Abstract The slow progression of Alzheimer’s disease (AD) poses a challenge for the quantification of early disease-driven cognitive decline. Here, we show that frequently administered remote and unsupervised digital cognitive assessments can detect differences in cognitive decline within 30 weeks in early AD. The sample comprised 202 individuals (52–85 years old) recruited from longitudinal observational studies, who were cognitively unimpaired (CU, n = 152) or had a diagnosis of mild cognitive impairment (MCI, n = 50). Participants self-administered remote tasks testing memory precision for objects and scenes, associative memory, and familiarity-dependent memory. The MCI group showed greater decline than the CU group in the familiarity-dependent task, while stratifying the MCI group by beta-amyloid (Aβ) status (n = 21 Aβ−; n = 24 Aβ+) revealed greater change in memory precision for objects and familiarity-dependent memory in the MCI Aβ+ group. A 30-week change in the remote familiarity-dependent task was correlated with a multi-year change in annual in-person neuropsychological assessments. In conclusion, frequent remote cognitive testing is a promising tool to feasibly capture and monitor subtle and short-term cognitive decline.

BackgroundAnticholinergic side effects of pharmacological treatment are a risk factor for cognitive decline in older people. Here, we aimed to assess the effect of anticholinergic burden of treatment on longitudinal rates of cognitive change and atrophy in functionally related brain regions in people from the Alzheimer’s disease (AD) spectrum.MethodsWe determined associations of anticholinergic burden of pharmacological treatment with rates of global cognition, episodic memory and executive function decline as well as basal forebrain and hippocampus atrophy in participants of the memory clinic based DELCODE cohort, spanning the range from cognitively normal through subjective cognitive decline, mild cognitive impairment and AD dementia. We had 794 cases with neuropsychological outcomes, and a subset of 703 cases with MRI outcomes. Effects were assessed using mixed effect models in a Bayesian framework using prior-insensitive cross-validated Bayes factors (CV-BF) and parameter estimates.ResultsWe found moderate evidence for an association of anticholinergic burden with baseline levels of cognitive impairment for the PACC5 as a global cognitive function score (CV-BF = 9.0) with more impairments with higher burden, but not with basal forebrain and hippocampus volumes, and weak evidence for an association of anticholinergic burden with longitudinal rates of change in the trail-making test B as an executive function score (CV-BF = 2.5), but not for other cognitive scores and not for brain volumes.ConclusionIn the presence of prodromal or manifest AD, in a memory clinic-based cohort anticholinergic burden had only a modest effect on cognitive decline and no effect on atrophy in brain regions that are related to the cholinergic system.

Mild cognitive impairment (MCI), a prodromal stage of Alzheimer's disease (AD), remains undiagnosed in > 90% of individuals, delaying access to timely evaluation and interventions. Self-administered digital cognitive assessments (SA-DCAs) offer scalable approaches for early detection, yet their real-world validation and clinical readiness remain uncertain. We developed a use-case-specific framework to evaluate SA-DCAs intended for community and primary-care MCI screening and applied it to a comprehensive scoping review of published evidence (2012-2025). Among 79 identified SA-DCAs, only four tools met predefined framework criteria across nine eligible studies. Common limitations included restricted population representativeness, inconsistent diagnostic performance reporting, limited biomarker anchoring, and reliance on prefiltered cohorts. Overall, the current evidence base is methodologically heterogeneous and incomplete for clinical deployment. The proposed framework characterizes requirements including anchoring strength, prevalence-adjusted performance reporting, and representative sampling establishing a foundation for advancing robust real-world evidence needed to translate SA-DCAs from research to clinical practice.

Brain clearance pathways via the interstitial fluid and cerebrospinal fluid are essential for maintaining homeostasis and removing metabolic waste. Disruptions in these processes are linked to neurodegenerative diseases, particularly Alzheimer’s Disease (AD), where an imbalance between amyloid β (Aβ) production and clearance leads to cerebral amyloidosis. Although AD research continues to rely heavily on murine models, including APPswe/PS1dE9 mice, it remains unclear to what extent these models replicate the deficits in brain clearance observed in human AD. We examined AD pathology in transgenic APPswe/PS1dE9 (tg) mice and wildtype (wt) littermates at six, nine and twelve months using immunohistochemistry for Aβ plaques (6E10+) and activated microglia (Iba1+). To assess integrative parameters contributing to brain clearance, we applied two complementary approaches. First, we performed daytime-dependent intrahippocampal injections of a fluorescent tracer in six-month-old tg and wt mice, followed by histological analysis of tracer dispersion 1 h and 3 h after injection. Second, we present a novel, longitudinal and non-invasive approach utilizing washout kinetics of intravenously administered gadolinium-based contrast agents (GBCA) as an integrative proxy to assess brain clearance dynamics. In this novel approach DOTAREM® was administered intravenously and imaging was done with anesthetized tg and wt mice at nine and twelve months of age. Integrity of the blood-brain-barrier was further evaluated via western blot analysis of occludin protein expression. Despite tg mice displayed characteristic AD-like phenotype, with significant, age-dependent Aβ plaque deposition and neuroinflammation no significant differences in the integrative GBCA washout kinetics were detected between wt and tg mice, suggesting that no major deficits in the combined clearance pathways are present. However, contrast-agent uptake showed a general age-dependent increase consistent with aging-related changes known from human studies. Our findings indicate that the APPswe/PS1dE9 mouse model does not exhibit evidence of major integrative brain clearance impairment, despite this being a hallmark of sporadic AD in humans. These results suggest limitations of this model for studying clearance-related mechanisms of late-onset AD, which should be acknowledged when planning future studies using this murine model.

BackgroundLow socioeconomic status is associated with a higher risk for a number of health conditions, including cognitive impairment.ObjectiveWhile the association with economic indicators has been well researched, aim of this paper was to investigate specifically the association between non-economic social deprivation and cognitive functioning later in life.MethodsParticipants without objective cognitive impairment at baseline (n = 91 controls, n = 106 with subjective cognitive decline) of the multicenter DZNE-Longitudinal Cognitive Impairment and Dementia Study (DELCODE) filled out a posthoc survey on social deprivation. Factor analysis identified four non-economic domains of social deprivation (Residential area conditions, Accessibility, Neighborhood support, Household distress). Cognition was assessed via the CERAD neuropsychological battery. Analyses were conducted using linear regression models as well as maximum-likelihood mixed-effects models adjusted for age, gender, years of education, marital status, depression, BMI, heart disease, hypercholesterolemia, hypertension, and diabetes.ResultsAt the time of the survey, cognitive functioning was significantly lower among participants with high deprivation (b = -0.21, p = 0.037). These participants also had a faster rate of cognitive decline (b = -0.07, p = 0.006). Analyses of the deprivation sub-scores revealed statistically significant effects only for Accessibility (b = -0.31, p = 0.001; rate of cognitive decline b = -0.07, p = 0.004).ConclusionsOur findings suggest that non-economic social deprivation, especially in terms of access to a grocery store, pharmacy, postal office, and bus stop, might be relevant for maintaining cognitive functioning in old age. Further research should explore potential mechanisms of how these environmental conditions affect brain aging and neurodegenerative processes.

Alzheimer’s disease (AD) is characterised by the accumulation of β-amyloid (Aβ) and tau proteins, resulting in neurodegeneration and cognitive decline. Although Aβ and tau disrupt synaptic function, the association linking these molecular pathologies to network-level dysfunction and memory impairment remains poorly understood. Here, we investigated the effects of Aβ and tau pathology (CSF Aβ42/40 ratio and tau phosphorylated at position 181, p-tau-181, respectively) on effective connectivity related to memory encoding, which may provide a link between synaptic pathology and cognitive outcomes. Functional magnetic resonance imaging (fMRI) during visual memory encoding was acquired from 205 participants in the multicentric DZNE Longitudinal Cognitive Impairment and Dementia Study (DELCODE) across the AD spectrum. Effective connectivity was assessed using Dynamic Causal Modelling (DCM) of task-fMRI data, focusing on the parahippocampal place area (PPA), hippocampus (HC), and precuneus (PCU)—regions central to memory encoding. Disruptions in connectivity between temporal and parietal lobes were associated with both memory impairment and indices of AD pathology. Specifically, reduced positive effective connectivity from the PCU to the PPA and from the HC to the PCU were linked to higher p-tau-181 levels, with an amplification effect observed in the presence of amyloid accumulation for the latter connectivity. The disruption from the PCU to the PPA was found to be associated with decreased memory performance. Together, these findings indicate that temporo-parietal connectivity is associated with both AD molecular pathology and, for a subset of connections, with memory performance.

Abstract Cerebrospinal fluid amyloid beta 42, total tau, and phosphorylated tau 181 are well accepted markers of Alzheimer’s disease. These biomarkers better reflect disease pathogenesis compared to clinical diagnosis. Here, we perform a genome wide association study meta-analysis including 18,948 individuals of European ancestry and identify 12 genome-wide significant loci across all three biomarkers, eight of them novel. We replicate the association of biomarkers with APOE , CR1 , GMNC/CCDC50 and C16orf95/MAP1LC3B . Novel loci include BIN1 for amyloid beta and GNA12, MS4A6A, SLCO1A2 with both total tau and phosphorylated tau 181, as well as additional loci on chr. 8, near ANGPT1 and chr. 9 near SMARCA2 . We also demonstrate that these variants have significant association with Alzheimer’s disease risk, disease progression and/or brain amyloidosis. The associated genes are implicated in lipid metabolism independent of APOE , coupled with autophagy and brain volume regulation driven by total tau and phosphorylated tau 181 dysregulation.

Background:While genetic factors strongly influence brain aging trajectories, variants conferring cognitive resilience remain poorly characterized. The neurokinin-3 receptor (NK3-R), encoded by Tachykinin Receptor 3 (TACR3), modulates cholinergic signaling in memory circuits vulnerable to aging. Previous studies linked the non-WT expression of the TACR3 variant rs2765 with cognitive decline and reduced volume of the hippocampus and basal forebrain, but systematic replication and mechanistic validation were lacking. Methods:We investigated rs2765 in the preregistered AgeGain cohort of cognitively healthy older adults (n=188) with independent validation in the ADNI cohort (n=809) which includes persons with and without Alzheimer's Disease (AD) that show healthy cognition, mild cognitive impairment or dementia. Analyses integrated structural neuroimaging, longitudinal cognitive assessments, epigenetic aging (PhenoAge), genome-wide methylation profiling, and mechanistic validation through luciferase assays and cross-species protein expression studies. Results:The infrequent protective rs2765 WT variant, found in 12.8% of Europeans, conferred 49% slower cognitive decline (p = 0.002) for amyloid-positive individuals of the ADNI cohort and 3.7 years younger epigenetic age (p = 0.013, 95% CI: 0.79-6.67 years) in the cognitively healthy AgeGain cohort. WT carriers showed larger hippocampal and basal forebrain volumes across cohorts, with Allen Brain Atlas integration revealing these outcomes to occur exclusively in regions where TACR3 expression positively correlated with gray matter volume. Mechanistically, the non-WT variant ameliorated RBMX-mediated post-transcriptional regulation, reducing NK3-R protein expression by 25-40% in vitro and ex vivo murine brain slice models. Senescence-accelerated mice exhibited reduced endogenous NK3-R expression, phenocopying the predicted functional consequences of the variant. In AgeGain participants, genome-wide methylation profiling identified 2,313 differentially methylated CpGs affecting 228 pathways spanning glutamatergic signaling, acetylcholine receptor pathways, chromatin remodeling, and angiogenesis, suggesting coordinated molecular reprogramming from synaptic function to systemic aging. Conclusions:rs2765 WT confers resilience to age- and AD-related cognitive decline through RBMX-dependent regulation of NK3-R expression, with effects of remarkable size cascading from memory to systemic aging. rs2765 genotyping could stratify individuals for NK3-R modulator therapy (e.g., fezolinetant or senktides) and identify those maintaining function despite pathological burden, complementing APOE-based risk assessment in precision geromedicine.

The basal forebrain (BF), home to cholinergic neurons essential for attention and memory undergoes structural changes in sporadic Alzheimer's Disease and in at-risk individuals, contributing to cognitive decline. To investigate whether BF volume is reduced in preclinical autosomal-dominant Alzheimer's disease (ADAD), we studied cognitively-unimpaired carriers of the PSEN1 E280A mutation from the Colombian kindred, the largest ADAD cohort with a single mutation, known for early cognitive decline (mild cognitive impairment at age 44, dementia at 49). Age was used as a proxy for disease progression to analyze BF volume and its relationship with age and cognitive performance. This study included 127 cognitively-unimpaired individuals from the PSEN1 Colombian kindred (60 carriers, 67 non-carriers; mean-age: 30.67 ± 6.65 years; mean-education: 12.24 ± 3.06 years). Unimpaired status was defined by Functional Assessment Staging (FAST) scores <2. Participants underwent structural MRI and cognitive testing, with BF volumes measured using a cholinergic nuclei map. Cognition was assessed with the Mini-Mental State Examination (MMSE) and the CERAD Word List Learning (WLL) task. BF volume differences between groups were assessed using a t-test, and partial Pearson correlations (adjusted for sex, education, and intracranial volume) were used to evaluate relationships between BF volume, age, and cognition. BF volumes did not differ significantly between carriers (693.83 ± 68.3 mm3) and non-carriers (694.00 ± 65.24 mm3) ( p  =  0.82). Age was negatively correlated with BF volume in the overall sample ( r  = -0.41, p  =  2.7e-06), carriers ( r  = -0.41, p  =  0.001), and non-carriers ( r  = -0.44, p  =  2.6e-04). However, BF volume showed no significant correlation with MMSE or WLL in the overall sample, carriers, or non-carriers. These findings suggest that Alzheimer's-related volumetric changes in the basal forebrain may not manifest during the early preclinical stages of ADAD. This highlights the importance of future studies incorporating longitudinal measures to track BF changes over time, spanning the continuum from preclinical to clinical stages. Such research could provide critical insights into the temporal dynamics of BF involvement and its potential as a marker for early detection or therapeutic target in ADAD.

IntroductionNormal aging is associated with alterations of functional connectivity in brain neuronal networks. Altered network connectivity may be associated with accelerated cognitive decline. Physical activity is considered a beneficial lifestyle factor for maintaining cognitive health. Higher intensities of physical activity may induce structural and functional changes in the brain, particularly in regions involved in cognitive functions. However, the underlying neural mechanisms are not widely investigated. Our aim was to examine the association between resting-state functional connectivity of brain networks previously associated with cognitive and motor functions, physical activity and cognitive performance in healthy older adults.MethodsWe analyzed resting-state fMRI, physical activity and neuropsychological data of 149 healthy older adults (mean age: 68 years). Physical activity was measured by using actigraphs worn for 7 days and categorized into moderate-to-vigorous activity. Euclidean norm minus one values used to represent mean overall physical activity. We used a hypothesis driven seed-based approach and data-driven independent component analysis to examine brain network activity of a priori selected brain regions and networks.ResultsNo significant associations were found in the seed-based analyses. The independent component analyses showed spatially restricted effects of moderate-to-vigorous physical activity in frontal regions of the default mode and salience networks, at p < 0.01 uncorrected.ConclusionDifferent physical activity intensities were not significantly associated with resting-state functional connectivity of various brain networks in a sample of healthy older adults. This finding contrasts with the results of previous cross-sectional studies.

OBJECTIVES:Amyotrophic lateral sclerosis (ALS) and Alzheimer's disease (AD) share neuropathological features, including tau, amyloid, and TDP-43 pathology. This study investigated whether AD-related pathological changes are associated with cognitive impairment ALS. METHODS:Cerebrospinal fluid (CSF total-tau, phosphorylated-tau, beta-amyloid) and plasma biomarkers (TDP-43; neurofilament light chain [NfL]) were analyzed in 192 individuals with ALS or ALS with frontotemporal dementia (ALS-FTD) and 100 healthy controls. Cognitive performance was assessed using the Edinburgh Cognitive and Behavioral ALS Screen (ECAS). Group comparisons and regression analyses examined associations between biomarker profiles and cognitive status. Autopsy data were available for a subset of participants. RESULTS:Compared with healthy controls, patients with ALS - particularly those with cognitive impairment (ALSci) or ALS-FTD - showed elevated AD-related biomarkers. Significant differences in beta-amyloid levels were observed between healthy controls (HCs) and patients with ALSci, but not between controls and cognitively unimpaired patients. CSF p-tau and total-tau levels were strongly associated with domain-specific cognitive performance. In contrast, plasma extracellular vesicle TDP-43 and NfL showed weak or no association with cognition. In vivo biomarkers alone reliably distinguished cognitive impairment only in ALSci and ALS-FTD. Postmortem analyses showed no strong association between ABC scores or overall TDP-43 burden and cognitive state; however, temporal and hippocampal TDP-43 burden was associated with cognitive dysfunction. INTERPRETATION:Our findings suggest that tau-related CSF biomarkers, particularly p-tau and total-tau, are associated with cognitive deficits in ALS, indicating that AD-related pathology might be associated to cognitive decline in ALS. However, postmortem data showed even stronger relation of TDP43 pathology to cognitive deficits in ALS. ANN NEUROL 2026;100:123-138.

Background Early and accurate detection of Alzheimer’s disease (AD) is essential for timely intervention and development of disease-modifying treatments. The DZNE-Longitudinal Cognitive Impairment and Dementia Study (DELCODE) provides a deeply phenotyped cohort covering preclinical and early clinical stages, including subjective cognitive decline (SCD) and mild cognitive impairment (MCI). Astrocyte reactivity and its biomarkers, particularly glial fibrillary acidic protein (GFAP), have gained increasing attention in AD research; however, the relationship between GFAP and amyloid in early disease, as well as its potential prognostic value beyond its association with amyloid status, remains insufficiently understood. Objectives To evaluate the performance of CSF and plasma GFAP across early disease stages, compare these measures according to amyloid status, and assess the prognostic value of GFAP for clinical progression across diagnostic stages during longitudinal follow-up. Setting This study used data from the multicenter DELCODE cohort in Germany, including participants with available plasma and/or CSF samples and standardized clinical, cognitive, imaging, and biomarker assessments. Measurements GFAP concentrations in plasma and CSF were quantified using validated immunoassay platforms. Standard CSF AD biomarkers and ApoE genotype were measured using established assays. Amyloid status was defined by the CSF Aβ42/40 ratio. Longitudinal follow-up occurred annually for up to ∼10 years, with clinical conversion determined according to NIA-AA criteria. Results Plasma and CSF GFAP increased across the AD continuum, with higher levels in MCI and AD (p < 0.001). Plasma GFAP showed a stronger association with amyloid status than CSF GFAP across all groups. In MCI, plasma GFAP combined with age and ApoE4 yielded an AUC of 0.87. Elevated plasma GFAP predicted increased risk of conversion to MCI (HR = 2.19, p < 0.001; adjusted HR = 1.70, p = 0.0056) and AD dementia (HR = 3.5; adjusted HR = 2.49 both p < 0.001). Conclusion Plasma GFAP is a sensitive, minimally invasive biomarker with diagnostic relevance for amyloid detection and prognostic relevance for clinical progression in early AD.

ABSTRACT Phosphorylated tau181 (p‐tau181), an Alzheimer's disease biomarker, was recently evaluated in amyotrophic lateral sclerosis (ALS). We investigated plasma p‐tau181 in 202 ALS/ALS‐FTD patients and 94 healthy controls, assessing cognitive performance, motor function, and longitudinal dynamics. Plasma p‐tau181 and NfL were significantly elevated in ALS, with p‐tau181 increasing over 1 year while NfL remained stable. Neither marker correlated with cognitive performance, and only NfL was associated with disease severity and progression. Plasma p‐tau181 was higher in patients with predominant lower motor neuron involvement. The results indicate that p‐tau181 reflects peripheral processes in ALS, providing a complementary, mechanistically distinct biomarker from NfL.

Research on visual episodic memory impairment in Alzheimer's disease often focuses on memory processes rather than the specific content of image being remembered. We previously showed that patients with mild cognitive impairment (MCI), a transitional stage that may precede Alzheimer's disease, can memorize certain images well, indicating that episodic memory is not uniformly impaired. Conversely, other specific images could not be memorized by MCI patients and were instead diagnostic for distinguishing MCI from healthy older adults. In this study, we investigate whether poor memory for these diagnostic images relates to impaired neural processing in specific brain regions and Alzheimer's biomarker pathology. We assessed 64 healthy controls and 48 MCI participants from the DZNE Longitudinal Cognitive Impairment and Dementia Study. Participants performed a visual scene memory task during fMRI and provided CSF biomarker data for amyloid and tau. Diagnostic images demonstrated significantly larger behavior-biomarker correlations (total tau, phospho-tau, Aβ42/Aβ40, and Aβ42/phospho-tau) compared with nondiagnostic images. This suggests memory for these specific diagnostic images is more affected by Alzheimer's disease pathology. The fMRI data revealed an interaction effect between group membership (healthy control/MCI) and image diagnosticity (diagnostic/nondiagnostic). MCI participants exhibited higher activation in specific scene-processing regions (parahippocampal place area, retrosplenial cortex, and occipital place area) for diagnostic compared with nondiagnostic images. Healthy controls, however, showed no processing differences between diagnostic and nondiagnostic images. These findings suggest MCI individuals may engage in inefficiently heightened encoding activation for diagnostic images. Our results show that special "diagnostic" images exist that can reliably reveal underlying amyloid and tau pathology alongside altered neural activity in scene regions.

Structural MRI is routinely acquired in the clinical assessment of Alzheimer's disease, yet quantitative morphometric indices derived from these scans remain largely confined to research settings. Here we present PHASE-AD - a framework that translates such indices into clinically interpretable classifications of atrophy subtype and stage that jointly capture atrophy progression while accounting for inter-individual atrophy heterogeneity. PHASE-AD is trained on MRI scans from 8,415 participants and robustly captures limbic-predominant and hippocampal-sparing atrophy subtypes that were identified across seven independent datasets. Two cross-validation schemes revealed high robustness across different field strength and scanner manufacturer configurations. Atrophy classifications were associated with diverging clinical profiles and tau accumulation patterns. In prospective designs mirroring contemporary AD trials, they stratified longitudinal cognitive trajectories and outperformed semi-quantitative visual MRI assessments as a clinically established comparator. These findings support the integration of automated atrophy subtyping and staging into clinical practice and pharmacological trials.

Abstract Background Potentially modifiable lifestyle and psychological factors may influence Alzheimer’s disease (AD)-related brain pathology and cognitive function, thereby influencing cognitive resilience in late life. Objective This cross-sectional study investigated associations and pathways between lifestyle and psychological factors related to cognitive reserve and psychological debt, AD-related biomarkers, and cognitive function, as well as potential differences in these associations between AD risk groups. Methods In total, 298 non-demented older adults (mean age = 69.5 years, 44% women) of the DELCODE study were included. Structural equation modeling was used to assess the associations between the constructs of cognitive reserve (education, occupational complexity, leisure activity participation) and psychological debt (depression and anxiety symptoms, neuroticism, sleep quality), manifest AD-related biomarkers (cerebrospinal fluid [CSF] amyloid-beta [Aβ] 42, splenial white matter hyperintensities [WMH], hippocampal volume), and latent cognitive function of increased AD risk (Preclinical Alzheimer’s Cognitive Composite [PACC]). In the structural equation model, biomarkers were transformed such that higher values indicated greater AD-related brain pathology and age was included as a covariate. Multigroup analyses assessed moderations by established AD risk modifiers, namely sex and apolipoprotein ε4 (APOE ε4) genotype. Results In the total sample, higher cognitive reserve was associated with better cognitive function (p = .005), independent of AD-related biomarkers. Higher cognitive reserve was associated with lower psychological debt (p = .035); however, neither construct showed a significant association with the AD-related biomarkers (p ≥ .177). AD-related biomarkers of CSF Aβ42 (p = .021), splenial WMH (p = .044), and hippocampal neurodegeneration (p = .007) were each independently associated with lower cognitive function. Most associations were comparable between AD risk groups stratified by sex and APOE ε4 genotype. The relationships between cognitive reserve and psychological debt, and between CSF Aβ42 and splenial WMH were stronger in APOE ε4 non-carriers than in carriers (all p ≤ .020). Conclusions Cognitive reserve emerges as a key resilience pathway, supporting late-life cognition independently of AD-related pathology, with largely consistent effects across AD risk groups. The role of psychological debt warrants longitudinal investigation, particularly in vulnerable older populations. Trial registration German Clinical Trials Register: DRKS00007966, Registered: 4 May 2015.

We propose five recommendations to make AI-based research studies more suitable for a clinical readership. First, authors should justify the added value of complex and potentially more opaque AI approaches. Second, rigorous description of input data, diagnostic criteria, and preprocessing is essential to avoid biased or clinically irrelevant outcomes. Third, benchmarking against clinically relevant performance thresholds should be established a priori. Fourth, method sections should combine an accessible lay summary with detailed technical supplement. Fifth, model explainability is encouraged to mitigate opacity. These recommendations aim to support AI research that is methodologically robust and interpretable for AD researchers.

Microglia are implicated in the progression of Alzheimer's disease (AD) pathology from its earliest stages, suggesting that cerebrospinal fluid (CSF) microglia profiling across clinical AD stages can aid in treatment development and monitoring. We analyzed two CSF cohorts (n = 834) that span from unimpaired controls to preclinical and dementia AD stages, identifying 109 dysregulated microglia-related proteins. Enrichment analyses revealed innate immune processes and cellular recruitment in preclinical AD, whereas AD dementia revealed adaptive immunity and macrophage responses. Next, we aligned the in vivo microglia protein profiles with ex vivo-derived microglial transcriptomic signatures, such as disease-associated microglia phenotypes. Transcriptomic signatures were not specific to either clinical stage but spanned both. We classified an 18-protein panel highlighting distinct changes between the preclinical and dementia stages. Our findings underscore the potential of microglia-based biomarker research for AD staging, offering insights into microglia dynamics in clinical AD stages and how transcriptomic signatures translate to proteomic profiles.

The role of the peripheral immune system in Alzheimer’s Disease (AD) remains insufficiently resolved, limiting the understanding of systemic disease effects and mechanisms. Here, we employed three high-resolution single-cell techniques, including flow cytometry, single-cell RNA- and ATAC-sequencing, to investigate peripheral immunity in AD dementia and earlier stages of the AD trajectory in over 100 patients. We identified reduced humoral immune responses in AD, characterized by a diminished B cell compartment displaying an impaired activation phenotype. Classical monocytes expanded in mild cognitive impairment and early AD dementia, acquiring a NF-kB/AP-1-mediated low-grade inflammation phenotype. Our findings link peripheral dysregulation in innate and adaptive immunity at cell frequency, transcriptional and epigenetic levels to the AD trajectory and provide insights into distinct phenotypes that define AD progression in contrast to healthy aging across cohorts. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement This work was supported by the National Dementia Strategy for Germany. MDB is supported by the Helmholtz Association and the German Research Foundation (DFG) (SFB1454 project number 432325352, IGK2168/2 project number 272482170). LB, ACA, MB, TU, JLS and MDB are members of the excellence cluster ImmunoSensation2 (EXC2151 project number 390873048). ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: All participants or their representatives provided informed consent. Ethics committees of the medical faculties of all participating sites - the ethical committees of Berlin (Charité, University Medicine), Bonn, Cologne, Göttingen, Magdeburg, Munich (Ludwig-Maximilians-University), Rostock, and Tübingen - gave ethical approval for this work. The process was led and coordinated by the ethical committee of the medical faculty of the University of Bonn. The registration number of the trial at the ethical committee in Bonn is 117/13 (please refer to Jessen, F., Spottke, A., Boecker, H., Brosseron, F., Buerger, K., Catak, C., Fliessbach, K., Franke, C., Fuentes, M., Heneka, M.T., et al., 2018. Design and first baseline data of the DZNE multicenter observational study on predementia Alzheimers disease (DELCODE). Alzheimers Res Ther 10, 15. 10.1186/s13195-017-0314-2). The ethics committee of the medical faculty of the University of Bonn gave ethical approval for the analysis of single-cell RNA-seq data under 227/19. DELCODE and DESCRIBE were conducted in accordance with the Helsinki Declaration from 1975. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes scRNA-seq, scATAC-seq and MCFC data are deposited at the DZNE Clinical Research Platform. Code will be deposited at Zenodo and will be available as of the date of publication. Any additional information required for data reanalysis is available from the lead contact upon request.

Alzheimer's disease (AD) is a major cause of dementia and cognitive decline. Here, we assessed how episodic memory (EM) network dysfunction, a hallmark of AD, is related to the longitudinal progression of AD biomarkers, neurodegeneration and cognition using data from the DZNE DELCODE study. This data set includes over 1000 longitudinal functional magnetic resonance imaging measurements of EM network function. We related activation and deactivation of EM to individual disease progression scores from a disease progression model. Voxel-wise analyses revealed widespread loss of deactivation and activation with disease progression. Trajectories for the loss of deactivation were nonlinear, associated with amyloid- and tau-positivity and visually preceded trajectories of cognitive decline. The relationship between deactivation and cognitive decline was partly independent of neurodegeneration. Our results provide evidence that synaptic dysfunction and neurodegeneration are independent drivers of cognitive decline, providing a rationale for targeting synaptic dysfunction along the AD cascade.