Apraxia screening predicts Alzheimer pathology in frontotemporal dementia

Clinicopathologic and Neuroimaging Correlations of Nonverbal Oral Apraxia in Patients With Neurodegenerative Disease

BACKGROUND AND OBJECTIVES

Nonverbal oral apraxia (NVOA) is the inability to plan, sequence, and execute voluntary oromotor movements in the absence of weakness. In the context of neurodegenerative disease, it remains unclear whether it is linked to a specific underlying pathologic, clinical, or neuroimaging finding. Thus, we aimed to assess the clinicopathologic and neuroimaging associations of NVOA.

METHODS

We conducted a retrospective study of autopsy-confirmed patients previously assessed through an NVOA evaluation tool with a previously published cutpoint to screen for NVOA. We compared demographic and clinical characteristics and postmortem pathology between those who developed NVOA and those who did not. We also compared clinicopathologic characteristics in mild vs greater than mild NVOA and early vs late-emerging NVOA. SPM12 was used to assess patterns of gray matter loss in NVOA vs non-NVOA with age and sex included as covariates.

RESULTS

A total of 104 patients (median age at symptom onset 63 years, 43% female) were included in the study. 63 (60.6%) developed NVOA. NVOA appeared at a median of 4.3 years from symptom onset. 29% developed NVOA within the first 3 years. Primary progressive apraxia of speech and the nonfluent variant of primary progressive aphasia were the most common baseline diagnoses in the NVOA group while progressive supranuclear palsy (PSP) syndrome and logopenic progressive aphasia (LPA) were the most common in patients without NVOA. Atrophy of the left lateral and medial posterior frontal cortex was related to NVOA. The most common pathologies associated with NVOA were PSP (36.5%) and corticobasal degeneration (CBD) (33.3%). In patients without NVOA, PSP (26.8%) and other pathologies (26.8%) were the most frequent. 11% of patients with NVOA had persistently mild NVOA and were more likely to have baseline diagnoses of LPA, PSP syndrome, or semantic dementia. The most frequent pathologies in this group were Alzheimer disease and PSP. The pathologic associations of greater than mild NVOA were CBD and PSP.

DISCUSSION

NVOA is present in several clinical syndromes. It is most associated with PSP and CBD. NVOA is a manifestation of left lateral and medial posterior frontal cortex damage rather than a particular pathology.

Research Criteria for the Behavioral Variant of Alzheimer Disease: A Systematic Review and Meta-analysis

Importance

The behavioral variant of Alzheimer disease (bvAD) is characterized by early and predominant behavioral deficits caused by AD pathology. This AD phenotype is insufficiently understood and lacks standardized clinical criteria, limiting reliability and reproducibility of diagnosis and scientific reporting.

Objective

To perform a systematic review and meta-analysis of the bvAD literature and use the outcomes to propose research criteria for this syndrome.

Data Sources

A systematic literature search in PubMed/MEDLINE and Web of Science databases (from inception through April 7, 2021) was performed in duplicate.

Study Selection

Studies reporting on behavioral, neuropsychological, or neuroimaging features in bvAD and, when available, providing comparisons with typical amnestic-predominant AD (tAD) or behavioral variant frontotemporal dementia (bvFTD).

Data Extraction and Synthesis

This analysis involved random-effects meta-analyses on group-level study results of clinical data and systematic review of the neuroimaging literature. The study was performed following Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.

Main Outcomes and Measures

Behavioral symptoms (neuropsychiatric symptoms and bvFTD core clinical criteria), cognitive function (global cognition, episodic memory, and executive functioning), and neuroimaging features (structural magnetic resonance imaging, [18F]fluorodeoxyglucose-positron emission tomography, perfusion single-photon emission computed tomography, amyloid positron emission tomography, and tau positron emission tomography).

Results

The search led to the assessment of 83 studies, including 13 suitable for meta-analysis. Data were collected for 591 patients with bvAD. There was moderate to substantial heterogeneity and moderate risk of bias across studies. Cases with bvAD showed more severe behavioral symptoms than tAD (standardized mean difference [SMD], 1.16 [95% CI, 0.74-1.59]; P < .001) and a trend toward less severe behavioral symptoms compared with bvFTD (SMD, -0.22 [95% CI, -0.47 to 0.04]; P = .10). Meta-analyses of cognitive data indicated worse executive performance in bvAD vs tAD (SMD, -1.03 [95% CI, -1.74 to -0.32]; P = .008) but not compared with bvFTD (SMD, -0.61 [95% CI, -1.75 to 0.53]; P = .29). Cases with bvAD showed a nonsignificant difference of worse memory performance compared with bvFTD (SMD, -1.31 [95% CI, -2.75 to 0.14]; P = .08) but did not differ from tAD (SMD, 0.43 [95% CI, -0.46 to 1.33]; P = .34). The neuroimaging literature revealed 2 distinct bvAD neuroimaging phenotypes: an AD-like pattern with relative frontal sparing and a relatively more bvFTD-like pattern characterized by additional anterior involvement, with the AD-like pattern being more prevalent.

Conclusions and Relevance

These data indicate that bvAD is clinically most similar to bvFTD, while it shares most pathophysiological features with tAD. Based on these insights, we propose research criteria for bvAD aimed at improving the consistency and reliability of future research and aiding the clinical assessment of this AD phenotype.

Longitudinal Speech Biomarkers for Automated Alzheimer's Detection

We introduce a novel audio processing architecture, the Open Voice Brain Model (OVBM), improving detection accuracy for Alzheimer's (AD) longitudinal discrimination from spontaneous speech. We also outline the OVBM design methodology leading us to such architecture, which in general can incorporate multimodal biomarkers and target simultaneously several diseases and other AI tasks. Key in our methodology is the use of multiple biomarkers complementing each other, and when two of them uniquely identify different subjects in a target disease we say they are orthogonal. We illustrate the OBVM design methodology by introducing sixteen biomarkers, three of which are orthogonal, demonstrating simultaneous above state-of-the-art discrimination for two apparently unrelated diseases such as AD and COVID-19. Depending on the context, throughout the paper we use OVBM indistinctly to refer to the specific architecture or to the broader design methodology. Inspired by research conducted at the MIT Center for Brain Minds and Machines (CBMM), OVBM combines biomarker implementations of the four modules of intelligence: The brain OS chunks and overlaps audio samples and aggregates biomarker features from the sensory stream and cognitive core creating a multi-modal graph neural network of symbolic compositional models for the target task. In this paper we apply the OVBM design methodology to the automated diagnostic of Alzheimer's Dementia (AD) patients, achieving above state-of-the-art accuracy of 93.8% using only raw audio, while extracting a personalized subject saliency map designed to longitudinally track relative disease progression using multiple biomarkers, 16 in the reported AD task. The ultimate aim is to help medical practice by detecting onset and treatment impact so that intervention options can be longitudinally tested. Using the OBVM design methodology, we introduce a novel lung and respiratory tract biomarker created using 200,000+ cough samples to pre-train a model discriminating cough cultural origin. Transfer Learning is subsequently used to incorporate features from this model into various other biomarker-based OVBM architectures. This biomarker yields consistent improvements in AD detection in all the starting OBVM biomarker architecture combinations we tried. This cough dataset sets a new benchmark as the largest audio health dataset with 30,000+ subjects participating in April 2020, demonstrating for the first time cough cultural bias.

Deep Learning-Based Classification and Voxel-Based Visualization of Frontotemporal Dementia and Alzheimer's Disease

Frontotemporal dementia (FTD) and Alzheimer's disease (AD) have overlapping symptoms, and accurate differential diagnosis is important for targeted intervention and treatment. Previous studies suggest that the deep learning (DL) techniques have the potential to solve the differential diagnosis problem of FTD, AD and normal controls (NCs), but its performance is still unclear. In addition, existing DL-assisted diagnostic studies still rely on hypothesis-based expert-level preprocessing. On the one hand, it imposes high requirements on clinicians and data themselves; On the other hand, it hinders the backtracking of classification results to the original image data, resulting in the classification results cannot be interpreted intuitively. In the current study, a large cohort of 3D T1-weighted structural magnetic resonance imaging (MRI) volumes (n = 4,099) was collected from two publicly available databases, i.e., the ADNI and the NIFD. We trained a DL-based network directly based on raw T1 images to classify FTD, AD and corresponding NCs. And we evaluated the convergence speed, differential diagnosis ability, robustness and generalizability under nine scenarios. The proposed network yielded an accuracy of 91.83% based on the most common T1-weighted sequence [magnetization-prepared rapid acquisition with gradient echo (MPRAGE)]. The knowledge learned by the DL network through multiple classification tasks can also be used to solve subproblems, and the knowledge is generalizable and not limited to a specified dataset. Furthermore, we applied a gradient visualization algorithm based on guided backpropagation to calculate the contribution graph, which tells us intuitively why the DL-based networks make each decision. The regions making valuable contributions to FTD were more widespread in the right frontal white matter regions, while the left temporal, bilateral inferior frontal and parahippocampal regions were contributors to the classification of AD. Our results demonstrated that DL-based networks have the ability to solve the enigma of differential diagnosis of diseases without any hypothesis-based preprocessing. Moreover, they may mine the potential patterns that may be different from human clinicians, which may provide new insight into the understanding of FTD and AD.

Limb and face apraxias in frontotemporal dementia: A systematic scoping review

PURPOSE

To investigate the literature for frequencies, profiles and neural correlates of limb and face apraxias in frontotemporal dementia (FTD).

METHOD

The search conducted in Ovid Medline, PsycINFO and Scopus yielded 487 non-duplicate records, and 43 were included in the final analysis.

RESULTS

Apraxias are evident in diverse forms in all clinical variants of FTD within the first four years of the disease. Face apraxia and productive limb apraxia co-occur in the behavioural and nonfluent variants. The logopenic variant resembles Alzheimer's disease in terms of pronounced parietal limb apraxia and absence of face apraxia. The semantic variant exhibits conceptual praxis deficits together with relatively preserved imitation skills. Concerning the genetic variants of FTD, productive limb apraxia is common among carriers of the progranulin gene mutation, and subtle gestural alterations have been documented among carriers of the chromosome 9 open reading frame 72 gene mutation before the expected disease onset. The data on neural correlations suggest that the breakdown of praxis results from bilateral cortical and subcortical damage in FTD and that Alzheimer-type pathology of the cerebrospinal fluid increases the severity of limb apraxia in all of the variants. Face apraxia correlates with degeneration of the medial and superior frontal cortices.

CONCLUSIONS

Each of the clinical variants of FTD exhibits a characteristic profile of apraxias that may support early differentiation between the variants and from Alzheimer's disease. However, the screening procedures developed for stroke populations seem insufficient, and a multifaceted assessment tool is needed. Although valid and practical tests already exist for dementia populations, a concise selection of test items that covers all of the critical domains is called for.

Psychological and Cognitive Markers of Behavioral Variant Frontotemporal Dementia-A Clinical Neuropsychologist's View on Diagnostic Criteria and Beyond

Behavioral variant frontotemporal dementia (bvFTD) is the second leading cognitive disorder caused by neurodegeneration in patients under 65 years of age. Characterized by frontal, insular, and/or temporal brain atrophy, patients present with heterogeneous constellations of behavioral and psychological symptoms among which progressive changes in social conduct, lack of empathy, apathy, disinhibited behaviors, and cognitive impairments are frequently observed. Since the histopathology of the disease is heterogeneous and identified genetic mutations only account for ~30% of cases, there are no reliable biomarkers for the diagnosis of bvFTD available in clinical routine as yet. Early detection of bvFTD thus relies on correct application of clinical diagnostic criteria. Their evaluation however, requires expertise and in-depth assessments of cognitive functions, history taking, clinical observations as well as caregiver reports on behavioral and psychological symptoms and their respective changes. With this review, we aim for a critical appraisal of common methods to access the behavioral and psychological symptoms as well as the cognitive alterations presented in the diagnostic criteria for bvFTD. We highlight both, practical difficulties as well as current controversies regarding an overlap of symptoms and particularly cognitive impairments with other neurodegenerative and primary psychiatric diseases. We then review more recent developments and evidence on cognitive, behavioral and psychological symptoms of bvFTD beyond the diagnostic criteria which may prospectively enhance the early detection and differential diagnosis in clinical routine. In particular, evidence on specific impairments in social and emotional processing, praxis abilities as well as interoceptive processing in bvFTD is summarized and potential links with behavior and classic cognitive domains are discussed. We finally outline both, future opportunities and major challenges with regard to the role of clinical neuropsychology in detecting bvFTD and related neurocognitive disorders.