Cost-effectiveness of Alzheimer's disease CSF biomarkers and amyloid-PET in early-onset cognitive impairment diagnosis

Artificial intelligence detection of cognitive impairment in older adults during walking

INTRODUCTION

To detect early cognitive impairment in community-dwelling older adults, this study explored the viability of artificial intelligence (AI)-assisted linear acceleration and angular velocity analysis during walking.

METHODS

This cross-sectional study included 879 participants without dementia (female, 60.6%; mean age, 73.5 years) from the 2011 Comprehensive Gerontology Survey. Sensors attached to the pelvis and left ankle recorded the triaxial linear acceleration and angular velocity while the participants walked at a comfortable speed. Cognitive impairment was determined using Mini-Mental State Examination scores. Deep learning models were used to discern the linear acceleration and angular velocity data of 12,302 walking strides.

RESULTS

The models' average sensitivity, specificity, and area under the curve were 0.961, 0.643, and 0.833, respectively, across 30 testing datasets.

DISCUSSION

AI-enabled gait analysis can be used to detect signs of cognitive impairment. Integrating this AI model into smartphones may help detect dementia early, facilitating better prevention.

Highlights

Artificial intelligence (AI)-enabled gait analysis can be used to detect the early signs of cognitive decline.This AI model was constructed using data from a community-dwelling cohort.AI-assisted linear acceleration and angular velocity analysis during gait was used.The model may help in early detection of dementia.

Model-Based Economic Evaluations of Interventions for Dementia: An Updated Systematic Review and Quality Assessment

BACKGROUND

There has been an increase in model-based economic evaluations of interventions for dementia. The most recent systematic review of economic evaluations for dementia highlighted weaknesses in studies, including lack of justification for model assumptions and data inputs.

OBJECTIVE

This study aimed to update the last published systematic review of model-based economic evaluations of interventions for dementia, including Alzheimer's disease, with a focus on any methodological improvements and quality assessment of the studies.

METHODS

Systematic searches in eight databases, including PubMed, Cochrane, Embase, CINAHL, PsycINFO, EconLit, international HTA database, and the Tufts Cost-Effectiveness Analysis Registry were undertaken from February 2018 until August 2022. The quality of the included studies was assessed using the Philips checklist and the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) 2022 checklist. The findings were summarized through narrative analysis.

RESULTS

This review included 23 studies, comprising cost-utility analyses (87%), cost-benefit analyses (9%) and cost-effectiveness analyses (4%). The studies covered various interventions, including pharmacological (n = 10, 43%), non-pharmacological (n = 4, 17%), prevention (n = 4, 17%), diagnostic (n = 4, 17%) and integrated (n = 1, 4%) [diagnostics-pharmacologic] strategies. Markov transition models were commonly employed (65%), followed by decision trees (13%) and discrete-event simulation (9%). Several interventions from all categories were reported as being cost effective. The quality of reporting was suboptimal for the Methods and Results sections in almost all studies, although the majority of studies adequately addressed the decision problem, scope, and model-type selection in their economic evaluations. Regarding the quality of methodology, only a minority of studies addressed competing theories or clearly explained the rationale for model structure. Furthermore, few studies systematically identified key parameters or assessed data quality, and uncertainty was mostly addressed partially.

CONCLUSIONS

This review informs future research and resource allocation by providing insights into model-based economic evaluations for dementia interventions and highlighting areas for improvement.

Predicting Alzheimer's disease CSF core biomarkers: a multimodal Machine Learning approach

Introduction

Alzheimer's disease (AD) is a progressive neurodegenerative disorder. Current core cerebrospinal fluid (CSF) AD biomarkers, widely employed for diagnosis, require a lumbar puncture to be performed, making them impractical as screening tools. Considering the role of sleep disturbances in AD, recent research suggests quantitative sleep electroencephalography features as potential non-invasive biomarkers of AD pathology. However, quantitative analysis of comprehensive polysomnography (PSG) signals remains relatively understudied. PSG is a non-invasive test enabling qualitative and quantitative analysis of a wide range of parameters, offering additional insights alongside other biomarkers. Machine Learning (ML) gained interest for its ability to discern intricate patterns within complex datasets, offering promise in AD neuropathology detection. Therefore, this study aims to evaluate the effectiveness of a multimodal ML approach in predicting core AD CSF biomarkers.

Methods

Mild-moderate AD patients were prospectively recruited for PSG, followed by testing of CSF and blood samples for biomarkers. PSG signals underwent preprocessing to extract non-linear, time domain and frequency domain statistics quantitative features. Multiple ML algorithms were trained using four subsets of input features: clinical variables (CLINVAR), conventional PSG parameters (SLEEPVAR), quantitative PSG signal features (PSGVAR) and a combination of all subsets (ALL). Cross-validation techniques were employed to evaluate model performance and ensure generalizability. Regression models were developed to determine the most effective variable combinations for explaining variance in the biomarkers.

Results

On 49 subjects, Gradient Boosting Regressors achieved the best results in estimating biomarkers levels, using different loss functions for each biomarker: least absolute deviation (LAD) for the Aβ42, least squares (LS) for p-tau and Huber for t-tau. The ALL subset demonstrated the lowest training errors for all three biomarkers, albeit with varying test performance. Specifically, the SLEEPVAR subset yielded the best test performance in predicting Aβ42, while the ALL subset most accurately predicted p-tau and t-tau due to the lowest test errors.

Conclusions

Multimodal ML can help predict the outcome of CSF biomarkers in early AD by utilizing non-invasive and economically feasible variables. The integration of computational models into medical practice offers a promising tool for the screening of patients at risk of AD, potentially guiding clinical decisions.

Incremental value of amyloid PET in a tertiary memory clinic setting in China

INTRODUCTION

The objective of this study is to investigate the incremental value of amyloid positron emission tomography (Aβ-PET) in a tertiary memory clinic setting in China.

METHODS

A total of 1073 patients were offered Aβ-PET using 18F-florbetapir. The neurologists determined a suspected etiology (Alzheimer's disease [AD] or non-AD) with a percentage estimate of their confidence and medication prescription both before and after receiving the Aβ-PET results.

RESULTS

After disclosure of the Aβ-PET results, etiological diagnoses changed in 19.3% of patients, and diagnostic confidence increased from 69.3% to 85.6%. Amyloid PET results led to a change of treatment plan in 36.5% of patients. Compared to the late-onset group, the early-onset group had a more frequent change in diagnoses and a higher increase in diagnostic confidence.

DISCUSSION

Aβ-PET has significant impacts on the changes of diagnoses and management in Chinese population. Early-onset cases are more likely to benefit from Aβ-PET than late-onset cases.

HIGHLIGHTS

Amyloid PET contributes to diagnostic changes and its confidence in Chinese patients. Amyloid PET leads to a change of treatment plans in Chinese patients. Early-onset cases are more likely to benefit from amyloid PET than late-onset cases.

A 3D convolutional neural network to classify subjects as Alzheimer's disease, frontotemporal dementia or healthy controls using brain 18F-FDG PET

With the arrival of disease-modifying drugs, neurodegenerative diseases will require an accurate diagnosis for optimal treatment. Convolutional neural networks are powerful deep learning techniques that can provide great help to physicians in image analysis. The purpose of this study is to introduce and validate a 3D neural network for classification of Alzheimer's disease (AD), frontotemporal dementia (FTD) or cognitively normal (CN) subjects based on brain glucose metabolism. Retrospective [18F]-FDG-PET scans of 199 CE, 192 FTD and 200 CN subjects were collected from our local database, Alzheimer's disease and frontotemporal lobar degeneration neuroimaging initiatives. Training and test sets were created using randomization on a 90 %-10 % basis, and training of a 3D VGG16-like neural network was performed using data augmentation and cross-validation. Performance was compared to clinical interpretation by three specialists in the independent test set. Regions determining classification were identified in an occlusion experiment and Gradient-weighted Class Activation Mapping. Test set subjects were age- and sex-matched across categories. The model achieved an overall 89.8 % accuracy in predicting the class of test scans. Areas under the ROC curves were 93.3 % for AD, 95.3 % for FTD, and 99.9 % for CN. The physicians' consensus showed a 69.5 % accuracy, and there was substantial agreement between them (kappa = 0.61, 95 % CI: 0.49-0.73). To our knowledge, this is the first study to introduce a deep learning model able to discriminate AD and FTD based on [18F]-FDG PET scans, and to isolate CN subjects with excellent accuracy. These initial results are promising and hint at the potential for generalization to data from other centers.

Cost-effectiveness comparison between blood biomarkers and conventional tests in Alzheimer's disease diagnosis

Dementia management has evolved with drugs such as lecanemab, shifting management from palliative care to early diagnosis and intervention. However, the administration of these drugs presents challenges owing to the invasiveness, high cost and limited availability of amyloid-PET and cerebrospinal fluid tests for guiding drug administration. Our manuscript explores the potential of less invasive blood biomarkers as a diagnostic method, with a cost-effectiveness analysis and a comparison with traditional tests. Our findings suggest that blood biomarkers are a cost-effective alternative, but with lower accuracy, indicating the need for multiple specific biomarkers for precision. This underscores the importance of future research on new blood biomarkers and their clinical efficacy.

Machine learning prediction of future amyloid beta positivity in amyloid-negative individuals

BACKGROUND

The pathophysiology of Alzheimer's disease (AD) involves β -amyloid (A β ) accumulation. Early identification of individuals with abnormal β -amyloid levels is crucial, but A β quantification with positron emission tomography (PET) and cerebrospinal fluid (CSF) is invasive and expensive.

METHODS

We propose a machine learning framework using standard non-invasive (MRI, demographics, APOE, neuropsychology) measures to predict future A β -positivity in A β -negative individuals. We separately study A β -positivity defined by PET and CSF.

RESULTS

Cross-validated AUC for 4-year A β conversion prediction was 0.78 for the CSF-based and 0.68 for the PET-based A β definitions. Although not trained for the clinical status-change prediction, the CSF-based model excelled in predicting future mild cognitive impairment (MCI)/dementia conversion in cognitively normal/MCI individuals (AUCs, respectively, 0.76 and 0.89 with a separate dataset).

CONCLUSION

Standard measures have potential in detecting future A β -positivity and assessing conversion risk, even in cognitively normal individuals. The CSF-based definition led to better predictions than the PET-based definition.

Agreement of cerebrospinal fluid biomarkers and amyloid-PET in a multicenter study

Core Alzheimer's disease (AD) cerebrospinal fluid (CSF) biomarkers have shown incomplete agreement with amyloid-positron emission tomography (PET). Our goal was to analyze the agreement between AD CSF biomarkers and amyloid-PET in a multicenter study. Retrospective multicenter study (5 centers). Participants who underwent both CSF biomarkers and amyloid-PET scan within 18 months were included. Clinical diagnoses were made according to latest diagnostic criteria by the attending clinicians. CSF Amyloid Beta1-42 (Aβ1-42, A), phosphorliated tau 181 (pTau181, T) and total tau (tTau, N) biomarkers were considered normal (-) or abnormal ( +) according to cutoffs of each center. Amyloid-PET was visually classified as positive/negative. Agreement between CSF biomarkers and amyloid-PET was analyzed by overall percent agreement (OPA). 236 participants were included (mean age 67.9 years (SD 9.1), MMSE score 24.5 (SD 4.1)). Diagnoses were mild cognitive impairment or dementia due to AD (49%), Lewy body dementia (22%), frontotemporal dementia (10%) and others (19%). Mean time between tests was 5.1 months (SD 4.1). OPA between single CSF biomarkers and amyloid-PET was 74% for Aβ1-42, 75% for pTau181, 73% for tTau. The use of biomarker ratios improved OPA: 87% for Aβ1-42/Aβ1-40 (n = 155), 88% for pTau181/Aβ1-42 (n = 94) and 82% for tTau/Aβ1-42 (n = 160). A + T + N + cases showed the highest agreement between CSF biomarkers and amyloid-PET (96%), followed by A-T-N- cases (89%). Aβ1-42/Aβ1-40 was a better marker of cerebral amyloid deposition, as identified by amyloid tracers, than Aβ1-42 alone. Combined biomarkers in CSF predicted amyloid-PET result better than single biomarkers.

Required improvements for cerebrospinal fluid-based biomarker tests of Alzheimer's disease

INTRODUCTION

Cerebrospinal fluid (CSF) biomarkers represent a well-established tool for diagnosing Alzheimer's disease (AD), independently from the clinical stage, by reflecting the presence of brain amyloidosis (A+) and tauopathy (T+). In front of this important achievement, so far, (i) CSF AD biomarkers have not yet been adopted for routine clinical use in all Centers dedicated to AD, mainly due to inter-lab variation and lack of internationally accepted cutoff values; (ii) we do need to add other biomarkers more suitable to correlate with the clinical stage and disease monitoring; (iii) we also need to detect the co-presence of other 'non-AD' pathologies.

AREAS COVERED

Efforts to establish standardized cutoff values based on large-scale multi-center studies are discussed. The influence of aging and comorbidities on CSF biomarker levels is also analyzed, and possible solutions are presented, i.e. complementing the A/T/(N) system with markers of axonal damage and synaptic derangement.

EXPERT OPINION

The first, mandatory need is to reach common cutoff values and defined (automated) methodologies for CSF AD biomarkers. To properly select subjects deserving CSF analysis, blood tests might represent the first-line approach. In those subjects undergoing CSF analysis, multiple biomarkers, able to give a comprehensive and personalized pathophysiological/prognostic information, should be included.

Global biomarker trends in Alzheimer's research: a bibliometric analysis

Alzheimer's disease (AD) has no effective treatment, although antibody drugs targeting beta-amyloid, mainly aducanumab, have produced useful clinical results. Biomarkers can be used to determine drug regimens effectively and to monitor the effects of drugs. A concept in which biomarkers reflect disease states is emerging. Although several AD biomarker studies have been reported, measurement methods and target molecules are still being validated, and various biomarkers are being explored. This study analyzed trends in research on AD biomarkers using bibliometric methods, revealing an exponential increase in research reports in this field, with the US most active in research. Analysis of the 'Burst' biomarkers using CiteSpace revealed that networks centered on authors, rather than networks among countries, drive new research trends in this area.