The Journal of Prevention of Alzheimer's Disease

BackgroundCognitively unimpaired (CU) adults vary substantially in their risk of developing mild cognitive impairment (MCI), yet most subtyping approaches focus on downstream neurobiological or cognitive markers rather than upstream, modifiable risk factors. We aimed to identify clinically meaningful subgroups of CU adults defined by integrated comorbid, behavioral, and social risk profiles, and to evaluate heterogeneity in both incident MCI risk and cardiometabolic treatment effects. MethodsWe conducted a prospective cohort study of 121,322 CU adults aged [≥]50 years from the All of Us Research Program. Baseline comorbidities, lifestyle behaviors, and social determinants of health were jointly modeled using the Bayesian Mixed Integrative Data Subtyping framework, which integrates binary and continuous modalities via modality-specific likelihoods and shared latent constructs. Subtype-specific risk of incident MCI was assessed using multivariable Cox proportional hazards models adjusting for demographics and baseline medication use. A double/debiased machine learning interactive regression model with inverse probability of censoring weights to mitigate bias from informative censoring was implemented to estimate the average treatment effects of antihypertensive agents, Glucagon-Like Peptide (GLP) receptor agonists, and non-GLP antidiabetic medications on time to MCI. ResultsFour distinct subtypes were identified: I low-risk healthy aging, II behavioral/social vulnerability, III cardiometabolic-depressive multimorbidity, and IV mixed social-medical vulnerability profiles. Compared with Subtype I, Subtype III demonstrated the highest risk of incident MCI (HR: 3.69, 95% CI: 3.14-4.33), followed by Subtype IV and Subtype II. In treatment effect analyses, antihypertensive use was associated with a modest prolongation of MCI-free survival overall (time ratio:1.04, 95% CI: 1.03-1.06), with the largest benefit observed in Subtype III (time ratio: 1.14, 95% CI: 1.09-1.19). Non-GLP antidiabetic therapies were similarly associated with modest overall delay, with significant benefits in Subtypes I and III. GLP-class therapies were not associated with overall delay but showed a significant association in Subtype III. ConclusionsIntegrative subtyping based on comorbid, behavioral, and social risk factors reveals clinically meaningful heterogeneity in both cognitive risk and treatment response. Aligning dementia prevention strategies with dominant vulnerability pathways may enhance the effectiveness and equity of population-level precision prevention.

IntroductionThe eligibility of anti-amyloid disease-modifying therapies (DMTs) and their integration into clinical practice in some institutions requires a specific range of Mini-Mental State Examination (MMSE) scores. Reliance on this pencil-and-paper psychometric instrument imposes operational burdens and risks perpetuating health disparities due to the tests known educational and cultural biases. This study evaluates the efficacy of the Digital Clock and Recall (DCR) - a rapid, FDA-listed digital cognitive assessment - to crosswalk to MMSE scores using machine learning, thereby offering a faster, scalable, and equitable mechanism for patient triage. MethodsWe conducted a retrospective analysis using data from the multi-site Bio-Hermes-001 study (NCT04733989, N=945). Participants were clinically classified as cognitively unimpaired, mild cognitive Impairment, or probable Alzheimers dementia. We trained a Poisson elastic net regression model using age and multimodal digital features derived from the DCR (including drawing kinematics and voice acoustics) to predict MMSE scores. The model was tested for generalizability using an independent external validation cohort from the Apheleia study (NCT05364307, N=238). ResultsThe machine learning model predicted MMSE scores with a root mean squared error (RMSE) of 2.31 in the training cohort. This error margin falls within the established test-retest reliability range of the manual MMSE itself (2-4 points), suggesting the prediction is statistically non-inferior to human administration. External validation in the Apheleia cohort demonstrated robust generalizability (RMSE = 2.62). Crucially, the model exhibited demographic fairness, maintaining consistent accuracy across Race (White RMSE = 2.34; Non-White RMSE = 2.14) and Ethnicity (Hispanic RMSE = 2.26; Non-Hispanic RMSE = 2.31). DiscussionMachine learning can leverage multimodal features from the DCR to accurately and equitably crosswalk to MMSE scores in support of current guidelines, transforming a time-intensive manual test into a rapid, automated assessment. By deploying this "digital triage" engine, where traditional assessments are still used for DMT eligibility, healthcare systems can streamline the identification of DMT-eligible patients, reduce specialist referral bottlenecks, and ensure that access to life-altering therapies is determined by pathology rather than demography.

Abstract/SummaryFinancial exploitation of older adults is an increasingly prevalent public health concern, yet few have characterized fraud prevalence longitudinally or evaluated whether financial exploitation vulnerability measures prospectively predict fraud outcomes. Using data from the Health and Retirement Study, we examined fraud prevalence across a 14-year period and tested whether the Perceived Financial Vulnerability Scale (PFVS) predicts subsequent fraud victimization among older adults. Fraud prevalence increased steadily over time, rising from 5.0% in 2008 (347 of N=6,920) to a peak of 10.2% in 2022 (448 of N=4,380). Higher PFVS scores measured in 2018 were associated with greater odds of fraud victimization reported in 2022 (OR=1.62, 95% CI [1.25-2.15], p<.001). Most individuals who later reported fraud fell within the highest group of PFVS scores up to five years earlier. Together, these findings highlight financial exploitation as an emerging aging-related vulnerability and support the PFVS as a brief indicator of future fraud risk.

ObjectivesTo characterize contemporary pharmacologic treatment patterns for Alzheimers disease and related dementias (ADRD) among U.S. long-term care residents and to examine facility- and resident-level factors associated with treatment. DesignRetrospective, observational study. Setting and ParticipantsElectronic health record data from 1,675,873 long-term care residents in the PointClickCare Life Sciences database included 359,801 with a documented ADRD diagnosis in skilled nursing facilities in the U.S. (January-April 2025). MethodsResidents were classified as treated/untreated based on receipt of guideline-directed ADRD therapy, consistent with Alzheimers Association guidelines. Analyses incorporated demographics, comorbidities, medication burden, and facility characteristics. Multivariate logistic regression estimated odds of receiving guideline-concordant therapy. ResultsOverall, 72.5% of residents with ADRD received [&ge;]1 pharmacologic treatment recommended for ADRD. Treatment was most common among residents with Lewy body dementia (83.9%) and early-onset Alzheimers disease (82.3%) and least frequent among residents aged [&ge;]90 years (65.1%), Black/African American residents (66.8%), and those with cerebral degeneration (66.8%). Treated residents exhibited higher medication burden (mean 4.4 vs 3.3). Diagnoses for other chronic conditions as well as specific ADRD subtypes strongly impacted probability of treatment; diabetes and hyperlipidemia were associated with lower odds of treatment, whereas ADRD subtypes strongly predicted treatment. Conclusions and ImplicationsMore than one-quarter of residents with ADRD remain untreated with guideline-recommended pharmacotherapy, and treatment varied significantly by non-clinical predictors. These findings underscore the need to investigate and understand possible treatment disparities, optimize polypharmacy management, and discover new ADRD treatments, as current options are often ineffective with many side effects. Brief SummaryThis study used real-world data from electronic health records (EHR) to understand treatment patterns of those with Alzheimers disease and related dementias (ADRD) in U.S. long-term care facilities. International Classification of Diseases Tenth Revision, Clinical Modification (ICD-10) codes were used to identify ADRD diagnoses and medication orders were used to identify treatment. From January to April 2025, there were 359,801 with a documented ADRD diagnosis in skilled nursing facilities. Over 25% of those with ADRD did not have a medication order for a guideline-recommended pharmacological treatment. Comorbidities of diabetes and hyperlipidemia were associated with lower odds of receiving ADRD treatment, suggesting concerns related to adverse drug reactions and competing clinical priorities. The use of cognitive and disease-modifying therapies was low compared to behavioral/psychiatric medications; this finding suggests a need for more effective and safe drugs that target the root causes of ADRD opposed to the behavioral and psychiatric complications. Taken together, the results of this study call for targeted interventions to address disparities in treatment, enhanced clinical decision-making support regarding polypharmacy, and improved pharmacological options for those with ADRD.

INTRODUCTIONPlasma biomarkers for Alzheimers disease (AD) pathology promise scalable diagnostic access, yet their performance in diverse, population-representative cohorts remains uncharacterized. We evaluated equity and transportability of plasma amyloid-tau-neurodegeneration (ATN) biomarkers in a nationally representative U.S. aging cohort. METHODSCross-sectional analysis of 4,427 adults aged [&ge;]50 years from the 2016 Health and Retirement Study Venous Blood Study. Plasma biomarkers (A{beta}42/40, pTau181, NfL, GFAP) were classified using established ATN criteria. Survey weights produced population-representative estimates. Outcomes included biomarker-cognition associations, fairness metrics (sensitivity, specificity, predictive values) stratified by race/ethnicity and sex, and education-stratified analyses. RESULTSAmong 4,427 participants representing 36.6 million U.S. adults (weighted: 68 years, 55% female, 79% White), survey-weighted analysis revealed tau as the only biomarker maintaining robust cognitive associations ({beta}=-0.74, p<0.001), while amyloid ({beta}=0.11, p=0.43) and neurodegeneration ({beta}=-0.27, p=0.08) lost significance. White participants demonstrated 12-percentage-point higher sensitivity than Black participants (23.4% vs. 11.4%), with Black women showing lowest sensitivity (8.8%). Educational attainment modified biomarker effects: low-education groups showed paradoxical positive amyloid associations ({beta}=0.74, p=0.01) and amplified neurodegeneration effects ({beta}=-1.02, p=0.006). Race-specific optimal cutpoints differed by 40%. Vascular comorbidity burden was higher in Black (82%) and Hispanic (73%) versus White (65%) participants, yet associations persisted after vascular adjustment. DISCUSSIONPlasma ATN biomarkers demonstrate significant equity gaps and differential transportability across demographic subgroups. The 12-percentage-point sensitivity disparity and education-dependent effect modification highlight barriers to equitable implementation. Population-based validation with fairness metrics should be prerequisite for clinical deployment.

Background and ObjectivesProgression from mild cognitive impairment (MCI) to Alzheimers Disease and Related Dementias (AD/ADRD) varies widely across individuals, yet the mechanisms underlying this heterogeneity remain unclear. Identifying clinical and social determinants influencing this transition could enable earlier intervention. While cardiovascular and social risk factors are established contributors to dementia incidence, their role in progression from MCI to dementia may differ. Few studies using real world clinical data have evaluated these potential determinants of MCI progression. MethodsUsing electronic health records (EHR) from patients with incident MCI at UCSF Health (2010-2024), we evaluated cardiovascular (blood pressure [BP], body mass index [BMI], and type II diabetes) and social (marital status, language preference, race/ethnicity, and neighborhood disadvantage) risk factors for rate of progression from MCI to AD/ADRD. Covariate-adjusted Cox proportional hazards models estimated hazard ratios for incident AD/ADRD, with evaluation of interactions by sex. ResultsAmong 6,529 patients, higher systolic BP was associated with AD/ADRD incidence (HR per 10 mmHg: 1.09, 95% CI: 1.05-1.14). BMI was inversely associated with incidence in both males (HR: 0.94; 95% CI: 0.92-0.97) and females (HR:0.98; 95% CI: 0.96-0.99). Compared to married individuals, widowed patients had a higher hazard of progression (HR: 1.15; 95% CI: 1.00-1.32). Spanish-speaking (HR: 1.38; 95% CI: 1.04-1.81), Chinese-speaking (HR: 1.19; 95% CI: 1.00-1.42), and "Other non-English" speaking patients (HR:1.24; 95% CI: 1.03-1.51) had a higher hazard of progression compared to English speakers. Latinx (HR:1.22; 95% CI: 1.01-1.48) and Asian patients (HR:1.14, 95% CI: 1.00-1.30; p=0.04) also had higher hazards of progression compared to White patients. Neighborhood disadvantage was not significantly associated with disease progression. DiscussionCardiovascular and social factors independently influence dementia progression, with some sex-specific patterns. Integrating clinical and social indicators highlights the potential of EHR data to identify high-risk patients earlier in the care continuum and support equitable dementia prevention.

Objective: Behavioral and social factors (BSFs) substantially influence the risk, onset, and progression of Alzheimer disease and related dementias (ADRD). A systematic representation of their interplay is essential for advancing prevention and targeted interventions. However, BSF-related knowledge is scattered across heterogeneous sources, limiting scalable evidence synthesis and computational analysis. To address this, we created a Behavioral Social Data and Knowledge Ontology for ADRD (BSOAD) to represent and integrate BSFs with respect to ADRD. Material and Methods: BSOAD was developed following established ontology design principles, prioritizing reuse of existing ontology elements to ensure semantic interoperability. It was built upon the Social Determinants of Health Ontology (SDoHO) and the Drug-Repurposing Oriented Alzheimer Disease Ontology (DROADO). BSF-related classes were enriched with ICD 10 CM Z55 Z65 codes and ADRD related classes with AD Onto. Relationships between BSFs and ADRD were derived through literature mining. Ontology quality was evaluated through Hootation based expert review and an LLM assisted framework assessing structural coverage and semantic coherence. Results: BSO AD contains 2275 classes, 153 object properties, and 49 data properties. Expert review demonstrated strong rational agreement (0.95), with disagreements resolved through discussion. LLM-based evaluation showed high category coverage rates ([&ge;] 0.97) and robust semantic alignment with the relevant literature (average completeness = 0.79; conciseness = 0.94). Discussion and Conclusion: BSOAD is, to our knowledge, the first ontology to systematically represent BSFs and hierarchically model their interrelationships in ADRD. It establishes a semantic backbone for computational analysis and knowledge integration. The LLM assisted evaluation framework demonstrates the feasibility of scalable, automated ontology assessment.

Mouse lifespan studies are slow and costly, limiting the number of interventions that have demonstrated robust anti-aging effects. This highlights the need for rapid early-stage screening tools capable of assessing both efficacy and potential side effects. Here, we present a short-term performance assay designed to rapidly profile functional benefits and early toxicity of longevity interventions in mice. Over an 8-week period, mice received one of five candidate anti-aging treatments: 17-estradiol, rapamycin + Smer28, berberine + resveratrol, sildenafil and pinealon. The protocol longitudinally monitored body weight and temperature, and food intake, alongside post-treatment assessments of grip strength, locomotor activity, Y-maze cognition, social behavior, and hematological and urinary parameters. The screen revealed compound-specific phenotypes: 17-estradiol induced significant weight loss, increased grip strength, and dorsal alopecia, consistent with metabolic remodeling. Sildenafil reduced basal body temperature and preserved locomotor activity. Berberine + resveratrol decreased food intake and fasting glucose without major changes in physical performance, resembling caloric restriction-like metabolic effects. Rapamycin + Smer28 modestly improved strength and sociability but induced anemia in 2 of 5 mice, indicating potential dose-dependent toxicity. Pinealon showed a trend toward improved working memory without detectable adverse effects. This multi-parametric approach enables discover healthspan extending interventions facilitating prioritization and dose refinement before committing to full lifespan studies. Finally, to our knowledge, this represents the first comprehensive preclinical aging study in mice fully funded through tokenized decentralized science (DeSci), demonstrating how community-governed, on-chain funding can support resource-intensive in vivo research.

BackgroundWhen creating risk prediction models for time-to-event data, methods that incorporate time are typically used. Random survival forests (RSF), an extension of random forests (RF), are one such class of models. We compared RSF to RF in the context of time-to-event outcomes in the ASPirin in Reducing Events in the Elderly (ASPREE) randomized controlled trial. We hypothesize that RSF will have superior discrimination and calibration versus RF. MethodsParticipants from ASPREE residing outside the US or with missing data were excluded. A total of 2,291 participants were assigned 1:1 into training and test sets. RF and RSF models were trained using a total of 115 measures as candidate predictors. The outcome of interest was the earliest of incident dementia, physical disability, or death. ResultsThe primary endpoint occurred in 10.5% of participants. Discrimination was similar between the models: sensitivity ([~]0.75), specificity ([~]0.57), positive predictive value ([~]0.17), time dependent AUC ([~]0.71), and Harrells concordance ([~]0.73). Calibration was likewise similar, Brier score ([~]0.09). DiscussionThe RF and RSF models exhibited comparable discrimination and calibration. We conclude that RSF may not always lead to more accurate predictions of outcomes compared to RF. Further examination in different clinical trial cohorts is needed to better understand the context in which adding time into outcomes risk modeling adds value.

BackgroundDespite national efforts to improve research inclusion, people from underserved communities remain underrepresented in dementia trials. Barriers occur at the point of initial engagement and also within the participation pathway itself, as the structure and burden of early screening procedures can discourage continuation. ACCESS D (Advancing Community Collaboration and Engagement Strategies in Dementia) aims to address these challenges by testing a community-based model that combines co-produced engagement events, low-burden research activities, and real-time support from the South Central Ambulance Service (SCAS), a trusted, community-visible healthcare workforce. Methods and analysisACCESS D is a 12-month mixed-methods feasibility study recruiting 100 adults aged 50-90 years with either (i) a diagnosis of mild cognitive impairment or dementia or (ii) a self- or proxy-reported memory concern affecting daily life. The study will deliver 12-18 co-produced community outreach events in non-clinical settings, supported by SCAS research paramedics and nurses. Following consent, participants will complete a core questionnaire and may optionally take part in one or more low-burden research activities designed to provide supported, first-hand experience of dementia research. Feasibility outcomes, including pathway progression and opt-in to future dementia research contact, will be descriptively summarised and stratified using NIHR INCLUDE-aligned underserved characteristics. Qualitative interviews and focus groups with participants and staff will examine acceptability, perceived value, barriers and enablers, and implementation learning, analysed using thematic analysis and integrated with quantitative findings. Ethics and disseminationThe study has received a favourable opinion from the Southwest Frenchay Research Ethics Committee and Health Research Authority approval (IRAS 361074). Findings will be disseminated via peer-reviewed publications, conference presentations and co-produced lay outputs for community partners and participants. These outputs will be accompanied by an implementation toolkit for research teams and a visual summary for potential participants. Strengths and limitations of this studyO_LIThis study tests a community based, co-produced delivery model that embeds real research participation within outreach, addressing both awareness and structural barriers to dementia research participation. C_LIO_LIOutreach is delivered by a trusted, community visible healthcare workforce, enabling real time support and reducing psychological, practical and digital barriers for underserved groups. C_LIO_LIThe mixed methods feasibility design combines quantitative indicators of reach, progression and resource use with qualitative insights into participant and staff experience, generating actionable learning for scale up. C_LIO_LIOutcomes are explicitly equity-stratified using NIHR INCLUDE aligned characteristics, allowing early assessment of differential reach and engagement across underserved populations. C_LIO_LIAs a single region feasibility study, findings may have limited immediate generalisability. However, the study is designed to generate transferable implementation insights and inform a future multi-site evaluation. C_LI

IntroductionPhosphorylated tau-217 (p-tau 217) is widely used as a plasma-based biomarker for Alzheimers Disease (AD) detection, demonstrating superior accuracy for detecting brain amyloid pathology. However, 30-50% of patients fall within an intermediate diagnostic "gray zone" where biomarker results are indeterminate, often decreasing physician confidence and requiring subsequent diagnostic workup. To address this, we developed a two-stage machine learning algorithm GRAD: Gatekeeper & Reflex for Alzheimers Disease to increase clinical confidence and reduce the AD health economic burden. MethodsWe initially analyzed 320 participants from the Alzheimers Disease Neuroimaging Initiative (ADNI) with plasma biomarkers and amyloid PET. We then built a two-stage machine learning classifier mimicking real clinical workflow where the stage 1 Gatekeeper used the gold-standard marker: p-tau217 with respective 25%/75% probability thresholds. The stage 2 Reflex step applied Random Forest multi-marker classification (p-tau 217, AB42/40, NFL, GFAP) for difficult-to-diagnose gray zone cases. To ensure statistical robustness, leave-one-out cross-validation with bootstrap confidence intervals was used. We externally validated the GRAD architecture on 1,644 A4 Study participants, with MRI enhancement analysis in 1,044 gray zone cases. To measure cost-effectiveness we compared our GRAD-staged testing to universal PET. ResultsThe models Gatekeeper resolved 55.6% of ADNI cases with 88.8% accuracy (NPV 91.8%, PPV 85.0%). The complete pipeline achieved AUC 0.867 (95% CI: 0.825-0.904), with 80.6% sensitivity, 80.0% specificity, LR+ 4.03, LR-0.24. For the difficult-to-diagnose gray zone cases, the Reflex machine learning model achieved AUC 0.755. In our A4 validation, the predictions correlated strongly with centiloid (r= 0.693). Expanding beyond plasma biomarkers, MRI integration improved gray zone classification from AUC 0.829 to 0.853 (p=0.014). The cost modeling analysis projected a 67% reduction in spending versus the current standard of universal PET. DiscussionOur clinically-staged diagnostic algorithm, GRAD, provides actionable classifications for the majority of patients while routing uncertain cases for additional workup. The GRAD framework offers a practical, cost-effective approach for implementing plasma biomarkers in clinical practice. Future iterations of this framework, with integration of novel biomarkers like MTBR-tau243 present a significant opportunity to alleviate the AD health-economic burden and eliminate expensive but unnecessary diagnostic measures. HighlightsO_LIGRAD: Two-stage "Gatekeeper + Reflex for Alzheimers Disease" algorithm resolves indeterminate plasma p-tau217 or gray zone patients with AUC of 0.755. C_LIO_LIOverall AUC of 0.867 (95% CI: 0.825-0.904) validated via leave-one-out cross-validation C_LIO_LIExternal validation in A4 Study demonstrates strong correlation with amyloid burden (r=0.693) C_LIO_LIMRI volumetric integration provides significant incremental value ({Delta}AUC=+0.025, p=0.014) C_LIO_LIProjected 67-71% cost reduction compared to universal PET screening C_LI Research in ContextO_ST_ABSLiterature ReviewC_ST_ABSWe searched PubMed, Google Scholar, and medRxiv databases for studies up to December 2025 that examined plasma p-tau217 diagnostic accuracy as well as "gray zone" management of patients. While several studies demonstrate area-under-the-curve (AUC) of >0.90, these studies largely compare cognitively normal individuals to those with established AD dementia with maximal biomarker separation [1-6]. The gray zone problem, affecting 30-50% of tested individuals, remains unaddressed in the vast majority of clinical implementation frameworks [7,8]. More recent work has established probability-based interpretation [9], but more cohesive algorithms for gray zone resolution through multi-marker integration remain rare if present. Furthermore, the health economic impacts of such resolution have not been fully established. InterpretationOur two-stage algorithm provides a workflow with clinical implementation potential, analogous to established laboratory medicine (i.e TSH with reflex free T4 testing). By first identifying high-confidence cases through univariate p-tau217 (55.6% resolution at 88.8% accuracy), and then applying multi-marker classification only to uncertain cases, we are able to achieve optimal resource utilization while simultaneously maintaining diagnostic accuracy. The finding that MRI usage provides statistically significant improvement ({Delta}AUC=+0.025) has practical implications given the fact that there is a reasonable level of MRI availability in clinical settings. Future DirectionsWhile this work accomplishes several key priorities, future work is required to validate them in diverse clinical populations. In addition, integration of other plasma markers (ex. MTBR-tau243), development of clinical decision support tools, reimbursement mechanisms, and longitudinal validation for treatment monitoring will be necessary to ensure the appropriate infrastructure exists to support providers and patients. Preliminary evidence suggests that %p-tau217 (the ratio of phosphorylated to total tau-217) and MTBR-tau243, a mass spectrometry-based marker of tau tangle pathology, may substantially improve gray zone classification by capturing complementary aspects of tau biology not reflected in absolute p-tau217 concentrations alone, which is a direction that future technical work should examine further.

Improving diversity in U.S. Alzheimers disease (AD) research is a pressing need. By 2050, Hispanic and Latino Americans will comprise 30% of the population. Hispanics are 1.5 times more likely and Blacks are twice as likely to develop AD compared to Whites, yet both remain vastly underrepresented in clinical trials research. Aging and AD research mentorship of underrepresented STEM undergraduates is designed to promote entry into related professions by students committed to decreasing disparities in AD research participation and clinical care. The NIA-funded MADURA program recruited 93 students from backgrounds historically underrepresented in STEM majors and/or from NIH-defined disadvantaged backgrounds. Trainees were placed in aging/AD research labs and received weekly training and mentorship from faculty research PIs and other types of supervisors (postdoctoral researchers, graduate students, research assistant staff...) Our study examined student ratings of the program and mentor behaviors, using a program-specific survey and the Mentoring Competency Assessment-21 (MCA-21). Trainees were highly satisfied with both mentoring relationships and the overall program. Student rated MCA-21 competency areas were quite high for both P.I.s and other types of research mentors. However, there were striking differences in associations between competencies and relationship and program satisfaction, by mentor type. For PI mentors, no MCA-21 competencies were associated with relationship satisfaction, but five of six competencies were associated with relationship satisfaction for other mentor types. Similarly, no PI mentor competencies were significantly correlated with overall placement satisfaction, but all six competencies were correlated with overall placement satisfaction for other mentor types. The authors discuss the likelihood of differing student expectations of faculty PI versus other types of research mentors, recommendations for assessing role-specific student expectations (including functions primarily possible only for senior faculty PIs), and utilizing nearer-peer plus PI faculty mentors to comprehensively address the gamut of mentee needs.

IntroductionAlzheimers disease (AD) prevalence rises with societal ageing. In clinical care, behavioral and psychological symptoms of dementia (BPSD)--including depression, agitation/aggression, apathy, and sleep disturbance--worsen patients quality of life and substantially increase caregiver burden, more significantly than the cognitive symptoms. Standard BPSD assessments rely on caregiver-rated questionnaires that are cross-sectional and may be biased when caregivers are themselves older adults. Device-based measures (e.g., research-grade wrist actigraphy) can provide objective longitudinal data and novel features. In parallel, therapeutic wearables may improve sleep and mood in adults, and might improve BPSD if accepted by people living with dementia. This study aims to assess the feasibility and acceptability of two wearables (Geneactiv actigraphy and Re-Timer circadian regulator) among AD patients with significant BPSD and their caregivers in Taiwan. MethodsThis dyadic pilot study will recruit 20 participants (n=10 AD patients; n=10 caregivers) from outpatient services and affiliated day-care/dementia hubs in rural Taiwan. Participants will wear Geneactiv continuously for 8 weeks and Re-Timer [&ge;]30 min/day for 4 weeks. Device-based data will be processed with GGIR, a well validated R-package designed for processing accelerometer data. Questionnaire assessments include Pittsburgh Sleep Quality Index, PSQI (PSQI), Neuropsychiatry Inventory Questionnaire (NPI-Q), Caregiver Burden Inventory (CBI), and a semi-structured interview based on the Taiwanese version of Quebec User Evaluation of Satisfaction with Assistive Technology (T-QUEST) at prespecified timepoints. DiscussionWearable devices may facilitate the measurement and treatment of specific BPSD, as well as reduce caregiver burden. If proven feasible even in rural Taiwan where both digital and health literacy and resources are limited, this model will inform how device-based dementia care model can be considered and applied in the context of global ageing. Ethics & registrationThis protocol has been approved by the China Medical University and Hospital Research Ethics Committee (CMUH114-REC3-072), and pre-registered in ClinicalTrials.gov (NCT07249918).

BackgroundEarly diagnosis of dementia can significantly improve care planning and patient outcomes while delaying progression. Machine learning algorithms can identify patterns in clinical and neuroimaging data that may aid in the early detection of dementia risk factors. ObjectiveTo evaluate the performance of the ensemble machine learning pipeline for classifying dementia status utilizing demographic, clinical, and imaging features, and to identify the most predictive variables contributing to model accuracy. MethodsA cross-sectional study analyzed 373 MRI scans from 150 subjects aged 60-98 years. Variables included cognitive scores (MMSE, CDR), volumetric brain measures (eTIV, nWBV, ASF), demographic features (age, sex, education), and socioeconomic status. After preprocessing and imputing missing values with random forests, tree-based variable selection was performed, and the dataset was split into training and test sets, with 5-fold cross-validation used for model validation. An ensemble of 8 machine learning models was used to classify patients as demented or non-demented. ResultsModel performance was assessed using the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, precision, F1 score, and Matthews Correlation Coefficient (MCC). Random Forest achieved the highest AUC (0.963), while MLP demonstrated the highest accuracy (94.6%), F1-score (0.943), and MCC (0.893). CDR, MMSE, and ASF were identified as the top predictors. Performance was robust across folds in 5-fold CV, and feature importance analyses supported clinical relevance. ConclusionsEnsemble ML approaches offer high predictive performance in dementia classification. ML frameworks have the potential to be integrated into diagnostic support tools, enabling more accurate and earlier detection of dementia using clinical and imaging data.

BackgroundSubjective cognitive decline (SCD) is common in older adults and may precede mild cognitive impairment (MCI). Whether longitudinal changes in self- or study partner (SP)-reported SCD improve early identification of individuals at risk for clinical progression, particularly along the Alzheimers disease (AD) biological continuum, remains unclear. MethodsWe pooled data from two longitudinal observational cohorts (DELCODE and ADNI). Cognitively unimpaired (CU) participants were recruited through public advertisement or memory clinics and included if baseline amyloid status, [&ge;]2 SCD assessments, and clinical follow-up were available. SCD was assessed using the Everyday Cognition questionnaire (self- and SP-report). Linear mixed-effects models examined longitudinal associations between SCD trajectories, baseline AD biomarkers, and progression to incident MCI. Multivariable Cox proportional hazards models tested whether one-year changes in SCD predicted subsequent progression. FindingsAmong 770 participants (median age 69{middle dot}9years [IQR 66{middle dot}0-74{middle dot}6]; 52{middle dot}6% women; median follow-up 5{middle dot}0years [4{middle dot}0-7{middle dot}0]), amyloid-positive individuals and those who progressed to MCI showed steeper longitudinal increases in both SCD reports. In amyloid-positive participants, only increases in SP-reported SCD differentiated progressors from non-progressors. One-year increases in SP-reported SCD predicted a higher risk of subsequent MCI compared with unchanged scores (hazard ratio 3{middle dot}24 [95%CI 1{middle dot}73-6{middle dot}07]), with effects confined to amyloid-positive participants. InterpretationLongitudinal increases in SP-reported cognitive difficulties, particularly over short intervals, are associated with near-term progression to MCI in amyloid-positive CU older adults. SP-based longitudinal monitoring may represent a low-burden approach to support earlier clinical surveillance in aging populations. FundingGerman Center for Neurodegenerative Diseases, US National Institutes of Health.

ObjectiveAccurate and scalable dementia phenotyping from electronic health records (EHRs) is foundational for population-level research, risk prediction, and learning health system interventions. Traditional rule- and keyword-based approaches are limited by inconsistent documentation and inability to capture clinical nuance. We aim to develop and evaluate a framework that leverages large language models (LLMs) with retrieval-augmented generation (RAG) to overcome these limitations and improve dementia identification from real-world EHR data. MethodsUsing EHR data from the Mass General Brigham health system, we first assembled a cohort of adults with potential dementia based on diagnosis codes, problem lists, dementia-related medications, and free-text note mentions. A subset of candidate cases underwent detailed manual chart review to assign gold-standard dementia status. With this labeled sample, we implemented and compared three approaches for dementia ascertainment: (1) a rule-based classifier leveraging structured EHR data, (2) large language models (LLMs) applied to keyword-filtered clinical note excerpts, and (3) a RAG-based LLM framework that integrates retrieved, context-rich note snippets. Within each approach, we evaluated multiple configurations of embedding models, retrieval methods, LLMs, structured-data inclusion, and prompts to identify the best-performing classifier. Performance was assessed using standard classification metrics, including sensitivity, specificity, positive predictive value (PPV), and F1 score, and supplemented by qualitative error analyses to characterize common sources of false positives and false negatives across methods. ResultsThe RAG-based classifier achieved the highest performance (F1=0.933, sensitivity=91.1%, PPV=95.5%) compared to rule-based (F1=0.823, sensitivity=81.1%, PPV=83.5%) and keyword-filtered LLM (F1=0.903, sensitivity=91.7%, PPV=88.6%). Including ICD codes alongside free text in the RAG-based LLM pipeline significantly reduced the PPV and modestly decreased F-1 score. Error analysis revealed that structured-code dependence contributed to false positives, whereas unrecognized contextual cues in notes drove false negatives. ConclusionA RAG-based LLM pipeline without structured ICD codes improved dementia ascertainment from EHR data compared with ICD-based rules and keyword-based filtering. This approach can enhance dementia case identification and support patient care, predictive modeling and risk analysis.

BackgroundThe benefits of amyloid-{beta} monoclonal antibodies for Alzheimers disease are tempered by the risk of amyloid-related imaging abnormalities (ARIA). Detailed real-world characterization of ARIA, including incidence, timing, radiologic severity and localization, natural history, and risk factors, is essential to optimize treatment safety. This study provides a comprehensive description of ARIA in a large real-world cohort of patients receiving lecanemab. MethodsIn this retrospective cohort study from the Mass General Brigham Alzheimers Therapeutic Program, we analyzed data from 468 patients with early Alzheimers disease who were at least 90 days from their first lecanemab infusion, including those whose treatment was modified or discontinued. ARIA was monitored using a standardized MRI protocol. High-dimensional analysis of baseline clinical, laboratory, and biomarker variables was performed using univariate correlations and Cox proportional hazards models with data-driven cutpoints. FindingsThe overall incidence of any ARIA was 25.2%, with ARIA-H occurring in 22.4% of patients and ARIA-E in 12.2%. Symptoms developed in 4.5% of all patients. Both subtypes demonstrated significant occipital involvement, with ARIA-H showing additional frontotemporal predominance. ARIA-H was typically mild and persistent with rare radiologic resolution, whereas ARIA-E was transient, resolving with a mean time to resolution of 75.6 days. Key baseline predictors of ARIA-H included CSF A{beta}42 [&le;]683.8 pg/mL (HR 6.39), [&ge;]1 microhemorrhage (HR 2.55), and at least one APOE {varepsilon}4 allele (HR 1.77). ARIA-E was predicted by diastolic blood pressure >75 mmHg (HR 3.12), serum chloride >105 mmol/L (HR 2.79), at least one APOE {varepsilon}4 allele (HR 2.51), and serum sodium >141 mmol/L (HR 1.70). ARIA-Mixed was associated with elevated serum chloride >105 mmol/L (HR 3.92), at least one APOE {varepsilon}4 allele (HR 2.87), and diastolic blood pressure >75 mmHg (HR 2.60). InterpretationThis comprehensive real-world characterization of ARIA, together with the identification of novel modifiable risk factors for ARIA-E, including elevated diastolic blood pressure and high-normal serum electrolytes, enables personalized risk assessment and tailored monitoring, thereby advancing the safe implementation of disease-modifying Alzheimers therapies. FundingFunding for this project was provided by the Mass General Neuroscience Transformative Scholar Award.

BackgroundEmerging evidence suggests that microbiota play a role in Alzheimers disease (AD) pathology and cognitive performance. Interventions targeting the oral-brain axis may offer neuroprotective benefits, particularly during the early stages of cognitive impairment. This randomized controlled trial (RCT) investigated whether a multistrain probiotic supplement could modulate AD-related plasma biomarkers and cognitive function in older adults with early cognitive impairment. MethodsParticipants from the Gwangju Alzheimers Disease and Related Dementia (GARD) Cohort in Korea were enrolled in a double-blind, randomized, placebo-controlled trial. Older adults with early cognitive impairment were randomized to receive either a multistrain probiotic supplement (KL-P301) or a placebo for 24 weeks. Plasma pTau181, glial fibrillary acidic protein (GFAP), and neurofilament light chain (NfL) were quantified at baseline and follow-up. Cognitive and clinical assessments included the Clinical Dementia Rating (CDR), Mini-Mental State Examination (MMSE), CERAD neuropsychological battery, Stroop test, and Rey-Osterrieth Complex Figure (ROCF). Treatment effects were analyzed using paired t-tests, linear mixed models, and ANCOVA adjusted for baseline and demographic covariates. ResultsOf the 87 participants analyzed (probiotic, n=40; placebo, n=47), the probiotic group exhibited a significant reduction in plasma pTau181 levels compared with the placebo group (p < 0.001). While GFAP and NfL levels remained stable in the probiotic group, the placebo group showed significant longitudinal increases (p = 0.014 and p = 0.041, respectively). Clinically, the probiotic group demonstrated improved CDR (p = 0.010), primarily driven by the memory domain. Domain-specific cognitive analyses revealed that the probiotic group significantly improved in visuospatial construction (Constructional Praxis, p = 0.036; ROCF copy, p = 0.027) and maintained stable constructional recall, whereas the placebo group showed a significant decline (p = 0.025). No significant between-group differences were observed in MMSE, verbal memory, or executive/attentional functions. ConclusionThe multistrain probiotic supplement reduced tau-related pathology and neuroinflammation-associated biomarkers and selectively preserved visuospatial construction and visual memory in older adults with early cognitive impairment. These findings suggest that modulating the oral-immune-brain axis with multistrain probiotics represents a viable, non-pharmacological strategy to slow AD-related pathological progression and cognitive decline in early-stage patients.

INTRODUCTIONAs {beta}-amyloid(A{beta}) accumulates, tau pathology spreads beyond medial temporal(MTL) regions into the neocortex, though some older adults resist this progression, or caTAUstrophe. Given previous evidence of higher tau levels in women, we tested whether tau resistance presents differently in A{beta}+ men and women. METHODSEmploying data from 872 A{beta}+ older adults(NTest=579) across three cohorts, we estimated resistance, defined as the deviation from participants expected level of neocortical tau. Models predicting the expected tau levels were trained separately in females(NTrain=172) and males(NTrain=121) experiencing typical tau progression to assess sex-specific resistance. RESULTSRelative feature importance in female-only and male-only expectation models differed in 97.7% of variables(p[FDR]<0.001). Moreover, age and A{beta} burden predicted male resistance, while CDR, latent PACC, and adjusted hippocampal volume were predictors for both sexes. DISCUSSIONOur study highlights the impact of sex-specific characteristics in the prediction of neocortical tau resistance. Understanding sex-specific resistance pathways informs targeted Alzheimers interventions.

INTRODUCTIONBilingualism is among several lifestyle factors associated with protection against cognitive decline, yet the biological mechanisms through which it exerts these effects remain poorly understood. METHODSWe compared neuropsychological functioning and biofluid markers of brain health between active (n = 280) and passive (n = 287) Spanish-Catalan bilinguals with biomarker-confirmed Alzheimers disease (AD). RESULTSActive bilinguals outperformed passive bilinguals on tests assessing attention/executive functions, language, and visuospatial/visuomotor functioning, demonstrating resilience given the same AD biological stage across participants. Active bilinguals also exhibited significant differences in cerebrospinal fluid and plasma biomarkers of amyloid burden and neuroinflammation, suggesting both resilience and resistance to AD pathophysiologic mechanisms. DISCUSSIONThe protective effects of bilingual experience may engage both resilience and resistance to AD pathophysiology mechanisms. These results underscore the importance of capturing bilingualism in aging cohorts and the study of how lifestyle and sociocultural factors shape the biological expression of neurodegenerative disease.