Disease Neuroimaging Initiative Cohort Research Articles

Association of β-Amyloid, Microglial Activation, Cortical Thickness, and Metabolism in Older Adults Without Dementia.

Plasma β-amyloid42 (Aβ42)/Aβ40 levels have shown promise in identifying Aβ-PET positive individuals. This study explored the concordance and discordance of plasma Aβ42/Aβ40 positivity (Plasma±) with CSF Aβ42/Aβ40 positivity (CSF±) and Aβ-PET positivity (PET±) in older adults without dementia. Associations of Aβ deposition, cortical thickness, glucose metabolism, and microglial activation were also investigated. We selected participants without dementia who had concurrent plasma Aβ42/Aβ40 and Aβ-PET scans from the Alzheimer's Disease Neuroimaging Initiative cohort. Participants were categorized into Plasma±/PET± based on thresholds of composite 18F-florbetapir (FBP) standardized uptake value ratio (SUVR) ≥1.11 and plasma Aβ42/Aβ40 ≤0.1218. Aβ-PET-negative individuals were further divided into Plasma±/CSF± (CSF Aβ42/Aβ40 ≤0.138), and the concordance and discordance of Aβ42/Aβ40 in the plasma and CSF were investigated. Baseline and slopes of regional FBP SUVR were compared among Plasma±/PET± groups, and associations of regional FBP SUVR, FDG SUVR, cortical thickness, and CSF soluble Triggering Receptor Expressed on Myeloid Cell 2 (sTREM2) levels were analyzed. One hundred eighty participants (mean age 72.7 years, 51.4% female, 96 cognitively unimpaired, and 84 with mild cognitive impairment) were included. We found that the proportion of Plasma+/PET- individuals was 6.14 times higher (odds ratio (OR) = 6.143, 95% confidence interval (CI) 2.740-16.185, p < 0.001) than that of Plasma-/PET+ individuals, and Plasma+/CSF- individuals showed 8.5 times larger percentage (OR = 8.5, 95% CI: 3.031-32.974, p < 0.001) than Plasma-/CSF+ individuals in Aβ-PET-negative individuals. Besides, Plasma+/PET- individuals exhibited faster (p < 0.05) Aβ accumulation predominantly in bilateral banks of superior temporal sulcus (BANKSSTS) and supramarginal, and superior parietal cortices compared with Plasma-/PET- individuals, despite no difference in baseline FBP SUVRs. In Plasma+/PET+ individuals, higher CSF sTREM2 levels correlated with slower BANKSSTS Aβ accumulation (standardized β (βstd) = -0.418, 95% CI -0.681 to -0.154, p = 0.002). Conversely, thicker cortical thickness and higher glucose metabolism in supramarginal and superior parietal cortices were associated with faster (p < 0.05) CSF sTREM2 increase in Plasma+/PET- individuals rather than in Plasma+/PET+ individuals. These findings suggest that plasma Aβ42/Aβ40 abnormalities may predate CSF Aβ42/Aβ40 and Aβ-PET abnormalities. Higher sTREM2-related microglial activation is linked to thicker cortical thickness and higher metabolism in early amyloidosis stages but tends to mitigate Aβ accumulation primarily at relatively advanced stages.

Abstract P250: Vascular Contributions to Alzheimer's Disease

Introduction: Alzheimer’s disease (AD) is a debilitating neurodegenerative disorder that affects over 50 million people worldwide. The role of vascular risk factors have been described as a contributor to and a major comorbidity of AD pathogenesis; yet there is discordant evidence of this complex relationship between cardiovascular diseases (CVD) and its risk factors (like hypertension, hyperlipidemia, obesity, etc.) with AD pathophysiology and cognitive decline. AD associated Genome-wide association studies (GWAS) and derived polygenic risk scores (AD PRS) have uncovered some significant common genetic variants relating to CVD risk factors; however, we are still yet to unravel the comprehensive impact of these vascular genetic contributions to AD pathogenesis and cognitive decline. Method: In this study, we used 1800 participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort spanning the full spectrum of clinical and pathological diagnoses of the AD spectrum, as our target dataset used for polygenic risk score calculation. Thereafter, we obtained GWAS summary statistics for cumulative cardiovascular risk by searching the Polygenic Score (PGS) GWAS Catalog using CVD as the ‘mapped trait” and then calculated cardiovascular-based PRS using HRS-based imputed GWAS data of the ADNI cohort. We conducted association analysis using a multivariate linear regression model with appropriate covariates (age, sex, APOE, education, etc.) between CVD-related polygenic risk score (CVD-PRS) with the different clinical AD diagnoses groups (Cognitively Normal/CN, Significant Memory Concerns/SMC, Mild Cognitive Impairment/MCI, Alzheimer's Disease/AD). We also performed association analysis of this CVD-PRS with white matter hyperintensity (WMHI) values, a vascular biomarker, at baseline using a linear regression model. Furthermore, we extended our analyses of CVD-related PRS with cognitive performance at baseline using composite scores for memory (MEM) and executive functioning (EF). Results and Conclusion: We found significantly strong correlation of CVD-related PRS with only certain AD Diagnoses groups (Significant Memory Concerns: SMC and Alzheimer's: AD) in the AD continuum. Additionally, we found strong associations (p &lt; 0.05) of this cardiovascular based- polygenic risk score with baseline memory as well as White Matter Hyperintensity (WMH) in our sample, specifically the group with a clinical diagnosis of AD. Furthermore, we are performing longitudinal association analysis of CVD-related PRS with longitudinal cognitive decline using a linear mixed effects model. Our findings will help us understand 1) how a cumulative CVD-related PRS associates with different AD diagnoses subgroups as well as vascular biomarkers for AD and 2) how CVD-related PRS associates with cognitive performance and longitudinal cognitive decline.

Estimating Dementia Onset: AT(N) Profiles and Predictive Modeling in Mild Cognitive Impairment Patients.

Mild Cognitive Impairment (MCI) usually precedes the symptomatic phase of dementia and constitutes a window of opportunities for preventive therapies. The objective of this study was to predict the time an MCI patient has left to reach dementia and obtain the most likely natural history in the progression of MCI towards dementia. This study was conducted on 633 MCI patients and 145 subjects with dementia through 4726 visits over 15 years from Alzheimer Disease Neuroimaging Initiative (ADNI) cohort. A combination of data from AT(N) profiles at baseline and longitudinal predictive modeling was applied. A data-driven approach was proposed for categorical diagnosis prediction and timeline estimation of cognitive decline progression, which combined supervised and unsupervised learning techniques. A reduced vector of only neuropsychological measures was selected for training the models. At baseline, this approach had high performance in detecting subjects at high risk of converting from MCI to dementia in the coming years. Furthermore, a Disease Progression Model (DPM) was built and also verified using three metrics. As a result of the DPM focused on the studied population, it was inferred that amyloid pathology (A+) appears about 7 years before dementia, and tau pathology (T+) and neurodegeneration (N+) occur almost simultaneously, between 3 and 4 years before dementia. In addition, MCI-A+ subjects were shown to progress more rapidly to dementia compared to MCI-A- subjects. Based on proposed natural histories and cross-sectional and longitudinal analysis of AD markers, the results indicated that only a single cerebrospinal fluid sample is necessary during the prodromal phase of AD. Prediction from MCI into dementia and its timeline can be achieved exclusively through neuropsychological measures.

Multi-Task Learning and Sparse Discriminant Canonical Correlation Analysis for Identification of Diagnosis-Specific Genotype-Phenotype Association.

The primary objective of imaging genetics research is to investigate the complex genotype-phenotype association for the disease under study. For example, to understand the impact of genetic variations over the brain functions and structure, the genotypic data such as single nucleotide polymorphism (SNP) is integrated with the phenotypic data such as imaging quantitative traits. The sparse models, based on canonical correlation analysis (CCA), are popular in this area to find the complex bi-multivariate genotype-phenotype association, as the number of features in genotypic and/or phenotypic data is significantly higher as compared to the number of samples. However, the sparse CCA based methods are, in general, unsupervised in nature, and fail to identify the diagnose-specific features those play an important role for the diagnosis and prognosis of the disease under study. In this regard, a new supervised model is proposed to study the complex genotype-phenotype association, by judiciously integrating the merits of CCA, linear discriminant analysis (LDA) and multi-task learning. The proposed model can identify the diagnose-specific as well as the diagnose-consistent features with significantly lower computational complexity. The performance of the proposed method, along with a comparison with the state-of-the-art methods, is evaluated on several synthetic data sets and one real imaging genetics data collected from Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. In the current study, the SNP as genetic data and resting state functional MRI ( fMRI) as imaging data are integrated to find the complex genotype-phenotype association. An important finding is that the proposed method has better correlation value, improved noise resistance and stability, and also has better feature selection ability. All the results illustrate the power and capability of the proposed method to find the diagnostic group-specific imaging genetic association, which may help to understand the neurodegenerative disorder in a more comprehensive way.

PROGNOSTIC VALUE OF AT(N) CLASSIFICATION IN CSF USING CLUSTERING ANALYSIS

AbstractBackgroundThe biological definition of Alzheimer’s disease (AD) and the use of biomarkers according to the AT(N) system may improve the clinical characterization of patients and the sequence of events during the clinical course. However, some studies have raised challenges in the clinical application of AT(N) system. Unbiased, data‐driven techniques such as unsupervised machine learning may help optimize it. We aimed to optimize the classification using unsupervised machine learning algorithms (clustering analysis).MethodWe included 421 patients from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort and 165 from our centre. We included subjective memory complaints (SMC), early mild cognitive impairment due to AD (EMCI), late MCI due to AD (LMCI) and MCI without evidence of neurodegeneration (MCI‐NN) patients according to clinical manifestations, neuropsychological test and CSF biomarkers from our center and the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. We clustered the biomarkers separately and all together using the KMeans algorithm for each cohort and compared them in terms of diagnosis, AT(N) system, biomarker values distribution, and risk of progression to dementia.ResultThe optimal number of clusters for all the biomarkers together was three. These clusters differed significantly in both cohorts according to diagnosis, AT(N) categories, biomarker values distribution, and progression to dementia. We found amyloid biomarkers to be more dichotomic, while Tau biomarkers were more continuous. Intergroup comparisons revealed the following groups: 1) non‐defined or unrelated to AD subjects, with no risk of progression to dementia; 2) early stages and/or more delayed risk of conversion to dementia subjects, and 3) more severe cognitive impairment subjects within the early stages of AD and with faster progression to dementia. Findings were largely similar between the two cohorts.ConclusionThis new 3‐group classification, developed with a data‐driven approach, represents a novel perspective to assess the risk of conversion to dementia in a much more simplified way and is complementary to the AT(N) system classification

Clinical Variables Contributing to Identification of Mild Cognitive Impairment Individuals With Defined ATN Profile

AbstractBackgroundThe framework globally used to classify Alzheimer’s Disease stages for research purposes includes three core biomarkers: A) Aβ burden; T) tau pathology; and N) neuronal injury. A lack of consensus exists in linking cognitive and behavioral tests used in everyday clinical practice to biological stages of dementia as defined by the ATN classification. The aim of the current study is to identify the clinical variables that best classify individuals into biologically defined cognitive profiles.MethodA total of 751 subjects, from the Alzheimer’s Disease Neuroimaging Initiative cohort, were categorized into seven ATN groups: 112 cognitively unimpaired with ATN‐; 46 cognitively unimpaired with ATN or AT+; 65 cognitively unimpaired with T, N or TN+; 128 amnestic Mild Cognitive Impairment (aMCI) with ATN‐; 223 aMCI with ATN or AT+; 94 aMCI with T, N or TN+ and 83 dementia with ATN or AT+. The groups for which the features exhibited differences during data exploration were classified employing 28 cognitive, behavioral, and demographic features using a supervised machine learning algorithm, specifically, a random forest classifier in combination with a synthetic oversampling technique.ResultWe found that the selected features were not able to distinguish between all seven groups and in particular between the three cognitively unimpaired groups. Therefore, classification was only performed for the three ATN‐defined aMCI groups achieving 58% overall accuracy. The Logical Delayed Recall was the most relevant feature (explaining 17% of the variance), followed by the Alzheimer’s Disease Assessment Scale‐Cognitive Subscale 13 (16%), Everyday Cognition Informant (8%), and Alzheimer’s Disease Assessment Scale 4 (8%). Marital status, Everyday Cognition Patient and sex were not relevant (0%).ConclusionCognitive and behavioral tests were not able to accurately distinguish cognitively unimpaired individuals with different biological profiles. Although accuracy could be improved, our first results showed that a test of delayed memory, one of general cognition, and one on activities of daily living are relevant variables for distinguishing between patients with aMCI and different biological profiles. During diagnosis, treatment monitoring, or in research contexts, these variables might be informative in differentiating the three ATN‐defined aMCI profiles, especially when biomarkers are not available.

Generative deep learning model‐based manifold construction for ATN biomarkers

AbstractBackgroundGenerative models have the potential to be beneficial in neuroimaging, as they can produce new data samples that closely resemble the ones they were trained on. This can assist in addressing the scarcity of brain scan data, which is a significant issue in neuroscience.MethodIn this study, we developed a one‐ring convolution‐based generative adversarial network (GAN) to generate ATN biomarkers, which operate on an icosahedron cortical mesh structure on a cortical surface. Data from the Alzheimer’s Disease Neuroimaging Initiative cohort was used in the study, including AV45, AV1451, and T1 magnetic resonance imaging (MRI) scans from 2283 subjects. The cortical thickness of all T1 MRI images was calculated using a Freesurfer pipeline and resampled into icosahedron subdivision level 3. AV45 and AV1451 scans were then converted into standard uptake value ratio (SUVR) and mapped to the cortical surface before being resampled into icosahedron subdivision level 3. For each modality, the proposed GAN was trained and evaluated using density and coverage metrics, which measure the fidelity and diversity of generated samples. Also, we validated the efficacy of the constructed ATN manifolds using the data augmentation scenario. Specifically, we evaluated the performance of tau stage classifiers using cortical thickness data with and without augmenting tau PET data from our manifolds.ResultOur proposed GAN achieved high performance in density and coverage. Specifically, the density and coverage for the amyloid SUVR were 1.0 and 0.97, for the tau SUVR were 1.34 and 1.0, and for the cortical thickness were 1.66 and 0.98, respectively, showing high performance in generating realistic and diverse samples for all modalities. The effectiveness of our proposed GAN was further demonstrated to show improved performance in the Tau Braak stage classification experiment when using the generated data. When augmented with generated data, the classification accuracy was improved from 0.77 to 0.84 in terms of AUC scores.ConclusionOur proposed GAN achieved high performance in density and coverage, generating realistic and diverse samples for each ATN biomarker, and demonstrated improved performance in a Tau Braak stage classification experiment when using the generated data.

Distinct CSF biomarker‐associated DNA methylation in Alzheimer’s disease and cognitively normal subjects

AbstractBackgroundGrowing evidence has demonstrated that DNA methylation (DNAm) plays an important role in Alzheimer’s disease and that DNAm differences can be detected in the blood of AD subjects. Because access to human brain tissues is primarily limited to postmortem samples, most studies have correlated blood DNAm with clinical diagnosis of AD in living individuals. However, as the pathophysiological process of AD can begin many years before the onset of clinical symptoms, there is often disagreement between neuropathology in the brain and clinical phenotypes. Thus, the association of blood DNAm with changes in AD neuropathology, rather than with clinical data, would provide more relevant information on AD pathogenesis.MethodCSF biomarkers are well‐established AD endophenotypes and correlate significantly with neuropathology scores measured on postmortem brain samples. In this study, we performed a comprehensive analysis to identify blood DNA methylation associated with CSF pathological biomarkers for AD in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. Our study included matched samples of whole blood DNA methylation, CSF Aβ42, phosphorylated tau181 (pTau181), and total tau (tTau) biomarkers data, measured on the same subjects and at the same clinical visits from a total of 202 subjects (123 cognitively normal, 79 AD cases).ResultWe identified a number of novel associations between blood DNAm and CSF Aβ42, pTau181, and total tau biomarkers, demonstrating that changes in various pathological processes in the CSF are also reflected in the blood epigenome. Overall, the CSF biomarker‐associated DNAm is relatively distinct in cognitively normal (CN) and AD subjects, highlighting the importance of analyzing omics data measured on cognitively normal subjects (which includes preclinical AD subjects) to identify diagnostic biomarkers, and considering disease stages in the development and testing of AD treatment strategies. Moreover, our pathway analysis of CSF biomarker associated‐DNAm from CN subjects revealed that biological processes associated with early brain impairment relevant to AD are marked by DNAm in the blood.ConclusionOur study provides a valuable resource for future mechanistic and biomarker studies of DNAm in AD.

Prediction of Conversion from Mild Cognitive Impairment to Alzheimer's Disease Using Amyloid PET and Brain MR Imaging Data: A 48-Month Follow-Up Analysis of the Alzheimer's Disease Neuroimaging Initiative Cohort.

We developed a novel quantification method named "shape feature" by combining the features of amyloid positron emission tomography (PET) and brain magnetic resonance imaging (MRI) and evaluated its significance in predicting the conversion from mild cognitive impairment (MCI) to Alzheimer's disease (AD) in the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. From the ADNI database, 334 patients with MCI were included. The brain amyloid smoothing score (AV45_BASS) and brain atrophy index (MR_BAI) were calculated using the surface area and volume of the region of interest in AV45 PET and MRI. During the 48-month follow-up period, 108 (32.3%) patients converted from MCI to AD. Age, Mini-Mental State Examination (MMSE), cognitive subscale of the Alzheimer's Disease Assessment Scale (ADAS-cog), apolipoprotein E (APOE), standardized uptake value ratio (SUVR), AV45_BASS, MR_BAI, and shape feature were significantly different between converters and non-converters. Univariate analysis showed that age, MMSE, ADAS-cog, APOE, SUVR, AV45_BASS, MR_BAI, and shape feature were correlated with the conversion to AD. In multivariate analyses, high shape feature, SUVR, and ADAS-cog values were associated with an increased risk of conversion to AD. In patients with MCI in the ADNI cohort, our quantification method was the strongest prognostic factor for predicting their conversion to AD.

Genome-Wide Association Analysis across Endophenotypes in Alzheimer's Disease: Main Effects and Disease Stage-Specific Interactions.

The underlying genetic susceptibility for Alzheimer's disease (AD) is not yet fully understood. The heterogeneous nature of the disease challenges genetic association studies. Endophenotype approaches can help to address this challenge by more direct interrogation of biological traits related to the disease. AD endophenotypes based on amyloid-β, tau, and neurodegeneration (A/T/N) biomarkers and cognitive performance were selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort (N = 1565). A genome-wide association study (GWAS) of quantitative phenotypes was performed using an SNP main effect and an SNP by Diagnosis interaction (SNP × DX) model to identify disease stage-specific genetic effects. Nine loci were identified as study-wide significant with one or more A/T/N endophenotypes in the main effect model, as well as additional findings significantly associated with cognitive measures. These nine loci include SNPs in or near the genes APOE, SRSF10, HLA-DQB1, XKR3, and KIAA1671. The SNP × DX model identified three study-wide significant genetic loci (BACH2, EP300, and PACRG-AS1) with a neuroprotective effect in later AD stage endophenotypes. An endophenotype approach identified novel genetic associations and provided insight into the molecular mechanisms underlying the genetic associations that may otherwise be missed using conventional case-control study designs.

An interpretable Alzheimer's disease oligogenic risk score informed by neuroimaging biomarkers improves risk prediction and stratification.

Stratification of Alzheimer's disease (AD) patients into risk subgroups using Polygenic Risk Scores (PRS) presents novel opportunities for the development of clinical trials and disease-modifying therapies. However, the heterogeneous nature of AD continues to pose significant challenges for the clinical broadscale use of PRS. PRS remains unfit in demonstrating sufficient accuracy in risk prediction, particularly for individuals with mild cognitive impairment (MCI), and in allowing feasible interpretation of specific genes or SNPs contributing to disease risk. We propose adORS, a novel oligogenic risk score for AD, to better predict risk of disease by using an optimized list of relevant genetic risk factors. Using whole genome sequencing data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort (n = 1,545), we selected 20 genes that exhibited the strongest correlations with FDG-PET and AV45-PET, recognized neuroimaging biomarkers that detect functional brain changes in AD. This subset of genes was incorporated into adORS to assess, in comparison to PRS, the prediction accuracy of CN vs. AD classification and MCI conversion prediction, risk stratification of the ADNI cohort, and interpretability of the genetic information included in the scores. adORS improved AUC scores over PRS in both CN vs. AD classification and MCI conversion prediction. The oligogenic model also refined risk-based stratification, even without the assistance of APOE, thus reflecting the true prevalence rate of the ADNI cohort compared to PRS. Interpretation analysis shows that genes included in adORS, such as ATF6, EFCAB11, ING5, SIK3, and CD46, have been observed in similar neurodegenerative disorders and/or are supported by AD-related literature. Compared to conventional PRS, adORS may prove to be a more appropriate choice of differentiating patients into high or low genetic risk of AD in clinical studies or settings. Additionally, the ability to interpret specific genetic information allows the focus to be shifted from general relative risk based on a given population to the information that adORS can provide for a single individual, thus permitting the possibility of personalized treatments for AD.

Associations of circulating saturated long-chain fatty acids with risk of mild cognitive impairment and Alzheimer’s disease in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort

No study has examined the associations between peripheral saturated long-chain fatty acids (LCFAs) and conversion from mild cognitive impairment (MCI) to Alzheimer's disease (AD). This study aimed to examine whether circulating saturated LCFAs are associated with both risks of incident MCI from cognitively normal (CN) participants and incident AD progressed from MCI in the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. We conducted analysis of data from older adults aged 55-90 years who were recruited at 63 sites across the USA and Canada. We examined associations between circulating saturated LCFAs (i.e., C14:0, C16:0, C18:0, C20:0) and risk for incident MCI in CN participants, and incident AD progressed from MCI. 829 participants who were enrolled in ADNI-1 had data on plasma saturated LCFAs, of which 618 AD-free participants were included in our analysis (226 with normal cognition and 392 with MCI; 60.2% were men). Cox proportional-hazards models were used to account for time-to-event/censor with a 48-month follow-up period for the primary analysis. Other than C20:0, saturated LCFAs were associated with an increased risk for AD among participants with MCI at baseline (Hazard ratios (HRs)=1.3 to 2.2, P=0.0005 to 0.003 in fully-adjusted models). No association of C20:0 with risk of AD among participants with MCI was observed. No associations were observed between saturated LCFAs and risk for MCI among participants with normal cognition. Saturated LCFAs are associated with increased risk of progressing from MCI to AD. This finding holds the potential to facilitate precision prevention of AD among patients with MCI. National Institutes of Health.

Deep learning analysis of UPLC-MS/MS-based metabolomics data to predict Alzheimer's disease

ObjectiveMetabolic biomarkers can potentially inform disease progression in Alzheimer's disease (AD). The purpose of this study is to identify and describe a new set of diagnostic biomarkers for developing deep learning (DL) tools to predict AD using Ultra Performance Liquid Chromatography Mass Spectrometry (UPLC-MS/MS)-based metabolomics data. MethodsA total of 177 individuals, including 78 with AD and 99 with cognitive normal (CN), were selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort along with 150 metabolomic biomarkers. We performed feature selection using the Least Absolute Shrinkage and Selection Operator (LASSO). The H2O DL function was used to build multilayer feedforward neural networks to predict AD. ResultsThe LASSO selected 21 metabolic biomarkers. To develop DL models, the 21 biomarkers identified by LASSO were imported into the H2O package. The data was split into 70% for training and 30% for validation. The best DL model with two layers and 18 neurons achieved an accuracy of 0.881, F1-score of 0.892, and AUC of 0.873. Several metabolomic biomarkers involved in glucose and lipid metabolism, in particular bile acid metabolites, were associated with APOE-ε4 allele and clinical biomarkers (Aβ42, tTau, pTau), cognitive assessments [the Alzheimer's Disease Assessment Scale-cognitive subscale 13 (ADAS13), the Mini-Mental State Examination (MMSE)], and hippocampus volume. ConclusionsThis study identified a new set of diagnostic metabolomic biomarkers for developing DL tools to predict AD. These biomarkers may help with early diagnosis, prognostic risk stratification, and/or early treatment interventions for patients at risk for AD.

A trichotomy method for defining homogeneous subgroups in a dementia population.

Diagnosis of dementia in the aging brain is confounded by the presence of multiple pathologies. Mixed dementia (MX), a combination of Alzheimer's disease (AD) proteins with vascular disease (VD), is frequently found at autopsy, and has been difficult to diagnose during life. This report develops a method for separating the MX group and defining preclinical AD (presence of AD factors with normal cognition) and preclinical VD subgroups (presence of white matter damage with normal cognition). Clustering was based on three diagnostic axes: (1) AD factor (ADF) derived from cerebrospinal fluid proteins (Aβ42 and pTau), (2) VD factor (VDF) calculated from mean free water and peak width of skeletonized mean diffusivity in the white matter, and (3) Cognition (Cog) based on memory and executive function. The trichotomy method was applied to an Alzheimer's Disease Neuroimaging Initiative cohort (N = 538). Eight biologically defined subgroups were identified which included the MX group with both high ADF and VDF (9.3%) and a preclinical VD group (3.9%), and a preclinical AD group (13.6%). Cog is significantly associated with both ADF and VDF, and the partial-correlation remains significant even when the effect of the other variable is removed (r(Cog, ADF/VDF removed) = 0.46, p < 10-28 and r(Cog, VDF/ADF removed) = 0.24, p < 10-7 ). The trichotomy method creates eight biologically characterized patient groups, which includes MX, preclinical AD, and preclinical VD subgroups. Further longitudinal studies are needed to determine the utility of the 3-way clustering method with multimodal biological biomarkers.

Lower Body Mass Index at Baseline Is Related to Steeper Cognitive Decline in the Alzheimer's Disease Neuroimaging Initiative Cohort.

Midlife obesity is a risk factor for dementia, whereas obesity in older age may be protective of cognition, a phenomenon known as the "obesity paradox." The mechanisms underlying this phenomenon and the relationship between body mass index (BMI) and cognitive function over time remain unclear. In 1399 adults with and without mild cognitive impairment (median age 73.6 years) from the Alzheimer's Disease Neuroimaging Initiative, we modeled the effects of baseline BMI on within-person trajectories of cognitive decline using Latent Growth Curve Modeling. We also tested if the effects of BMI on cognitive decline are global or specific to memory, executive function, or language. Higher baseline BMI was associated with better memory ( βBMI = 0.06, p < .05) and worse executive function ( βBMI = -0.05, p < .05) and not associated with language. Independent of baseline diagnosis, higher baseline BMI was associated with slower rate of decline in executive function, memory, and language ( βBMI = 0.13, 0.12, and 0.12, respectively; p < .01). Higher BMI was not associated with the intercept ( βBMI = 0.04, p = .059) or change ( βBMI = 0.04, p = .415) in a global cognitive factor. We found that higher baseline BMI was associated with slower cognitive decline in participants with and without mild cognitive impairment diagnosis. Higher BMI in this context seems to be protective of cognitive function for people at risk for dementia. Our findings also support domain-specific effects of obesity on various cognitive functions rather than a final common pathway.

Prevalence and Clinical Implications of a β-Amyloid–Negative, Tau-Positive Cerebrospinal Fluid Biomarker Profile in Alzheimer Disease

Knowledge is lacking on the prevalence and prognosis of individuals with a β-amyloid-negative, tau-positive (A-T+) cerebrospinal fluid (CSF) biomarker profile. To estimate the prevalence of a CSF A-T+ biomarker profile and investigate its clinical implications. This was a retrospective cohort study of the cross-sectional multicenter University of Gothenburg (UGOT) cohort (November 2019-January 2021), the longitudinal multicenter Alzheimer Disease Neuroimaging Initiative (ADNI) cohort (individuals with mild cognitive impairment [MCI] and no cognitive impairment; September 2005-May 2022), and 2 Wisconsin cohorts, Wisconsin Alzheimer Disease Research Center and Wisconsin Registry for Alzheimer Prevention (WISC; individuals without cognitive impairment; February 2007-November 2020). This was a multicenter study, with data collected from referral centers in clinical routine (UGOT) and research settings (ADNI and WISC). Eligible individuals had 1 lumbar puncture (all cohorts), 2 or more cognitive assessments (ADNI and WISC), and imaging (ADNI only) performed on 2 separate occasions. Data were analyzed on August 2022 to April 2023. Baseline CSF Aβ42/40 and phosphorylated tau (p-tau)181; cognitive tests (ADNI: modified preclinical Alzheimer cognitive composite [mPACC]; WISC: modified 3-test PACC [PACC-3]). Exposures in the ADNI cohort included [18F]-florbetapir amyloid positron emission tomography (PET), magnetic resonance imaging (MRI), [18F]-fluorodeoxyglucose PET (FDG-PET), and cross-sectional tau-PET (ADNI: [18F]-flortaucipir, WISC: [18F]-MK6240). Primary outcomes were the prevalence of CSF AT biomarker profiles and continuous longitudinal global cognitive outcome and imaging biomarker trajectories in A-T+ vs A-T- groups. Secondary outcomes included cross-sectional tau-PET. A total of 7679 individuals (mean [SD] age, 71.0 [8.4] years; 4101 male [53%]) were included in the UGOT cohort, 970 individuals (mean [SD] age, 73 [7.0] years; 526 male [54%]) were included in the ADNI cohort, and 519 individuals (mean [SD] age, 60 [7.3] years; 346 female [67%]) were included in the WISC cohort. The prevalence of an A-T+ profile in the UGOT cohort was 4.1% (95% CI, 3.7%-4.6%), being less common than the other patterns. Longitudinally, no significant differences in rates of worsening were observed between A-T+ and A-T- profiles for cognition or imaging biomarkers. Cross-sectionally, A-T+ had similar tau-PET uptake to individuals with an A-T- biomarker profile. Results suggest that the CSF A-T+ biomarker profile was found in approximately 5% of lumbar punctures and was not associated with a higher rate of cognitive decline or biomarker signs of disease progression compared with biomarker-negative individuals.

A data-driven approach to complement the A/T/(N) classification system using CSF biomarkers.

The AT(N) classification system not only improved the biological characterization of Alzheimer's disease (AD) but also raised challenges for its clinical application. Unbiased, data-driven techniques such as clustering may help optimize it, rendering informative categories on biomarkers' values. We compared the diagnostic and prognostic abilities of CSF biomarkers clustering results against their AT(N) classification. We studied clinical (patients from our center) and research (Alzheimer's Disease Neuroimaging Initiative) cohorts. The studied CSF biomarkers included Aβ(1-42), Aβ(1-42)/Aβ(1-40) ratio, tTau, and pTau. The optimal solution yielded three clusters in both cohorts, significantly different in diagnosis, AT(N) classification, values distribution, and survival. We defined these three CSF groups as (i) non-defined or unrelated to AD, (ii) early stages and/or more delayed risk of conversion to dementia, and (iii) more severe cognitive impairment subjects with faster progression to dementia. We propose this data-driven three-group classification as a meaningful and straightforward approach to evaluating the risk of conversion to dementia, complementary to the AT(N) system classification.

Identifying healthy individuals with Alzheimer’s disease neuroimaging phenotypes in the UK Biobank

BackgroundIdentifying prediagnostic neurodegenerative disease is a critical issue in neurodegenerative disease research, and Alzheimer’s disease (AD) in particular, to identify populations suitable for preventive and early disease-modifying trials. Evidence from genetic and other studies suggests the neurodegeneration of Alzheimer’s disease measured by brain atrophy starts many years before diagnosis, but it is unclear whether these changes can be used to reliably detect prediagnostic sporadic disease.MethodsWe trained a Bayesian machine learning neural network model to generate a neuroimaging phenotype and AD score representing the probability of AD using structural MRI data in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) Cohort (cut-off 0.5, AUC 0.92, PPV 0.90, NPV 0.93). We go on to validate the model in an independent real-world dataset of the National Alzheimer’s Coordinating Centre (AUC 0.74, PPV 0.65, NPV 0.80) and demonstrate the correlation of the AD-score with cognitive scores in those with an AD-score above 0.5. We then apply the model to a healthy population in the UK Biobank study to identify a cohort at risk for Alzheimer’s disease.ResultsWe show that the cohort with a neuroimaging Alzheimer’s phenotype has a cognitive profile in keeping with Alzheimer’s disease, with strong evidence for poorer fluid intelligence, and some evidence of poorer numeric memory, reaction time, working memory, and prospective memory. We found some evidence in the AD-score positive cohort for modifiable risk factors of hypertension and smoking.ConclusionsThis approach demonstrates the feasibility of using AI methods to identify a potentially prediagnostic population at high risk for developing sporadic Alzheimer’s disease.

Multimodal multitask learning for predicting MCI to AD conversion using stacked polynomial attention network and adaptive exponential decay

Early identification and treatment of moderate cognitive impairment (MCI) can halt or postpone Alzheimer’s disease (AD) and preserve brain function. For prompt diagnosis and AD reversal, precise prediction in the early and late phases of MCI is essential. This research investigates multimodal framework-based multitask learning in the following situations: (1) Differentiating early mild cognitive impairment (eMCI) from late MCI and (2) predicting when an MCI patient would acquire AD. Clinical data and two radiomics features on three brain areas deduced from magnetic resonance imaging were investigated (MRI). We proposed an attention-based module, Stack Polynomial Attention Network (SPAN), to firmly encode clinical and radiomics data input characteristics for successful representation from a small dataset. To improve multimodal data learning, we computed a potent factor using adaptive exponential decay (AED). We used experiments from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort study, which included 249 eMCI and 427 lMCI participants at baseline visits. The proposed multimodal strategy yielded the best c-index score in time prediction of MCI to AD conversion (0.85) and the best accuracy in MCI-stage categorization (83.19%). Moreover, our performance was equivalent to that of contemporary research.

Neuropsychiatric inventory questionnaire scores are predictive of dementia progression

AbstractBackgroundIn ageing adults, the most common form of dementia is Alzheimer’s disease (AD). AD pathogenesis involves the accumulation of beta‐amyloid (Aβ) protein aggregation in plaques and tau proteins in neurofibrillary tangles that are associated with a decreased number of synapses in the brain, altered neuronal function and cell death via neurotoxicity, as well as learning and memory deficits. Clinically, the presence and severity of neuropsychiatric symptoms that AD patients present with can be reliably measured by the Neuropsychiatric Inventory Questionnaire (NPIQ). This study aims to explore the association between baseline mental health and the severity of AD progression.MethodTo measure baseline mental health, NPIQ scores were used while the change in deq was used to represent the severity of AD progression within the TARCC cohort. deq is a homolog of d, a reliable latent dementia proxy that represents cognitive correlates of functional status and is specific for distinguishing cases with AD from other dementia‐related presentations. Most crucially, the deq homolog has been adjusted for equivalence across ethnicities. Using stratified analysis and structured equation models, the association between baseline mental health and change in was investigated.ResultThe first model exploring NPIQ and the change in deq was only adjusted for baseline deq; it showed NPIQ explaining 19% of the variance in delta deq and was statistically significant at p = 0.016 for Non‐Hispanic Whites. With the second model, age, sex, and education were adjusted for in addition to baseline deq. NPIQ was shown to explain 25% of the variance in delta deq while being statistically significant at p = 0.037 for Non‐Hispanic Whites. This model was replicated in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort; NPIQ was shown to be predictive for delta deq at p = 0.013.ConclusionIn Non‐Hispanic Whites, worse baseline mental health has been shown to predict increased severity and progression of AD. This makes it a clinical therapeutic target with the possible benefit of impacting the course of AD in patients. The fascinating interplay between mental health and its relationship to Alzheimer’s disease should be studied further with an additional focus on ethnicity.