Alzheimer's Disease Conversion Research Articles

Enrichment for clinical trials of early AD: Combining genetic risk factors and plasma p-tau as screening instruments.

Identifying low-cost, minimally-invasive screening instruments for Alzheimer's disease (AD) trial enrichment will improve the efficiency of AD trials. A total of 685 cognitively normal (CN) individuals and individuals with mild cognitive impairment (MCI) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were grouped according to cutoffs of genetic risk factor (G) polygenic hazard score (PHS) and tau pathology (T) plasma phosphorylated tau-181 (p-tau181) into four groups: G+T+, G-T-, G+T-, and G-T+. We assessed the associations between group level and longitudinal cognitive decline and AD conversion. Power analyses compared the estimated sample size required to detect differences in cognitive decline. The G+T+ group was associated with faster cognitive decline and higher AD risk. Clinical trials enrolling G+T+ participants would benefit from significantly reduced sample sizes compared with similar trials using only single makers as an inclusion criterion. The combination of two low-cost, minimally-invasive measures-genetics and plasma biomarkers-would be a promising screening procedure for clinical trial enrollment. Participants with unimpaired or mildly impaired cognition were grouped based on cutoffs on genetic risk factors (G: polygenic hazardous score [PHS]) and Alzheimer's pathology (T: baseline plasma phosphorylated tau-181 [p-tau181]). Participants with high PHSs and plasma p-tau181 levels (G+T+) were at risk of faster cognitive decline and AD progression. The combination of PHS and plasma p-tau181 could enhance clinical trial enrichment more effectively than using single biomarkers.

Prognostic model for predicting Alzheimer’s disease conversion using functional connectome manifolds

BackgroundEarly detection of Alzheimer’s disease (AD) is essential for timely management and consideration of therapeutic options; therefore, detecting the risk of conversion from mild cognitive impairment (MCI) to AD is crucial during neurodegenerative progression. Existing neuroimaging studies have mostly focused on group differences between individuals with MCI (or AD) and cognitively normal (CN), discarding the temporal information of conversion time. Here, we aimed to develop a prognostic model for AD conversion using functional connectivity (FC) and Cox regression suitable for conversion event modeling.MethodsWe developed a prognostic model using a large-scale Alzheimer’s Disease Neuroimaging Initiative dataset, and it was validated using external data obtained from the Open Access Series of Imaging Studies. We considered individuals who were initially CN or had MCI but progressed to AD and those with MCI with no progression to AD during the five-year follow-up period. As the exact conversion time to AD is unknown, we inferred this information using imputation approaches. We generated cortex-wide principal FC gradients using manifold learning techniques and computed subcortical-weighted manifold degrees from baseline functional magnetic resonance imaging data. A penalized Cox regression model with an elastic net penalty was adopted to define a risk score predicting the risk of conversion to AD, using FC gradients and clinical factors as regressors.ResultsOur prognostic model predicted the conversion risk and confirmed the role of imaging-derived manifolds in the conversion risk. The brain regions that largely contributed to predicting AD conversion were the heteromodal association and visual cortices, as well as the caudate and hippocampus. Our risk score based on Cox regression was consistent with the expected disease trajectories and correlated with positron emission tomography tracer uptake and symptom severity, reinforcing its clinical usefulness. Our findings were validated using an independent dataset. The cross-sectional application of our model showed a higher risk for AD than that for MCI, which correlated with symptom severity scores in the validation dataset.ConclusionWe proposed a prognostic model predicting the risk of conversion to AD. The associated risk score may provide insights for early intervention in individuals at risk of AD conversion.

The plasma miRNAome in ADNI: Signatures to aid the detection of at-risk individuals.

MicroRNAs are short non-coding RNAs that control proteostasis at the systems level and are emerging as potential prognostic and diagnostic biomarkers for Alzheimer's disease (AD). We performed small RNA sequencing on plasma samples from 847 Alzheimer's Disease Neuroimaging Initiative (ADNI) participants. We identified microRNA signatures that correlate with AD diagnoses and help predict the conversion from mild cognitive impairment (MCI) to AD. Our data demonstrate that plasma microRNA signatures can be used to not only diagnose MCI, but also, critically, predict the conversion from MCI to AD. Moreover, combined with neuropsychological testing, plasma microRNAome evaluation helps predict MCI to AD conversion. These findings are of considerable public interest because they provide a path toward reducing indiscriminate utilization of costly and invasive testing by defining the at-risk segment of the aging population. We provide the first analysis of the plasma microRNAome for the ADNI study. The levels of several microRNAs can be used as biomarkers for the prediction of conversion from MCI to AD. Adding the evaluation of plasma microRNA levels to neuropsychological testing in a clinical setting increases the accuracy of MCI to AD conversion prediction.

Potential Inflammatory Markers Related to the Conversion to Alzheimer’s Disease in Female Patients With Late-Life Depression

BackgroundInflammation has been postulated as a mediating factor in the development of Alzheimer’s disease (AD) pathology. We investigated candidate inflammatory markers related to conversion to AD among patients with depression. MethodsA longitudinal study was conducted with older women with depression who were at least 55 years of age, with a mean follow-up period of 5.73 years. At baseline, 9 inflammatory cytokines were measured using the immunoreactivity method. During follow-up, patients with depression who complained of cognitive impairment were evaluated and diagnosed with AD conversion. Association of the cytokines with conversion to AD was analyzed using multivariable Cox proportional hazards regression with adjusting covariates. For clinical applicability, the optimal cutoff value was determined using the minimum p value approach for the conversion to AD and was used to plot an AD-free survival curve. ResultsAmong 132 participants, 34 patients with depression (25.76%) developed AD during their follow-up period. Higher levels of interleukin (IL) 1β at baseline (hazard ratio = 3.30 [95% CI, 1.11–9.78], p = .031) and lower levels of IL-10 (p < .001) were significantly associated with an increased risk of progression to AD. The survival curve plotted by the cutoff value of ≥0.25 pg/mL for IL-1β and ≤0.15 pg/mL for IL-10 suggested adjusted hazard ratios of 8.96 (95% CI, 3.48–23.09; p < .001) for IL-1β and 10.99 (p < .001) for IL-10, respectively. ConclusionsThis study demonstrated that IL-1β and IL-10 were associated with conversion to AD among patients with late-life depression, suggesting their potential as predictive markers of the transition to AD from depression.

Predicting Longitudinal Cognitive Decline and Alzheimer's Conversion in Mild Cognitive Impairment Patients Based on Plasma Biomarkers.

The increasing burden of Alzheimer's disease (AD) emphasizes the need for effective diagnostic and therapeutic strategies. Despite available treatments targeting amyloid beta (Aβ) plaques, disease-modifying therapies remain elusive. Early detection of mild cognitive impairment (MCI) patients at risk for AD conversion is crucial, especially with anti-Aβ therapy. While plasma biomarkers hold promise in differentiating AD from MCI, evidence on predicting cognitive decline is lacking. This study's objectives were to evaluate whether plasma protein biomarkers could predict both cognitive decline in non-demented individuals and the conversion to AD in patients with MCI. This study was conducted as part of the Korean Longitudinal Study on Cognitive Aging and Dementia (KLOSCAD), a prospective, community-based cohort. Participants were based on plasma biomarker availability and clinical diagnosis at baseline. The study included MCI (n = 50), MCI-to-AD (n = 21), and cognitively unimpaired (CU, n = 40) participants. Baseline plasma concentrations of six proteins-total tau (tTau), phosphorylated tau at residue 181 (pTau181), amyloid beta 42 (Aβ42), amyloid beta 40 (Aβ40), neurofilament light chain (NFL), and glial fibrillary acidic protein (GFAP)-along with three derivative ratios (pTau181/tTau, Aβ42/Aβ40, pTau181/Aβ42) were analyzed to predict cognitive decline over a six-year follow-up period. Baseline protein biomarkers were stratified into tertiles (low, intermediate, and high) and analyzed using a linear mixed model (LMM) to predict longitudinal cognitive changes. In addition, Kaplan-Meier analysis was performed to discern whether protein biomarkers could predict AD conversion in the MCI subgroup. This prospective cohort study revealed that plasma NFL may predict longitudinal declines in Mini-Mental State Examination (MMSE) scores. In participants categorized as amyloid positive, the NFL biomarker demonstrated predictive performance for both MMSE and total scores of the Korean version of the Consortium to Establish a Registry for Alzheimer's Disease Assessment Packet (CERAD-TS) longitudinally. Additionally, as a baseline predictor, GFAP exhibited a significant association with cross-sectional cognitive impairment in the CERAD-TS measure, particularly in amyloid positive participants. Kaplan-Meier curve analysis indicated predictive performance of NFL, GFAP, tTau, and Aβ42/Aβ40 on MCI-to-AD conversion. This study suggests that plasma GFAP in non-demented participants may reflect baseline cross-sectional CERAD-TS scores, a measure of global cognitive function. Conversely, plasma NFL may predict longitudinal decline in MMSE and CERAD-TS scores in participants categorized as amyloid positive. Kaplan-Meier curve analysis suggests that NFL, GFAP, tTau, and Aβ42/Aβ40 are potentially robust predictors of future AD conversion.

Potential prognostic value of CSF-targeted proteomics across the Alzheimer’s disease continuum

BackgroundCore biomarkers for Alzheimer’s disease (AD), such as Aβ42 and tau, have demonstrated high prognostic accuracy but do not fully capture the complex pathophysiology of AD. In this study, our objective was to identify novel cerebrospinal fluid (CSF) biomarkers using proteomics across the entire AD continuum to predict conversion to AD and explore their involvement in AD pathogenesis.MethodsA cohort of 186 cognitively normal (CN), 127 subjective memory complaint (SMC), 79 early mild cognitive impairment (EMCI), 249 late MCI (LMCI), and 132 AD individuals was analyzed, with a follow-up period of over 3 years for non-AD participants. CSF 65 peptides, as well as hippocampal and entorhinal volumes were analyzed, and cognitive function was evaluated using the 13-item cognitive subscale of the Alzheimer’s Disease Assessment Scale (ADAS-Cog 13). Cox proportional hazards models and mediation analysis were performed to investigate associations and causal relationships.ResultsDuring the follow-up, approximately one-fourth (146/580) of the non-AD participants progressed to AD. After adjusting for baseline diagnosis (CN to LMCI) and other variables, multivariable Cox regression analysis identified three peptides (VAELEDEK, VSFELFADK, and VVSSIEQK) as significant predictors of conversion to AD. Incorporating these three peptides into the initial model significantly improved the C-statistic from 0.82 to 0.85 for predicting AD conversion, surpassing the predictive ability of Aβ42 and P-tau. Moreover, hippocampal and entorhinal volumes mediated 30.3–53.8% of the association between the three peptides and ADAS-Cog 13 scores.ConclusionsThese findings underscore the potential of these three peptides as robust prognostic biomarker candidates for AD conversion across the entire AD continuum, with a mechanism involving the mediation of hippocampal and entorhinal volumes.

A multimodal machine learning model for predicting dementia conversion in Alzheimer’s disease

Alzheimer’s disease (AD) accounts for 60–70% of the population with dementia. Mild cognitive impairment (MCI) is a diagnostic entity defined as an intermediate stage between subjective cognitive decline and dementia, and about 10–15% of people annually convert to AD. We aimed to investigate the most robust model and modality combination by combining multi-modality image features based on demographic characteristics in six machine learning models. A total of 196 subjects were enrolled from four hospitals and the Alzheimer’s Disease Neuroimaging Initiative dataset. During the four-year follow-up period, 47 (24%) patients progressed from MCI to AD. Volumes of the regions of interest, white matter hyperintensity, and regional Standardized Uptake Value Ratio (SUVR) were analyzed using T1, T2-weighted-Fluid-Attenuated Inversion Recovery (T2-FLAIR) MRIs, and amyloid PET (αPET), along with automatically provided hippocampal occupancy scores (HOC) and Fazekas scales. As a result of testing the robustness of the model, the GBM model was the most stable, and in modality combination, model performance was further improved in the absence of T2-FLAIR image features. Our study predicts the probability of AD conversion in MCI patients, which is expected to be useful information for clinician’s early diagnosis and treatment plan design.

Intelligent prediction of Alzheimer’s disease via improved multifeature squeeze-and-excitation-dilated residual network

This study aimed to address the issue of larger prediction errors existing in intelligent predictive tasks related to Alzheimer’s disease (AD). A cohort of 487 enrolled participants was categorized into three groups: normal control (138 individuals), mild cognitive impairment (238 patients), and AD (111 patients) in this study. An improved multifeature squeeze-and-excitation-dilated residual network (MFSE-DRN) was proposed for two important AD predictions: clinical scores and conversion probability. The model was characterized as three modules: squeeze-and-excitation-dilated residual block (SE-DRB), multifusion pooling (MF-Pool), and multimodal feature fusion. To assess its performance, the proposed model was compared with two other novel models: ranking convolutional neural network (RCNN) and 3D vision geometrical group network (3D-VGGNet). Our method showed the best performance in the two AD predicted tasks. For the clinical scores prediction, the root-mean-square errors (RMSEs) and mean absolute errors (MAEs) of mini-mental state examination (MMSE) and AD assessment scale–cognitive 11-item (ADAS-11) were 1.97, 1.46 and 4.20, 3.19 within 6 months; 2.48, 1.69 and 4.81, 3.44 within 12 months; 2.67, 1.86 and 5.81, 3.83 within 24 months; 3.02, 2.03 and 5.09, 3.43 within 36 months, respectively. At the AD conversion probability prediction, the prediction accuracies within 12, 24, and 36 months reached to 88.0, 85.5, and 88.4%, respectively. The AD predication would play a great role in clinical applications.

Impaired olfactory identification in dementia-free individuals is associated with the functional abnormality of the precuneus

ObjectiveOlfactory dysfunction indicates a higher risk of developing dementia. However, the potential structural and functional changes are still largely unknown. MethodsA total of 236 participants were enrolled, including 45 Alzheimer's disease (AD) individuals and 191dementia-free individuals. Detailed study methods, comprising neuropsychological assessment and olfactory identification test (University of Pennsylvania smell identification test, UPSIT), as well as structural and functional magnetic resonance imaging (MRI) were applied in this research. The dementia-free individuals were divided into two sub-groups based on olfactory score: dementia-free with olfactory dysfunction (DF-OD) sub-group and dementia-free without olfactory dysfunction (DF-NOD) sub-group. The results were analyzed for subsequent intergroup comparisons and correlations. The cognitive assessment was conducted again three years later. Results(i) At dementia-free stage, there was a positive correlation between olfactory score and cognitive function. (ii) In dementia-free group, the volume of crucial brain structures involved in olfactory recognition and processing (such as amygdala, entorhinal cortex and basal forebrain volumes) are positively associated with olfactory score. (iii) Compared to the DF-NOD group, the DF-OD group showed a significant reduction in olfactory network (ON) function. (iv) Compared to DF-NOD group, there were significant functional connectivity (FC) decline between PCun_L(R)_4_1 in the precuneus of posterior default mode network (pDMN) and the salience network (SN) in DF-OD group, and the FC values decreased with falling olfactory scores. Moreover, in DF-OD group, the noteworthy reduction in FC were observed between PCun_L(R)_4_1 and amygdala, which was a crucial component of ON. (v) The AD conversion rate of DF-OD was 29.41%, while the DF-NOD group was 12.50%. The structural and functional changes in the precuneus were also observed in AD and were more severe. ConclusionsIn addition to the olfactory circuit, the precuneus is a critical structure in the odor identification process, whose abnormal function underlies the olfactory identification impairment of dementia-free individuals.

Prediction Of Mild Cognitive Impairment to Alzheimer’s Disease Conversion Via Machine Learning

Alzheimer’s Disease (AD) is a progressive neurodegenerative condition characterized by deterioration of cognitive functions. Although there does not exist a cure for AD, early diagnosis and intervention during stages of Mild Cognitive Impairment (MCI) can be incredibly beneficial in slowing its development. However, predicting the conversion from MCI to AD remains a challenging task. This paper aims to provide an overview of the current research on predicting MCI to AD conversion, with a focus on machine learning methodologies to aid feature extraction and classification. Through a literature review, we aim to offer insights into the latest state-of-the-art techniques to predict MCI to AD conversion as well as prevailing trends in this field including deep learning, transfer learning, contrastive learning, and graph neural networks. We conclude from our study that utilizing AD/HC classes for feature extraction is a promising approach for generating discriminative features for stable MCI and progressive MCI classification. In addition, employing multi-modality models is instrumental in attaining a robust validation framework for the early diagnosis of AD.

“Back to Braak”: Role of Nucleus Reuniens and Subcortical Pathways in Alzheimer’s Disease Progression

BackgroundPatients with Alzheimer’s Disease (AD) exhibit structural alterations of the thalamus that correlate with clinical symptoms. However, given the anatomical complexity of this brain structure, it is still unclear whether atrophy affects specific thalamic nuclei and modulates the clinical progression from a prodromal stage, known as Mild Cognitive Impairment (MCI), to full-fledged AD.ObjectivesTo characterize the structural integrity of distinct thalamic nuclei across the AD spectrum, testing whether MCI patients who convert to AD (c-MCI) show a distinctive pattern of thalamic structural alterations compared to patients who remain stable (s-MCI).DesignInvestigating between-group differences in the volumetric features of distinct thalamic nuclei across the AD spectrum.SettingProdromal and clinical stages of AD.ParticipantsWe analyzed data from 84 healthy control subjects (HC), 58 individuals with MCI, and 102 AD patients. The dataset was obtained from the AD Neuroimaging Initiative (ADNI-3) database. The MCI group was further divided into two subgroups depending on whether patients remained stable (s-MCI, n=22) or progressed to AD (s-MCI, n=36) in the 48 months following the diagnosis.MeasurementsA multivariate analysis of variance (MANOVA) assessed group differences in the volumetric features of distinct thalamic nuclei obtained from magnetic resonance (MR) images. A stepwise discriminant function analysis identified which feature most effectively predicted the conversion to AD. The corresponding predictive performance was evaluated through a Receiver Operating Characteristic approach.ResultsAD and c-MCI patients showed generalized atrophy of thalamic nuclei compared to HC. In contrast, no significant structural differences were observed between s-MCI and HC subjects. Compared to s-MCI, c-MCI individuals displayed significant atrophy of the nucleus reuniens and a trend toward significant atrophy in the anteroventral and laterodorsal nuclei. The discriminant function analysis confirmed the nucleus reuniens as a significant predictor of AD conversion, with a sensitivity of 0.73 and a specificity of 0.69.ConclusionsIn line with the pathophysiological relevance of the nucleus reuniens proposed by seminal post-mortem studies on patients with AD, we confirm the pivotal role of this nucleus as a critical hub in the clinical progression to AD. We also propose a theoretical model to explain the evolving dysfunction of subcortical brain networks in the disease process.

Modified RCTU Score: A Semi-Quantitative, Visual Tool for Predicting Alzheimer's Conversion from aMCI.

This research evaluated the modified RCTU score, derived from amyloid PET scans, for predicting the progression from amnestic Mild Cognitive Impairment (aMCI) to Alzheimer's Disease (AD). aMCI patients underwent baseline evaluations, including amyloid PET. AD conversion was identified through neuropsychological tests after observation. The RCTU was modified by segmenting frontal, parietal, and temporal lobes into left and right, resulting in seven areas. Scores from both modified and conventional RCTU were analyzed and compared. Among 45 patients, 12 progressed to AD (over 17.8 ± 6.8 months). AD converters showed higher scores in modified RCTU scores. Modified RCTU score had strong correlations with amyloid SUVR (r > 0.7). Modified RCTU sum score was the significant covariate of AD conversion. Modified RCTU could determine the asymmetry of amyloid deposits. We demonstrated that symmetric deposits of amyloid showed a higher risk for AD conversion when analyzed using modified RCTU. The modified RCTU score is a promising method for predicting AD conversion, correlating strongly with amyloid SUVR.

Early Diagnosing and Transformation Prediction of Alzheimer's Disease Using Multi-Scaled Self-Attention Network on Structural MRI Images with Occlusion Sensitivity Analysis.

Structural magnetic resonance imaging (sMRI) is vital for early Alzheimer's disease (AD) diagnosis, though confirming specific biomarkers remains challenging. Our proposed Multi-Scale Self-Attention Network (MUSAN) enhances classification of cognitively normal (CN) and AD individuals, distinguishing stable (sMCI) from progressive mild cognitive impairment (pMCI). This study leverages AD structural atrophy properties to achieve precise AD classification, combining different scales of brain region features. The ultimate goal is an interpretable algorithm for this method. The MUSAN takes whole-brain sMRI as input, enabling automatic extraction of brain region features and modeling of correlations between different scales of brain regions, and achieves personalized disease interpretation of brain regions. Furthermore, we also employed an occlusion sensitivity algorithm to localize and visualize brain regions sensitive to disease. Our method is applied to ADNI-1, ADNI-2, and ADNI-3, and achieves high performance on the classification of CN from AD with accuracy (0.93), specificity (0.82), sensitivity (0.96), and area under curve (AUC) (0.95), as well as notable performance on the distinguish of sMCI from pMCI with accuracy (0.85), specificity (0.84), sensitivity (0.74), and AUC (0.86). Our sensitivity masking algorithm identified key regions in distinguishing CN from AD: hippocampus, amygdala, and vermis. Moreover, cingulum, pallidum, and inferior frontal gyrus are crucial for sMCI and pMCI discrimination. These discoveries align with existing literature, confirming the dependability of our model in AD research. Our method provides an effective AD diagnostic and conversion prediction method. The occlusion sensitivity algorithm enhances deep learning interpretability, bolstering AD research reliability.

Predicting Four-Year's Alzheimer's Disease Onset Using Longitudinal Neurocognitive Tests and MRI Data Using Explainable Deep Convolutional Neural Networks.

Prognosis of future risk of dementia from neuroimaging and cognitive data is important for optimizing clinical management for patients at early stage of Alzheimer's disease (AD). However, existing studies lack an efficient way to integrate longitudinal information from both modalities to improve prognosis performance. In this study, we aim to develop and evaluate an explainable deep learning-based framework to predict mild cognitive impairment (MCI) to AD conversion within four years using longitudinal whole-brain 3D MRI and neurocognitive tests. We proposed a two-stage framework that first uses a 3D convolutional neural network to extract single-timepoint MRI-based AD-related latent features, followed by multi-modal longitudinal feature concatenation and a 1D convolutional neural network to predict the risk of future dementia onset in four years. The proposed deep learning framework showed promising to predict MCI to AD conversion within 4 years using longitudinal whole-brain 3D MRI and cognitive data without extracting regional brain volumes or cortical thickness, reaching a balanced accuracy of 0.834, significantly improved from models trained from single timepoint or single modality. The post hoc model explainability revealed heatmap indicating regions that are important for predicting future risk of AD. The proposed framework sets the stage for future studies for using multi-modal longitudinal data to achieve optimal prediction for prognosis of AD onset, leading to better management of the diseases, thereby improving the quality of life.

Orthogonal Mixed-Effects Modeling for High-Dimensional Longitudinal Data: An Unsupervised Learning Approach.

The linear mixed-effects model is commonly utilized to interpret longitudinal data, characterizing both the global longitudinal trajectory across all observations and longitudinal trajectories within individuals. However, characterizing these trajectories in high-dimensional longitudinal data presents a challenge. To address this, our study proposes a novel approach, Unsupervised Orthogonal Mixed-Effects Trajectory Modeling (UOMETM), that leverages unsupervised learning to generate latent representations of both global and individual trajectories. We design an autoencoder with a latent space where an orthogonal constraint is imposed to separate the space of global trajectories from individual trajectories. We also devise a cross-reconstruction loss to ensure consistency of global trajectories and enhance the orthogonality between representation spaces. To evaluate UOMETM, we conducted simulation experiments on images to verify that every component functions as intended. Furthermore, we evaluated its performance and robustness using longitudinal brain cortical thickness from two Alzheimer's disease (AD) datasets. Comparative analyses with state-of-the-art methods revealed UOMETM's superiority in identifying global and individual longitudinal patterns, achieving a lower reconstruction error, superior orthogonality, and higher accuracy in AD classification and conversion forecasting. Remarkably, we found that the space of global trajectories did not significantly contribute to AD classification compared to the space of individual trajectories, emphasizing their clear separation. Moreover, our model exhibited satisfactory generalization and robustness across different datasets. The study shows the outstanding performance and potential clinical use of UOMETM in the context of longitudinal data analysis.

Insulin-like growth factor binding protein-2 in at-risk adults and autopsy-confirmed Alzheimer brains.

Insulin, insulin-like growth factors (IGF) and their receptors are highly expressed in the adult hippocampus. Thus, disturbances in the insulin-IGF signalling pathway may account for the selective vulnerability of the hippocampus to nascent Alzheimer's disease (AD) pathology. In the present study, we examined the predominant IGF-binding protein in the CSF, IGFBP2. CSF was collected from 109 asymptomatic members of the parental history-positive PREVENT-AD cohort. CSF levels of IGFBP2, core AD and synaptic biomarkers were measured using proximity extension assay, ELISA and mass spectrometry. Cortical amyloid-beta (Aβ) and tau deposition were examined using 18F-NAV4694 and flortaucipir. Cognitive assessments were performed during up to 8 years of follow-up, using the Repeatable Battery for the Assessment of Neuropsychological Status. T1-weighted structural MRI scans were acquired, and neuroimaging analyses were performed on pre-specified temporal and parietal brain regions. Next, in an independent cohort, we allocated 241 dementia-free ADNI-1 participants into four stages of AD progression based on the biomarkers CSF Aβ42 and total-tau (t-tau). In this analysis, differences in CSF and plasma IGFBP2 levels were examined across the pathological stages. Finally, IGFBP2 mRNA and protein levels were examined in the frontal cortex of 55 autopsy-confirmed AD and 31 control brains from the Quebec Founder Population (QFP) cohort, a unique population isolated from Eastern Canada. CSF IGFBP2 progressively increased over 5 years in asymptomatic PREVENT-AD participants. Baseline CSF IGFBP2 was positively correlated with CSF AD biomarkers and synaptic biomarkers, and negatively correlated with longitudinal changes in delayed memory (P = 0.024) and visuospatial abilities (P = 0.019). CSF IGFBP2 was negatively correlated at a trend-level with entorhinal cortex volume (P = 0.082) and cortical thickness in the piriform (P = 0.039), inferior temporal (P = 0.008), middle temporal (P = 0.014) and precuneus (P = 0.033) regions. In ADNI-1, CSF (P = 0.009) and plasma (P = 0.001) IGFBP2 were significantly elevated in Stage 2 [CSF Aβ(+)/t-tau(+)]. In survival analyses in ADNI-1, elevated plasma IGFBP2 was associated with a greater rate of AD conversion (hazard ratio = 1.62, P = 0.021). In the QFP cohort, IGFBP2 mRNA was reduced (P = 0.049); however, IGFBP2 protein levels did not differ in the frontal cortex of autopsy-confirmed AD brains (P = 0.462). Nascent AD pathology may induce an upregulation in IGFBP2 in asymptomatic individuals. CSF and plasma IGFBP2 may be valuable markers for identifying CSF Aβ(+)/t-tau(+) individuals and those with a greater risk of AD conversion.

Predicting mild cognitive impairments from cognitively normal brains using a novel brain age estimation model based on structural magnetic resonance imaging.

Brain age prediction is a practical method used to quantify brain aging and detect neurodegenerative diseases such as Alzheimer's disease (AD). However, very few studies have considered brain age prediction as a biomarker for the conversion of cognitively normal (CN) to mild cognitive impairment (MCI). In this study, we developed a novel brain age prediction model using brain volume and cortical thickness features. We calculated an acceleration of brain age (ABA) derived from the suggested model to estimate different diagnostic groups (CN, MCI, and AD) and to classify CN to MCI and MCI to AD conversion groups. We observed a strong association between ABA and the 3 diagnostic groups. Additionally, the classification models for CN to MCI conversion and MCI to AD conversion exhibited acceptable and robust performances, with area under the curve values of 0.66 and 0.76, respectively. We believe that our proposed model provides a reliable estimate of brain age for elderly individuals and can identify those at risk of progressing from CN to MCI. This model has great potential to reveal a diagnosis associated with a change in cognitive decline.

Deep learning for risk-based stratification of cognitively impaired individuals

Quantifying the risk of progression to Alzheimer's disease (AD) could help identify persons who could benefit from early interventions. We used data from the Alzheimer's Disease Neuroimaging Initiative (ADNI, n= 544, discovery cohort) and the National Alzheimer's Coordinating Center (NACC, n= 508, validation cohort), subdividing individuals with mild cognitive impairment (MCI) into risk groups based on cerebrospinal fluid amyloid-β levels and identifying differential gray matter patterns. We then created models that fused neural networks with survival analysis, trained using non-parcellated T1-weighted brain MRIs from ADNI data, to predict the trajectories of MCI to AD conversion within the NACC cohort (integrated Brier score: 0.192 [discovery], and 0.108 [validation]). Using modern interpretability techniques, we verified that regions important for model prediction are classically associated with AD. We confirmed AD diagnosis labels using postmortem data. We conclude that our framework provides a strategy for risk-based stratification of individuals with MCI and for identifying regions key for disease prognosis.

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.

A single baseline amyloid PET could be sufficient for predicting Alzheimer’s disease conversion in mild cognitive impairment

AbstractBackgroundBaseline amyloid burden in mild cognitive impairment (MCI) has been linked to conversion to Alzheimer’s disease (AD), but the comparison of baseline and longitudinal changes in amyloid burden for predicting AD remains unresolved. The objectives of this study aimed to compare the prognostic ability of baseline and longitudinal changes in amyloid burden in MCI patients.MethodSeventy‐five individuals with MCI were recruited and examined annually by clinical interviews for a mean follow‐up of 24 months (Range 11.6 – 42.0). [18F]Florbetaben positron emission tomography (PET) scans were performed. T1‐weighted 3D volumes were acquired for co‐registration, and to define regions of interest (ROIs). We examined whether baseline and longitudinal amyloid burden changes can improve AD conversion by Cox proportional hazard model analysis and receiver operating characteristic (ROC) curve analysis.ResultCox proportional hazards model analysis showed that baseline amyloid burden was significantly associated with increased risk of conversion to AD (hazard ratio [HR] = 10.0; 95% confidence interval [CI], 1.15 – 85.39; p = 0.04), but longitudinal amyloid burden changes was not (HR = 0.2; 95% CI, 0.04 – 1.51; p = 0.13). When predicting AD, longitudinal amyloid burden changes had better ROC accuracy of 65.2% (95% CI, 48.4 – 82.0%) than baseline amyloid burden of 59.6% (95% CI, 40.3 – 79.0%), without statistical significance in pairwise comparison.ConclusionA single baseline amyloid PET could be sufficient in the prediction of AD conversion in MCI.