.0) | 3 (30.0) | 2 (20.0) | 1 (17.0) |
| Mild AD | 5 (83.0) | 7 (70.0) | 8 (80.0) | 5 (83.0) |
| CDR global score a | 0.5 ± 0 | 0.6 ± 0.16 | 0.5 ± 0 | 0.5 ± 0 |
| CDR-SB scorea | 2.3 ± 0.9 | 3.5 ± 1.1 | 2.7 ± 1.0 | 2.8 ± 0.8 |
| APOE status, n (%) | ||||
| ε2/ε3 | 0 | 1 (10.0) | 2 (20.0) | 0 |
| ε3/ε3 | 1 (16.7) | 1 (10.0) | 4 (40.0) | 1 (16.7) |
| ε3/ε4 | 4 (66.7) | 5 (50.0) | 2 (20.0) | 1 (16.7) |
| ε4/ε4 | 1 (16.7) | 3 (30.0) | 2 (20.0) | 4 (66.7) |
| CSF biomarkersa | ||||
| Aß42/Aß40 ratio (ref. >0.1) | 0.043 ± 0.009 | 0.041 ± 0.006 | 0.044 ± 0.007 | 0.043 ± 0.010 |
| Total tau, pg/mL (ref. <290) | 635 ± 180 | 712 ± 283 | 580 ± 198 | 755 ± 243 |
| P-tau181, pg/mL (ref. <61) | 87.7 ± 25.1 | 99.6 ± 42.5 | 76.5 ± 26.1 | 100.8 ± 34.6 |
| ARIA-H, microhaemorrhage, n (%) | 0 | 6 (60.0) | 1 (10.0) | 0 |
AD Alzheimer’s disease, APOE apolipoprotein E, ARIA‑H amyloid-related imaging abnormalities–haemorrhage/hemosiderin deposits, CDR Clinical Dementia Rating, CSF cerebrospinal fluid, MCI mild cognitive impairment, MMSE Mini‑Mental State Examination, SB sum of boxes
a = mean ± standard deviation
Randomisation
In Part A1, randomisation was conducted in blocks and was restricted such that at least three participants received the 125 µg dose of ALZ-101 before any participant was randomised to the 250 µg dose level. Randomisation was not stratified. No randomisation was used for Part B or Part A2. Study participants, study partners, and study personnel remained blinded to treatment allocation in Part A1 throughout the study, but were aware that active treatment was administered to every participant in the extension phase (Part B). Preparation of the study drug was performed by an unblinded study team member, while drug administration in Parts A1 and B was conducted by blinded personnel. In Part A2, knowledge of active treatment was maintained by all due to the open-label design.
Study assessments
Safety and tolerability were evaluated from adverse events; clinical laboratory testing (chemistry, haematology, urinalysis) from blood, urine, and CSF; vital signs (heart rate, blood pressure, body temperature, and body weight); 12-lead electrocardiogram (ECG); physical and neurological examinations; 3T brain MRI; and the Columbia-Suicide Severity Rating Scale (C-SSRS) [50]. The Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change (ADCS-CGIC) [51] scale also was used as a safety assessment. Safety MRI scans of the brain were performed once during the screening period, three times during the treatment period, and once during the follow-up period. The results of each MRI evaluation were reviewed before each subsequent dosing. Possible clinical signs of inflammation (e.g., fever, increases in inflammatory markers in laboratory tests) were assessed throughout the study. After the first dose of study medication, participants were hospitalised for 24 h for observation as a precaution. An independent Data and Safety Monitoring Board periodically reviewed available safety data.
Exploratory assessments included changes in cognition and function assessed with the Alzheimer’s Disease Composite Score (ADCOMS) [52], CDR – Sum of Boxes (CDR-SB) [53], MMSE, ADAS-Cog13 [54], Neuropsychological Test Battery (NTB) [55], Alzheimer’s Disease Cooperative Study – Activities of Daily Living (ADCS-ADL) [56], and Neuropsychiatric Inventory (NPI) [57].
Primary outcomes
The primary outcomes were safety and tolerability of ALZ-101 compared with placebo, based on the incidence of adverse events (AEs), AEs of special interest (AESIs), including injection-related events (IREs) and ARIAs, and other safety assessments. The number of participants with clinically significant worsening of cognitive or functional outcomes on the ADCS-CGIC scale (scores of 6 and 7) was recorded.
Secondary outcomes
Immunogenicity of ALZ-101 was evaluated by measuring Aβ-specific antibodies in serum using an enzyme-linked immunosorbent assay (ELISA; Supplementary Material). The IgG isotype antibodies were detected with europium-labeled Protein G, and IgA and IgM antibodies were measured using goat anti-IgA and mouse anti-IgM antibodies. An in-study baseline cut point was calculated from the mean signal of drug-naïve participant samples + 3.09 SD. Values equal to or above the cut point were scored positive and titrated until the signal was below this cut point. The last dilution value above the cut point was the reported titre. Endpoints included Aβ-specific antibody titre of post-baseline samples (if baseline sample was negative) or titre fold increase (ratio of any post-baseline Aβ-specific antibody titre to baseline antibody titre in serum); area under serum Aβ-specific antibody titre curve from Week 0 to Week 20, maximum titre level (Cmax), and time to peak (tmax); and relative abundance of Aβ-specific antibody subclasses (IgG, IgM, and IgA) in serum. Responders were defined as a positive post-baseline result for Aβ-specific antibodies (if baseline sample was negative) or post-baseline antibody titre at least four times the baseline antibody titre in serum.
Exploratory outcomes
CSF samples were collected at screening, 4 weeks after the last immunization in Parts A1 and A2 (Week 20), and 4 weeks after the last immunization in Part B (Week 20 of Part B). Biomarker measurements were performed at the Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Sweden, in one round of experiments using one batch of reagents by board-certified laboratory technicians who were blinded to clinical data. Intra-assay coefficients of variation were below 10%.
Pre-specified biomarkers in CSF included levels of Aβ42 and Aβ40 and the ratio Aβ42/40, neurofilament light chain (NfL), total tau protein (T-tau), phosphorylated tau protein 181 (P-tau181), neurogranin, soluble triggering receptor expressed on myeloid cells 2 (sTREM-2), and Chitinase‑3‑like protein 1 (YKL-40). CSF NfL, neurogranin, and sTREM2 concentrations were measured using in-house ELISAs as previously described [58–60]. CSF YKL-40 concentration was measured using a commercially available sandwich immunoassay with electrochemiluminescence detection (Meso Scale Discovery, Rockville, MD, USA), and CSF Aβ42 and Aβ40 were measured with the LUMIPULSE G assay (Fujirebio) [61]. CSF NfL, T-tau, P-tau181, neurogranin, and Aβ assays are validated for routine clinical use, whereas CSF YKL-40 and sTREM2 assays are for research use only.
Post hoc analyses included measurement of CSF synaptosomal-associated protein 25 kDa (SNAP-25), as well as plasma NfL and plasma phosphorylated tau protein 217 (P-tau217) in blood samples collected at baseline (Week 0), 4 weeks (Week 20), and 26 weeks (Week 42) following the last administration in Part A1. For Part B, plasma levels of these biomarkers were measured at 4 weeks (Week 20 of Part B) and 26 weeks (Week 42 of Part B) after the last administration. CSF SNAP-25 and plasma biomarkers were not evaluated on participants in the 400 µg dose arm (Part A2). Plasma NfL levels were measured using Single molecule array (Simoa) technology and commercially available assays on an HD-X instrument, according to instructions from the assay manufacturer (Quanterix, Billerica, MA). Plasma P-tau217 was measured as previously described [62]. The method used for CSF SNAP-25 quantification has also been described previously [63]. Assays for plasma NfL and P-tau217 are validated for routine clinical use whereas that for CSF SNAP-25 is for research use only.
The pattern of possible T-cell activation induced by ALZ-101 in peripheral blood mononuclear cells (PBMCs) was evaluated in all participants. Blood samples for preparation of PBMCs were obtained at baseline and at 4 and 26 weeks after the last dose of Part A1 (Week 20 and Week 42) for the placebo, 125 µg and 250 µg of ALZ-101 groups, and at baseline and 4 weeks after the last dose of Part A2 (Week 20) for the 400 µg of ALZ-101 group. An enzyme-linked immunospot (ELISpot) assay was set up and validated on in-study samples to measure cell viability and responses to stimulation with a custom-synthesised mix of overlapping 15-mer Aβ peptide fragments with a four-residue overlap or with an anti-CD3 antibody as positive control. Interferon-gamma (IFNγ) and interleukin-4 (IL-4) were used as indicators of Th1 and Th2 responses. See Supplementary Material for details.
Changes in cognition and function were assessed at screening, Week 20 of Part A1 and A2, and Week 42 of Part A1, as well at Week 20 and 42 of Part B. Scales included the ADCOMS, CDR-SB, MMSE, ADAS-Cog13, NTB, ADCS-ADL, and NPI.
Statistical analysis
No formal sample size calculations were performed as the primary objective was to evaluate the safety and tolerability of ALZ-101. The probability of at least one AE in 10 participants was 0.80 given a true event probability (i.e., incidence of AE) of 0.15. For AEs with a true event rate of 0.06 and a group size of 10, the probability of any event was 0.46. With a total of 20 participants included in the two ALZ-101 treatment groups (125 µg and 250 µg), the probability was 0.96 that any event would occur given a true event probability of 0.15. Given a true event rate of 0.06 and 20 participants, the probability of any event observed was 0.71.
In Part A1, data were analysed for the safety analysis set, which was all participants who received at least one dose of ALZ-101 or placebo; the full-analysis set (FAS) was all randomised participants who had a valid baseline assessment and at least one valid post-baseline assessment of the Aβ-specific antibody titre. For Part A2 and Part B, the FAS was all participants in the safety analysis who had at least one valid assessment of the Aβ-specific antibody titre.
Data were analysed using descriptive summary statistics. Analysis of the titre-based responder outcome at 4 weeks after the fourth dose in Part A1 in relation to treatment with ALZ-101/placebo was based on observed cases. The response probability for participants treated with ALZ-101 for each dose group was estimated with 95% confidence intervals (CI) based on the binomial distribution.
The evaluation of the time course in the Aβ-specific antibody titre was based on observed cases for all participants. Estimates of the temporal development in the Aβ-specific antibody titre and in the titre fold increase within each dose group and the comparison of the different dose groups at each time point were based on mixed models for repeated measurements (MMRM) of log-transformed values. The MMRM included treatment group, time point and the interaction between treatment group and time point as fixed factors and repeated measures within subject over time points.
Exploratory clinical efficacy endpoints comprised cognitive and functional scales (ADCOMS, CDR‑SB, CDR‑GS, MMSE, ADAS‑Cog13, NTB, ADCS‑ADL, NPI). In Part A1, change from baseline in each scale was analysed using a MMRM with treatment group, time point, and treatment‑by‑time interaction as fixed effects, baseline score as covariate, and an unstructured within‑patient covariance; if convergence failed, an autoregressive structure was used, as prespecified in the statistical analysis plan. The study was primarily designed for evaluation of safety; no formal efficacy hypotheses or adjustments for multiple comparisons were prespecified or performed, and P values for clinical scales are therefore nominal and interpreted as exploratory.
All statistical analyses were performed using SAS® (Version 9.4 or higher, SAS Institute Inc., Cary, NC, USA).
Exploratory external‑comparator analysis of ADCOMS
Because all placebo‑treated participants subsequently received ALZ‑101 in the extension phase, no long‑term internal placebo comparator was available for ADCOMS. Therefore, a non‑prespecified exploratory external‑comparator analysis was conducted using published data from a mild‑AD subgroup in the Alzheimer’s Disease Neuroimaging Initiative (ADNI; [64]). Participants randomised to ALZ‑101 125 or 250 µg in Part A1 were pooled (n = 20). Group‑level baseline comparability with the ADNI mild‑AD cohort was assessed using ADCOMS, MMSE, ADAS‑Cog13, and CDR‑SB (Supplementary Table 1). Annual ADCOMS slopes for the pooled ALZ‑101 group were estimated with a linear mixed‑effects model (LMM) including fixed effects of time and treatment and random intercepts and slopes. For the ADNI cohort, published mean ADCOMS values and standard deviations at approximately 0, 6, 12, and 24 months were used to fit a generalised least‑squares (GLS) linear model and derive an annual slope and its variance. The difference between slopes was tested with a z‑statistic. This analysis was exploratory, did not adjust for individual‑level prognostic factors, and all P values are nominal. Further details are provided in the Supplementary Material.
Results
Of 63 participants screened, 33 (52.4%) were involved in the study between 29 October 2021 and 27 January 2025 (Fig. 2). In Part A1, 20 participants were randomised to ALZ-101 and 6 to placebo, of which 23 (88.5%) continued into Part B. Six participants were recruited into the A2 arm. One participant was withdrawn from the study before dosing. All 32 participants dosed were included in the safety and FAS populations.
Fig. 2.
Patient disposition
Apart from an imbalance in individuals with ARIA-H at baseline, where six of the seven were randomised to the 125 µg arm and one to the 250 µg arm, baseline characteristics (Table 1) were generally comparable across treatment groups with a mean (SD) age of 68.5 (6.4) years. Twelve (37.5%) of the participants were female. At screening, 25 (78.1%) participants were classified as mild AD and seven (21.9%) as MCI. The mean (SD) MMSE score was 23.4 (2.7) and 26.4 (3.0) for the mild AD and MCI individuals, respectively, and 24.1 (3.03) for the trial population as a whole. The mean (SD) CDR-SB score was 2.98 (0.90) and 2.50 (1.44) for the mild AD and MCI populations, respectively, and 2.89 (1.02) for the trial population as whole. The majority of participants (n = 22; 68.8%) were APOE ε4 carriers — 12 (37.5%) heterozygotes and 10 (31.3%) homozygotes — and 10 participants (31.2%) were non-carriers. Concomitant AChEI was used by 27 (84.4%) participants and memantine by 18 (56.3%) participants.
Safety and tolerability
All participants experienced at least one AE, and a total of 483 AEs were recorded (Table 2). Overall, 448 (93%) AEs were assessed as mild in severity, 34 AEs were of moderate severity, and one severe AE was reported. Two (2) serious AEs, squamous cell carcinoma of the lip and atrial fibrillation, were reported for participants treated with 250 µg doses of ALZ-101. Both were judged as unlikely related to study drug and were not considered as serious drug reactions. All participants in the placebo, ALZ-101 125 µg, and ALZ-101 250 µg groups had at least one AE assessed as related (i.e., probably or possibly related) to study drug, mainly IREs. In the ALZ-101 400 µg group, four participants (66.7%) had 18 AEs considered related to treatment. A listing of all AEs is found in Supplementary Table 2. The reported AESIs were mostly systemic or local IREs.
Table 2.
Treatment-emergent adverse events (safety analysis set)
| Placebo (n = 6) | ALZ-101 125 μg (n = 10) | ALZ-101 250 μg (n = 10) | ALZ-101 400 μg (n = 6) | Total (n = 32) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Subjects (%) | Events | Subjects (%) | Events | Subjects (%) | Events | Subjects (%) | Events | Subjects (%) | Events | |
| Any AE | 6 (100) | 101 | 10 (100) | 173 | 10 (100) | 164 | 6 (100) | 45 | 32 (100) | 483 |
| Any AE leading to withdrawal from treatment | 0 | 0 | 0 | 0 | 1 (10) | 1 | 0 | 0 | 1 (3) | 1 |
| Any SAE | 0 | 0 | 0 | 0 | 2 (20) | 2 | 0 | 0 | 2 (6) | 2 |
| Any non-serious AE | 6 (100) | 101 | 10 (100) | 173 | 10 (100) | 162 | 6 (100) | 45 | 32 (100) | 481 |
| Any AESI | 6 (100) | 35 | 9 (90) | 67 | 9 (90) | 57 | 4 (66.7) | 11 | 28 (87.5) | 170 |
| Severity | ||||||||||
| mild | 3 (50) | 94 | 6 (60) | 160 | 4 (40) | 151 | 4 (67) | 43 | 17 (53) | 448 |
| moderate | 3 (50) | 7 | 3 (30) | 12 | 6 (60) | 13 | 2 (33) | 2 | 14 (44) | 34 |
| severe | 1 (10) | 1 | 1 (3) | 1 | ||||||
| Causality | ||||||||||
| not related | 15 | 41 | 33 | 1 (17) | 19 | 1 (3) | 108 | |||
| unlikely related | 36 | 50 | 66 | 1 (17) | 8 | 1 (3) | 160 | |||
| possibly related | 23 | 36 | 2 (20) | 23 | 1 (17) | 12 | 3 (9) | 94 | ||
| probably related | 6 (100) | 27 | 10 (100) | 46 | 8 (80) | 42 | 3 (50) | 6 | 27 (84) | 121 |
Groups are presented according to initial allocation to treatment groups in Part A (A1+A2). During Part B, participants in the placebo group received 4 doses of 250 µg of ALZ-101 and participants in the active ALZ-101 groups in Part A1 received 2 doses of 250 µg of ALZ-101
AE adverse event, AESI adverse event of special interest, SAE serious adverse event
A total of 16 asymptomatic ARIA-H events, of which two were superficial siderosis, occurred in nine participants: in two initially randomised to placebo (33%), six initially randomised to 125 µg (60%), and one initially randomised to 250 µg (10%), with no cases in the 400 µg group (Table 3). In three participants, the MRI-detected ARIA-H led to temporary interruption of study drug administration; treatment was resumed once the findings were judged stable on follow-up MRI. One participant (17%) who had received placebo in Part A developed both an ARIA-H and an ARIA-E event at Week 42 of Part B. On subsequent MRIs, the ARIA-E decreased in size and eventually resolved completely.
Table 3.
Overview of ARIA findings (safety analysis set)
| Treatment group (Part A) | Participant (APOE carrier status) | ARIA-H (micr. haem.) | ARIA-H (sup. sid.) | ARIA-E | ||||
|---|---|---|---|---|---|---|---|---|
| # At screening | Part A* | Part B* | Part A | Part B | Part A | Part B | ||
| Placebo | AD01 (ε4/ε4) | 0 | W42 | W90 (W42**) | W90 (W42**) | |||
| W90 (W42**) | ||||||||
| W98 (W50**) | ||||||||
| AD02 (ε3/ε3) | 0 | W61 | N/A | N/A | N/A | |||
| 125 µg | AD03 (ε3/ε4) | 1 | W42 | |||||
| AD04 (ε2/ε3) | 1 | W52 | ||||||
| AD05 (ε4/ε4) | 1 | W42 | W42 | W79 | ||||
| AD06 (ε3/ε4) | 1 | W72 | W115 | |||||
| W119 | ||||||||
| AD07 (ε3/ε4) | 1 | W42 | ||||||
| AD08 (ε3/ε4) | 1 | W33*** | N/A | N/A | N/A | |||
| 250 µg | AD09 (ε3/ε4) | 4 | W7*** | |||||
| Total | 10 | 8 | 6 | 1 | 1 | 0 | 1 | |
ARIA‑E (amyloid-related imaging abnormalities–edema/effusion) describes MRI-detected parenchymal vasogenic edema and/or sulcal effusions. ARIA‑H (amyloid-related imaging abnormalities–haemorrhage/hemosiderin deposits) encompasses MRI findings of cerebral microhaemorrhages and superficial hemosiderin deposition (superficial siderosis)
N/A Not applicable as the individual did not participate in Part B
* Weeks since inclusion in Part A
** Weeks since first administration of ALZ-101 in the placebo treatment group
*** Increase in size of a pre-existing lesion
No significant mean changes for any clinical laboratory results, vital signs or ECG parameters (Supplementary Table 3) were observed within or between any treatment groups at any time point, but individual abnormalities were recorded. In the ALZ-101 dose groups, three participants experienced clinically significant ECG abnormalities reported as AEs, including ventricular bigeminy. Sinus bradycardia was observed in one participant in each of the 125 µg and 250 µg dose groups, with the former case being associated with presyncope. One participant in the 125 µg group reported mild suicidal ideation without suicidal behavior on the C-SSRS at Week 30, which was considered an AE of depression; the symptoms resolved by 4 months.
Immunogenicity
During ELISA measurement of IgG antibody responses to ALZ-101 vaccination, it was observed that 50 individual serum samples from a healthy, drug-naïve population (acquired from BioIVT, USA, and used to establish and validate the assay ahead of study start) exhibited higher responses to the ALZ-101 antigen compared to samples from drug-naïve participants enrolled in the clinical study (data not shown). This elevated baseline response was attributed to a high level of naturally occurring antibodies to the antigen in the healthy population. High background levels of anti-ALZ-101 IgG antibodies were confirmed in a follow-up experiment using sera from healthy young individuals (BioIVT; n = 10; average age of 26.7 ± 1.5 (SD)) and healthy elderly individuals (BioIVT; n = 10; average age of 66.4 ± 3.6 (SD)) (Supplementary Fig. 2). Notably, healthy younger individuals exhibited lower reactivities against Aβ42CC oligomers than healthy elderly individuals. In fact, the younger individuals had values comparable to those of AD participants from study ALZ-C-001. Specifically, the mean ELISA signal for sera diluted 1:10 was 5304 ± 2071 (SD) for the younger individuals and 8698 ± 5215 (SD) for the older healthy individuals, while AD participants in study ALZ-C-001 had a mean signal of 4952 ± 3255 (SD; n = 32). The difference between healthy younger and healthy older was assessed using a pooled t-test and found to be 3394 (95% CI, -334 to 7122; P = 0.072). A Welch’s t-test comparing the difference between healthy older and AD patient samples found it to be 3746 (95% CI, -101 to 7593; P = 0.055).
Consequently, the initial cut-off value for assessing antibody titres and responder rates was inaccurate, necessitating adoption of an in-study cut-off value (as described in the Methods section) to more accurately evaluate antibody responses in vaccinated participants. However, since the baseline cut-off values in the ELISA were established using low serum dilutions (1:10), where naturally occurring antibodies contribute significantly to the signal also in the study population, there is a risk of underestimating true vaccine-induced antibody titres. Titres were reported only for dilutions yielding signals above the baseline cut-off, with acknowledgment of the limitation that pre-existing antibodies could impact the sensitivity and interpretation of the vaccine response.
ALZ-101 induced Aβ-specific antibody responses above the background antibody levels for IgG, with weaker responses for IgM and IgA (Supplementary Fig. 3–5). The oligomer specificity of the polyclonal IgG response to ALZ-101 vaccination was assessed post hoc by a pre-incubation experiment using non-aggregated Aβ42 and the ALZ-101 antigen on participant samples (n = 2), with the monoclonal antibody ALZ-201 included as a control. Pre-adsorption with increasing concentrations of non-aggregated Aβ42 did not substantially alter the ELISA signal for either the immune sera or the antibody, whereas pre-adsorption with the ALZ-101 antigen completely abolished it for both, suggesting that the binding profiles of the polyclonal IgG response and the monoclonal antibody ALZ-201 were very similar (Supplementary Fig. 6). These exploratory data should be interpreted with caution due to the small sample size, but are consistent with an oligomer‑preferring antibody response.
In the placebo group, Aβ-specific IgG titres remained negative, with the exception of isolated false positives at Weeks 30 and 42 in two different participants, and until Week 6 in Part B (Fig. 3). One participant each in the 250 µg and 400 µg groups had positive titres at baseline. In Part A1, all participants on active doses developed detectable IgG titres by Week 20 in the 125 µg group and by Week 30 in the 250 µg group. Mean IgG titres increased until Week 30, reaching geometric mean titres (GMT) of 99.5 and geometric SDs (GSD) of 4.89 in the 125 µg group and 92.4 (GSD 2.89) in the 250 µg group. In Part A2 (400 µg), mean IgG titres increased up to the last sampling time point (Week 20), when the highest individual titre (640) and the highest GMT of 114.8 (GSD 4.35) were recorded.
Fig. 3.
Total IgG geometric mean titres (GMT) for the main study (Part A1/A2; left panel) and the extension phase (Part B, right panel). In the Part B panel, patients were grouped by their treatment allocation in the main study. The main study dosing schedule was Weeks 0, 4, 8, and 16. Part B followed the same regimen, with patients originally randomised to active treatment receiving placebo injections at Weeks 4 and 8. Titres were defined as the last positive data point exceeding the mean signal of naturally occurring antibodies in drug-naïve patients by more than 3.09 standard deviations (SD). Data are presented as log10-transformed GMTs, with a constant offset of 0.01 added prior to transformation to accommodate zero and near-zero values. Error bars represent ±1 SD on the log-transformed scale
No statistically significant differences in IgG titres were observed between the 125 µg and 250 µg groups in Part A1 (Supplementary Table 4). Even though the mean IgG response for the 250 µg dose group seemed to be slightly higher than for the 125 µg group at all earlier time points up to Week 20, with a notably larger difference at Week 12 (mean difference 3.81; 95% CI 0.98–14.91; P = 0.054), there was not a pronounced dose–response gradient between 125 and 250 µg in this small Phase 1b study. Although the 400 µg of ALZ-101 group was not included in the statistical analyses of IgG titres, the data suggest that increasing the dose from 250 µg per injection to 400 µg per injection did not elevate IgG titres (Supplementary Table 5). All three dose levels were thus sufficient to induce a robust IgG response and further dose escalation is unlikely to increase peak antibody titres. Taken together, these findings support that all three doses are at, or very near, the plateau of the antibody dose–response. In the ALZ-101 125 µg and 250 µg groups, all participants fulfilled responder criteria during Part A1 while in the ALZ-101 400 µg group (Part A2), five of six (93.3%) participants were responders. In the placebo group, four of the five (80%) participants who continued into Part B and were treated with ALZ-101 were responders in the extension phase. Within this limited dataset, we did not observe a discernible effect of immunosenescence on vaccine‑induced IgG responses (data not shown). Nevertheless, the relatively narrow age span and lack of younger participants mean that potential age‑related effects on immunogenicity cannot be reliably excluded.
In general, IgM titres were low but fulfilled the responder definition in two (33.3%), six (60%), and three (50.0%) participants in the placebo, 250 µg, and 400 µg ALZ-101 groups, respectively (Supplementary Fig. 4). Detectable IgM levels also were seen in the 125 ug group but no subject satisfied the responder criteria. Several participants had positive IgM titres at baseline. The low IgM titre levels observed, combined with evidence of false-positive signals in the placebo group and at baseline, likely reflect a combination of weak vaccine-induced IgM responses and assay sensitivity limitations. For IgA, there were no positive titres at baseline, with weak responses observed in six (60%), six (60%), and three (50%) participants in the 125 µg, 250 µg, and 400 µg groups, respectively (Supplementary Fig. 5). In the extension part, two out of the five participants (40%) initially allocated to receiving placebo became IgA responders. No statistically significant treatment differences were observed in IgM and IgA titre levels or responder rates between the different dose levels (not shown).
Exploratory CSF biomarkers
No significant differences were detected between treatment groups for changes in CSF biomarker concentrations. Mean baseline values (at screening) and changes from baseline for pre-specified CSF biomarkers (full analysis set) are provided in Supplementary Table 6.
All treatment groups showed increases in mean CSF NfL levels from screening (approximately 9 weeks before randomization) to the second CSF sampling time point at 29 ± 2 weeks (standard error of the mean; SEM) post-screening (4 weeks after the fourth administration), with the largest increase seen in the 400 µg dose group (Fig. 4). The exposure of approximately 10 weeks to anti-ALZ-101 antibodies prior to the second CSF sampling point was thus insufficient to affect CSF NfL levels. However, at the third CSF sampling time point of the extension phase, 83 ± 3 weeks (SEM) post-screening (4 weeks after the fourth administration in Part B), participants receiving active treatment in Part A1 exhibited notably smaller mean CSF NfL increases compared to those initially on placebo.
Fig. 4.
Mean (standard error) change from baseline screening values for cerebrospinal fluid (CSF) neurofilament light chain (NfL). Screening samples were collected approximately 9 weeks before the first study drug administration. Patients were grouped according to their treatment allocation in the main study, although all patients received 250 µg ALZ-101 in the extension phase (Part B). The 125 µg and 250 µg dose arms in the main study, which displayed similar titres, have been combined for presentation. One patient in the 125-µg dose group was identified as an outlier and excluded from the mean calculation (see Supplementary Information for a plot including all data points). Variability in the x-axis is due to differences in the timing of transition from the main study into Part B
In the 125 µg treatment group, one participant exhibited pronounced decreases in CSF NfL at the second and third sampling points (− 520 pg/mL and − 770 pg/mL, respectively), which were inconsistent with the values observed in other study participants, leading to outlier analyses (Supplementary Fig. 7). Collectively, these analyses consistently highlighted a single 125 µg participant with an unusually high baseline (3590 pg/mL) and extreme subsequent decreases, raising the possibility that the observed trend for this individual reflects natural resolution or stabilization of an acute neurological insult, vascular injury, or an inflammatory process present at baseline, rather than a direct pharmacological effect. This outlier was excluded in Fig. 4 (see Supplementary Fig. 8, for a similar plot of all data) but not from the formal statistical analyses.
The placebo‑allocated group showed a larger increase in CSF NfL from Part A1 (mean change from baseline 41 ± 24 pg/mL; n = 6) to Part B (182 ± 73 pg/mL; n = 5), where the corresponding paired within‑subject mean change from Part A1 to Part B was 123 ± 60 pg/mL (SEM; n = 5; P = 0.11). In contrast, the 125 µg and 250 µg groups showed minimal net changes over the same interval: for 125 µg (including the outlier), mean changes were − 6 ± 59 pg/mL in Part A1 (n = 10) and − 36 ± 125 pg/mL in Part B (n = 7), yielding a paired within‑subject mean change of 2 ± 54 pg/mL (SEM; n = 7; P = 0.97), and for 250 µg, 63 ± 32 pg/mL in Part A1 (n = 10) and 68 ± 73 pg/mL in Part B (n = 9), yielding a paired mean change of 0.9 ± 64 pg/mL (SEM; n = 9; P = 0.99). When the single 125 µg outlier was excluded, mean changes were similar to those observed for the 250 µg group: 51 ± 18 pg/mL in Part A1 (n = 9) and 87 ± 29 pg/mL in Part B (n = 6), corresponding to a paired within‑subject mean change of 44 ± 40 pg/mL (SEM; n = 6; P = 0.32). A combined active‑dose group (125 and 250 µg in Part A1, excluding the 125 µg outlier) had mean changes of 57 ± 19 pg/mL in Part A1 (n = 19) and 76 ± 44 pg/mL in Part B (n = 15), yielding a paired within‑subject mean change of 18 ± 41 pg/mL (SEM; n = 15; P = 0.66). P values are two‑sided, approximate, using a paired t-test against 0.
CSF T‑tau (Fig. 5) showed small and highly variable changes over time across all treatment groups. In the placebo‑allocated group, the mean change from baseline was 39 ± 23 pg/mL in Part A1 (n = 6) and 37 ± 52 pg/mL in Part B (n = 5), with a corresponding paired within‑subject mean change from Part A1 to Part B of − 9.4 ± 34.7 pg/mL (SEM; n = 5; P = 0.80). The 125 µg and 250 µg groups displayed similar patterns: for 125 µg, mean changes were 28.6 ± 14.3 pg/mL in Part A1 (n = 10) and 26.7 ± 51.8 pg/mL in Part B (n = 7), yielding a paired within‑subject mean change of 5.7 ± 46.5 pg/mL (SEM; n = 7; P = 0.91), and for 250 µg, 23.3 ± 12.2 pg/mL in Part A1 (n = 10) and 22.1 ± 17.7 pg/mL in Part B (n = 9), yielding a paired mean change of 0.7 ± 14.7 pg/mL (SEM; n = 9; P = 0.96). In the pooled active‑dose group (125 and 250 µg in Part A1), the mean change from baseline was 26.0 ± 9.2 pg/mL in Part A1 (n = 20) and 24.1 ± 23.7 pg/mL in Part B (n = 16), corresponding to a paired within‑subject mean change of 2.9 ± 21.1 pg/mL (SEM; n = 16; P = 0.89). P values are two‑sided, approximate, using a paired t-test against 0.
Fig. 5.
Mean (standard error) changes from baseline screening values for cerebrospinal fluid (CSF) biomarkers: phosphorylated tau181 (P-tau181), total tau (T-tau), neurogranin, and SNAP-25. Screening samples were collected approximately 9 weeks prior to the first administration of study drug. Patients were grouped according to their treatment allocation in the main study, although all patients received 250 µg ALZ-101 in the extension phase (Part B). Variation in the x-axis reflects differences in timing of patient transitions from the main study into Part B
CSF P‑tau181 (Fig. 5) also showed small and variable changes over time. In the placebo‑allocated group, the mean change from baseline was 7.3 ± 2.3 pg/mL in Part A1 (n = 6) and 4.6 ± 7.9 pg/mL in Part B (n = 5), with a corresponding paired within‑subject mean change from Part A1 to Part B of − 4.2 ± 7.2 pg/mL (SEM; n = 5; P = 0.59). The 125 µg and 250 µg groups showed similar patterns: for 125 µg, mean changes were 5.0 ± 2.3 pg/mL in Part A1 (n = 10) and 0.7 ± 7.1 pg/mL in Part B (n = 7), yielding a paired within‑subject mean change of − 3.7 ± 6.5 pg/mL (SEM; n = 7; P = 0.59), and for 250 µg, 3.2 ± 1.7 pg/mL in Part A1 (n = 10) and 2.6 ± 2.8 pg/mL in Part B (n = 9), yielding a paired mean change of − 0.7 ± 2.5 pg/mL (SEM; n = 9; P = 0.80). In the pooled active‑dose group (125 and 250 µg in Part A1), the mean change from baseline was 4.1 ± 1.4 pg/mL in Part A1 (n = 20) and 1.8 ± 3.4 pg/mL in Part B (n = 16), corresponding to a paired within‑subject mean change of − 2.0 ± 3.1 pg/mL (SEM; n = 16; P = 0.52). P values are two‑sided, approximate, using a paired t-test against 0.
In Part A1, CSF P-tau181 and T-tau showed numerically smaller increases in the active dose groups compared with placebo, with the largest rises in the placebo group, intermediate changes at 125 µg, and the smallest at 250 µg; this inverse pattern appeared to partly extend to the open-label 400 µg arm (P-tau181: 1.8 ± 3.6 pg/mL; T-tau: 26.3 ± 26.6 pg/mL; SEM; n = 6; Fig. 5). In contrast, changes in CSF NfL at Part A1 did not exhibit a similar pattern (Fig. 4). Similar observations were made in Part B: when the placebo group received active treatment, P-tau181 and T-tau displayed an early stabilization pattern resembling that seen in the original active arms during Part A, whereas NfL continued to rise numerically in the original placebo group but appeared to level off in the continued active-dose groups.
Neurogranin, SNAP-25, YKL-40, and sTREM-2 remained relatively stable with small and variable changes across all groups and time points, showing no clear progression or treatment-related divergence between Part A1 and Part B (Fig. 5).
Plasma levels of NfL (Supplementary Fig. 9) and P-tau217 (Supplementary Fig. 10) showed small but favourable changes after active treatment compared with placebo, even if group differences were not statistically significant.
Exploratory cognitive and functional assessments
At baseline, mean cognitive and functional outcomes were comparable across the treatment groups and consistent with early AD severity (Supplementary Table 7). Some variability between participants was observed during the study, but no unexpected or sudden impairment in cognitive or clinical status was noted. Overall changes in these outcomes were generally small over the 42-week course of the placebo-controlled part of the study: ADAS-Cog13 worsened by only a few points on average in all dose groups, with similar small mean changes at Week 20 and Week 42 and no consistent separation between placebo and ALZ-101. ADCS-ADL showed modest functional decline, global CDR scores remained essentially stable, and NPI and NTB scores fluctuated only slightly around baseline, while ADCOMS changes were small and remained rather stable throughout the semi-blinded extension (Part B).
No statistically significant differences between treatment groups were found for any measures except the least squares (LS) mean difference in ADCOMS scores at Week 20 (approximately 29 weeks after the first assessment at screening), where the 125 µg ALZ-101 group showed a -0.10 difference (95% CI: -0.20 to 0.00; P = 0.040) relative to placebo, indicating lesser cognitive decline in the ALZ-101 group (Supplementary Table 8). For the 250 µg group, a similar but smaller effect was observed, with an LS mean difference in ADCOMS relative placebo at Week 20 of -0.07 (95% CI: -0.16 to 0.01; P = 0.095). By Week 42, effect sizes shrunk and 95% CIs included zero for both doses: -0.09 (95% CI: -0.21 to 0.03; P = 0.148) and − 0.04 (95% CI: -0.16 to 0.07; P = 0.444) for the 125 µg and 250 µg dose groups over placebo, undermining the robustness and persistence of the Week 20 signal.
Of note, as illustrated by arithmetic mean ADCOMS scores (Fig. 6), most participants receiving ALZ-101 in Parts A1 and B did not experience clinically meaningful cognitive decline during the entire study, and exhibited slower rates of cognitive decline compared to historical ADNI data [64] (Fig. 7). In Fig. 6, active dose groups in Part A1 are plotted together because titres were similar across the two arms (see Supplementary Fig. 12, for a plot of all treatment groups separately). A meaningful decline was defined as an increase in ADCOMS of approximately 0.05 points for patients with MCI due to AD and 0.10 points for those with mild AD dementia [65].
Fig. 6.
Mean (standard error) changes from baseline in ADCOMS scores. Baseline values represent screening assessments approximately 9 weeks prior to the first study drug administration. Patients were grouped according to treatment allocation in the main study, although all patients received 250 µg ALZ-101 in the extension phase (Part B). The 125-µg and 250-µg arms from the main study have been combined for presentation (see Supplementary Information for separated plots). Variation in the x-axis reflects differences in timing of patient transitions from the main study into Part B
Fig. 7.
Mean (standard deviation) ADCOMS scores over time for the combined ALZ‑C‑001 active dose groups (125 µg and 250 µg) and ADNI historical controls with late MCI or AD dementia at baseline. Sample sizes (n) contributing to each mean are indicated adjacent to the corresponding data points for both ALZ-C-001 and ADNI groups. ADNI data were extracted from Evenden et al. 2024 and used in a post hoc, exploratory external‑comparator analysis alongside ALZ‑C‑001 ADCOMS data. Variation in the x-axis in the ALZ-C-001 data set reflects differences in timing of patient transitions from the main study into Part B
Numerically lower ADCOMS scores were observed in the 400 µg open‑label arm at Week 20 and also at Week 20 of Part B, when both participants and raters knew that all participants were receiving active treatment (Fig. 6). While some expectancy influence is plausible under these conditions, ADCOMS was explicitly designed to minimise measurement noise and maximise sensitivity to underlying disease progression [52], so unblinded knowledge of active treatment would be expected to have only a modest impact on its values; however, the literature does not provide a precise quantitative estimate of such an effect. To assess (in a non-prespecified post hoc analysis) whether the later semi‑blinded phase materially influenced the observed long‑term trajectory, we compared annualised ADCOMS slopes for the combined 125 µg + 250 µg group derived from weighted linear regressions of mean ADCOMS over time during the double‑blind, placebo‑controlled Part A1 (from screening at -9 weeks to Week 42, spanning ∼51 weeks) and over the full ∼105‑week period (Parts A1 and B), using the number of participants at each visit as weights to approximate frequency‑weighted least squares and to obtain the slopes. The estimated annual ADCOMS increase ± SE was 0.0454 ± 0.0043 points during Part A1 and 0.0267 ± 0.0098 points over the full 0–105‑week period; although the full‑period slope was numerically flatter, the difference between the two estimates (0.0187 ADCOMS points/year, SE 0.0107; z ∼ 1.7; P = 0.09) did not reach conventional statistical significance, arguing against a strong long-term placebo effect in this data set.
A non-prespecified exploratory comparison with historical controls from published ADNI data ([64]; Fig. 7) was undertaken to provide context for longer‑term ADCOMS changes under ALZ‑101 treatment. At baseline, the pooled ALZ‑101 125 µg and 250 µg groups and the ADNI AD cohort showed closely similar cognitive severity on ADCOMS and MMSE: mean ADCOMS 0.547 vs. 0.574 (median 0.569), and MMSE ∼24.1 vs. 23.2, respectively. ADAS‑Cog13 was somewhat higher (worse) in ALZ‑C‑001 (∼35 vs. 30.3), while CDR‑SB indicated somewhat greater functional impairment in ADNI (4.42 vs. ∼3.1), and ADNI participants were on average older (∼75 vs. ∼70 years). The LMM fitted to ALZ‑C‑001 yielded an estimated annual ADCOMS change of 0.024 points (95% CI − 0.007 to 0.054; P = 0.10) in the pooled ALZ‑101 125 and 250 µg group, indicating no statistically significant progression over the observed time frame. In contrast, the GLS model fitted to ADNI AD mean ADCOMS values produced an estimated annual change of 0.166 points (95% CI 0.144 to 0.188; P < 0.0001), consistent with robust disease progression over 24 months. The between‑group difference in annual slopes (ALZ‑101 minus ADNI AD) was − 0.141 (95% CI − 0.178 to − 0.105; P < 0.0001), indicating a markedly slower rate of ADCOMS worsening in the ALZ‑101 group compared with the historical ADNI AD trajectory (Supplementary Fig. 12; Supplementary Table 9). These results suggest that, despite similar baseline composite severity, patients receiving ALZ‑101 may have experienced slower ADCOMS worsening than expected from the ADNI mild‑AD natural‑history trajectory; however, this external comparison is exploratory and should be interpreted with caution.
Exploratory, non-prespecified correlation analyses between changes from baseline CSF biomarkers and ADCOMS scores were weak and inconsistent (|r| ∼ 0.3) in this small sample (data not shown), in line with the limited power and known instability of correlation estimates at this sample size [66].
Exploratory T-cell stimulations
T-cell activation induced by ALZ-101 in PBMCs show that stimulation with Aβ peptide fragments resulted in low levels of response compared to anti-CD3 positive controls. For IL-4, a tendency for increased responses was observed with ALZ-101 compared to placebo, while IFN-γ responses were more randomly distributed between ALZ-101 and placebo (Supplementary Fig. 13).
Discussion
The primary objective of this first-in-human study was to evaluate the safety and tolerability of the novel Aβ–targeted vaccine ALZ-101 in participants with early AD. ALZ-101 is comprised of a protein-engineered oligomeric Aβ42 antigen, Aβ42CC [43]. Non-clinical studies using the monoclonal antibody ALZ-201, derived from ALZ-101-vaccinated mice, confirmed that the Aβ42 oligomers specifically targeted by both ALZ-101 and ALZ-201 are disease-relevant in AD brain tissue [28]. The antigen is formulated with aluminium hydroxide and administered intramuscularly. To our knowledge, this represents the first clinical evaluation of a vaccine targeting oligomer-specific, non-generic epitopes on Aβ in early AD.
Beyond establishing safety and immunogenicity, our study results should be interpreted in the context of converging non-clinical evidence that Aβ oligomers are key pathogenic drivers in AD and represent an attractive therapeutic target (reviewed in [20, 21]), but as discussed in the Introduction, there is currently no clear precedent for how oligomer-specific disease modification should manifest in patients. Non-clinical studies suggest that most pathogenic Aβ species in human brain are rare, diffusible oligomers that differ qualitatively from bulk synthetic assemblies and plaque‑bound aggregates [22, 23, 27], concentrating neurotoxicity into a very small fraction of total Aβ that is likely not effectively targeted by current immunotherapy treatments raised against the N-terminal of Aβ [28]. The only late-stage clinical trial results available for a compound specifically addressing oligomer toxicity in early AD individuals is on valiltramiprosate, an oral small-molecule Aβ oligomerisation inhibitor. In the Phase 3 APOLLOE4 trial in APOE ε4/ε4 participants with early AD, valiltramiprosate did not significantly slow clinical decline in the overall population or in individuals with mild AD, but significantly reduced hippocampal atrophy overall (by 18%) and showed greater slowing of hippocampal (by 26%) and cortical atrophy in the MCI subgroup, where nominally significant clinical benefits were also observed [34]. Valiltramiprosate was also not associated with an increased risk of ARIA compared with placebo in this trial. Compared with most Aβ targeting strategies, valiltramiprosate thus shows a contrasting ARIA and MRI‑volume profile in early AD: A systematic review and meta-analysis of Aβ targeting therapies concluded that secretase inhibitors (that reduce Aβ production) consistently show increased hippocampal and whole‑brain atrophy (respectively by ∼19.6% and ∼12.8% over placebo), ARIA‑inducing monoclonal antibodies (e.g. plaque-targeting antibodies aducanumab, lecanemab, donanemab, gantenerumab, bapineuzumab) show increased ventricular enlargement (by ∼38.7% over placebo) with a strong correlation between ARIA frequency and ventricular expansion (r∼0.86), and no anti‑Aβ drug class in this analysis clearly reduced brain atrophy compared to placebo [67]. Some of these therapies (aducanumab, lecanemab, and donanemab) nonetheless have significant beneficial effects on the rate of cognitive decline in early AD individuals [12–14] whereas the potential efficacy of valiltramiprosate currently appears confined to APOE ε4/ε4 individuals at the MCI stage of early AD, and these signals have not yet been corroborated by a positive primary clinical endpoint in a confirmatory Phase 3 trial [34].
ALZ-101 showed an acceptable safety profile in this elderly AD patient population, who typically presents multiple co-morbidities. No deaths or serious drug reactions were attributed to ALZ-101, and no critical safety signals emerged during the study. Additionally, there were no clinically meaningful differences identified in clinical laboratory tests, vital signs, or ECGs.
IREs and ARIAs were classified as AESIs. Throughout all study phases (A1, A2, and B), 170 AESIs were reported in 28 participants, representing 88% of the study population. The majority of these AESIs were either systemic or related to local injection site reactions (ISRs). ISRs occurred in most participants and were mainly injection site pain, redness (erythema), and warmth. Although minor differences were noted in the frequency and severity of these local reactions between treatment groups, no clinically meaningful differences were observed, indicating that blinding of the study medication was maintained. The most frequent systemic injection-related events included fatigue and headache.
The MRI protocol in ALZ‑C‑001 was designed to match the detection sensitivity of ARIA surveillance recommended for anti‑amyloid monoclonal antibody trials [47]. We used 3T MRI exclusively, high‑resolution 3D FLAIR as the primary ARIA‑E sequence, and susceptibility‑sensitive T2*‑weighted imaging with high‑resolution T2* gradient-recalled echo (GRE) at 3T and susceptibility‑weighted angiography on GE systems for ARIA‑H, along with centralised neuroradiology reads and strict eligibility cut‑offs for baseline microhaemorrhages (> 4) and presence of superficial siderosis. We recognise that susceptibility‑weighted imaging is more sensitive than conventional 2D T2* GRE for detecting small hemosiderin deposits and cerebral microbleeds [68], and that use of such a sequence might substantially increase apparent ARIA‑H severity [69]. However, all pivotal anti‑amyloid monoclonal antibody trials and the ARIA‑H grading scheme [70] derived from them are based on 2D T2* GRE, and GRE/T2* is considered adequate for ARIA‑H surveillance in this context.
Sixteen asymptomatic ARIA-H events — including two superficial siderosis events in the same participant — were detected in nine participants (Table 3). Importantly, seven participants had ARIA-H lesions at baseline across the entire study population of whom six were randomised to the 125 µg group and one to the 250 µg group (Table 1), and all seven experienced at least one additional ARIA-H event during the trial irrespective of APOE genotype. The remaining two participants who developed ARIA-H lesions had no ARIA-H detected at baseline but developed lesions during Part A1; notably, both of these participants were in the placebo group during Part A1. The occurrence of ARIA-H in this study aligns with reported data in the literature. ARIA-H microhaemorrhages occur in asymptomatic individuals in the general population, with prevalence increasing with age and especially among APOE ε4 carriers [71]. Compared to healthy elderly control individuals, brain microhaemorrhages are more frequent in individuals with MCI (20–43%) and AD (18–32%) [72–75]. They also are more prevalent in individuals with progressive MCI (31–54%) versus those with stable MCI (36%), possibly reflecting more advanced small vessel or AD-related pathology [76]. In the overall ALZ-C-001 study population, 22 participants (69%) were APOE ε4 carriers (12 ε3/ε4, 10 ε4/ε4).
Background incidence rates of ARIA-H in similar patient populations as those in the current study can be estimated from large clinical trials with comparable MRI monitoring. Reported annual ARIA-H microhaemorrhage incidence rates for placebo cohorts vary by study, ranging from approximately 3.3% to 8.6% per year in donanemab Phase 2 and Phase 3 studies, respectively [13, 77], 5.1% in a lecanemab Phase 3 study [14], about 5.5% in gantenerumab Phase 3 studies [78], and roughly 4.5% in aducanumab Phase 3 studies [12]. In study ALZ-C-001, six individuals were monitored with MRI for 20 weeks, three for 42 weeks, and 23 for an average of 104 weeks, equaling a total of 50.8 person‑years. Using a Poisson model with the literature benchmark range of 3.3%-8.6% annual incidence (mean 5.4%, SD = 2.07%), expected rates across 50.8 person-years range from 1.67 to 4.37, with the mean at 2.74, falling short of the observed 9 cases in ALZ-C-001. Even a 95% upper prediction interval (~ 9.2%; 4.67 expected cases) falls short, indicating that the observed rate exceeds the literature distribution (one-sided P < 0.01 at mean; P∼0.11 at upper range). Given the small sample size of this study, it is not possible to definitively conclude whether ALZ-101 significantly affects ARIA-H risk. However, baseline ARIA-H lesions are known to significantly increase the risk of subsequent ARIA-H events, reflecting an inherent susceptibility linked to underlying cerebrovascular amyloid pathology. In this context, the observation that ARIA-H events in ALZ-101-treated participants occurred exclusively in participants with pre-existing lesions, as well as typically occurring late in the study, supports interpreting these occurrences as reflecting a predisposition rather than a novel drug-induced effect. However, treatment-mediated amplification of pre-existing cerebrovascular vulnerability remains a possibility requiring larger studies with stratified analyses to resolve.
One asymptomatic ARIA-E event occurred in Part B, in a participant randomised to placebo in Part A1, after receiving four doses of 250 µg ALZ-101. The ARIA-E decreased in size in the next two MRI scans to resolve completely by approximately 12 weeks. In placebo groups of clinical trials on early AD individuals, the average annual incidence of ARIA-E has been consistently low (approximately 1.3% (± 0.5 SD)) and in line with the observations made in this study [12–14, 77, 78]. Based on a total MRI follow-up of 50.8 person-years in the ALZ-C-001 trial, the Poisson-modeled probability of observing at least one ARIA-E event at an annual incidence of 1.3% (SD = 0.5%; range 0.8–1.8%) is 48% (range 33–60%). The single observed ARIA-E (at 104 weeks), yielding an empirical annual rate of ~ 2.0%, is thus plausibly attributable to chance alone.
Results obtained with a validated ELISA method, using the antigen for capture and Protein G for detection, indicated that naturally occurring IgG antibodies against the Aβ42CC oligomer were commonly present in drug-naïve individuals, with higher levels observed in healthy elderly individuals than in younger individuals. Signals in healthy elderly individuals (mean age 66.4 ± 3.6 years) were on average 1.6-fold greater (P = 0.072) than those in healthy young individuals (mean age 26.7 ± 1.5 years) and, notably, 1.8-fold greater (P = 0.055) than those observed in participants of this study (mean age 68.6 ± 6.4 years).
Research on naturally occurring antibodies in individuals with AD spans several decades, yet results have been contradictory. Some studies report lower antibody levels in AD compared to controls [79–81], others find similar levels [82–84], and yet others observe higher antibody concentrations [85–88]. Notably, one study [89] reported that antibody levels directed against a linear N-terminal fragment of Aβ increase progressively in AD, whereas levels against a linear mid-domain Aβ fragment decrease. We hypothesise that these inconsistent findings mainly stem from the use of Aβ peptide assemblies or fragments that do not accurately present the toxic Aβ conformation involved in the disease, whereas ex vivo studies strongly suggest that the Aβ42CC oligomers used in this study do present a very toxic Aβ conformation [28].
Based on converging observations from this study and prior work on naturally occurring anti‑Aβ antibodies — including the finding that mid‑domain antibodies decline with increasing amyloid burden and cognitive impairment [89] — we speculate that the high baseline IgG signals against the Aβ42CC oligomer in healthy elderly individuals may reflect an age‑related increase in the overall burden of oligomeric Aβ42 in the brain, while recognizing that only a subset of these oligomers is likely to be neurotoxic [27, 28]. Given the presence of meningeal immune cells and lymphatic drainage from the central nervous system (CNS) to cervical lymph nodes [90], there exists a pathway for CNS antigenic components, such as oligomeric Aβ42, to reach peripheral immune sites and stimulate natural protective antibody production. This likely involves innate-like B-1 cells, although a contribution from adaptive B-2 cells cannot be excluded. The observation that these antibodies are of the IgG isotype, with relatively greater blood-brain-barrier permeability compared to IgM, combined with their unexpectedly high levels indeed supports a functional role within the CNS. Defective functioning of this natural antibody response against neurotoxic Aβ42 could therefore increase the risk of developing AD.
Pre-existing antibodies — including those that occur naturally — can exert both inhibitory and enhancing effects on B cell responses via mechanisms such as epitope masking, complement activation, and Fc receptor engagement, collectively shaping the magnitude, breadth, and affinity of humoral immunity to vaccination [91]. Importantly, no reduction in ALZ-101 efficacy via epitope masking or enhanced antigen clearance was observed in this study, as ALZ-101, formulated with 0.3% aluminium hydroxide and administered intramuscularly, successfully induced an adaptive immune response against oligomeric Aβ42CC. While no evidence of epitope masking or enhanced antigen clearance by pre-existing antibodies was observed, the presence of innate anti-idiotype antibodies prior to vaccination cannot be excluded by antigen-binding ELISA alone and could theoretically contribute to underestimation of vaccine-induced responses. Nonetheless, we found that vaccination produced robust and statistically significant immune responses as demonstrated by significant increases in anti-ALZ-101 IgG levels (Fig. 3). The robust increases in IgG levels were consistent with the pharmacological profile of ALZ-101 and findings from preclinical studies. IgG antibody titres typically appeared after three doses of ALZ-101 given 4 weeks apart. In Part B, booster doses given to participants vaccinated in Part A1 generated rapid and sustained IgG production, suggesting formation of memory B-cells. Antibodies were detected in most participants as late as 52 weeks from the last booster dose in Part B.
Of the 32 enrolled participants, 31 received the study drug (one placebo participant in Part A1 withdrew before entering Part B), with 29 meeting the predefined responder criteria. Two placebo-treated participants fulfilled IgG responder criteria at 14 and 26 weeks after their last administration of aluminium hydroxide in Part A1, indicating that these were false positive signals. Only two participants failed to meet the responder criteria in response to receiving active treatment: one placebo-treated participant who had a false positive signal in Part A1 failed to develop titres in Part B, and one participant in the 400 µg cohort (Part A2). Additionally, one participant dosed with 125 µg in Part A1 and initially categorised as a responder, did not develop titres during Part B, suggesting that the titres observed in Part A1 were false positives. Overall, three participants were thus ultimately classified as non‑responders: 1/6 (17%) in the 400 µg group, 1/15 (7%) at the 250 µg dose level (one placebo‑treated individual who failed to develop titres after receiving 250 µg in Part B), and 1/10 (10%) at the 125 µg dose level (a participant whose apparent response in Part A1 was later judged to be a false positive, as no titres were observed upon re‑dosing in Part B). The occurrence of a small fraction of false positive titres may be attributed to chance or to the intrinsic immunostimulatory properties of aluminium hydroxide. Importantly, these isolated false positives do not affect the conclusion that ALZ-101 induces an immune response in a majority of treated individuals, resulting in a high overall responder rate in the order of 90%.
Serum IgA appeared earlier in the vaccination cycle, as did IgM, with titres very low in comparison with IgG. For IgM, positive titres were observed in five drug-naïve participants. For an intramuscular vaccine, the observation of low IgA and IgM levels combined with a strong IgG response aligns with typical vaccine-induced immune dynamics, reflecting a predominantly systemic immune focus rather than a mucosal one.
That the ALZ-101 vaccine induces a humoral immune response dominated by IgG antibodies was supported by PBMC re-stimulation experiments, which showed a trend toward increased IL-4 responses, indicative of Th2 activity, while IFNγ responses were variable and inconsistent across the ALZ-101 and placebo treatment groups. It should be noted that both types of T helper cells, Th1 and Th2, are involved in a typical immune response but that ELISpot assays measuring IFNγ often display greater sensitivity than those measuring IL-4, with IFNγ-producing cells being more readily detected upon stimulation. Since IL-4 assay sensitivity or dynamic range is limited, this may underestimate Th2 activity. In the IFNγ ELISpot assay, the anti-CD3 positive control yielded spot counts that were too numerous to quantify in several samples, including at baseline. Taken together, the analytical PBMC results, the high IgG titres, and the choice of adjuvant (aluminium hydroxide) all point to a low risk of inducing a detrimental Th1 cellular immune response following ALZ-101 vaccination.
Examination of cognitive and functional assessments was exploratory, and rather few statistically significant findings were observed which were not consistently replicated across scales or doses as expected given the Phase 1 trial design, with a small sample size and limited statistical power for efficacy. Across ADAS‑Cog13, ADCS‑ADL, CDR‑Global, NPI, and NTB, mean baseline values were balanced between groups and longitudinal changes were small, variable, and without consistent separation between placebo and ALZ‑101 dose levels at any time point in Part A1 and throughout Part B (Supplementary Tables 7 and 8). ADAS‑Cog13 showed only a few points of worsening on average, ADCS‑ADL declined modestly, global CDR scores remained near baseline, and NPI and NTB scores fluctuated only slightly around baseline, indicating slow disease progression but no robust or sustained treatment signal on these measures. However, no substantial worsening of cognitive or functional outcomes was observed in vaccinated participants over the duration of the study in the ADCOMS score (Fig. 6), which is a composite outcome measure specifically developed to detect early-stage cognitive and functional changes in early AD (MCI due to AD and mild AD) [52] and whose trajectory was directionally consistent with CDR-SB but appeared more favourable, in line with its greater sensitivity and lower noise in early AD [52, 64]. Post-hoc sensitivity analyses based on weighted linear regressions of mean ADCOMS over the placebo-controlled Part A1 and the full study also suggested that the slow long‑term ADCOMS progression observed during Part B is unlikely to be explained solely by expectancy effects arising after unblinding in Part B.
The absence of a dedicated placebo group for the entire duration of the study limits the ability to draw definitive conclusions regarding putative disease modification. In addition to the randomised within‑trial comparisons, we therefore compared long‑term ADCOMS trajectories with an external natural‑history reference from a mild‑AD ADNI cohort [64]. In this post hoc, non‑randomised analysis, the pooled 125 µg and 250 µg ALZ‑101 group had an estimated annual ADCOMS slope of 0.024 (95% CI − 0.007 to 0.054), whereas the ADNI cohort showed an estimated slope of 0.166 (95% CI 0.144 to 0.188), corresponding to a slope difference of − 0.141 (95% CI − 0.178 to − 0.105; P < 0.0001). This pattern is compatible with slower clinical worsening under ALZ‑101 than in the ADNI mild‑AD cohort, but the comparison relies on aggregate ADNI data, lacks individual‑level matching and multiplicity control, and was not prespecified; it should therefore be regarded as hypothesis‑generating only and cannot be taken as evidence of efficacy.
No statistically significant treatment-associated changes were observed in the CSF biomarker analyses. CSF was sampled at screening (approximately 9 weeks before the first dose) and at Week 20, so the Part A change interval spans roughly 29 weeks, of which only about 20 weeks are on randomised treatment, and any biological effect would be expected to emerge gradually after serological titres develop (around Weeks 4–8). Within this time frame, the first on‑study CSF assessment in Part A is therefore likely to be substantially influenced by the underlying baseline trajectory rather than by any pharmacodynamic effect. By contrast, the CSF sampling at around Week 84 in Part B provides a longer observation window during which a putative disease‑modifying effect might be expected to manifest as a flatter biomarker trajectory in groups that received active treatment in Part A. Interpretation is, however, limited by the small sample sizes, particularly in participants initially randomised to placebo (n = 6 in Part A1; n = 5 in Part B).
As expected, the Aβ42/Aβ40 ratio was unaffected by ALZ-101 vaccination in Part A and Part B. In addition, ALZ-101 treatment did not induce any neuroinflammatory reactions in the CNS as measured by YKL-40, reflecting glial activation in the brain, and sTREM-2, a marker of microglial activation, at any of the time points.
For the AD biomarkers P-tau181 and T-tau, the active dose groups tended to show lower rates of progression at the Part A1 time point compared with placebo, with the largest numerical increases observed in the placebo group, intermediate changes at 125 µg, and the smallest at 250 µg; this pattern appeared to partly extend to the open-label 400 µg arm (Fig. 5). Absolute changes remained small overall. In Part B, mean changes from baseline were largely stable compared with Part A1 values, with no evidence of progressive increases and small between-period differences (typically on the order of a few pg/mL). These trajectories are consistent with relative stabilization — or modest blunting — of CSF P-tau181 and T-tau early in the study that remained during continued ALZ-101 exposure. Neurogranin and SNAP-25, in contrast, showed only small mean changes with large within-group variability and no consistent dose- or treatment-related pattern across Part A1 and Part B (Fig. 5). The implications of these observations in the context of disease modification are not yet clear. Once AD pathology is established and patients are in the symptomatic phase, the spontaneous longitudinal dynamic range of CSF T‑tau and P‑tau181 over typical trial durations appears relatively small which may limit the sensitivity of these measures as progression endpoints in mild AD even though absolute levels remain clearly abnormal and diagnostically informative. In a recent study on mild AD–spectrum populations, T‑tau increased on the order of roughly 25–30 pg/mL per year and P‑tau181 increased about 3–4 pg/mL per year [92], whereas in our ALZ‑101 trial mean paired changes in T‑tau and P‑tau181 from Part A1 to Part B (spanning ∼55 weeks) were close to zero in the active‑treatment groups, suggesting a potential deviation from the progressive natural‑history slopes — particularly for T‑tau — that warrants confirmation in a larger, prospectively powered trial.
CSF P‑tau181 and T‑tau biomarkers nevertheless remain useful pharmacodynamic readouts of plaque‑removing agents such as aducanumab [12] and lecanemab [14], which in clinical studies have consistently reduced CSF P‑tau181 and T‑tau and have also been reported to lower CSF neurogranin [14]. However, despite these biomarker shifts, the overall impact of plaque‑removing therapies on clinical disease progression has been modest, suggesting that acute changes in CSF tau species and neurogranin may largely reflect rapid plaque clearance — possibly mediated by microglial phagocytosis of plaque‑associated dystrophic neurites, as supported by neuropathological observations in lecanemab‑treated brain tissue [93] — rather than a rapid upstream effect on the rate of neurodegeneration. Conversely, experimental kinetic studies have shown that the production rate of soluble tau in human CSF is positively correlated with amyloid plaque burden, indicating that amyloid plaques are key drivers of soluble tau production [94]. Therapies that directly remove plaques would thus be expected to rapidly reduce soluble tau output. By contrast, a non‑plaque‑targeting therapy such as ALZ‑101 would be anticipated to influence tau production only indirectly and over longer timescales, leading to more gradual and modest changes in CSF T‑tau and P‑tau181 rather than the acute tau biomarker drops observed with plaque‑clearing antibodies.
In addition, plaque‑removing agents have generally not demonstrated robust or consistent effects on CSF or plasma NfL, a marker of neuroaxonal injury, despite clear effects on amyloid and tau biomarkers [14, 95, 96]. In contrast, ALZ-101 treatment appeared to stabilise CSF NfL levels in the extension phase for those exposed to study drug in the main study but not for those randomised to receive placebo injections (Fig. 4). The rate of neuronal damage, as reflected by NfL concentrations, thus appears to be slower in participants vaccinated with ALZ-101 early in the study compared to those initially on placebo, for which exposure to the study drug was limited to the extension phase only. These data support a mechanism of action of ALZ-101 that is distinct from plaque-lowering agents, consistent with evidence that the neurotoxicity of brain-derived Aβ is driven predominantly by its soluble oligomeric forms rather than by plaques.
Taken together, CSF P‑tau181, T‑tau, and NfL all showed numerically stable or slightly attenuated trajectories over time rather than progressive increases, and a similar pattern was observed for ADCOMS, such that both biomarker and cognitive/functional readouts pointed in the same direction — toward an absence of detectable disease worsening under ALZ‑101 exposure.
Limitations
Limitations of this study include the small sample size; the study was not powered to detect statistically significant effects on biomarkers or on cognitive/functional outcomes at individual dose levels. Additionally, the relatively short duration of follow-up for many participants may have limited the ability to observe significant changes. Future studies are needed in the target population, with longer placebo-controlled durations and sufficient power to detect clinically meaningful and statistically significant effects on relevant clinical endpoints.
Conclusions
This study demonstrates that ALZ-101, a novel vaccine candidate based on a stabilised Aβ oligomer immunogen utilizing Aβ42CC peptide technology, showed no clear safety signals and induced sustained antibody production specifically targeting oligomeric Aβ species in participants with early AD. The high responder rate of 90% supports the immunogenic robustness of ALZ-101 and suggests potential generalizability across broader patient populations. Notably, high levels of similar protective IgG antibodies are naturally present in healthy elderly individuals but are reduced in AD patients, suggesting a protective role for this antibody subset. Although no significant clinical treatment effects were observed in the short placebo-controlled phase, data from the long-term extension suggest ALZ-101 may stabilise cognitive (ADCOMS) and biomarker outcomes, including CSF NfL, P-tau181, and T-tau. These findings suggest that ALZ-101 may exert a distinct modulatory effect on AD pathology, likely through its Aβ oligomer-specific polyclonal antibody response, which warrants further clinical development of the drug.
Supplementary Information
Acknowledgements
The authors would like to acknowledge the editorial assistance of Richard Perry, PharmD, in the preparation of this manuscript under the direction of the authors, which was supported by Alzinova AB, Mölndal, Sweden. The authors also want to acknowledge the study sites, participants, and caregivers who participated in this study.
Abbreviations
- AChEI
Acetylcholinesterase inhibitor
- AD
Alzheimer’s disease
- ADAS-Cog13
Alzheimer’s Disease Assessment Scale – Cognitive Subscale 13
- ADCOMS
Alzheimer’s Disease Composite Score
- ADCS-ADL
Alzheimer’s Disease Cooperative Study – Activities of Daily Living
- ADCS-CGIC
Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change
- ADNI
Alzheimer’s Disease Neuroimaging Initiative
- AE
Adverse event
- AESI
AE of special interest
- APOE
Apolipoprotein E
- Aβ
Amyloid-β
- ARIA
Amyloid-related imaging abnormalities
- ARIA-E
Amyloid-related imaging abnormalities – edema/effusion
- ARIA-H
Amyloid-related imaging abnormalities – haemorrhage/hemosiderin deposits
- C-SSRS
Columbia-Suicide Severity Rating Scale
- CDR
Clinical Dementia Rating
- CDR-SB
Clinical Dementia Rating – Sum of Boxes
- CI
Confidence interval
- CNS
Central nervous system
- CSF
Cerebrospinal fluid
- DWI
Diffusion-weighted imaging
- ECG
Electrocardiogram
- ELISA
Enzyme-linked immunosorbent assay
- ELISpot
Enzyme-linked immunospot
- FAS
Full analysis set
- FLAIR
Fluid‑attenuated inversion recovery
- GLS
Generalised least‑squares
- GMT
Geometric mean titre
- GRE
Gradient‑recalled echo
- GSD
Geometric standard deviation
- IFNγ
Interferon-γ
- IL-4
Interleukin-4
- IRE
Injection-related event
- ISR
Injection-site reaction
- LMM
Linear mixed‑effects model
- LS
Least squares
- MCI
Mild cognitive impairment
- MMRM
Mixed models for repeated measurements
- MMSE
Mini-Mental State Examination
- MRI
Magnetic resonance imaging
- NfL
Neurofilament light chain
- NPI
Neuropsychiatric Inventory
- NTB
Neuropsychological Test Battery
- PBMC
Peripheral blood mononuclear cell
- P-tau181
Phosphorylated tau protein 181
- P-tau217
Phosphorylated tau protein 217
- SEM
Standard error of the mean
- SNAP-25
Synaptosomal-associated protein 25
- sTREM-2
Soluble triggering receptor expressed on myeloid cells 2
- T-tau
Total tau protein
- YKL-40
Chitinase-3-like protein 1
Authors’ contributions
ZL, AS, SP, NS, MS and JR contributed in the design and interpretation of clinical study results; AS and SP wrote the final manuscript; HZ assisted in biomarker data collection and interpretation; RP established MRI protocols; JR was the clinical study’s principal investigator.
Funding
This study was funded by Alzinova AB, Mölndal, Sweden.
Data availability
Data for this study are reported in the manuscript and Data Supplement. Additional data may be available upon written request to the Sponsor. The protocol and statistical analysis plan are available upon request.
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with international and national regulations governing biomedical research involving human subjects, including the Declaration of Helsinki, Good Clinical Practice, and all applicable Finnish national regulations. Prior to initiation, the study protocol and informed consent materials received approval from the National Committee on Medical Research Ethics (TUKIJA). All participants provided written informed consent before any study procedures were undertaken.
Consent for publication
Not applicable.
Competing interests
HZ has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp & Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave; has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD; is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of CERimmune Therapeutics (outside submitted work).
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Gustavsson A, Norton N, Fast T, Frölich L, Georges J, Holzapfel D, Kirabali T, Krolak-Salmon P, Rossini PM, Ferretti MT, Lanman L, Chadha AS, van der Flier WM. Global estimates on the number of persons across the Alzheimer’s disease continuum. Alzheimers Dement. 2023;19:658–70. [DOI] [PubMed] [Google Scholar]
- 2.Serrano-Pozo A, Frosch MP, Masliah E, Hyman BT. Neuropathological alterations in Alzheimer disease. Cold Spring Harb Perspect Med. 2011;1:a006189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Busche MA, Hyman BT. Synergy between amyloid-β and tau in Alzheimer’s disease. Nat Neurosci. 2020;23:1183–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Shulman M, Kong J, O’Gorman J, Ratti E, Rajagovindan R, Viollet L, Huang E, Sharma S, Racine AM, Czerkowicz J, Graham D, Li Y, Hering H, Haeberlein SB. TANGO: a placebo-controlled randomized phase 2 study of efficacy and safety of the anti-tau monoclonal antibody gosuranemab in early Alzheimer’s disease. Nat Aging. 2023;3:1591–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Florian H, Wang D, Arnold SE, Boada M, Guo Q, Jin Z, Zheng H, Fisseha N, Kalluri HV, Rendenbach-Mueller B, Budur K, Gold M. Tilavonemab in early Alzheimer’s disease: results from a phase 2, randomized, double-blind study. Brain. 2023;146:2275–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Monteiro C, Toth B, Brunstein F, Bobbala A, Datta S, Ceniceros R, Sanabria Bohorquez SM, Anania VG, Wildsmith KR, Schauer SP, Lee J, Dolton MJ, Ramakrishnan V, Abramzon D, Teng E. Lauriet investigators. Randomized Phase II Study of the Safety and Efficacy of Semorinemab in Participants With Mild-to-Moderate Alzheimer Disease: Lauriet. Neurology. 2023;101:e1391–401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fleisher AS, Munsie LM, Perahia DGS, Andersen SW, Higgins IA, Hauck PM, Lo AC, Sims JR, Brys M, Mintun M, PERISCOPE-ALZ Site Investigators. Assessment of Efficacy and Safety of Zagotenemab: Results From PERISCOPE-ALZ, a Phase 2 Study in Early Symptomatic Alzheimer Disease. Neurology. 2024;102:e208061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Novak P, Kovacech B, Katina S, Schmidt R, Scheltens P, Kontsekova E, Ropele S, Fialova L, Kramberger M, Paulenka-Ivanovova N, Smisek M, Hanes J, Stevens E, Kovac A, Sutovsky S, Parrak V, Koson P, Prcina M, Galba J, Cente M, Hromadka T, Filipcik P, Piestansky J, Samcova M, Prenn-Gologranc C, Sivak R, Froelich L, Fresser M, Rakusa M, Harrison J, Hort J, Otto M, Tosun D, Ondrus M, Winblad B, Novak M, Zilka N. ADAMANT: a placebo-controlled randomized phase 2 study of AADvac1, an active immunotherapy against pathological tau in Alzheimer’s disease. Nat Aging. 2021;1:521–34. [DOI] [PubMed] [Google Scholar]
- 9.Kovacech B, Cullen NC, Novak P, Hanes J, Kontsekova E, Katina S, Parrak V, Fresser M, Vanbrabant J, Feldman HH, Winblad B, Stoops E, Vanmechelen E, Zilka N. Post hoc analysis of ADAMANT, a phase 2 clinical trial of active tau immunotherapy with AADvac1 in patients with Alzheimer’s disease, positive for plasma p-tau217. Alzheimers Res Ther. 2024;16(1):254. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Edwards AL, Collins JA, Junge C, Kordasiewicz H, Mignon L, Wu S, Li Y, Lin L, DuBois J, Hutchison RM, Ziogas N, Shulman M, Martarello L, Graham D, Lane R, Budd Haeberlein S, Beaver J. Exploratory Tau Biomarker Results From a Multiple Ascending-Dose Study of BIIB080 in Alzheimer Disease: A Randomized Clinical Trial. JAMA Neurol. 2023;80:1344–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shulman M, Wu S, Ziogas N, Edwards A, Collins J, Lin L, Tien I, Curiale G, Li Y, Mummery C, Lane R, Junge C, Beaver J, Tian Y, Landen J, Bullain S, Gallagher D. Exploratory analyses of clinical outcomes from the BIIB080 phase 1b study in mild Alzheimer’s disease. Nat Aging. 2026;6:445–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Budd Haeberlein S, Aisen PS, Barkhof F, Chalkias S, Chen T, Cohen S, Dent G, Hansson O, Harrison K, von Hehn C, Iwatsubo T, Mallinckrodt C, Mummery CJ, Muralidharan KK, Nestorov I, Nisenbaum L, Rajagovindan R, Skordos L, Tian Y, van Dyck CH, Vellas B, Wu S, Zhu Y, Sandrock A. Two Randomized Phase 3 Studies of Aducanumab in Early Alzheimer’s Disease. J Prev Alzheimers Dis. 2022;9:197–210. [DOI] [PubMed] [Google Scholar]
- 13.Sims JR, Zimmer JA, Evans CD, Lu M, Ardayfio P, Sparks J, Wessels AM, Shcherbinin S, Wang H, Monkul Nery ES, Collins EC, Solomon P, Salloway S, Apostolova LG, Hansson O, Ritchie C, Brooks DA, Mintun M, Skovronsky DM. Donanemab in Early Symptomatic Alzheimer Disease: The TRAILBLAZER-ALZ 2 Randomized Clinical Trial. JAMA. 2023;330:512–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.van Dyck CH, Swanson CJ, Aisen P, Bateman RJ, Chen C, Gee M, Kanekiyo M, Li D, Reyderman L, Cohen S, Froelich L, Katayama S, Sabbagh M, Vellas B, Watson D, Dhadda S, Irizarry M, Kramer LD, Iwatsubo T. Lecanemab in Early Alzheimer’s Disease. N Engl J Med. 2023;388:9–21. [DOI] [PubMed] [Google Scholar]
- 15.Daly T, Kepp KP, Imbimbo BP. Are lecanemab and donanemab disease-modifying therapies? Alzheimers Dement. 2024;20:6659–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Espay AJ, Kepp KP, Herrup K. Lecanemab and Donanemab as Therapies for Alzheimer’s Disease: An Illustrated Perspective on the Data. eNeuro. 2024;11:ENEURO0319–232024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Liu KY, Schneider LS, Howard R. The need to show minimum clinically important differences in Alzheimer’s disease trials. Lancet Psychiatry. 2021;8:1013–6. [DOI] [PubMed] [Google Scholar]
- 18.Barakos J, Purcell D, Suhy J, Chalkias S, Burkett P, Marsica Grassi C, Castrillo-Viguera C, Rubino I, Vijverberg E. Detection and Management of Amyloid-Related Imaging Abnormalities in Patients with Alzheimer’s Disease Treated with Anti-Amyloid Beta Therapy. J Prev Alzheimers Dis. 2022;9:211–20. [DOI] [PubMed] [Google Scholar]
- 19.Salloway S, Wojtowicz J, Voyle N, Lane CA, Klein G, Lyons M, Rossomanno S, Mazzo F, Bullain S, Barkhof F, Bittner T, Schneider A, Grundman M, Aldea R, Boada M, Smith J, Doody R. Amyloid-Related Imaging Abnormalities (ARIA) in Clinical Trials of Gantenerumab in Early Alzheimer Disease. JAMA Neurol. 2025;82:19–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Huang YR, Liu RT. The Toxicity and Polymorphism of β-Amyloid Oligomers. Int J Mol Sci. 2020;21:4477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ivanova MI, La Rosa C, Ramamoorthy A. Toxic mechanisms of amyloid oligomers and therapeutic strategies. Protein Sci. 2026;35:e70519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Jin M, Shepardson N, Yang T, Chen G, Walsh D, Selkoe DJ. Soluble amyloid beta-protein dimers isolated from Alzheimer cortex directly induce Tau hyperphosphorylation and neuritic degeneration. Proc Natl Acad Sci U S A. 2011;108:5819–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zott B, Simon MM, Hong W, Unger F, Chen-Engerer HJ, Frosch MP, Sakmann B, Walsh DM, Konnerth A. A vicious cycle of β amyloid-dependent neuronal hyperactivation. Science. 2019;365:559–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Klyubin I, Betts V, Welzel AT, Blennow K, Zetterberg H, Wallin A, Lemere CA, Cullen WK, Peng Y, Wisniewski T, Selkoe DJ, Anwyl R, Walsh DM, Rowan MJ. Amyloid beta protein dimer-containing human CSF disrupts synaptic plasticity: prevention by systemic passive immunization. J Neurosci. 2008;28:4231–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Shankar GM, Li S, Mehta TH, Garcia-Munoz A, Shepardson NE, Smith I, Brett FM, Farrell MA, Rowan MJ, Lemere CA, Regan CM, Walsh DM, Sabatini BL, Selkoe DJ. Amyloid-beta protein dimers isolated directly from Alzheimer’s brains impair synaptic plasticity and memory. Nat Med. 2008;14:837–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Yang T, Li S, Xu H, Walsh DM, Selkoe DJ. Large Soluble Oligomers of Amyloid β-Protein from Alzheimer Brain Are Far Less Neuroactive Than the Smaller Oligomers to Which They Dissociate. J Neurosci. 2017;37:152–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hong W, Wang Z, Liu W, O’Malley TT, Jin M, Willem M, Haass C, Frosch MP, Walsh DM. Diffusible, highly bioactive oligomers represent a critical minority of soluble Aβ in Alzheimer’s disease brain. Acta Neuropathol. 2018;136:19–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Sandberg A, Berenjeno-Correa E, Rodriguez RC, Axenhus M, Weiss SS, Batenburg K, Hoozemans JJM, Tjernberg LO, Scheper W. Aβ42 oligomer-specific antibody ALZ-201 reduces the neurotoxicity of Alzheimer’s disease brain extracts. Alzheimers Res Ther. 2022;14:196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Englund H, Sehlin D, Johansson AS, Nilsson LN, Gellerfors P, Paulie S, Lannfelt L, Pettersson FE. Sensitive ELISA detection of amyloid-beta protofibrils in biological samples. J Neurochem. 2007;103:334–45. [DOI] [PubMed] [Google Scholar]
- 30.Sevigny J, Chiao P, Bussière T, Weinreb PH, Williams L, Maier M, Dunstan R, Salloway S, Chen T, Ling Y, O’Gorman J, Qian F, Arastu M, Li M, Chollate S, Brennan MS, Quintero-Monzon O, Scannevin RH, Arnold HM, Engber T, Rhodes K, Ferrero J, Hang Y, Mikulskis A, Grimm J, Hock C, Nitsch RM, Sandrock A. The antibody aducanumab reduces Aβ plaques in Alzheimer’s disease. Nature. 2016;537:50–6. [DOI] [PubMed] [Google Scholar]
- 31.Savage MJ, Kalinina J, Wolfe A, Tugusheva K, Korn R, Cash-Mason T, Maxwell JW, Hatcher NG, Haugabook SJ, Wu G, Howell BJ, Renger JJ, Shughrue PJ, McCampbell A. A sensitive aβ oligomer assay discriminates Alzheimer’s and aged control cerebrospinal fluid. J Neurosci. 2014;34:2884–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Williams AD, Sega M, Chen M, Kheterpal I, Geva M, Berthelier V, Kaleta DT, Cook KD, Wetzel R. Structural properties of Abeta protofibrils stabilized by a small molecule. Proc Natl Acad Sci U S A. 2005;102:7115–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Au DF, Ostrovsky D, Fu R, Vugmeyster L. Solid-state NMR reveals a comprehensive view of the dynamics of the flexible, disordered N-terminal domain of amyloid-β fibrils. J Biol Chem. 2019;294:5840–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Abushakra S, Power A, Watson D, Porsteinsson A, Sabbagh M, MacSweeney E, Cohen S, Boada Rovira M, Doraiswamy PM, Liang E, Flint S, Kesslak JP, McLaine R, Albayrak A, Schaefer J, Yu J, Tolar L, Dickson S, Hey JA, Tolar M. Clinical Efficacy, Safety and Imaging Effects of Oral Valiltramiprosate in APOEε4/ε4 Homozygotes with Early Alzheimer’s Disease: Results of the Phase III, Randomized, Double-Blind, Placebo-Controlled, 78-Week APOLLOE4 Trial. Drugs. 2025;85:1455–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Gilman S, Koller M, Black RS, Jenkins L, Griffith SG, Fox NC, Eisner L, Kirby L, Rovira MB, Forette F, Orgogozo JM. AN1792(QS-21)-201 Study Team. Clinical effects of Abeta immunization (AN1792) in patients with AD in an interrupted trial. Neurology. 2005;64:1553–62. [DOI] [PubMed] [Google Scholar]
- 36.Pride M, Seubert P, Grundman M, Hagen M, Eldridge J, Black RS. Progress in the active immunotherapeutic approach to Alzheimer’s disease: clinical investigations into AN1792-associated meningoencephalitis. Neurodegener Dis. 2008;5:194–6. [DOI] [PubMed] [Google Scholar]
- 37.Triplett O, Varda N, Decourt B, Vasconcellos R, Sabbagh MN. Active Immunization Targeting Amyloid β for the Treatment of Alzheimer’s Disease. Neurodegener Dis. 2025;25(4):202–17. doi: 10.1159/000546287. Epub 2025 May 10. Erratum in: Neurodegener Dis. 2025;25:228. [DOI] [PMC free article] [PubMed]
- 38.Lee M, Bard F, Johnson-Wood K, Lee C, Hu K, Griffith SG, Black RS, Schenk D, Seubert P. Abeta42 immunization in Alzheimer’s disease generates Abeta N-terminal antibodies. Ann Neurol. 2005;58:430–5. [DOI] [PubMed] [Google Scholar]
- 39.Dalgediene I, Lasickiene R, Budvytyte R, Valincius G, Morkuniene R, Borutaite V, Zvirbliene A. Immunogenic properties of amyloid beta oligomers. J Biomed Sci. 2013;20:10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Farlow MR, Andreasen N, Riviere ME, Vostiar I, Vitaliti A, Sovago J, Caputo A, Winblad B, Graf A. Long-term treatment with active Aβ immunotherapy with CAD106 in mild Alzheimer’s disease. Alzheimers Res Ther. 2015;7:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Hull M, Sadowsky C, Arai H, Le Prince Leterme G, Holstein A, Booth K, Peng Y, Yoshiyama T, Suzuki H, Ketter N, Liu E, Ryan JM. Long-Term Extensions of Randomized Vaccination Trials of ACC-001 and QS-21 in Mild to Moderate Alzheimer’s Disease. Curr Alzheimer Res. 2017;14:696–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Yu HJ, Dickson SP, Wang PN, Chiu MJ, Huang CC, Chang CC, Liu H, Hendrix SB, Dodart JC, Verma A, Wang CY, Cummings J. Safety, tolerability, immunogenicity, and efficacy of UB-311 in participants with mild Alzheimer’s disease: a randomised, double-blind, placebo-controlled, phase 2a study. EBioMedicine. 2023;94:104665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sandberg A, Luheshi LM, Söllvander S, Pereira de Barros T, Macao B, Knowles TP, Biverstål H, Lendel C, Ekholm-Petterson F, Dubnovitsky A, Lannfelt L, Dobson CM, Härd T. Stabilization of neurotoxic Alzheimer amyloid-beta oligomers by protein engineering. Proc Natl Acad Sci U S A. 2010;107:15595–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Dubnovitsky A, Sandberg A, Rahman MM, Benilova I, Lendel C, Härd T. Amyloid-β protofibrils: size, morphology and synaptotoxicity of an engineered mimic. PLoS ONE. 2013;8:e66101. [DOI] [PMC free article] [PubMed]
- 45.Monsonego A, Zota V, Karni A, Krieger JI, Bar-Or A, Bitan G, Budson AE, Sperling R, Selkoe DJ, Weiner HL. Increased T cell reactivity to amyloid beta protein in older humans and patients with Alzheimer disease. J Clin Invest. 2003;112:415–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR Jr, Kawas CH, Klunk WE, Koroshetz WJ, Manly JJ, Mayeux R, Mohs RC, Morris JC, Rossor MN, Scheltens P, Carrillo MC, Thies B, Weintraub S, Phelps CH. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7:263–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Sperling RA, Jack CR Jr, Black SE, Frosch MP, Greenberg SM, Hyman BT, Scheltens P, Carrillo MC, Thies W, Bednar MM, Black RS, Brashear HR, Grundman M, Siemers ER, Feldman HH, Schindler RJ. Amyloid-related imaging abnormalities in amyloid-modifying therapeutic trials: recommendations from the Alzheimer’s Association Research Roundtable Workgroup. Alzheimers Dement. 2011;7:367–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993;43(11):2412–4. [DOI] [PubMed] [Google Scholar]
- 49.Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98. [DOI] [PubMed] [Google Scholar]
- 50.Posner K, Brown GK, Stanley B, Brent DA, Yershova KV, Oquendo MA, Currier GW, Melvin GA, Greenhill L, Shen S, Mann JJ. The Columbia-Suicide Severity Rating Scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am J Psychiatry. 2011;168:1266–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Schneider LS, Olin JT, Doody RS, Clark CM, Morris JC, Reisberg B, Schmitt FA, Grundman M, Thomas RG, Ferris SH. Validity and reliability of the Alzheimer’s Disease Cooperative Study-Clinical Global Impression of Change. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis Assoc Disord. 1997;11(Suppl 2):S22–32. [DOI] [PubMed] [Google Scholar]
- 52.Wang J, Logovinsky V, Hendrix SB, Stanworth SH, Perdomo C, Xu L, Dhadda S, Do I, Rabe M, Luthman J, Cummings J, Satlin A. ADCOMS: a composite clinical outcome for prodromal Alzheimer’s disease trials. J Neurol Neurosurg Psychiatry. 2016;87:993–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. A new clinical scale for the staging of dementia. Br J Psychiatry. 1982;140:566–72. [DOI] [PubMed] [Google Scholar]
- 54.Kueper JK, Speechley M, Montero-Odasso M. The Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog): Modifications and Responsiveness in Pre-Dementia Populations. A Narrative Review. J Alzheimers Dis. 2018;63:423–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Harrison J, Minassian SL, Jenkins L, Black RS, Koller M, Grundman M. A neuropsychological test battery for use in Alzheimer disease clinical trials. Arch Neurol. 2007;64:1323–9. [DOI] [PubMed] [Google Scholar]
- 56.Galasko D, Bennett D, Sano M, Ernesto C, Thomas R, Grundman M, Ferris S. An inventory to assess activities of daily living for clinical trials in Alzheimer’s disease. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis Assoc Disord. 1997;11(Suppl 2):S33–9. [PubMed] [Google Scholar]
- 57.Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, Gornbein J. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology. 1994;44:2308–14. [DOI] [PubMed] [Google Scholar]
- 58.Gaetani L, Höglund K, Parnetti L, Pujol-Calderon F, Becker B, Eusebi P, Sarchielli P, Calabresi P, Di Filippo M, Zetterberg H, Blennow K. A new enzyme-linked immunosorbent assay for neurofilament light in cerebrospinal fluid: analytical validation and clinical evaluation. Alzheimers Res Ther. 2018;10:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Kvartsberg H, Portelius E, Andreasson U, Brinkmalm G, Hellwig K, Lelental N, Kornhuber J, Hansson O, Minthon L, Spitzer P, Maler JM, Zetterberg H, Blennow K, Lewczuk P. Characterization of the postsynaptic protein neurogranin in paired cerebrospinal fluid and plasma samples from Alzheimer’s disease patients and healthy controls. Alzheimers Res Ther. 2015;7:40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Öhrfelt A, Axelsson M, Malmeström C, Novakova L, Heslegrave A, Blennow K, Lycke J, Zetterberg H. Soluble TREM-2 in cerebrospinal fluid from patients with multiple sclerosis treated with natalizumab or mitoxantrone. Mult Scler. 2016;22:1587–95. [DOI] [PubMed] [Google Scholar]
- 61.Gobom J, Parnetti L, Rosa-Neto P, Vyhnalek M, Gauthier S, Cataldi S, Lerch O, Laczo J, Cechova K, Clarin M, Benet AL, Pascoal TA, Rahmouni N, Vandijck M, Huyck E, Le Bastard N, Stevenson J, Chamoun M, Alcolea D, Lleó A, Andreasson U, Verbeek MM, Bellomo G, Rinaldi R, Ashton NJ, Zetterberg H, Sheardova K, Hort J, Blennow K. Validation of the LUMIPULSE automated immunoassay for the measurement of core AD biomarkers in cerebrospinal fluid. Clin Chem Lab Med. 2021;60:207–19. [DOI] [PubMed] [Google Scholar]
- 62.Gonzalez-Ortiz F, Ferreira PCL, González-Escalante A, Montoliu-Gaya L, Ortiz-Romero P, Kac PR, Turton M, Kvartsberg H, Ashton NJ, Zetterberg H, Harrison P, Bellaver B, Povala G, Villemagne VL, Pascoal TA, Ganguli M, Cohen AD, Minguillon C, Contador J, Suárez-Calvet M, Karikari TK, Blennow K. A novel ultrasensitive assay for plasma p-tau217: Performance in individuals with subjective cognitive decline and early Alzheimer’s disease. Alzheimers Dement. 2024;20:1239–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Nilsson J, Ashton NJ, Benedet AL, Montoliu-Gaya L, Gobom J, Pascoal TA, Chamoun M, Portelius E, Jeromin A, Mendes M, Zetterberg H, Rosa-Neto P, Brinkmalm A, Blennow K. Quantification of SNAP-25 with mass spectrometry and Simoa: a method comparison in Alzheimer’s disease. Alzheimers Res Ther. 2022;14:78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64.Evenden D, Prosser A, Michopoulou S, Kipps C. for the Alzheimer’s Disease Neuroimaging Initiative. ADCOMS sensitivity versus baseline diagnosis and progression phenotypes. Alzheimers Dement (Amst). 2024;16:e12540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Tahami Monfared AA, Houghton K, Zhang Q, Mauskopf J. Alzheimer’s Disease Neuroimaging Initiative. Staging Disease Severity Using the Alzheimer’s Disease Composite Score (ADCOMS): A Retrospective Data Analysis. Neurol Ther. 2022;11:413–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Schönbrodt FD, Perugini M. At what sample size do correlations stabilize? J Res Pers. 2013;47:609–12. [Google Scholar]
- 67.Alves F, Kalinowski P, Ayton S. Accelerated Brain Volume Loss Caused by Anti-β-Amyloid Drugs: A Systematic Review and Meta-analysis. Neurology. 2023;100:e2114–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68.Yang SJ, Lim JS, Choi Y, Kim HS, Kim SJ, Lee JH, Suh CH. Comparing Susceptibility-Weighted Imaging and T2* Gradient-Recalled Echo for Cerebral Microbleeds Detection: A Systematic Review and Meta-Analysis. J Clin Neurol. 2026;22:193–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 69.Sima DM, Phan TV, Franceschi AM, Gibbs WN, Barkhof F, Scheltens P, Salloway S, Cummings J, Van Hecke W, Smeets D. Monitoring of amyloid related imaging abnormalities: SWI vs T2*-GRE. J Prev Alzheimers Dis. 2025;12:100220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70.Cogswell PM, Andrews TJ, Barakos JA, Barkhof F, Bash S, Benayoun MD, Chiang GC, Franceschi AM, Jack CR Jr, Pillai JJ, Poussaint TY, Raji CA, Ramanan VK, Tanabe J, Tanenbaum L, Whitlow CT, Yu FF, Zaharchuk G, Zeinah M, Benzinger TS. ASNR Alzheimer, ARIA, and Dementia Study Group. Alzheimer Disease Anti-Amyloid Immunotherapies: Imaging Recommendations and Practice Considerations for Monitoring of Amyloid-Related Imaging Abnormalities. AJNR Am J Neuroradiol. 2025;46:24–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Yates PA, Villemagne VL, Ellis KA, Desmond PM, Masters CL, Rowe CC. Cerebral microbleeds: a review of clinical, genetic, and neuroimaging associations. Front Neurol., van der Flier WM, Sluimer JD, Leys D, Barkhof F, Scheltens P. Prevalence and severity of microbleeds in a memory clinic setting. Neurology. 2006;66:1356-60. [DOI] [PubMed]
- 72.Cordonnier C, van der Flier WM, Sluimer JD, Leys D, Barkhof F, Scheltens P. Prevalence and severity of microbleeds in a memory clinic setting. Neurology. 2006;66:1356–60. [DOI] [PubMed] [Google Scholar]
- 73.Goos JD, Henneman WJ, Sluimer JD, Vrenken H, Sluimer IC, Barkhof F, Blankenstein MA, Scheltens PH, van der Flier WM. Incidence of cerebral microbleeds: a longitudinal study in a memory clinic population. Neurology. 2010;74:1954–60. [DOI] [PubMed] [Google Scholar]
- 74.Hanyu H, Tanaka Y, Shimizu S, Takasaki M, Abe K. Cerebral microbleeds in Alzheimer’s disease. J Neurol. 2003;250:1496–7. [DOI] [PubMed] [Google Scholar]
- 75.Pettersen JA, Sathiyamoorthy G, Gao FQ, Szilagyi G, Nadkarni NK, St George-Hyslop P, Rogaeva E, Black SE. Microbleed topography, leukoaraiosis, and cognition in probable Alzheimer disease from the Sunnybrook dementia study. Arch Neurol. 2008;65:790–5. [DOI] [PubMed] [Google Scholar]
- 76.Kirsch W, McAuley G, Holshouser B, Petersen F, Ayaz M, Vinters HV, Dickson C, Haacke EM, Britt W 3rd, Larseng J, Kim I, Mueller C, Schrag M, Kido D. Serial susceptibility weighted MRI measures brain iron and microbleeds in dementia. J Alzheimers Dis. 2009;17:599–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 77.Mintun MA, Lo AC, Duggan Evans C, Wessels AM, Ardayfio PA, Andersen SW, Shcherbinin S, Sparks J, Sims JR, Brys M, Apostolova LG, Salloway SP, Skovronsky DM. Donanemab in Early Alzheimer’s Disease. N Engl J Med. 2021;384:1691–704. [DOI] [PubMed] [Google Scholar]
- 78.Bateman RJ, Smith J, Donohue MC, Delmar P, Abbas R, Salloway S, Wojtowicz J, Blennow K, Bittner T, Black SE, Klein G, Boada M, Grimmer T, Tamaoka A, Perry RJ, Turner RS, Watson D, Woodward M, Thanasopoulou A, Lane C, Baudler M, Fox NC, Cummings JL, Fontoura P, Doody RS. GRADUATE I and II Investigators and the Gantenerumab Study Group. Two Phase 3 Trials of Gantenerumab in Early Alzheimer’s Disease. N Engl J Med. 2023;389:1862–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 79.Qu BX, Gong Y, Moore C, Fu M, German DC, Chang LY, Rosenberg R, Diaz-Arrastia R. Beta-amyloid auto-antibodies are reduced in Alzheimer’s disease. J Neuroimmunol. 2014;274:168–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 80.Sohn JH, So JO, Hong HJ, Kim JW, Na DR, Kim M, Kim H, Nam E, Ha HJ, Kim YH, Mook-Jung I. Identification of autoantibody against beta-amyloid peptide in the serum of elderly. Front Biosci (Landmark Ed). 2009;14:3879–83. [DOI] [PubMed] [Google Scholar]
- 81.Weksler ME, Relkin N, Turkenich R, LaRusse S, Zhou L, Szabo P. Patients with Alzheimer disease have lower levels of serum anti-amyloid peptide antibodies than healthy elderly individuals. Exp Gerontol. 2002;37:943–8. [DOI] [PubMed] [Google Scholar]
- 82.Britschgi M, Olin CE, Johns HT, Takeda-Uchimura Y, LeMieux MC, Rufibach K, Rajadas J, Zhang H, Tomooka B, Robinson WH, Clark CM, Fagan AM, Galasko DR, Holtzman DM, Jutel M, Kaye JA, Lemere CA, Leszek J, Li G, Peskind ER, Quinn JF, Yesavage JA, Ghiso JA, Wyss-Coray T. Neuroprotective natural antibodies to assemblies of amyloidogenic peptides decrease with normal aging and advancing Alzheimer’s disease. Proc Natl Acad Sci U S A. 2009;106:12145–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 83.Klaver AC, Coffey MP, Smith LM, Bennett DA, Finke JM, Dang L, Loeffler DA. ELISA measurement of specific non-antigen-bound antibodies to Aβ1–42 monomer and soluble oligomers in sera from Alzheimer’s disease, mild cognitively impaired, and noncognitively impaired subjects. J Neuroinflammation. 2011;8:93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 84.Xu W, Kawarabayashi T, Matsubara E, Deguchi K, Murakami T, Harigaya Y, Ikeda M, Amari M, Kuwano R, Abe K, Shoji M. Plasma antibodies to Abeta40 and Abeta42 in patients with Alzheimer’s disease and normal controls. Brain Res. 2008;1219:169–79. [DOI] [PubMed] [Google Scholar]
- 85.Gruden MA, Davidova TB, Malisauskas M, Sewell RD, Voskresenskaya NI, Wilhelm K, Elistratova EI, Sherstnev VV, Morozova-Roche LA. Differential neuroimmune markers to the onset of Alzheimer’s disease neurodegeneration and dementia: autoantibodies to Abeta((25–35)) oligomers, S100b and neurotransmitters. J Neuroimmunol. 2007;186:181–92. [DOI] [PubMed] [Google Scholar]
- 86.Gustaw KA, Garrett MR, Lee HG, Castellani RJ, Zagorski MG, Prakasam A, Siedlak SL, Zhu X, Perry G, Petersen RB, Friedland RP, Smith MA. Antigen-antibody dissociation in Alzheimer disease: a novel approach to diagnosis. J Neurochem. 2008;106:1350–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 87.Gustaw-Rothenberg KA, Siedlak SL, Bonda DJ, Lerner A, Tabaton M, Perry G, Smith MA. Dissociated amyloid-beta antibody levels as a serum biomarker for the progression of Alzheimer’s disease: a population-based study. Exp Gerontol. 2010;45:47–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 88.Nath A, Hall E, Tuzova M, Dobbs M, Jons M, Anderson C, Woodward J, Guo Z, Fu W, Kryscio R, Wekstein D, Smith C, Markesbery WR, Mattson MP. Autoantibodies to amyloid beta-peptide (Abeta) are increased in Alzheimer’s disease patients and Abeta antibodies can enhance Abeta neurotoxicity: implications for disease pathogenesis and vaccine development. Neuromolecular Med. 2003;3:29–39. [DOI] [PubMed] [Google Scholar]
- 89.Liu YH, Wang J, Li QX, Fowler CJ, Zeng F, Deng J, Xu ZQ, Zhou HD, Doecke JD, Villemagne VL, Lim YY, Masters CL, Wang YJ. Association of naturally occurring antibodies to β-amyloid with cognitive decline and cerebral amyloidosis in Alzheimer’s disease. Sci Adv. 2021;7:eabb0457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 90.Norris GT, Kipnis J. Immune cells and CNS physiology: Microglia and beyond. J Exp Med. 2019;216:60–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 91.Cyster JG, Wilson PC. Antibody modulation of B cell responses-Incorporating positive and negative feedback. Immunity. 2024;57:1466–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 92.de Leeuw DM, Trieu C, Vromen EM, Blujdea ER, Verberk IMW, Duits FH, Teunissen CE, Pijnenburg YAL, van der Flier WM, van Harten AC, Visser PJ, Tijms BM. Cerebrospinal Fluid Amyloid and Tau Biomarker Changes Across the Alzheimer Disease Clinical Spectrum. JAMA Netw Open. 2025;8:e2519919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 93.Castellani RJ, Shanes ED, McCord M, Reish NJ, Flanagan ME, Mesulam MM, Jamshidi P. Neuropathology of Anti-Amyloid-β Immunotherapy: A Case Report. J Alzheimers Dis. 2023;93:803–13. [DOI] [PubMed] [Google Scholar]
- 94.Sato C, Barthélemy NR, Mawuenyega KG, Patterson BW, Gordon BA, Jockel-Balsarotti J, Sullivan M, Crisp MJ, Kasten T, Kirmess KM, Kanaan NM, Yarasheski KE, Baker-Nigh A, Benzinger TLS, Miller TM, Karch CM, Bateman RJ. Tau Kinetics in Neurons and the Human Central Nervous System. Neuron. 2018;97:1284–e12987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 95.Pontecorvo MJ, Lu M, Burnham SC, Schade AE, Dage JL, Shcherbinin S, Collins EC, Sims JR, Mintun MA. Association of Donanemab Treatment With Exploratory Plasma Biomarkers in Early Symptomatic Alzheimer Disease: A Secondary Analysis of the TRAILBLAZER-ALZ Randomized Clinical Trial. JAMA Neurol. 2022;79:1250–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 96.Bittner T, Tonietto M, Klein G, Belusov A, Illiano V, Voyle N, Delmar P, Scelsi MA, Gobbi S, Silvestri E, Barakovic M, Napolitano A, Galli C, Abaei M, Blennow K, Barkhof F. Biomarker treatment effects in two phase 3 trials of gantenerumab. Alzheimers Dement. 2025;21:e14414. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data for this study are reported in the manuscript and Data Supplement. Additional data may be available upon written request to the Sponsor. The protocol and statistical analysis plan are available upon request.