AD/PD 2026 Conference Alzheimer's Preclinical and Clinical Developments

Resting-state EEG as a Tool to Monitor Early Electrophysiological Changes During Lecanemab Infusion

Yumei Liu, Shaoxia Fan, Shuhan Fan, Zian Pei, Zhifan Zhou, Xue Shi, Yi Guo

1Shenzhen Bay laboratory, Institute Of Neurological And Psychiatric Disorders, Shenzhen, China, 2Shenzhen People's Hospital, Department Of Neurology, shenzhen, China, 3Southern University of Science and Technology, Department Of Electronic And Electrical Engineering, shenzhen, China

Aims: Lecanemab, an anti-amyloid-β monoclonal antibody, slows cognitive decline in Alzheimer’s disease (AD). However, its early electrophysiological effects remain unclear. This study examined early resting-state electroencephalograms (rsEEG) changes after Lecanemab infusion and assessed whether EEG could predict adverse effects.

Methods: We prospectively collected rsEEG from AD patients before their first Lecanemab infusion, and at days 1, 3, 7, 1 month, and 3 months post-infusion. Of 32 eligible patients, usable rsEEG was obtained from 25, 20, 18, 22, and 19, respectively. Power spectral density and source localization changes were analyzed.

Results: At baseline, higher theta power correlated with lower MMSE (ρ= -0.38) and MoCA (ρ= -0.42), and higher ADL scores (ρ = 0.43). Lower alpha power was linked to higher CDR (ρ= -0.38) and lower MMSE (ρ = 0.38). DTABR(delta + theta/alpha +beta) correlated positively with CDR (ρ= 0.41) and ADL (ρ= 0.40), and negatively with MMSE (ρ = -0.43) and MoCA (ρ = -0.45). Longitudinally, delta power increased, peaking at day 7 (0.23 ± 0.02 vs. 0.28 ± 0.03; Cohen’s d = 0.89), with sources localized to bilateral parietal/temporal and left occipital regions; these changes attenuated by 1 and 3 months. Delta activity and DTABR correlated with Hypertension-Coronary score (ρ = 0.51) and cognitive decline duration (ρ = 0.55). Patients with adverse events (ARIA, infusion reactions, dizziness, allergy) had lower baseline alpha power than those without (0.26 ± 0.02 vs. 0.31 ± 0.03; η²p = 0.137).

Conclusions: Lecanemab induces early electrophysiological changes in AD, marked by increased delta activity and DTABR within the first week. RsEEG is sensitive in detecting these alterations and may serve as a noninvasive biomarker for both treatment monitoring and predicting adverse events.

AI-Driven Digital Speech Biomarkers Detect Therapeutic Effects of Transcranial Pulse Stimulation in Alzheimer’s Disease

Daniel Wiechmann,2, Ali Günes, Elma Kerz, Yu Qiao, Martin Köhne, Ulrich Sprick

1University of Amsterdam, Institute For Logic, Language And Computation, Amsterdam, Germany, 2Exaia Technologies, Aachen, Germany, 3Alexius/Josef Krankenhaus, Department Forschung Und Entwicklung, Neurostimulationszentrum, Neuss, Germany

Aims: To develop interpretable AI models using digital biomarkers derived from spontaneous speech to differential diagnosis of Alzheimer’s disease (AD) and to individually quantify therapeutic effects of transcranial pulse stimulation (TPS) on disease progression.

Methods: End-to-end automatic classification models were trained using spontaneous speech data from 40 patients with clinically diagnosed AD (mean age = 73.8 ± 9.9 years) and 27 healthy controls. Diagnosis was based on current German S3 guidelines, including comprehensive neuropsychological assessments (CERAD Plus including MMSE, Stroop-Test) and structural MRI. Interpretable AI identified the most discriminative speech features associated with AD pathology. Digital speech biomarkers were then used to quantify TPS-related modulation of disease trajectory related to 14 TPS- treated patients, based on paired recordings acquired 268.7 ± 152.9 days apart.

Results: Speech biomarkers showed strong correlations with cognitive performance (CERAD r = 0.72; MMSE r = 0.73) and outperformed the Stroop test in capturing cognitive impairment. Classification accuracy for AD versus healthy controls achieved F1 = 0.81. At the group level, patients receiving TPS showed stable or slightly improved linguistic performance throughout the follow-up period, contrasting with the ~3.3-point annual MMSE decline expected in untreated Alzheimer’s disease. TPS intervention also revealed pronounced individual variability in t0/t1-speech biomarker trajectories, indicating heterogeneous patient responsiveness. Key discriminative speech features identified by the interpretable AI models yield mechanistic insights into TPS-related neurophysiological effects.

Conclusions: This study introduces a novel framework integrating interpretable AI with speech-derived digital biomarkers as precision tools for AD diagnosis and monitoring. Findings demonstrate that TPS exerts personalized, measurable effects on disease progression, detectable through sensitive speech-based metrics which surpass traditional cognitive assessments.

Effectiveness, Safety and Biomarker Dynamics of Lecanemab in Chinese Alzheimer's Disease Population: a Multicenter Real-World Study

Wang Liao, Qun Yu, Bin Chen, Hualin Chen, Chenyang Wang, Yang Li, Ying Yang, Ziyu She, Qinggan Zeng, Zengqiang Zhang, Guihua Li, Shujun Feng, Han Lin, Jipeng Ouyang, Jun Liu

1The Second Affiliated Hospital of Guangzhou Medical University, Guangzhou, China, 2Department of Neurology, Department of Pediatric Neurology, Zhujiang Hospital of Southern Medical University, Guangzhou, China, 3Department of Neurology, the First Affiliated Hospital of Jinan University Medical College, Guangzhou, China, 4Department of Neurology, Zhongshan People's Hospital, Guangdong, Zhongshan, China, 5Department of Neurology, Heyou Hospital, Shunde District, Foshan City, Guangdong, China., Foshan, China, 6Hainan Hospital of Chinese PLA Genera Hospital, Sanya, China, 7Department of Neurology, The Affiliated Guangdong Second Provincial General Hospital of Jinan University, Guangzhou, China, 8Department of Neurology, Guangdong Provincial People’s Hospital, Guangzhou, China, 9Department of Neurology, Houjie Hospital of Dongguan, Dongguan, China, 10Department of Neurology, The Eighth Affiliated Hospital of Southern Medical University (The First People’s Hospital of Shunde Foshan), F, China

Aims: Lecanemab, a monoclonal antibody targeting soluble amyloid-beta (Aβ) protofibrils, was introduced in China in June 2024. However, its real-world effectiveness and safety in the Chinese population remain to be fully elucidated, and optimal strategies for monitoring therapeutic response are yet to be established. This study aims to evaluate the effectiveness and safety of Lecanemab in clinical practice in China.

Methods: This was a prospective, multicenter real-world study conducted across 23 centers. To better assess the treatment effect, a matched cohort from the Alzheimer' s Disease Neuroimaging Initiative was included for comparison. Participants received biweekly intravenous infusions of Lecanemab at a dose of 10 mg/kg. Comprehensive assessments were conducted, including neuropsychological evaluations, plasma biomarkers (Aβ42, Aβ40, P-Tau217, P-Tau181, else), and magnetic resonance imaging. For patients who provided consent, both Aβ and Tau-positron emission tomography scan imaging was performed after six months to evaluate pathological clearance.

Results: As of August 2025, 261 participants who received at least one injection were enrolled (59.0% female; mean age 69.74 ± 8.92 years; baseline MMSE 18.22 ± 6.1; 79.7% had CDR-GS at 0.5~1). Most plasma biomarkers exhibited significant treatment-related changes, with 82 participants completed assessments at 3 months, and 50 completed at 6 months. Among 29 patients with PET-CT reassessment, mean summary Aβ decreased 42.9±6.2 centiloid and 41.4% converted to negative status. Notably, APOE ε4 non-carriers achieved greater clinical and biomarker improvements. The most frequent adverse event was infusion-related reactions (11.1%). Amyloid-related imaging abnormalities occurred in 24 (9.2%) patients, with 4 (1.6%) being symptomatic, which did not differ significantly among participants with mild cognitive impairment (MCI) and moderate AD.

Conclusions: Lecanemab was generally well-tolerated and the improvements in blood biomarkers suggest that it may have a beneficial impact on disease progression.

Regional Amyloid Burden and Clearance Across AD Clinical Phenotypes after 6 Months of Anti-amyloid Therapy

Alma Ghirelli, Edoardo Gioele Spinelli, Giordano Cecchetti, Giulia Rugarli, Stefano Pisano, Federico Coraglia, Ana Maria Samanes Gajate, Andrea Panzacchi, Gino Pepe, Elisa Canu, Veronica Castelnovo, Elisa Sibilla, Anna Gilioli, Chiara Tripodi, Fabiola Freri, Paolo Vezzulli, Sonia Calloni, Andrea Falini, Camilla Ferri, Arturo Chiti, Massimo Filippi, Federica Agosta

1IRCSS San Raffaele Scientific Institute, Neurology, Milan, Italy, 2IRCCS San Raffaele Scientific Institute, Neurology, Milan, Italy, 3IRCCS San Raffaele Hospital, Milano, Italy, 4IRCCS San Raffaele Hospital, Nuclear Medicine, Milano, Italy, 5Neurology Unit, Neurorehabilitation Unit, and Neurophysiology Service, Irccs Ospedale San Raffaele, Milan, Italy

Aims: To assess baseline amyloid burden and regional SUVR patterns across AD phenotypes and their changes after 6 months of anti-amyloid therapy.

Methods: Twenty AD patients received lecanemab (n=8) or donanemab (n=12) for at least 6 months. Mean age was 69 years; 60% female; mean disease duration 2.9 years. Fourteen patients had mild cognitive impairment (MCI) and six mild dementia. Clinical phenotypes included classic AD (n=10), early-onset AD (EOAD; n=8), primary progressive aphasia (PPA; n=1) and posterior cortical atrophy (PCA; n=1). Six patients (30%) were APOE ε4 heterozygous. Baseline MMSE was 24.8, CDR 0.6, CDR-SB 3. Baseline amyloid-PET centiloids averaged 87.1 (±23), with regional SUVRs extracted using Combinostics (reference: total cerebellum). ANCOVA models assessed SUVR differences across clinical phenotypes, adjusting for age and disease duration. Eleven patients completed 6-month follow-up PET scan.

Results: Baseline centiloids did not differ between MCI (83.4) and mild dementia (95.6; p=0.29), including after adjustment. The highest baseline regional SUVRs were in anterior cingulate (1.68), followed by precuneus, prefrontal and parietal regions. After covariate adjustment, EOAD showed higher SUVRs than other phenotypes in prefrontal (p=0.01), lateral temporal (p=0.05), parietal (p=0.02), sensorimotor (p=0.02), and caudate (p=0.03) areas. At follow-up, six patients (all receiving donanemab; p=0.03) had PET negativity (<12 centiloids). Mean centiloid reduction was 64.7 for donanemab and 50.7 for lecanemab. MCI showed greater clearance than mild dementia (71.6 vs. 32.1; p=0.042), attenuated after adjustment (p=0.055). Regional clearance was higher in MCI than mild AD in prefrontal, parietal, occipital regions, and precuneus. No significant global or regional SUVR differences at 6 months were observed across phenotypes or treatment groups.

Conclusions: Baseline and 6-month amyloid reduction varied by clinical phenotype. EOAD had higher baseline SUVRs. MCI had greater reductions than mild dementia.

Immunotherapy for Parkinson’s Disease and Alzheimer’s Disease: a Promising Disease- Modifying Therapy

Anns Mahboob, Hasan Ali, Aljazi Alnaimi, Mahmoud Yousef, Mlaak Rob, Nawaf Al-Muhannadi, Kalana Senevirathne, Ali Chaari

1Weill-Cornell Medicine-Qatar, Doha, Qatar, 2Weill Cornell Medicine - Qatar, Doha, Qatar, 3Weill Cornell Medicine - Qatar, Medical, Doha, Qatar

Aims: Alzheimer’s disease (AD) and Parkinson’s disease (PD) are two of the most prevalent neurodegenerative disorders, contributing substantially to global morbidity, mortality, and socioeconomic burden. Current therapeutic strategies largely offer symptomatic relief without altering disease progression. This review aimed to evaluate the potential of antibody-based immunotherapies, particularly monoclonal antibodies (mAbs), as disease-modifying approaches for AD and PD, focusing on their mechanistic rationale, clinical progress, and future promise.

Methods: A narrative review of recent preclinical and clinical studies was conducted, emphasizing immunotherapies that target pathological proteins implicated in AD and PD. Key targets included amyloid-beta (Aβ) and tau (τ) in AD, and alpha- synuclein (α-syn) in PD. Data were synthesized to assess therapeutic efficacy, safety, and translational challenges, with particular attention to trial design, biomarker development, and patient stratification.

Results: Emerging evidence suggests that mAbs targeting Aβ, tau, and α-syn hold promise in modifying disease trajectories. Clinical trials of Aβ-directed antibodies in AD have demonstrated reductions in amyloid burden and modest cognitive benefits, though safety concerns, particularly amyloid-related imaging abnormalities (ARIA), remain significant. For PD, α-syn–directed therapies are in earlier stages of development, with encouraging preclinical findings but limited clinical validation to date. Across both diseases, barriers to success include limited blood–brain barrier penetration, variable patient responses, and challenges in measuring disease-modifying outcomes.

Conclusions: Monoclonal antibody–based immunotherapies represent a promising frontier for disease modification in AD and PD. While progress has been made, particularly in AD, substantial challenges remain before these therapies can be widely implemented. Future directions include combination approaches, refinement of biomarkers for early diagnosis and treatment monitoring, and the development of personalized therapeutic strategies to optimize efficacy and safety.

ALDH2 Activation Alleviates Neuroinflammation and Neuronal Apoptosis

MECHANISMS, PATHOPHYSIOLOGY / MICROGLIA Wu Kuan-Hsuan, Ching-Chi Chiu,2

1Graduate Institute of Biomedical Science, College of Medicine, Chang Gung University, Taoyuan, Taiwan, 2Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, Taoyuan, Taiwan

Aims: Aldehyde dehydrogenase 2 (ALDH2) is a mitochondrial enzyme involved in removing reactive aldehydes generated during oxidative stress, such as 4-hydroxy-2-nonenal (4-HNE). ALDH2 also suppresses activation of the NLRP3 inflammasome pathway. ALDA-1, a small-molecule ALDH2 activator, has demonstrated anti-inflammatory effects, potentially offering therapeutic benefits in neurodegenerative diseases. TOMM40, a mitochondrial outer membrane protein, has been associated with increased AD risk. In our previous study, we identified two TOMM40 missense mutations (F113L and F131L) from early-onset AD patients, both of which were found to impair mitochondrial function. In this study, we investigated the neuroprotective effects of ALDA-1 in BV2 microglial cells expressing wild-type or mutant TOMM40.

Methods: BV2 microglial cells were transfected with wild-type or mutant (F113L or F131L) TOMM40 expression plasmids. BV2 microglia were pre-treated with ALDA-1 for 6 hours to assess its effects on mitochondrial function and inflammatory response. Mitochondrial ROS levels and mitochondrial membrane potential were measured using MitoSOX and TMRE molecular probes, respectively. Pro- inflammatory cytokines were quantified using ELISA. To assess neurotoxicity, conditioned medium from treated or untreated BV2 cells was applied to HT22 hippocampal neurons, and cell viability was evaluated using the CCK-8 assay.

Results: ALDA-1 restored ALDH2 activity and reduced 4-HNE accumulation in BV2 microglia expressing mutant TOMM40, thereby attenuating microglial activation and lowering the levels of pro- inflammatory cytokines (IL-1β, TNF-α, IL-6). In addition, ALDA-1 significantly reduced mitochondrial ROS production and restored mitochondrial membrane potential. Importantly, conditioned medium from ALDA-1-treated BV2 cells significantly alleviated the cell death of HT22 hippocampal neurons induced by conditioned medium from BV2 microglia expressing mutant TOMM40.

Conclusions: These findings suggest that ALDH2 activation by ALDA-1 may serve as a potential therapeutic agent for preventing or treating AD by targeting microglial mitochondrial dysfunction and neuroinflammation.

PRIMUS-AD: Design of the Phase 2 Study Treating Early AD Patients with the Novel Drug Candidate Pri-002

DEVELOPMENT, CLINICAL TRIALS / AGGREGATION INHIBITORS Dagmar Jürgens, Gerhard Tischler, Alexander Brener, Tina Bartsch, Gunther Kauselmann, Knut Adermann, Kathrin Zeiger, Katja Lindner, Julie-Anne Gabelich, Dieter Willbold,7, Oliver Peters,9

1Priavoid GmbH, Clinical Development, Düsseldorf, Germany, 2Prinnovation GmbH, Clinical Operation, Düsseldorf, Germany, 3Prinnovation GmbH, Ceo, Düsseldorf, Germany, 4Prinnovation GmbH, Clinical Development, Düsseldorf, Germany, 5Priavoid GmbH, Cmc, Düsseldorf, Germany, 6Priavoid GmbH, Düsseldorf, Germany, 7Heinrich-Heine-Universität Düsseldorf, Institut Für Physikalische Biologie, Düsseldorf, Germany, 8Priavoid GmbH, Chief Medical Officer, Düsseldorf, Germany, 9Charite, 3department Of Psychiatry And Neuroscience, Memory Clinic, Berlin, Germany

Aims: The drug candidate PRI-002 is an all-D-enantiomeric peptide and was developed to disassemble toxic Aβ oligomers into harmless Aβ monomers. It is anticipated that PRI-002 reduces neurotoxicity and restores synaptic plasticity in early stages of AD. In studies in healthy volunteers and in patients with early AD the orally administered PRI-002 has demonstrated to be safe and well tolerable.

Methods: The PRImus-AD study (EU clinical trial number 2022-503148-41) is a randomized, double- blind, placebo-controlled study to assess safety and efficacy of PRI-002 in patients with MCI or mild dementia due to Alzheimer’s Disease (AD). CSF or Amyloid-PET is used to confirm AD pathology. The trial is conducted in six European countries at 40 study sites. Eligible patients were randomly assigned (1:1:1) to receive 300 or 600 mg PRI-002 per day or placebo following an adaptive design for 48 to 96 weeks of treatment duration. Stratification factors are disease state (MCI or mild dementia) and APOE e4 status. A follow-up assessment is planned 12 weeks after the end of treatment. In addition to the primary objective of safety, efficacy will be assessed based on the change from baseline to week 48 in global outcome as measured by CDR-SB.

Results: 540 patients in total were screened. A screening failure rate of 42 % resulted in 304 eligible patients randomly assigned to the treatment arms. A blinded sample size recalculation confirmed the initial calculation of subjects to be enrolled. ARIA monitoring is closely performed for the first 90 patients until week 24.

Conclusions: Conclusion: The PRIMUS-AD phase 2 study enrolled 304 patients with MCI and mild dementia due to AD to assess safety and therapeutic benefit of PRI-002. Results are expected in Q3 2026.

Lecanemab for Early Alzheimer’s Disease: 48-month Results for APOE Ε4 Non- Carriers or Heterozygotes from the Clarity AD Open-Label Extension

DEVELOPMENT, CLINICAL TRIALS / AMYLOID CLEARANCE Lutz Froelich, Shobha Dhadda, Michio Kanekiyo, Amanda Goodwin, Mark Hodgkinson, Steven Hersch, Michael Irizarry, Lynn Kramer

1Medical Faculty Mannheim, University of Heidelberg, Central Institute of Mental Health, Mannheim, Germany, 2Eisai Inc., Nutley, United States of America

Aims: In Clarity AD, the Aβ-directed antibody lecanemab reduced markers of amyloid and slowed decline of cognition and function in early Alzheimer’s disease (AD). Lecanemab is associated with amyloid-related imaging abnormalities (ARIA), particularly in ApoE ε4 carriers. Herein, we report 48- month findings from the ongoing Clarity AD open-label extension (OLE) study for ApoE ε4 non-carriers and heterozygotes.

Methods: Clarity AD is an 18-month, randomized study (Core) followed by an OLE phase in individuals with early AD. Clinical (CDR-SB, ADAS-Cog14, and ADCS-MCI-ADL) outcomes and safety were evaluated from OLE data out to 48 months. Continued lecanemab treatment in the OLE was compared to matched control data from Alzheimer's Disease Neuroimaging Initiative (ADNI) data and the Swedish BioFINDER1 study. Results were evaluated in ApoE ε4 non-carriers and heterozygotes in Clarity AD.

Results: In the ApoE ε4 non-carrier and heterozygote population, 1521 participants were treated with lecanemab across the Clarity AD Core and OLE. No new safety signals were observed. ARIA rates were low and after 6 months, were similar to ARIA rates on placebo. Lecanemab treatment continued to delay progression through 48 months compared to the matched control datasets, with differences in CDR-SB increasing over 18 months (ADNI:0.50; BioFINDER:0.66), 36 months (ADNI:0.87; BioFINDER:1.59), and 48 months (ADNI:1.53; BioFINDER:2.37). Consistent rates of clinical stability or improvements were observed across assessments regardless of baseline tau levels. Across the 48- month period lecanemab reduced the relative risk of progression to next stage of disease versus matched placebo controls.

Conclusions: In ApoE ε4 non-carriers and heterozygote treated with lecanemab through 48 months, there was an increasing treatment difference compared to matched control data, consistent with a disease-modifying effect. The long-term safety profile of lecanemab was confirmed.

Demographic, Cognitive, and Biomarker Patterns in MCI and Alzheimer’s Disease (AD) Among Patients Initiating Donanemab in the Real-World

DEVELOPMENT, CLINICAL TRIALS / AMYLOID CLEARANCE Ann Hartry, Andrew Spargo, Akash Bhushan, Vamsi Chandra Kasivajjala, Aparajeeta Priyadarasani, Anand Srirampurapu

1Eli Lilly and Company, Indianapolis, United States of America, 2Neurodiscovery AI, Real-world Evidence & Research Analytics, Bangalore, India, 3Neurodiscovery AI, Alpharetta, United States of America, 4Neurodiscovery AI, London, United Kingdom

Aims: Baseline characterization of patients initiating donanemab in the real world can provide insights into the application of appropriate use recommendations in routine care. This analysis summarizes pre-treatment MCI and AD diagnoses, MMSE scores, biomarker testing, and APOE genotype patterns.

Methods: We performed a retrospective study of health record data (NeuroDiscovery) from patients receiving at least one dose of donanemab through Q3 2025. Pre-treatment diagnoses for MCI (G31.84) and AD (G30.x) were captured, and disease severity was classified using ICD-linked codes and MMSE categories. Age and gender were taken from the earliest pre-treatment record. Biomarker testing, including amyloid PET, CSF, and plasma measures, and APOE results were summarized.

Results: The donanemab-treated cohort included 580 patients, with a balanced gender distribution (55.9% female) and a predominantly White population. Overall, 83.9% of patients were aged 71 years or older. MMSE data were available for 54% of patients, with scores consistent with MCI (MMSE>23) in 49.5% of cases and with mild AD (MMSE 20-23) in 44.1%. Amyloid and P-tau biomarker results were available for nearly all patients. APOE genotyping was available for approximately half of the initiating patients, showing a higher proportion of non-ε4 carriers, followed by ε4 carriers and a smaller group of ε4 homozygous individuals. ICD-based classification at treatment initiation showed that approximately two-thirds of patients had an AD diagnosis, with the remaining one-third classified as MCI. The mean time from the most recent biomarker to treatment initiation was 52 days, and the mean time from the first biomarker to treatment was 225 days.

Conclusions: This Real-world evidence study provides demographic and diagnostic distributions to support safety and effectiveness analyses as well as guidance for more efficient paths to appropriate treatment.

Biomarker-positive APOE4 Carriers with MCI Show Disease Stability over 4-years of Valiltramiprosate/alz-801 Treatment: MMSE Responder Analysis from the Alz-801 Phase 2 Long-Term Extension Study

DEVELOPMENT, CLINICAL TRIALS / AMYLOID CLEARANCE Patrick Kesslak, Susan Abushakra, Katerina Sheardova, Jakub Hort, Niels Prins, Margaret Bray, Adem Albayrak, Aidan Power, Martin Tolar, John Hey

1Alzheon, Framingham, United States of America, 2International Clinical Research Center, Brno, Czech Republic, 3Second Faculty of Medicine, Charles University and Motol University Hospital, Neurology, Prague, Czech Republic, 4Brain Research Center Amsterdam, Amsterdam, Netherlands

Aims: Oral valiltramiprosate/ALZ-801 is an inhibitor of amyloid oligomer formation in development as a disease-modifying treatment for Alzheimer’s disease (AD). In this Phase 2 single-arm study in APOE4 carriers with Early AD (Hey 2024), valiltramiprosate showed cognitive stability over 156 weeks/3 years in the MCI subgroup (Hey, ADPD 2026). Disease stability by MMSE was analyzed over 208 weeks.

Methods: Phase 2, single-arm, study enrolled 53 APOE3/4 and 31 APOE4/4 Early AD subjects (MMSE ≥22, CDR-G= 0.5, 1) with positive amyloid-PET or CSF AD biomarkers; MCI defined as MMSE >26. MMSE was monitored over 208 weeks. Disease stability (responder) was defined as ≤ 4-point change from baseline to 104, 156 and 208 weeks.

Results: 84 subjects enrolled; 51% female, mean age 69 years, MMSE 26.0, 36/84 had MCI at baseline; 70, 65 and 51 subjects completed 2, 3 and 4 years on valiltramiprosate. Observed case analysis shows % MCI responders at each timepoint. Adverse events in >10% of subjects included COVID-19, nausea, urinary tract infection, decreased appetite, decreased weight and vomiting. There were no ARIA-E cases. MCI Stable Stable Stable Subjects/ APO4 MMSE Week 104 MMSE Week 156 MMSE Week 208 genotype (n, %) (n, %) (n, %) MCI Both 26/31 (84%) 21/30 (70%) 14/24 (58%) genotypes 14/17 (82%) 12/14 12/17 (71%) 9/13 6/13 (46%) 8/11 APOE3/4 APOE4/4 (86%) (69%) (73%)

Conclusions: Valiltramiprosate treatment of APOE4-carrier MCI subjects with confirmed AD pathology showed high proportion of subjects with stable disease stage over 2-4 years (MMSE drop ≤4). Cognitive stability in APOE4/4 MCI group remained high with 86% and 73% at 2 and 4 years. Consistent with the Phase 3 results (Abushakra, ADPD2026), valiltramiprosate shows promising benefit-risk profile in homozygotes with MCI.

NADALS Basket Trial: Drug Repurposing with Baricitinib, an FDA-approved JAK Inhibitor, in AD and ALS Patients

DEVELOPMENT, CLINICAL TRIALS / NEW CLINICAL TRIAL DESIGNS; SIMULATION OF PROGRESS- DIGITAL TWINS Mark Albers, Eric Macklin, Shayda Daneshvari, Mark Appleman, Ashlyn Butkowski, Taylor Stimpson, Lia Tamburello, Brenda Thornell, Reena Maharaj, Liam Dwyer, Yueyue Xiong, Kento Kawai, Patrick Reeves, Ruxandra Sirbulescu, Mark Poznansky, Merit Cudkowicz, Doreen Ho, James Berry, Pia Kivisäkk, Steven Arnold

1Massachusetts General Hospital, Department Of Neurology, Charlestown, United States of America, 2Massachusetts General Hospital, Department Of Neurology, Boston, United States of America, 3Massachusetts General Hospital, Vaccine And Immunotherapy Center, Charlestown, United States of America, 4Healey & AMG Center for ALS, Massachusetts General Hospital, Harvard Medical School, Boston, United States of America

Aims: Olumiant® (baricitinib) was FDA/EMA-approved for rheumatoid arthritis, COVID, and alopecia areata. Assessment of this potent JAK inhibitor on biomarkers in AD could provide insights into the role of neuroinflammation in AD disease pathophysiology and progression. We tested both pharmacokinetic and pharmacodynamic biomarkers of baricitinib’s action in AD patients and compared them with a parallel arm of ALS patients as well as an asymptomatic carrier of a pathogenic C9ORF72 gene.

Methods: NADALS was a prospective, observational, longitudinal, US study that enrolled adults with AD who initiated oral baricitinib. Participants were followed for 24 weeks to assess changes in established and candidate biomarkers, in relation to AD progression and baricitinib treatment. Assessments at screening, baseline and following 8 weeks (2 mg), and 16 and 24 weeks (4 mg), included biomarkers of AD pathology (Aβ, phospho-tau, and tau), inflammation, neuronal injury/death, smell function, cognitive function, and exploratory measures, including 32-marker phenotyping of immune cells in peripheral blood and CSF / serum proteomics with the SOMAlogic®, the Olink panel, and the NuLisa panel. Psychometric assessments were performed, and adverse events related to study procedures were tracked.

Results: 13 AD patients were consented, of whom 7 AD patients provided ≥1 CSF sample. 6 participants completed study procedures, 0 participants died, and 1 participant prematurely stopped participation. Biomarker and clinical data will be presented.

Conclusions: NADALS aimed to characterize neuroinflammatory and neurodegenerative biomarkers in baricitinib-treated patients with AD to inform an understanding of the association of brain exposure of baricitinib with peripheral and central inflammatory and neurodegenerative biomarkers in people with AD, ALS and asymptomatic gene carriers of C9ORF72.

A Dual 6-month & 18-month Randomized, Placebo-controlled, Double-Blind, Clinical Trial Investigating Efficacy and Safety of Buntanetap in Early Alzheimer's Participants

DEVELOPMENT, CLINICAL TRIALS / NEW CLINICAL TRIAL DESIGNS; SIMULATION OF PROGRESS- DIGITAL TWINS Cheng Fang, Melissa Gaines, Sarah Mccallum, Maria Maccecchini Annovis Bio, Malvern, United States of America

Aims: In a Phase 2b dose-ranging study in participants with mild to moderate Alzheimer’s disease (AD) (NCT05686044), buntanetap (also known as Posiphen) demonstrated a statistically significant cognitive improvement in plasma pTau217-positive participants with mild AD. These cognitive benefits were supported by biomarker findings showing that buntanetap reduced neurotoxic aggregation proteins, decreased inflammation, and improved neuronal function. To validate these effects, a new pivotal clinical trial in early AD was initiated in April 2025 (NCT06709014).

Methods: A prospective, double-blind, randomized, placebo-controlled Phase 3 study is currently ongoing. A total of 760 participants with early AD will be enrolled across 83 sites in the United States. This dual- study will include efficacy readout at both 6-month and 18-month, to evaluate buntanetap’s symptomatic and disease-modifying treatment effects.

Results: We will present updates on site activation status, screening procedures, screen failure rates and key drivers, enrollment pace, and the projected timeline for database lock.

Conclusions: In summary, the Phase 3 study in early AD is ongoing. We will share our progress to date, along with key insights and lessons learned throughout the conduct of the trial.

Data-Driven Approaches to Representative US Enrolment Planning in the Trontier Studies of Trontinemab in Early Symptomatic Alzheimer’s Disease

DEVELOPMENT, CLINICAL TRIALS / PERSONALIZED MEDICINES, SEX /RACE, AI, AND COMBINATION THERAPY Jessica Colby-Milley, Daniel Gautieri, Michelle Johnson, Kelly Markham-Coultes, Angeliki Thanasopoulou, Frank Perilli, Erin Beck, Ruth Croney

1F. Hoffmann-La Roche Ltd., Mississauga, Canada, 2SiteRx Inc., New York, United States of America, 3Roche Products Ltd., Welwyn Garden City, United Kingdom, 4F. Hoffmann-La Roche Ltd., Basel, Switzerland

Aims: Ensuring clinical trial populations represent the real-world diversity of people living with Alzheimer's disease (AD) is crucial to developing broadly beneficial therapies. This necessitates accessible trials that account for recruitment variations across demographics such as race, ethnicity, age, and sex. To achieve representative US enrolment in the Phase III trials of trontinemab in early symptomatic AD, TRONTIER 1 & TRONTIER 2, a data-driven approach for US site selection and recruitment planning was implemented.

Methods: A SiteRx real-world database, encompassing healthcare provider, patient, site and recruitment funnel data from over 50 AD studies were analyzed. Sites were characterized using a composite measure of "enrolment potential" based on proximate volume of people with AD (community and site database) and historical enrolment. Proximate volume was overlaid on a map of the US to identify geographies with high volume and density of underrepresented AD populations. The planned site list was overlaid to inform site selection activities. Screening funnel modelling, based on 13 previous AD trials with comparable inclusion/exclusion parameters, was performed to predict participant recruitment dynamics and calculate screen fail rates across protocol screening steps and demographic groups (race, ethnicity, age, and sex).

Results: Site characterization identified approximately 150 AD trial sites with strong representative enrolment potential. Heat mapping data identified 15 US geographies where trial sites could be included to increase trial access to all demographic groups. Screening funnel modelling created a tool to predict the number of participants needed at each stage of screening to achieve representative enrolment, and identified conversion rate variations across demographic groups.

Conclusions: Characterizing AD trial sites and modelling the screening funnel with real-world data provided valuable tools for data-driven US site landscaping and representative enrollment planning in the TRONTIER studies.

Blood-Based Biomarkers in Alzheimer’s Disease: Correlation Analysis, Longitudinal Trends, and the Effects of Donepezil

BIOMARKERS, DIAGNOSTICS / CSF, BLOOD, BODY FLUID BIOMARKERS William Baek Parkside Medical Group, Neurology, Upland, United States of America

Aims: To investigate 1. Correlations between MMSE scores and p-tau181, p-tau217, and Abeta42/40 ratio 2. Longitudinal trends in p-tau181, p-tau217, and Abeta42/40 ratio 3. If treatment affects p- tau181, p-tau217, and Abeta42/40 ratio

Methods: This was a retrospective chart review of 90 subjects diagnosed with AD between 2011 and 2025. At least 2 MMSE scores and p-tau181 results were required. Both patients with pre- and post- treatment BBM levels were included. Retrospective longitudinal correlation (Pearson, Spearman) analysis, per-subject linear slopes, mixed-effects models, paired pre→post changes (Wilcoxon) with p-values were used.

Results: 29 subjects were included in the final analysis. Cross-sectional correlations were observed between MMSE and nearest BBM values; r ≈ −0.34 for p-tau181 and MMSE, r ≈ −0.31 for p-tau217 and MMSE, and r ≈ +0.40 for Abeta 42/40 and MMSE. Correlation within each subject was close to zero. Both p-tau181 and p-tau217 showed an increasing trend over time. Abeta42/40 ratio showed a decreasing trend. Among subjects with pretreatment baseline BBMs, Donepezil was associated with statistically significant reductions in p-tau181 and increases in Abeta42/40 ratio. p-tau217 did not show a significant change.

Conclusions: Higher tau levels relate to worse and higher Abeta 42/40 ratio to better cognition. Both p-tau181 and p-tau217 showed an increasing trend over time, whereas Abeta42/40 ratio showed a decreasing trend, supporting the underlying pathology. No statistically significant correlations between MMSE and any BBM were observed at individual time points, though the direction of changes aligned with disease progression. Donepezil reduced p-tau181 levels and increased Abeta42/40 ratio, suggesting a change in the natural course of AD.

Evaluating Diffusion MRI as a Biomarker of Treatment Response in Amyloid- Targeting Alzheimer’s Disease Trials

BIOMARKERS, DIAGNOSTICS / MULTIMODAL IMAGING Susanna Gobbi, Antonio Cerdán Cerdá,3, Muhamed Barakovic, Erica Silvestri,4, Matteo Tonietto, Nicola Spotorno, Gregory Klein,6, Stefano Magon

1Roche Pharma Research and Early Development, Neuroscience and Rare Diseases Biomarkers, Basel, Switzerland, 2Instituto de Neurociencias, Consejo Superior de Investigaciones Científicas & Universidad Miguel Hernández, Sant Joan d’Alacant, Spain, 3Roche Pharma Research and Early Development, Basel, Switzerland, 4A4P Consulting Ltd., Sandwich, United Kingdom, 5Clinical Memory Research Unit, Department Of Clinical Sciences Malmö, Malmö, Sweden, 6F. Hoffmann-La Roche Ltd., Basel, Switzerland

Aims: To investigate whether diffusion MRI (dMRI) can detect changes resulting from amyloid- targeting immunotherapies.

Methods: This study analyzed multimodal imaging data (PET, volumetric and dMRI) from a subset of participants of the phase 3 gantenerumab GRADUATE I and II trials. Despite not meeting primary clinical endpoints, graduate is a rich dataset to investigate imaging association. The relationship between free-water corrected dMRI metrics (mean diffusivity (MD(FWC)), fractional anisotropy (FA(FWC)), free-water fraction (FWF)), amyloid load, and brain volume were assessed in region-of-interest (ROI) (cortical gray, white matter, Mayo and Dickerson composite regions). Baseline associations were tested using Pearson's correlations. Longitudinally, individual slopes for dMRI metrics, global centiloids, and volume change (Jacobian integration) were calculated using linear mixed-effect models. Analyses included: 1) within-group “slope-slope” partial Pearson’s correlations to related dMRI changes to amyloid and volume changes, 2) region-wise Spearman’s correlations of net treatment effects (gantenerumab minus placebo) on amyloid removal and dMRI metrics, and 3) linear models on dMRI slopes to assess the main treatment effect (gantenerumab vs. placebo). All the metrics and the slopes were adjusted for age, sex, ApoE ε4 status, and Amyloid-Related Imaging Abnormalities (ARIA).

Results: Participants in the gantenerumab-treated arm who exhibited faster global amyloid removal tended to show a decrease in FA(FWC) across most ROIs, along with increased rates of MD(FWC) in Mayo (Figure 1). Regions with greater net amyloid removal due to treatment demonstrated moderate greater relative net increases in MD(FWC) and FWF (Figure 2). The treatment analysis reported higher FWF rates in gantenerumab compared to placebo in white matter and Mayo (Figure 3).

Conclusions: This study shows novel longitudinal findings that dMRI metrics are dynamically altered by amyloid-targeting therapies.

Effectiveness and Safety of Neuralcim® in Mild/moderate Clinical Alzheimer's Syndrome Amnesic Variant in the Context of Routine Medical Practice.

TARGETS, MECHANISMS FOR TREATMENT / ANTI-INFLAMMATORY Eliseo Capote Leyva, Saily Sosa Pérez, Nelky Urrutia Amable, Leslie Pérez, Carmen Viada, Yaima Llanes García, Julio López Arguelles, Aleima V. Rodríguez Carvajal, Marbelys Guevara Rodríguez, Larry Medina Hernández, Katiuska Iglesias Mónagas, Adrián Pedroso Ríos, Odalmis Chacón Borrero, Elizabeth Colina Ávila, Mónica Rodriguez Montalván, Yoenia Debora Morales, Luis A. Rodríguez Sánchez, Melitza De La C. Pupo Machado, Pedro E. Labrada Aguilera, Mabel R. Oduardo Aguilar, Roberto L. Lotti Mesa, Niorvis Martin Abiague, Marité García Llano, Tania Crombet, Kalet León, Mayra Ramos Suzarte, Teresita Rodríguez

1Center of Molecular Immunology, Clinical Trials, Havana, Cuba, 2Hospital Iván Portuondo, Artemisa, Cuba, 3Hospital Comandante Pinares, Artemisa, Cuba, 4Hospital Gustavo Aldereguía Lima, Cienfuegos, Cuba, 5Hospital Arnaldo Milian Castro, Villa Clara, Cuba, 6Hospital Camilo Cienfuegos, Santi Spiritus, Cuba, 7Hospital Antonio Luaces, Ciego de Ávila, Cuba, 8Hospital Manuel Ascunce Domenech, Camagüey, Cuba, 9Hospital Saturnino Lora, Santiago de Cuba, Cuba, 10Hospital Juan Bruno Zayas, Santiago de Cuba, Cuba, 11Hospital Faustino pérez, Matanzas, Cuba, 12Hospital Roberto Rodríguez, Ciego de Ávila, Cuba, 13Hospital Ernesto Guevara, Las Tunas, Cuba, 14Hospital Lucía Iñiguez, Holguín, Cuba, 15Hospital Carlos Manuel de Céspedes, Granma, Cuba, 16Hospital Celia Sánchez Manduley, Granma, Cuba, 17Hospital Agustinho Neto, Guantánamo, Cuba, 18Hospital Abel Santamaría, Pinar del Río, Cuba

Aims: To evaluate the effectiveness and safety of NeuralCIM® in patients with mild-to-moderate clinical Alzheimer's syndrome, amnesic variant, according to the 2011 National Institute on Aging - Alzheimer's Association (NIA-AA) criteria and under routine medical practice conditions.

Methods: A multicenter, prospective, observational study was conducted under real-world conditions. NeuralCIM® was administered intranasally at 0.5 mg three times per week. 388 patients who reached week 52 of treatment and 187 patients who reached week 78 were evaluated. At baseline and at weeks 52 and 78, the following scales or instruments were performed: ADAScog11, CDR, GDS, MoCA, NPI, Cornell, Katz, Lawton, CIBIS (at baseline) and CIBIC plus (week 52 and 78).

Results: At week 52, 66.0% of patients had improvement (≤ -4 points) and stability (≥ -3 and ≤1 point) according to ADAScog11. These results were maintained at week 78 with 63% of patients with improvement and stability. This contrasts with the natural evolution of the disease where without treatment almost 90% of patients worsen (≥2 points). Regarding the remaining instruments used (CDR, GDS, MoCA, NPI, Cornell, and CIBIC plus), 60% or more of patients showed improvement or maintained the same scores at both weeks 52 and 78. Katz and Lawton showed that more than 80% of patients maintained the same level of dependence with which they began this study in relation to basic and instrumental activities of daily living. This contrasts with the progression towards worsening in this neurodegenerative disease. The majority of adverse events were non-serious and mild intensity. Serious adverse events related to NeuralCIM® accounted for 4.7% of the total adverse events.

Conclusions: NeuralCIM® shows encouraging results and a good safety profile in the treatment of mild-moderate clinical Alzheimer's syndrome, amnesic variant.

Longitudinal Neuropsychological Changes in Global and Domain-specific Cognition in Early Alzheimer’s Disease Treated with Lecanemab: a Prospective Study

TARGETS, MECHANISMS FOR TREATMENT / IMMUNOTHERAPY Naoko Kawabe, Satoshi Hosoki, Chiharu Watanabe, Nobuyuki Ohara, Kota Maekawa, Kazuma Ota, Daichi Imamura, Kotaro Haruna, Kunitaka Matsuishi, Hiroshi Komatsu, Kentaro Iwata, Nobuo Kohara,2, Michi Kawamoto

1Kobe City Medical Center General Hospital, Department Of Rehabilitation, Kobe, Japan, 2Kobe City Medical Center General Hospital, Department Of Neurology, Kobe, Japan, 3Kobe City Medical Center General Hospital, Department Of Psychiatry, Kobe, Japan

Aims: To clarify domain-specific longitudinal changes in cognitive function in patients with early Alzheimer's disease (AD) treated with lecanemab, an anti–beta-amyloid monoclonal antibody, as these changes remain insufficiently characterized in real-world practice.

Methods: We assessed early AD patients with confirmed amyloid pathology who received lecanemab at our institution between March 2024 and July 2025. Neuropsychological assessments included the Mini-Mental State Examination (MMSE), Clinical Dementia Rating–Sum of Boxes (CDR-SB), Logical Memory (LM), Frontal Assessment Battery (FAB), Rey–Osterrieth Complex Figure (ROCF), Trail Making Test, and letter fluency. Scores at baseline (BL), 6 months (6M), and 12 months post-treatment were compared using a linear mixed-effects model with particular attention to changes from BL to 6M. Subscale scores were examined using the Wilcoxon signed-rank test.

Results: Twenty patients were included (mean age 75.8 years [SD 6.4]; 40% male; mean education 13.1 years [SD 2.9]; mean CDR-SB 3.0 [SD 0.9]; APOE4: 50% heterozygotes, 15% non-carriers, 35% unknown). At 6 months post-treatment, MMSE declined (change −1.15, 95% CI −2.11 to −0.19, p=0.01), with a decrease in the Orientation subscale (p=0.02). CDR-SB also worsened (change +0.83, 95% CI 0.27–1.39, p=0.002), mainly due to declines in the Orientation (p=0.02) and Judgement (p=0.01) subscales. Among memory measures, LM delayed recall declined (change −0.83, 95% CI −1.43 to −0.23, p=0.004). Only FAB improved (change +1.14, 95% CI 0.08–2.20, p=0.03), with an increase in the Luria Motor Series subtest (p=0.03). Other measures showed no significant change.

Conclusions: In early AD patients treated with lecanemab, global cognition declined, particularly in orientation, and verbal memory also declined, whereas executive function showed potential improvement, most notably in motor programming.

A Multicenter Phase 2a Clinical Study to Evaluate Safety, PK and PD of Ntrx-07 in Subjects with MCI or Mild to Moderate AD

TARGETS, MECHANISMS FOR TREATMENT / MICROGLIA Tony Giordano, Joseph Foss, Marianna Kiraly NeuroTherapia, Inc, Cleveland, United States of America

Aims: NTRX-07, a novel cannabinoid CB2 agonist, has been demonstrated in numerous animal models to reduce inflammation, enhance amyloid clearance, and improve learning and memory. A 90 mg dose of NTRX-07 is being compared to placebo in subjects with AD following 28-days of treatment. While the primary objectives are safety and tolerability, CSF biomarkers of neuroinflammation and neuronal function, along with free-water MRI, qEEG and P300 ERP are being collected for analysis pre- and post- treatment.

Methods: This is a 28-day, placebo controlled, double-masked Phase 2a clinical trial.

Results: The trial was initiated in May of 2026. To date, 26 of the 28 subjects have been randomized, with 18 having completed the 28-days of treatment. No dose limiting toxicities have been observed. It is anticiapted that the trial will be completed in December f 2025, with top line data available in January of 2026. This data will be presented at the conference.

Conclusions: With more than half the subject randomized and approximately 1/3 completing the study, the treatment has been safe and well tolerated. Biomarker results, qEEG, P300 ERP, free water MRI and PK data is being collected and will be available in January 2026.

Memantine in the Treatment of Dementia in Hungary

AND COGNITIVE DYSFUNCTION / OTHER Petra Halmos, Dániel Bereczki, Balázs Dobi, Péter Vinnai, Tibor Kovács

1Semmelweis University, Doctoral College János Szentágothai Neurosciences Division, Budapest, Hungary, 2Semmelweis University, Department Of Neurology, Budapest, Hungary, 3Semmelweis University, Hun-ren Se Neuroepidemiological Research Group, Budapest, Hungary

Aims: The aim of this retrospective cohort study was to provide a comprehensive overview of the current state of the healthcare system, placing emphasis on pharmacological treatment - especially memantine - utilisation patterns and their association with survival outcomes.

Methods: Using the NEUROHUN database - containing inpatient, outpatient care data and pharmacy reports submitted for reimbursement - we conducted an analysis covering the entire population of Hungary between 2010 and 2020. Descriptive statistics, disease incidence and prevalence were calculated, followed by Kaplan-Meier analyses and Cox proportional hazard regression to evaluate survival outcomes. Sex, age at diagnosis, dementia subtypes, diagnosis switch, calculated adherence and persistence, and psychotropic drug use were also included as independent variables. All analyses were performed in R (version 4.5.1).

Results: We identified 398,584 patients with a diagnosis of dementia between 2010 and 2020. The mean age at diagnosis was 75.22 years (SD = 12.110), and 63.7% of patients were female. Of the total cohort, only 13.3% (n=53,070) received anti-dementia medication (memantine, donepezil, or rivastigmine); 12,757 (24.0%) of these were on memantine monotherapy, the group with the highest rate of psychotropic prescriptions. Kaplan–Meier analysis showed the shortest median survival in untreated patients (3.105 years, 95% CI: 3.083 - 3.129), followed by those on memantine monotherapy among treated groups (3.704 years, 95% CI: 3.606 - 3.784). The multivariable Cox proportional hazards model revealed that patients receiving a combination of memantine and donepezil had the lowest mortality risk (HR: 0.581, 95% CI: 0.566-0.597) compared to the untreated group.

Conclusions: These findings suggest a high proportion of delayed diagnoses, as memantine is indicated in advanced dementia stages. Timely diagnosis (particularly in the era of emerging disease-modifying therapies), along with evidence-based healthcare planning, is therefore inevitable.

Discovery of IGS01, a Highly Selective Muscarinic Acetylcholine M4 Receptor Positive Allosteric Modulator (PAM) with a Favorable DDI Risk Profile for Psychosis

NEURODEGENERATIVE DISEASES / DRUG DEVELOPMENT, CLINICAL TRIALS Junliang Hao, Yishan Sun, Chao Wang, Ye Tian, Jiayin Li, Qiong Tan, Pengcheng Huang, Q. Kevin Fang Ignis Therapeutics, Shanghai, China

Aims: Differentiated from standard-of-care antipsychotics, muscarinic acetylcholine M4 receptor modulators have sparked significant interest for their potential to provide a superior benefit-risk profile in treating psychosis. Positive and selective modulation of the M4 receptor fine-tunes neural circuits underlying psychosis while minimizing gastrointestinal and other side effects associated with stimulation of other muscarinic subtypes. The selective M4 PAM emraclidine demonstrated efficacy signals in Phase 1b, but missed endpoints in two Phase 2 studies, triggering multifaceted re-thinking on better ways to harness the M4-PAM MoA. One noteworthy aspect of emraclidine is its exclusive metabolism by CYP3A4, which raises concerns about potential drug-drug interactions (DDIs) and limits its use in geriatric patients (e.g. AD psychosis). To overcome this limitation, Ignis initiated a program to identify a selective M4 PAM with reduced DDI liability.

Methods: A SAR campaign guided by an internally resolved cryo-EM structure of the emraclidine-M4- Gi complex was conducted. Compounds were advanced through a flow scheme that focused on ADME properties and DDI risk assessment.

Results: IGS01 displayed comparable potency to emraclidine in a human M4 cAMP assay with minimal activities at other muscarinic receptors and across a 98-target secondary pharmacology panel. In preclinical studies, IGS01 dose-dependently reversed amphetamine induced hyperlocomotion and rescued amphetamine induced sensory gating deficits in rats. It exhibited favorable PK properties with low to moderate in vivo clearance (m/r/d/mky). Importantly, IGS01 was metabolized by multiple CYP isoforms with a fm,cyp3A4 < 0.5, suggesting significantly lower DDI risk compared to emraclidine.

Conclusions: IGS01 has completed GLP toxicology studies en route to an IND application. It represents a new opportunity to harness M4 activation for the treatment of psychosis in geriatric patients that demands a favorable DDI risk profile.

Long-Term Safety of Bepranemab: Data from Phase Ii/ib Open-Label Extension Studies in Prodromal–mild Alzheimer’s Disease and Progressive Supranuclear Palsy

CLINICAL TRIALS / TAU CLEARANCE Matthew Barton, Tim Buchanan, Alexander Klein, William Byrnes, Irene Rebollo Mesa, Steven De Bruyn, Bart Van Den Steen, Hans Van Tricht, Katie Taiyari, Jos Bloemers, Saori Shimizu, Nancy Strong, Ute Massow, Ralph Maguire, Colin Ewen

1UCB, Asset Leadership, Bepranemab, Raleigh, United States of America, 2UCB, Clinical Science, Braine-l’Alleud, Belgium, 3UCB, Medical Neurology Therapeutic & Disease Area Leadership, Monheim am Rhein, Germany, 4UCB, Global Clinical Development, Raleigh, United States of America, 5UCB, Biometrics & Data Science, Madrid, Spain, 6UCB, Clinical Science, Braine-l'Alleud, Belgium, 7UCB, Global Clinical Science & Operations, Braine-l'Alleud, Belgium, 8UCB, Biometrics & Data Science, Slough, United Kingdom, 9UCB, Patient-centered Outcomes Research, Breda, Netherlands, 10UCB, Global Clinical Development, Brussels, Belgium, 11UCB, Biometrics & Data Science, Smyrna, United States of America, 12UCB, Global Clinical Development, Monheim am Rhein, Germany, 13UCB, Patient Solutions, Braine-l'Alleud, Belgium, 14UCB, Early Solutions, Slough, United Kingdom

Aims: Bepranemab is a recombinant, humanised, full-length immunoglobulin G4 monoclonal antibody, targeting a mid-region tau epitope essential for tau aggregation. We describe results from two open-label extension (OLE) studies assessing long-term safety of bepranemab in Alzheimer’s disease (TOGETHER OLE [NCT04867616]) and in progressive supranuclear palsy (PSP), a rapidly progressing neurodegenerative disorder (PSP002 [NCT04658199]).

Methods: TOGETHER was a, global, multicentre, placebo-controlled, Phase II study. Following the 80- week double-blind period, participants entering the 48-week OLE continued the same bepranemab dose, or if previously on placebo, were re-randomised 1:1 to bepranemab 45/90 mg/kg (intravenous [IV] every 4 weeks [Q4W]) for ≤44 weeks, followed by a safety follow-up period (20 weeks after the last infusion). The OLE focused on long-term safety, including treatment-emergent adverse events (TEAEs) from OLE Baseline to study end. PSP002 was an OLE of PSP003 (NCT04185415), a double-blind, placebo-controlled, Phase Ib study of bepranemab (90 mg/kg, IV Q4W over 48 weeks) in PSP. In PSP002, participants (aged ≥40 years) met Movement Disorder Society-PSP criteria for possible/probable PSP and completed the PSP003 treatment period. Exclusion criteria included meeting mandatory withdrawal and/or stopping criterion in PSP003. PSP002 primary endpoint: incidence of TEAEs from OLE Baseline to Month 60.

Results: TOGETHER: Of the 466 participants randomised into TOGETHER, 387 participants were dosed during the OLE period and 312 completed the OLE period. PSP002: Of the 25 participants randomised into PSP003, 19 entered the OLE; the PSP002 study is ongoing with a small number of participants. Safety results from the TOGETHER and PSP002 OLEs will be presented.

Conclusions: Results from the TOGETHER and PSP002 OLEs will provide further information about the long-term safety of bepranemab.

Nidb-3101: a Third Generation Anti-TAU Biparatopic Antibody Designed for Enhanced Therapeutic Efficacy in AD

MECHANISMS FOR TREATMENT / IMMUNOTHERAPY Jean-Philippe Courade, Eric Hatterer, Lucile Broyer, Fabien Cusin, Mathieu Clavel, Jeremy Bourguignon, Margot Bonnet-Gamard, Pauline Malinge, Franck Gueneau, Valery Moine, Nadia Anceriz, Bart Klein, Sara Pecoraro, Thuur Zuidweg, Simsek Naz, Kristin Piche, Wiep Scheper, Wilma Van De Berg,6

1Discoveric bio alpha, Pfäffikon, Switzerland, 2Light Chain Bioscience-Novimmune S.A,, Geneva, Switzerland, 3Vrije Universiteit, Section Functional Genomics, Dept Human Genetics, Amsterdam Umc, Amsterdam, Netherlands, 4Section Clinical Neuroanatomy and Biobanking, Department of Anatomy and Neurosciences, Amsterdam UMC, Vrije Universiteit, Amsterdam, Netherlands, 5StressMarq Biosciences Inc., Victoria BC, Canada, 6Amsterdam Neuroscience, Program Neurodegeneration, Amsterdam, Netherlands

Aims: Emerging phase 2 data from second-generation anti-Tau antibodies indicate significant effects on Tau pathology and cognitive decline (discussed by Höglinger, abstract #776 ). This abstract summarizes key findings from these developments and presents the rationale and design of NIDB- 3101, a novel third-generation fully human IgG1 anti-Tau biparatopic bispecific antibody (bsAb).

Methods: NIDB-3101 was engineered to bind two distinct Tau epitopes (confirmed with overlapping peptide binding and crystallography), improving target coverage (Figure 1A). The bsAb was optimized for superior affinity compared to benchmarks, Fc-engineered for extended half-life via FcRn recycling and reduced neuroinflammation risk through low Fc-effector function. Binding and functional characterizations were conducted using human recombinant Tau proteins, and PK/PD modeling was employed to assess the in vivo dose efficacy of NIDB-3101 in Alzheimer's patients compared to benchmarks. Binding and function using Human AD extracts is presented by Van de Berg et al., abstract #1046

Results: NIDB-3101 demonstrated high-affinity binding to Tau fibrils (Figure1B), surpassing monospecific controls and benchmarks. ELISA and SPR confirmed binding to recombinant human Tau fibrils, while electron microscopy revealed a higher ratio of antibodies to Tau fibrils compared to benchmark antibodies. In a cellular aggregation assay, NIDB-3101 exhibited superior functional inhibition. Fc mutations effectively reduced FcγR binding and increased FcRn affinity at pH 6, enhancing recycling. Integrating these biological data facilitated the prediction of NIDB-3101's half- life and PK/PD modeling, indicating efficacy at lower doses than benchmarks. Figure 1: (A) NIDB-3101 recognizing MTBR and C-terminal Tau. (B) NIDB-3101 binding (ELISA) to recombinant 2N4R Tau fibrils. NIDB-3101 displays an enhanced binding potency over benchmarks.

Conclusions: Biparatopic NIDB-3101's high-affinity binding to diverse Tau species, half-life extension mutations, and low effector function position it as a promising therapeutic candidate for Alzheimer's disease.

The Mechanisms of Tmem106b Protective Variant in Cellular and Pathological Processes of Neurodegeneration

DISEASES / CELL, MOLECULAR AND SYSTEMS BIOLOGY Sini-Pauliina Juopperi, Roosa-Maria Willman, Henna Martiskainen, Nadine Huber, Susanna Kemppainen, Sami Heikkinen, Anssi Lipponen, Annakaisa Haapasalo, Ville Leinonen,4, Tarja Malm, Mikko Hiltunen

1Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland, 2A. I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio, Finland, 3Unit of Neurosurgery, Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland, 4Department of Neurosurgery, Kuopio University Hospital, Kuopio, Finland

Aims: Transmembrane protein 106B (TMEM106B) is a lysosomal membrane protein expressed in neurons and glia. Recent studies have shown that the C-terminal domain of TMEM106B forms amyloid filaments in the brain of patients with various neurodegenerative diseases (NDDs), including tauopathies, β-amyloid-, α-synuclein-, and TDP-43-proteinopathies, as well as in elderly individuals without brain disease, in an age dependent manner. Whether the filaments contribute to pathogenesis or are only bystanders in disease, remains unclear. Genetic variants in TMEM106B are linked to an altered risk for frontotemporal dementia and Alzheimer’s disease. A threonine-to-serine substitution at position 185 (TMEM106B T185S; rs3173615 C>G) has been associated with a protective effect. The cellular mechanisms underlying the risk-modulating effect are still elusive.

Methods: Here, cellular mechanisms of the TMEM106B rs3173615 variant will be investigated in human induced pluripotent stem cell (iPSC) -derived microglia (iMG), brain biopsies of idiopathic normal pressure hydrocephalus (iNPH) patients, and postmortem brain samples of NDD patients. iPSCs were generated from skin fibroblasts of individuals who are homozygous carriers of the protective allele (GG) and non-carriers (CC). The iPSCs will be differentiated into iMG and treated with lipopolysaccharide (LPS), followed by omics-analyses and functional experiments. Cortical biopsies from iNPH patients and postmortem brain samples from NDD patients carrying rs3173615 CC, CG, or GG, will be examined using immunohistochemical stainings including IBA1, β-amyloid, CD68, LAMP1, GFAP, synapsin I, TDP-43, and TMEM106B antibodies.

Results: Staining of inferior temporal cortex samples of 78 NDD patients with TMEM106B C-terminal antibody revealed that TMEM106B accumulation was significantly increased in later Braak stages V-VI compared to stages I-IV.

Conclusions: Forthcoming analyses aim to clarify how the rs3173615 variant influences TMEM106B function, filament formation, and its role in neurodegeneration.

Safety and ARIA Analyses of the Oral Anti-amyloid Oligomer Agent Valiltramiprosate in Apoe4/4 Homozygotes with Early AD: Results of the 78-week Phase 3 Apolloe4 Trial

AGGREGATION INHIBITORS David Watson, Patrick Kesslak, Jerome Barakos, Susan Abushakra, Sharon Cohen, Emer Macsweeney, Karim Bennys, Mercè Boada, Earvin Liang, Rosalind Mclaine, Susan Flint, Aidan Power, John Hey, Martin Tolar

1Alzheimer’s Research and Treatment Center, Wellington, United States of America, 2Alzheon, Framingham, United States of America, 3Clario Inc, Philadelphia, United States of America, 4Toronto Memory Program, Toronto, Canada, 5Re:Cognition Health, London, United Kingdom, 6Montpellier university, Montpellier, France, 7Ace Alzheimer Center Barcelona, Barcelona, Spain

Aims: Valiltramiprosate/ALZ-801, a valine-conjugated prodrug of tramiprosate that prevents Aβ toxic oligomer formation, was evaluated in APOE4/4 homozygotes with Early AD. APOE4 is a risk factor for cerebral amlyloid angiopathy and ARIA. This 78-week placebo-controlled trial included subjects with any number of baseline microhemorrhages (MH) and several superificial siderosis lesions.

Methods: APOE4/4 homozygotes with Early AD (MMSE ≥22) were randomized 1:1 (placebo:265 mg BID), MRIs were conducted every 26 weeks, with centralized evaluations for ARIA-E/H. Anti-platelets were allowed, anti-coagulants were exclusionary.

Results: Safety population (N=325); 51% females, 89% Caucasian, mean age 68 years, MMSE 25.6; 298 had screening/baseline and 290 had post-treatment MRI. At baseline, 32% had >1 MH (placebo, 32%, active, 31%); 11% of placebo and 8% active had siderosis (maximum 4 and 5 lesions, respectively); and 5% per/arm had ≥ 10 MH. ARIA-E occurred in 5 subjects/arm (3%) and macrohemorrhages in 2 subjects/arm. Increase in MH from low to high category (0, 1-4, 5-9, ≥10 MH) occurred in 14% placebo and 9% active. Radiographically all ARIA-E were mild except 1 severe (placebo) and 1 moderate (active), and all ARIA-E/H events were asymptomiatic. AEs with incidence >5% and >2x placebo rate were nausea (26%), weight decrease (14%), decreased appetite and vomiting (10% each); with lower headache, falls and dizziness rates in active arm.

Conclusions: Valiltramiprosate was well tolerated with favorable safety over 78 weeks consistent with prior studies in >3000 AD patients. ARIA-E rates were similar between valiltramiprosate and placebo arms (~3%), ARIA-H was lower than placebo, and no ARIA events were symptomatic. Coupled with promising efficacy in MCI subjects, valiltramiprosate provides a favorable benefit-risk profile for high-risk APOE4/4 patients with unmet need for safe and effective disease-modifying treatments.

Positive Effects of Oral Valiltramiprosate on Grey and White Matter Microstructural Integrity by DTI in Apoe4/4 Homozygotes with Early AD: Apolloe4 78-week Phase 3 Results

AMYLOID CLEARANCE Earvin Liang, Susan Abushakra, P Doraiswamy, Duygu Tosun-Turgut, David Watson, Aidan Power, Sharon Cohen, Emer Macsweeney, Jennifer Lynch, Chris Conklin, Madhura Ingalhalikar, John Hey, Patrick Kesslak, Susan Flint, Martin Tolar

1Alzheon, Framingham, United States of America, 2Duke University, Durham, United States of America, 3University of California, San Francisco, Department Of Radiology And Biomedical Imaging, San Francisco, United States of America, 4Alzheimer’s Research and Treatment Center, Wellington, United States of America, 5Toronto Memory Program, Toronto, Canada, 6Re:Cognition Health, London, United Kingdom, 7NeuroClin Glasgow Ltd, Glasgow, United Kingdom, 8Clario, San Mateo, United States of America

Aims: Valiltramiprosate/ALZ-801, an oral anti-amyloid oligomer agent was evaluated in 78-week Phase 3 trial in APOE4/4 homozygotes with Early AD. In overall population valiltramiprosate did not show clinical efficacy but showed significant slowing of brain atrophy on volumetric MRI. Pre- specified MCI subgroup showed positive clinical and vMRI effects. Brain microstructural integrity was evaluated using diffusion tensor imaging (MRI-DTI).

Methods: Study enrolled 325 homozygotes (162 placebo, 163 valiltramiprosate); ~40% were MCI (MMSE >26) and 60% Mild AD (MMSE, 22-26). ADAS-Cog13 and CDR-SB were primary and key secondary outcomes, respectively; hippocampal volume (HV) was the main vMRI outcome. Mean water diffusivity (DTI-MD) in grey and white matter (GM,WM) were centrally analyzed by Clario Inc. Lower DTI-MD indicates tissue preservation. MMRM was primary analysis; Pearson’s correlations were conducted.

Results: Clinical/vMRI datasets included 320/290 subjects respectively; DTI included 85 MCI, 123 Mild subjects. In overall population, valiltramiprosate effects on MD showed positive trends in 4 GM regions (p<0.1), and significant WM effects [corpus callosum genu (GCC) (p<0.001), fornix (p=0.007), corpus callosum (p=0.023), and corona radiata (p=0.037)]. MCI showed GM effects in caudate (p=0.0095) and cingulate cortex (p=0.015); WM effects in GCC/whole CC/splenium CC (p=0.003/0.014/0.037), fornix (p=0.032), and corona radiata (p=0.042). In MCI active arm, MD effects in several GM/WM areas correlated significantly with clinical (correlations ≥0.34, p<0.03) and volumetric measures (≥ 0.56, p<0.001).

Conclusions: In the pre-specified MCI group, valiltramiprosate reduced mean water diffusivity in grey/white matter regions, most prominent in areas affected early in MCI (cingulate cortex/ fornix/corpus callosum). These effects correlated significantly with its positive clinical/brain-volume benefits, suggesting preservation of the brain’s microstructural integrity. Together, these imaging findings suggest slowing of neurodegeneration, and support valiltramiprosate’s efficacy in APOE4/4 MCI patients.

Valiltramiprosate Effects in the Pre-specified MCI Group Show Significant Correlations Between Clinical and Brain Volume Benefits: Phase-3 Apolloe4 Results in Early AD Apoe4/4 Homozygotes

AMYLOID CLEARANCE Aidan Power, P Doraiswamy, Duygu Tosun-Turgut, Susan Abushakra, Earvin Liang, Marwan Sabbagh, Anton Porsteinsson, David Watson, Emer Macsweeney, Jorge Matías-Guiu, John Hey, Susan Flint, Patrick Kesslak, Luc Bracoud, Martin Tolar

1Alzheon, Framingham, United States of America, 2Duke University, Durham, United States of America, 3University of California, San Francisco, Department Of Radiology And Biomedical Imaging, San Francisco, United States of America, 4Barrow Neurological Institute, Phoenix, United States of America, 5University of Rochester School of Medicine and Dentistry, Rochester, United States of America, 6Alzheimer’s Research and Treatment Center, Wellington, United States of America, 7Re:Cognition Health, London, United Kingdom, 8Hospital Clínico San Carlos. San Carlos Health Research Institute (IdISSC), Universidad Complutense de Madrid, Neurology, MADRID, Spain, 9Clario, Lyon, France

Aims: Valiltramiprosate, an oral small molecule inhibitor of amyloid-beta (Aβ) oligomer formation, was evaluated in a 78-week trial in APOE4/4 homozygotes with Early AD. Since hippocampus atrophy occurs early in APOE4/4 homozygotes, hippocampal volume (HV) was the main imging outcome. Correlations of valiltramiprosate effects on vMRI with its effects on clinical outcomes are presented over 78 weeks.

Methods: This double-blind trial randomized 325 subjects 1:1 (valiltramiprosate 265 mg BID/ placebo); stratified by MCI (MMSE 27-30) or Mild AD (MMSE 22-26). ADAS-Cog13 was the primary with key secondary/secondary functional outcomes (CDR-SB, DAD). Volumetric MRI measures, including HV, cortical thickness/whole brain volumes (CT, WBV) were collected every 26 week and centrally analyzed. MMRM was main analysis model; analysis by disease stages was pre-specified. Spearman’s correlations are presented.

Results: In overall population, primary and secondary clinical outcomes were not significant but all vMRI outcomes showed significant atrophy slowing versus placebo (HV, 18% less atrophy, p=0.017). Prespecified, MCI (MMSE 27-30, N=125), showed positive effects/trends on ADAS-Cog13 (52%, p=0.041), DAD (96%, p=0.016) and CDR-SB (102%, p=0.053); with significant atrophy slowing in HV (26%), CT (35%), both p< 0.01, and in WBV (22%, p=0.027). In active arm, subject-level correlations between HV-ADAS (r=-0.40), HV-CDR (r=-0.45) were both significant (p<0.005); clinical correalations for CT, WBV were also significant (coefficients 0.34-0.49).

Conclusions: In APOLLOE4 pre-specified MCI group, valiltramiprosate showed nominally significant cognitive and functional benefits with significant atrophy slowing in all brain compartments (HV 26%, WBV 22%). Hippocampal and whole-brain volume effects showed significant subject-level correlations with cognitive and functional benefits over 78 weeks (correlations 0.45-0.49). In the MCI population with adequate neuronal reserve, valiltramiprosate effects on neurodegeneration translate to meaningful clinical efficacy with favorable safety in APOE4/4 homozygotes.

Tml-6: a Novel Drug Targeting at Autolysosomal Machinery in Phase 2 Trial of Alzheimer’s Disease (AD) and Parkinson’s Disease (PD)

AMYLOID CLEARANCE Ih-Jen Su,2, Ching-Chi Chiu, Chien-Hong Lin, Chia-Yu Hsu, Hui-Chen Wang, Ho-Hui Su Chang, Feng-Shiun Shie

1Department of Biotechnology and Food Technology, Southern Taiwan University of Science and Technology, Tainan City, Taiwan, 2Merry Life Biomedical Company, Ltd., Tainan City, Taiwan, 3Department of Medical Biotechnology and Laboratory Science, College of Medicine, Chang Gung University, Taoyuan, Taiwan, 4Center for Neuropsychiatric Research, National Health Research Institutes, Miaoli County, Taiwan

Aims: Background: Autolysosomal dysfunction is increasingly recognized as a key mechanism for neurodegenerative disorders such as AD and PD pathogenesis. TML-6, a synthetic curcumin analog, targets this mechanism by restoring lysosomal acidification and promoting β-amyloid and α-synuclein clearance. Objective: To evaluate the therapeutic effects of TML-6 in AD/PD animal models and in the Phase 2 trial design of AD and PD.

Methods: The lysosomal integrity was damaged by the lysosomotropic reagent l-leucyl-l-leucine methyl ester (LLOMe) or bafilomycin A1, and Aβ and α-synuclein was used to evaluate the phagocytic lysosomal responses in primary neural cell and microglia cultures. The effect of TML-6 on a PD animal model was conducted. The Phase 2 AD trial is conducted and will recruit 210 subjects with MCI or mild dementia of AD. Participants will be treated with oral TML-6. Endpoints include neuropsychological assessments and blood biomarkers. The phase 2 results of TML-6 in AD will be extended to PD treatment in 2028.

Results: TML-6 demonstrated the ability to rescue lysosomal function and facilitate both intracellular, extracellular Aβ and α-synuclein clearance in neurons and microglia. The effect of TML-6 on PD animal model revealed promising behavior improvement and α-synuclein reduction. Following Phase 1 data showing favorable safety and pharmacokinetics in elderly subjects, a global Phase 2 trial for AD (U.S., Sweden, Taiwan) will be conducted after FDA approval. The trial success of phase 2 in AD will then be extended to phase 2 trial in 2028 for PD.

Conclusions: TML-6 exhibits a multi-target action mechanism by driving autolysosomal regulation to remove the mutant protein accumulation in neuron and microglia, meeting the new era strategy to be developed as potential breakthrough drugs for neurodegenerative diseases in AD and PD therapy.

Protocol for a Multicenter, Prospective, Real-World Study: Lecanemab in Patients with Early Alzheimer's Disease(LEAD) in China

AMYLOID CLEARANCE Hui Zhao, Limoran Tang, Yun Xu

1Afflicated Drum Tower HospitaL, Department Of Neurology, NANJING, China, 2Department of Neurology, Afflicated Drum Tower Hospita, NANJING, China

Aims: LEAD is an investigator- led,multicenter, prospective, real-world study to avaluate effectiveness and safety of Lecanemab. The primary objective was to assess the effectiveness of lecanemab under real-world clinical practice in early Alzheimer's diseasepatients in China. The secondary objectives included evaluating the safety of lecanemab for the management of AD in Chinese patients in terms of adverse events (AEs), serious AEs along with discontinuation of lecanemab.

Methods: This multicentre, prospective cohort trial included participants 40 to 90 years old with early AD (MCI or mild dementia due to AD) having an indication for lecanemab. Each participant had evidence of PET or by CSF. Baseline data on demographics, medical history, quality of life, cognitive score, plasma biomarkers, Apoe and neuroimaging indicators were assessed and a treatment with intravenous lecanemab (10 mg per kilogram of body weight each 2 weeks) was started. The patients will be followed up after three, six and twelve months including safety visits, provide blood samples to measure AD biomarkers at months. All cognitive assessments will be repeated at month 6 and 12. The last study visit will be at week 52, when all baseline measurements will be repeated.

Results: From August 2024 to present, the research is still ongoing, After 6 months of treatment, MMSE (0.821, p=0.013) and ADL (-4.605, p<0.001) showed significant improvement compared to the baseline, the plasma pTau 217/Aβ 42 and pTau 181/Aβ 42 ratios were significantly reduced (p<0.001) and the plasma Aβ 42 concentration was significantly increased (p<0.001). Adverse reactions occurred in 22 cases (ARIA-H 7%, ARIA-E 2%, infusion-related reactions (fever 3%, allergy 1%)).

Conclusions: The existing results suggest that Lecanemab treatment for early Alzheimer's disease (AD) shows clinical symptom and plasma biomarkers improvement at 6 month.

Valiltramiprosate Reverses Neurodenegeration and Improves Clinical Outcomes in Subgroup of Early Symptomatic AD in Homozygotes: Results from 78 Week Phase 3 Apolloe4 Trial

NEUROPROTECTIVE & MITOCHONDRIAL COMPOUNDS Jeremy Yu, Susan Abushakra, Hugo Geerts, Shaina Short, Jean Schaefer, Aidan Power, Patrick Kesslak, Martin Tolar, John Hey

1Alzheon, Framingham, United States of America, 2Certara, Radnor, United States of America

Aims: Progressive hippocampal atrophy is a hallmark of Alzheimer’s disease (AD). Valiltramiprosate/ALZ-801 is an oral, small-molecule inhibitor of amyloid-oligomer formation in development as a disease-modifying treatment for early symptomatic AD. While the APOLLOE4 Phase 3 trial did not meet the primary cognitive endpoint in the overall Early AD population, valiltramiprosate showed significant slowing of atrophy in hippocampus and all other brain regions. Also, the prespecified Mild Cognitive Impairment (MCI) subgroup (MMSE >26) showed meaningful, (nominally) significant clinical benefits that correlated with significant reduction in hippocampal atrophy. Responder analysis was conducted on a subgroup of super-responders who showed increased hippocampal volume (HV) over baseline at 78 weeks.

Methods: APOLLOE4 was a randomized, double-blind, 78-week, placebo-controlled Phase 3 trial conducted in 325 APOE4/4 homozygotes (MMSE 22-30). Valiltramiprosate 265 mg or placebo was administered BID. Primary outcome was ADAS-Cog13, and HV by vMRI was the main imaging outcome. Super-responders were defined as an increase in HV at 78 weeks over baseline.

Results: Early symptomatic completers with ≥26 MMSE included 10/64 and 2/67 subjects in active and placebo arms, respectively. This represents 16% active and 3% placebo super-responders. One active super-responder had an MMSE of 24. The valiltramiprosate super-responders (6M/5F) showed a larger HV (Δ +66.0±20.4 mm3) at 78 weeks and exhibited younger age, higher baseline MMSE, and larger baseline HV than overall early symptomatic population. The HV-rescuing benefit was associated with improved ADAS-Cog13 and CDR-SB.

Conclusions: Valiltramiprosate reversed hippocampal atrophy and stabilized cognition/function over 78 weeks in a subgroup of early symptomatic AD subjects. The increased HV in this subgroup suggests a beneficial effect of valiltramiprosate on neurogenesis/neuroplasticity and supports its potential as a preventive treatment for APOE4/4 presymptomatic AD patients.

Baseline Characteristics Among Early Initiators of Donanemab in the United States: a Real-World Study

AI, OTHER Oralee Varnado, Andrew Spargo, Chandrasekar Gopalakrishnan, Stevan Severtson, Kathryn Starzyk, Amy Kovacik, Ann Hartry

1Eli Lilly and Company, Medical Affairs, Indianapolis, United States of America, 2OM1, Inc, Department Of Real-world Evidence & Research Analytics, Boston, United States of America

Aims: To date, there are no published real-world studies of donanemab. This study’s purpose was to describe baseline demographic and clinical characteristics of early donanemab initiators.

Methods: Data for this descriptive, new user cohort study were derived from the OM1 Real World Data Cloud (OM1, Inc., Boston, MA), a real-world dataset with linked claims and electronic medical record (EMR) data on patients in the US. The study population included adult patients initiating donanemab between July 2024 and September 2025, the first 15 months following FDA approval. Patients were also required to have a recorded diagnosis of mild cognitive impairment or Alzheimer's disease and at least one claim or EMR encounter in the 365 days prior to treatment initiation. Baseline characteristics of donanemab initiators were tabulated and included demographics, insurance type, clinical comorbidities, medication use and use of diagnostic testing procedures.

Results: The final eligible cohort comprised 1,143 patients [mean age in years (standard deviation (SD)) = 76.6 (6.4)] who initiated donanemab. Patients were predominantly female (53.4%), white race (93.7%), and resided in the South (74.8%). 63.9% had a diagnosis of MCI and 81.5% had a diagnosis of AD, reflecting real-world coding patterns and the inability to capture exact disease severity. The first dose for most patients (71.6%) was 350 mg, consistent with the modified dosing regimen. Comorbidities were frequent with most common being dyslipidemia (78.9%), hypertension (71.3%) and prior malignancy (27.7%).

Conclusions: This cohort study provides insights into the early initiators of donanemab in routine care. Since the characteristics of the study population reflect the specific data source used, findings may not be broadly generalizable. Additional analyses of donanemab utilization and monitoring during follow-up will be presented at the conference.

Chronological Evaluation of Brain Changes in Alzheimer’s Patients Receiving Anti-aβ Treatment with Synthetic MRI

STRUCTURAL MRI, MR SPECTROSCOPY Yuki Morita, Akira Matsushima, Yoshiki Sekijima Shinshu University School of Medicine, Neurology & Rheumatology, Matsumoto city, Japan

Aims: To investigate the effect of anti-Aβ therapy by quantifying annual brain volume changes using Synthetic MRI (SyMRI).

Methods: Patients diagnosed with Alzheimer’s disease (AD) and treated with Lecanemab or Donanemab were enrolled. SyMRI was performed at baseline and subsequently at 1, 2, 3, 6, and 12 months post-treatment initiation. Quantitative analysis using SyMRI NEURO® included measurements of brain parenchymal volume (BPV), intracranial volume (ICV), brain parenchymal fraction (BPF; calculated as BPV/ICV), gray matter volume fraction (GM), white matter volume fraction (WM), and myelin volume fraction (MY). Longitudinal changes were analyzed using a generalized linear mixed model, with time since treatment initiation as the fixed effect.

Results: A total of 24 patients were enrolled; 13 completed the 6-month follow-up, and 6 completed the 12-month evaluation. At baseline, the mean (range) age, Mini-Mental State Examination-Japanese (MMSE-J) score, and BPF were 69.5 years (53–87), 24.6 (21–29), and 75.8% (69.9–82.9), respectively. There was a moderate association between BPF and MMSE-J (r = 0.463, 95% confidence interval: 0.113–0.783). The mean coefficient of variation (CV) for ICV across time points was 0.67%. Annualized changes in MMSE-J, BPF, GM, WM, and MY were 0.7 ± 0.7, -1.4% ± 0.2%, 1.5% ± 0.6%, -2.0% ± 1.0%, and -0.3% ± 0.4% (mean ± standard error), respectively. Three participants experienced ARIA-H within one year, all cases being mild without significant differences in baseline parameters compared to those without ARIA.

Conclusions: SyMRI-based measurements showed low CV, indicating strong test-retest reliability. The annual decline in BPF matched rates previously reported for mild cognitive impairment patients and GM remained more stable than WM.

Plasma GFAP as a Biomarker for Anti-inflammatory Response and Cognitive Improvement in Phase 1 Trials of Troculeucel for Alzheimer’s Disease

DIAGNOSTICS: CSF- AND BLOOD-BASED BIOMARKERS Paul Song, Sean Hong, Yoonmi Kang, Hank Lee, Juan Mata, Harry Chung, Jesse Carr, Clemente Humberto Zuniga Gil, Blanca Isaura Acosta Gallo, Rufino Menchaca Diaz, Cesar Alejandro Amescua, Katia Betito, Lucia Hui

1NKGen Biotech, Santa Ana, United States of America, 2Behavioral Research Specialists, LLC, Glendale, United States of America, 3Tijuana General Hospital, Tijuana, Mexico, 4Hospital Angeles, Tijuana, Mexico

Aims: To assess plasma GFAP levels as a biomarker for anti-inflammatory effects and cognitive improvement following treatment with troculeucel, an autologous, non-genetically modified NK cell therapy with enhanced cytotoxicity and activating receptor expression, in Alzheimer’s disease (AD) patients enrolled in Phase 1 studies.

Methods: In a Phase 1 dose-escalation study (NCT04678453; n=10), troculeucel was administered intravenously every three weeks for 3 months (4 doses at 1B, 2B, or 4B). In a separate Phase 1 study (NCT06189963; n=3), moderate AD patients received 6B cells every three weeks for 12 months. Plasma GFAP levels, an inflammatory marker in neurodegenerative disease, along with cognitive scores (ADCOMS and CDR-SB) were collected at baseline and 3 months post-treatment. Correlation analyses were performed to evaluate relationships between GFAP levels and cognitive performance.

Results: At baseline, plasma GFAP levels showed strong, statistically significant positive correlations with cognitive scores: CDR-SB (r=0.60, p=0.045) and ADCOMS (r=0.64, p=0.030). After 3 months of troculeucel treatment, correlations remained strong: CDR-SB (r=0.57, p=0.056) and ADCOMS (r=0.66, p=0.022). The consistency in correlation strength and direction across both timepoints indicates that plasma GFAP levels are closely associated with cognitive status in AD patients.

Conclusions: These findings suggest plasma GFAP is a responsive biomarker for monitoring the anti- inflammatory effects of troculeucel and its cognitive benefits in AD. The data support troculeucel’s therapeutic potential and highlight plasma GFAP as a valuable indicator of treatment efficacy in early- phase clinical trials.

Sabirnetug Biomarker Treatment Responses: Exploratory Evaluation of the CNS Disease Panel Nulisa-SEQ-TM

DIAGNOSTICS: CSF- AND BLOOD-BASED BIOMARKERS Hugo Vanderstichele, June Kaplow, Maddelyn Hyland, James Doherty, Jasna Jerecic Acumen Pharmaceuticals, Newton, United States of America

Aims: Biofluid biomarkers are used to evaluate efficacy and mechanism of action of disease- modifying drugs and are transformative in the diagnosis of Alzheimer’s disease (AD). We conducted an evaluation of biomarker changes after treatment with the AβO-selective antibody sabirnetug (ACU193) in the phase 1 randomized, placebo-controlled INTERCEPT-AD study in early symptomatic AD using the NULISAseq™ CNS Disease Panel v1 to confirm and expand our understanding of changes in neuronal protein levels.

Methods: Biomarkers were measured in CSF and EDTA-plasma pre- and post-dose. Percent change in NPQ units from baseline to endpoint for the 127 analytes was compared between treatment groups and placebo using two-sample Welch’s t-tests. No adjustments were made for multiplicity therefore statistical significance is considered nominal. Spearman’s rank correlation analyses assessed relationships amongst and between protein proteoforms (Aß, tau, phosphorylated tau, neurofilament) over time across all dose groups as well as the correlation of ELISA and NULISA results.

Results: Published biomarker results from INTERCEPT-AD were confirmed and extended with NULISA. A treatment effect was shown for pTau217 in CSF. In EDTA-plasma, treatment effects were observed for GFAP, Aß38, Aß40, Aß42, and for the ratio of tau/Aß42, pTau181/Aß42, and pTau217/Aß42. Positive correlations were obtained in CSF between and amongst the different Aß (Aß38, Aß40, Aß42), tau (total Tau, pTau181, pTau217, pTau231), and neurofilament (NF-L, NF-H) proteoforms. A negative correlation was seen between Aß42/Aß40 and pTau181. Previously reported results correlated with NPQ values measured with NULISA.

Conclusions: AT(N) biomarker correlations were identified between different proteoforms of Aß and pTau. Aß38, pTau231 and NFL-H/L, amongst other candidate proteins, were identified as novel pharmacodynamic biomarkers. The results further support the downstream pathophysiological changes affected by sabirnetug treatment in AD.

Longitudinal Effects of Alzheimer’s Disease Medication on Cognitive Outcomes

FOR TREATMENT / ABETA, TRUNCATED & PGLU-ABETA Jenny Jyoung Lee, Anisha Das, Yuanjia Wang, Annie Lee

1Ewha Womans University, Department Of Data Science, Seoul, Korea, Republic of, 2Columbia University, Department Of Neurology, New York, United States of America, 3Columbia University, Department Of Biostatistics, New York, United States of America

Aims: The long-term effectiveness of Alzheimer’s disease (AD) medications on cognition and brain health remains uncertain, particularly when accounting for the dynamic nature of treatment exposure. To address this gap, we leveraged longitudinal data from the Harvard Aging Brain Study– Healthy Diversity (HABS-HD) cohort to examine the effect of longitudinal AD medication exposure on changes in cognition, dementia severity, and white matter hyperintensity (WMH) volume using data- driven approach.

Methods: Eligible HABS-HD participants with mild cognitive impairment were analyzed. Cognitive outcomes included CDR Sum of Boxes (CDR-Sum), MMSE, Digit Symbol Substitution Test (DDST), and WMH volume. Longitudinal trajectories were modeled with linear mixed-effects model among AD medication initiators or discontinuers (n=148). As treatment decisions are influenced by worsening cognition, MSMs were applied to address time-varying confounding and to achieve baseline covariate balance, enabling causal effect estimation of continued treatment versus no treatment in a larger sample (n=367).

Results: AD medication users exhibited slower cognitive decline and disease progression, with less worsening of CDR-Sum (β=–3.83), higher MMSE (β=5.96), improved DSST (β=1.39), and attenuated WMH progression (β=–1.55). In the larger cohort, MSMs confirmed consistent effects, including lower CDR-Sum (α=–4.26), higher MMSE (α=5.28), and improved DSST (α=1.12). Overall, continued AD medication use was associated with broadly protective cognitive and neuroimaging profiles, underscoring its potential to mitigate decline and modify disease progression.

Conclusions: Our study demonstrates that initiation or discontinuation of AD medications was associated with slower short-term cognitive decline, and continued use had a causal effect on sustaining cognitive function and slowing neuroimaging progression over the longer term. These results highlight the clinical value of both early intervention and continued adherence, and also illustrates how causal modeling approaches can enhance the evaluation of AD therapies.

:: Decoding Heterogeneity in A4: Explainable ML Finds Solanezumab Responsive Subgroups in Preclinical AD

FOR TREATMENT / AI, OTHER Jan Sedway, Bessi Qorri, Mike Tsay, Christian Cumbaa, Paul Leonchyk, Larry Alphs, Luca Pani,3, Joseph Geraci,5

1NetraMark, Toronto, Canada, 2Department of Psychiatry and Behavioral Sciences, Leonard M. Miller School of Medicine, University Of Miami, Coral Gables, United States of America, 3Department of Biomedical, Metabolic and Neural Sciences, University Of Modena And Reggio Emilia, Modena, Italy, 4Department of Pathology and Molecular Medicine, Queen's University, Kingston, Canada, 5Centre for Biotechnology and Genomic Medicine, Medical College Of Georgia, Augusta, United States of America

Aims: To test whether a novel machine learning method can surface patient subpopulations with differential response within a negative Alzheimer’s disease trial and to deconstruct those populations into explainable and unexplainable subgroups.

Methods: A sub-insight learning approach was applied to the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) dataset to identify subpopulations defined by 2–4 variables predictive of treatment response. A4 enrolled clinically asymptomatic older adults with amyloid pathology randomized to solanezumab or placebo, with the primary outcome being change in the Preclinical Alzheimer Cognitive Composite (PACC) Total Score through Week 240. A subsample comprising participants with cognitive impairment at follow-up (n=51) and without (n=100) was analyzed.

Results: Two subpopulations demonstrated less decline with solanezumab at Week 240. Subgroup 1 (n=20, 26% of the testing sample) was defined by higher right amygdala volume and higher baseline PACC Digit Symbol Substitution Test scores (Cohen’s d=1.1, p=.03). Subgroup 2 (n=9, 12% of the testing sample) was defined by higher right superior temporal lobe volume and higher baseline PACC Digit Symbol Substitution Test scores (Cohen’s d=2.3, p=.03). These findings suggest that individuals with higher brain volume and higher cognitive function at baseline may benefit most from solanezumab.

Conclusions: An explainable ML strategy identified clinically interpretable profiles that may be preferential responders to solanezumab, despite the overall negative A4 result. These exploratory findings, based on limited sample sizes, suggest that greater regional brain volume and stronger psychomotor speed/attention at baseline may enrich for treatment benefit. Independent replication, presentation of absolute subgroup counts, and adjustment for multiplicity are warranted; if confirmed, this approach could support precision enrichment in future preclinical AD trials.

Incidence and Characteristics of Amyloid-related Imaging Abnormalities (ARIA) with Lecanemab in the US Veterans Affairs Healthcare System (VAHS)

DYSFUNCTION / OTHER Ying Wang,2, Peter Morin, Vanesa Andreu Arasa, Xavier Cambi, Wen Hu, Joel Reisman, Yuehan Li, Dan Berlowitz, Raymond Zhang, Ran Gao, Babak Haji, Amir Abbas Tahami Monfared,9, Quanwu Zhang, Weiming Xia,5,6

1Wentworth Institute of Technology, Boston, United States of America, 2VA Bedford Healthcare System, Geriatric Research Education And Clinical Center, Bedford, United States of America, 3Boston University, Neurology, Boston, United States of America, 4Boston University, Radiology, Boston, United States of America, 5Boston University, Pharmacology, Physiology & Biophysics, Boston, United States of America, 6University of Massachusetts Lowell, Department Of Biological Sciences, Kennedy College Of Sciences, Lowell, United States of America, 7University of Massachusetts, Lowell, United States of America, 8Eisai, Inc, Nutley, United States of America, 9McGill University, Montreal, Canada

Aims: ARIA may occur in patients treated with anti-amyloid therapies. This study reports ARIA incidence and patient characteristics among Veterans treated with lecanemab (LEC) in the VAHS.

Methods: Veterans receiving LEC infusions through July 2025 in VAHS and community care, but not Medicare-covered beneficiaries, were included. ARIA was ascertained by radiologists. Clinical notes were reviewed under neurologist/neuroradiologist guidance to report ARIA incidence and attributes. Structured data were assessed to describe patient characteristics.

Results: 117 Veterans receiving LEC was identified with mean age 73.7 years, 96.6% male, 90.6% white, and 90.6% non-Hispanic. Top baseline comorbidities included hypertension (64.1%), hypercholesterolemia (47.9%), diabetes (28.2%), cerebrovascular disease (14.5%), and cancer (8.5%). All patients were APOE4 non-carriers or heterozygous APOE4 carriers (VAHS excludes APOE ε4 homozygotes). By the study date, 18.8% patients received LEC for ≤3 months, 18.8% for 3–6 months, 44.4% for 7–12 months, and 18.0% for more than 1 year. After LEC initiation, 9.4% patients had ARIA, including 7.7% ARIA-E, 8.5% ARIA-H, 0.9% isolated ARIA-E ,1.7% isolated ARIA-H, and 4.3% symptomatic ARIA. Mean time to ARIA was 118.0±84.1 days (median=77 days). As of July 2025, ARIA-E was resolved in 4/9 cases and ARIA-H was generally stabilized. Infusions resumed in 3 cases. One emergency visit and no hospitalization were documented within 30 days of ARIA events.

Conclusions: In this VAHS cohort, absolute risk of ARIA during the study observation period was 9.4% (11/117). Most ARIA events were mild or moderate and a minority were symptomatic. Data interpretation should consider modest sample size, variable treatment exposure, and incomplete observation of infusion pauses versus discontinuations. This study supports the feasibility of ARIA monitoring and management for LEC therapy in routine practice.

Reversal of Biperiden-induced Cognitive Impairment by Donepezil: a Proof-of- Pharmacology Tool for Muscarinic Modulators in Healthy Elderly

DRUG DEVELOPMENT, CLINICAL TRIALS Massimo Magliocca, Andriy Gorbenko, Laura Borghans, Kiefer Comassi, Gabriel Jacobs, Philip Kremer

1Centre For Human Drug Research (CHDR), Neurology, Leiden, Netherlands, 2Centre For Human Drug Research (CHDR), Psychiatry, Leiden, Netherlands

Aims: To investigate whether the pharmacodynamic (PD) profile induced by intravenous (IV) biperiden can be reversed by the acetylcholinesterase inhibitor donepezil, thereby validating the IV biperiden challenge model for proof-of-pharmacology studies with novel muscarinic acetylcholine receptor (mAChR) modulators.

Methods: Double-blind, placebo-controlled, randomized, two-way crossover study in healthy elderly (≥65 to 80 years). Participants received 10 mg oral donepezil or placebo at the 0-hour timepoint, followed by a 1-hour IV infusion of 2.6 mg biperiden starting at the 3-hour timepoint. A validated test battery assessed pro- and anti-cholinergic effects up to 24 hours post-dose. Pharmacokinetic non- compartmental analysis of both drugs was conducted.

Results: 16 participants completed the study. A single dose of 10 mg donepezil improved adaptive tracking (1.69% [0.23–3.15], p=0.025) and N-back reaction times (one-back: −53.59 ms [−100.39, −6.79], p=0.026) at the 2.5-h timepoint (pre-biperiden), and reversed the PD effects induced by 2.6 mg IV biperiden at the 4-h timepoint. Reversals were observed in adaptive tracking (3.05% [1.56–4.53], p=0.0002), indicating enhanced sustained attention; N-back reaction times (one-back: −60.00 ms [−105.70, −14.30], p=0.012; zero-back: −65.71 ms [−101.33, −30.09], p=0.0006), suggesting improved processing speed; and immediate visual verbal learning (up to 1.9 [0.8–2.9], p=0.0016), indicating effects on learning. Additionally, compared to placebo, donepezil significantly reduced alpha, delta, and theta pharmaco-encephalography band amplitudes under both eyes-open and eyes-closed conditions.

Conclusions: The model demonstrated low and acceptable intersubject (∼30–35%) and intrasubject (∼18–24%) variability in plasma Cmax and AUC0–last, further supporting its reliability for repeated- measures study designs. These findings validate the 2.6 mg IV biperiden challenge model as a robust tool for proof-of-pharmacology studies of novel mAChR modulators in early-stage clinical development.

Two Parallel Phase 3 Trials to Evaluate X/t for the Treatment of Agitation Associated with Alzheimer’s Disease (ADAGIO-1/ADAGIO-2)

DRUG DEVELOPMENT, CLINICAL TRIALS Ian Ramsay, Ronald Marcus, Yanqiu Shao, George Grossberg, Frank Schieber, Rosalinda Sepulveda

1Bristol Myers Squibb, Princeton, United States of America, 2Saint Louis University School of Medicine, Psychiatry And Behavioral Neuroscience, St. Louis, United States of America

Aims: Alzheimer’s disease (AD) is the most common cause of dementia. Agitation develops in approximately half of patients with AD, contributing to disease progression, poor quality of life, mortality risk, increased risk of institutionalization, and caregiver burden. With only 1 FDA-approved treatment for AD agitation, further treatment options are warranted. The M1/M4-preferring muscarinic agonist, xanomeline, previously showed antipsychotic activity in placebo-controlled trials in people with AD. Xanomeline combined with the peripherally restricted muscarinic receptor antagonist trospium chloride (X/T) is FDA approved to treat schizophrenia. In post hoc analyses of pooled data from the randomized, placebo-controlled EMERGENT trials, X/T showed improvement in agitation in participants with acute schizophrenia exacerbations. These data suggest that X/T may be effective in treating AD symptoms. Our aim is to evaluate the safety and efficacy of X/T for the treatment of agitation in AD.

Methods: ADAGIO-1 (NCT07011732) and ADAGIO-2 (NCT07011745) are identical, ongoing, 14-week, phase 3, randomized, double-blind, placebo-controlled, parallel-group trials of participants aged 55- 90 years with AD diagnosis including biomarker confirmation of AD pathology and at least 2 weeks of agitation prior to screening. Agitation is defined by Neuropsychiatric Inventory (NPI)/NPI-Nursing Home and Clinical Global Impressions-Severity (CGI-S) scales scores greater or equal to 4 and by Cohen-Mansfield Agitation Inventory-International Psychogeriatric Association (CMAI-IPA) criteria at screening and baseline visits.

Results: The trials are currently recruiting to enroll approximately 704 participants total. The primary endpoint is change from baseline to end of treatment in CMAI-IPA total score. Secondary outcome measures include change from baseline in CGI-S scale. Safety and additional efficacy measures will also be evaluated.

Conclusions: X/T has the potential to be a new treatment for AD agitation based on muscarinic receptor agonism.

Development of All-d-peptides for the Disassembly of Toxic TAU Aggregates

TREATMENT / PROTEIN AGGREGATION, NFT, MISFOLDING, CHAPERONES Nadine Rösener, Kira Allmeroth, Bettina Kass, Mathias Wendt, Sarah Schemmert, Joana Petushi, Hévila Brognaro, Karolina Korostov, Gina Hartmann, Brian Gensenmüller, Sahra Russek, Susanne Wegmann, Christian Betzel, Jacek Biernat,4, Eckhard Mandelkow,4, Antje Willuweit, Dieter Willbold,5

1Priavoid GmbH, Düsseldorf, Germany, 2Deutsches Zentrum f. Neurodegenerative Erkrankungen (DZNE), c/o Charite University Medicine Berlin, Berlin, Germany, 3Institute of Biochemistry and Molecular Biology, Laboratory for Structural Biology of Infection and Inflammation, University of Hamburg, c/o DESY, Department Of Chemistry, Hamburg, Germany, 4Deutsches Zentrum f. Neurodegenerative Erkrankungen (DZNE) und Universitätsklinikum Bonn, Klinik für Neurodegenerative Erkrankungen, Bonn, Germany, 5Heinrich-Heine-Universität Düsseldorf, Institut Für Physikalische Biologie, Düsseldorf, Germany

Aims: Tau protein oligomers and fibrillar aggregates exert neurotoxicity in Alzheimer’s disease (AD) and other tauopathies. Development of compounds that directly disassemble these Tau assemblies into harmless physiological Tau monomers will disrupt their self-replication and toxicity and will enable disease-modifying treatments for AD and other tauopathies. Here, we report the development of all-D-enantiomeric peptide ligands that bind monomeric full-length Tau with high affinity, thereby stabilizing the physiological intrinsically disordered structure and preventing initiation of aggregation as well as eliminating already existing aggregates by disassembling them into monomers in a thermodynamic mode of action (MoA).

Methods: Mirror-image phage display, next generation sequencing (NGS), surface plasmon resonance (SPR), Thioflavin-T aggregation assay (ThT-assay), cell-viability assay, seeded Tau aggregation in cells, chloroalkane penetration assay (CAPA-assay), time-resolved fluorescence energy transfer (TR-FRET)

Results: We introduce our drug development toolbox of selecting all-D-enantiomeric compounds targeting Tau and verifying their thermodynamic MoA. The compound TF2D-5a was identified by mirror-image phage display, NGS and rational design [1]. It shows binding in the nanomolar range characterized by SPR and reduced full-length Tau aggregation in a heparin-free seeded Thioflavin-T (ThT) assay. We demonstrate cell penetration by the CAPA-assay and validate TF2D-5a’s inhibitory effect in cells using the TauRD-CFP/YFP biosensor cell system. Another important tool is a TR-FRET assay for demonstrating ex vivo target engagement using postmortem human brain homogenate samples of AD patients. [1] Altendorf, T. et al., Int J Mol Sci 24, 2161 (2023).

Conclusions: The present work provides promising results for the development of lead compounds with a thermodynamic MoA for the future treatment of AD and other tauopathies.

Acetylated TAU as a Driver of TBI-AD: Therapeutic Targeting of P300/cbp and Sirtuin- 1 Pathways

TREATMENT / TAU, PHOSPHORYLATION, TRUNCATION Vivek Saraswat, Andrew Pieper, Edwin Vázquez-Rosa, Sarah Barker Case Western Reserve University, Psychiatry, Cleveland, United States of America

Aims: To investigate the role of acetylated tau (ac-tau) in traumatic brain injury–accelerated Alzheimer’s disease (TBI-AD) and evaluate therapeutic strategies targeting tau acetylation as a means to prevent neurodegeneration.

Methods: 1. Established a TBI-AD model using amyloid-driven 5xFAD mice. 2. Assessed cognitive function, microglial activation, and blood–brain barrier (BBB) integrity following TBI. 3. Tested two ac- tau–lowering interventions: diflunisal (p300/CBP inhibition) and P7C3-A20 (NAD⁺ restoration/Sirt1 activation). 4. Analyzed clinical data from TBI patients prescribed NSAIDs with or without p300/CBP inhibitory activity.

Results: TBI selectively induced ac-tau in 5xFAD mice, causing cognitive decline, microglial activation, and BBB dysfunction. Both diflunisal and P7C3-A20 normalized ac-tau, preventing neurodegeneration and restoring BBB integrity. Clinical data showed reduced dementia incidence in TBI patients prescribed p300/CBP-inhibiting NSAIDs compared to non-targeting NSAIDs.

Conclusions: Acetylated tau is a central driver of TBI-AD and a tractable therapeutic target. Dual- pathway interventions demonstrate that reducing tau acetylation, rather than targeting amyloid, effectively prevents TBI-AD progression. Conserved microglial signatures link tau pathology to neuroimmune dysfunction and provide translational readouts. These findings support both repurposing existing NSAIDs and developing next-generation ac-tau modulators as disease-modifying strategies for TBI-accelerated dementia.

Choline Alphoscerate in the Therapy of Mild Cognitive Impairment and Associated Vascular Damage. One-Year Results of the CARL Study

Enea Traini, Anna Carotenuto, Mohammed Hossein, Vincenzo Andreone, Francesco Amenta

1University of Camerino, School Of Pharmacy, Camerino, Italy, 2A. Cardarelli Hospital, Neurology And Stroke Unit-neurology, Naples, Italy

Aims: Brain atrophy, mainly involving the hippocampus, is related to progression of cognitive impairment and conversion from mild cognitive dysfunction to overt dementia. Choline alphoscerate is a cholinergic precursor officially approved for the treatment of degenerative-involutive and cerebrovascular psychogenic syndromes characterized by age-related memory deficits, confusion, reduced motivation, attentional decline, affective instability and behavioral disturbances.

Methods: The CARL (Choline Alphoscerate in Mild Cognitive Impairment) study evaluated the efficacy of choline alphoscerate in patients with MCI and associated vascular damage, in slowing the hippocampal, entorhinal, cortical atrophy, the ventricular dilation and the progression of cognitive and behavioural symptoms associated with MCI and concomitant vascular brain injury. In this double- blind, placebo-controlled trial, 60 elderly MCI subjects (exhibiting MRI evidence of vascular-related brain changes) were randomized 1:1 to receive either choline alphoscerate 1,200 mg/day or placebo. High-resolution T1-weighted MRI scans were acquired and analyzed at the baseline and after 12- month; hippocampal, entorhinal, neocortical volumes and ventricular dimensions were quantified using FreeSurfer v7.4.1. Cognitive, functional and behavioural status were evaluated by a comprehensive neuropsychological battery at study entry and during one year of follow-up.

Results: The hippocampal region shows signs of degeneration before the onset of cognitive symptoms and represents the first region to degenerate. After one year, patients treated with choline alphoscerate demonstrated a significantly lower rate of hippocampal and entorhinal atrophy, and less ventricular dilation, compared with placebo. Neuropsychological evaluation showed a relevant difference between the two groups mainly in behavioral tests with a significative reduction in apathy in the choline alphoscerate treated group.

Conclusions: These findings reinforce the efficacy of choline alphoscerate in improving behavioural symptoms and reveal a significant efficacy in attenuating the progression of early neurodegenerative changes in MCI with vascular involvement.