Poster gallery

Federica Ribaldi 1,2, Christian Chicherio 1,3, Augusto J Mendes 1,2, Laurene Pereira 1, Ozge Sayin 1,2, Marie-Louise Montandon 1, Claudio Bassetti 4,5, Giovanni B Frisoni 1,2

1 Geneva Memory Center, Department of Rehabilitation and Geriatrics, Geneva University Hospitals, Geneva, Switzerland
2 Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland
3 Center for Interdisciplinary Study of Gerontology and Vulnerability (CIGEV), University of Geneva, Geneva, Switzerland
4 Department of Neurology, Bern University Hospital (Inselspital) and University of Bern, Bern 3010, Switzerland
5 Sleep-Wake-Epilepsy Centre, Bern University Hospital (Inselspital) and University of Bern, Bern 3010, Switzerland

Dementia prevention is a critical public health challenge, profoundly impacting individual quality of life, family well-being, and societal costs. The World Health Organization recognizes dementia prevention as a public health priority; however, standardized procedures to address this priority remain limited. At the Geneva Memory Center of Geneva University Hospitals, we developed and implemented an integrated plan to promote cognitive health and prevent cognitive decline, including community education, general practitioner involvement, and specialized care initiatives for dementia prevention. The first component, the Prevention Memory Workshop ("Atelier mémoire de prévention"), provides education, social and cognitive stimulation for the general public. This initiative includes structured activities aimed at increasing awareness about brain health and providing practical recommendations and tools for dementia prevention. The second component, the Lemanic Program on Cognitive Impairment and Dementia ("Cours Lémanique sur les Démences et les Troubles Cognitifs"), engages general practitioners as key stakeholders, promoting best practices for dementia prevention in the Lemanic region and French-speaking Switzerland. The third component, Brain Health Services for dementia prevention (dBHS), provides a comprehensive secondary prevention approach. dBHS offers risk assessment, risk communication, and personalized strategies for cognitively unimpaired individuals at risk of developing dementia. Utilizing advanced tools, dBHS identifies individuals at increased dementia risk and develops tailored intervention plans incorporating lifestyle modifications and biological interventions. This abstract describes the development and implementation of these initiatives, highlighting comprehensive dementia prevention strategies. These programs may serve as models for adoption in other European countries, addressing the global challenge of dementia prevention.

Marc Züst

University Hospital of Old Age Psychiatry and Psychotherapy, UPD Bern

Sleep and loneliness are increasingly recognized as modifiable risk factors for dementia. Positive social interactions have been linked to better sleep, cognitive function, and physiological health. This project tests the hypothesis that engaging older adults in social activities, specifically collaborative board game nights using the award-winning game Dorfromantik, can enhance sleep quality, memory, and plasma-based biomarkers of brain health.

The study will follow a between-subject design comparing collaborative versus solo game play. Outcome measures include polysomnography, episodic memory testing, high-density EEG synchrony between players, slow-wave–spindle coupling during sleep, and biomarker levels (saliva oxytocin; plasma Aβ, pTau, GFAP, NfL). Social bonding will be assessed via self-report and standardized video analysis.
A parallel rodent study is envisioned, using group versus single housing and behavioral assays to explore learning, mood, and social behavior. This translational approach could shed light on the neural and biochemical mechanisms linking social engagement to sleep and brain health.

If successful, this project could provide a low-cost model for social-based sleep enhancement with real-world public health applications, including integration into care facilities and community programs. Further innovation may arise from digital tools to support and monitor such interventions.

Esther Brill, Stefan Klöppel

University Psychiatric Services (UPD) Bern

Background: Dementia constitutes a growing global public health concern, underscoring the urgent need for effective and sustainable preventive strategies. Multidomain lifestyle interventions, encompassing physical and cognitive training, dietary modifications, and social engagement are commonly recommended to mitigate dementia risk. However, long-term maintenance of behavioral changes requires strong motivational and volitional processes. Value-congruent behaviors enhance intrinsic motivation through meaningful and stable behavior-outcome associations. Despite the efficacy of well-structured lifestyle-interventions, long-term adherence remains optimizable, with fewer than half of participants completing all components.

Methods: We designed a digitally-embedded value-based motivational framework, integrating smartphone-based momentary assessments, reminders, and just-in-time adaptive interventions (JITAIs). A feasibility study (n = 15) was conducted in older adults with subjective or mild cognitive impairment to evaluate the acceptability and usability of these digital tools. Our framework is designed to align individual values with lifestyle goals, employing motivational interviewing and goal-setting techniques. After initial implementation as a manualized clinical consultation in a memory clinic setting, we now evaluate the effects of smartphone-delivered personalized, value-based communication styles compared to neutral statements on the intended adherence to dementia preventive behaviors.

Results: The feasibility pilot highlights that smartphone-based JITAIs are both acceptable and well tolerated among older adults with cognitive concerns. The digital interventions demonstrated good adherence and high user acceptability. Clinical observations indicate that value-based motivational approaches may enhance participants’ intention to adhere to recommended lifestyle changes. The evaluation of the effects of positively-framed, value-based digital messages on lifestyle adherence is currently underway.

Conclusion: Digitally delivered value-based motivational approaches and positively framed communication show considerable promise in supporting sustained engagement in dementia-preventive lifestyle behaviors. These interventions are clinically relevant, scalable and resource-efficient, making them well-suited for broader public health implementation. By targeting long-term behavior change, this framework may contribute to healthy aging and inform future strategies for dementia risk reduction.

Laurene Da Silva Pereira, Christian Chicherio, Giovanni B Frisoni, Federica Ribaldi

Geneva Memory Center, Department of Rehabilitation and Geriatrics, Geneva University Hospitals, Geneva, Switzerland
Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland
Center for Interdisciplinary Study of Gerontology and Vulnerability (CIGEV), University of Geneva, Geneva, Switzerland

As the global population ages, dementia becomes a critical public health concern. According to the 2019 Global Burden of Disease Study, 50 million people worldwide currently live with dementia—a figure projected to reach 152 million by 2050. In response, the World Health Organization has prioritized dementia prevention. Recent evidence suggests that up to 45% of dementia cases are attributable to fourteen modifiable risk factors, including social isolation, low levels of education, physical inactivity, and cardiovascular health issues, highlighting the potential of targeted preventive interventions.

This study proposed a structured primary prevention program for the general population aged from 50 and above. The intervention included seven interactive workshops aimed at raising awareness of dementia prevention strategies, promoting evidence-based lifestyle changes, and fostering community engagement. The workshops focused on three core pillars: continued education, cognitive training, and social engagement. Sessions covered essential aspects of aging, including distinctions between physiological and pathological aging from cognitive, psychological, and physical perspectives, along with the role of physical activity and nutrition.

To evaluate participant satisfaction and perceived impact, a questionnaire was completed after the
workshop series. Key domains included feasibility, knowledge acquisition, cognitive self-perception, and motivation for lifestyle changes. Follow-up assessments at 3 and 9 months used the same questionnaire to evaluate longer-term outcomes.

Among 41 participants, baseline data showed 94% satisfaction with the workshop’s feasibility, and 98% found the content informative for dementia prevention. Additionally, 88% reported reduced concern about age-related memory decline, and 85% indicated the program supported changes in their daily life. Abstentions and negative responses (1% to 13%) were primarily linked to the program’s limited duration and technical difficulties with computerized cognitive training.

This pilot initiative demonstrates the feasibility of a structured, evidence-based primary prevention
program and its potential to enhance awareness and empower participants with tools to support dementia prevention.

Korian Wicki, Marina Wunderlin, Marc Alain Züst

University Hospital of Old Age Psychiatry and Psychotherapy, University of Bern

Recently, disturbed slow wave sleep (SWS) has been identified as an early, modifiable risk factor for dementia. SWS is crucial for memory and metabolic clearance functions, and lack of SWS causes these critical functions to suffer, which in turn worsens cognitive decline. A vicious cycle forms between cognitive decline and loss of SWS. Improving SWS could be a way to break this cycle by providing the much-needed opportunity for the brain to recuperate and ameliorate cognitive decline. We aim to enhance slow wave activity in older adults using phase-locked auditory stimulation (PLAS). Our laboratory-based results show that the magnitude of the physiological response to PLAS predicts improvement of memory functions and metabolic clearance. However, larger, long-term studies in an ecologically valid setting are needed to assess the efficacy of PLAS for the improvement of sleep, memory, and metabolic clearance with the goal of preventing cognitive decline. Laboratory-based studies quickly become economically and logistically unfeasible to achieve this goal.

We propose a blinded, sham-controlled cross-over study utilizing home-use devices to study the effect of PLAS on memory functions in 60 older adults with cognitive impairment. Participants will undergo real- and sham-PLAS in the comfort of their own homes across a 12-week study period. Cognitive performance will be assessed using engaging “serious games”, and blood will be sampled before- and after each experimental period to test levels of dementia-related biomarkers and their response to PLAS.

We expect PLAS to enhance sleep, leading to downstream effects on memory performance and metabolic clearance. Using a novel approach allowing brain-age estimation from sleep-electrophysiology, we hypothesize to see a “rejuvenating” effect of PLAS, restoring an electrophysiological profile typically seen in younger brains.

This study could pave the way for PLAS-capable home-use devices as an affordable, non-invasive tool to combat cognitive decline and could lead to novel preventative applications for memory clinics, relieving their clinical burden and improving public health.

Marina Wunderlin 1, Korian Wicki 1,2, Marc Alain Züst 1

1 University Hospital of Old Age Psychiatry and Psychotherapy, University of Bern, 3000 Bern 60, Switzerland

2 Graduate School for Health Sciences, University of Bern, 3012 Bern, Switzerland

Slow wave activity, the signature of deep sleep, has consistently been linked to amyloid-beta (Aβ), a pathophysiological marker of neurodegeneration. Less is known about how Aβ relates to specific microstructural processes within deep sleep, such as the coupling of slow waves and spindles, where better functioning reflects younger age, increased memory, and less brain atrophy.

47 older adults (agemean = 70.5 (0.68)) ranging in cognitive functioning completed one adaptation and one baseline night. A subsample (n=39, agemean = 70.5 (0.74)) additionally underwent a three-night acoustic stimulation intervention to boost slow wave activity. Blood samples post-baseline and post-intervention were analyzed for Aβ 1-42/1-40-ratio.

Irrespective of cognitive functioning levels, slow wave–spindle coupling was the best predictor for baseline Aβ, better than slow wave activity, age or cognitive functioning. Specifically, better Aβ-levels were linked to a coupling physiology resembling a younger brain. While intervention-induced increases in slow wave activity were linked to a beneficial Aβ-response across all cognitive levels, increases in slow wave–spindle coupling benefited Aβ-response exclusively in cognitively impaired individuals.

Our results suggest a link between SW–spindle coupling and Aβ that goes beyond slow wave activity. This hints towards a potential specific function of SW–spindle coupling related to the early pathophysiology of neurodegeneration.

Sayin Özge 1,2, Augusto J. Mendes 1,2, Federica Ribaldi 1,2, Giovanni B. Frisoni 1,2

1 Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland

2 Geneva Memory Center, Department of RehabilitaPon and Geriatrics, Geneva University Hospitals, Geneva, Switzerland

It is estimated that modifiable risk factors (RFs) account for approximately 45% of dementia cases worldwide. While several studies have examined the extent to which these RFs contribute to dementia incidence in other regions, such as Brazil, Denmark, China, USA, Switzerland's specific context has not been investigated. This is of particular interest because recently it has been shown that dementia incidence is declining in many high-income countries. Therefore, we aimed to calculate the population attributable fraction (PAF) in the Swiss population who have access to high-quality health care system to inform effective dementia prevention strategies. We will use the Swiss National Health Survey, a nationally representative dataset, in which potentially modifiable dementia RFs were collected every five years from 1992 to 2022. The PAF for each risk factor (RF) will be estimated based on its prevalence and the associated relative risks from prior meta-analyses. We will adjust for the communality between RFs and calculate overall weighted PAF, focusing on variations across cantons and over time.

Result: We expect that the PAF for dementia in Switzerland will show how the potentially modifiable RFs have contributed across the years to dementia incidence. We anticipate that the overall PAF might be lower than in low-income countries, reflecting Switzerland's high quality of health care system. In addition, due to the lifestyle changes in Switzerland, we expect to see a decrease in weighted PAF between 1992 and 2022. The harmonization of the data across years is ongoing and the results will be presented at AAIC in July 2025. This study is essential for understanding how potentially modifiable factors contribute to dementia risk in Switzerland. By identifying the contribution of the RFs, we will reveal the potential for dementia prevention in the Swiss context. These findings will offer crucial evidence to guide public health policies and targeted interventions, helping to reduce the burden of dementia in Switzerland.

Andrea C. Hausheer, Fabian Moser, Patrick Manser, Eling D. de Bruin

MOVE-IT, Department of Health, Eastern Switzerland University of Applied Sciences, St.Gallen, Switzerland

Sedentary lifestyles accelerate brain ageing, while regular physical activity can mitigate cognitive impairment and reduce dementia risk (Tari et al., 2025). We investigate and compare 24-hours movement behaviors in residents with moderate to major neurocognitive disorders (MNCD) against guidelines (Ross et al., 2020).
Residents with MNCD (Sonnweid AG (Wetzikon (CH)) were screened on “major neurocognitive disorders” according DSM-5® (APA, 2013), and Mini Mental State Examination (MMSE) ≥5-26 (Trivedi, 2017). Following informed consent assessments ran from September 2024 onwards. Outcomes include the amount of physically activity (PA) and sedentary behavior (SB) over 24-hours.
Out of 44 residents approached, 8 declined participation, 3 presented with MMSE < 5, and 1 person dropped out for health reasons. Data of 32 residents (13 male) show participants had a Mini-Mental Status Exam of median (IQR) 17.5 [13.75-20.0] points. SB median (IQR) was 12.98 [11.03-15.27] h/day, median (IQR) light PA 3.33 [2.49-4.18] h/day, median (IQR) moderate PA 0.19 [0.06-0.36] h/day, median (IQR) vigorous PA 0 [0-0.004] h/day, total sleep time median (IQR) 4.88 [3.85-7.19] h/day, sleep latency median (IQR) 0.31 [0.25-0.34] h/day, wake up after sleep onset median (IQR) 3.41 [2.79-4.41] h/day. Sleep efficiency median (IQR) was 61 [54-69] %.
This study gives quantified insights into PA, SB, and sleep of Swiss residents with MNCD. Compared with guidelines-based recommended PA (21.5 min/day moderate to vigorous activity), sleep (getting 7 to 8 hours of good-quality sleep) and SB (8 h/day or less) (Ross et al., 2020) our participants show unfavorable values. This warrants research into how long periods of sitting can be broken up and PA & sleep can be improved through dedicated programs.

Andra Bria

CAS in Brain Health University of Bucharest, Social Science

Language shapes perception, and nowhere is this more evident than in medical terminology. The term ""dementia"—derived from Latin meaning "away from" mind"—carries centuries of stigma, fear, and implications of irreversible decline. This linguistic framing influences how society views these conditions, how healthcare professionals approach treatment, and critically, how diagnosed individuals perceive their own futures.

While medical science has advanced our understanding of neurodegenerative conditions, recognizing spectrums, subtypes, and potential interventions, our terminology remains rooted in outdated paradigms. This talk explores the psychological and social implications of the term "dementia" and proposes a framework for developing language that acknowledges reality while preserving dignity and hope.

Drawing from research in neurolinguistics, health psychology, and patient-centered care, I'll examine how terminology affects treatment engagement, caregiver expectations, research funding, and public health messaging. Case studies from other medical fields that have successfully updated terminology (e.g., from "mental retardation" to "intellectual disability") provide valuable insights.

I propose a pathway toward terminology that emphasizes brain health rather than deficit, accurately reflects our evolving understanding of these conditions as potentially modifiable, and empowers rather than diminishes those affected. Potential alternatives include "neurocognitive condition," "cognitive health challenge," or condition-specific terms that reflect underlying pathology rather than symptoms.

This linguistic evolution must balance scientific accuracy with psychological impact. By reconceptualizing these conditions through language, we can reduce stigma, encourage earlier diagnosis, promote prevention efforts, and maintain human dignity throughout the process. The words we choose don't just reflect our understanding—they shape it. As our knowledge evolves, so too should our language, creating space for both medical reality and human hope.

Volpara, G. 1, Ribaldi, F. 2,3, Mendes, A.J. 2,3, Chicherio, C. 3,4, Cappa, S.F. 1,5, Frisoni, G.B. 2,3

1 ICoN Cognitive Neuroscience center, University Institute of Advanced Studies, Pavia, Italy.
2 Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland.
3 Geneva Memory Center, Department of Rehabilitation and Geriatrics, Geneva University Hospitals, Geneva, Switzerland.
4 Center for Interdisciplinary Study of Gerontology and Vulnerability (CIGEV), University of Geneva, Geneva, Switzerland
5 IRCCS Istituto Auxologico, Milan, Italy.

Introduction: Subjective Cognitive Decline (SCD) refers to the self-perception of cognitive difficulties despite normal performance on standard neuropsychological tests (Jessen et al., 2020). SCD is associated with an elevated risk of future cognitive impairment and dementia. This study aimed to examinate metacognition in cognitively unimpaired individuals with and without SCD (Ribaldi et al., 2021).

Materials: Subjects completed a neuropsychological test battery and a prospective–retrospective memory task consisting of a lexical decision phase followed by a recall phase. The task yielded measures of performance prediction (Judgments of Learning, JoL) and metacognitive accuracy (concordance between JoL and actual performance): assessed before the task (predictive) and after (postdictive) (Cauvin et al., 2019).

Methods: This study included 139 cognitively unimpaired individuals from the Geneva Memory Center cohort. SCD (n=80) self-reported complaints without objective deficits; volunteers (n=59) referred no complaints or deficits. All were free from neurological or psychiatric comorbidities. They completed a metacognitive task.

Results: SCD were significantly older (68 years [62 – 74]) than volunteers (62 years [57 – 68], p=.003). No differences occurred in overall task accuracy components (prospective: p=.468; retrospective: p=.415). However, SCD showed significantly lower predictive JoL in both the prospective (p=.005) and retrospective (p=.033) components. In contrast, postdictive JoL scores did not differ between groups (prospective: p=.135; retrospective: p=.400).
In the prospective memory task SCD showed reduced prospective metacognitive accuracy (p=.034), while no differences were observed in postdictive accuracy (p=.106).

Discussion: SCD show a reduced predictive Jol and predictive metacognitive accuracy in prospective memory. This did not occur after task execution, suggesting a possible metacognitive realignment. These evidences lead to hypothesize impairment of predictive metacognition as a potential early marker of cognitive decline.

Conclusion: SCD individuals have impaired predictive metacognitive accuracy despite similar performance, suggesting that this specific impairment may represent an early marker of cognitive decline.

Konstantinos Ioannis Zervas, Tom Willems, Shawn Hiew, Katharina Henke

University of Bern, Bern, Switzerland

Episodic memories fade rapidly after learning, with the majority forgotten within 24 hours, yet we hypothesized that memory traces (engrams) of the forgotten information persist within the human brain’s episodic memory system. We investigated the long-term fate of 96 newly learned face-object associations over a six-month period in 40 young, healthy participants. Using 7T whole-brain fMRI, we tracked memory traces as they are formed and retrieved 30 minutes, 24 hours, one week, and six months later. During each retrieval trial, participants rated their memory confidence. Responses rated as ‘guessed’ were classified as forgotten. At 30 minutes post-encoding, 33% of associations were forgotten; this slightly decreased to 31% at 24 hours, before rising to 38% at one week and 46% at six months. Although retrieval accuracy for guessed responses remained at chance across all intervals, underlying memory traces persisted within the episodic memory network, including hippocampal and prefrontal regions. Both their reactivation and hippocampal–prefrontal connectivity significantly correlated with guessing retrieval accuracy. Memory traces of forgotten associations closely matched the location and shape of those supporting consciously accessible memories, but they were thinner and showed greater changes over time. In contrast, traces associated with accessible memories remained more stable and showed strong recruitment of the neocortex as early as 30 minutes following encoding. These findings suggest that ostensibly forgotten memories persist in the brain although they are no longer accessible to conscious awareness. Thus, forgetting may reflect a loss of conscious access rather than a loss of the underlying information itself.

Tom Willems 1, Konstantinos Zervas 1, Finn Rabe 1,4, Andrea Federspiel 2,3, Katharina Henke 1

1 University of Bern, Institute of Psychology, Bern, Switzerland
2 Department of diagnostic and interventional Neuroradiology, Inselspital, University Hospital Bern, University of Bern, Bern, Switzerland
3 Translational Imaging Center Bern, Swiss Institute for Translational and Entrepreneurial Medicine, University of Bern, Bern, Switzerland
4 Department of Adult Psychiatry and Psychotherapy, University of Zurich, Zurich, Switzerland.

For a long time, forgetting has been taken as the dissipation of the neural memory traces (engrams). However, recent engram research in mice, however, suggests that the engrams of forgotten memories do persist. This raises the question whether engrams underlying human episodic memories also persist despite forgetting? And do forgotten memories influence human behavior implicitly? To address this question, 40 men and women learned 96 face-object pairs. Using high-resolution functional magnetic resonance imaging at 7 tesla we mapped the fate of the 96 memories at the systems level from learning to a 30-minute and onward to a 24-hour memory test. Upon each retrieval attempt, participants indicated whether they remembered or forgot the memory. Univariate and multivariate analyses of the functional brain data revealed that the engrams of forgotten memories remain implemented in the episodic memory network and continue to influence the accuracy of guessing responses at test. The engrams of forgotten memories were implemented more deeply within bilateral hippocampus overnight, while consciously accessible memories were neocorticalized overnight. The engrams of both consciously accessible and inaccessible (forgotten) memories were shifted by overnight consolidation within right hippocampus and anterior cingulate gyrus such that pattern dissimilarities supported correct retrieval responses. We provide strong evidence that forgotten human episodic memories remain implemented in the episodic memory network and continue to influence behavior implicitly. Contrary to what traditional models of memory, conscious accessibility is not a prerequisite for episodic memory.

Bouhour Camille 1, Rochas Vincent 2,3, Brunet Denis 4, Toussas Konstantin 1, Unschuld Paul 3,5, Bréchet Lucie 1

1 Department of Clinical Neurosciences, University of Geneva, Geneva, Switzerland
2 FCBG Fondation Campus Biotech Geneva, Campus Biotech, Geneva, Switzerland
3 CIBM Center for Biomedical Imaging, Cognitive and Affective Neuroimaging section, University of Geneva, Geneva, Switzerland
4 Geriatric Psychiatry Service, University Hospitals of Geneva (HUG), Thônex, Switzerland
5 Department of Psychiatry, University of Geneva, Geneva, Switzerland

Mild cognitive impairment (MCI) is the transitory stage between normal aging and dementia, affecting 10-20% of adults aged 65 and older. Despite its prevalence, the mechanisms underlying the MCI stage of cognitive decline remain largely unclear. Resting-state electroencephalography (EEG) power spectral analysis is a widely used technique to investigate functional brain changes associated with cognitive decline and aging. Previous research indicates that Alzheimer's Disease (AD) is characterized by increased power in slow-wave (delta and theta) and decreased power in fast-wave (alpha and beta) frequencies. Similar, though less consistent, spectral patterns have been observed in individuals with MCI. In contrast, healthy aging is generally associated with reduced slow-wave and alpha activity, alongside increased beta power. In this preliminary study, we performed a power spectrum analysis of resting-state EEG recordings from three groups: individuals with MCI, healthy older adults, and healthy younger adults. High-density EEG recordings (257 channels) were obtained from 30 participants in total (N=10 per group) during a 5-minute resting-state with eyes closed and analyzed using the Welch method across frequencies between 1 to 70 Hz. Preliminary uncorrected results revealed a reduction in alpha power in both MCI and healthy older adults over occipital and temporal regions, consistent with typical age-related neural changes. MCI patients also showed increased theta power compared to healthy controls over frontal and right temporal regions, indicating EEG slowing. Interestingly, beta power increased in MCI patients compared to healthy controls over frontal, parietal, and occipital regions, in contrast to previously reported reductions in AD. These findings suggest that early cognitive decline in MCI may primarily involve increases in slow-wave activity, while decreases in fast-wave activity may occur later during progression toward AD.

Micaela Borsa 1,2, Felix Pompizii 1,2, Carolina Gutierrez Herrera 1,2, Antoine Adamantidis 1,2

1 Zentrum für Experimentelle Neurologie, Department of Neurology, Inselspital University Hospital Bern, Bern, Switzerland.
2 Department of Biomedical Research, University of Bern, Bern, Switzerland.

The intricate relationship between Alzheimer’s Disease (AD) and sleep is bidirectional since sleep disturbances are increasingly recognized as both potential markers of AD and factors that may exacerbate the disease progression. In particular, reduced rapid eye movement (REM, also called paradoxical) sleep in individuals with AD is thought to contribute to cognitive decline and the AD progression. Recently, we have identified a somatodendritic decoupling in the dorsal prefrontal cortex (dPFC) during REM sleep supporting synaptic plasticity and emotional memory consolidation. We hypothesized that disruptions in this mechanism may contribute to AD-associated cognitive impairments. To investigate this, we characterized the dPFC somatodendritic decoupling in AD mice models across the sleep-wake cycle using simultaneous 2-photon calcium imaging and electrophysiological recordings in head-restrained, spontaneously sleeping mice. We found a complete absence of dPFC somatodendritic decoupling during REM sleep in 8-13 months old 5xFAD mice, primarily due to decreased dendritic activity of pyramidal (PYR) cells. Notably, age-dependent differences emerged with 5xFAD mice older than 1 year showing increased PYR somatic and dendritic activity during REM sleep compared to younger mice, possibly reflecting compensatory mechanisms for REM sleep deficits. Interestingly, a similar trend was observed in the PYR somatic activity of 5xFAD wild-type animals older than 1 year but not in 3-month-old BL6 mice, suggesting an age-related, AD-independent decrease in activity during non-REM (NREM) sleep. While the underlying mechanisms remain unclear, we propose that REM sleep-specific restoration of the dPFC somatodendritic decoupling could serve as a potential target for improving memory functions in AD.

Evelina Volovičeva 1, Stephanie Bernhard 2, Sandra Goetze 3, Alaa Othman 4, Arnold von Eckardstein 5, Jérôme Robert 6

1 University Hospital Zurich, University of Zurich, BioMed PhD program
2 University Hospital Zurich
3 ETH Zurich
4 Functional Genomics Center Zurich
5 University Hospital Zurich, University of Zurich
6 University Hospital Zurich, University of Zurich

Alzheimer’s disease (AD) is the leading cause of dementia in elderly adults, with 55 million cases recorded worldwide in 2019. Genetic variations of apolipoprotein E (APOE) affect the risk of AD development, with the APOE4 allele being detrimental, APOE3 neutral, and APOE2 protective. Yet, the relationship between APOE genotype, apoE particle composition, and apoE function in AD remains elusive.

Previous studies indicated that the role of apoE in the brain depends on particles’ cellular origin. Notably, the composition of apoE particles is also cell-type-specific, with pericyte-secreted apoE containing more cholesterol than astrocyte-secreted apoE. This creates a challenge for studying the roles of different apoE isoforms in AD pathogenesis.

We propose a genotype- and cell-type-specific approach to investigate the composition and function of apoE particles. We hypothesize that both the APOE genotype and cell type of origin influence the lipid and protein composition of apoE particles, which in turn modulates their role in AD pathology.

To test our hypothesis, we have differentiated the apoE-secreting cells of the brain: astrocytes, pericytes, and microglia from human induced pluripotent stem cells (iPSCs). By including apoE-secreting cells with different APOE genotypes (derived from isogenic iPSCs), we will investigate the influence of genetic variants on apoE particle composition in a cell-type-specific manner. We will investigate the protein and lipid compositions of apoE particles secreted by each differently genotyped cell type using liquid chromatography-mass spectrometry (LC-MS). Next, we will assess the role of these particles in AD by characterizing their effect on amyloid beta and tau accumulation in neurons.

By the end of our project, we aim to answer key questions: how the composition of apoE particles is linked to the cell type of origin and APOE genotype, and how these particles affect cerebrovascular health in the context of AD.

Linh Tran, Urban Deutsch, Britta Engelhardt, Steven T. Proulx

Theodor Kocher Institute, University of Bern, Switzerland

There has been a renewed focus on the role of central nervous system (CNS) fluids and barriers in neurological disorders such as cerebral amyloid angiopathy (CAA). A body of suggestive evidence has indicated a link between impaired amyloid-β (Aβ) clearance and CAA formation, highlighting a fundamental need to improve our insights into the mechanisms of brain clearance. Nonetheless, the mechanisms for fluid circulation and solute exchange in the brain parenchyma remain highly debated and poorly understood. In many studies, full consideration of the brain barriers that compartmentalize the CNS is lacking. With this project, we aim to provide new transgenic mouse models marking the brain barriers that will help us to fill these critical knowledge gaps between CNS fluid clearance and CAA progression. Two novel specific dual-reporter mouse strains: one visualizing CNS vasculature using Claudin5-GFP (cerebral blood vessels) and Prox1-TdTom (lymphatic vessels) and one marking the brain borders with Aqp4-mRuby3 (for astrocyte endfeet of the glia limitans) and VE-cadherin-GFP (for the leptomeninges and blood vessels), were successfully crossed with an ArcAβ transgenic mouse line, which has resulted in two novel triple-transgenic models of cerebral amyloidosis. In the first stage, Aβ deposition was observed in correlation with changes of brain barrier marker fluorescent expression using ex vivo imaging on decalcified skull or fixed brain sections at different ages (early vs chronic stage of disease). Secondly, solute drainage pathways from the intersititial fluid space were investigated through evaluation of tracer distribution at various timepoints after infusion into different brain regions and the CSF space. Our preliminary results demonstrate that our novel brain barrier reporter mice enable new perspectives on the understanding of brain clearance in rodent CAA models. Going forward, our histological findings will be validated with multiple in vivo imaging modalities, including two-photon microscopy and synchrotron-based X-ray imaging.

O. Braissant 1, V.A. McLin 2, D. Sessa 2 and K. Pierzchala 3,4

1 Service of Clinical Chemistry, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland,
2 Swiss Pediatric Liver Center, Department of Pediatrics, Gynecology and Obstetrics, University Hospitals Geneva and University of Geneva, Geneva, Switzerland,
3 CIBM Center for Biomedical Imaging, Switzerland,
4 CIBM PCI EPFL Metabolic Imaging Section, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland.

Background: Increasing evidence suggests that liver dysfunction may contribute to Alzheimer’s disease (AD) pathogenesis. Although type C hepatic encephalopathy (HE) is associated with cognitive impairments, particularly in attention and memory, the potential overlap with AD remains largely unexplored. While studies have confirmed a link between cirrhosis and cognitive decline, the role of the gut-liver-brain axis in neurodegeneration has only recently regained scientific attention, highlighting the importance of liver health for central nervous system function. However, the underlying mechanisms remain poorly understood.

Objective: To investigate the potential connection between liver failure and the development or progression of Alzheimer’s disease.

Methods: Chronic liver disease (CLD) was induced in rats via bile duct ligation (BDL). Alzheimer’s pathology was assessed through Congo red staining for amyloid beta (Aβ) and Gallyas staining for tau. Immunohistochemistry was used to evaluate changes in aquaporin (water) channels (Aqp1, Aqp4, Aqp9) and glial fibrillary acidic protein (GFAP) expression. Blood levels of neurodegeneration markers (neurofilament light chain (NfL), Aβ, p-tau, t-tau, GFAP, and myelin oligodendrocyte glycoprotein (MOG)) and bile acids were measured and compared between BDL and SHAM-operated control rats.

Results: BDL rats exhibited hallmark AD pathology, including intracellular Aβ deposition and tau abnormalities characteristic of pre-tangle and neurofibrillary tangle (NFT) stages. These changes were accompanied by dysregulation of brain water channels and significant alterations in circulating neurodegeneration biomarkers and bile acid profiles compared to SHAM controls.

Conclusion: Our findings highlight the importance of liver health for central nervous system function and suggest a contributory role in cognitive decline, potentially increasing the risk of Alzheimer’s disease. These results support the integration of liver function assessment into the diagnostic process for dementia and underscore the importance of a systemic approach to understanding AD pathogenesis.

Spyridoula Mitsikosta, Jürgen Ripperger, Urs Albrecht

Dept. of Biology, University of Fribourg

Cyclin-dependent kinase 5 (Cdk5) is a proline-directed serine/threonine protein kinase that plays a vital role in the cell cycle and the central nervous system. Cdk5 is controlled by neuron-specific activators p35 and p39 and it becomes active upon direct binding, forming a heterodimer complex. Cdk5 has emerged as a key mediator in Tau phosphorylation and its hyperactivation promotes the amyloid plaque formation that leads to Alzheimer’s disease. Moreover, Cdk5 regulates the mammalian circadian clock by phosphorylating key clock components such as CLOCK and PER2, and recent studies have shown that its activity is downregulated by light. In this study, we focus on the repression of the Cdk5 by light and more specifically, we aim to identify the light modification site that is responsible for the Cdk5 activity. For this reason, we test various mutations on the identified phosphorylation sites that would have an effect on the activity of the Cdk5. For the in vitro studies we are using forskolin as a light mimic in NIH 3T3 fibroblast mouse cells which have been transfected with the Cdk5 mutants.

Saina Asadi 1, Maria Giulia Preti 2,3, Dimitri Van De Ville 2,3, Giovanni Frisoni 4, Valentina Garibotto 5, Gilles Allali 6, Sara Stampacchia 7, Alessandra Griffa 6,2

1 Department of Radiology, University Hospital of Lausanne and University of Lausanne, Lausanne (CHUV-UNIL), Vaud, Switzerland

2Neuro-X Institute, Ecole Polytechnique Fédérale De Lausanne (EPFL), Geneva, Switzerland

3Department of Radiology and Medical Informatics, Faculty of Medicine, University of Geneva, Switzerland

4Memory Clinic and LANVIE - Laboratory of Neuroimaging of Aging, University Hospitals and University of Geneva, Geneva, Switzerland

5Diagnostic Department, Division of Nuclear Medicine and Molecular Imaging, University Hospitals of Geneva, Geneva, Switzerland

6Leenaards Memory Center, Department of Clinical Neurosciences, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

7Laboratory of Cognitive Neuroscience, Neuro-X Institute, Faculty of Life Sciences, Ecole Polytechnique Fédérale de Lausanne, Geneva, Switzerland

Alzheimer’s disease is characterized by progressive amyloidopathy and tauopathy. Recent evidence highlights greater-than-expected inter-individual variability in tau accumulation patterns across brain regions, closely reflecting clinical heterogeneity. However, the mechanisms underlying tauopathy spread remain poorly understood. A leading hypothesis posits that neural connections serve as conduits for the spread of aggregated tau across brain regions, with neural activity exacerbating tau release and spreading. Variability in white matter connections (structural connectivity) and neural activity synchronization (functional connectivity) may therefore contribute to differences in tau accumulation and clinical progression across individuals.

In this study, we used multimodal brain imaging and advanced analytical methods to investigate whether tau spreading is driven by structural connectivity, functional connectivity, or their combination. We conducted a cross-sectional analysis of 198 Memory Clinic patients (71.5 ± 7.7yo; 101 females) assessed at the Memory Center of Geneva University Hospitals. Participants underwent amyloid-PET, tau-PET, magnetic resonance imaging, and neuropsychological evaluation. Reference structural, functional, and partialized functional connectivity—a measure reflecting functional synchronization through monosynaptic structural pathways—were derived from 100 healthy subjects from the Human Connectome Project. Using graph signal processing tools, an emerging area in neuroscience research, we assessed the similarity between regional tau accumulation and the underlying brain connectivity networks. At the group level, we found that tau distribution closely followed structure-function connectivity patterns, particularly as summarized by partialized functional connectivity. At the individual level, a score quantifying tau spreading through the structure-function brain network was associated with Braak stage and with cognitive impairment, in both amyloid-positive and cognitively impaired amyloid-negative individuals.

These findings from a Memory Clinic cohort support the hypothesis that tau spreading is mediated by both structural and functional brain connectivity. Anticipating the pathways of tau spreading could inform patients and caregivers about the likely clinical progression and support the development of personalized interventions.

Aurélie Bussy 1,3, Camille Cathala 3,4, Ferath Kherif 3, Antoine Lutti 3, Bogdan Draganski 1,2,3

1 Universitätsklinik für Neurologie, Inselspital Bern, University of Bern, Switzerland

2 Institute for diagnostic and interventional neuroradiology, Inselspital Bern, University of Bern, Switzerland

3 Laboratory for Research in Neuroimaging LREN, Centre for Research in Neurosciences, Department of Clinical Neurosciences, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.

4 Signal Processing Laboratory, Ecole Polytechnique Fédérale de Lausanne, Switzerland

Introduction: Cerebral small vessel disease, a major contributor to stroke and cognitive decline with aging, commonly appears as white matter hyperintensities (WMH) on MRI. WMHs vary in location and microstructure, reflecting pathology such as demyelination and axonal loss. Cardiovascular risk factors (CVR)—including systolic blood pressure (SBP), heart rate, and cholesterol—are linked to WMH burden. This study examines how longitudinal CVR influences WMH volume and microstructure using quantitative MRI (qMRI).

Methods: We analyzed data from 153 participants (86 women) in the BrainLaus study, a sub-cohort of CoLaus|PsycoLaus (Lausanne, Switzerland), each with at least three CVR assessments over an average of 14.45 years. CVR metrics included SBP, DBP, BMI, bioimpedance, WHR, heart rate, glucose, insulin, and lipid profiles. MRI was performed on a 3T Siemens Prisma using FLAIR, T1, multi-echo FLASH, and DWI sequences. WMHs were segmented with a ResUNet-based algorithm. Quantitative maps (MTsat, R1, R2*) were generated. CVR trajectories were modeled using splines, and area under the curve (AUC) was computed and normalized per participant. Statistical analyses included FDR-corrected linear models with sex interactions.

Results: Higher CVR (e.g., SBP, LDL, WHR, glucose) was associated with increased WMH volumes, with stronger effects in women for SBP, LDL, total cholesterol, and triglycerides. qMRI revealed sex-specific microstructural changes: in women, lipid-related CVR correlated with increased extracellular water (MD, ISOVF); in men, hypertension and heart rate were linked to lower myelin content (MT, R1) and increased g-ratio, indicating myelin thinning.

Discussion: CVR impacts WMH differently in men and women, suggesting distinct underlying mechanisms. These findings support sex-specific strategies for managing CVR to preserve white matter health and reduce cognitive decline risk.

Valentin Zufferey 1, Jeanne Espourteille 1, Enea Parietti 1, Héloïse Guenon 1, Nicolas Preitner 2, Morvane Colin 3, Kevin Richetin 1,4

1 Centre for Psychiatric Neurosciences (CNP), Lausanne University Hospital (CHUV) - University of Lausanne (UNIL), 1015 Lausanne, Switzerland
2 AC Immune SA, 1015 Lausanne, Switzerland
3 Univ. Lille, Inserm, CHU Lille, LilNCog—Lille Neuroscience & Cognition, 59000 Lille, France
4 Leenaards Memory Centre, Lausanne University Hospital (CHUV) - University of Lausanne (UNIL), 1011 Lausanne, Switzerland

Astrocytes, integral to brain homeostasis and pathology, exhibit mitochondrial sensitivity to extracellular pathological material, positioning them as ideal biosensors for neurological disorders. Richetin et al. (2020) demonstrated that tau accumulation in astrocytes disrupts mitochondrial dynamics, correlating with synaptic dysfunction and cognitive decline in Alzheimer's disease models. Building on this, Perbet et al. (2023) revealed that brain-derived extracellular vesicles (BD-EVs) transfer neuron-derived tau isoforms (3R/4R) to astrocytes, inducing isoform-specific mitochondrial responses: 3R-tau fragments mitochondria, while 4R-tau increases branching. Leveraging lentiviral-delivered MitoTimer biosensors, our lab established a live-imaging platform to monitor mitochondrial turnover and redox states in astrocytes, thus enabling real-time detection of pathology-induced changes. Using this approach, we detected pathological BD-EVs in human biofluids (plasma, cerebrospinal fluid), with mitochondrial signatures distinguishing Pick's disease (3R-tau) from progressive supranuclear palsy (4R-tau). Furthermore, synthetic tau extracellular paired-helical filaments (ePHF-tau) were detected by astrocytes in a 3D neuron-astrocyte co-culture, where glial cells displayed a much stronger response compared to neurons. The phagocytic activity of astrocytes, their anatomical positioning at blood-brain interfaces, and their mitochondrial plasticity make them sensitive "sniffer cells" for brain-derived pathological material, which can be specifically and non-invasively isolated from biofluids like plasma or saliva. While current limitations include temporal data complexity and intercellular variability, scalability improvements could enable high-throughput screening for diagnostics and therapeutic monitoring in the future.

Aurélien Riou, Anne Eckert, Amandine Grimm

Research Cluster Molecular & Cognitive Neuroscience, Department of Biomedicine, University of Basel, Basel, Switzerland
Neurobiology Laboratory for Brain Aging and Mental Health, Psychiatric University Clinics, Basel, Switzerland

Tauopathies are a group of neurodegenerative diseases characterized by the pathological accumulation of abnormal tau protein. A consequence of tau pathologies is mitochondrial dysfunctions, which affect essential processes such as mitochondrial transport, bioenergetic, and dynamic. Given the high energy demands of neurons, tau-induced mitochondrial impairment significantly contributes to neuronal vulnerability and degeneration. Studies have revealed that cells can transfer mitochondria between them to help energy-deficient cells. This process, known as intercellular mitochondrial transfer, can occur via two different pathways: an indirect transfer via extracellular vesicles and a direct transfer via tunneling nanotubes and gap junctions. Considering the impact of pathological tau on mitochondrial transport and cytoskeletal dynamics, we hypothesized that abnormal tau protein negatively affects intercellular mitochondrial transfer. In this study, we investigated intercellular mitochondrial transfer between astrocytic and neuronal cells under tauopathy conditions. Using A172 astrocytic cells and SH-SY5Y neuronal cells bearing or not the P301L tau mutation, we observed that abnormal tau surprisingly enhances mitochondrial transfer from astrocytic to neuronal cells, predominantly through contact-dependent mechanisms. While the fate of transferred mitochondria was essentially unchanged, a trend toward impaired mitophagy was observed in neuronal cells bearing the P301L mutation. To summarise, our data highlight a novel pathway by which abnormal tau protein impacts mitochondrial function, namely the transfer of astrocytic mitochondria to neuronal cells

Bommarito Giulia 1, Griffa Alessandra 1,3, Michel Patrik 2, Hall Caroline 1, Salvioni Chiabotti Paolo 1, Pistocchi Silvia 4, Aleman-Gomez Yasser 4, Damian, Daniel 1, Jreige Mario 5, Prior John 5, Dunet Vincent 4, Rouaud Olivier 1, Hagmann Patric 4, Allali Gilles 1

1 Leenaards Memory Center, Department of clinical neurosciences, Lausanne University Hospital and University of Lausanne (CHUV-UNIL), chem. de Mont-Paisible 16, 1011, Lausanne, Switzerland

2 Stroke Center, Neurology Service, Department of clinical neurosciences, Lausanne University Hospital and University of Lausanne (CHUV-UNIL), Rue du Bugnon 46, 1011, Lausanne, Switzerland

3 Medical Image Processing Laboratory, Neuro-X Institute, École Polytechnique Fédérale De Lausanne (EPFL), Chemin des Mines 9, 1202, Geneva, Switzerland

4 Department of Diagnostic and Interventional Radiology, Lausanne University Hospital and University of Lausanne (CHUV-UNIL), Rue du Bugnon 46, 1011, Lausanne, Switzerland

5 Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne (CHUV-UNIL), Rue du Bugnon 46, 1011, Lausanne, Switzerland

Introduction: Enlarged perivascular spaces (ePVS) are one of the imaging features of cerebral small vessel disease, a common co-pathology in patients with dementia with Lewy bodies (DLB) and Alzheimer’s disease (AD). Moreover, they are part of the brain glymphatic system. In this study we characterized ePVS in DLB and AD cohorts referred to a memory center.

Methods: In this retrospective study, we collected data from patients with a clinical DLB diagnosis or clinico-biological AD diagnosis, evaluated at our memory center (Leenaards Memory Center, Lausanne University Hospital, Switzerland). Patients were assessed for ePVS at the level of the basal ganglia (BG) or centrum semiovale (CSO), using established guidelines and scores. Differences in prevalence between the two groups were investigated. Regression model was used to evaluate the association between ePVS cognition as expressed by Montreal Cognitive Assessment (MoCA) test.

Results: We included 71 DLB and 71 age- and gender-matched AD patients. Both patients with DLB and AD presented with a high ePVS score at the level of the CSO (median (interquartile range)= 3.0 (1.0) on a score ranging from 0 to 4). At the BG level, DLB patients presented with a median score of 2.0 (1.75), while AD patients presented with a median score of 1.0 (1.0). When compared, DLB patients had higher ePVS load in the BG (p= .011) and CSO (p= .004). No association was found between ePVS and MoCA.

Discussion: DLB and AD patients present with a high load of CSO ePVS. At both CSO and BG level, DLB patients present with a greater ePVS load compared to AD, possibly related to glymphatic system dysfunction. EPVS do not seem to impact on cognition on these patients.

Aatmika Barve 1,5†, Jeanne Espourteille 1†, Valentin Zufferey 1, Elodie Leroux 2, Romain Perbet 2, Séverine Bégard 2, Raphaëlle Caillierez 2, Claude-Alain Maurage 3, Nicolas Toni 1, Luc Buée 2, Morvane Colin 2, Kevin Richetin 1,4

1 Lausanne University Hospital (CHUV) and University of Lausanne, Department of Psychiatry, Center for Psychiatric Neurosciences, 1008 Prilly-Lausanne, Switzerland
2 University of Lille, Institut National de la Recherche (INSERM), CHU-Lille, UMR-S 1172, Lille Neu roscience & Cognition, Lille, France
3 Univ. Lille, CHU-Lille, CRB/CIC1403, Centre de Resources Biologiques du Centre d’Investigation Clinique, Lille, France
4 Lausanne University Hospital (CHUV) and University of Lausanne, Department of Clinical Neuro science (DNC), Leenaards Memory Center, 1011 – Lausanne, Switzerland
5 Department of Neurosciences and Movement Sciences, University of Fribourg, Fribourg, Switzerland

Tauopathies are neurodegenerative disorders characterized by the abnormal intracellular aggregation of different tau protein isoforms. To date, there is limited understanding and few elements to distinguish the cellular consequences associated with the accumulation of tau 3R and 4R isoforms. To further our understanding of the underlying mechanisms behind these specific disruptions in tau homeostasis, we investigated the content of extracellular vesicles present in
the brain parenchyma of patients diagnosed with 3R and 4R tauopathies, and compared these profiles to those of healthy controls.

Marta Vilademunt 1,2, Luana Piano-Vieira 2, Fabio Grieco 1,2, Charline Carron 1,2, Thibault Sprenger 1,2, Nicolas Toni 1,2

1 Lausanne University Hospital (CHUV)
2 Lausanne University (UNIL)

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder marked by cognitive decline, memory loss, and behavioral changes. A key pathological feature of AD is the disruption of adult hippocampal neurogenesis (AHN), a process essential for learning, memory, and emotional regulation. The neurogenic niche, where AHN occurs, is tightly regulated by glial cells—particularly microglia and astrocytes.

Microglia influence neurogenesis through immune surveillance and cytokine secretion, while astrocytes support neuronal development and synaptic integration. Although both cell types are essential for maintaining niche homeostasis, their interactions in the context of AD remain poorly understood.

Our research aims to elucidate the glial crosstalk within the neurogenic niche and explore therapeutic strategies to restore its function in AD. Previous findings from our lab demonstrated that astrocyte-secreted molecules attenuate microglial activation, reduce pro-inflammatory cytokine release, and enhance neurogenesis.

In the APP/PS1 mouse model of AD, we observed early impairments in AHN, with a more pronounced and earlier onset in females (4 months) compared to males (6 months). This was characterized by a decrease in doublecortin-positive (DCX+) immature neurons, despite increased proliferation. Building on this, we compared the effect of astrocyte-derived molecules and Toll-like receptor inhibition on APP/PS1 mice, on the neurogenic niche, adult neurogenesis and cognitive function.

Our preliminary data suggest that targeting astrocyte-microglia interactions can rebalance the neurogenic niche, promote neurogenesis, and potencially alleviate cognitive deficits in AD. This study highlights the therapeutic potential of modulating glial communication to combat neurodegenerative diseases.

Tetiana Serdiuk 1, Virginie Redeker 2, Jimmy Savistchenko 2, Sandesh Neupane 3, Walther Haenseler 4, Yanick Fleischmann 5, Sabrina Keller 1, Cinzia Tiberi 6, Ruxandra Bachmann-Gagescu 4,7,8, Matthias Gstaiger 1, Thomas Braun 6, Roland Riek 5, Steve Gentleman 9, Adriano Aguzzi 3, Natalie de Souza 1, Ronald Melki 2, Paola Picotti 1

1 Institute of Molecular Systems Biology, Department of Biology, ETH Zurich, Zurich, Switzerland
2 Institut Francois Jacob (MIRCen), CEA, and Laboratory of Neurodegenerative Diseases, CNRS, Université Paris-Saclay, Fontenay-Aux-Roses, France
3 Institute of Neuropathology, University of Zurich, Switzerland
4 URPP Adaptive Brain Circuits in Development and Learning, University of Zurich, Zurich, Switzerland
5 Laboratory of Physical Chemistry, ETH Zurich, Zurich, Switzerland
6 Biozentrum, University of Basel, Switzerland
7 Department of Molecular Life Sciences, University of Zurich, Switzerland
8 Institute of Medical Genetics, University of Zurich, Schlieren, Switzerland
9 Department of Brain Sciences, Hammersmith Hospital, Imperial College London, UK

The aggregation of the protein alpha-synuclein (αSyn) is a hallmark of several neurodegenerative diseases, including Parkinson’s disease (PD), Dementia with Lewy bodies (DLB), and Multiple System Atrophy (MSA). Here, we aim to characterize the structure-function relationship of disease-specific αSyn fibrils at a proteome-wide level with a novel structural proteomics method.

We employ the Limited proteolysis coupled with mass spectrometry (LiP-MS) to detect structural changes and altered protein-protein interactions across thousands of proteins, with a resolution of ~10-20 amino acids. This enables LiP-MS to provide structural insights that remain inaccessible to other omics approaches.

In vitro, in neurons and directly in native patient brain homogenates, we show that pathogenic αSyn from distinct synucleinopathies (PD, DLB and MSA) exhibit structural differences. Furthermore, we found that fibrillar structural differences are associated with distinct structural responses in neuronal proteomes and pathways. Selected hits were validated using CRISPR-based tools.
Our findings reveal that fibrillar structural differences directly influence fibril interactomes and neuronal responses.

Augusto J. Mendes 1,2, Federica Ribaldi 1,2, Michela Pievani 3, Cecilia Boccalini 4, Valentina Garibotto 4,5,6, Giovanni B. Frisoni 1,2

1 Laboratory of Neuroimaging of Aging (LANVIE),University of Geneva, Switzerland;
2 Geneva Memory Center, Department of Rehabilitation andGeriatrics, GenevaUniversityHospitals,Switzerland;
3 Laboratory of Alzheimer’s Neuroimaging and Epidemiology (LANE), IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy;
4 Laboratory of Neuroimaging and Innovative Molecular Tracers (NIMTlab), Geneva University Neurocenter and Faculty of Medicine, University of Geneva, Switzerland;
5 Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospitals, Switzerland;
6 CIBM Center for Biomedical Imaging, Geneva, Switzerland.

Background and Objectives:The amyloid cascade hypothesis suggests Alzheimer’s disease (AD) progresses from amyloid deposition to tau pathology, neurodegeneration, and cognitive impairment. The 2024 Alzheimer’s Association (AA-2024) criteria introduce a two-dimensional biological-clinical staging system to account for copathologies and resilience. We estimated the proportion of individuals whose trajectories align with the amyloid cascade.

Methods: Cross-sectional data from 256 amyloid-positive participants (mean age 72.7 years; 51% female) in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) were classified in a 4×4 biological/clinical matrix. Biological stages (A to D) were based on amyloid and tau PET imaging, using five cutoff methods from Jack et al. (2017). Clinical stages ranged from cognitively unimpaired to dementia. Participants were categorized as (1) compliant (aligned biological and clinical stages), (2) resilient (advanced biological stage, milder clinical symptoms), or (3) copathologic (early biological stage, advanced clinical symptoms). Observed distributions were compared with hypothetical zero- and high-penetrance scenarios using χ² tests.

Results: Between 31%–36% of individuals were compliant with the amyloid cascade, depending on the tau-PET cutoff. Compliance rates were higher than in a zero-penetrance model but lower than in a high-penetrance model (p < 0.01). Copathologic and resilient patterns varied significantly (17%–63% and 6%–52%, respectively; p < 0.001).

Conclusion: Only about one-third of amyloid-positive individuals exhibited strict compliance with the amyloid cascade sequence. These findings emphasize the considerable influence of resilience and copathologies in modulating disease progression beyond the traditional amyloid-first model.

Anna Chauveau 1, P. Salvioni Chiabotti 2, D. Berdous 1, C. Gaillard 1, G. Allali 2, O. Rouaud 2, S. Perriot 1, 3

1 Lemanic Institute for Regenerative Medicine, Epalinges, Switzerland
2 Leenaards Memory Center, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland
3 Centre for Psychiatric Neurosciences, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

Mutations in the progranulin (GRN) gene are a major cause of Frontotemporal Lobar Degeneration (FTLD). To date, over 65 variants have been identified, mostly nonsense or splice-site mutations, resulting in a loss of function with progranulin (PGRN) haploinsufficiency and TDP-43 neuropathology. However, it has currently been not proven that missense mutations may also cause FTLD.
Here, having diagnosed a patient with FTLD presenting borderline levels of PGRN and a novel missense GRN mutation, we aim to assess the pathogenicity of this novel variant using human induced pluripotent stem cells (hiPSCs) as a model.

For this study, we reprogrammed hiPSCs from the aforementionned patient’s cells. Then, we generated and phenotyped neurons and microglia derived from these hiPSCs. Last, using CRISPR-Cas9 technology, we are correcting the GRN missense mutation in the patient-derived hiPSCs and introducing the mutation in control hiPSCs to firmly demonstrate its pathogenicity.

Characterization of patient-derived neurons revealed several hallmarks of GRN-FTLD (cytoplasmic TDP-43, lysosomal dysregulation, and cell death). We are now characterizing patient-derived microglia and their role in neuroinflammation and neurotoxicity. These experiments demonstrate a disease-associated phenotype in hiPSC-derived brain cells from this patient. Last, characterization of CRISPR-edited cells will allow us to definitely uncover whether this missense mutation is the cause of the patient’s pathology.

Our model provides for the first time biological insights into the pathogenic role of missense GRN mutations and offers a model to test potential treatments on patient-derived neurons and microglia.

Paolo Salvioni Chiabotti 1, Olivier Rouaud 1, Marc Sollberger 2,3, Sylvain Perriot 4

1 Leenaards Memory Center, Department of Clinical Neuroscience, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

2 Memory Clinic, University Department of Geriatric Medicine FELIX PLATTER, Basel, Switzerland

3 Department of Neurology, University Hospital Basel, Switzerland

4 Center for Psychiatric Neuroscience, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

Human induced pluripotent stem cells (hiPSCs) hold a great potential for treating neurodegenerative diseases, such as frontotemporal dementia (FTD) – the second cause of neurodegenerative dementia in individuals younger than 60 - due to their ability to differentiate into various cell types found in the central nervous system. By generating patient-specific hiPSCs and directing their differentiation into neuronal populations, personalized disease models can be created, offering unique opportunities for drug screening, disease mechanism understanding, and potential cell-based therapies.

After an initial single case study where we thoughtfully described a novel missense progranulin (GRN) mutation in a behavioral variant FTD (bvFTD) with biological hallmarks of TDP43-FTD with an hiPSC model, we introduced this technique with various selected FTD patients in Lausanne, with phenotypes ranging from progressive supranuclear palsy (PSP) to semantic dementia (SD).

Having established this technique by bringing together the clinical data of the Leenaards Memory Center in Lausanne University Hospital and the innovative development of the iPSC platform in the Center for Psychiatric Neuroscience (CPN) in Lausanne University Hospital, we aim at implementing it at a national level bringing together the clinical expertise of members of the Swiss-FTD network to our local knowledge.

Berdugo-Vega Gabriel, Sierra Cesar, Calati Veronika, Orsat Jules, Gräff Johannes

Laboratory of Neuroepigenetics, Brain Mind Institute, School of Life Sciences, Ecole Polytechnique Federale Lausanne (EPFL); Lausanne, 1015, Switzerland.

Counteracting cognitive decline is an important goal in regenerative medicine. Recently, partial cellular reprogramming has emerged as a promising strategy to restore cellular function and promote tissue regeneration, but whether this approach can be targeted to sparse cell types responsible for cognitive frailty is not known. In aged mice and mouse models of Alzheimer’s Disease (AD) engram cells responsible for memory formation are functionally impaired, but the underlying molecular mechanisms of this impairment and potential means to rescue it have not been explored. We found that partial reprogramming of engram cells in aged mice and models of AD by OSK-mediated gene therapy restored engram reactivation, reversed the expression of age and disease-related cellular markers and re-established aberrant epigenetic and transcriptional alterations related to synaptic plasticity. Importantly, irrespective of the brain area targeted or the behavioural paradigm employed, it also recovered learning and memory capacities towards those observed in healthy young animals, suggesting cognitive rejuvenation. These results posit that partial reprogramming targeted to sparse cell populations in the brain can be exploited to restore cognition during aging and disease.

Roberto Frigerio 1, Nils Briel 2, Francesca Torrini 1, Jonathan Garlipp 1, Alessandro Gori 3, Tobias Weiss 2, Paolo Arosio 1

1 Department of Chemistry and Applied Biosciences, Institute for Chemical and Bioengineering ETH Zurich Zurich 8093, Switzerland
2 Department of Neurology, University Hospital Zürich, 8091, Switzerland
3 Consiglio Nazionale delle Ricerche Istituto di Scienze e Tecnologie Chimiche “Giulio Natta” (SCITEC) Milan 20131, Italy

Extracellular vesicles (EVs), including brain-derived EVs, hold significant promise for liquid biopsy applications, particularly in the diagnosis and monitoring of neurodegenerative diseases such as Alzheimer's, Parkinson's, and Amyotrophic Lateral Sclerosis. However, their heterogeneous nature presents a major bottleneck to clinical translation, complicating their efficient and selective isolation from complex biofluids.1
Here, we introduce a robust EV isolation method based on phase-separated zwitterionic (ZW) coacervates. These coacervates form across a broad range of pH and ionic conditions, ensuring compatibility with diverse biological fluids.2 Their intrinsic antifouling properties reduce nonspecific binding, enabling the selective capture of EVs when functionalized with affinity probes. We demonstrate this approach using a membrane-sensing peptide—a pan-specific probe targeting the high-curvature lipid bilayer of EVs—to overcome the variability in EV surface marker expression.3 This strategy is applied to pull down and release EVs from untreated serum, cerebrospinal fluid, and urine, achieving high yields and preserving vesicle integrity.
We demonstrate the method's utility both as a preparative step for downstream analysis and as a one-pot assay using flow cytometry, significantly streamlining pre-analytical workflows.2 Both approaches are employed to profile EV-associated biomarkers and to distinguish clinical cohorts of neurodegenerative disease patients. Overall, functionalized ZW coacervates offer a powerful platform for selective EV isolation and direct analysis from complex samples, advancing the development of EV-based diagnostics.

Arthur Paquis, Carla Cristina Barbosa Martinho, Giovani B. Frisoni

Centre de la Mémoire, Hôpitaux Universitaire de Genève (HUG)

The Biobank at Geneva University Hospital’s Memory Center is a key infrastructure supporting translational research in neurodegenerative diseases, particularly Alzheimer’s disease and related disorders. To date, the biobank has collected over 10,000 cerebrospinal fluid (CSF) samples, 40,000 blood-derived samples(including plasma, serum, and PAXgene), 18,000 stool samples, and 4,000 DNA samples extracted from both blood and stool. These biospecimens are linked to comprehensive clinical datasets from more than 2,000 participants.
Participants are categorized into diagnostic groups including Mild Cognitive Impairment (MCI), dementia, and cognitively healthy controls, with detailed medical histories, neuropsychological assessments, and imaging data systematically recorded. This rich dataset enables longitudinal studies on disease progression, biomarker discovery, and personalized medicine approaches.
The biobank operates under a robust ethical and legal framework, ensuring informed consent, data anonymization, and full compliance with Swiss federal regulations and international standards. It is certified under ISO quality management systems and adheres to the best practices promoted by the Swiss Biobanking Platform (SBP), the national infrastructure for biobanking excellence. These certifications guarantee high-quality sample handling, traceability, and interoperability with national and European research networks.
This poster outlines the operational model of the biobank, including patient recruitment, sample logistics, data integration, and governance. The infrastructure is supported by secure IT systems and a multidisciplinary team committed to scientific rigor and ethical integrity. By providing access to high-quality, well-annotated biospecimens, the biobank serves as a cornerstone for cutting-edge research in memory disorders and contributes to the advancement of precision medicine.

Lily Henry, Thomas Schneider, Ansgar Felbecker, Peter Nirmalraj

1 Transport at Nanoscale Interfaces Laboratory, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, CH-8600, Switzerland.

2 Department of Neurology, Cantonal Hospital St. Gallen, St. Gallen, CH-9007, Switzerland

3 Universitätsklinik für Neurologie, Inselspital, University of Bern, Bern

Amyloid β (Aβ) peptides play a central role in Alzheimer’s disease (AD), with isoforms Aβ-40 and Aβ-42 exhibiting distinct aggregation pathways. Clarifying the interactions between the two isoforms and differences in protein aggregate morphology in body fluids is key for monitoring disease progression and identifying potential therapeutic targets. We present a multiscale imaging approach combining atomic force microscopy and super-resolution fluorescence microscopy to resolve and chemically distinguish aggregates formed when Aβ-40 and Aβ-42 are co-incubated in both phosphate-buffered saline (PBS) and synthetic human cerebrospinal fluid (CSF). At the single-particle level, we show that Aβ-40 exhibits an inhibitory effect on Aβ-42 aggregation, likely through co-localization of Aβ-40 oligomers along the fibrillar structures of Aβ-42. These findings are supported by FTIR and ThT kinetic aggregation assays. Importantly, we extended our approach to human patient CSF samples and demonstrated that this method can differentiate between the two amyloid isoforms in CSF of patients at both early and late stages of Alzheimer's disease and observe a correlation between aggregate morphology and patients' clinical evaluation. Finally, we are adapting the developed protocols to screen amyloid and tau isoforms in the plasma of patients at various of decline in memory and cognition.

Konstantin Toussas 1, Damien Marie 2, Stefania Konstantopoulou 1, Camille Bouhour 1, Paul G. Unschuld 3,4, Lucie Bréchet 1,3

1 Department of Clinical Neurosciences, University of Geneva, Geneva, Switzerland
2 CIBM Center for Biomedical Imaging, Cognitive and Affective Neuroimaging section, University of Geneva, Geneva, Switzerland
3 Geriatric Psychiatry Service, University Hospitals of Geneva (HUG), Thônex, Switzerland
4 Department of Psychiatry, University of Geneva, Geneva, Switzerland

Mild Cognitive Impairment (MCI) is characterized by cognitive decline, particularly in memory, with relatively preserved daily functioning. A key feature of MCI is an accelerated grey matter (GM) volume loss in the fronto-temporal regions, including the hippocampus. Yet, the link between region-specific brain atrophy and behavioral symptoms remains unclear. As part of the MemStim randomized clinical trial, we examined GM volume differences using voxel-based morphometry on T1-weighted MRI scans from 33 individuals with MCI and 58 gender and age-matched healthy controls (HC). As expected, MCI patients performed worse than controls on the Montreal Cognitive Assessment (MoCA), a standardized diagnostic tool for MCI. Exploratory whole-brain analysis revealed widespread GM frontotemporal atrophy in MCI patients, centered on the left hippocampus-amygdala complex. Significant GM reductions were also found in the anterior cingulate cortex, orbitofrontal gyrus, and inferior temporal gyrus. Several cognitive scores were significantly associated with these brain region clusters. Notably, the left hippocampus cluster strongly correlated with episodic memory scores of the Logical Memory Story B test. The right amygdala cluster moderately correlated with the Rey-Osterrieth Complex Figure test scores and the 3-Objects-3-Places test. These preliminary findings highlight the utility of morphometric approaches in characterizing structural brain changes in MCI patients and suggest the value of further investigations into cortical thickness, pathological protein burden, and alterations in network connectivity.

Elif Harput 1, Cecilia Boccalini 1, Max Scheffler 6, Jeanne Espourteille 1, Karl-Olof Lovblad 8, Francois R. Jornayvaz 7, Giovanni B. Frisoni 2,3, Valentina Garibotto 1,4,5

1 Laboratory of Neuroimaging and Innovative Molecular Tracers (NIMTlab), Geneva University Neurocentre and Faculty of Medicine, University of Geneva, Geneva 1205, Switzerland

2 Geneva Memory Centre, Department of Rehabilitation and Geriatrics, Geneva University Hospitals, Geneva 1205, Switzerland

3 Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva 1205, Switzerland

4 Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospitals, Geneva 1205, Switzerland

5 Centre for Biomedical Imaging, University of Geneva, Geneva 1205, Switzerland

6 Division of Radiology, Geneva University Hospitals, Geneva, Switzerland

7 Service of Endocrinology, Diabetes, Nutrition and Patient Therapeutic Education, Geneva University Hospital, Geneva, Switzerland

8 Neuroradiology, Geneva University Hospitals, Geneva, Switzerland

Type 2 diabetes is a risk factor for cognitive decline and dementia, through structural and vascular brain changes that may act in synergy with Alzheimer's disease (AD).This study aimed to assess the contributions of AD biomarkers and cerebrovascular changes to cognition in diabetic vs non-diabetic individuals.

We included 185 subjects from Geneva Memory Center with available amyloid- and tau-PET, MRI exam with T1 and FLAIR sequences, and Mini-Mental State Examination (MMSE). Amyloid and tau status were defined based on PET visual assessment. Indices of vascular pathology, namely white matter lesion volumes were extracted from MRI images and age-related white matter changes (ARWMC) were scored. Four subgroups were defined based on amyloid positivity/negativity and diabetic/non-diabetic status. To evaluate differences across groups, analyses of variance were performed controlling for age, sex, and diagnostic stage (subjective cognitive decline [SCD], mild cognitive impairment [MCI], and dementia). The moderating effect of diabetes on the relationship between MMSE and biomarkers was tested using linear regression analyses with interaction terms, adjusting for the same covariates.

The prevalence of diabetes was 25% (46 subjects). Diabetic individuals showed higher frequencies of MCI and dementia (p=0.04), and tau-PET positivity (p=0.03), and presented higher ARWMC scores (p=0.005). In amyloid-negative individuals, MMSE was significantly lower in diabetic individuals than in those without diabetes (p=0.0001). In amyloid-positive individuals, ARWMC scores were significantly higher in diabetics than non-diabetic subjects (p=0.001). No significant interaction effects with diabetes were found in the whole group, suggesting that the association between AD biomarkers or vascular pathology and cognitive performance is not modulated by the presence of diabetes.

Our findings suggest that diabetes is positively associated with cognitive impairment, even in non-AD patients, and higher vascular burden, independently of AD biomarkers. Diabetes may contribute to structural and cognitive impairment even in the absence of synergistic effects with AD.

Arzu Çöltekin

Institute of Interactive Technologies, School of Computer Science, FHNW University of Applied Sciences and Arts Northwestern Switzerland

Recent developments in computer- and data science domains, specifically the lean applications that can run on mobile devices such as mobile phones and tablets, new artificial intelligence (machine- and deep learning) paradigms, and a new appreciation of designing systems that enable human-AI collaborations offer exciting possibilities also in the world of dementia prediction, monitoring and possibly interventions.

In this talk, we offer insights from two efforts: One is from cognitive data obtained through a mobile app that is re-envisions cognitive testing augmented reality (AR) and other gamified feature (the Altoida App). Our studies reveal that the app picks up signs of conversion from mild cognitive impairment (MCI) to Alzheimer's Disease (AD), and cognitive improvements observed in cardiac patients after a surgery, both of which demonstrate evidence that continuous monitoring is possible and offers unique insights. The second effort is an exploratory study on the open source ADNI data, where we examine how those who change from MCI to AD differ from those who remain MCI, using visual analytics and machine learning as our main methods. We consolidate the insights from the two efforts into a summary of useful digital protocols and biomarkers for dementia research.

Lisa Hering, Sasha Mukhija, Simon J. Schreiner, Franz Lehner, Jan Loosli, Tobias Weiss, Hans H. Jung*, Nils Briel*

1 University of Zurich, Faculty of Medicine, Zurich, Switzerland

2 University Hospital Zurich, Department of Neurology, Zurich, Switzerland

3 Zurich Neuroscience Center, University of Zurich, Zurich, Switzerland

*contributed equally

Objective: Amyotrophic Lateral Sclerosis (ALS) and Frontotemporal Dementia (FTD) lack established biomarkers for early and accurate diagnosis, relying predominantly on extensive clinical assessments. The phosphorylated-tau to total-tau ratio (pTau:tTau) has been proposed to discriminate ALS and/or FTD (ALS|FTD) from their mimics. We aimed to identify an optimal set of cerebrospinal fluid (CSF), serum, and available clinical biomarkers that can accurately differentiate ALS|FTD from related neurodegenerative diseases – specifically 4R-tauopathy, Alzheimer's disease, Lewy body diseases, and normal pressure hydrocephalus – as well as from controls.

Methods: This retrospective, cross-sectional study analyzed data from 239 patients with suspected neurodegenerative diseases from the Department of Neurology, University Hospital Zurich, who underwent lumbar puncture between 2016-2025. Diagnoses were established based on international consensus criteria or the absence of clinical suspicion of neurodegenerative disease for controls. Extracted data included demographics, disease duration, cognitive performance, CSF biomarker profiles, and relevant comorbidities including head trauma, metabolic, cerebrovascular, autoimmune, and neoplastic disorders. We used XGBoost feature importance to prioritize predictors.

Results: The pTau:tTau ratio was lowest in ALS|FTD (mean=0.109±0.033), followed by 4R-tauopathy (mean=0.129±0.034, p<0.05), Alzheimer's disease (mean=0.162±0.021, p<0.001), and remaining groups (mean=0.134±0.035, p<0.001). In ALS|FTD patients, reduced pTau:tTau ratio correlated with increased CSF-serum albumin ratio and less severe medial temporal lobe atrophy. Compared with using the pTau:tTau ratio alone (AUC=0.75), a multivariate model incorporating the pTau:tTau ratio alongside temporal disease characteristics, immunological and amyloid CSF markers, serum albumin, and radiological white matter lesion burden achieved a considerably more accurate ALS|FTD differentiation (AUC= 0.87).

Conclusions: This study contextualises the pTau:tTau ratio with blood-brain barrier dysfunction and mesiotemporal structural integrity in ALS|FTD. Integrating CSF biomarkers with clinical and demographic features significantly enhances diagnostic precision beyond conventional approaches. Yet, clinical-biological heterogeneity of ALS and FTD variants might drive additional variation. Prospective validation in larger cohorts and earlier disease stages will be essential.

Adrien W. Schmid 1,6, Caroline E. Herron 2, Bea Gil 2, Tara Spires-Jones 3, Jamie Rose 3, Nikita Rudinskiy 4, Bradley T. Hyman 4, Matthew Frosh 5.

1 Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland (former)
2 Conway Institute, University College of Dublin, Ireland
3 UK Dementia Research Institute and Centre for Discovery Brain Sciences at The University of Edinburgh, Edinburgh, Scotland.
4 Harvard Medical School, Mass General Hospital, MA, USA
5 Massachusetts General Hospital and Harvard Medical School, Massachusetts General Institute for Neurodegenerative Disease, MA, USA
6 CSL Behring AG

Aim: The modest therapeutic efficacy reported with some anti-amyloid-beta (Aβ) antibodies emphasizes the strong need for an improvement of next generation disease modifying-therapies for Alzheimer’s disease (AD).

Findings: Using advanced technologies in high resolution mass spectrometry (MS) we were able to shed light on an enigma associated with SDS-PAGE stable Aβ oligomers. Here we provide strong analytical proof of novel Aβ target hotspots present in neurotoxic Aβ oligomers. We show that the early appearance of specific Aβ species, rapidly seed the formation of oligomers, which eventually leads to the formation of stable complexes of larger Aβ entities. We applied quantitative MS approaches to elucidate the composition of these Aβ structures and provide for the first-time quantitative data on the composition of Aβ oligomers, which can now be translated and applied for the analysis of human post-mortem brain tissue or CSF samples from AD patients. Comparative antibody binding studies with aducanumab strongly suggest, that there exist Aβ target “loopholes” which can be exploited for the development of new therapeutic antibodies with binding properties against novel target hotspots present in these oligomers. These findings support a novel, unparalleled approach for targeting early, pathological Aβ species prior to the appearance of large, neurotoxic assemblies. We provide here a first example of a new class of monoclonal antibody (clone: JD1), with a unique binding profile that clearly differentiates to the binding properties of aducanumab or lecanemab.

Conclusion: Our findings highlight the strong need for identifying additional, molecular mechanisms associated with the early events of Aβ peptide aggregation. This in turn will allow the improvement of antibody-specific target engagement of early, neurotoxic Aβ oligomers present in the AD brain or periphery and therefore attenuate or even prevent the pathological progression to severe dementia.

L. Fiorillo 1,2, G. Lombardi 1,3, N. La Porta 4,5, L. Arnaud 1, A. Castelnovo 6,3,7, A. Kaelin-Lang 1,3,8 and S. Galati 1,3

1 Parkinson Disease and Movement Disorder Center, Neurocenter of Southern Switzerland, Ente Ospedaliero Cantonale (luigi.fiorillo@eoc.ch)
2 Institute of Digital Technologies for Personalized Healthcare, University of Applied Sciences and Arts of Southern Switzerland
3 Faculty of Biomedical Sciences, Università Della Svizzera Italiana
4Institute of Information Systems and Networking, University of Applied Sciences and Arts of Southern Switzerland
5 Faculty of Informatics, Università della Svizzera Italiana
6 Sleep Medicine Unit, Neurocenter of Italian Switzerland, Ente Ospedaliero Cantonale
7 Department of Psychiatric Neurophysiology, University Hospital of Psychiatry, University of Bern
8 Department of Neurology, Inselspital, Bern University Hospital

Introduction: Slow-wave activity (SWA) during slow-wave sleep (SWS) facilitates synaptic downscaling, while theta activity during wakefulness reflects synaptic potentiation. Levodopa-induced dyskinesia (LID) in Parkinson’s disease (PD) is linked to impaired synaptic homeostasis, but whether this stems from defective downscaling, impairing wake-related synaptic buildup, remains unclear. This study mainly focuses on analyzing theta activity during wakefulness to investigate the build-up process in resting state.

Methods: We compared actigraphy and high-density EEG in 12 healthy volunteers and three PD cohorts: de novo (n=12), advanced non-dyskinetic (n=13), and dyskinetic (n=11). Participants underwent one-week actigraphy monitoring (i.e., GeneActiv sensors, with sleep metrics extracted via GGIR algorithms) and morning/evening resting-state EEG. Theta power (4–8 Hz) were analyzed using both linear models (LM) and linear mixed-effects models (LMM), adjusted for confounders (e.g., age, gender, LEDD and disease duration). To validate the spatial distribution of significant effects, we applied non-parametric permutation tests combined with Threshold-Free Cluster Enhancement (TFCE).

Results: Dyskinetic patients exhibited reduced actigraphy-estimated sleep efficiency (p<0.001), prolonged sleep latency (p<0.05), and increased wake after sleep onset (p=0.01) versus controls. Theta power was higher in dyskinetic patients across all cortical regions during morning sessions (LM-TFCE-corrected p<0.05), with absent diurnal scaling across prefrontal, temporal, and occipital regions (LMM-TFCE-corrected p<0.05) – contrasting controls and non-dyskinetic groups showing progressive theta accumulation.

Conclusions: These findings demonstrate for the first time a bidirectional synaptic homeostasis failure in LID: impaired SWA-mediated downscaling coexists with saturated morning theta activity, indicating defective wake-related synaptic potentiation. Our results highlight sleep modulation as a therapeutic priority. By enhancing slow-wave sleep, can we restore the synaptic downscaling, and potentially break the cycle of theta saturation and dyskinesia initiation and/or progression?

Kevin Richetin, Aatmika Barve, Shiraz Ferjani, Elodie Ristorcelli, Valentin Zufferey, Paolo Salvioni Chiabotti, Giulia Bommarito, Éric Morel, Alessandra Griffa, Daniel Damian, Mirco Nasuti, Jean-François Demonet, Olivier Rouaud, Gilles Allali

Leenards Memory Center, Lausanne

Since 10 years, the Leenaards Memory Center, the University Memory Clinic of the Lausanne University Hospital has developed CLEMENS, a mixed-use registry designed to support both clinical care and research in neurodegenerative cognitive disorders. To date, more than 8,000 patients have been included in this registry.

While the availability of data varies across individuals, a substantial number of cases include neuropsychological assessments, morphometric brain MRI, CSF biomarker data (Aβ, Tau), and a structured diagnostic classification based on the BNA model (stage, syndrome, etiology). Among them, over 4,500 patients have signed the general research consent, making them eligible for inclusion in the EVINDER project, which aims to enhance clinical stratification by integrating plasma biomarkers.
In collaboration with the CHUV Genomic Biobank, more than 900 plasma samples have been identified, collected during medical care, sometimes on the day of diagnosis, sometimes several years earlier. Although not all samples have been extracted or analyzed yet, preliminary plasma assays for P-Tau217, GFAP, and NfL already reveal significant variations across BNA profiles, highlighting their potential as biological stratification tools. To support clinical implementation, an interactive platform is under development.

This tool will allow CLM physicians to retrospectively explore plasma biomarker levels in relation to BNA classifications drawn from the CLEMENS registry. This direct integration of biological and clinical data paves the way toward precision medicine in neurodegenerative cognitive disorders and future plasma biomarkers dosages.

Paola Milanese 1, Georgios Foustoukos 1, Lila Banterle 1, Laura M.J. Fernandez 1, Matthias Prigge 2, Anita Lüthi 1

1 Department of Fundamental Neurosciences, University of Lausanne, Switzerland
2 Leibniz Institute for Neurobiology, University of Magdeburg, Germany

The onset of Alzheimer’s disease (AD) and of abnormal sleep patterns are signaled by initially minute but ultimately disastrous physiological brain malfunctions. The Locus Coeruleus (LC) belongs to the brain areas that show such malfunctions in the early stages of AD and generates large noradrenaline (NA) signals that are key for wake cognition but also for the sleep architecture in mice. Therefore, insights on how sleep disturbances relate to LC malfunctions could provide sensitive signatures of the earliest stages of AD.

We explore the real-time functionality of the noradrenergic system in APP/PS1, and APP/MAPT knock-in mice with or without neuromelanin expression, using dual fiber photometry in freely behaving animals that express biosensors for free NA levels in the thalamus or the hippocampus (CA1) and for neuronal Ca2+ fluctuations in LC neurons. In combination with polysomnography (EEG/EMG), we record local field potential (LFP) activity in the somatosensory cortex (S1) and in CA1 to assess vigilance states in mice aged 2-18 months to cover early and late stages of AD.

We investigate sleep architecture parameters, such as time spent in each vigilance state and bout durations. We track how Ca2+ activity in the LC and its NA output relate to sleep properties. Preliminary data indicate that depending on the mouse line, REMS bouts tend to shorten with aging, indicators such as LC calcium and EMG activity show unusual activity during some REMS bouts, suggesting that REMS might be dysregulated from the prodromal phase. Ongoing work will associate histological signatures of AD progression with these sleep features.

Decoding the patterns of LC’s dysfunction during the prodromal phase of AD and their consequences on sleep could prove essential to early diagnosis of AD and for interventions.

Anasua Mukhopadhyay 1, Wachara Chanakul 1, Yu-Noel Larpin 1, René Köffel 2, Saurabh Awasthi 3, Anna Protopopova 1, Alessandro Ianiro 1,4, Michael Mayer 1

1 Adolphe Merkle Institute, University of Fribourg, Fribourg, Switzerland.
2 Institute for Chemistry and Biotechnology, Zurich University of Applied Sciences (ZHAW), Wädenswil, Switzerland.
3 Department of Biotechnology, National Institute of Pharmaceutical Education and Research, Raebareli (NIPER-R), Lucknow-226002, Uttar Pradesh, India
4 Department of Chemistry, Katholieke Universiteit Leuven, Leuven, Belgium

Biological nanopores are emerging as powerful sensing tools for single-molecule analysis of nucleic acids, peptides, and proteins. However, the limited size of existing biological nanopores presents a long-standing challenge for transporting large-size, full-length proteins in their natively folded state. Here, we introduce a stable, low-noise, cylindrical transmembrane pore formed by self-assembly of Pneumolysin (PLY) toxin. In situ, assembly of the PLY nanopore occurred in a single step directly on a lipid bilayer upon application of a potential difference of 100 mV. Electrical resistance measurements revealed a diameter of approximately 20 ± 2 (N = 50) nm of membrane-inserted PLY nanopores. This exceptionally large nanopore enabled accurate single-molecule resistive-pulse sensing for the estimation of the volume and shape of folded proteins, ranging in size from FAB (≈ 50 kDa) to tetramers of concanavalin A (4 x 28 kDa = 112 kDa). We used PLY pores to detect differences in the volume of single proteins within a mixture and estimated the size distribution of tau protein (monomer ≈ 45 kDa) and its oligomers from dimers to hexamers in solution. By combining volume analysis with single-particle shape approximation, we uncovered details of Tau oligomerization at nanomolar concentrations. The novel PLY nanopore shows strong promise for advancing the quantification and characterization of heterogeneous amyloid oligomers as biomarkers.

Arben Miftari 1, Fabrizio Pizzagalli 2, Giulia Bommarito 1, Stéphane Armand 3, Frédéric Assal 4, Dimitri Van de Ville 5,6, Alessandra Griffa 1,6, Gilles Allali 1

1 Leenaards Memory Centre, Department of Clinical Neurosciences, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

2 Department of Neurosciences “Rita Levi Montalcini”, University of Turin, Turin, Italy

3 Kinesiology Laboratory, Geneva University Hospitals and University of Geneva, Geneva, Switzerland

4 Division of Neurology, Department of Clinical Neurosciences, Faculty of Medicine, Geneva University Hospitals, University of Geneva, Geneva, Switzerland

5 Neuro-X Institute, École Polytechnique Fédérale De Lausanne, Geneva, Switzerland

6 Medical Image Processing Laboratory, Neuro-X Institute, École Polytechnique Fédérale De Lausanne – EPFL, Geneva, Switzerland Authors affiliation:

Introduction: Idiopathic normal pressure hydrocephalus (iNPH) is the leading cause of reversible dementia in the elderly, is characterized mainly by gait disturbances along with ventricular enlargement, and can be treated with shunting procedure. Neuroradiological features rely on visual assessment, including sulcal characteristics. This study applies automatic sulcal-based morphometry to characterize the iNPH sulcal phenotype and to distinguish iNPH patients who respond or not to the cerebrospinal fluid tap test (CSF-TT), a prognostic test to predict response to shunting.

Methods: The study included 32 patients (mean age: 78.9y / Females: 14) diagnosed with iNPH and 41 healthy controls (HC) (mean age: 74.9y / Females: 30). Quantitative gait assessments were performed before and after the CSF-TT to identify responders (Resp) and non-responders (nResp). Sulcal morphology was evaluated using MRI, focusing on depth, width, length, and surface area. A generalized linear model (GLM) identified the iNPH sulcal phenotype, and a Support Vector Machine (SVM) classifier was applied to distinguish iNPH patients from controls, as well as Resp from nResp.

Results: The GLM analysis identified sulcal depth and opening as the main features characterizing the iNPH phenotype, with respect to HC. Eight core sulci contributed the most, including compressed central, superior frontal, and frontal intraparietal bilateral sulci, as well as flattened left calcarine and posterior lateral fissures. An SVM classifier trained on these features effectively differentiated iNPH patients from HC (AUC: 0.933) but had limited accuracy for Resp vs. nResp (AUC: 0.556).

Discussion/Conclusion: This study identified an iNPH neuroradiological phenotype based on sulcal morphology, emphasizing depth and opening as key markers. An SVM classifier trained on these features accurately distinguished iNPH patients from HCs but was less effective for Resp vs. nResp. Further studies are needed to explore more advanced sulcal landmarks in iNPH.

Alexandra Rhally 1, Johanna Cuony 1, Kevin Battistini 2, Marjolaine Uginet 1, Jonathan Worley 3, Alexander Johannes Lobrinus 3, Patrice H. Lalive 1, 3, 4 , Lukas Sveikata 1

1 Neurology Department, Geneva University Hospitals, Switzerland

2 Neuroradiology Department, Geneva University Hospitals, Switzerland

3 Clinical Pathology Department, Geneva University Hospitals, Switzerland

4 Department of Pathology and Immunology, Faculty of Medicine, University of Geneva, Switzerland

INTRODUCTION: We report the case of a patient admitted for new-onset epileptic seizure, secondary to acute cerebral micro- and macro-hemorrhages, with no identifiable predisposing factors.

CASE REPORT: A 74-year-old woman with a history of progressive memory impairment over several months was hospitalized following a generalized epileptic seizure.

Brain MRI revealed subcortical edema, lobar and subarachnoid macro- and micro-hemorrhages, as well as multifocal cortical siderosis. Cerebrospinal fluid (CSF) analysis showed elevated protein levels (1.32 g/L), lymphocytic pleocytosis (16G/L; 62% lymphocytes), and positive Alzheimer's biomarkers (tau: 1700 ng/L, Aβ42/Aβ40 ratio: 0.05). Comprehensive immuno-infectious screening and peripheral blood flow cytometry were unremarkable.

Inflammatory cerebral amyloid angiopathy (CAA-ri) was suspected, prompting treatment with methylprednisolone (1 g/day for 3 days, followed by a tapering schedule) and lacosamide (100 mg twice daily). The patient showed initial clinical improvement and was transferred to a rehabilitation unit.

Two weeks later, she experienced a decline in consciousness and developed right-sided hemiparesis. Imaging revealed recurrent left frontal hemorrhage with increased microbleeds.

At the family’s request, palliative care was initiated, and the patient passed away eight days later.

Postmortem examination revealed a high-grade intravascular B-cell lymphoma (IVL) with extensive necrotic-hemorrhagic foci, alongside neuropathological evidence of Alzheimer’s disease but no vascular amyloid deposits.

DISCUSSION: While diffuse lobar microbleeds and cortical siderosis are hallmark features of CAA-ri, they can also occur in IVL, making them non-specific findings. IVL can closely mimic the clinico-radiological presentation of microangiopathic disorders due to the presence of malignant cells within small blood vessels, resulting in microvascular dysregulation and disruption of the blood-brain barrier.

CONCLUSION: IVL is a rare and highly morbid condition that remains challenging to diagnose in vivo due to its nonspecific clinical and radiological features. In cases of suspected CAA-ri, a brain biopsy should be considered, as it may be crucial in guiding appropriate treatment strategies.

Albanese, E. 1, Annoni, A. 1, Fiordelli, M. 1, Graf, G. 2, De Benedetti, A. 3, Puhan, MA. 2, Frei, A. 2, Pacifico, D. 1

1 Institute of Public Health, Faculty of Biomedical Sciences, Università della Svizzera Italiana, Lugano, Switzerland
2 Epidemiology Biostatistics and Prevention Institute, University of Zurich, Zurich, Switzerland
3 Dipartimento della sanità e della socialità, Ticino Canton, Switzerland

Background: The accuracy of the Global Burden of Disease (GBD) estimates and projections for dementia in Switzerland remains uncertain. These modelling methods rely on calibrations derived from the Health and Retirement Study Aging, Demographics, and Memory Study (HRS ADAMS) criteria, which have not been validated in European contexts, including Switzerland. We used empirical data from the SwissDEM epidemiological study to quantify discrepancies between observed dementia prevalence and model-based GBD estimates.

Methods: SwissDEM is a population-based, one-phase, cross-sectional study that employed standardized protocols developed by the 10/66 Dementia Research Group. Dementia cases were identified using a locally validated version of the 10/66 diagnostic algorithm. Data were collected up to 2022 in community-dwelling and institutionalized older adults in the cantons of Ticino and Zurich. We compared SwissDEM with GBD estimates using demographic data of the Swiss national census.

Results: Despite participation rates below 50%, socio-demographic characteristics of the SwissDEM sample closely matched national data, suggesting high representativeness and good generalizability. In 2022, the SwissDEM study estimated 163,474 dementia cases among older adults in Switzerland—approximately 20,000 more than the GBD 2019 estimate (142,105; 95% UI: 121,551–164,647). Differences were especially pronounced among the oldest age groups (>85 years).

Conclusion: Empirical assessments in population representative samples provide dementia prevalence estimates that are more accurate than those of indirect modelling approaches like the GBD. The GBD approach likely underestimates actual dementia prevalence, particularly in nursing homes, and by missing milder cases among octogenarians and nonagenarians. Our findings underscore the necessity of robust epidemiological studies to accurately measure and monitor the true dementia burden, ensuring informed public health planning and intervention strategies.

V. Treyer 1, M. Huellner 1, M. Ahnen 2, J. Fischer 2, M. Jehl 2, E. Mikhaylova 2, I. Sacco 2, J. Streb 2

1 Department of Nuclear Medicine, University Hospital Zurich/University of Zurich, Zurich, Switzerland
2 Positrigo AG, Zurich, Switzerland

Introduction: The logistical and financial burdens of conventional full-body PET scanners limit patient access and operational efficiency in neuroimaging. This work demonstrates Real-World and Clinical Trial Data assessment of NeuroLF, a compact, brain-dedicated PET system designed to overcome these barriers. NeuroLF designed to reduce the system's footprint, cost, and infrastructure requirements, enabling flexible, high-throughput deployment in diverse clinical environments.

Methods & Data Sources: We present a consolidated analysis of the NeuroLF's clinical performance, drawing from two principal sources. Firstly, under a clinical research agreement with ProScan Imaging (Naples, FL, USA), the system is being utilized in routine diagnostic practice. Demonstrating its capacity for high-throughput operation, over 150 patient scans were performed within the first three months post-installation. For this conference, we present an interim analysis of a representative cohort of approximately 10 patients who were imaged with amyloid tracers (mainly Vizamyl). Secondly, we present initial data from the ExploreBPET study, an ongoing, explorative, multicenter clinical trial evaluating the NeuroLF system in Leipzig and Zurich. This trial is enrolling patients utilizing a range of tracers including FDG, FET, and agents for tau and amyloid pathology.

Preliminary Results: Data from both the real-world clinical setting in Naples and the formal ExploreBPET trial will be presented. The analysis focuses on diagnostic comparisons, image quality, and operational metrics in a high-volume workflows. The results demonstrate robust system performance and consistent, high-quality imaging across different tracers and international sites.

Conclusion: This multicenter, real-world evidence demonstrates the successful clinical implementation and versatility of the NeuroLF system. The data confirm its suitability for both routine diagnostic use in a high-throughput commercial imaging center and for advanced research applications within a formal, multicenter trial. The NeuroLF system represents a validated and robust solution for expanding patient access to essential neurological PET imaging.

Enfal Nur Celik 1, Maosheng Ye 2, Deng-Feng Zhang 2, Jana Nussbaumer 1, Benjamin F. Comes 1, Junlong Wang 3, Uwe Konietzko 1, Daniel Razansky 4, Christian Behl 5, Axel Rominger 3, Roger M. Nitsch1, Zhude Tu 6, Christoph Hock 1, Ruiqing Ni 1,3

1 Institute for Regenerative Medicine, University of Zurich, Zurich, Switzerland;
2 State Key Laboratory of Genetic Evolution & Animal Models, Kunming, Yunnan, China.
3 Department of Nuclear Medicine, Inselspital University of Bern, Bern, Switzerland
4 Institute for Biomedical Engineering, University of Zurich & ETH Zurich, Zurich, Switzerland
5 Institute of Pathobiochemistry, University Medical Center of the Johannes Gutenberg University, Mainz, Germany
6 Department of Radiology, Washington University School of Medicine, St. Louis, Missouri, USA.

Background: The sigma-1 receptor (σ1R) has emerged as a critical player in Alzheimer's disease (AD), serving as both a potential biomarker and a therapeutic target. This study aimed to investigate alterations in regional σ1R expression in relation to amyloid-beta and tau pathologies, synaptic loss, gliosis, and autophagy in the brains of patients with AD and animal models.

Methods: Immunohistochemical staining for σ1R, LC3b, GABARAP, and LAMP1 was conducted on postmortem tissue slices from frontal, temporal, and entorhinal cortices and the hippocampus from 44 AD patients and 45 nondemented controls (NCs) from the brains of P301L tau, APPPS1 and arcAb amyloidosis mice. Proteomics, single-cell RNA analysis and transcriptomics were performed on the brains of APP/PS1 mice. Relationships between σ1R expression and the levels of amyloid-b, phospho-tau, synaptic density markers, and astrocytic and microglial markers were examined.

Results: Compared with that in the NC group, the level of σ1R expression in the hippocampus and frontal cortex but not in the entorhinal or temporal cortices in the AD group was lower. σ1R levels were negatively correlated with amyloid-beta, phospho-tau, and Braak stages but positively correlated with synaptic density (SV2A and synaptophysin), GFAP and Iba1. The colocalization of σ1R with AT-8 and the autophagy markers LC3b, GABARAP, and LAMP1 was observed in the brains of P301L mice and AD patients. Reduced σ1R expression and SIGMAR1 level was observed in the hippocampus of APP/PS1.

Conclusions: Our findings provide postmortem evidence of reduced σ1R expression in the frontal cortex and hippocampus of AD patients and in APPPS1 mice. Furthermore, σ1R expression is negatively correlated with the levels of amyloid-beta and phospho-tau but positively correlated with the levels of the synaptic markers GFAP and Iba1, highlights its importance and potential as a biomarker and therapeutic target in AD.

Chen Wang 1, Daniel Damian 2, Aurélien Lathuilière 1, Olivier Rouaud 2, Olivier Philippe 3, Oana Gabriela Simionescu 4, Nicolae Sanda 5, Samia Hakimi 6, Tatiana Massardi 6, Rebecca Dreher 7, Enver Lleshi 8, Oscar Daher 9, Yves Kuhne 9, Christian Lovis 10, Gilles Allali 2, Giovanni B. Frisoni 1

1 Memory Centre, Department of Rehabilitation and Geriatrics, Geneva University Hospitals and University of Geneva, Geneva, Switzerland

2 Leenaards Memory Centre, Department of Clinical Neurosciences, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland

3 Centre Mémoire, Réseau Hospitalier Neuchâtelois, Neuchâtel, Switzerland

4 Centre de Consultation de la Mémoire, Hôpital du Valais, Sion, Switzerland

5 Memory Centre, Neurology Department, Fribourg Cantonal Hospital, Fribourg and University of Geneva, Geneva, Switzerland

6 Réseau Fribourgeois de Santé Mentale, Centres de soins hospitaliers, Switzerland

7 Centre Mémoire de la Côte, Réseau Santé La Côte, Rolle, Switzerland

8 Centre Mémoire de l’Est Vaudois, Réseau Santé Haut Léman, Rennaz, Switzerland

9 Centre Mémoire et Gériatrie du Nord Broye, Montagny-près-Yverdon, Switzerland

10 Division of Medical Information Sciences, Geneva University Hospitals and Department of Radiology and Medical Informatics, University of Geneva, Geneva, Switzerland

Background: Biologics for Alzheimer’s disease (AD) demand real-world data (RWD) infrastructures to assess safety, effectiveness, and system-level impact. The ROMENS registry was developed to address fragmented data collection across memory clinics in French-speaking Switzerland.

Objective: To establish a scalable, multi-center registry for RWD collection on AD biologics and supporting clinical and policy decision-making.

Methods: Eight memory clinics participate in ROMENS, under cantonal ethics approvals. A working group of clinicians and researchers defined core variables aligned with ALZ-NET and InRAD, supported by a unified ROMENS Dictionary. An automated pipeline—built with SQL and Python—extracts structured data from Electronic Patient Records (EPRs) and uploads them into REDCap. A real-time dashboard, developed in Python using Dash, enables flexible cohort definition using an unlimited number of combinable filters (via intersection or union logic), supporting both exploratory and hypothesis-driven research.

Results: ROMENS collects a 37-variable Minimum Dataset and over 200 Extended Dataset variables across key domains: demographics, diagnostics, neuropsychology, medication use, and safety events (SAEs, ARIA). It currently includes 8,712 patients across 8 clinics. The dashboard facilitates customized cohort selection with user-defined inclusion/exclusion criteria.

Conclusion: ROMENS provides a replicable model for embedding automated RWD collection into routine dementia care. By integrating with EPRs and enabling flexible data access, it supports longitudinal tracking and real-world evaluation of AD therapies. The geographical scope of the system is presently confined to the Romandie region, and the quality of the data is commensurate with the consistency of the EPR systems that are contributing to it.

Data Deposition: ROMENS data are governed by the consortium. Access is subject to ethical approval and data use agreements."

Miriam Rabl

Psychiatric University Hospital Zurich and University of Zurich

BACKGROUND: Multiple pathophysiological processes have been described in Alzheimer's disease (AD). Their inter-individual variations, complex interrelations, and relevance for clinical manifestation and disease progression remain poorly understood. The Omics-AD study’s main scope is to perform an in-depth multi-modal characterization in people with pre-clinical and early AD to a) better understand pathophysiological changes of AD and b) identify and validate new biomarkers for AD and AD-related outcomes, including cognitive decline and neuropsychiatric symptoms (NPS).

METHOD: Memory-clinic patients and volunteers had been included in a prospective brain aging study at four Swiss memory clinics. Comprehensive clinical and neuropsychological data was collected using validated instruments at baseline and follow-up visits. Paired blood and CSF samples along with structural MRI were obtained at baseline. CSF markers of the core AD pathology have been used to determine the presence of AD pathology. Untargeted omics and targeted molecular analysis will be performed and integrated in multi-modal, multi-omics data analysis.

RESULT: A total of 434 participants were included by December 2024. Of these, 200 (46.1%) were cognitively unimpaired (of which 80 had subjective cognitive decline, SCD) and 234 (53.9%) were cognitively impaired (203 with mild cognitive impairment, MCI, and 31 with mild AD dementia). Half of the participants presented with NPS as measured by the Neuropsychiatric Inventory Questionnaire, while the most common symptoms were irritability (19%) and depression (18%). In total, 45.6% (n = 173) subjects were amyloid positive (25.7% of NC, 23.2% of SCD, 61.7% of MCI, and 75.9% of mild AD dementia).

CONCLUSION: This brain aging, multi-centric, and well characterised cohort unique in Switzerland will allow for single- and multi-omics analysis, to investigate in depth molecular and biological pathway alterations, and their relations with clinical manifestation and progression. It will further serve for the validation of new biomarkers to diagnose and monitor relevant disease processes.

Noëla Rémond-Derbez 1, Fosco Bernasconi 2, Jevita Poteegadoo 2, John Prior 4, Mario Jreige 4, Lada Kohoutova 2, Olaf Blanke 2, Gilles Allali

1 Leenards Memory Center, Departement of Clinical Neurosciences, Vaudois University Hospital Center (CHUV), 1011 Lausanne, Switzerland

2 Laboratory of Cognitive Neuroscience, Neuro-X Institute & Brain Mind Institute, Swiss Federal Institute of Technology (EPFL), 1202 Geneva, Switzerland

3 Department of Clinical Neurosciences, Geneva University Hospital, 1205 Geneva, Switzerland

4 Nuclear Medicine and Molecular imaging, Oncology interdisciplinary Center, Vaudois University Hospital Center (CHUV), 1011 Lausanne, Switzerland.

Introduction: Lewy body disease (DLB) is the second most common cause of neurodegenerative dementia. Early identification of individuals at risk for DLB is crucial to maximize the efficacy of neuroprotective interventions. Hallucinations are early and key symptoms of DLB and previous research, particularly from studies in Parkinson's disease, suggests they may be associated with cognitive decline.

Aim: We aimed at investigating neural correlates of hallucinations in DLB, their characteristics and evolution, as well as their link with cognitive impairment and decline.

Method: A literature search was conducted using Ovid MEDLINE, resulting in the inclusion of 42 studies. All studies investigated minor or complex hallucinations in DLB patients, using fMRI, PET, DAT or EEG imaging.

Results: Hallucinations in DLB are associated with complex large-scale network alterations, involving visual, attentional, memory-related, default mode and salience networks. Most studies reported reduced functional connectivity both within and between networks. However, the role of striatal dopamine uptake, as assessed by DAT imaging was inconsistent. Hallucinations were typically evaluated using brief questionnaires, while some studies provided valuable insights into their characteristics. Direct comparisons between DLB patients with and without hallucinations remain limited, however the available studies offer support for a distinct clinical profile. Several findings suggest a possible association between neural basis of hallucinations and cognition. These results underscore the need for more detailed and longitudinal investigations.

Perspectives: Further studies investigating hallucinations as a new biomarker are needed to foster early and differential diagnosis, as well as improve prognosis of cognitive decline. To address these gaps, we have recently initiated a 5-year longitudinal study including 40 DLB patients, 40 non amnestic MCI and 40 healthy controls. Our study focuses fMRI and EEG imaging, includes diverse neuropsychiatric and neuropsychological assessments, as well as a robotic paradigm to induce presence hallucinations in a controlled setting.

Beatrice Pizzarotti 1, Giulia Bommarito 2, Paolo Salvioni Chiabotti 2, Olivier Rouaud 2, Mahmoud Messerer 3, Gilles Allali 2

1 Neurology Service, Department of Clinical Neurosciences, University Hospital of Lausanne and University of Lausanne, 1011 Lausanne, Switzerland;

2 Leenaards Memory Center, Department of Clinical Neurosciences, University Hospital of Lausanne and University of Lausanne, 1011 Lausanne, Switzerland;

3 Neurosurgery Service, Department of Clinical Neurosciences, University Hospital of Lausanne and University of Lausanne, 1011 Lausanne, Switzerland

Introduction: Anti-amyloid monoclonal antibodies have been recently approved for Alzheimer’s disease (AD). Idiopathic normal pressure hydrocephalus (iNPH) is often associated with AD. Although the presence of any degenerative comorbidity constitutes an exclusion criterion for anti-amyloid therapy, the frequent co-occurrence of these two conditions, coupled with the potential reversibility of iNPH, may warrant a dedicated assessment

Objectives: We aimed to identify the prevalence and characteristics of iNPH patients with AD-positive biomarkers potentially eligible for anti-amyloid treatments.

Methods: We retrospectively identified iNPH patients according to international criteria. Demographic, clinical, CSF and MRI data were collected. AD positive status was defined as A+T+ and exclusion criteria for anti-amyloid drugs were evaluated.

Results: Among 81 iNPH patients (78.9 y.o (±5.1)) referred from Neurosurgery Service to Leenaards Memory Center of the Lausanne University Hospital (Switzerland) from May 2019 to December 2023, 11 (13.6%) of iNPH patients presented AD positive biomarkers and 7 (8.6%) were eligible for anti-amyloid drugs. Among ineligible patients, 2 (18%) had more than 4 microbleeds at brain MRI, 1 (9%) for CDR>1 and 1 (9%) had active anticoagulant treatment. Clinical and radiological features were similar between iNPH patients eligible or ineligible to treatment.

Discussion: Among iNPH patients with AD comorbidity, the majority could be eligible to AAT according to Clarity AD inclusion criteria. Apart from iNPH radiologic signs, the main exclusion criteria were based on imaging findings, as in previous studies; we found similar percentage (around 9%) of patients excluded due to dementia severity and/or anticoagulant treatment. Our study did not find any clinical or radiological features that correlate with the presence of AD biomarkers in iNPH.

Conclusions: While only a minority of iNPH patients qualify for anti-amyloid drugs, it is essential to develop a clinical strategy to stratify the therapeutical approaches for these patients.

Audrey Gabelle 1, Stephen Macfarlane 2, Timo Grimmer 3, Luca Villa 4, Elizabeth Gordon 4, Thomas Jubault 4, Nicolas Guizard 4, Kun Jin 5, William R. Chezem 5, Christopher U. Missling 5, Marwan N. Sabbagh 6

1 University of Montpellier, Montpellier University Hospital
2 The Dementia Centre, HammondCare, Melbourne, Victoria
3 Technical University of Munich, School of Medicine and Health, Klinikum rechts der Isar, Munich
4 QYNAPSE SAS, Paris
5 Anavex Life Sciences, New York, New York
6 Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona

The ANAVEX2-73-AD-004 phase IIb/III trial was a randomized, double-blind, placebo-controlled study lasting 48 weeks. It evaluated the efficacy and safety of blarcamesine, an oral sigma-1 receptor agonist aimed at restoring autophagy, in patients with early-stage Alzheimer's disease (AD).

A total of 508 patients were randomized to receive either blarcamesine (30 mg or 50 mg; n = 338) or placebo (n = 170) once daily by mouth.
The primary endpoints were changes in ADAS-Cog13 and ADCS-ADL scores. Secondary endpoints included CDR-SB score and biomarkers such as the plasma Aβ42/40 ratio and brain volume measured by MRI. Analyses were conducted using mixed models for repeated measures, Welch’s t-test, and general linear models.
Among the 462 participants in the intent-to-treat (ITT) population (average age: 73.7 years; 48.7% women), 73.1% completed the trial. The blarcamesine group showed statistically significant improvements on the ADAS-Cog13 (−2.027; P = 0.008) and CDR-SB (−0.483; P = 0.010). The ADCS-ADL showed a positive trend, though it was not statistically significant. Treatment positively impacted the plasma Aβ42/40 ratio and reduced brain atrophy. A genetic analysis revealed an even greater benefit (49.8%) in patients with the wild-type SIGMAR1 gene. The safety profile was favorable.

Blarcamesine may represent a promising new oral therapeutic option for early-stage AD, either as a complement to or an alternative to current anti-amyloid treatments.

Tom Vlaar, Bernadette Meyer, Lars van der Heide and Ioana M. ILIE

University of Amsterdam, the Netherlands

The proteins of the Bcl-2 family play crucial roles in regulating apoptosis. It is divided into pro-survival and pro-apoptotic proteins that determine cellular fate. In particular, Bax is a crucial executor of apoptosis as its activation initiates the apoptotic phenotype. Hence, targeting this protein represents an attractive therapeutic approach, which can aid in regulating apoptotic signalling and potentially contribute to the development of novel therapies against cancer and neurodegenerative diseases.

Here, we introduce a digital paradigm, which relies on rational design and computer simulations to develop and validate peptide-based agents that bind to Bax, thereby inhibiting its apoptotic properties. The peptides are rationally designed and optimized to bind to Bax starting from the crystal structures of affimers in a complex with Bcl-2 proteins. Next, molecular dynamics (MD) simulations are employed to probe the stability of the Bax–peptide complexes and to estimate the binding free energies. The results show that the designed peptides bind with high affinity to Bax. Two of the designed peptides bind in the canonical hydrophobic groove (BH1 domain) of Bax and one peptide binds to the outside of the BH3 domain (α2-helix). Notably, the peptides restrict the flexibility of the α1–α2 loop, modulating the trigger bottom site associated with toxicity.

All in all, the results highlight the potential of these peptides as valuable tools for further exploration in modulating apoptotic pathways and set the structural foundation for a machine learning powered engine for peptide design.

Xue Wang, Rebecca Sternke-Hoffmann, Pereira Curia Guadalupe, Xun Sun, Jinghui Luo

Department of Biology and Chemistry, Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland

Parkinson disease (PD) is one of the most prevalent neurodegenerative diseases, characterized by the abnormal alpha-synuclein (α-syn) aggregation. While amyloid protein cascade hypothesis remains a main explanation for PD pathology, no chemical drugs targeting α-syn oligomers are success in clinic, largely due to their transient and heterogeneous nature. Recent studies have identified a distinct intermediate state in α-syn aggregation: liquid-like condensates at nanoscale known as nanoclusters, which are more dynamic and less cytotoxic than traditional oligomers. In this study, we demonstrate that metformin and its analogues can form larger α-syn nanoclusters, then more easily directing their fate toward autophagic degradation rather than fibril formation. Interestingly, we also show that in the presence of metformin, α-syn could rescue the role of LC3 in stabilizing membranes during autophagosome formation. These findings offer novel insight into the aggregation landscape of α-syn, suggesting that nanoclusters represent a promising therapeutic target. Moreover, our results highlight the potential of biguanide-based compounds as dual-function agents—modulating α-syn phase behavior while enhancing its clearance via autophagy—for future PD treatment strategies.

Jaspreet Singh Saini 1, Sophie Mutel 1, Moira Marizzoni 2, Arthur Paquis 1, Thomas Rio Frio 3, Valentina Garibotto 4, Giovanni Frisoni 1

1 Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland
1 Geneva Memory Centre, Department of Rehabilitation and Geriatrics, Geneva University Hospitals, Geneva, Switzerland
2 Laboratory of Neuroimaging and Alzheimer's Epidemiology, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy
3 Department of Medical Genetics, Hôpitaux Universitaires de Genève
4 Laboratory of Neuroimaging and Innovative Molecular Tracers (NIMTlab), Geneva University Neurocentre and Faculty of Medicine, University of Geneva, Geneva, Switzerland

Alzheimer’s disease (AD) is the leading cause of dementia, affecting 55.2 million people globally. The highest regional burdens were reported in the Western Pacific and Europe, according to the World Health Organization, 2019. AD is characterized by cognitive decline hallmark neuropathological features, including amyloid-beta plaques and phosphorylated tau tangles (pTau), and genetic risk factors such as carrying the Apolipoprotein E (APOE) ε4 allele. Recent studies have uncovered pathophysiological links between the gut microbiome and brain amyloidosis, particularly via bacterial lipopolysaccharides and short-chain fatty acids, highlighting the relevance of gut microbiota in AD pathology.

In this study, we investigate associations between gut microbiota composition and key Alzheimer’s disease features—amyloid, pTau, and APOE genotype—alongside clinical assessments of cognitive status. Comprehensive microbial profiles were generated from stool-derived DNA using advanced shotgun metagenomic sequencing from 551 participants at the Memory Center of Geneva University Hospital. The cohort included individuals spanning the cognitive spectrum, from unimpaired to cognitively impaired.

Microbial abundance and prevalence profiles were obtained using read-based taxonomic profilers mOTUs and Sylph. These profiles were incorporated into a Bayesian probabilistic modelling to examine associations with AD pathology and related clinical indicators. This study hypothesizes that bacterial taxa significantly reduced in (1) cognitively impaired individuals, (2) those with amyloid and pTau positivity, and (3) carriers of the APOE ε4 allele (e.g., e2/ε3 vs e3/ε4 genotypes) may serve as promising candidates for targeted probiotic interventions.

Our findings revealed several species within the genera Dysosmobacter, Blautia, Bacteroidetes, Lawsonibacter, and Faecalibacterium emerging as top candidates. Results also suggest the presence of potentially novel species within these genera, warranting further phylogenomic analysis to resolve species- and strain-level diversity. Future pan-genome analyses will aim to elucidate the metabolic potential of these probiotic candidates. Altogether, this study lays the groundwork for microbiome-based interventions in Alzheimer’s disease, advancing the path toward precision probiotic treatment.

Inês Dias 1,2,3, Christian R. Baumann 1,3,4, Daniela Noain 1,3,4

1 Department of Neurology, University Hospital Zurich (USZ), Switzerland
2 Department of Health Sciences and Technology (D-HEST), ETH Zurich, Switzerland
3 Neuroscience Center Zurich (ZNZ), Switzerland
4 Center of Competence Sleep and Health, UZH, Switzerland

Slow-wave sleep (SWS) alterations are critical features of neurodegenerative diseases such as Alzheimer’s (AD) and Parkinson’s (PD) disease, with disrupted slow-wave activity (SWA) exacerbating pathological processes. We developed and validated mouse closed-loop auditory stimulation (mCLAS) as a novel, non-invasive method to selectively target SWS and modulate SWA dynamics, providing a tool to address disease-specific impairments in neurodegeneration.

Our findings reveal that precise tracking of a 2 Hz component of slow waves, combined with phase-specific targeting, optimizes mCLAS efficacy. A 30° up-phase target produced significant 15–30% increases in SWA in AD (Tg2576) mice, whereas a 40° target enhanced SWA by 30–35% in PD (M83) mice. These parameters were tailored to maximize stimulation precision and enhance SWA across models, resulting in an adaptive modulation of pathological slow-wave characteristics. In AD mice, mCLAS reduced type I (global) slow-wave amplitude and slope while enhancing type II (local) slow-wave recruitment and synchrony, supporting sleep continuity and local network connectivity, while counteracting pathological sleep fragmentation. Contrastingly, mCLAS increased type I slow-wave incidence in PD mice, restoring pathologically impaired global synchrony and rebalancing 24-hour slow-wave dynamics while reducing type II prevalence toward healthy patterns. Overall, mCLAS selectively prompted the acute modulation of distinctive slow-wave types, boosting sleep homeostatic regulation mechanisms in both disease models. Moreover, preliminary results in the AD model suggest that mCLAS prevents behavioural deficits, reduces pathological protein deposition, and mitigates degeneration, highlighting the neuroprotective effect of SWS modulation via mCLAS.

These findings provide strong support for mCLAS as a translational tool for addressing sleep disturbances and underlying pathology in neurodegeneration. Hence, mCLAS may become the future cornerstone of SWA-based therapies for neurodegenerative diseases.

Aline Broeglin, Anne Eckert, Amandine Grimm

Research Cluster Molecular & Cognitive Neuroscience, Department of Biomedicine, University of Basel, Basel, Switzerland
Neurobiology Laboratory for Brain Aging and Mental Health, Psychiatric University Clinics, Basel, Switzerland

Tauopathies are neurodegenerative diseases characterized by an abnormal accumulation of the protein tau in cells, leading to significant neuronal death and cognitive impairment. An important feature of these tau-related diseases is the deficits in bioenergetic functions due to the impairment of mitochondrial integrity. Mitochondria are involved in many cellular processes, including adenosine triphosphate (ATP) synthesis. They are critical in the brain, which consumes much of the body's energy. It has been shown that mitochondrial dysfunctions precede cognitive deficits in these diseases. Therefore, mitochondria are considered a therapeutic target to prevent the development of brain disorders.
A new therapeutic approach is to consider mitochondria as the treatment itself with mitochondrial transplantation. This involves isolating healthy mitochondria and transplanting them into damaged organs or cells using various methods. Our project aims to investigate the therapeutic potential of mitochondrial transplantation in a cellular model of tauopathies.

We used the neuroblastoma cell line SH-SY5Y as recipient cells for the transplantation and the astrocytic cell line A172 as a donor of healthy mitochondria. Cellular and molecular biology tools, as well as fluorescence microscopy experiments, were used during the investigations.

We show that the functional integrity of isolated mitochondria is maintained after the isolation process, and that isolated mitochondria can enter the recipient cells. An optimal treatment concentration was determined, and an increase in cellular viability and bioenergetic functions of the healthy and tauopathy cellular model was observed. Moreover, an improvement in the neurite outgrowth has been detected in both cellular models.

Mitochondrial transplantation offers a new therapeutic approach in tauopathies by considering mitochondria as the treatment itself. Further investigations are needed to explore the cellular mechanisms underlying this phenomenon.

George Kurian 1, Paolo Salvioni Chiabotti 2, Caroline Hall 2, Noemie Lejay 2, Jérémy Perriraz 1, Olivier Rouaud 2, Adrien Depeursinge 1,3, Marie Nicod Lalonde 1, Niklaus Schaefer 1, Gilles Allali 2, John O. Prior 1, Mario Jreige 1

1 Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.
2 Leenaards Memory Center, Department of Clinical Neurosciences, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.
3 Institute of Informatics, School of Management, HES-SO Valais-Wallis University of Applied Sciences and Arts Western Switzerland, Sierre, Switzerland.

Background: Quantification of I-123-ioflupane uptake using modern SPECT/CT improves diagnostic accuracy for disorders affecting the nigrostriatal pathway. This study assessed whether absolute and relative standardized uptake values (SUV) could distinguish dementia with Lewy bodies (DLB) from non-DLB in patients with suspected DLB and explored associations with core clinical features, including visual hallucinations (VH) and REM sleep behavior disorder (RBD).

Methods: Seventy-four patients (mean age 71.5 ± 9.26 years; 39% female) were retrospectively included. All underwent I-123-ioflupane DAT-SPECT with both iterative (Flash 3D) and quantitative reconstructions. SUVmax, SUVmean, and relative SUV (rSUV) values were extracted for the caudate, putamen, and striatum. Uptake metrics were compared between DLB and non-DLB groups and analyzed in relation to clinical features, including visual hallucinations (VH) and REM sleep behavior disorder (RBD).

Results: Visual assessment classified 28/74 scans (38%) as abnormal (sensitivity 90%, specificity 80%, AUC 0.846). Quantitative SUVmax in the striatum and putamen showed the highest diagnostic performance (AUC up to 0.83). Striatal SUVmax remained an independent predictor of DLB in multivariable analysis (OR = 0.58, p = 0.003). Patients with VH had significantly lower striatal SUVmax than those without (p = 0.004), with an optimal cutoff of ≤6.0g/mL (AUC = 0.70, sensitivity 64.3%, specificity 80.4%). No significant differences were observed for RBD.

Conclusions: Quantitative I-123-ioflupane uptake assessment using SUV measures offers clinically relevant diagnostic value for differentiating DLB from other neurodegenerative diseases. It also helps in identifying patients with visual hallucinations, supporting the broader integration of SUV-based dopaminergic imaging into clinical workflows.

Hadrien M. Lalive 1,2, Alice Accorroni 1, Dylan Donayre 1, Umberto Nencha 1, Chen Wang 1,2, Augusto J Mendes 1,2,3, Federica Ribaldi 1,2,3, Giovanni B Frisoni 1,2, Aurélien Lathuilière 1,2

1 Geneva Memory Center, Department of Rehabilitation and Geriatrics, Geneva University Hospitals, Geneva, Switzerland.

2 Faculty of Medicine, University of Geneva, Switzerland.

3 Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland.

OBJECTIVE: To longitudinally evaluate pain, anxiety and adverse events (AEs) associated with lumbar puncture (LP) in older adults, and to investigate their association with patients’ perceptions, clinical characteristics, and Alzheimer’s disease (AD) cerebrospinal fluid (CSF) biomarker levels.

BACKGROUND: LP is a common procedure in memory clinics to assess AD biomarkers and is generally considered safe and well-tolerated due to a low complication rate. However, patient discomfort before, during, and after LP may be underrecognized and underreported.

METHODS: We prospectively recruited 109 consecutive Memory Clinic patients (mean age 70.6 ± 8.7 years; 51% female) undergoing their first LP for cognitive assessment at the Geneva University Hospitals Memory Center (January 2024 – January 2025). Pain, anxiety, and AEs were assessed using structured surveys and the Beck Anxiety Inventory at four time points: before LP, immediately after, one hour post-procedure, and during a follow-up call 24-72 hours later. Categorical variables were analyzed with Cochrane’s Q test, non-parametric continuous variables were analyzed using Wilcoxon’s or Friedman’s tests, and multivariate regression was performed to examine associations between patient characteristics, CSF biomarkers, anxiety, AEs, and pain.

RESULTS: Most participants reported minimal anxiety throughout the procedure. While mild headaches, back pain, and fatigue were common, severe AE occurred in only one participant (postdural puncture headache). AEs were not associated with cognitive status or CSF biomarkers, but were more frequent in females, suggesting a potential role for anatomical differences. Pain during LP was lower than patients expected, averaging less than 3/10, and post-procedural pain was low, strongly linked to the presence of adverse effects. Older patients reported less pain than younger patients. Patient satisfaction was high, with 80% of patients willing to undergo LP again.

CONCLUSIONS:
LP is well tolerated in older adults regardless of baseline anxiety, cognitive status, or AD CSF biomarkers. These findings support the use of LP for AD biomarker assessment in memory clinics.

Alexa Holfelder 1,2, Esther Brill 1,2, Michael Falkner 3, Christine Krebs 1, Rachid Guerchouche 4,5, Minh Tran-Duc 4,5, Stefan Klöppel 1, Jacob Lahr 1

1 University Psychiatric Services Bern, University Hospital of Old Age Psychiatry and Psychotherapy, Bern, Switzerland

2 Graduate School for Health Sciences, University of Bern, Bern, Switzerland

3 ARTORG Center for Biomedical Engineering Research, University of Bern, Bern, Switzerland

4 Cognition Behavior Technology (CoBTeK) Lab, FRIS-Université Côte d’Azur, Nice, France

5 Institut National de Recherche en Informatique et en Automatique (INRIA), Valbonne, France

Background: With an aging population, cognitive decline and dementia present significant healthcare challenges. Accurate diagnosis and early intervention are crucial, yet access to specialized dementia screening clinics remains limited - particularly in rural and remote areas. Telemedicine offers a promising avenue to expand access to early cognitive assessments. This pilot study explores the subjective experiences of older adults undergoing remote cognitive testing, focusing on how they perceive and interact with digital technology in the context of person-centered dementia care. It also examines clinicians’ perspectives on the platform’s usability and the quality of remote interactions.

Methods: A secure videoconferencing platform was developed to administer a range of standardised neuropsychological assessments remotely. The study captures the perspectives of older adults completing the assessments and clinicians delivering them. Usability and acceptability were evaluated using 10-point Likert scale questionnaires, supported by structured qualitative interviews with clinical staff.

Results: Preliminary findings from 10 participants indicate high satisfaction with the remote assessment experience, including ease of use (m=8.33), video quality (m=8.39), and audio clarity (m=8.19). Clinicians, however, noted lower satisfaction due to technical setup issues (m=4.13) and challenges in communication with older individuals through the platform (m=5.5). Qualitative insights are currently being collected and will undergo thematic analysis upon completion of data collection in July 2025.

Conclusion: These early findings suggest that telemedicine is a promising and acceptable approach for cognitive assessment among elderly, particularly in areas with limited in-person services. While participants reported positive experiences, clinician feedback highlights areas for improvement. Enhancing platform usability and interaction quality will be critical to supporting the integration of remote cognitive assessments into routine care and advancing equitable access to dementia services across diverse populations.

Greta Rizzi, Anna Messina, Rebecca Amati, Anna Maria Annoni, Emiliano Albanese & Maddalena Fiordelli

Institute of Public Health, Faculty of Biomedical Sciences, Università della Svizzera italiana

Background and Objectives: In Ticino, the Italian-speaking region of Switzerland, most people with dementia live at home, with informal and formal caregivers playing a critical role in daily care. Metaphorical language is often used to articulate the complex nature of care, offering valuable insights into the lived experiences of caregivers. This study explores how caregivers in Ticino spontaneously use metaphors in their narratives to conceptualize dementia and their role.

Research Design and Methods: We conducted six focus groups with caregivers of people with dementia (N=6 formal, N=13 informal) as part of the Swiss adaptation of the World Health Organization’s iSupport for dementia program. A qualitative content analysis was performed focusing on metaphors’ semantic and linguistic complexity and contextual meanings. Metaphors were categorized into three themes: (1) about dementia, (2) about caregiving, and (3) about the caregiver-person with dementia relationship.

Results: Caregivers frequently framed people with dementia as “children” or employed natural elements to depict their burden. Metaphors revealed different caregiving perspectives: some emphasized an individualistic approach aimed at independence while "patching" problems or "drowning" in challenges; others reflected a community-cantered moral dedication view, like the metaphor of a “mission”; others captured the complex dynamics of power and dependency in the caregiver-person with dementia relationship.

Discussion and Implications: Unlike commonly cited metaphors in dementia literature, metaphors used by caregivers in Ticino were deeply rooted in local context. Overall, metaphors provided a multifaced view of caregiving, illuminating tensions between self-care and care for others. Some metaphors foster an empathetic approach to care, while others (i.e., infantilization) risk reinforcing stigma or idealizing self-sacrifice. This study highlights the importance of reflecting on caregiving language in training and support programs to improve mutual understanding, challenge stigma, and promote person-cantered approaches.

Giuliana Crippa 1,2, Severin Pinilla 1, Regula Blaser 3, Kathy Haas 3, Franziska Zúñiga 4, Stefan Klöppel 1

1 University Hospital of Old Age Psychiatry and Psychotherapy, Bern, Switzerland

2 Graduate School for Health Sciences, University of Bern, Bern, Switzerland

3 Institute on Ageing, University of Applied Sciences, Bern, Switzerland

4 Nursing Science (INS), Department Public Health, University of Basel, Basel, Switzerland

Background: Dementia imposes a significant burden on healthcare resources, driven by the need for ongoing, intensive support as the condition advances. This burden is further compounded by additional challenges such as behavioral and psychological symptoms of dementia (BPSD) and delirium, which often contribute to more complex and resource-intensive care trajectories. Effective dementia care requires timely recognition and tailored management of symptoms such as BPSD and delirium. By focusing on tertiary prevention, it is possible to prevent escalation, enhance quality of life, and reduce strain on healthcare resources. Despite the critical role of effective dementia care, the absence of clear indicators to assess its quality hinders efforts to evaluate outcomes and drive systematic improvements across care settings.

Methods: Clinicians, advanced practice nurses, dementia care specialists and researchers practicing in either the canton of Bern or Zurich will be recruited for this Delphi study. The development process is guided by the National Dementia Platform, the German S3-Guidelines on Dementia and the Swiss Recommendations for the Diagnosis and Treatment of BPSD. The newly developed indicators will undergo validation through theoretical examinations and observation processes of individuals engaged in dementia care.

Results: In total, 20 experts will be recruited to participate in the Delphi study, with the objective of establishing indicators for 18 pre-existing criteria derived from the above mentioned sources. Preliminary results are anticipated by the fall of 2025.

Conclusion: Establishing indicators of competent dementia care is critical for improving care quality and system efficiency. These indicators enable the healthcare system to track performance, identify areas for improvement, and implement targeted interventions that align with best practices. By operationalizing complex care needs into measurable outcomes, driving continuous quality enhancement and ensuring person-centered care for people living with dementia.

Stefanie Steininger

Clienia Klinik Schlössli, Oetwil am See

Importance: With the increasing prevalence of dementia syndromes and polymorbidity, palliative care (both early and end-of-life care) is becoming an increasingly important issue for the quality of life and treatment of patients with dementia. Therefore, geriatric psychiatry faces new challenges in implementing palliative care to optimise treatment options for patients with dementia, especially in an inpatient setting.

Aim: How can palliative care be integrated into inpatient geriatric psychiatry, which medications are feasible and where are the limitations?

Design, setting and participants: A total of 211 patients were admitted to a neuropsychiatric ward in 2024, with a focus on dementia syndromes, BPSD and delirium. An early palliative care plan was developed in 25%. Four patients received end-of-life palliative care.

Results: Intensive family care, spiritual care, development and implementation of an early palliative care plan, and close medical and nursing care, including bolus medication administration via subcutaneous catheter at the end of life, were defined as feasible. With regard to medication, the recommendations of Palliativmedizin Ostschweiz were considered feasible for the most part.
Patients who required a specialised palliative care setting for pain control, dyspnoea or i.v. sedation could not be managed.
Overall, the introduction of palliative care was appreciated by patients and their carers.

Conclusion: Palliative care, knowledge and management are essential in geriatric psychiatry. Early advance care planning and end-of-life care may be feasible in an inpatient setting, with limitations. However, there are serious limitations: Palliative care in geriatric psychiatry can never be an alternative to specialist inpatient palliative care. But it can be a valuable addition. Palliative care in psychiatry is a taboo that is long overdue. It represents an evolution in geriatric psychiatry and should be given greater consideration in future care planning.