Early-stage Alzheimer Research Articles

Pathways to selective vulnerability in neuromodulatory subcortical structures during the early pathogenic stages of Alzheimer’s disease

AbstractBackgroundThe two main catecholaminergic members of the isodendritic core nuclei are the substantia nigra (SN) and the locus coeruleus (LC). Despite cytoarchitectural and neurochemical similarities between the two nuclei, the LC is significantly more vulnerable to early degenerative changes in tauopathies like Alzheimer’s disease (AD) as compared to the SN. Here, we seek to use the gradient of vulnerability in these catecholaminergic nuclei to identify pathways associated with vulnerability in early AD pathological stages.MethodThis cohort includes individuals with no neuropathologic entities and Alzheimer’s disease proteinopathic stages. Individuals with TDP‐43, Lewy Body, or Chronic Traumatic Encephalopathy pathologic entities were excluded. The LC and SN was dissected from frozen human brainstem tissue using macroscopic anatomic features. To compare the LC and SN, only cases below Braak III with sufficient RNA from both nuclei were included. The Nanostring nCounter platform was used to probe specific pathways related to neurodegenerative processes. DBH and SLC6A3 served as confirmatory readouts of accurate anatomic sampling of the LC and SN. Differential gene expression analysis was used to evaluate patterns of gene expression that differentiate the two nuclei. Annotated gene sets were evaluated using global significance scores (GSS), a representation of cumulative t‐statistics for relevant genes.ResultThe LC and SN were collected from 26 cases (Figure 1) eligible for this analysis. Accurate dissections were confirmed by differential expression of DBH (p = 1.38e‐9) and SLC6A3 (p = 4.5e‐16). Among the differentially expressed genes (Figure 2), genes associated with microglia activation (GSS = 2.35), oxidative stress (GSS = 2.17), and lipid metabolism (GSS = 2.00) significantly differentiated the LC and SN.ConclusionThe LC and SN serve as examples of differentially vulnerable nuclei at the earliest neuropathologic stages of AD. Cell intrinsic factors, physiological demands, vascular contributions, and exposure to extrinsic factors like toxins are thought to contribute to the LC’s early vulnerability. Here, we show that neuroinflammation and oxidative stress pathways are likely central to differential vulnerability at the earliest stages of Alzheimer’s disease and may represent targets for intervention in prodromal stages.

In vivo tau deposition in the brain stem nuclei and circadian rest‐activity rhythm change in non‐demented older adults

AbstractBackgroundCircadian rest‐activity rhythm (RAR) alterations are common in the early stage of Alzheimer’s disease (AD) disease. Given that the brainstem nuclei, the locus coeruleus (LC) and dorsal raphe nucleus (DRN) in particular, play key roles in modulating arousal and the sleep‐wake cycle, tau pathology in the brain stem nuclei, which occurs even before cortical tau progression, may relate to circadian RAR alteration in older individuals. Therefore, we aimed to test the hypothesis that in vivo tau deposition in the LC and DRN correlates with circadian RAR in non‐demented older adults.MethodA total of 63 non‐demented older adults from the Korean Brain Aging Study for Early Diagnosis and Prediction of Alzheimer’s Disease (KBASE) were included. All participants underwent [11C] PiB PET and [18F] AV1451‐PET. Circadian RAR variables were measured using actigraphy for 8 consecutive days. Tau deposition of the LC and DRN (LC‐tau and DRN‐tau, respectively) was measured in the regions of interest (ROI) masks for each participant prepared in native space using the Harvard arousal network atlas for the LC and DRN. For the purpose of comparison, tau deposition in the substantia nigra (SN‐tau) was also measured in the neuromelanin sensitive MRI‐based ROI for the SN.ResultIncreased LC‐tau and DRN‐tau were significantly correlated with delayed acrophase (ß = 0.303, p = 0.024 for LC‐tau and ß = 0.398, p = 0.003 for DRN‐tau; Table 1 & Figure 1). In contrast, SN‐tau did not related with any RAR variables. Even after controlling for global tau deposition, the significance did not changed (ß = 0.321, p = 0.021 for LC‐tau and ß = 0.367, p = 0.008 for DRN‐tau, Table 1). In addition, Aß positivity showed a significant moderation effect on the relationship between LC‐tau with acrophase. Subgroup analyses showed that increased LC‐tau significantly related with delayed acrophase in Aß‐positivegroup (ß = 0.427, p = 0.004), but did not in Aß‐negative group (ß = 0.099, p = 0.615)(Figure 2)ConclusionOur findings support the possibility that the tau pathology in the LC and DRN underlie circadian RAR alteration, more specifically delayed phase, in non‐demented older adults. In addition, such relationship appears to be more prominent in individuals with brain Aß pathology.

A comparison of machine learning based‐composite cognitive test scores to track cognitive decline in early stages of Alzheimer’s disease dementia: ADOPIC study

AbstractBackgroundClinical trials of early Alzheimer’s disease (AD) dementia use cognition as a primary outcome and therefore cognitive measures that accurately reflect disease progression and represent multiple cognitive domains are required. Current cognitive endpoints are computed by averaging standardized change from baseline scores (i.e., preclinical Alzheimer cognitive composite (PACC)). Here we compare the PACC to composite scores for which the combination of multiple performance scores has been optimized using machine learning (ML) algorithms in a large, harmonised data set, namely ADOPIC.MethodsA dataset harmonised across various cognitive scores (Table 1) was used to construct composite scores using ML‐based algorithms: a manifold learning dimension reduction technique (UMAP), principal component analysis (PCA) and Latent variable analysis (LVA). Data from ADNI (n = 1470), AIBL (n = 1105) and OASIS participants (n = 412, Table 2) with ≥3 assessments ≤5 years before clinical progression/last visit were included. Participants were classified clinically as; stable cognitively unimpaired (CU), CU progressing to mild cognitive impairment (MCI) or dementia, stable MCI, MCI progressing to dementia, or AD dementia. For UMAP, all stable participants (including dementia) were used in 4‐fold cross‐validation training sets. However, for testing UMAP and unsupervised models (PCA and LVA) all clinical groups (except for dementia) were used. The validity of ML‐based composites was challenged by modelling mean (SD) change over time in the non‐demented clinical groups, using linear mixed model (LMM) analysis. Signal‐to‐noise ratios (SNRs) were calculated (mean change/SD change) for each composite. The mean change was measured in progressives relative to stable participants of each group.ResultsEach ML‐based cognitive composite showed sensitivity to cognitive decline in the progressor groups (Figure 1 A). For the MCI progressors, PCA and UMAP composites had significantly higher SNRs than PACC (P<0.01; Figure 1B), however, LVA performance was not significantly better than PACC. For the CU progressor group, SNRs for PCA and LVA did not show significant differences with PACC and UMAP performed significantly worse than PACC (P<0.01).ConclusionML‐based cognitive composite score computed using PCA provides a practical solution to track cognitive decline with improved performance in tracking cognitive decline in MCI progressors compared to PACC, while being comparable in CUs.

Age‐related and AD pathology‐related differences in neural activation and neural specificity in memory

AbstractBackgroundBeta‐amyloid (Aß) and tau deposition is spatially selective in the early stages of Alzheimer’s Disease (AD), with Aß depositing predominately in posterior‐medial (PM) regions and tau accumulating in the medial temporal lobe (MTL) with increasing age before spreading to anterior‐temporal (AT) regions. Functionally, AT and PM are linked to object and scene processing, respectively, and the hippocampus in the MTL is important for integration and forming associations between objects and scenes. Given the differential vulnerability of AT, PM, and MTL to Aß and tau, we examined the behavioral and neural differences in memory for objects, scenes, and object‐in‐scene pairs between young and cognitively normal older adults with and without AD pathologies.MethodTwenty‐three young (19‐34 yrs) and 32 older (60‐91 yrs) participants completed an incidental encoding task on object, scene, and object‐in‐scene pair images during fMRI and a recognition test outside the scanner (Fig.1). Thirty older participants underwent PiB and FTP PET that measured Aß and tau. We conducted univariate whole‐brain analysis and multi‐voxel pattern analysis (MVPA) of AT, PM, and the hippocampus. The MVPA trains a support vector machine to classify the stimulus category (object, scene, pair) based on the ROI’s activation pattern. A higher classification accuracy represents informative neural representation that reflects functional specificity of the region.ResultBehaviorally, older people had worse object (ps<.027) and pair (ps<.017) memory. Greater temporal meta‐ROI tau was related to worse object (p = .032) and pair (p = .022) memory. Whole‐brain analyses showed activations in AT and PM networks for object and scene processing, and additional frontal, temporal, and occipital regions for pair processing (Fig.2A‐C) where older adults, compared to younger adults, showed activation reductions in fusiform, parahippocampal, precuneus, and occipital regions (Fig.2D). The MVPA revealed age‐related reductions in classification accuracy for pairs (Fig.3A) and that greater Aß and tau were related to lower accuracy in the hippocampus (Fig.3B).ConclusionWe found evidence supporting specific vulnerabilities in object and associative memory in older adults correlated with tau pathology. FMRI findings suggest both lower neural activation and lower neural specificity in older adults when processing complex associations. AD pathology appears to contribute to lower neural specificity in the hippocampus.

Naturalistic Assessment of Reaction Time Variability in Older Adults at Risk for Alzheimer’s Disease

AbstractBackgroundBreakdowns in attentional control may underscore cognitive changes seen in the earliest stages of Alzheimer’s disease (AD). One way to measure these changes is to assess variability in cognitive performance. Variability in cognition may be most evident when repeatedly sampled over hours or days and in different environmental contexts or at different times of day. We analyzed data from an ongoing smartphone‐based study of cognition in AD which utilizes extremely brief cognitive tests taken remotely over several days in natural environments.MethodOlder adults (N = 319, aged 75 +/‐ 5 years) completed repeated cognitive testing on the Ambulatory Research in Cognition (ARC) platform and had CSF or PET amyloid biomarkers and APOE genotype available. RTSD was assessed in a brief processing speed task by taking the SD of correct RTs. All models adjusted for baseline age, education, and gender.ResultAveraging across sessions, symptomatic participants (indicated by Clinical Dementia Rating® (CDR®) > 0) showed more variability than asymptomatic (CDR 0) participants (p = 0.009). APOE ε4 carriers showed more variability than non‐carriers, although this was not significant (p = 0.067). More fine‐grained models considered RTSD within each session (e.g. from trial‐to‐trial) using mixed effects models. Symptomatic participants exhibited greater trial‐to‐trial RTSD compared to asymptomatic participants (p = 0.001). APOE ε4 carriers also exhibited greater trial‐to‐trial RTSD compared to non‐carriers (p = 0.004). When limiting the sample to asymptomatic participants to determine if there were changes in the preclinical stage of AD, there remained a strong APOE ε4 effect, with carriers exhibiting greater trial‐to‐trial RTSD than non‐carriers (p = 0.005). Interestingly, amyloid was not significant in any models.ConclusionOlder adults at risk for AD show increased RTSD even in brief assessments. These subtle fluctuations in attention were most evident in APOE e4 carriers at very short time intervals, and less so when averaged across trials and sessions. Overall, these findings suggest that fluctuations in attention in preclinical AD are most evident from moment to moment, highlighting the need for fine‐grained and repeated assessment of attentional control.

In vivo tau deposition in the brain stem nuclei and circadian rest‐activity rhythm change in non‐demented older adults

AbstractBackgroundCircadian rest‐activity rhythm (RAR) alterations are common in the early stage of Alzheimer’s disease (AD) disease. Given that the brainstem nuclei, the locus coeruleus (LC) and dorsal raphe nucleus (DRN) in particular, play key roles in modulating arousal and the sleep‐wake cycle, tau pathology in the brain stem nuclei, which occurs even before cortical tau progression, may relate to circadian RAR alteration in older individuals. Therefore, we aimed to test the hypothesis that in vivo tau deposition in the LC and DRN correlates with circadian RAR in non‐demented older adults.MethodA total of 63 non‐demented older adults from the Korean Brain Aging Study for Early Diagnosis and Prediction of Alzheimer’s Disease (KBASE) were included. All participants underwent [11C] PiB PET and [18F] AV1451‐PET. Circadian RAR variables were measured using actigraphy for 8 consecutive days. Tau deposition of the LC and DRN (LC‐tau and DRN‐tau, respectively) was measured in the regions of interest (ROI) masks for each participant prepared in native space using the Harvard arousal network atlas for the LC and DRN. For the purpose of comparison, tau deposition in the substantia nigra (SN‐tau) was also measured in the neuromelanin sensitive MRI‐based ROI for the SN.ResultIncreased LC‐tau and DRN‐tau were significantly correlated with delayed acrophase (β = 0.303, p = 0.024 for LC‐tau and β = 0.398, p = 0.003 for DRN‐tau; Table 1 & Figure 1). In contrast, SN‐tau did not related with any RAR variables. Even after controlling for global tau deposition, the significance did not changed (β = 0.321, p = 0.021 for LC‐tau and β = 0.367, p = 0.008 for DRN‐tau, Table 1). In addition, Aβ positivity showed a significant moderation effect on the relationship between LC‐tau with acrophase. Subgroup analyses showed that increased LC‐tau significantly related with delayed acrophase in Aβ‐positivegroup (β = 0.427, p = 0.004), but did not in Aβ‐negative group (β = 0.099, p = 0.615)(Figure 2)ConclusionOur findings support the possibility that the tau pathology in the LC and DRN underlie circadian RAR alteration, more specifically delayed phase, in non‐demented older adults. In addition, such relationship appears to be more prominent in individuals with brain Aβ pathology.

The electronic Person Specific Outcome Measure (ePSOM) development program in the US: A survey to understand what matters to individuals most about their brain health

AbstractBackgroundIt is essential both drug and lifestyle‐based interventions aimed to delay the onset of advanced cognitive decline deliver a meaningful outcome for the patient. In the early stages of Alzheimer’s disease and related dementias (ADRDs), patient‐reported outcome (PRO) measures should be used in parallel with biological investigations of ADRDs to provide a complementary endpoint offering further proof of intervention’s effectiveness from the patients’ point of view. The aim of the electronic Person‐Specific Outcome Measure (ePSOM) programme is to develop a scalable, easily accessible and efficient PRO tool for monitoring personally meaningful outcomes in early ADRD.MethodWe designed and ran an online nation‐wide study to understand what matters to people about their brain health in the US. The ePSOM US survey started in Jan 2023 and collects primarily free‐text responses along with sociodemographic information and self‐reported neurodegenerative disease diagnosis. Respondents are presented with five domains and asked to describe in their own words their brain‐health priorities. Free‐text data will be analysed using Natural Language Processing (NLP) techniques to identify the most common brain‐health priorities by key sociodemographic groups. We will also compare these results with previously published UK results to look for any notable differences between the US and UK respondents’ priorities.ResultThe survey aims to recruit n = 6,000 respondents. Data collection started early 2023 and is planned to be complete by Spring/Summer 2023. The previously published ePSOM UK study (Saunders et al., 2021) with n = 5,808 respondents resulted in over 80,000 free‐text responses and 184 unique priority themes about brain health. We will present the preliminary results of the US study at AAIC and contextualising the results in light of new pharmacological treatments approved for AD.ConclusionThis large‐scale population‐based study will offer an evidence base for what outcomes constitute a meaningful impact of new treatments. Previous work from our group suggests that brain health priorities shift along the preclinical, prodromal and dementia continuum. The ultimate aim of the ePSOM programme is to design an electronic outcomes app which would allow an individual to define the most important outcomes against which we should measure treatment success across a range of severities.

Exploratory research on Neurogranin ratios for biomarkers of neurodegeneration

AbstractBackgroundNeurogranin (Ng) is a cerebrospinal fluid biomarker for synaptic dysfunction at early stage of Alzheimer’s disease (AD). It has been reported that some truncated forms from Ng is increased in AD brain tissue. However, the clinical significance of Ng forms in plasma has remained unclear. We have developed a simultaneous detection method of a plurality of Ng forms in plasma by using immunoprecipitation combined with matrix‐assisted laser desorption ionization time‐of‐flight mass spectrometry (IP‐MALDI‐MS). In this study, we applied the IP‐MALDI‐MS technique to investigate blood‐based biomarkers for neurodegeneration.MethodPlasma samples were obtained from cognitively normal (CN), mild cognitive impairment (MCI), Alzheimer’s dementia (AD) and non‐Alzheimer’s dementia (non‐AD) individuals in National Center for Geriatrics and Gerontology. These subjects were classified as Aβ positive and negative based on PET with PIB tracer. Ng forms were immunoprecipitated from the samples spiked by stable isotope‐labeled (SIL) Ng peptides using magnetic beads coupled with anti‐Ng monoclonal antibody. After the IP was performed two times consecutively, Ng forms eluted from antibody beads were applied on the MALDI plate. SIL‐Ng peptides were used as internal standards for normalizing an intensity of each endogenous Ng form in plasma.ResultWe calculated intensity ratio of a Ng form‐to‐other Ng form (Ng ratio) in all combinations of peaks corresponding to mass of Ng forms. To investigate the possibility of Ng as a biomarker for neurodegeneration, we screened the Ng ratios with the difference between CN and Aβ positive AD or Aβ negative non‐AD. Eleven Ng ratios were significantly higher in the Aβ positive AD and Aβ negative non‐AD groups than the CN group. In ROC analyses, AUC values greater than 0.750 were found in discriminating the Aβ positive AD or Aβ negative non‐AD group from the CN group. Among the eleven Ng ratios, five were significantly correlated with MMSE (ρ←0.400).ConclusionSome Ng forms cleaved at or nearly the IQ domain have been found in the AD brain, which suggests the function loss of Ng in synaptic plasticity. Increasing Ng ratios in plasma may be biomarker candidates for synaptic dysfunction involved in the change of Ng forms in the brain.

Amyloid predicts longitudinal atrophy in non‐demented individuals: Results from the AMYPAD Prognostic & Natural History study

AbstractBackgroundWhile the association between tau deposition on PET and brain atrophy on MRI is well documented in the literature, the relation between amyloid deposition and neurodegeneration in the earliest stages of Alzheimer’s disease (AD) is less studied. We investigated this relationship in a large cohort of non‐demented individuals.MethodWe included 1355 participants from the AMYPAD Prognostic & Natural History study (PNHS) with available MRI and amyloid‐PET. Among those, 753 had longitudinal MRI, with a mean follow‐up time of 3.47 years (SD = 1.55). The FreeSurfer 7.0.1 longitudinal pipeline was used to measure gray matter (GM) thickness and volumes in 40 regions (Reuter et al., 2012; Desikan et al., 2006). From PET scans, cortical amyloid burden was assessed globally using the Centiloid (CL) method. MRI‐derived regional volumes and thickness were harmonized across sites with neuroCombat (Fortin et al., 2018); all variables were Z‐scored.Linear mixed‐effect models were used to investigate the effect of amyloid burden at baseline and its interaction with time on regional volume and thickness measurements. Models were corrected for age, sex, and CDR. P‐values were FDR‐adjusted.ResultCohort characteristics are shown in Table 1. Decreasing cortical thickness and volumes were observed over time in most regions. At baseline, higher amyloid burden was related to smaller volumes of the hippocampus, amygdala, putamen, and lateral ventricles; and thinner entorhinal cortex, precuneus and posterior cingulate cortex. At follow‐up, subjects with higher baseline amyloid burden showed greater loss of volume and thickness in mainly temporal and parietal regions, as well as amygdala and hippocampal volume (Figure 1).ConclusionIn the largely asymptomatic AMYPAD PNHS cohort, we show that amyloid burden at baseline is predictive of future neurodegeneration, especially affecting temporal volume and parietal thickness. Cortical thickness was slightly more sensitive towards amyloid burden at baseline, while loss of cortical volume was greater than loss of thickness.

Network changes and compensation mechanisms of wake‐promoting neurons across AD continuum

AbstractBackgroundSleep is a complex process regulated by various neural mechanisms, including those promoting wakefulness, those promoting sleep, and those regulating the circadian clock. Recent research has revealed that wake‐promoting neurons, specifically those found in the noradrenergic locus ceruleus (LC), the orexinergic (ORX) neurons in the lateral hypothalamic area (LHA), and the histaminergic neurons in the hypothalamic tuberomammillary nucleus (TMN), are particularly vulnerable to the AD‐tau pathology found in the early stages of Alzheimer’s disease (AD). These studies have uncovered how these neurons accumulate AD‐tau, the extent to which they die in AD, and specific correlations with polysomnographic readings. Notably, a complex pattern of neuronal dysregulation within and among these wake‐promoting neurons has been observed, suggesting the existence of robust compensatory mechanisms in an attempt to overcome dysfunction. Thus, understanding the vulnerability of these nuclei to AD requires network‐based studies. These studies are crucial for informing disease‐specific treatment for sleep disturbance and identifying factors underlying the vulnerability of these neurons.MethodsResults from several ongoing experiments in the Grinberg Lab, including methods such as quantitative pathology, molecular profiling and clinicopathological correlationResultsFor example, the hypocretin (orexin) receptor 1 (HCRTR1) shows upregulation in orexin neurons from early AD stages, while no changes are observed in the LC or TMN. Additionally, while the LC shows an early pattern of upregulation of genes associated with neuroinflammation, early dysregulation in the ORX is related to synapse‐related genes. Furthermore, circadian genes such as CRY1 and CRY2 show upregulation in the ORX but downregulation in the TMN during AD progression.ConclusionThis talk will focus on current knowledge, including unpublished data, based on studies in humans, on the vulnerability of wake‐promoting neurons in AD, with a particular emphasis on similarities and contrasts in the patterns of dysregulation among these nuclei during AD progression, in conjunction with changes in the suprachiasmatic nucleus, the circadian clock, and regional expression of circadian genes.

N6-methyladenosine-modified circRIMS2 mediates synaptic and memory impairments by activating GluN2B ubiquitination in Alzheimer's disease

BackgroundSynaptic degeneration occurs in the early stage of Alzheimer's disease (AD) before devastating symptoms, strongly correlated with cognitive decline. Circular RNAs (circRNAs) are abundantly enriched in neural tissues, and aberrant expression of circRNAs precedes AD symptoms, significantly correlated with clinical dementia severity. However, the direct relationship between circRNA dysregulation and synaptic impairment in the early stage of AD remains poorly understood.MethodsHippocampal whole-transcriptome sequencing was performed to identify dysregulated circRNAs and miRNAs in 4-month-old wild-type and APP/PS1 mice. RNA antisense purification and mass spectrometry were utilized to unveil interactions between circRIMS2 and methyltransferase 3, N6-adenosine-methyltransferase complex catalytic subunit (METTL3). The roles of circRIMS2/miR-3968 in synaptic targeting of UBE2K-mediated ubiquitination of GluN2B subunit of NMDA receptor were evaluated via numerous lentiviruses followed by morphological staining, co-immunoprecipitation and behavioral testing. Further, a membrane-permeable peptide was used to block the ubiquitination of K1082 on GluN2B in AD mice.ResultscircRIMS2 was significantly upregulated in 4-month-old APP/PS1 mice, which was mediated by METTL3-dependent N6-methyladenosine (m6A) modification. Overexpression of circRIMS2 led to synaptic and memory impairments in 4-month-old C57BL/6 mice. MiR-3968/UBE2K was validated as the downstream of circRIMS2. Elevated UBE2K induced synaptic dysfunction of AD through ubiquitinating K1082 on GluN2B. Silencing METTL3 or blocking the ubiquitination of K1082 on GluN2B with a short membrane-permeable peptide remarkably rescued synaptic dysfunction in AD mice.ConclusionsIn conclusion, our study demonstrated that m6A-modified circRIMS2 mediates the synaptic and memory impairments in AD by activating the UBE2K-dependent ubiquitination and degradation of GluN2B via sponging miR-3968, providing novel therapeutic strategies for AD.

A slice selection guided deep integrated pipeline for Alzheimer’s prediction from Structural Brain MRI

Alzheimer’s disease, a progressive form of dementia, has risen to become the fifth leading cause of death among individuals aged 65 and older. The diagnosis of Alzheimer’s is both time-consuming and costly, involving radiologists and clinical experts at multiple stages, which presents a significant challenge in the medical field. Moreover, cases of Alzheimer’s and dementia often go undiagnosed or misdiagnosed worldwide. To address this issue, medical experts meticulously analyze patients’ structural MRI (sMRI) scans to identify potential abnormalities linked to Alzheimer’s or other forms of dementia. Recognizing the devastating impact of this disease on people’s lives, Artificial Intelligence (AI) researchers have been dedicated to developing automated solutions for early-stage Alzheimer’s diagnosis in recent years, aiming to support medical practitioners in their efforts. Despite the application of various AI-driven solutions that use sMRI data for Alzheimer’s diagnosis, there are still research gaps that need attention. These gaps include the need for guided slice selection and the development of a simpler yet effective integrated pipeline where each stage of the process is fully automated, eliminating the need for medical practitioner intervention. In this study, we propose an integrated automated solution that incorporates a guided machine learning-based selection process using K-Means++ leading to a Gradient Boosting-based method for identifying the 16 most relevant 2-dimensional sMRI slices from 3-dimensional sMRI data. This step is crucial for accurate Alzheimer’s classification. Furthermore, we introduce a deep learning architecture that combines EfficientNetV2S-based transfer learning with densely-learned features in an optimized manner. To evaluate the effectiveness of our proposed deep-integrated architecture, we used two benchmark datasets from ADNI and OASIS, conducting rigorous experimental analysis and validation. The results demonstrated that our integrated architecture outperformed all other experimented architectures, achieving a 20-Fold Cross Validation Accuracy of 83.64% (CN vs AD), 82.69% (CN vs MCIc), and 71.40% (CN vs MCInc) on the ADNI dataset, and 91.54% (CN vs AD) on the OASIS dataset. This signifies the potential of our approach in improving Alzheimer’s diagnosis accuracy and offers hope for early detection and intervention in this debilitating disease.

A Comprehensive Study on Early Alzheimer’s Disease Detection through Advanced Machine Learning Techniques on MRI Data

Alzheimer’s Disease (AD) is a neurodegenerative condition affecting predominantly elderly individuals, repre- senting the most common cause of dementia. Early clinical manifestations of AD include selective memory impairment, and while certain symptomatic improvements can be achieved through treatment, there is currently no cure. Magnetic Resonance Imaging (MRI) is utilized for brain imaging to assess suspected AD patients, providing results that include local and global brain atrophy. Some studies suggest that MRI features can predict the rate of AD decline and guide future treatments. However, to reach this stage, clinicians and researchers must employ machine learning techniques for accurate prediction of the progression from mild cognitive impairment to dementia. We propose the development of a reliable model to assist clinicians in achieving this and predicting early-stage Alzheimer’s Disease.

Clinical Features of Subjective Cognitive Decline in The Early Stages of Alzheimer’s Disease

Subclinical stage of the disease precedes the clinical stage of moderate cognitive decline in Alzheimer's disease (AD). Subjective cognitive decline (SCD) — a condition in which the level of cognitive function habitual for the subject gradually begins to decrease. In 2021, researchers from the National Institute on Aging and the Alzheimer's Association (NIA-AA) clinical criteria for the diagnosis of SCD have been proposed, as well as features that increase the probability of preclinical stage AD in patients with moderate cognitive impairment have been identified. Patients should be offered a complex of examinations — questionnaires regarding the impact of memory impairment on current cognitive activity (forgetfulness, searching for things, difficulty finding words, etc.), testable self-report of cognitive dynamics, neuropsychological testing and diagnosis of pathopsychological changes such as depression and anxiety. It would appear that counselling in the form of interviews and/or testing of persons able to provide relevant information about the patient should be included in the examination of patients with complaints of memory disorders, regardless of their degree of severity. It may be necessary to conduct a survey on the patient’s daily activity, ability to self-service (score, orientation, planning, control and so on), as well as to obtain information about any memory-related changes that have become visible to others, because it is the data from the partner/relative that increase the predictive value of the diagnostic. The modern approach to the study of cognitive functions in elderly people without dementia in the long-term is certainly able to help identify people with a high risk of developing AD.

1 Detection of Cognitive Subtypes Within Cognitively Normal Older Adults Using Hierarchical Community Detection

Objective:Identifying individuals at the earliest stages of Alzheimer’s Disease (AD) would enable development and study of interventions prior to onset of symptoms. However, differentiating age-related cognitive changes from subtle pathological changes remains a challenge in the field. Methods that would enable earlier detection of AD in elders with no subjective or objective cognitive concerns (i.e., individuals in the preclinical stage) would be of great interest. Community detection, a metric founded in graph theory, offers an alternative approach for characterizing subtle heterogeneity within aging samples and has the potential to inform cognitive variability in aging.Participants and Methods:Using a hierarchical community detection, we examined whether cognitive subtypes could be identified in 226 cognitively normal older adults (from the Alzheimer’s Disease Neuroimaging Initiative [ADNI] study). Cognitive profiles of each community were characterized first using MANOVAs to examine the relationship between community membership and 12 age-, gender-, and education-corrected neuropsychological variables. Pair-wise comparisons were examined for significant main effects. We then examined whether these subtypes were related to biomarkers (cortical volumes, fluorodeoxyglucose (FDG)-positron emission tomography (PET) hypometabolism) or clinical progression. All p values were corrected for multiple comparisons.Results:Three communities (i.e., cognitive subtypes) were identified within the healthy aging sample. The first and largest community identified (N = 106) was characterized by a relative weakness on a single measure visuospatial executive function. Both the second (N = 76) and third community (N = 44) scored significantly lower on immediate, delayed, and recognition memory relative to the first community. The third community was characterized by a relative weakness in category fluency and speeded visual sequencing as well (p < .000). The three communities did not differ on age, gender, education, race, or ethnicity. Community membership was associated with entorhinal volume (with the second and third communities having significantly smaller entorhinal volumes than the first community), though community membership was not significantly associated with other biomarkers examined. Conversion rate reached trend level significance at 12 month follow up (more converters in the third community).Conclusions:Hierarchical community detection is an alternative method for characterizing neuropsychological variation and it appears sensitive to relatively small differences that may be observed in a normal aging sample. While the sample size was relatively small, this approach shows promise for potentially leading to earlier detection of cognitive decline among individuals classified to be aging normally (e.g., community 3).

2 Cognitive Heterogeneity and Risk of Progression in Data-Driven Subtle Cognitive Decline Phenotypes

Objective:There is increasing recognition of cognitive and pathological heterogeneity in early-stage Alzheimer’s disease and other dementias. Data-driven approaches have demonstrated cognitive heterogeneity in those with mild cognitive impairment (MCI), but few studies have examined this heterogeneity and its association with progression to MCI/dementia in cognitively unimpaired (CU) older adults. We identified cluster-derived subgroups of CU participants based on comprehensive neuropsychological data and compared baseline characteristics and rates of progression to MCI/dementia or a Dementia Rating Scale (DRS) of <129 across subgroups.Participants and Methods:A hierarchical cluster analysis was conducted using 11 baseline neuropsychological test scores from 365 CU participants in the UCSD Shiley-Marcos Alzheimer’s Disease Research Center (age M=71.93 years, SD=7.51; 55.9% women; 15.6% Hispanic/Latino/a/x/e). A discriminate function analysis was then conducted to test whether the individual neuropsychological scores predicted cluster-group membership. Cox regressions examined the risk of progression to consensus diagnosis of MCI or dementia, or to DRS score <129, by cluster group.Results:Cluster analysis identified 5 groups: All-Average (n=139), Low-Visuospatial (n=46), Low-Executive (n=51), Low-Memory/Language (n=83), and Low-All Domains (n=46). The discriminant function analysis using the neuropsychological measures to predict group membership into these 5 clusters correctly classified 85.2% of the participants. Subgroups had unique demographic and clinical characteristics. Relative to the All-Average group, the Low-Visuospatial (hazard ratio [HR] 2.39, 95% CI [1.03, 5.56], p=.044), Low-Memory/Language (HR 4.37, 95% CI [2.24, 8.51], p<.001), and Low-All Domains (HR 7.21, 95% CI [3.59, 14.48], p<.001) groups had greater risk of progression to MCI/dementia. The Low-Executive group was also twice as likely to progress to MCI/dementia compared to the AllAverage group, but did not statistically differ (HR 2.03, 95% CI [0.88,4.70], p=.096). A similar pattern of results was found for progression to DRS score <129, with the Low-Executive (HR 2.82, 95% CI [1.26, 6.29], p=.012), Low-Memory/Language (HR 3.70, 95% CI [1.80, 7.56], p<.001) and Low-All Domains (HR 5.79, 95% CI [2.74, 12.27], p<.001) groups at greater risk of progression to a DRS score <129 than the All-Average group. The Low-Visuospatial group was also twice as likely to progress to DRS <129 compared to the All-Average group, but did not statistically differ (HR 2.02, 95% CI [0.80, 5.06], p=.135).Conclusions:Our results add to a growing literature documenting heterogeneity in the earliest cognitive and pathological presentations associated with Alzheimer’s disease and related disorders. Participants with subtle memory/language, executive, and visuospatial weaknesses all declined at faster rates than the All-Average group, suggesting that there are multiple pathways and/or unique subtle cognitive decline profiles that ultimately lead to a diagnosis of MCI/dementia. These results have important implications for early identification of individuals at risk for MCI/dementia. Given that the same classification approach may not be optimal for everyone, determining profiles of subtle cognitive difficulties in CU individuals and implementing neuropsychological test batteries that assess multiple cognitive domains may be a key step towards an individualized approach to early detection and fewer missed opportunities for early intervention.

Prevalence and localization of nocturnal epileptiform discharges in mild cognitive impairment.

Recent evidence shows that identifying and treating epileptiform abnormalities in patients with Alzheimer's disease could represent a potential avenue to improve clinical outcome. Specifically, animal and human studies have revealed that in the early phase of Alzheimer's disease, there is an increased risk of seizures. It has also been demonstrated that the administration of anti-seizure medications can slow the functional progression of the disease only in patients with EEG signs of cortical hyperexcitability. In addition, although it is not known at what disease stage hyperexcitability emerges, there remains no consensus regarding the imaging and diagnostic methods best able to detect interictal events to further distinguish different phenotypes of Alzheimer's disease. In this exploratory work, we studied 13 subjects with amnestic mild cognitive impairment and 20 healthy controls using overnight high-density EEG with 256 channels. All participants also underwent MRI and neuropsychological assessment. Electronic source reconstruction was also used to better select and localize spikes. We found spikes in six of 13 (46%) amnestic mild cognitive impairment compared with two of 20 (10%) healthy control participants (P = 0.035), representing a spike prevalence similar to that detected in previous studies of patients with early-stage Alzheimer's disease. The interictal events were low-amplitude temporal spikes more prevalent during non-rapid eye movement sleep. No statistically significant differences were found in cognitive performance between amnestic mild cognitive impairment patients with and without spikes, but a trend in immediate and delayed memory was observed. Moreover, no imaging findings of cortical and subcortical atrophy were found between amnestic mild cognitive impairment participants with and without epileptiform spikes. In summary, our exploratory study shows that patients with amnestic mild cognitive impairment reveal EEG signs of hyperexcitability early in the disease course, while no other significant differences in neuropsychological or imaging features were observed among the subgroups. If confirmed with longitudinal data, these exploratory findings could represent one of the first signatures of a preclinical epileptiform phenotype of amnestic mild cognitive impairment and its progression.

48 Psychometric properties of DCTclock™ with commonly used neuropsychological tests and their combined ability to predict Beta-Amyloid Positron Emission Tomography Status

Objective:Sensitive and non-invasive methods of screening for early-stage Alzheimer’s disease (AD) are urgently needed. The digital clock drawing test (DCTclockTM) is an established and well-researched neuropsychological tool that can aid in early detection of dementia. Other simple, yet sensitive, neuropsychological measures able to detect early stages of AD include Trail Making Tests (TMT). We investigated the psychometric properties of DCTclockTM with TMT-A and TMT-B. We then sought to understand the degree to which neuropsychological tools (i.e., DCTclockTM, TMT-A, and B) versus the Montreal Cognitive Assessment (MoCA) predict beta-amyloid (Aß) positron emission tomography (PET) status (positive or negative) in cognitively normal individuals.Participants and Methods:Participants included a sample of cognitively normal older adults (n= 59, M age = 69.2, F = 64%) recruited from the Butler Memory and Aging Program. The Linus Health DCTclockTM uses a digital pen to capture traditional clock drawing test performance and advanced analytics to evaluate the drawing process for indicators of cognitive difficulty. DCTclockTM may have overlapping cognitive properties with TMT measures, like efficiency, processing speed, and spatial reasoning. We compared latency measures (i.e., process efficiency, clock face speed, average latency, and processing speed) and spatial reasoning of the DCTclockTM to z-scores of TMT-A and TMT-B to detect any overlapping psychometric properties. Verbal fluency was included for discriminant validity. We then ran logistic regressions on a subset of the sample to compare neuropsychological tests (DCTclockTM total score [score that captures overall performance], TMT-A/B, and verbal fluency) to the MoCA, a commonly used cognitive screening tool, in determining PET status.Results:Highly correlated (r > .7) DCTclockTM variables were excluded. We found statistically significant correlations between some DCTclockTM measures and TMT-A/B, like DCTclockTM drawing process efficiency and TMT-A and TMT-B (r= .45, p< .001, r=.29, p< .026, respectively), and DCTclockTM average latency and TMT-A and TMT-B (r=.3, p< .024, r= .26, p< .044, respectively). No statistically significant associations were found between any DCTclockTM measures and verbal fluency, or between DCTclockTM spatial reasoning and TMT-A/B. We then investigated the effect of these neuropsychological tests (DCTclockTM total score, TMT-A/B, verbal fluency) and age on the likelihood of PET positivity (subset of sample, total PET, n=31). The model was statistically significant (x2 (5) = 15.35, p< .01). The model explained 53% (Nagelkerke R2) of the variance in PET status and correctly classified 74.2% of cases. DCTclockTM was the only significant predictor (p< .02), after controlling for TMT-A, TMT-B, verbal fluency, and age. Comparatively, there was no effect of MoCA and age (total PET, n= 29) on the likelihood of PET positivity.Conclusions:Overall, these results suggest psychometric convergence on elements of DCTclockTM and TMT-A/B, while there was no association in spatial operations between DCTclockTM and TMT measures. Further, when compared to the MoCA, DCTclockTM and these commonly used neuropsychological tests (verbal fluency and TMT-A/B) were better predictors of PET status, primarily driven by the DCTclockTM. Digitized neuropsychological tools may provide additional metrics not captured by pen-and-paper tests that can detect AD-associated pathology.

STREM2 is associated with attenuated tau aggregate accumulation in the presence of amyloid-β pathology.

Triggering Receptor Expressed on Myeloid Cell 2 (TREM2) plays a crucial role in the transition of microglia from a state of homeostasis to a state associated with the disease. Mutations in TREM2 are strongly linked with a higher risk of developing neurodegenerative diseases, including Alzheimer's disease. There have been contradictory findings regarding the potential detrimental or protective effects of microglial activation and TREM2-related microglial responses in Alzheimer's disease. Although previous studies reported increased CSF soluble TREM2 (sTREM2) in different clinical stages of Alzheimer's disease, the exact association between Alzheimer's disease hallmarks such as amyloid-beta and tau pathology remains unclear. In the present study, I aimed to investigate the association between TREM2-related microglial responses and tau accumulation in the presence and absence of amyloid-beta pathology in order to give a better view of the role of microglial activation in Alzheimer's disease development. Imaging data of 178 non-demented participants including 107 amyloid-beta-negative participants, 71 amyloid-beta-positive were recruited from Alzheimer's disease Neuroimaging Initiative. The CSF sTREM2 was used as an in vivo indicator of microglial responses associated with TREM2. Furthermore, I used longitudinal tau-PET and resting-state functional MRI connectomes in order to investigate the association of TREM2-related microglial activation and tau spreading through functional connections. A higher level of sTREM2 was associated with slower tau aggregate accumulation in non-demented amyloid-beta-positive. Furthermore, measuring the tau spreading through inter-connected regions using functional MRI connectomes confirms that the TREM2-related microglial activity might be a protective factor against tau pathology in brain tissue. These findings demonstrate that in individuals with initial amyloid-beta abnormalities, TREM2-related microglial activation is linked to reduced regional accumulation of tau aggregates and also, spreading across inter-connected brain regions, as evaluated through functional MRI connectomes during the early stages of Alzheimer's disease.

UNDERSTANDING THE MAIN MEMORY DEFICITS IN THE EARLY STAGE OF ALZHEIMER'S DISEASE

This work aims to present an understanding of the main memory deficits in the early stages of Alzheimer's disease. For this, a bibliographical research was carried out in the Google Scholar database with keywords such as "early stage of Alzheimer's disease", "semantic memory" and "episodic memory". We explored the complexity of the disease, characterized by neurodegeneration and brain atrophy, with memory loss as a prominent symptom. We analyzed the phases of the disease, focusing on the initial phase subdivided into three stages. We detail each stage, highlighting its characteristics. We address biological changes, including the formation of senile plaques and neurofibrillary tangles, which disrupt synapses and affect communication between neurons. In the section on memory, we identify the affected types, mainly episodic and semantic memory, highlighting their importance in the early stages. We explore the relationship between biological changes and memory deficits. We conclude by emphasizing the complexity of the disease, the need for continuous research and the importance of early diagnosis. We discuss challenges in formulating universal treatments due to individual variation and societal changes. We highlight the urgency of future research to face the challenges of this devastating disease.