Corporate Presentation October 2026 1
Forward-Looking Statements 2 This presentation contains forward-looking statements within the meaning of The Private Securities Litigation Reform Act of 1995. Any statement describing Acumen’s goals, expectations, financial or other projections, intentions or beliefs is a forward-looking statement and should be considered an at-risk statement. Words such as “believes,” “expects,” “anticipates,” “could,” “would,” “seeks,” “aims,” “plans,” “potential,” “will” and similar expressions are intended to identify forward-looking statements, although not all forward-looking statements contain these identifying words. Forward-looking statements include statements concerning Acumen’s business, and Acumen’s ability to achieve its strategic and financial goals, including its projected use of cash, cash equivalents and marketable securities and the expected sufficiency of its cash resources into early 2027, the therapeutic potential of Acumen’s product candidate, sabirnetug (ACU193), including against other antibodies, the timing of anticipated topline results of ALTITUDE-AD, the potential for additional development to support a subcutaneous dosing option of sabirnetug, and the potential to develop a candidate to treat Alzheimer's Disease utilizing Enhanced Brain Delivery (EBDTM) technology. These statements are based upon the current beliefs and expectations of Acumen management, and are subject to certain factors, risks and uncertainties, particularly those inherent in the process of discovering, developing and commercializing safe and effective human therapeutics. Such risks may be amplified by the impacts of the COVID-19 pandemic. These and other risks concerning Acumen’s programs are described in additional detail in Acumen’s filings with the Securities and Exchange Commission (“SEC”), including in Acumen’s most recent Annual Report Form 10-K and future filings and reports by Acumen. Copies of these and other documents are available from Acumen. Additional information will be made available in other filings that Acumen makes from time to time with the SEC. These forward-looking statements speak only as of the date hereof, and Acumen expressly disclaims any obligation to update or revise any forward-looking statement, except as otherwise required by law, whether, as a result of new information, future events or otherwise. In this presentation, references to cash also include cash equivalents.
3 Large and growing market in need of additional treatment options Sabirnetug (ACU193): monoclonal antibody (mAb) highly selective for toxic AβOs with positive Phase 1 clinical trial results in AD patients; Phase 2 (IV) topline results expected November 2026 Enhanced Brain Delivery (EBD®) program to develop oligomer-targeted antibodies with BBB-penetrating technology; pre-clinical candidate (PCC) data announced early 2026 & lead candidate IND targeted for mid-2027 Experienced leadership team with extensive AD drug development experience Strong balance sheet supporting clinical development plans for sabirnetug (~$79M at 9/30/26) BBB: blood-brain-barrier Advancing Next Generation Treatments for Early Alzheimer’s Disease (AD) Targeting Toxic Amyloid Beta Oligomers (AβOs)
Early Alzheimer’s Disease (AD) Patient Population Represents Significant and Growing Market; Preclinical AD a Potential Future Addressable Population *Alzheimer's Association An estimated 7.2 million* Americans age 65 and older live with Alzheimer’s dementia today STAGES AND CHARACTERISTICS OF AD PROGRESSION Severe DementiaModerate DementiaMild DementiaMild Cognitive ImpairmentPreclinical AD ~5 Million* ~2 Million* Early Alzheimer’s Disease in the U.S. Cognitive Decline Behavioral and Psychological Complications Functional Decline This number is projected to grow to 13.8 million* by 2060 ~92% growth Biomarker changes (AβΟ+) 4
5 *ARIA-E: Amyloid-related imaging abnormalities - edema AD Treatment Landscape Continues to Evolve Approved Therapies Establish Momentum and Next-generation Approaches Expand Impact Future of AD Landscape Broadens Growth Prospects • New blood-based biomarkers and imaging tools will improve diagnostic and monitoring capabilities • Expansion into preclinical AD will significantly increase the addressable population • Combinations with other targets, such as tau, hold promise Current AD Treatments Offer Hope for People Living with AD • First approved anti-amyloid therapies steadily growing • Education and infrastructure buildout allowing for greater access to treatments by the addressable patient population • Increase efficacy • Reduce ARIA-E* risk • Provide more convenient delivery options • Target amyloid that occurs earlier in disease course Opportunities for New Anti-amyloid Therapies to Strengthen Risk-Benefit Tradeoff
6 Phase 1 Phase 2 Phase 3Preclinical Intravenous (IV) Subcutaneous (SC) Expect to report Phase 2 topline results in Nov. 2026 Phase 1 topline results announced Mar. 2025 support further development TfR-Brain transporter + anti-AβO mAb (SC) Option exercised to develop up to two EBD development candidates for the treatment of AD; lead candidate IND filing targeted for mid-2027 Sabirnetug (ACU193) Enhanced Brain Delivery (EBD®) 2026 a Pivotal Year for Acumen’s Pipeline Sabirnetug Global Phase 2 Study Results & Next Generation EBD Candidate Selection
Acumen Leadership Team Experienced in AD/Neuro Drug Development Acumen team has decades of experience in Alzheimer’s drug discovery and development AMY SCHACTERLE, PHD Chief Regulatory Officer, Head of Quality PAUL SHUGHRUE, PHD VP, Research & Strategy JASNA JERECIC, PHD Disease Area Strategy Lead ERIC SIEMERS, MD Chief Medical Officer MATT ZUGA Chief Financial Officer & Chief Business Officer RUSSELL BARTON Chief Operating Officer DANIEL O'CONNELL Chief Executive Officer SIEW TIN GAN Head of Clinical Operations 7 JULIE BOCKENSTETTE Chief People Officer JAMES DOHERTY, PHD President & Chief Development Officer LIEAN SCHENK SVP, Head of CMC LAURA ROSEN, MD, PHD SVP, Clinical Development DEREK MEISNER, JD Chief Legal Officer
Milestones Achieved and Upcoming ~$79M Cash, cash equivalents and marketable securities as of Sept. 30, 2026 8 MILESTONES TIMING STATUS Initiation of ALTITUDE-AD Phase 2 trial 2Q2024 Completion of enrollment of ALTITUDE-AD 1Q2025 Phase 1 subcutaneous topline results 1Q2025 EBD® non-clinical data package Early 2026 ALTITUDE-AD topline results November 2026 EBD lead candidate IND filing Mid-2027 Acumen expects its cash runway to extend into early 2027
Amyloid Beta Oligomers in AD Sabirnetug (ACU193): monoclonal antibody (mAb) highly selective for toxic AβOs
Soluble AβOs Contribute to Pathophysiological Processes Associated with Alzheimer’s Disease • Toxic consequences of soluble Aβ oligomer production include synapse dysfunction and loss, tau hyperphosphorylation, immune cell activation and functional impairment • Reduced neuronal toxicity and intervention at the synaptic level may prevent irreversible neuronal cell death Aberrant Tau Phosphorylation Selective Neuron Death Oxidative & ER Stress Insulin Resistance ChAT Loss Synaptic Deterioration Plasticity Dysfunction Receptor Redistribution Impact on Astrocytes/Microglia Disrupted Ca2+ homeostasis Inhibition of Axonal Transport 10 Adapted from Blomeke et al. 2024 Model of Aβ oligomer levels • Soluble Aβ forms appear early in the course of disease pathophysiology • Production of toxic soluble Aβ persists after plaque removal
Sabirnetug: Potential Next Generation Immunotherapy for Early AD 11 • Humanized, affinity matured mAb developed to target toxic Aβ oligomers • IgG2 subclass mAb with reduced effector function Designed for Improved Efficacy & Safety Regulatory Interactions • FDA Fast Track designation for the treatment of early AD • Phase 2 implemented as a registration- quality study Positive Ph1 in AD Patients & Encouraging Ph2 Enrollment • Successful Phase 1 exclusively in early AD patients Safety, target engagement, biomarker effects • Phase 2 (n=542) enrollment complete in March 2025; topline results expected November 2026 Monomers Oligomers Protofibrils Fibrils Plaques Soluble Amyloid Insoluble Amyloid sabirnetug IgG2 Neurotoxic
Sabirnetug
Q2W: Dosing every two weeks; Q4W: Dosing every four weeks. PART A: SINGLE-ASCENDING DOSE n = 8 per cohort (32 total) 6:2 per cohort PART B: MULTIPLE-ASCENDING DOSE n = 10 per cohort (30 total) 3 administrations of drug or PBO 8:2 per cohort COHORT 1: 2 mg/kg sabirnetug or Placebo 2mg COHORT 2: 10 mg/kg sabirnetug or Placebo 10mg COHORT 3: 25 mg/kg sabirnetug or Placebo 25mg COHORT 4: 60 mg/kg sabirnetug or Placebo 60mg COHORT 5: 10 mg/kg sabirnetug or Placebo (Q4W)* 10mg COHORT 6: 60 mg/kg sabirnetug or Placebo (Q4W)* 60mg COHORT 7: 25 mg/kg sabirnetug or Placebo (Q2W)* ≥ 1wk ≥ 1wk ≥ 1wk ≥ 1wk ≥ 1wk ≥ 1wk 25mg ≥ 1wk INTERCEPT-AD: A Randomized Placebo Controlled Phase 1 in Early AD Patients 13
Doses Approaching Maximal Target Engagement Support Sabirnetug AβO Mechanism and Helped Guide Dose Selection for Next Study Phase 14 Single & Multiple Dose Cohorts - Exposure Response Relationship (Emax Model) Emax: 22.71 AU/mL Complex EC50: 136 ng/mL sabirnetug 0 600 1200 1800 0 10 20 30 CSF [ACU193] (ng/mL) A C U 19 3- A βO C om pl ex (A U /m L) SAD 2 mg/kg SAD 10 mg/kg SAD 25 mg/kg SAD 60 mg/kg MAD 10 mg/kg Q4W MAD 60 mg/kg Q4W MAD 25 mg/kg Q2W Sabirnetug-AβO Complex TARGET ENGAGEMENT *One patient from Cohort 5 (10 mg/kg Q4W) excluded because only received one administration of drug (study drug discontinued after lacunar infarct). E. Siemers, et al. INTERCEPT-AD, a phase 1 study of intravenous sabirnetug in participants with mild cognitive impairment or mild dementia due to Alzheimer's disease. JPAD 2025. • Acumen developed a novel assay to measure the complex of AβOs bound to sabirnetug in cerebrospinal fluid • Observed target engagement with oligomers that increased across a range of doses • Achieved saturation point between 25 mg/kg and 60 mg/kg
Importance of Key Fluid Biomarkers Associated with AD Pathology 15 Tau Pathology: pTau181 pTau217 Neuronal Injury: Total tau Synaptic Injury: Neurogranin VAMP2 Amyloid Pathology: Aβ 42/40 Astrocytic Activation: GFAP • Biomarkers from cerebrospinal fluid and plasma capture neuronal, synaptic, and axonal injury and reflect the cumulative outcome of different pathological substrates in AD1 • Evidence suggests that biomarkers are likely to be better predictors of the underlying pathology of AD than imaging alone2 • After just three administrations of sabirnetug, patients with early AD demonstrated improvements in biomarkers associated with AD pathology 1. Tarawneh, R. Biomarkers: Our Path Towards a Cure for Alzheimer Disease. Biomarker Insights Volume 15: 1–15. 2020; 2. Blennow K, Zetterberg H. The Past and the Future of Alzheimer's Disease Fluid Biomarkers. J Alzheimer's Dis. 2018;62(3):1125-1140. Aβ oligomer Amyloid plaque
Consistent Improvement in CSF Amyloid, Tau and Synaptic Biomarkers Indicate Downstream Pharmacology of Sabirnetug After Only Three Doses 16 CSF BIOMARKERS pTau181 NeurograninAβ 42/40 Ratio VAMP2 Tau pathologyAmyloid pathology Synaptic injury n = 8 subjects/treated group; 6 subjects in pooled placebo (PBO); p-values from unpaired, 2-sided Student’s t test N orm alization A D Progression N orm alization A D Progression N orm alization A D Progression N orm alization A D Progression E. Cline, et al, Biofluid biomarker changes following treatment with sabirnetug (ACU193) in INTERCEPT-AD, a phase 1 trial in early Alzheimer’s disease. JPAD 2025.
Sabirnetug Demonstrates Potential for Best-in-Class Safety Profile Compelling Overall Safety Profile, with Low Incidence of ARIA-E in the INTERCEPT-AD study 17 SAFETY Limited incidence of ARIA-E • 10 mg/kg Q4W: 1 asymptomatic case • 25 mg/kg Q2W: 1 asymptomatic case • 60 mg/kg Q4W: 2 asymptomatic cases; 1 symptomatic case No ARIA-E observed in ApoE4 homozygotes (n=6), despite comprising 13% of study • Differentiated from other antibodies that have ARIA-E rates ~30% to ~40% in participants who are E4-homozygotes Broad therapeutic index with convenient monthly dosing • Safety profile may support attractive benefit/risk option for large portion of patients 5 0 Total ARIA-E cases, or ~10% Cases of ARIA-E in ApoE4 homozygotes N=6 0 Deaths, SAEs Related to Study Drug INTERCEPT-AD Phase 1 Safety Data E. Siemers, et al. INTERCEPT-AD, a phase 1 study of intravenous sabirnetug in participants with mild cognitive impairment or mild dementia due to Alzheimer's disease. JPAD 2025. ApoE4: ε4 allele of Apolipoprotein E
Potential for Differentiated Efficacy Potential for Differentiated Safety INTERCEPT-AD Phase 1 Data Support Potential for Sabirnetug to Offer Next Generation Efficacy and Safety 18 Compelling safety profile with low incidence of ARIA-E Absence of ARIA-E observed in ApoE4 homozygotes Broad potential therapeutic index with convenient monthly dosing First mAb to demonstrate selective target engagement of AβOs (most toxic form of Aβ) Rapid, meaningful plaque reduction comparable to the current market front-runners at similar timepoints Improvement of AD biomarkers in CSF and plasma are a strong indication of downstream effects Key Takeaways from INTERCEPT-AD
Ra nd om iz at io n 1: 1: 1 Sabirnetug 35mg/kg Q4W (n ~180) Sabirnetug 50mg/kg Q4W* (n ~180) Placebo Q4W (n ~180) Primary Endpoint Change in iADRS1 at 18 months Secondary Endpoints1 CDR-SB, ADAS- Cog13, ADCS-ADL, AD biomarkers 12-month open label extension Objective: To evaluate the clinical efficacy, safety and tolerability of sabirnetug Patient population: Patients with early AD (MCI or mild dementia due to early AD) ALTITUDE-AD Phase 2 Study Results Expected in Late 2026 1. iADRS: Integrated Alzheimer's Disease Rating Scale; CDR-SB: Clinical Dementia Rating – Sum of Boxes; ADAS-cog: Alzheimer's Disease Assessment Scale – Cognitive Subscale; ADCS-ADL: Alzheimer's Disease Cooperative Study – Activities of Daily Living 19 *Titration of sabirnetug 35mg/kg Q4W for two doses. • 542-participant study fully enrolled in March 2025 (U.S., Canada, U.K., Germany, Spain) • Open label extension portion of ALTITUDE-AD initiated in November 2025 • Topline results expected in November 2026 Key Highlights
Subcutaneous Formulation Well-Tolerated in Phase 1 Healthy Volunteer Study Potential to Dose Once-Weekly with Single Injection to Broaden Patient Access and Increase Treatment Convenience Topline results announced in March 2025 show systemic exposure supports further clinical development of the subcutaneous formulation 20 Phase 1 study to compare the pharmacokinetics of subcutaneous form of sabirnetug to the IV form Outcomes: • Safety • Subcutaneous bioavailability • Information on flat dosing Population: • Healthy volunteers • Age matched to AD population in sabirnetug Phase 1 (INTERCEPT-AD) study IV dose (single dose) (n = 12) Subcutaneous dose (4 doses, 1/week) (n = 16) Phase 1 Subcutaneous Healthy Volunteer Study Announced partnership with Halozyme in November 2023 to develop subcutaneous dosing option for sabirnetug using Halozyme’s drug delivery technology, ENHANZE®
Sabirnetug IP & Market Exclusivity • Exclusive, perpetual, irrevocable, worldwide, royalty-free license from Merck to its Amyloid Derived Diffusible Ligand (ADDL) IP including issued sabirnetug patents • Sabirnetug Global IP estate: Issued patents in 19 countries Composition of matter patents and methods of use run into July 2031 Patent term extensions may be available, 3-5 years depending on jurisdiction • Biologics market exclusivity is expected for sabirnetug as a novel biologic drug US provides 12 years market exclusivity for novel biologics Europe provides 10 years of market exclusivity for novel biologics • EBD program includes Acumen wholly-owned AβO-selective mAbs expected to generate novel IP including Composition of Matter Patent(s) 21
Enhanced Brain Delivery (EBD®)
23 Delivering Antibodies to the Brain via the Transferrin Receptor (TfR) Aim to Widen the Therapeutic Window to Increase Efficacy, Safety, and Convenience • Increased brain exposure may enhance efficacy • Wider capillary distribution may reduce ARIA-E risk • Lower dose volumes enable convenient subcutaneous administration Enhanced Brain Delivery (EBD) Opportunities: Oligomer-targeted antibody TfR-targeted carrier Collaboration, option and license agreement announced in July 2025 with JCR Pharmaceuticals Acumen holds exclusive option to license and develop up to two candidates under terms of agreement Image courtesy of JCR Pharmaceuticals
Acumen's EBD May Improve Efficacy, Safety and Delivery Compared to Non-AβO Selective Antibodies Differentiated Cargo (AβO-selective mAb) 24 Differentiated Aβ targeting Validated TfR-binding Validated BBB Carrier Technology (TfR-targeting) Acumen EBD Candidates • Targets synaptotoxic AβΟ species • Robust fluid biomarker results in Phase 1b INTERCEPT-AD trial • Demonstrated low ARIA-E overall in Phase 1b, including no ARIA- E observed in APOE 4/4 carriers • Low rate of infusion related reactions observed to date clinically • Candidate mAbs: ACU193 and ACU234 • Enhance brain penetration versus native antibody to enable low- volume SC delivery • Potential for lower ARIA rate due to TfR prevalence on capillary bed • Little to no anemia observed with JCR technologies = Enhanced efficacy: Leverage improved brain delivery/ distribution and Aβ oligomer targeting Best-in-class safety: Lower ARIA, anemia, IRRs Designed for SC administration: Low dose, formulation, stability ApoE4: ε4 allele of Apolipoprotein E
Acumen Explored Broad Parameters to Develop a TfR-Mediated AβO Product 25 TfR Affinity Architecture Valency AβO Selectivity scFv or VHH Heavy Chain vs Light Chain Monovalent Bivalent Affinity Range Selectivity Range SC Properties 14,000x 5,000x Dosing, Formulation, Stability TfR: transferrin receptor; scFv: single chain variable fragment antibodies; VHHs: variable heavy domain antibodies; SC: subcutaneous
Multiple Dose Cohorts* Half-life/Dosing Interval Once monthly dosing Dosing Exposure 35 mg/kg and 50 mg/kg in Phase 2 IV Target Engagement Maximal TE of AβOs observed Phase 1 Results Provide Foundation for EBD Program 26 Single & Multiple Dose Cohorts - Exposure Response Relationship (Emax Model) Emax: 22.71 AU/mL Complex EC50: 136 ng/mL sabirnetug 0 600 1200 1800 0 10 20 30 CSF [ACU193] (ng/mL) AC U1 93 -A βO C om pl ex (A U/ m L) SAD 2 mg/kg SAD 10 mg/kg SAD 25 mg/kg SAD 60 mg/kg MAD 10 mg/kg Q4W MAD 60 mg/kg Q4W MAD 25 mg/kg Q2W *One patient from Cohort 5 (10 mg/kg Q4W) excluded because only received one administration of drug (study drug discontinued after lacunar infarct). E. Siemers, et al. INTERCEPT-AD, a phase 1 study of intravenous sabirnetug in participants with mild cognitive impairment or mild dementia due to Alzheimer's disease. JPAD 2025.
Mouse Surrogate Antibodies Developed to Explore EBD and Target Engagement in a Mouse Model of Alzheimer’s Disease 27 The fusion of anti-oligomer antibodies and J-Brain Cargo (anti-TfR scFv) did not alter the ability of constructs to bind TfR or alter the preferential binding to AβOs Cline et al., CTAD, 2025 234 0 nM Oligomer Binding Monomer Binding 2.182 nM 3.761 µM ACU193 0.33 nM 5.34 nM 3.52 nM 0.595 nM 0.448 nM 2.475 nM 6.914 µM 5.871 µM 8.6 µM mTfR: murine transferrin receptor; mscFv: murine single chain variable fragment antibodies 0.319 * *Cline et al., AAIC, 2025
Animals dosed IV at 2 mpk Cline et al., CTAD, 2025 • Higher brain exposure was observed for all fusion proteins compared to ACU193: o mBBB.193-1 exhibited ~35-fold higher brain exposure at 3 hours o mBBB.193-2 and mBBB.234-2 exhibited ~17-fold higher brain exposure at 24 hours • mBBB.193-2 and mBBB.234-2 showed the greatest cumulative exposure: o 32- to 55-fold higher AUC than ACU193 In Wild-Type Mice, All EBD Fusion Proteins Showed Increased Brain Exposure Compared to Sabirnetug (ACU193) ~ 35x @ 3 hrs ~ 17x @ 24 hrs 28
Surrogate Antibodies Show Enhanced Delivery to the AD Mouse Brain, when Compared with ACU193 and ACU234 AD Mouse Model (ATRE10) Animals dosed at IV 2 mpk Cline et al., CTAD, 2025 ~ 63X ~ 68X ~ 65X 29 • The brain exposure of the three fusion proteins was 63- to 68-fold higher compared to ACU193 and ACU234 • Higher concentrations of all three fusion proteins seen in AD mouse brain compared to wild-type mice may be due to target engagement of AβOs and retention in brain
Robust Data Package Supporting EBD Construct Selection Inclusive of NHP Primate Data Anemia Risk Brain Distribution Stability & Manufacturability Pharmacokinetics TfR Affinity 30 Aβ Oligomer Binding
Key Takeaways from NHP Study Results 31 Enhanced Brain Penetration EBD candidates achieved 14-40x higher brain levels in non-human primates compared to native antibodies 24 hours after dosing Lead clinical candidate IND filing targeted for 2027 Subcutaneous Dosing Capability Favorable stability profile and enhanced brain delivery support a path to subcutaneous administration with low-volume devices Low Anemia Risk • Hematology data in non-human primates indicate low potential for anemia: At 24 hours after SC dosing, EBD candidates demonstrated no observed change in red blood cell count, hematocrit, hemoglobin or reticulocyte count • No adverse events observed
Significant and growing Alzheimer’s population in need of additional treatment options Synaptotoxic AβOs appear early in Alzheimer’s Disease and contribute to its pathophysiological processes; sabirnetug demonstrates high selectivity for AβOs in AD patients Positive Phase 1 data strengthen potential for sabirnetug to offer next-generation efficacy and safety Significant interest in ALTITUDE-AD, a Phase 2 study investigating sabirnetug, evidenced by rapid enrollment EBD® program augments portfolio optionality, leveraging Acumen capabilities and assets Acumen Positioned to Deliver Potential Next-Gen Treatment for Early AD 32 Key Takeaways Next Steps EBD program pre-clinical candidate (PCC) data announced in early 2026 Topline results from ALTITUDE-AD Phase 2 IV study evaluating the clinical efficacy and safety of sabirnetug in patients with MCI or mild dementia due to Alzheimer’s expected in November 2026 Lead clinical candidate IND filing for EBD program targeted for mid-2027
Appendix 33
Selection of Non-clinical Sabirnetug Data
Amyloid Beta Oligomers (AβOs) in Alzheimer’s Disease Pathology Aβ monomers are normally produced, non- toxic and orders of magnitude more prevalent than other forms of amyloid. Aβ plaques are relatively inert but serve as reservoirs for toxic species of amyloid: oligomers and protofibrils. Neurotoxic Soluble Amyloid Insoluble Amyloid
Literature Selection: Soluble AβOs Contribute to Pathophysiological Processes Associated with Alzheimer’s Disease Disrupted Ca2+ homeostasis Demuro et al, 2005 De Felice et al, 2007 Alberdi et al, 2010 Wang et al, 2018 ChAT loss Heinitz et al, 2006 Nunes-Tavares et al, 2012 Inhibition of axonal transport Pigino et al, 2009 Poon et al, 2009 Decker et al, 2010 Oxidative stress Longo et al, 2000 Sponne et al, 2003 Tabner et al, 2005 De Felice et al, 2007 ER stress Resende et al, 2008 Nishitsuji et al, 2009 Insulin resistance Zhao et al, 2008 Zhao et al, 2009 Ma et al, 2009 De Felice et al, 2009 Impact on astrocytes/microglia Hu et al, 1998 Jimenez et al, 2008 Sondag et al, 2009 Tomiyama et al, 2010 Aberrant Tau phosphorylation De Felice et al, 2008 Ma et al, 2009 Tomiyama et al, 2010 Zempel et al, 2010 Bloom, 2014 Forny-Germano et al, 2020 Wakeman et al, 2022 Darricau et al, 2023 Selective neuron death Lambert et al, 1998 Kim et al, 2003 Florent et al, 2006 Ryan et al, 2009 Lee et al, 2017 Komura, 2019 Plasticity dysfunction Lambert et al, 1998 Walsh et al, 2002 Wang et al, 2002 Townsend et al, 2006 Yasumoto et al, 2019 Synapse deterioration Zhao et al, 2006 Lacor et al, 2007 Shankar et al, 2007 Wu et al, 2010 Brito-Moreira et al, 2017 Actor-Engel et al, 2021 Sackmann & Hallbeck, 2020 Limegrover et al, 2021 Receptor Redistribution Snyder et al, 2005 Roselli et al, 2005 Lacor et al, 2007 Zhao et al, 2008 36
Differences in the Profile of Sabirnetug Relative to Approved anti-Aβ DMTs: Greater AβO selectivity against Aβ monomers and linear fibrillar species Aducanumab Lecanemab Sabirnetug 1×10-10 1×10-9 1×10-8 1×10-7 1×10-6 1×10-5 Sabirnetug vs Lecanemab and Aducanumab A ffi ni ty - K D (M ) Aβ1-40 monomer Aβ oligomers 8750x 222x 79.6x In cr ea se d bi nd in g af fin ity Greater selectivity for soluble Aβ species isolated from human AD brain Greater selectivity against monomeric Aβ Cline et al, AAIC 2025. • Eluates from affinity chromatography using: o r-mab158 36% globular, 64% fibrillar, Aβ species o sabirnetug 99% globular, 1% fibrillar Aβ species • lecanemab 222x higher affinity to AβO than monomer • sabirnetug 8750x higher affinity to AβO than monomer De Leon et al, AAIC 2025. There have been no head-to-head clinical trials between the product candidates listed above. Study designs and protocols for each product candidate were different, and as a result, results may not be comparable between product candidates. 37
Vascular Aβ Binding Profile in a Murine Model of Vascular CAA Immunohistochemical comparison of vascular Aβ binding in the APP:hE4 mouse brain • Sabirnetug (ACU193) and lecanemab showed qualitatively similar overall staining in both cortex and cerebellum • But sabirnetug had significantly less cerebellar vascular labeling than lecanemab (p = 0.0001) Cortex Cerebellum Grenon, Martine et al., Ex vivo comparison of ACU193 and lecanemab reveals binding differences in mouse brain, Alzheimer's & Dementia, 2026. There have been no head-to-head clinical trials between the product candidates listed above. Study designs and protocols for each product candidate were different, and as a result, results may not be comparable between product candidates. 38
Sabirnetug is an IgG2 Antibody, Which May Lower Risk of ARIA and IRR IgG1 antibodies (lecanemab, donanemab) strongly bind Fcγ1 receptors and C1q, triggering complement activation and antibody- dependent cellular phagocytosis (ADCP), causing inflammation and triggering strong cytokine release. IgG2 antibodies, like sabirnetug, weakly bind Fcγ1 receptors and avoid complement activation, potentially reducing the risk of inflammation and cytokine release. Maverakis E et al. Glycans in the immune system and The Altered Glycan Theory of Autoimmunity: a critical review. J Autoimmun., 2015. Wang, J., et al. A systemic view of Alzheimer disease — Insights from amyloid-β metabolism beyond the brain. Nature Reviews Neurology, 2017. Shashidharamurthy R, et al. Dynamics of the interaction of human IgG subtype immune complexes with cells expressing R and H allelic forms of a low-affinity Fcγ receptor CD32A, 2009. Potential Impact on Safety: • IgG1-based anti-Aβ DMTs (lecanemab, donanemab) show substantial inflammatory risk, leading to ARIA and hypersensitivity reactions • Including IRRs and anaphylaxis • This may limit the doses that can be used to drive efficacy • IgG2-based sabirnetug may enable amyloid clearance with reduced inflammation, cytokine release, and ADCP involvement • IgG2 Fc attenuation may lower the frequency and severity of hypersensitivity reactions or ARIA Differences in effector coupling may influence inflammatory risk profile 39
Phase 1 INTERCEPT-AD
Sabirnetug drug specific capture (anti-sabirnetug idiotype mAb) 41 Target Engagement Assessed by Measuring Sabirnetug-AβO Complex in CSF AβO selective detection (anti-AβΟ mAb) Only Sabirnetug-AβO complex is measurable • Novel assay configuration tailored to selectively detect sabirnetug-AβO complex in CSF as direct measure of target engagement • Translated for clinical use from a preclinical assay developed by Merck that showed sabirnetug engages target AβOs in transgenic mouse brain (tg2576) in dose dependent manner MSD S-Plex (Turbo) Immunoassay TARGET ENGAGEMENT
Target Engagement of Sabirnetug with AβOs is Dose Proportional 42 Single Dose Cohorts Multiple Dose Cohorts* 2 mg/kg 10 mg/kg 25 mg/kg 60 mg/kg 0 10 20 30 *One patient from Cohort 5 (10 mg/kg Q4W) excluded because only received one administration of drug (study drug discontinued after lacunar infarct). TARGET ENGAGEMENT p = 0.001 p = 0.03 p = 0.0007No Significant Differences E. Siemers, et al. INTERCEPT-AD, a phase 1 study of intravenous sabirnetug in participants with mild cognitive impairment or mild dementia due to Alzheimer's disease. JPAD 2025.
Nearly All Sabirnetug-Treated Patients in High Dose MAD Cohorts Showed Reductions in Plaque Load After Three Doses at 63 or 70 days 43 Plaque load based on florbetapir PET 0 15 30 45 60 75 0 25 50 75 100 125 25 mg/kg Q2W MAD Days C en til oi ds 20 21 22 23 24 25 26 27 0 15 30 45 60 75 0 25 50 75 100 125 60 mg/kg Q4W MAD Days C en til oi ds 28 29 30 31 32 33 34 35 25 mg/kg Q2W MAD 60 mg/kg Q4W MAD Mean reduction in amyloid plaque ∆ (absolute value, centiloids) 13.7 ∆ (%, centiloids) 20.6% Mean reduction in amyloid plaque ∆ (absolute value, centiloids) 18.1 ∆ (%, centiloids) 25.6% PLAQUE REDUCTION E. Siemers, et al. INTERCEPT-AD, a phase 1 study of intravenous sabirnetug in participants with mild cognitive impairment or mild dementia due to Alzheimer's disease. JPAD 2025.
Sabirnetug Serum Exposure is Dose Proportional Without Accumulation 44 Single Dose Cohorts Multiple Dose Cohorts PHARMACOKINETICS Estimated serum terminal T1/2 of 5-7 days E. Siemers, et al. INTERCEPT-AD, a phase 1 study of intravenous sabirnetug in participants with mild cognitive impairment or mild dementia due to Alzheimer's disease. JPAD 2025.
Sabirnetug CSF Exposure is Dose and Dose-Regimen Proportional 45 Single Dose Cohorts Multiple Dose Cohorts* *One patient from Cohort 5 (10 mg/kg Q4W) excluded because only received one administration of drug (study drug discontinued after lacunar infarct). PHARMACOKINETICS E. Siemers, et al. INTERCEPT-AD, a phase 1 study of intravenous sabirnetug in participants with mild cognitive impairment or mild dementia due to Alzheimer's disease. JPAD 2025.
Trend Toward Normalizing Plasma Biomarkers with 10 mg/kg and 60 mg/kg Q4W 46 MAD PBO 10 m g/kg Q 4W 25 m g/kg Q 2W 60 m g/kg Q 4W -100 -50 0 50 100 150 Pl as m a G FA P (% c ha ng e fr om b as el in e) MAD PBO 10 m g/kg Q 4W 25 m g/kg Q 2W 60 m g/kg Q 4W -50 0 50 100 Pl as m a pT au 18 1 (% c ha ng e fr om b as el in e) MAD PBO 10 m g/kg Q 4W 25 m g/kg Q 2W 60 m g/kg Q 4W -100 -50 0 50 100 150 Pl as m a pT au 21 7 (% c ha ng e fr om b as el in e) 1- 6 w k po st -d os in g • Plasma measurements of glial fibrillary acidic protein (GFAP), pTau181, and pTau217 in 10 mg/kg Q4W & 60 mg/kg Q4W groups were lower than placebo • More impact to fluid biomarkers was observed with longer dosing duration o The 25 mg/kg Q2W cohort differed in dose and sample timing, with drug on board for less time than the 10 mg/kg & 60 mg/kg Q4W cohorts PLASMA BIOMARKERS pTau181GFAP pTau217 A D Progression N orm alization n = 8 subjects/treated group; 6 subjects in pooled placebo (PBO); p-values from unpaired, 2-sided Student’s t test E. Cline, et al, Biofluid biomarker changes following treatment with sabirnetug (ACU193) in INTERCEPT-AD, a phase 1 trial in early Alzheimer’s disease. JPAD 2025.
SAD MAD 10 mg/kg Cohorts 2, 5 25 mg/kg Cohorts 3, 7 60 mg/kg Cohorts 4, 6 NO ARIA-E Asymptomatic ARIA-E Symptomatic ARIA-E Discontinued ApoE D21 D140 3,4 PBO PBO 3,3 3,3 3,4 3,4 PBO PBO 3,4 3,4 3,4 ApoE D28 D63 D126 3,4 3,3 3,3 4,4 4,4 PBO PBO PBO 3,3 3,4 3,4 3,4 PBO PBO PBO 3,3 ApoE D28 D70 D196 2,3 3,3 3,3 4,4 3,3 PBO PBO PBO 3,4 4,4 3,4 3,3 3,4 PBO PBO PBO ApoE D21 D140 3,3 3,3 PBO PBO 4,4 3,3 2,4 3,3 PBO PBO 3,4 3,3 ApoE D28 D70 D98 3,3 3,4 3,4 3,4 3,4 3,4 PBO PBO PBO 3,3 3,4 PBO PBO PBO 4,4 4,4 ApoE D21 D140 4,4 PBO PBO 3,4 3,4 PBO PBO 3,3 3,3 3,4 2,4 3,4 2 mg/kg Cohort 1 ApoE D21 D140 3,4 3,3 PBO PBO 3,4 2,3 3,4 PBO PBO 3,3 3,3 3,3 ARIA-E Summary for INTERCEPT-AD 47 PBO: Patient on placebo SAFETY
Phase 2 ALTITUDE-AD
Simulated CSF Target Engagement at Steady-State for ALTITUDE-AD Doses • CSF target engagement was simulated at a candidate list of doses given Q4W at steady-state • Notable diminishing differentiation as dose increases • Doses were selected with peak-trough variation in mind: select doses based on trough (end of dosing interval) CSF engagement upper dose: 50 mg/kg Q4W lower dose: 35 mg/kg Q4W Ph2 Dosing Strategy (ALTITUDE-AD) TIME 49