No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Prediction of neuropathologic lesions from clinical data
Alzheimer's & Dementia
Abstract
Introduction:
Post-mortem analysis provides definitive diagnoses of neurodegenerative diseases; however, only a few can be diagnosed during life.
Methods:
This study employed statistical tools and machine learning to predict 17 neuropathologic lesions from a cohort of 6518 individuals using 381 clinical features (Table S1). The multisite data allowed validation of the model's robustness by splitting train/test sets by clinical sites. A similar study was performed for predicting Alzheimer's disease (AD) neuropathologic change without specific comorbidities.
Results:
Prediction results show high performance for certain lesions that match or exceed that of research annotation. Neurodegenerative comorbidities in addition to AD neuropathologic change resulted in compounded, but disproportionate, effects across cognitive domains as the comorbidity number increased.
Discussion:
Certain clinical features could be strongly associated with multiple neurodegenerative diseases, others were lesion-specific, and some were divergent between lesions. Our approach could benefit clinical research, and genetic and biomarker research by enriching cohorts for desired lesions.
... Indeed, binary categorization of CR captures only the most extreme examples and neglects its likely variable expression by different people and across different cognitive domains. To operationalize CR, we used our previously described (Phongpreecha et al., 2023a) neuropathologic damage estimate derived using a machine learning (ML) approach that combined 17 lesions (including plaques, tangles, LBs, hippocampal sclerosis, and TDP-43) into a single index, along with assessments of cognitive function as determined by neuropsychological tests or clinical assessment. Our model simplifies the relationship into linear equations: ...
The cellular and molecular distinction between brain aging and neurodegenerative disease begins to blur in the oldest old. Approximately 15–25% of observations in humans do not fit predicted clinical manifestations, likely the result of suppressed damage despite usually adequate stressors and of resilience, the suppression of neurological dysfunction despite usually adequate degeneration. Factors during life may predict the clinico-pathologic state of resilience: cardiovascular health and mental health, more so than educational attainment, are predictive of a continuous measure of resilience to Alzheimer’s disease (AD) and AD-related dementias (ADRDs). In resilience to AD alone (RAD), core features include synaptic and axonal processes, especially in the hippocampus. Future focus on larger and more diverse cohorts and additional regions offer emerging opportunities to understand this counterforce to neurodegeneration. The focus of this review is the molecular basis of resilience to AD.
... Greater efforts are needed to profile biomarkers with various pathophysiological mechanisms in DLB in large collaborative longitudinal studies. Machine learning techniques and emerging high-throughput proteomic profiling platforms such as Olink and SomaScan may also prove useful for identifying co-pathologies and discovering novel mechanistic pathways implicated in LBD [116,117]. ...
Purpose of review:
Currently, no disease modifying therapies (DMTs) have been approved for use in dementia with Lewy bodies (DLB). Clinical trials face difficulties due to the clinical and neuropathological heterogeneity of the condition with a diverse array of neuropathogenic mechanisms contributing to the clinical phenotype. The purpose of this review is to describe how recent advances in the development of biofluid biomarkers may be used in clinical trials to tackle some of these challenges.
Recent findings:
Biomarkers are essential both to support the accurate diagnosis of DLB and to delineate the influence of coexisting pathologies. Recent advances in the development of α-synuclein seeding amplification assays (SAA) allow accurate identification of α-synuclein from the prodromal stages in DLB. Additionally, validation of plasma phosphorylated tau assays in DLB is ongoing and offers an accessible biomarker to indicate the existence of AD co-pathology. Use of biomarkers for diagnosis and group stratification in clinical trials of DLB is growing and likely to be of increasing importance in the future.
Summary:
In vivo biomarkers can enhance patient selection in clinical trials allowing greater diagnostic accuracy, a more homogeneous trial population, and stratification by co-pathology to create subgroups most likely to derive therapeutic benefit from DMTs.
Background:
Alzheimer's disease neuropathologic changes (AD-NC) are important to identify people with high risk for AD dementia (ADD) and subtyping ADD.
Objective:
Develop imputation models based on clinical measures to infer AD-NC.
Methods:
We used penalized generalized linear regression to train imputation models for four AD-NC traits (amyloid-β, tangles, global AD pathology, and pathologic AD) in Rush Memory and Aging Project decedents, using clinical measures at the last visit prior to death as predictors. We validated these models by inferring AD-NC traits with clinical measures at the last visit prior to death for independent Religious Orders Study (ROS) decedents. We inferred baseline AD-NC traits for all ROS participants at study entry, and then tested if inferred AD-NC traits at study entry predicted incident ADD and postmortem pathologic AD.
Results:
Inferred AD-NC traits at the last visit prior to death were related to postmortem measures with R2 = (0.188,0.316,0.262) respectively for amyloid-β, tangles, and global AD pathology, and prediction Area Under the receiver operating characteristic Curve (AUC) 0.765 for pathologic AD. Inferred baseline levels of all four AD-NC traits predicted ADD. The strongest prediction was obtained by the inferred baseline probabilities of pathologic AD with AUC = (0.919,0.896) for predicting the development of ADD in 3 and 5 years from baseline. The inferred baseline levels of all four AD-NC traits significantly discriminated pathologic AD profiled eight years later with p-values < 1.4×10-10.
Conclusions:
Inferred AD-NC traits based on clinical measures may provide effective AD biomarkers that can estimate the burden of AD-NC traits in aging adults.
Background
We have proposed that cognitive resilience (CR) counteracts brain damage from Alzheimer’s disease (AD) or AD-related dementias such that older individuals who harbor neurodegenerative disease burden sufficient to cause dementia remain cognitively normal. However, CR traditionally is considered a binary trait, capturing only the most extreme examples, and is often inconsistently defined.
Methods
This study addressed existing discrepancies and shortcomings of the current CR definition by proposing a framework for defining CR as a continuous variable for each neuropsychological test. The linear equations clarified CR’s relationship to closely related terms, including cognitive function, reserve, compensation, and damage. Primarily, resilience is defined as a function of cognitive performance and damage from neuropathologic damage. As such, the study utilized data from 844 individuals (age = 79 ± 12, 44% female) in the National Alzheimer’s Coordinating Center cohort that met our inclusion criteria of comprehensive lesion rankings for 17 neuropathologic features and complete neuropsychological test results. Machine learning models and GWAS then were used to identify medical and genetic factors that are associated with CR.
Results
CR varied across five cognitive assessments and was greater in female participants, associated with longer survival, and weakly associated with educational attainment or APOE ε4 allele. In contrast, damage was strongly associated with APOE ε4 allele (P value < 0.0001). Major predictors of CR were cardiovascular health and social interactions, as well as the absence of behavioral symptoms.
Conclusions
Our framework explicitly decoupled the effects of CR from neuropathologic damage. Characterizations and genetic association study of these two components suggest that the underlying CR mechanism has minimal overlap with the disease mechanism. Moreover, the identified medical features associated with CR suggest modifiable features to counteract clinical expression of damage and maintain cognitive function in older individuals.
Introduction:
Neuropsychiatric symptoms (NPS) are common in Lewy body disease (LBD), but their etiology is poorly understood.
Methods:
In a population-based post mortem study neuropathological data was collected for Lewy body (LB) neuropathology, neurofibrillary tangles (NFT), amyloid beta burden, TDP-43, lacunar infarcts, cerebral amyloid angiopathy (CAA), and hyaline atherosclerosis. Post mortem interviews collected systematic information regarding NPS and cognitive status. A total of 1038 cases were included: no pathology (NP; n = 761), Alzheimer's disease (AD; n = 189), LBD (n = 60), and AD+LBD (n = 28).
Results:
Hallucinations were associated with higher LB Braak stages, while higher NFT Braak staging was associated with depression, agitation, and greater number of symptoms in the Neuropsychiatric Inventory. Cases with dual AD+LBD pathology had the highest risk of hallucinations, agitation, apathy, and total symptoms but a multiplicative interaction between these pathologies was not significant.
Discussion:
LB and AD pathology contribute differentially to NPS likely with an additive process contributing to the increased burden of NPS.
Aducanumab (Aduhelm) is approved in the United States for the treatment of patients with mild cognitive impairment due to Alzheimer's disease or mild AD dementia. Aducanumab Appropriate Use Recommendations (AURs) have been published and have helped guide best practices for use of aducanumab. As real-world use has occurred and more information has accrued, the AURs require refinement. We update the AURs to better inform appropriate patient selection and improve shared decision-making, safety monitoring, and risk mitigation in treated patients. Based on evolving experience we emphasize the importance of detecting past medical conditions that may predispose to amyloid related imaging abnormalities (ARIA) or may increase the likelihood of ARIA complications including autoimmune or inflammatory conditions, seizures, or disorders associated with extensive white matter pathology. The apolipoprotein E ε4 (APOE4) genotype is strongly associated with ARIA and exhibits a gene dose effect. We recommend that clinicians perform APOE genotyping to better inform patient care decisions, discussions regarding risk, and clinician vigilance concerning ARIA. As most ARIA occurs during the titration period of aducanumab, we suggest performing MRI before the 5th, 7th, 9th, and 12th infusions to improve detection. Uncommonly, ARIA may be recurrent or serious; we suggest additional parameters for treatment discontinuation taking these observations into account. It is important to continue to learn from the real-world use of aducanumab and the AURs will continue to evolve as new information becomes available. This AUR update does not address efficacy, price, or insurance coverage and is provided to assist clinicians to establish best practices for use of aducanumab in the treatment of patients with mild cognitive impairment and mild Alzheimer's dementia.
Introduction:
We examined the association between Alzheimer's disease (AD) and type 2 diabetes mellitus (DM) and hypothesized that diabetes is associated with an increased pathological burden in clinically and pathologically diagnosed AD.
Methods:
All data were obtained from the Uniform Data Set (UDS) v3, the Neuropathology Data Set, and the Researcher's Data Dictionary-Genetic Data from the National Alzheimer's Coordinating Center. The dataset (37 cases with diabetes and 1158 cases without) relies on autopsy-confirmed data in clinically diagnosed AD patients who were assessed for diabetes type in form A5 or D2 during at least one visit. Differences in scores were explored using a general linear model. Effect sizes were calculated using sample means and standard deviations (Cohen's d).
Results:
The presence of diabetes was associated with a lower Thal phase of amyloid plaques (A score; 4.6 ± 0.79 vs. 4.3 ± 0.85, P < .05) and lower Braak stage for neurofibrillary degeneration (B score; 5.58 ± 0.72 vs. 5.16 ± 0.96, P < 0.05) but not for density of neocortical neuritic plaques (CERAD score-C score). The National Institute on Aging-Alzheimer's Association Alzheimer's disease neuropathologic change (ABC score) was not different between AD+DM and AD-DM.
Discussion:
This pilot study found a significantly lower Thal phase of amyloid plaques and Braak stage for neurofibrillary degeneration in AD-confirmed individuals with diabetes compared to those without. Thus type 2 DM is not associated with increased AD pathology in clinically and pathologically confirmed cases of AD.
Lewy bodies (LBs) and limbic-predominant age-related TDP-43 encephalopathy neuropathologic change (LATE-NC) are common in older persons and associated with cognitive impairment. However, little is known about the relationship between LBs and LATE-NC and their combined roles in cognitive impairment and Alzheimer’s dementia in community-dwelling participants. The study included 1670 community-based participants (mean age-at-death, 89.5 years (SD = 6.65); 69% females) who underwent annual assessments of cognition to create summary measures of global cognition and cognitive domains and evaluation for Alzheimer’s dementia. Systematic neuropathologic evaluations were performed to assess LBs, LATE-NC, and Alzheimer’s disease (AD) pathology. We excluded cases with pathologically confirmed frontotemporal lobar degeneration in this study. Logistic and linear regression analyses were used, adjusted for demographics and AD pathology. LBs were present in 428 (25.6%) decedents (29 nigra-predominant, 165 limbic-type, and 234 neocortical-type) while 865 (51.7%) decedents exhibited LATE-NC (307 stage 1, 167 stage 2, and 391 stage 3). LBs combined with LATE-NC were common (15% of all participants) and in those with Alzheimer’s dementia (25%). Neocortical-type, but not nigral-predominant or limbic-type LBs increased the odds of stage 2/3 LATE-NC (odds ratio = 1.70; 95% confidence interval = 1.26–2.30). The association between neocortical-type LBs and stage 2/3 LATE-NC was stronger in those under 90 years of age and in women. In analyses of cognition and Alzheimer’s dementia, LATE-NC and neocortical-type LBs, separately, were related to lower global cognition, five specific cognitive domains, and an increased odds of Alzheimer’s dementia, above and beyond the AD pathology. Limbic-type LBs were related to lower global cognition, and the domains of episodic, working, and semantic memory, and increased odds of Alzheimer’s dementia. Furthermore, there was no interaction between limbic/neocortical-type LBs and LATE-NC on cognitive function, cognitive domains, or Alzheimer’s dementia. These findings suggest that neocortical-type LBs are associated with LATE-NC, specifically in the younger old and in women. Limbic/neocortical-type LBs and LATE-NC have separate and additive effects on cognitive function and odds of Alzheimer’s dementia.
Supplementary Information
The online version contains supplementary material available at 10.1186/s40478-021-01260-0.
Background:
The ultimate validation of a clinical marker for Alzheimer's disease (AD) is its association with AD neuropathology.
Objective:
To identify clinical measures that predict pathology, we evaluated the relationships of the picture version of the Free and Cued Selective Reminding Test (pFCSRT + IR), the Mini-Mental State Exam (MMSE), and the Clinical Dementia Rating scale Sum of Boxes (CDR-SB) to Braak stage.
Methods:
315 cases from the clinicopathologic series at the Knight Alzheimer's Disease Research Center were classified according to Braak stage. Boxplots of each predictor were compared to identify the earliest stage at which decline was observed and ordinal logistic regression was used to predict Braak stage.
Results:
Looking at the assessment closest to death, free recall scores were lower in individuals at Braak stage III versus Braak stages 0 and I (combined) while MMSE and CDR scores for individuals did not differ from Braak stages 0/I until Braak stage IV. The sum of free recall and total recall scores independently predicted Braak stage and had higher predictive validity than MMSE and CDR-SB in models including all three.
Conclusion:
pFCSRT + IR + IR scores may be more sensitive to early pathological changes than either the CDR-SB or the MMSE.
Objective
Temporal lobe epilepsy is a common form of epilepsy that might be amenable to surgery. However, magnetic resonance imaging (MRI)‐negative hippocampal sclerosis (HS) can hamper early diagnosis and surgical intervention for patients in clinical practice, resulting in disease progression. Our aim was to automatically detect and evaluate the structural alterations of HS.
Methods
Eighty patients with pharmacoresistant epilepsy and histologically proven HS and 80 healthy controls were included in the study. Two automated classifiers relying on clinically empirical and radiomics features were developed to detect HS. Cross‐validation was implemented on all participants, and specificity was assessed in the 80 controls. The performance, robustness, and clinical utility of the model were also evaluated. Structural analysis was performed to investigate the morphological abnormalities of HS.
Results
The computational model based on clinical empirical features showed excellent performance, with an area under the curve (AUC) of 0.981 in the primary cohort and 0.993 in the validation cohort. One of the features, gray‐white matter boundary blurring in the temporal pole, exhibited the highest weight in model performance. Another model based on radiomics features also showed satisfactory performance, with AUC of 0.997 in the primary cohort and 0.978 in the validation cohort. In particular, the model improved the detection rate of MRI‐negative HS to 96.0%. The novel feature of cortical folding complexity of the temporal pole not only played a crucial role in the classifier but also had significant correlation with disease duration.
Significance
Machine learning with quantitative clinical and radiomics features is shown to improve HS detection. HS‐related structural alterations were similar in the MRI‐positive and MRI‐negative HS patient groups, indicating that misdiagnosis originates mainly from empirical interpretation. The cortical folding complexity of the temporal pole is a potentially valuable feature for exploring the nature of HS.
Introduction:
Alzheimer's disease diagnosis requires postmortem visualization of amyloid and tau deposits. As brain atrophy can provide assessment of consequent neurodegeneration, our objective was to predict postmortem neurofibrillary tangles (NFT) from in vivo MRI measurements.
Methods:
All participants with neuroimaging and neuropathological data from the Alzheimer's Disease Neuroimaging Initiative, the National Alzheimer's Coordinating Center and the Rush Memory and Aging Project were selected (n = 186). Two hundred and thirty two variables were extracted from last MRI before death using FreeSurfer. Nonparametric correlation analysis and multivariable support vector machine classification were performed to provide a predictive model of Braak NFT staging.
Results:
We demonstrated that 59 of our MRI variables, mostly temporal lobe structures, were significantly associated with Braak NFT stages (P < .005). We obtained a 62.4% correct classification rate for discrimination between transentorhinal, limbic, and isocortical groups.
Discussion:
Structural neuroimaging may therefore be considered as a potential biomarker for early detection of Alzheimer's disease-associated neurofibrillary degeneration.
Limbic‐predominant age‐related TAR‐DNA‐binding protein‐43 (TDP‐43) encephalopathy with hippocampal sclerosis pathology (LATE‐NC+HS) is a neurodegenerative disorder characterized by severe hippocampal CA1 neuron loss and TDP‐43‐pathology, leading to cognitive dysfunction and dementia. Polymorphisms in GRN, TMEM106B and ABCC9 are proposed as LATE‐NC+HS risk factors in brain bank collections. To replicate these results in independent population‐representative cohorts, hippocampal sections from brains donated to three such studies (Cambridge City over 75‐Cohort [CC75C], Cognitive Function and Ageing Study [CFAS], and Vantaa 85+ Study) were stained with haematoxylin‐eosin (n=744) and anti‐pTDP‐43 (n=713), and evaluated for LATE‐NC+HS and TDP‐43 pathology. Single nucleotide polymorphism genotypes in GRN rs5848, TMEM106B rs1990622 and ABCC9 rs704178 were determined. LATE‐NC+HS (n=58) was significantly associated with the GRN rs5848 genotype (χ2(2)=20.61, p<0.001) and T‐allele (χ2(1)=21.04, p<0.001), and TMEM106B rs1990622 genotype (Fisher′s exact test, p<0.001) and A‐allele (χ2(1)=25.75, p<0.001). No differences in ABCC9 rs704178 genotype or allele frequency were found between LATE‐NC+HS and non‐ LATE‐NC+HS neuropathology cases. Dentate gyrus TDP‐43 pathology associated with GRN and TMEM106B variations, but the association with TMEM106B nullified when LATE‐NC+HS cases were excluded. Our results indicate that GRN and TMEM106B are associated with severe loss of CA1 neurons in the aging brain, while ABCC9 was not confirmed as a genetic risk factor for LATE‐NC+HS. The association between TMEM106B and LATE‐NC+HS may be independent of dentate TDP‐43 pathology. This article is protected by copyright. All rights reserved.
Background:
Neuropathology has demonstrated a high rate of comorbid pathology in dementia due to Alzheimer's disease (ADD). The most common major comorbidity is Lewy body disease (LBD), either as dementia with Lewy bodies (AD-DLB) or Alzheimer's disease with Lewy bodies (AD-LB), the latter representing subjects with ADD and LBD not meeting neuropathological distribution and density thresholds for DLB. Although it has been established that ADD subjects with undifferentiated LBD have a more rapid cognitive decline than those with ADD alone, it is still unknown whether AD-LB subjects, who represent the majority of LBD and approximately one-third of all those with ADD, have a different clinical course.
Methods:
Subjects with dementia included those with "pure" ADD (n = 137), AD-DLB (n = 64) and AD-LB (n = 114), all with two or more complete Mini Mental State Examinations (MMSE) and a full neuropathological examination.
Results:
Linear mixed models assessing MMSE change showed that the AD-LB group had significantly greater decline compared to the ADD group (β = -0.69, 95% CI: -1.05, -0.33, p<0.001) while the AD-DLB group did not (β = -0.30, 95% CI: -0.73, 0.14, p = 0.18). Of those with AD-DLB and AD-LB, only 66% and 2.1%, respectively, had been diagnosed with LBD at any point during their clinical course. Compared with clinically-diagnosed AD-DLB subjects, those that were clinically undetected had significantly lower prevalences of parkinsonism (p = 0.046), visual hallucinations (p = 0.0008) and dream enactment behavior (0.013).
Conclusions:
The probable cause of LBD clinical detection failure is the lack of a sufficient set of characteristic core clinical features. Core DLB clinical features were not more common in AD-LB as compared to ADD. Clinical identification of ADD with LBD would allow stratified analyses of ADD clinical trials, potentially improving the probability of trial success.
Neuropathologists assess vast brain areas to identify diverse and subtly-differentiated morphologies. Standard semi-quantitative scoring approaches, however, are coarse-grained and lack precise neuroanatomic localization. We report a proof-of-concept deep learning pipeline that identifies specific neuropathologies—amyloid plaques and cerebral amyloid angiopathy—in immunohistochemically-stained archival slides. Using automated segmentation of stained objects and a cloud-based interface, we annotate > 70,000 plaque candidates from 43 whole slide images (WSIs) to train and evaluate convolutional neural networks. Networks achieve strong plaque classification on a 10-WSI hold-out set (0.993 and 0.743 areas under the receiver operating characteristic and precision recall curve, respectively). Prediction confidence maps visualize morphology distributions at high resolution. Resulting network-derived amyloid beta (Aβ)-burden scores correlate well with established semi-quantitative scores on a 30-WSI blinded hold-out. Finally, saliency mapping demonstrates that networks learn patterns agreeing with accepted pathologic features. This scalable means to augment a neuropathologist’s ability suggests a route to neuropathologic deep phenotyping.
We describe a recently recognized disease entity, limbic-predominant age-related TDP-43 encephalopathy (LATE). LATE neuropathological change (LATE-NC) is defined by a stereotypical TDP-43 proteinopathy in older adults, with or without coexisting hippocampal sclerosis pathology. LATE-NC is a common TDP-43 proteinopathy, associated with an amnestic dementia syndrome that mimicked Alzheimer's-type dementia in retrospective autopsy studies. LATE is distinguished from frontotemporal lobar degeneration with TDP-43 pathology based on its epidemiology (LATE generally affects older subjects), and relatively restricted neuroanatomical distribution of TDP-43 proteinopathy. In community-based autopsy cohorts, ∼25% of brains had sufficient burden of LATE-NC to be associated with discernible cognitive impairment. Many subjects with LATE-NC have comorbid brain pathologies, often including amyloid-β plaques and tauopathy. Given that the 'oldest-old' are at greatest risk for LATE-NC, and subjects of advanced age constitute a rapidly growing demographic group in many countries, LATE has an expanding but under-recognized impact on public health. For these reasons, a working group was convened to develop diagnostic criteria for LATE, aiming both to stimulate research and to promote awareness of this pathway to dementia. We report consensus-based recommendations including guidelines for diagnosis and staging of LATE-NC. For routine autopsy workup of LATE-NC, an anatomically-based preliminary staging scheme is proposed with TDP-43 immunohistochemistry on tissue from three brain areas, reflecting a hierarchical pattern of brain involvement: amygdala, hippocampus, and middle frontal gyrus. LATE-NC appears to affect the medial temporal lobe structures preferentially, but other areas also are impacted. Neuroimaging studies demonstrated that subjects with LATE-NC also had atrophy in the medial temporal lobes, frontal cortex, and other brain regions. Genetic studies have thus far indicated five genes with risk alleles for LATE-NC: GRN, TMEM106B, ABCC9, KCNMB2, and APOE. The discovery of these genetic risk variants indicate that LATE shares pathogenetic mechanisms with both frontotemporal lobar degeneration and Alzheimer's disease, but also suggests disease-specific underlying mechanisms. Large gaps remain in our understanding of LATE. For advances in prevention, diagnosis, and treatment, there is an urgent need for research focused on LATE, including in vitro and animal models. An obstacle to clinical progress is lack of diagnostic tools, such as biofluid or neuroimaging biomarkers, for ante-mortem detection of LATE. Development of a disease biomarker would augment observational studies seeking to further define the risk factors, natural history, and clinical features of LATE, as well as eventual subject recruitment for targeted therapies in clinical trials.
Accumulation of abnormal tau in neurofibrillary tangles (NFT) occurs in Alzheimer disease (AD) and a spectrum of tauopathies. These tauopathies have diverse and overlapping morphological phenotypes that obscure classification and quantitative assessments. Recently, powerful machine learning-based approaches have emerged, allowing the recognition and quantification of pathological changes from digital images. Here, we applied deep learning to the neuropathological assessment of NFT in postmortem human brain tissue to develop a classifier capable of recognizing and quantifying tau burden. The histopathological material was derived from 22 autopsy brains from patients with tauopathies. We used a custom web-based informatics platform integrated with an in-house information management system to manage whole slide images (WSI) and human expert annotations as ground truth. We utilized fully annotated regions to train a deep learning fully convolutional neural network (FCN) implemented in PyTorch against the human expert annotations. We found that the deep learning framework is capable of identifying and quantifying NFT with a range of staining intensities and diverse morphologies. With our FCN model, we achieved high precision and recall in naive WSI semantic segmentation, correctly identifying tangle objects using a SegNet model trained for 200 epochs. Our FCN is efficient and well suited for the practical application of WSIs with average processing times of 45 min per WSI per GPU, enabling reliable and reproducible large-scale detection of tangles. We measured performance on test data of 50 pre-annotated regions on eight naive WSI across various tauopathies, resulting in the recall, precision, and an F1 score of 0.92, 0.72, and 0.81, respectively. Machine learning is a useful tool for complex pathological assessment of AD and other tauopathies. Using deep learning classifiers, we have the potential to integrate cell- and region-specific annotations with clinical, genetic, and molecular data, providing unbiased data for clinicopathological correlations that will enhance our knowledge of the neurodegeneration.
TAR DNA binding protein 43 (TDP-43) is a versatile RNA/DNA binding protein involved in RNA-related metabolism. Hyper-phosphorylated and ubiquitinated TDP-43 deposits act as inclusion bodies in the brain and spinal cord of patients with the motor neuron diseases: amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD). While the majority of ALS cases (90–95%) are sporadic (sALS), among familial ALS cases 5–10% involve the inheritance of mutations in the TARDBP gene and the remaining (90–95%) are due to mutations in other genes such as: C9ORF72, SOD1, FUS, and NEK1 etc. Strikingly however, the majority of sporadic ALS patients (up to 97%) also contain the TDP-43 protein deposited in the neuronal inclusions, which suggests of its pivotal role in the ALS pathology. Thus, unraveling the molecular mechanisms of the TDP-43 pathology seems central to the ALS therapeutics, hence, we comprehensively review the current understanding of the TDP-43's pathology in ALS. We discuss the roles of TDP-43's mutations, its cytoplasmic mis-localization and aberrant post-translational modifications in ALS. Also, we evaluate TDP-43's amyloid-like in vitro aggregation, its physiological vs. pathological oligomerization in vivo, liquid-liquid phase separation (LLPS), and potential prion-like propagation propensity of the TDP-43 inclusions. Finally, we describe the various evolving TDP-43-induced toxicity mechanisms, such as the impairment of endocytosis and mitotoxicity etc. and also discuss the emerging strategies toward TDP-43 disaggregation and ALS therapeutics.
Importance:
There are currently no proven treatments to reduce the risk of mild cognitive impairment and dementia.
Objective:
To evaluate the effect of intensive blood pressure control on risk of dementia.
Design, setting, and participants:
Randomized clinical trial conducted at 102 sites in the United States and Puerto Rico among adults aged 50 years or older with hypertension but without diabetes or history of stroke. Randomization began on November 8, 2010. The trial was stopped early for benefit on its primary outcome (a composite of cardiovascular events) and all-cause mortality on August 20, 2015. The final date for follow-up of cognitive outcomes was July 22, 2018.
Interventions:
Participants were randomized to a systolic blood pressure goal of either less than 120 mm Hg (intensive treatment group; n = 4678) or less than 140 mm Hg (standard treatment group; n = 4683).
Main outcomes and measures:
The primary cognitive outcome was occurrence of adjudicated probable dementia. Secondary cognitive outcomes included adjudicated mild cognitive impairment and a composite outcome of mild cognitive impairment or probable dementia.
Results:
Among 9361 randomized participants (mean age, 67.9 years; 3332 women [35.6%]), 8563 (91.5%) completed at least 1 follow-up cognitive assessment. The median intervention period was 3.34 years. During a total median follow-up of 5.11 years, adjudicated probable dementia occurred in 149 participants in the intensive treatment group vs 176 in the standard treatment group (7.2 vs 8.6 cases per 1000 person-years; hazard ratio [HR], 0.83; 95% CI, 0.67-1.04). Intensive BP control significantly reduced the risk of mild cognitive impairment (14.6 vs 18.3 cases per 1000 person-years; HR, 0.81; 95% CI, 0.69-0.95) and the combined rate of mild cognitive impairment or probable dementia (20.2 vs 24.1 cases per 1000 person-years; HR, 0.85; 95% CI, 0.74-0.97).
Conclusions and relevance:
Among ambulatory adults with hypertension, treating to a systolic blood pressure goal of less than 120 mm Hg compared with a goal of less than 140 mm Hg did not result in a significant reduction in the risk of probable dementia. Because of early study termination and fewer than expected cases of dementia, the study may have been underpowered for this end point.
Trial registration:
ClinicalTrials.gov Identifier: NCT01206062.
Background:
We developed multifactorial models for predicting incident dementia and brain pathology in the oldest old using the Vantaa 85+ cohort.
Methods:
We included participants without dementia at baseline and at least 2 years of follow-up (N = 245) for dementia prediction or with autopsy data (N = 163) for pathology. A supervised machine learning method was used for model development, considering sociodemographic, cognitive, clinical, vascular, and lifestyle factors, as well as APOE genotype. Neuropathological assessments included β-amyloid, neurofibrillary tangles and neuritic plaques, cerebral amyloid angiopathy (CAA), macro- and microscopic infarcts, α-synuclein pathology, hippocampal sclerosis, and TDP-43.
Results:
Prediction model performance was evaluated using AUC for 10 × 10-fold cross-validation. Overall AUCs were 0.73 for dementia, 0.64-0.68 for Alzheimer's disease (AD)- or amyloid-related pathologies, 0.72 for macroinfarcts, and 0.61 for microinfarcts. Predictors for dementia were different from those in previous reports of younger populations; for example, age, sex, and vascular and lifestyle factors were not predictive. Predictors for dementia versus pathology were also different, because cognition and education predicted dementia but not AD- or amyloid-related pathologies. APOE genotype was most consistently present across all models. APOE alleles had a different impact: ε4 did not predict dementia, but it did predict all AD- or amyloid-related pathologies; ε2 predicted dementia, but it was protective against amyloid and neuropathological AD; and ε3ε3 was protective against dementia, neurofibrillary tangles, and CAA. Very few other factors were predictive of pathology.
Conclusions:
Differences between predictors for dementia in younger old versus oldest old populations, as well as for dementia versus pathology, should be considered more carefully in future studies.
The objective of this article is to review and integrate interrelated areas of research on personality and Alzheimer’s disease (AD). Prospective studies indicate that individuals who score higher on conscientiousness (more responsible and self-disciplined) and lower on neuroticism (less anxious and vulnerable to stress) have a reduced risk of developing dementia, even in the presence of AD neuropathology. Personality is also related to measures of cognitive performance and cognitive decline, with effect sizes similar to those of other clinical, lifestyle, and behavioral risk factors. These associations are unlikely to be due to reverse causality: Long-term prospective data indicate that there are no changes in personality that are an early sign of the disease during the preclinical phase of AD. With the onset and progression of dementia, however, there are large changes in personality that are reported consistently by caregivers in retrospective studies and are consistent with the clinical criteria for the diagnosis of dementia. The review also discusses potential mechanisms of the observed associations and emphasizes the need for prospective studies to elucidate the interplay of personality traits with AD neuropathology (amyloid and tau biomarkers) in modulating the risk and timing of onset of clinical dementia. The article concludes with the implications of personality research for identifying those at greater risk of AD and the potential of personality-tailored interventions aimed at the prevention and treatment of AD.
Background: Alzheimer’s disease (AD) is the most common form of dementia. While neuropathological changes pathognomonic for AD have been defined, early detection of AD prior to cognitive impairment in the clinical setting is still lacking. Pioneer studies applying machine learning to magnetic-resonance imaging (MRI) data to predict mild cognitive impairment (MCI) or AD have yielded high accuracies, however, an algorithm predicting neuropathological change is still lacking. The objective of this study was to compute a prediction model supporting a more distinct diagnostic criterium for AD compared to clinical presentation, allowing identification of hallmark changes even before symptoms occur.
Methods: Autopsy verified neuropathological changes attributed to AD, as described by a combined score for Aβ-peptides, neurofibrillary tangles and neuritic plaques issued by the National Institute on Aging – Alzheimer’s Association (NIAA), the ABC score for AD, were predicted from structural MRI data with RandomForest (RF). MRI scans were performed at least 2 years prior to death. All subjects derive from the prospective Vienna Trans-Danube Aging (VITA) study that targeted all 1750 inhabitants of the age of 75 in the starting year of 2000 in two districts of Vienna and included irregular follow-ups until death, irrespective of clinical symptoms or diagnoses. For 68 subjects MRI as well as neuropathological data were available and 49 subjects (mean age at death: 82.8 ± 2.9, 29 female) with sufficient MRI data quality were enrolled for further statistical analysis using nested cross-validation (CV). The decoding data of the inner loop was used for variable selection and parameter optimization with a fivefold CV design, the new data of the outer loop was used for model validation with optimal settings in a fivefold CV design. The whole procedure was performed ten times and average accuracies with standard deviations were reported.
Results: The most informative ROIs included caudal and rostral anterior cingulate gyrus, entorhinal, fusiform and insular cortex and the subcortical ROIs anterior corpus callosum and the left vessel, a ROI comprising lacunar alterations in inferior putamen and pallidum. The resulting prediction models achieved an average accuracy for a three leveled NIAA AD score of 0.62 within the decoding sets and of 0.61 for validation sets. Higher accuracies of 0.77 for both sets, respectively, were achieved when predicting presence or absence of neuropathological change.
Conclusion: Computer-aided prediction of neuropathological change according to the categorical NIAA score in AD, that currently can only be assessed post-mortem, may facilitate a more distinct and definite categorization of AD dementia. Reliable detection of neuropathological hallmarks of AD would enable risk stratification at an earlier level than prediction of MCI or clinical AD symptoms and advance precision medicine in neuropsychiatry.
Clarifying the relationships between neuropsychiatric symptoms and Alzheimer's disease (AD)-related pathology may open avenues for effective treatments. Here, we investigate the odds of developing neuropsychiatric symptoms across increasing burdens of neurofibrillary tangle and amyloid-β pathology. Participants who passed away between 2004 and 2014 underwent comprehensive neuropathologic evaluation at the Biobank for Aging Studies from the Faculty of Medicine at the University of São Paulo. Postmortem interviews with reliable informants were used to collect information regarding neuropsychiatric and cognitive status. Of 1,092 cases collected, those with any non-Alzheimer pathology were excluded, bringing the cohort to 455 cases. Braak staging was used to evaluate neurofibrillary tangle burden, and the CERAD neuropathology score was used to evaluate amyloid-β burden. The 12-item neuropsychiatric inventory was used to evaluate neuropsychiatric symptoms and CDR-SOB score was used to evaluate dementia status. In Braak I/II, significantly increased odds were detected for agitation, anxiety, appetite changes, depression, and sleep disturbances, compared to controls. Increased odds of agitation continue into Braak III/IV. Braak V/VI is associated with higher odds for delusions. No increased odds for neuropsychiatric symptoms were found to correlate with amyloid-β pathology. Increased odds of neuropsychiatric symptoms are associated with early neurofibrillary tangle pathology, suggesting that subcortical neurofibrillary tangle accumulation with minimal cortical pathology is sufficient to impact quality of life and that neuropsychiatric symptoms are a manifestation of AD biological processes.
Background:
Previous evidence linking diabetes to Alzheimer's disease (AD) neuropathology is mixed and scant data are available from low- and middle-income countries.
Objective:
To investigate the association between diabetes and AD neuropathology in a large autopsy study of older Brazilian adults.
Methods:
In this cross-sectional study, diabetes was defined by diagnosis during life or use of antidiabetic medication. A standardized neuropathological examination was performed using immunohistochemistry. The associations of diabetes with Consortium to Establish and Registry for Alzheimer Disease (CERAD) scores for neuritic plaques and Braak-Braak (BB) scores for neurofibrillary tangles were investigated using multivariable ordinal logistic regression. We investigated effect modification of education, race, and APOE on these associations.
Results:
Among 1,037 subjects (mean age = 74.4±11.5 y; mean education = 4.0±3.7 y; 48% male, 61% White), diabetes was present in 279 subjects. Diabetes was not associated with BB (OR = 1.12, 95% CI = 0.81-1.54, p = 0.48) or with CERAD (OR = 0.97, 95% CI = 0.68-1.38, p = 0.86) scores on analyses adjusted for sociodemographic and clinical variables. We observed effect modification by the APOE allele ɛ4 on the association between diabetes mellitus and BB scores.
Conclusion:
No evidence of an association between diabetes and AD neuropathology was found in a large sample of Brazilians; however, certain subgroups, such as APOE allele ɛ4 carriers, had higher odds of accumulation of neurofibrillary tangles.
In the study of various diseases, heterogeneity among patients usually leads to different progression patterns and may require different types of therapeutic intervention. Therefore, it is important to study patient subtyping, which is grouping of patients into disease characterizing subtypes. Subtyping from complex patient data is challenging because of the information heterogeneity and temporal dynamics. Long-Short Term Memory (LSTM) has been successfully used in many domains for processing sequential data, and recently applied for analyzing longitudinal patient records. The LSTM units are designed to handle data with constant elapsed times between consecutive elements of a sequence. Given that time lapse between successive elements in patient records can vary from days to months, the design of traditional LSTM may lead to suboptimal performance. In this paper, we propose a novel LSTM unit called Time-Aware LSTM (T-LSTM) to handle irregular time intervals in longitudinal patient records. We learn a subspace decomposition of the cell memory which enables time decay to discount the memory content according to the elapsed time. We propose a patient subtyping model that leverages the proposed T-LSTM in an auto-encoder to learn a powerful single representation for sequential records of patients, which are then used to cluster patients into clinical subtypes. Experiments on synthetic and real world datasets show that the proposed T-LSTM architecture captures the underlying structures in the sequences with time irregularities.
The Dementia with Lewy Bodies (DLB) Consortium has refined its recommendations about the clinical and pathologic diagnosis of DLB, updating the previous report, which has been in widespread use for the last decade. The revised DLB consensus criteria now distinguish clearly between clinical features and diagnostic biomarkers, and give guidance about optimal methods to establish and interpret these. Substantial new information has been incorporated about previously reported aspects of DLB, with increased diagnostic weighting given to REM sleep behavior disorder and (123)iodine-metaiodobenzylguanidine (MIBG) myocardial scintigraphy. The diagnostic role of other neuroimaging, electrophysiologic, and laboratory investigations is also described. Minor modifications to pathologic methods and criteria are recommended to take account of Alzheimer disease neuropathologic change, to add previously omitted Lewy-related pathology categories, and to include assessments for substantia nigra neuronal loss. Recommendations about clinical management are largely based upon expert opinion since randomized controlled trials in DLB are few. Substantial progress has been made since the previous report in the detection and recognition of DLB as a common and important clinical disorder. During that period it has been incorporated into DSM-5, as major neurocognitive disorder with Lewy bodies. There remains a pressing need to understand the underlying neurobiology and pathophysiology of DLB, to develop and deliver clinical trials with both symptomatic and disease-modifying agents, and to help patients and carers worldwide to inform themselves about the disease, its prognosis, best available treatments, ongoing research, and how to get adequate support.
Objective:
Dementia with Lewy bodies (DLB) is associated with a range of cognitive and non-cognitive symptoms. We aimed to identify if some of these symptoms might aid early diagnosis of Lewy body disease in cases of mild cognitive impairment (MCI).
Methods:
Lewy body MCI (MCI-LB; n = 36), Alzheimer's disease MCI (MCI-AD; n = 21), DLB (n = 36), AD (n = 21) and control (n = 20) participants were recruited. An interview-based questionnaire about the presence of symptoms thought to be associated with Lewy body disease was completed by participants with, where possible, their carer/relative. The prevalence of each symptom was compared between MCI-LB and MCI-AD and between established DLB and AD, and a symptom scale based on these findings was devised.
Results:
Fluctuating concentration/attention; episodes of confusion; muscle rigidity; changes in hand-writing, gait and posture; falls; drooling; weak voice; symptoms of REM sleep behaviour disorder (RBD) and misjudging objects were more common in MCI-LB compared with MCI-AD, and also in DLB compared with AD. Hyposmia, tremor, slowness and autonomic symptoms were not specific to Lewy body disease. REM sleep behaviour disorder and hyposmia were reported to develop several years prior to the onset of cognitive symptoms in Lewy body disease. A 10-point symptom scale differentiated between MCI-LB and MCI-AD with a sensitivity of 83% and a specificity of 100%.
Conclusions:
Drooling, misjudging objects and symptoms related to parkinsonism, fluctuating cognition and RBD may be the most characteristic symptoms of MCI-LB. Slowness, tremor, autonomic symptoms and hyposmia are all common in MCI-LB but are not specific to the disease. Copyright © 2017 John Wiley & Sons, Ltd.
Hyperphosphorylated transactive response DNA-binding protein 43 (TDP-43, encoded by TARDBP) proteinopathy has recently been described in ageing and in association with cognitive impairment, especially in the context of Alzheimer’s disease pathology. To explore the role of mixed Alzheimer’s disease and TDP-43 pathologies in clinical Alzheimer’s-type dementia, we performed a comprehensive investigation of TDP-43, mixed pathologies, and clinical Alzheimer’s-type dementia in a large cohort of community-dwelling older subjects. We tested the hypotheses that TDP-43 with Alzheimer’s disease pathology is a common mixed pathology; is related to increased likelihood of expressing clinical Alzheimer’s-type dementia; and that TDP-43 pathologic stage is an important determinant of clinical Alzheimer’s-type dementia. Data came from 946 older adults with (n = 398) and without dementia (n = 548) from the Rush Memory and Aging Project and Religious Orders Study. TDP-43 proteinopathy (cytoplasmic inclusions) was present in 496 (52%) subjects, and the pattern of deposition was classified as stage 0 (none; 48%), stage 1 (amygdala; 18%), stage 2 (extension to hippocampus/entorhinal; 21%), or stage 3 (extension to neocortex; 14%). TDP-43 pathology combined with a pathologic diagnosis of Alzheimer’s disease was a common mixed pathology (37% of all participants), and the proportion of subjects with clinical Alzheimer’s-type dementia formerly labelled ‘pure pathologic diagnosis of Alzheimer’s disease’ was halved when TDP-43 was considered. In logistic regression models adjusted for age, sex, and education, TDP-43 pathology was associated with clinical Alzheimer’s-type dementia (odds ratio = 1.51, 95% confidence interval = 1.11, 2.05) independent of pathological Alzheimer’s disease (odds ratio = 4.30, 95% confidence interval = 3.08, 6.01) or other pathologies (infarcts, arteriolosclerosis, Lewy bodies, and hippocampal sclerosis). Mixed Alzheimer’s disease and TDP-43 pathologies were associated with higher odds of clinical Alzheimer’s-type dementia (odds ratio = 6.73, 95% confidence interval = 4.18, 10.85) than pathologic Alzheimer’s disease alone (odds ratio = 4.62, 95% confidence interval = 2.84, 7.52). In models examining TDP-43 stage, a dose-response relationship with clinical Alzheimer’s-type dementia was observed, and a significant association was observed starting at stage 2, extension beyond the amygdala. In this large sample from almost 1000 community participants, we observed that TDP-43 proteinopathy was very common, frequently mixed with pathological Alzheimer’s disease, and associated with a higher likelihood of the clinical expression of clinical Alzheimer’s-type dementia but only when extended beyond the amygdala.
Objectives:
Prominent impairment of visuospatial processing is a feature of dementia with Lewy bodies (DLB), and diagnosis of this impairment may help clinically distinguish DLB from Alzheimer's disease (AD). The current study compared autopsy-confirmed DLB and AD patients on the Hooper Visual Organization Test (VOT), a test that requires perceptual and mental reorganization of parts of an object into an identifiable whole. The VOT may be particularly sensitive to DLB since it involves integration of visual information processed in separate dorsal and ventral visual "streams".
Methods:
Demographically similar DLB (n=28), AD (n=115), and normal control (NC; n=85) participants were compared on the VOT and additional neuropsychological tests. Patient groups did not differ in dementia severity at time of VOT testing. High and Low AD-Braak stage DLB subgroups were compared to examine the influence of concomitant AD pathology on VOT performance.
Results:
Both patient groups were impaired compared to NC participants. VOT scores of DLB patients were significantly lower than those of AD patients. The diagnostic sensitivity and specificity of the VOT for patients versus controls was good, but marginal for DLB versus AD. High-Braak and low-Braak DLB patients did not differ on the VOT, but High-Braak DLB performed worse than Low-Braak DLB on tests of episodic memory and language.
Conclusions:
Visual perceptual organization ability is more impaired in DLB than AD but not strongly diagnostic. The disproportionate severity of this visual perceptual deficit in DLB is not related to degree of concomitant AD pathology, which suggests that it might primarily reflect Lewy body pathology. (JINS, 2016, 22, 1-11).
Objective:
To examine frequencies and relationships of 5 common neuropathologic abnormalities identified at autopsy with late-life cognitive impairment and dementia in 2 different autopsy panels.
Methods:
The Nun Study (NS) and the Honolulu-Asia Aging Study (HAAS) are population-based investigations of brain aging that included repeated cognitive assessments and comprehensive brain autopsies. The neuropathologic abnormalities assessed were Alzheimer disease (AD) neuropathologic changes, neocortical Lewy bodies (LBs), hippocampal sclerosis, microinfarcts, and low brain weight. Associations with screening tests for cognitive impairment were examined.
Results:
Neuropathologic abnormalities occurred at levels ranging from 9.7% to 43%, and were independently associated with cognitive impairment in both studies. Neocortical LBs and AD changes were more frequent among the predominantly Caucasian NS women, while microinfarcts were more common in the Japanese American HAAS men. Comorbidity was usual and very strongly associated with cognitive impairment. Apparent cognitive resilience (no cognitive impairment despite Braak stage V) was strongly associated with minimal or no comorbid abnormalities, with fewer neocortical AD lesions, and weakly with longer interval between final testing and autopsy.
Conclusions:
Total burden of comorbid neuropathologic abnormalities, rather than any single lesion type, was the most relevant determinant of cognitive impairment in both cohorts, often despite clinical diagnosis of only AD. These findings emphasize challenges to dementia pathogenesis and intervention research and to accurate diagnoses during life.
Dementia with Lewy bodies (DLB) is the second most common type of degenerative dementia following Alzheimer's disease (AD). DLB is clinically and pathologically related to Parkinson's disease (PD) and PD dementia, and the three disorders can be viewed as existing on a spectrum of Lewy body disease. In recent years there has been a concerted effort to establish the phenotypes of AD and PD in the prodromal phase (before the respective syndromes of cognitive and motor impairment are expressed). Evidence for the prodromal presentation of DLB is also emerging. This paper briefly reviews what is known about the clinical presentation of prodromal DLB before discussing the pathology of Lewy body disease and how this relates to potential biomarkers of prodromal DLB. The presenting features of DLB can be broadly placed in three categories: cognitive impairment (particularly nonamnestic cognitive impairments), behavioural/psychiatric phenomena (for example, hallucinations, rapid eye movement sleep behaviour disorder (RBD)) and physical symptoms (for example, parkinsonism, decreased sense of smell, autonomic dysfunction). Some noncognitive symptoms such as constipation, RBD, hyposmia and postural dizziness can predate the onset of memory impairment by several years in DLB. Pathological studies of Lewy body disease have found that the earliest sites of involvement are the olfactory bulb, the dorsal motor nucleus of the vagal nerve, the peripheral autonomic nervous system, including the enteric nervous system, and the brainstem. Some of the most promising early markers for DLB include the presence of RBD, autonomic dysfunction or hyposmia, (123)I-metaiodobenzylguanidine cardiac scintigraphy, measures of substantia nigra pathology and skin biopsy for α-synuclein in peripheral autonomic nerves. In the absence of disease-modifying therapies, the diagnosis of prodromal DLB is of limited use in the clinic. That said, knowledge of the prodromal development of DLB could help clinicians identify cases of DLB where the diagnosis is uncertain. Prodromal diagnosis is of great importance in research, where identifying Lewy body disease at an earlier stage may allow researchers to investigate the initial phases of dementia pathophysiology, develop treatments designed to interrupt the development of the dementia syndrome and accurately identify the patients most likely to benefit from these treatments.
TDP-43 immunoreactivity occurs in 19-57 % of Alzheimer's disease (AD) cases. Two patterns of TDP-43 deposition in AD have been described involving hippocampus (limbic) or hippocampus and neocortex (diffuse), although focal amygdala involvement has been observed. In 195 AD cases with TDP-43, we investigated regional TDP-43 immunoreactivity with the aim of developing a TDP-43 in AD staging scheme. TDP-43 immunoreactivity was assessed in amygdala, entorhinal cortex, subiculum, hippocampal dentate gyrus, occipitotemporal, inferior temporal and frontal cortices, and basal ganglia. Clinical, neuroimaging, genetic and pathological characteristics were assessed across stages. Five stages were identified: stage I showed scant-sparse TDP-43 in the amygdala only (17 %); stage II showed moderate-frequent amygdala TDP-43 with spread into entorhinal and subiculum (25 %); stage III showed further spread into dentate gyrus and occipitotemporal cortex (31 %); stage IV showed further spread into inferior temporal cortex (20 %); and stage V showed involvement of frontal cortex and basal ganglia (7 %). Cognition and medial temporal volumes differed across all stages and progression across stages correlated with worsening cognition and medial temporal volume loss. Compared to 147 AD patients without TDP-43, only the Boston Naming Test showed abnormalities in stage I. The findings demonstrate that TDP-43 deposition in AD progresses in a stereotypic manner that can be divided into five distinct topographic stages which are supported by correlations with clinical and neuroimaging features. Given these findings, we recommend sequential regional TDP-43 screening in AD beginning with the amygdala.
Clinico-pathological correlation studies and positron emission tomography amyloid imaging studies have shown that some individuals can tolerate substantial amounts of Alzheimer's pathology in their brains without experiencing dementia. Few details are known about the neuropathological phenotype of these unique cases that might prove relevant to understanding human resilience to Alzheimer's pathology. We conducted detailed quantitative histopathological and biochemical assessments on brains from non-demented individuals before death whose brains were free of substantial Alzheimer's pathology, non-demented individuals before death but whose post-mortem examination demonstrated significant amounts of Alzheimer's changes ('mismatches'), and demented Alzheimer's cases. Quantification of amyloid-β plaque burden, stereologically-based counts of neurofibrillary tangles, neurons and reactive glia, and morphological analyses of axons were performed in the multimodal association cortex lining the superior temporal sulcus. Levels of synaptic integrity markers, and soluble monomeric and multimeric amyloid-β and tau species were measured. Our results indicate that some individuals can accumulate equivalent loads of amyloid-β plaques and tangles to those found in demented Alzheimer's cases without experiencing dementia. Analyses revealed four main phenotypic differences among these two groups: (i) mismatches had striking preservation of neuron numbers, synaptic markers and axonal geometry compared to demented cases; (ii) demented cases had significantly higher burdens of fibrillar thioflavin-S-positive plaques and of oligomeric amyloid-β deposits reactive to conformer-specific antibody NAB61 than mismatches; (iii) strong and selective accumulation of hyperphosphorylated soluble tau multimers into the synaptic compartment was noted in demented cases compared with controls but not in mismatches; and (iv) the robust glial activation accompanying amyloid-β and tau pathologies in demented cases was remarkably reduced in mismatches. Further biochemical measurements of soluble amyloid-β species-monomers, dimers and higher molecular weight oligomers-in total brain homogenates and synaptoneurosomal preparations failed to demonstrate significant differences between mismatches and demented cases. Together, these data suggest that amyloid-β plaques and tangles do not inevitably result in neural system derangement and dementia in all individuals. We identified distinct phenotypic characteristics in the profile of brain fibrillar and soluble amyloid-β and tau accrual and in the glial response that discriminated demented and non-demented individuals with high loads of Alzheimer's pathology. Amyloid-β deposition in the form of fibrillar plaques and intimately related oligomeric amyloid-β assemblies, hyperphosphorylated soluble tau species localized in synapses, and glial activation emerged in this series as likely mediators of neurotoxicity and altered cognition, providing further insight into factors and pathways potentially involved in human susceptibility or resilience to Alzheimer's pathological changes.
Diagnostic processes of Alzheimer's disease (AD) are evolving. Knowledge about disease-specific biomarkers is constantly increasing and larger volumes of data are being measured from patients. To gain additional benefits from the collected data, a novel statistical modeling and data visualization system is proposed for supporting clinical diagnosis of AD. The proposed system computes an evidence-based estimate of a patient's AD state by comparing his or her heterogeneous neuropsychological, clinical, and biomarker data to previously diagnosed cases. The AD state in this context denotes a patient's degree of similarity to previously diagnosed disease population. A summary of patient data and results of the computation are displayed in a succinct Disease State Fingerprint (DSF) visualization. The visualization clearly discloses how patient data contributes to the AD state, facilitating rapid interpretation of the information. To model the AD state from complex and heterogeneous patient data, a statistical Disease State Index (DSI) method underlying the DSF has been developed. Using baseline data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), the ability of the DSI to model disease progression from elderly healthy controls to AD and its ability to predict conversion from mild cognitive impairment (MCI) to AD were assessed. It was found that the DSI provides well-behaving AD state estimates, corresponding well with the actual diagnoses. For predicting conversion from MCI to AD, the DSI attains performance similar to state-of-the-art reference classifiers. The results suggest that the DSF establishes an effective decision support and data visualization framework for improving AD diagnostics, allowing clinicians to rapidly analyze large quantities of diverse patient data.
Major discoveries have been made in the recent past in the genetics, biochemistry and neuropathology of frontotemporal lobar degeneration (FTLD). TAR DNA-binding protein 43 (TDP-43), encoded by the TARDBP gene, has been identified as the major pathological protein of FTLD with ubiquitin-immunoreactive (ub-ir) inclusions (FTLD-U) with or without amyotrophic lateral sclerosis (ALS) and sporadic ALS. Recently, mutations in the TARDBP gene in familial and sporadic ALS have been reported which demonstrate that abnormal TDP-43 alone is sufficient to cause neurodegeneration. Several familial cases of FTLD-U, however, are now known to have mutations in the progranulin (GRN) gene, but granulin is not a component of the TDP-43- and ub-ir inclusions. Further, TDP-43 is found to be a component of the inclusions of an increasing number of neurodegenerative diseases. Other FTLD-U entities with TDP-43 proteinopathy include: FTLD-U with valosin-containing protein (VCP) gene mutation and FTLD with ALS linked to chromosome 9p. In contrast, chromosome 3-linked dementia, FTLD-U with chromatin modifying protein 2B (CHMP2B) mutation, has ub-ir, TDP-43-negative inclusions. In summary, recent discoveries have generated new insights into the pathogenesis of a spectrum of disorders called TDP-43 proteinopathies including: FTLD-U, FTLD-U with ALS, ALS, and a broadening spectrum of other disorders. It is anticipated that these discoveries and a revised nosology of FTLD will contribute toward an accurate diagnosis, and facilitate the development of new diagnostic tests and therapeutics.
Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient-based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O(1). Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.
Consensus opinion characterizes dementia with Lewy bodies (DLB) as a progressive dementing illness, with significant fluctuations in cognition, visual hallucinations and/or parkinsonism. When parkinsonism is an early dominant feature, consensus opinion recommends that dementia within the first year is necessary for a diagnosis of DLB. If dementia occurs later, a diagnosis of Parkinson's disease with dementia (PDD) is recommended. While many previous studies have correlated the neuropathology in DLB with dementia and parkinsonism, few have analysed the relationship between fluctuating cognition and/or well-formed visual hallucinations and the underlying neuropathology in DLB and PDD. The aim of the present study was to determine any relationship between these less-studied core clinical features of DLB, and the distribution and density of cortical Lewy bodies (LB). The brains of 63 cases with LB were obtained over 6 years following population-based studies of dementia and parkinsonian syndromes. Annual, internationally standardized, clinical assessment batteries were reviewed to determine the presence and onset of the core clinical features of DLB. The maximal density of LB, plaques and tangles in the amygdala, parahippocampal, anterior cingulate, superior frontal, inferior temporal, inferior parietal and visual cortices were determined. Current clinicopathological diagnostic criteria were used to classify cases into DLB (n = 29), PDD (n = 18) or parkinsonism without dementia (n = 16) groups. Predictive statistics were used to ascertain whether fluctuating cognition or visual hallucinations predicted the clinicopathological group. Analysis of variance and regressions were used to identify any significant relationship(s) between the presence and severity of neuropathological and clinical features. Cognitive fluctuations and/or visual hallucinations were not good predictors of DLB in pathologically proven patients, although the absence of these features early in the disease course was highly predictive of PDD. Cases with DLB had higher LB densities in the inferior temporal cortex than cases with PDD. There was no association across groups between any neuropathological variable and the presence or absence of fluctuating cognition. However, there was a striking association between the distribution of temporal lobe LB and well-formed visual hallucinations. Cases with well-formed visual hallucinations had high densities of LB in the amygdala and parahippocampus, with early hallucinations relating to higher densities in parahippocampal and inferior temporal cortices. These temporal regions have previously been associated with visual hallucinations in other disorders. Thus, our results suggest that the distribution of temporal lobe LB is more related to the presence and duration of visual hallucinations in cases with LB than to the presence, severity or duration of dementia.
Background and objectives:
Limbic predominant age related TDP-43 encephalopathy neuropathologic change (LATE-NC) is a prevalent degenerative pathology in the oldest old who are the fastest growing segment of our population with the highest rates of dementia. We aimed to determine the relationship between LATE-NC and cognitive impairment and to identify its potential risk factors by studying its relationship with common past medical histories in an oldest old cohort.
Methods:
Participants from The 90+ Study with longitudinal evaluations and autopsy data were included. Dementia status and impairment in 5 main cognitive domains were determined at postmortem conferences leveraging all clinical and neuropsychological data blind to neuropathological diagnosis. Medical history information was obtained from patients and their informants. LATE-NC and Alzheimer's disease neuropathologic change (ADNC) were considered present in those with TDP-43 pathology in hippocampus and/or neocortex and those with high likelihood of ADNC according to NIA-AA guidelines respectively. We examined the association of degenerative pathologies with cognitive outcomes and multiple comparisons adjusted relationship of medical history variables with LATE-NC and ADNC using logistic regressions adjusted for age at death, sex, and education.
Results:
328 participants were included in this study. LATE-NC was present in 32% of the participants. It had a significant association with the presence of dementia (OR: 2.8, 95% CI: 1.7-4.6) and impairment in memory (OR: 3.0, 95% CI: 1.8-5.1), language (OR: 2.6, 95% CI: 1.6-4.3), and orientation (OR: 3.5, 95% CI: 2.1-5.9). The association with impaired orientation was unique to LATE-NC and the strength and significance of the other associations were comparable to ADNC. Furthermore, we found history of osteoarthritis (OR: 0.37, adjusted 95% CI: 0.21-0.66) and hypertension (OR: 0.52, adjusted 95% CI: 0.28-0.98) were associated with a reduced likelihood of LATE-NC, but not ADNC.
Discussion:
Our results suggest that LATE-NC is a prevalent degenerative pathology in the oldest old and has significant associations with dementia and impairment in cognitive domains with magnitudes that are comparable to ADNC. We also found that past medical histories of hypertension and osteoarthritis were associated with a lower likelihood of LATE-NC. This might help identify upstream mechanisms leading to this important pathology.
Background and objectives:
Evaluating and understanding the heterogeneity in dementia course has important implications for clinical practice, healthcare decision-making, and research. However, inconsistent findings have been reported with regard to the disease courses of the two most common dementias, Alzheimer's disease (AD) and Dementia with Lewy bodies (DLB). Using autopsy-confirmed diagnoses, we aimed to examine the disease trajectories in the years before death among dementia patients with pure AD, pure DLB, or mixed (AD and DLB) pathologies.
Methods:
The current retrospective longitudinal study included 62 participants with autopsy-confirmed diagnoses of pure AD (n=34), mixed AD and DLB (AD+DLB, n = 17), or pure DLB (n=11) from the Predictors 2 Cohort Study, a prospective, clinic-based, cohort of dementia patients. Generalized estimating equation models, with time zero at death, were used to examine the trajectory of cognition (Folstein Mini-Mental State Examination, MMSE), function (Activities of Daily Living, ADL), and dependence scale among patients with different autopsy-confirmed diagnosis (pure AD, AD+DLB, and pure DLB). The models were adjusted for age, sex, education, and baseline features including extrapyramidal signs, MMSE, ADL, and dependence scale.
Results:
The participants on average received 9.4±4.6 assessments at 6-month intervals during a mean 5.4±2.9 years of follow-up time. The three groups were similar in both cognition and function status at baseline. Cognition and function were highly correlated among AD+DLB patients but not in pure AD or pure DLB patients at baseline. Patients of the three groups all declined in both cognition and function but had different trajectories of decline. More specifically, the pure DLB patients experienced approximately double the rate of both cognitive decline and functional decline than the pure AD patients, and the mixed pathology group showed double the rate of functional decline as compared to pure AD patients.
Discussion:
In this longitudinal study, we found that among patients with dementia, those with Lewy body pathology experienced faster cognitive and functional decline than those with pure AD pathology.
Objective
MRI fails to reveal hippocampal pathology in 30-50% of temporal lobe epilepsy (TLE) surgical candidates. To address this clinical challenge, we developed an automated MRI-based classifier that lateralizes the side of covert hippocampal pathology in TLE.
Methods
We trained a surface-based linear discriminant classifier that uses T1-weighted (morphology) and T2-weighted as well as FLAIR/T1 (intensity) features. The classifier was trained on 60 TLE patients (mean age: 35.6; 58% female) with histologically-verified hippocampal sclerosis (HS). Images were deemed as MRI-negative in 42% of cases based on neuroradiological reading (40% based on hippocampal volumetry). The predictive model automatically labelled patients as left or right TLE. Lateralization accuracy was compared to electro-clinical data, including side of surgery. Accuracy of the classifier was further assessed in two independent TLE cohorts with similar demographics and electro-clinical characteristics (n=57; 58% MRI-negative).
Results
The overall lateralization accuracy was 93% (95%; CI 92% - 94%), regardless of HS visibility. In MRI-negative TLE, the combination of T2 and FLAIR/T1 intensities provided the highest accuracy both in the training (84%, area-under-the-curve (AUC): 0.95±0.02) and the validation cohorts (Cohort 1: 90%, AUC: 0.99; Cohort 2: 76%, AUC: 0.94).
Conclusion
This prediction model for TLE lateralization operates on readily available conventional MRI contrasts and offers gain in accuracy over visual radiological assessment. The combined contribution of decreased T1- and increased T2-weighted intensities makes the synthetic FLAIR/T1 contrast particularly effective in MRI-negative HS, setting the basis for broad clinical translation.
The aging brain is vulnerable to a wide array of neuropathologies. Prior work estimated that the three most studied of these, Alzheimer’s disease (AD), infarcts, and Lewy bodies, account for about 40% of the variation in late life cognitive decline. However, that estimate did not incorporate many other diseases that are now recognized as potent drivers of cognitive decline (e.g. limbic predominant age-related TDP-43 encephalopathy [LATE-NC], hippocampal sclerosis, other cerebrovascular conditions). We examined the degree to which person-specific cognitive decline in old age is driven by a wide array of neuropathologies.
1,164 deceased participants from two longitudinal clinical-pathologic studies, the Rush Memory and Aging Project and Religious Orders Study, completed up to 24 annual evaluations including 17 cognitive performance tests and underwent brain autopsy. Neuropathologic examinations provided 11 pathologic indices, including markers of AD, non-AD neurodegenerative diseases (i.e. LATE-NC, hippocampal sclerosis, Lewy bodies), and cerebrovascular conditions (i.e. macroscopic infarcts, microinfarcts, cerebral amyloid angiopathy, atherosclerosis, and arteriolosclerosis). Mixed effects models examined the linear relation of pathologic indices with global cognitive decline, and random change point models examined the relation of the pathologic indices with the onset of terminal decline and rates of preterminal and terminal decline.
Cognition declined an average of about 0.10 unit per year (estimate = -0.101, SE = 0.003, p < 0.001) with considerable heterogeneity in rates of decline (variance estimate for the person-specific slope of decline was 0.0094, p < 0.001). When considered separately, 10 of the 11 pathologic indices were associated with faster decline and accounted for between 2 and 34% of the variation in decline, respectively. When considered simultaneously, the 11 pathologic indices together accounted for a total of 43% of the variation in decline; AD-related indices accounted for 30–36% of the variation, non-AD neurodegenerative indices 4–10%, and cerebrovascular indices 3–8%. Finally, the 11 pathologic indices combined accounted for less than a third of the variation in the onset of terminal decline (28%) and rates of preterminal (32%) and terminal decline (19%).
Although age-related neuropathologies account for a large proportion of the variation in late life cognitive decline, considerable variation remains unexplained even after considering a wide array of neuropathologies. These findings highlight the complexity of cognitive aging and have important implications for the ongoing effort to develop effective therapeutics and identify novel treatment targets.
Cognitive impairment and its severe form dementia are increasingly prevalent in older adults and loom as a public health disaster unless effective interventions are developed. Cognitive impairment is a convergent trait caused by damage from an idiosyncratic mix of four prevalent diseases (Alzheimer disease; vascular brain injury; Lewy body diseases, such as Parkinson disease and dementia with Lewy bodies; and limbic-predominant age-related transactive response DNA-binding protein 43 encephalopathy) that is counterbalanced by individually varying resilience, which is comprised of reserve and compensation. Brain regional damage from each of these four prevalent diseases is generated by the net effect of injury and (mal)adaptive response and is accompanied by characteristic lesions. Existing therapeutics enhance resilience, whereas most agents under development target mechanisms of damage with only suppression of vascular brain injury yet to show therapeutic promise. We hope to anticipate future tailored interventions that target mechanisms of damage and thereby avert the oncoming surge of cognitive impairment and dementia in older adults. SIGNIFICANCE STATEMENT: Brain regional damage is generated by the net effect of injury and (mal)adaptive response. The extent to which signs and symptoms of such damage occur is influenced by an underlying resilience comprising reserve and compensation. Finding tailored interventions that target specific mechanisms of damage likely yields the most effective therapies.
Recently, a consensus working group provided new terminology for a common disease entity, limbic predominant age-related TDP-43 encephalopathy (LATE), and its neuropathological substrate (LATE-NC). LATE-NC not only often co-occurs with Alzheimer disease neuropathological change (ADNC), but also may present in isolation. The present study aimed to investigate potential risk factors and neuropathological characteristics associated with LATE-NC. A sample of 616 autopsied participants (>75 years at death), with TDP-43 immunohistochemical studies performed, was obtained from the National Alzheimer’s Coordinating Center. Logistic regression analyses examined associations between demographic, clinical and neuropathological characteristics and LATE-NC (TDP-43 in amygdala, hippocampus, or entorhinal/inferior temporal cortex) (alpha = 0.05). Adjusted models indicated that ADNC, hippocampal sclerosis (HS), arteriolosclerosis, and limbic or amygdala-predominant Lewy body disease (LBD), but not other LBD subtypes, were associated with higher odds of LATE-NC, whereas congestive heart failure (CHF) and motor problems as first predominant symptom were associated with lower odds of LATE-NC. Our findings corroborate previous studies indicating associations between LATE-NC and ADNC, HS, and arteriolosclerosis. Novel findings suggest the association with LATE-NC is restricted to amygdala/limbic-predominant subtype of LBD, and a possible protective (or competing risk) association with CHF. This study may inform future hypothesis-driven research on LATE-NC, a common brain disease of aging.
The shared role of amyloid-β (Aβ) deposition in cerebral amyloid angiopathy (CAA) and Alzheimer disease (AD) is arguably the clearest instance of crosstalk between neurodegenerative and cerebrovascular processes. The pathogenic pathways of CAA and AD intersect at the levels of Aβ generation, its circulation within the interstitial fluid and perivascular drainage pathways and its brain clearance, but diverge in their mechanisms of brain injury and disease presentation. Here, we review the evidence for and the pathogenic implications of interactions between CAA and AD. Both pathologies seem to be driven by impaired Aβ clearance, creating conditions for a self-reinforcing cycle of increased vascular Aβ, reduced perivascular clearance and further CAA and AD progression. Despite the close relationship between vascular and plaque Aβ deposition, several factors favour one or the other, such as the carboxy-terminal site of the peptide and specific co-deposited proteins. Amyloid-related imaging abnormalities that have been seen in trials of anti-Aβ immunotherapy are another probable intersection between CAA and AD, representing overload of perivascular clearance pathways and the effects of removing Aβ from CAA-positive vessels. The intersections between CAA and AD point to a crucial role for improving vascular function in the treatment of both diseases and indicate the next steps necessary for identifying therapies.
Recurrent neural networks (RNNs) have been widely adopted in research areas concerned with sequential data, such as text, audio, and video. However, RNNs consisting of sigma cells or tanh cells are unable to learn the relevant information of input data when the input gap is large. By introducing gate functions into the cell structure, the long short-term memory (LSTM) could handle the problem of long-term dependencies well. Since its introduction, almost all the exciting results based on RNNs have been achieved by the LSTM. The LSTM has become the focus of deep learning. We review the LSTM cell and its variants to explore the learning capacity of the LSTM cell. Furthermore, the LSTM networks are divided into two broad categories: LSTM-dominated networks and integrated LSTM networks. In addition, their various applications are discussed. Finally, future research directions are presented for LSTM networks.
Accurately predicting the underlying neuropathological diagnosis in patients with behavioural variant frontotemporal dementia (bvFTD) poses a daunting challenge for clinicians but will be critical for the success of disease-modifying therapies. We sought to improve pathological prediction by exploring clinicopathological correlations in a large bvFTD cohort. Among 438 patients in whom bvFTD was either the top or an alternative possible clinical diagnosis, 117 had available autopsy data, including 98 with a primary pathological diagnosis of frontotemporal lobar degeneration (FTLD), 15 with Alzheimer's disease, and four with amyotrophic lateral sclerosis who lacked neurodegenerative disease-related pathology outside of the motor system. Patients with FTLD were distributed between FTLD-tau (34 patients: 10 corticobasal degeneration, nine progressive supranuclear palsy, eight Pick's disease, three frontotemporal dementia with parkinsonism associated with chromosome 17, three unclassifiable tauopathy, and one argyrophilic grain disease); FTLD-TDP (55 patients: nine type A including one with motor neuron disease, 27 type B including 21 with motor neuron disease, eight type C with right temporal lobe presentations, and 11 unclassifiable including eight with motor neuron disease), FTLD-FUS (eight patients), and one patient with FTLD-ubiquitin proteasome system positive inclusions (FTLD-UPS) that stained negatively for tau, TDP-43, and FUS. Alzheimer's disease was uncommon (6%) among patients whose only top diagnosis during follow-up was bvFTD. Seventy-nine per cent of FTLD-tau, 86% of FTLD-TDP, and 88% of FTLD-FUS met at least 'possible' bvFTD diagnostic criteria at first presentation. The frequency of the six core bvFTD diagnostic features was similar in FTLD-tau and FTLD-TDP, suggesting that these features alone cannot be used to separate patients by major molecular class. Voxel-based morphometry revealed that nearly all pathological subgroups and even individual patients share atrophy in anterior cingulate, frontoinsula, striatum, and amygdala, indicating that degeneration of these regions is intimately linked to the behavioural syndrome produced by these diverse aetiologies. In addition to these unifying features, symptom profiles also differed among pathological subtypes, suggesting distinct anatomical vulnerabilities and informing a clinician's prediction of pathological diagnosis. Data-driven classification into one of the 10 most common pathological diagnoses was most accurate (up to 60.2%) when using a combination of known predictive factors (genetic mutations, motor features, or striking atrophy patterns) and the results of a discriminant function analysis that incorporated clinical, neuroimaging, and neuropsychological data.
Introduction:
Temporal lobe epilepsy (TLE) is the most frequent form of focal epilepsy in adults. Because approximately half of these patients develop drug resistance, epilepsy surgery designed to remove the epileptogenic zone is an excellent option in selected patients. Histopathological analyses of hippocampal specimens in TLE patients revealed 4 types of Ammon's horn sclerosis, which are correlated with long-term epileptological outcome. The aim of this study was the correlation of noninvasive, high-resolution, morphological magnetic resonance imaging (MRI) at an ultra-high-field (7 T) of the hippocampus in TLE patients with histopathological findings.
Methods:
High-resolution, T2-weighted FSE MRI in 14 patients with drug-resistant temporal lobe epilepsy was performed on a 7-T Magnetom using a 32-channel coil. Four independent investigators assessed the delineation and semiquantitative evaluation of volume, signal intensity, internal architecture, and overall grading of the hippocampal subfields CA1-4, as well as the presence of the dentate granule cell layer (DGCL), on MRI scans. Results were compared with semiquantitative evaluation of neuronal loss and astrogliosis in the histological sections of the surgical specimens.
Results:
Seven-tesla MR examinations were evaluable in 13 cases. Volume loss and signal intensity, as well as overall grading, showed a strong correlation between MRI and histology in individual CA regions. Furthermore, sensitivity and specificity values up to 100% were found for the detection of pathology in the CA subfields. The prediction of Ammon's horn sclerosis type was correct in up to 12 of 13 cases, whereas the dentate gyrus could not be delineated on MRI.
Discussion:
High-resolution, ultra-high-field MRI is a promising tool for the detection of subtle changes in the hippocampus in patients with temporal lobe epilepsy. Large cohorts will be necessary to confirm the predictive value of 7-T MRI in the preoperative evaluation of TLE patients.
We study the problem of attributing the prediction of a deep network to its input features, a problem previously studied by several other works. We identify two fundamental axioms---Sensitivity and Implementation Invariance that attribution methods ought to satisfy. We show that they are not satisfied by most known attribution methods, which we consider to be a fundamental weakness of those methods. We use the axioms to guide the design of a new attribution method called Integrated Gradients. Our method requires no modification to the original network and is extremely simple to implement; it just needs a few calls to the standard gradient operator. We apply this method to a couple of image models, a couple of text models and a chemistry model, demonstrating its ability to debug networks, to extract rules from a deep network, and to enable users to engage with models better.
The pathology-based classification of Alzheimer's disease (AD) and other neurodegenerative diseases is a work in progress that is important for both clinicians and basic scientists. Analyses of large autopsy series, biomarker studies, and genomics analyses have provided important insights about AD and shed light on previously unrecognized conditions, enabling a deeper understanding of neurodegenerative diseases in general. After demonstrating the importance of correct disease classification for AD and primary age-related tauopathy, we emphasize the public health impact of an underappreciated AD "mimic," which has been termed "hippocampal sclerosis of aging" or "hippocampal sclerosis dementia." This pathology affects >20% of individuals older than 85 years and is strongly associated with cognitive impairment. In this review, we provide an overview of current hypotheses about how genetic risk factors (GRN, TMEM106B, ABCC9, and KCNMB2), and other pathogenetic influences contribute to TDP-43 pathology and hippocampal sclerosis. Because hippocampal sclerosis of aging affects the "oldest-old" with arteriolosclerosis and TDP-43 pathologies that extend well beyond the hippocampus, more appropriate terminology for this disease is required. We recommend "cerebral age-related TDP-43 and sclerosis" (CARTS). A detailed case report is presented, which includes neuroimaging and longitudinal neurocognitive data. Finally, we suggest a neuropathology-based diagnostic rubric for CARTS.
Introduction:
Diabetes' relationship to specific neuropathologic causes of dementia is incompletely understood.
Methods:
We used logistic regression to evaluate the association between diabetes and infarcts, Braak stage, neuritic plaque score, and level of Alzheimer's neuropathologic changes in 2365 autopsied persons. In a subset of >1300 persons with available cognitive data, we examined the association between diabetes and cognition using Poisson regression.
Results:
Diabetes increased odds of brain infarcts (odds ratio [OR] = 1.57, P < .0001), specifically lacunes (OR = 1.71, P < .0001), but not Alzheimer neuropathology. Diabetes plus infarcts was associated with lower cognitive scores at end of life than infarcts or diabetes alone, and diabetes plus high level of Alzheimer's neuropathologic changes was associated with lower mini-mental state examination scores than the pathology alone.
Discussion:
This study supports the conclusions that diabetes increases the risk of cerebrovascular but not Alzheimer's pathology, and at least some of diabetes' relationship to cognitive impairment may be modified by neuropathology.
Objective: To investigate the association of hippocampal sclerosis (HS) with TAR-DNA binding protein of 43 kDa (TDP-43) and other common age-related pathologies, dementia, probable Alzheimer's disease (AD), mild cognitive impairment (MCI) and cognitive domains in community-dwelling older subjects. Methods: Diagnoses of dementia, probable AD and MCI in 636 autopsied subjects from the Religious Order Study and the Rush Memory and Aging Project were based on clinical evaluation and cognitive performance tests. HS was defined as severe neuronal loss and gliosis in the hippocampal CA1and/or subiculum. The severity and distribution of TDP-43 was assessed and other age-related pathologies were also documented. Results: HS was more common in those aged > 90 years (18.0%) compared to younger subjects (9.2%). HS cases commonly coexisted with TDP-43 pathology (86%), which was more severe (p < 0.001) in HS cases. Although, HS also commonly coexisted with AD and Lewy body (LB) pathology; only TDP-43 pathology increased the odds of HS (OR=2.63; 95% CI 2.07-3.34). In logistic regression models accounting for age, TDP-43 and other common age-related pathologies; HS cases had higher odds of dementia (OR=3.71; 95% CI=1.93-7.16), MCI and probable AD (OR=3.75; 95% CI=2.01-7.02). In linear regression models, including an interaction term for HS and TDP-43 pathology; HS with coexisting TDP-43 was associated with lower function in multiple cognitive domains while HS without TDP-43 did not have statistically significant associations. TDP-43 without HS was separately related to lower episodic memory. Interpretation: The combined role of hippocampal sclerosis and TDP-43 pathology are significant factors underlying global cognitive impairment and probable AD in older subjects. This article is protected by copyright. All rights reserved.
Copyright © 2015 American Neurological Association.
Objective:
The pathologic indices of Alzheimer disease, cerebrovascular disease, and Lewy body disease accumulate in the brains of older persons with and without dementia, but the extent to which they account for late life cognitive decline remains unknown. We tested the hypothesis that these pathologic indices account for the majority of late life cognitive decline.
Methods:
A total of 856 deceased participants from 2 longitudinal clinical-pathologic studies, Rush Memory and Aging Project and Religious Orders Study, completed a mean of 7.5 annual evaluations, including 17 cognitive tests. Neuropathologic examinations provided quantitative measures of global Alzheimer pathology, amyloid load, tangle density, macroscopic infarcts, microinfarcts, and neocortical Lewy bodies. Random coefficient models were used to examine the linear relation of pathologic indices with global cognitive decline. In subsequent analyses, random change point models were used to examine the relation of the pathologic indices with the onset of terminal decline and rates of preterminal and terminal decline (ie, nonlinear decline).
Results:
Cognition declined a mean of about 0.11 U per year (estimate = -0.109, standard error [SE] = 0.004, p < 0.001), with significant individual differences in rates of decline; the variance estimate for the individual slopes was 0.013 (SE = 0.112, p < 0.001). In separate analyses, global Alzheimer pathology, amyloid, tangles, macroscopic infarcts, and neocortical Lewy bodies were associated with faster rates of decline and explained 22%, 6%, 34%, 2%, and 8% of the variation in decline, respectively. When analyzed simultaneously, the pathologic indices accounted for a total of 41% of the variation in decline, and the majority remained unexplained. Furthermore, in random change point models examining the influence of the pathologic indices on the onset of terminal decline and the preterminal and terminal components of the cognitive trajectory, the common pathologic indices accounted for less than a third of the variation in the onset of terminal decline and rates of preterminal and terminal decline.
Interpretation:
The pathologic indices of the common causes of dementia are important determinants of cognitive decline in old age and account for a large proportion of the variation in late life cognitive decline. Surprisingly, however, much of the variation in cognitive decline remains unexplained, suggesting that other important determinants of cognitive decline remain to be identified. Identification of the mechanisms that contribute to the large unexplained proportion of cognitive decline is urgently needed to prevent late life cognitive decline.
The cerebral neuropathology of Type 2 diabetes (CNDM2) has not been positively defined. This review includes a description of CNDM2 research from before the 'Pubmed Era'. Recent neuroimaging studies have focused on cerebrovascular and white matter pathology. These and prior studies about cerebrovascular histopathology in diabetes are reviewed. Evidence is also described for and against the link between CNDM2 and Alzheimer's disease pathogenesis. To study this matter directly, we evaluated data from University of Kentucky Alzheimer's Disease Center (UK ADC) patients recruited while non-demented and followed longitudinally. Of patients who had come to autopsy (N = 234), 139 met inclusion criteria. These patients provided the basis for comparing the prevalence of pathological and clinical indices between well-characterized cases with (N = 50) or without (N = 89) the premortem diagnosis of diabetes. In diabetics, cerebrovascular pathology was more frequent and Alzheimer-type pathology was less frequent than in non-diabetics. Finally, a series of photomicrographs demonstrates histopathological features (including clinical-radiographical correlation) observed in brains of persons that died after a history of diabetes. These preliminary, correlative, and descriptive studies may help develop new hypotheses about CNDM2. We conclude that more work should be performed on human material in the context of CNDM2.