Unique regional patterns of amyloid burden predict progression to prodromal and clinical stages of Alzheimer's disease

Multi-instance learning attention model for amyloid quantification of brain sub regions in longitudinal cognitive decline

Amyloid PET scans help in identifying the beta-amyloid deposition in different brain regions. The purpose of this study is to develop a deep learning model that can automate the task of finding amyloid deposition in different regions of the brain only by using PET scan and without the corresponding MRI scan. 2647 18F-Florbetapir PET scans are collected from Alzheimer's Disease Neuroimaging Initiative (ADNI) from multiple centres taken over a period. A deep learning model based on multi-instance learning and attention is proposed which is trained and validated using 80% of the scans and the remaining 20% of the scans are used for testing the model. The performance of the model is validated using Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). The proposed model is further tested upon an external dataset consisting of 1413 18F-Florbetapir PET scans from the Anti-Amyloid Treatment in Asymptomatic Alzheimer's (A4) study. The proposed model achieves MAE of 0.0243 and RMSE of 0.0320 for summary Standardized Uptake Value Ratio (SUVR) based on composite reference region for ADNI test set. When tested on the A4-study dataset, the proposed model achieves MAE of 0.038 and RMSE of 0.0495 for summary SUVR based on the composite region. The results show that the proposed model provides less MAE and RMSE when compared with existing models. A graphical user interface is developed based on the proposed model where the predictions are made by selecting the files of 18F-Florbetapir PET scans.

Evaluation of ComBat harmonization for reducing across-tracer biases in regional amyloid PET analyses

Background

Differences in amyloid positron emission tomography (PET) radiotracer pharmacokinetics and binding properties lead to discrepancies in amyloid-β uptake estimates. Harmonization of tracer-specific biases is crucial for optimal performance of downstream tasks. Here, we investigated the efficacy of ComBat, a data-driven harmonization model, for reducing tracer-specific biases in regional amyloid PET measurements from [18F]-florbetapir (FBP) and [11C]-Pittsburgh Compound-B (PiB).

Methods

One-hundred-thirteen head-to-head FBP-PiB scan pairs, scanned from the same subject within ninety days, were selected from the Open Access Series of Imaging Studies 3 (OASIS-3) dataset. The Centiloid scale, ComBat with no covariates, ComBat with biological covariates, and GAM-ComBat with biological covariates were used to harmonize both global and regional amyloid standardized uptake value ratios (SUVR). Intraclass correlation coefficient (ICC) and mean standardized absolute error (MsAE) were computed to measure the absolute agreement between tracers. Additionally, longitudinal amyloid SUVRs from an anti-amyloid drug trial were simulated using linear mixed effects modeling. Differences in rates-of-change between simulated treatment and placebo groups were tested, and change in statistical power/Type-I error after harmonization was quantified.

Results

In the head-to-head tracer comparison, the best ICC and MsAE were achieved after harmonizing with ComBat with no covariates for the global summary SUVR. ComBat with no covariates also performed the best in harmonizing regional SUVRs. In the clinical trial simulation, harmonization with both Centiloid and ComBat increased statistical power of detecting true rate-of-change differences between groups and decreased false discovery rate in the absence of a treatment effect. The greatest benefit of harmonization was observed when groups exhibited differing FPB-to-PiB proportions.

Conclusions

ComBat outperformed the Centiloid scale in harmonizing both global and regional amyloid estimates. Additionally, ComBat improved the detection of rate-of-change differences between clinical trial groups. Our findings suggest that ComBat is a viable alternative to Centiloid for harmonizing regional amyloid PET analyses.

Identifying potential genetic epistasis implicated in Alzheimer's disease via detection of SNP-SNP interaction on quantitative trait CSF Aβ42

Although genome-wide association studies have identified multiple Alzheimer's disease (AD)-associated loci by selecting the main effects of individual single-nucleotide polymorphisms (SNPs), the interpretation of genetic variance in AD is limited. Based on the linear regression method, we performed genome-wide SNP-SNP interaction on cerebrospinal fluid Aβ42 to identify potential genetic epistasis implicated in AD, with age, gender, and diagnosis as covariates. A GPU-based method was used to address the computational challenges posed by the analysis of epistasis. We found 368 SNP pairs to be statistically significant, and highly significant SNP-SNP interactions were identified between the marginal main effects of SNP pairs, which explained a relatively high variance at the Aβ42 level. Our results replicated 100 previously reported AD-related genes and 5 gene-gene interaction pairs of the protein-protein interaction network. Our bioinformatics analyses provided preliminary evidence that the 5-overlapping gene-gene interaction pairs play critical roles in inducing synaptic loss and dysfunction, thereby leading to memory decline and cognitive impairment in AD-affected brains.

Using Neuroimaging to Study Cerebral Amyloid Angiopathy and Its Relationship to Alzheimer's Disease

Cerebral amyloid angiopathy (CAA) is characterized by amyloid-β aggregation in the media and adventitia of the leptomeningeal and cortical blood vessels. CAA is one of the strongest vascular contributors to Alzheimer's disease (AD). It frequently co-occurs in AD patients, but the relationship between CAA and AD is incompletely understood. CAA may drive AD risk through damage to the neurovascular unit and accelerate parenchymal amyloid and tau deposition. Conversely, early AD may also drive CAA through cerebrovascular remodeling that impairs blood vessels from clearing amyloid-β. Sole reliance on autopsy examination to study CAA limits researchers' ability to investigate CAA's natural disease course and the effect of CAA on cognitive decline. Neuroimaging allows for in vivo assessment of brain function and structure and can be leveraged to investigate CAA staging and explore its associations with AD. In this review, we will discuss neuroimaging modalities that can be used to investigate markers associated with CAA that may impact AD vulnerability including hemorrhages and microbleeds, blood-brain barrier permeability disruption, reduced cerebral blood flow, amyloid and tau accumulation, white matter tract disruption, reduced cerebrovascular reactivity, and lowered brain glucose metabolism. We present possible areas for research inquiry to advance biomarker discovery and improve diagnostics.

The Alzheimer's Disease Neuroimaging Initiative in the era of Alzheimer's disease treatment: A review of ADNI studies from 2021 to 2022

The Alzheimer's Disease Neuroimaging Initiative (ADNI) aims to improve Alzheimer's disease (AD) clinical trials. Since 2006, ADNI has shared clinical, neuroimaging, and cognitive data, and biofluid samples. We used conventional search methods to identify 1459 publications from 2021 to 2022 using ADNI data/samples and reviewed 291 impactful studies. This review details how ADNI studies improved disease progression understanding and clinical trial efficiency. Advances in subject selection, detection of treatment effects, harmonization, and modeling improved clinical trials and plasma biomarkers like phosphorylated tau showed promise for clinical use. Biomarkers of amyloid beta, tau, neurodegeneration, inflammation, and others were prognostic with individualized prediction algorithms available online. Studies supported the amyloid cascade, emphasized the importance of neuroinflammation, and detailed widespread heterogeneity in disease, linked to genetic and vascular risk, co-pathologies, sex, and resilience. Biological subtypes were consistently observed. Generalizability of ADNI results is limited by lack of cohort diversity, an issue ADNI-4 aims to address by enrolling a diverse cohort.

Predicting cognitive dysfunction and regional hubs using Braak staging amyloid-beta biomarkers and machine learning

Mild cognitive impairment (MCI) is a transitional stage between normal aging and early Alzheimer's disease (AD). The presence of extracellular amyloid-beta (Aβ) in Braak regions suggests a connection with cognitive dysfunction in MCI/AD. Investigating the multivariate predictive relationships between regional Aβ biomarkers and cognitive function can aid in the early detection and prevention of AD. We introduced machine learning approaches to estimate cognitive dysfunction from regional Aβ biomarkers and identify the Aβ-related dominant brain regions involved with cognitive impairment. We employed Aβ biomarkers and cognitive measurements from the same individuals to train support vector regression (SVR) and artificial neural network (ANN) models and predict cognitive performance solely based on Aβ biomarkers on the test set. To identify Aβ-related dominant brain regions involved in cognitive prediction, we built the local interpretable model-agnostic explanations (LIME) model. We found elevated Aβ in MCI compared to controls and a stronger correlation between Aβ and cognition, particularly in Braak stages III-IV and V-VII (p < 0.05) biomarkers. Both SVR and ANN, especially ANN, showed strong predictive relationships between regional Aβ biomarkers and cognitive impairment (p < 0.05). LIME integrated with ANN showed that the parahippocampal gyrus, inferior temporal gyrus, and hippocampus were the most decisive Braak regions for predicting cognitive decline. Consistent with previous findings, this new approach suggests relationships between Aβ biomarkers and cognitive impairment. The proposed analytical framework can estimate cognitive impairment from Braak staging Aβ biomarkers and delineate the dominant brain regions collectively involved in AD pathophysiology.

Multi-domain cognition dysfunction accompanies frontoparietal and temporal amyloid accumulation in the elderly

It is helpful to understand the pathology of Alzheimer's disease by exploring the relationship between amyloid-β accumulation and cognition. The study explored the relationship between regional amyloid-β accumulation and multiple cognitions and study their application value in the Alzheimer's disease diagnosis. 135 participants completed 18F-florbetapir Positron Emission Tomography (PET), structural MRI, and a cognitive battery. Partial correlation was used to examine the relationship between global and regional amyloid-β accumulation and cognitions. Then, a support vector machine was applied to determine whether cognition-related accumulation regions can adequately distinguish the cognitively normal controls (76 participants) and mild cognitive impairment (30 participants) groups or mild cognitive impairment and Alzheimer's disease (29 participants) groups. The result showed that amyloid-β accumulation regions were mainly located in the frontoparietal cortex, calcarine fissure, and surrounding cortex and temporal pole regions. Episodic memory-related regions included the frontoparietal cortices; executive function-related regions included the frontoparietal, temporal, and occipital cortices; and processing speed-related regions included the frontal and occipital cortices. Support vector machine analysis showed that only episodic memory-related amyloid-β accumulation regions had better classification performance during the progression of Alzheimer's disease. Assessing regional changes in amyloid, particularly in frontoparietal regions, can aid in the early detection of amyloid-related decline in cognitive function.

White matter hyperintensities influence distal cortical β-amyloid accumulation in default mode network pathways

BACKGROUND AND PURPOSE

Cerebral small vessel disease (SVD) has been suggested to contribute to the pathogenesis of Alzheimer's disease (AD). Yet, the role of SVD in potentially contributing to AD pathology is unclear. The main objective of this study was to test the hypothesis that WMHs influence amyloid β (Aβ) levels within connected default mode network (DMN) tracts and cortical regions in cognitively unimpaired older adults.

METHODS

Regional standard uptake value ratios (SUVr) from Aβ-PET and white matter hyperintensity (WMH) volumes from three-dimensional magnetic resonance imaging FLAIR images were analyzed across a sample of 72 clinically unimpaired (mini-mental state examination ≥26), older adults (mean age 74.96 and standard deviation 8.13) from the Alzheimer's Disease Neuroimaging Initiative (ADNI3). The association of WMH volumes in major fiber tracts projecting from cortical DMN regions and Aβ-PET SUVr in the connected cortical DMN regions was analyzed using linear regression models adjusted for age, sex, ApoE, and total brain volumes.

RESULTS

The regression analyses demonstrate that increased WMH volumes in the superior longitudinal fasciculus were associated with increased regional SUVr in the inferior parietal lobule (p = .011).

CONCLUSION

The findings suggest that the relation between Aβ in parietal cortex is associated with SVD in downstream white matter (WM) pathways in preclinical AD. The biological relationships and interplay between Aβ and WM microstructure alterations that precede overt WMH development across the continuum of AD progression warrant further study.

Occipital Amyloid Deposition Is Associated with Rapid Cognitive Decline in the Alzheimer's Disease Continuum

BACKGROUND

Clinical significance of additional occipital amyloid-β (Aβ) plaques in Alzheimer's disease (AD) remains unclear.

OBJECTIVE

In this study, we investigated the effect of regional Aβ deposition on cognition in patients on the AD continuum, especially in the occipital region.

METHODS

We retrospectively reviewed the medical record of 208 patients with AD across the cognitive continuum (non-dementia and dementia). Multivariable linear regression analyses were performed to determine the effect of regional Aβ deposition on cognitive function. A linear mixed model was used to assess the effect of regional deposition on longitudinal changes in Mini-Mental State Examination (MMSE) scores. Additionally, the patients were dichotomized according to the occipital-to-global Aβ deposition ratio (ratio ≤1, Aβ-OCC- group; ratio >1, Aβ-OCC+ group), and the same statistical analyses were applied for between-group comparisons.

RESULTS

Regional Aβ burden itself was not associated with baseline cognitive function. In terms of Aβ-OCC group effect, the Aβ-OCC+ group exhibited a poorer cognitive performance on language function compared to the Aβ-OCC- group. High Aβ retention in each region was associated with a rapid decline in MMSE scores, only in the dementia subgroup. Additionally, Aβ-OCC+ individuals exhibited a faster annual decline in MMSE scores than Aβ-OCC- individuals in the non-dementia subgroup (β= -0.77, standard error [SE] = 0.31, p = 0.013).

CONCLUSION

The present study demonstrated that additional occipital Aβ deposition was associated with poor baseline language function and rapid cognitive deterioration in patients on the AD continuum.

Clear-headed into old age: Resilience and resistance against brain aging-A PET imaging perspective

With the advances in modern medicine and the adaptation towards healthier lifestyles, the average life expectancy has doubled since the 1930s, with individuals born in the millennium years now carrying an estimated life expectancy of around 100 years. And even though many individuals around the globe manage to age successfully, the prevalence of aging-associated neurodegenerative diseases such as sporadic Alzheimer's disease has never been as high as nowadays. The prevalence of Alzheimer's disease is anticipated to triple by 2050, increasing the societal and economic burden tremendously. Despite all efforts, there is still no available treatment defeating the accelerated aging process as seen in this disease. Yet, given the advances in neuroimaging techniques that are discussed in the current Review article, such as in positron emission tomography (PET) or magnetic resonance imaging (MRI), pivotal insights into the heterogenous effects of aging-associated processes and the contribution of distinct lifestyle and risk factors already have and are still being gathered. In particular, the concepts of resilience (i.e. coping with brain pathology) and resistance (i.e. avoiding brain pathology) have more recently been discussed as they relate to mechanisms that are associated with the prolongation and/or even stop of the progressive brain aging process. Better understanding of the underlying mechanisms of resilience and resistance may one day, hopefully, support the identification of defeating mechanism against accelerating aging.

Machine Learning Model to Predict Diagnosis of Mild Cognitive Impairment by Using Radiomic and Amyloid Brain PET

PURPOSE

The study aimed to develop a deep learning model for predicting amnestic mild cognitive impairment (aMCI) diagnosis using radiomic features and amyloid brain PET.

PATIENTS AND METHODS

Subjects (n = 328) from the Alzheimer's Disease Neuroimaging Initiative database and the EudraCT 2015-001184-39 trial (159 males, 169 females), with a mean age of 72 ± 7.4 years, underwent PET/CT with 18 F-florbetaben. The study cohort consisted of normal controls (n = 149) and subjects with aMCI (n = 179). Thirteen gray-level run-length matrix radiomic features and amyloid loads were extracted from 27 cortical brain areas. The least absolute shrinkage and selection operator regression was used to select features with the highest predictive value. A feed-forward neural multilayer network was trained, validated, and tested on 70%, 15%, and 15% of the sample, respectively. Accuracy, precision, F1-score, and area under the curve were used to assess model performance. SUV performance in predicting the diagnosis of aMCI was also assessed and compared with that obtained from the machine learning model.

RESULTS

The machine learning model achieved an area under the receiver operating characteristic curve of 90% (95% confidence interval, 89.4-90.4) on the test set, with 80% and 78% for accuracy and F1-score, respectively. The deep learning model outperformed SUV performance (area under the curve, 71%; 95% confidence interval, 69.7-71.4; 57% accuracy, 48% F1-score).

CONCLUSIONS

Using radiomic and amyloid PET load, the machine learning model identified MCI subjects with 84% specificity at 81% sensitivity. These findings show that a deep learning algorithm based on radiomic data and amyloid load obtained from brain PET images improves the prediction of MCI diagnosis compared with SUV alone.

The role of ADAM10 in astrocytes: Implications for Alzheimer’s disease

Much of the early research into AD relies on a neuron-centric view of the brain, however, evidence of multiple altered cellular interactions between glial cells and the vasculature early in AD has been demonstrated. As such, alterations in astrocyte function are widely recognized a contributing factor in the pathogenesis of AD. The processes by which astrocytes may be involved in AD make them an interesting target for therapeutic intervention, but in order for this to be most effective, there is a need for the specific mechanisms involving astrocyte dysfunction to be investigated. “α disintegrin and metalloproteinase” 10 (ADAM10) is capable of proteolytic cleavage of the amyloid precursor protein which prevents amyloid-β generation. As such ADAM10 has been identified as an interesting enzyme in AD pathology. ADAM10 is also known to play a role in a significant number of cellular processes, most notable in notch signaling and in inflammatory processes. There is a growing research base for the involvement of ADAM10 in regulating astrocytic function, primarily from an immune perspective. This review aims to bring together available evidence for ADAM10 activity in astrocytes, and how this relates to AD pathology.

Predicting conversion of brain β-amyloid positivity in amyloid-negative individuals

Background

Cortical deposition of β-amyloid (Aβ) plaque is one of the main hallmarks of Alzheimer’s disease (AD). While Aβ positivity has been the main concern so far, predicting whether Aβ (−) individuals will convert to Aβ (+) has become crucial in clinical and research aspects. In this study, we aimed to develop a classifier that predicts the conversion from Aβ (−) to Aβ (+) using artificial intelligence.

Methods

Data were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort regarding patients who were initially Aβ (−). We developed an artificial neural network-based classifier with baseline age, gender, APOE ε4 genotype, and global and regional standardized uptake value ratios (SUVRs) from positron emission tomography. Ten times repeated 10-fold cross-validation was performed for model measurement, and the feature importance was assessed. To validate the prediction model, we recruited subjects at the Samsung Medical Center (SMC).

Results

A total of 229 participants (53 converters) from the ADNI dataset and a total of 40 subjects (10 converters) from the SMC dataset were included. The average area under the receiver operating characteristic values of three developed models are as follows: Model 1 (age, gender, APOE ε4) of 0.674, Model 2 (age, gender, APOE ε4, global SUVR) of 0.814, and Model 3 (age, gender, APOE ε4, global and regional SUVR) of 0.841. External validation result showed an AUROC of 0.900.

Conclusion

We developed prediction models regarding Aβ positivity conversion. With the growing recognition of the need for earlier intervention in AD, the results of this study are expected to contribute to the screening of early treatment candidates.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13195-022-01067-8.

Introducing a gatekeeping system for amyloid status assessment in mild cognitive impairment

Background

In patients with mild cognitive impairment (MCI), enhanced cerebral amyloid-β plaque burden is a high-risk factor to develop dementia with Alzheimer’s disease (AD). Not all patients have immediate access to the assessment of amyloid status (A-status) via gold standard methods. It may therefore be of interest to find suitable biomarkers to preselect patients benefitting most from additional workup of the A-status. In this study, we propose a machine learning–based gatekeeping system for the prediction of A-status on the grounds of pre-existing information on APOE-genotype ¹⁸F-FDG PET, age, and sex.

Methods

Three hundred and forty-two MCI patients were used to train different machine learning classifiers to predict A-status majority classes among APOE-ε4 non-carriers (APOE4-nc; majority class: amyloid negative (Aβ-)) and carriers (APOE4-c; majority class: amyloid positive (Aβ +)) from ¹⁸F-FDG-PET, age, and sex. Classifiers were tested on two different datasets. Finally, frequencies of progression to dementia were compared between gold standard and predicted A-status.

Results

Aβ- in APOE4-nc and Aβ + in APOE4-c were predicted with a precision of 87% and a recall of 79% and 51%, respectively. Predicted A-status and gold standard A-status were at least equally indicative of risk of progression to dementia.

Conclusion

We developed an algorithm allowing approximation of A-status in MCI with good reliability using APOE-genotype, ¹⁸F-FDG PET, age, and sex information. The algorithm could enable better estimation of individual risk for developing AD based on existing biomarker information, and support efficient selection of patients who would benefit most from further etiological clarification. Further potential utility in clinical routine and clinical trials is discussed.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00259-022-05879-6.