Imaging as a biomarker in drug discovery for Alzheimer's disease: is MRI a suitable technology?

Considerations for biomarker strategies in clinical trials investigating tau-targeting therapeutics for Alzheimer's disease

The use of biomarker-led clinical trial designs has been transformative for investigating amyloid-targeting therapies for Alzheimer's disease (AD). The designs have ensured the correct selection of patients on these trials, supported target engagement and have been used to support claims of disease modification and clinical efficacy. Ultimately, this has recently led to approval of disease-modifying, amyloid-targeting therapies for AD; something that should be noted for clinical trials investigating tau-targeting therapies for AD. There is a clear overlap of the purpose of biomarker use at each stage of clinical development between amyloid-targeting and tau-targeting clinical trials. However, there are differences within the potential context of use and interpretation for some biomarkers in particular measurements of amyloid and utility of soluble, phosphorylated tau biomarkers. Given the complexities of tau in health and disease, it is paramount that therapies target disease-relevant tau and, in parallel, appropriate assays of target engagement are developed. Tau positron emission tomography, fluid biomarkers reflecting tau pathology and downstream measures of neurodegeneration will be important both for participant recruitment and for monitoring disease-modification in tau-targeting clinical trials. Bespoke design of biomarker strategies and interpretations for different modalities and tau-based targets should also be considered.

Synthesis and evaluation of curcumin-based near-infrared fluorescent probes for detection of amyloid β peptide in Alzheimer mouse models

The abnormal accumulation of amyloid β protein (Aβ) is one of the most important causes of Alzheimer's disease (AD) and is usually a detecting biomarker. Curcumin and its derivatives have potential Aβ aggregate targeting ability; we synthesized a series of curcumin-based near-infrared fluorescence probes in this study. By characterizing the excitation wavelength and emission wavelength, the imaging characteristics of the investigation in the near-infrared light region were determined; with an increase in the concentration of the probe compounds, the fluorescence intensity showed an upward trend, demonstrating ideal optical characteristics. In vivo, imaging results showed that the synthesized probe compounds could penetrate the blood-brain barrier (BBB) and specifically bind to Aβ in the brain of APP/PS1 mice. Especially for compound 3b, the maximum emission wavelength was around 667 nm, and the fluorescence signal intensity in the brain of the APP/PS1 mice model was more than twice that of the wild control group at 120 min after administration, which could display Aβ pathological changes. The fluorescent probes designed in this study can become an effective tool for early AD diagnosis and visual detection.

History in perspective: How Alzheimer's Disease came to be where it is?

Treatment of Alzheimer's Disease (AD) remains an unsolved issue despite the pronounced global attention it has received from researchers over the last four decades. Determining the primary cause of the disease is challenging due to its long prodromal phase and multifactorial etiology. Regardless, academic disagreements amongst the scientific community have helped in making significant advancements in underpinning the molecular basis of disease pathogenesis. Substantial development in fluid and imaging biomarkers for AD led to a sharp turn in defining the disease as a molecular construct, dispensing its clinical definition. With conceptual progress, revisions in the diagnostic criteria of AD were made, culminating into the research framework proposed by National Institute on Aging and Alzheimer's Association in 2018 which unified different stages of the disease continuum, giving a common language of AT(N)¹ classification to researchers. With realization that dementia is the final stage of AD spectrum, its early diagnosis by means of cerebrospinal fluid biomarkers, Positron Emission Tomography and Magnetic Resonance Imaging of the brain holds crucial importance in discovering ways of halting the disease progression. This article maps the insights into the pathogenesis as well as the diagnostic criteria and tests for AD as these have evolved over time. A contextualized timeline of how the understanding of AD has matured with advancing knowledge allows future research to be directed and unexplored avenues to be prioritized.

Machine Learning-Based Framework for Differential Diagnosis Between Vascular Dementia and Alzheimer's Disease Using Structural MRI Features

Background and Objective: Vascular dementia (VaD) and Alzheimer's disease (AD) could be characterized by the same syndrome of dementia. This study aims to assess whether multi-parameter features derived from structural MRI can serve as the informative biomarker for differential diagnosis between VaD and AD using machine learning.

Methods: A total of 93 patients imaged with brain MRI including 58 AD and 35 VaD confirmed by two chief physicians were recruited in this study from June 2013 to July 2019. Automated brain tissue segmentation was performed by the AccuBrain tool to extract multi-parameter volumetric measurements from different brain regions. Firstly, a total of 62 structural MRI biomarkers were addressed to select significantly different features between VaD and AD for dimensionality reduction. Then, the least absolute shrinkage and selection operator (LASSO) was further used to construct a feature set that is fed into a support vector machine (SVM) classifier. To ensure the unbiased evaluation of model performance, a comparative study of classification models was implemented by using different machine learning algorithms in order to determine which performs best in the application of differential diagnosis between VaD and AD. The diagnostic performance of the classification models was evaluated by the quantitative metrics derived from the receiver operating characteristic curve (ROC).

Results: The experimental results demonstrate that the SVM with RBF achieved an encouraging performance with sensitivity (SEN), specificity (SPE), and accuracy (ACC) values of 82.65%, 87.17%, and 84.35%, respectively (AUC = 0.861, 95% CI = 0.820–0.902), for the differential diagnosis between VaD and AD.

Conclusions: The proposed computer-aided diagnosis method highlights the potential of combining structural MRI and machine learning to support clinical decision making in distinction of VaD vs. AD.

Automated Detection of Alzheimer's Disease Using Brain MRI Images- A Study with Various Feature Extraction Techniques

The aim of this work is to develop a Computer-Aided-Brain-Diagnosis (CABD) system that can determine if a brain scan shows signs of Alzheimer's disease. The method utilizes Magnetic Resonance Imaging (MRI) for classification with several feature extraction techniques. MRI is a non-invasive procedure, widely adopted in hospitals to examine cognitive abnormalities. Images are acquired using the T2 imaging sequence. The paradigm consists of a series of quantitative techniques: filtering, feature extraction, Student's t-test based feature selection, and k-Nearest Neighbor (KNN) based classification. Additionally, a comparative analysis is done by implementing other feature extraction procedures that are described in the literature. Our findings suggest that the Shearlet Transform (ST) feature extraction technique offers improved results for Alzheimer's diagnosis as compared to alternative methods. The proposed CABD tool with the ST + KNN technique provided accuracy of 94.54%, precision of 88.33%, sensitivity of 96.30% and specificity of 93.64%. Furthermore, this tool also offered an accuracy, precision, sensitivity and specificity of 98.48%, 100%, 96.97% and 100%, respectively, with the benchmark MRI database.

Join and Deep Ensemble Regression of Clinical Scores for Alzheimer's Disease Using Longitudinal and Incomplete Data<sup/>

Alzheimer's disease (AD) is a neurodegenerative disease with an irreversible and progressive process. Close monitoring of AD is essential for making adjustments in the treatment plan. Since clinical scores can indicate the disease status effectively, the prediction of the scores based on the magnetic resonance imaging (MRI data is highly desirable. Different from previous studies at a single time point, we propose to build a model to explore the relationship between MRI data and scores, thereby predicting longitudinal scores at future time points from the corresponding MRI data. The model incorporates three parts, correntropy regularized joint learning-based feature selection, deep polynomial network based feature encoding, and finally, support vector regression. The regression process is carried out for two scenarios. One is to use baseline data for predictions at future time points, and the other is to combine all the previous data for the prediction at the next time point. Meanwhile, the missing scores are filled in the second scenario to address the incompleteness presented in the data. The simulation results demonstrate that the proposed model accurately describes the relationship between MRI data and scores, and thus it can be effective in predicting longitudinal scores.

Changing Landscapes in the Neuroimaging of Dementia

Neuroimaging in dementia has advanced several folds in the past decade. It has evolved from diagnosing secondary causes of dementia to the current use in identifying primary dementia and aid in clinically perplexing situations. There has been a leap in the imaging technology that can virtually dissect the brain with a high degree of radiopathological correlation. The neuroimaging in dementia is classified into structural, functional, and molecular imaging. Structural imaging includes voxel-based morphometry and diffusion tensor imaging. Functional imaging includes 18F-fluorodeoxy glucose positron emission tomography imaging, 99mTc hexamethylpropyleneamineoxime single photon emission computed tomography imaging, and functional magnetic resonance imaging studies. Molecular imaging includes amyloid imaging, tau imaging, and translocated protein imaging. These advancements have led to using neuroimaging as a biomarker in assessing the progression and also in deciphering prognosis of the disease. In this article, we discuss the current clinical relevance of these neurological advancements.

Neuromarkers for Mental Disorders: Harnessing Population Neuroscience

Despite abundant research into the neurobiology of mental disorders, to date neurobiological insights have had very little impact on psychiatric diagnosis or treatment. In this review, we contend that the search for neuroimaging biomarkers-neuromarkers-of mental disorders is a highly promising avenue toward improved psychiatric healthcare. However, many of the traditional tools used for psychiatric neuroimaging are inadequate for the identification of neuromarkers. Specifically, we highlight the need for larger samples and for multivariate analysis. Approaches such as machine learning are likely to be beneficial for interrogating high-dimensional neuroimaging data. We suggest that broad, population-based study designs will be important for developing neuromarkers of mental disorders, and will facilitate a move away from a phenomenological definition of mental disorder categories and toward psychiatric nosology based on biological evidence. We provide an outline of how the development of neuromarkers should occur, emphasizing the need for tests of external and construct validity, and for collaborative research efforts. Finally, we highlight some concerns regarding the development, and use of, neuromarkers in psychiatric healthcare.

Taming Alzheimer's disease: New perspectives, newer horizons

Alzheimer's disease (AD) is the leading cause of dementia. However, current therapies do not prevent progression of the disease. New research into the pathogenesis of the disease has brought about a greater understanding of the "amyloid cascade" and associated receptor abnormalities, the role of genetic factors, and revealed that the disease process commences 10 to 20 years prior to the appearance of clinical signs. This greater understanding of the disease has prompted development of novel disease-modifying therapies (DMTs) which may prevent onset or delay progression of the disease. Using genetic biomarkers like apolipoprotein E (ApoE) ε4, biochemical biomarkers like cerebrospinal fluid (CSF) amyloid and tau proteins, and imaging biomarkers like magnetic resonance imaging (MRI) and positron emission tomography (PET), it is now possible to detect preclinical AD and also monitor its progression in asymptomatic people. These biomarkers can be used in the selection of high-risk populations for clinical trials and also to monitor the efficacy and side-effects of DMT. To validate and standardize these biomarkers and select the most reliable, repeatable, easily available, cost-effective and complementary options is the challenge ahead.

Neuroimaging Endpoints in Amyotrophic Lateral Sclerosis

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative, clinically heterogeneous syndrome pathologically overlapping with frontotemporal dementia. To date, therapeutic trials in animal models have not been able to predict treatment response in humans, and the revised ALS Functional Rating Scale, which is based on coarse disability measures, remains the gold-standard measure of disease progression. Advances in neuroimaging have enabled mapping of functional, structural, and molecular aspects of ALS pathology, and these objective measures may be uniquely sensitive to the detection of propagation of pathology in vivo. Abnormalities are detectable before clinical symptoms develop, offering the potential for neuroprotective intervention in familial cases. Although promising neuroimaging biomarker candidates for diagnosis, prognosis, and disease progression have emerged, these have been from the study of necessarily select patient cohorts identified in specialized referral centers. Further multicenter research is now needed to establish their validity as therapeutic outcome measures.

Structural Neuroimaging of the Medial Temporal Lobe in Alzheimer's Disease Clinical Trials

Atrophy in the medial temporal lobe (MTA) is being used as a criterion to support a diagnosis of Alzheimer's disease (AD). There are several structural neuroimaging approaches for quantifying MTA, including semiquantitative visual rating scales, volumetry (3D), planimetry (2D), and linear measures (1D). Current applications of structural neuroimaging in Alzheimer's disease clinical trials (ADCTs) incorporate it as a tool for improving the selection of subjects for enrollment or for stratification, for tracking disease progression, or providing evidence of target engagement for new therapeutic agents. It may also be used as a surrogate marker, providing evidence of disease-modifying effects. However, despite the widespread use of volumetric magnetic resonance imaging (MRI) in ADCTs, there are some important challenges and limitations, such as difficulties in the interpretation of results, limitations in translating results into clinical practice, and reproducibility issues, among others. Solutions to these issues may arise from other methodologies that are able to link the results of volumetric MRI from trials with conventional MRIs performed in routine clinical practice (linear or planimetric methods). Also of potential benefit are automated volumetry, using indices for comparing the relative rate of atrophy of different regions instead of absolute rates of atrophy, and combining structural neuroimaging with other biomarkers. In this review, authors present the existing structural neuroimaging approaches for MTA quantification. They then discuss solutions to the limitations of the different techniques as well as the current challenges of the field. Finally, they discuss how the current advances in AD neuroimaging can help AD diagnosis.

Finding effective biomarkers for pediatric traumatic brain injury

As traumatic brain injury (TBI) continues to affect children and young adults worldwide, research on reliable biomarkers grows as a possible aid in determining the severity of injury. However, many studies have revealed that diverse biomarkers such as S100B and myelin basic protein (MBP) have many limitations, such as their elevated normative concentrations in young children. Therefore, the results of these studies have yet to be translated to clinical applications. However, despite the setbacks of research into S100B and MBP, investigators continue to research viable biomarkers, notably glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase L1 (UCH-L1), as possible aids in medical decision making. Studies have revealed that GFAP and UCH-L1 actually are better predictors of injury progression than the before-mentioned biomarkers S100B and MBP. In addition, UCH-L1 has demonstrated an ability to detect injury while CT is negative, suggesting an ability to detect acute intracranial lesions. Here, we evaluate research testing levels of GFAP and UCH-L1 on children diagnosed with TBI and compare our results to those of other tested biomarkers. In a recent study done by Hayes et al., GFAP and UCH-L1 demonstrated the potential to recognize children with the possibility of poor outcome, allowing for more specialized treatments with clinical and laboratory applications. Although studies on GFAP and UCH-L1 have for the most part warranted positive results, further studies will be needed to confirm their role as reliable markers for pediatric TBI.

Serum Concentrations of Ubiquitin C-Terminal Hydrolase-L1 and Glial Fibrillary Acidic Protein after Pediatric Traumatic Brain Injury

Objective reliable markers to assess traumatic brain injury (TBI) and predict outcome soon after injury are a highly needed tool for optimizing management of pediatric TBI. We assessed serum concentrations of Glial Fibrillary Acidic Protein (GFAP) and Ubiquitin C-Terminal Hydrolase-L1 (UCH-L1) in a cohort of 45 children with clinical diagnosis of TBI (Glasgow Coma Scale [GCS] 3–15) and 40 healthy subjects, evaluated their associations with clinical characteristics and outcomes, and compared their performance to previously published data on two well-studied blood biomarkers, S100B and MBP. We observed higher serum levels of GFAP and UCH-L1 in brain-injured children compared with controls and also demonstrated a step-wise increase of biomarker concentrations over the continuum of severity from mild to severe TBI. Furthermore, while we found that only the neuronal biomarker UCH-L1 holds potential to detect acute intracranial lesions as assessed by computed tomography (CT), both markers were substantially increased in TBI patients even with a normal CT suggesting the presence of undetected microstructural injuries. Serum UCH-L1 and GFAP concentrations also strongly predicted poor outcome and performed better than S100B and MBP. Our results point to a role of GFAP and UCH-L1 as candidate biomarkers for pediatric TBI. Further studies are warranted.

Identification of Differentially Expressed Genes through Integrated Study of Alzheimer's Disease Affected Brain Regions

Background

Alzheimer’s disease (AD) is the most common form of dementia in older adults that damages the brain and results in impaired memory, thinking and behaviour. The identification of differentially expressed genes and related pathways among affected brain regions can provide more information on the mechanisms of AD. In the past decade, several studies have reported many genes that are associated with AD. This wealth of information has become difficult to follow and interpret as most of the results are conflicting. In that case, it is worth doing an integrated study of multiple datasets that helps to increase the total number of samples and the statistical power in detecting biomarkers. In this study, we present an integrated analysis of five different brain region datasets and introduce new genes that warrant further investigation.

Methods

The aim of our study is to apply a novel combinatorial optimisation based meta-analysis approach to identify differentially expressed genes that are associated to AD across brain regions. In this study, microarray gene expression data from 161 samples (74 non-demented controls, 87 AD) from the Entorhinal Cortex (EC), Hippocampus (HIP), Middle temporal gyrus (MTG), Posterior cingulate cortex (PC), Superior frontal gyrus (SFG) and visual cortex (VCX) brain regions were integrated and analysed using our method. The results are then compared to two popular meta-analysis methods, RankProd and GeneMeta, and to what can be obtained by analysing the individual datasets.

Results

We find genes related with AD that are consistent with existing studies, and new candidate genes not previously related with AD. Our study confirms the up-regualtion of INFAR2 and PTMA along with the down regulation of GPHN, RAB2A, PSMD14 and FGF. Novel genes PSMB2, WNK1, RPL15, SEMA4C, RWDD2A and LARGE are found to be differentially expressed across all brain regions. Further investigation on these genes may provide new insights into the development of AD. In addition, we identified the presence of 23 non-coding features, including four miRNA precursors (miR-7, miR570, miR-1229 and miR-6821), dysregulated across the brain regions. Furthermore, we compared our results with two popular meta-analysis methods RankProd and GeneMeta to validate our findings and performed a sensitivity analysis by removing one dataset at a time to assess the robustness of our results. These new findings may provide new insights into the disease mechanisms and thus make a significant contribution in the near future towards understanding, prevention and cure of AD.

Effects of rivastigmine on visual attention in subjects with amnestic mild cognitive impairment: A serial functional MRI activation pilot-study

A pilot study to investigate the effects of rivastigmine on the brain activation pattern due to visual attention tasks in a group of amnestic Mild Cognitive Impaired patients (aMCI). The design was an initial three-month double blind period with a rivastigmine and placebo arms, followed by a nine-month open-label period. All patients underwent serial functional magnetic resonance imaging (fMRI) at baseline, and after three and six months of follow-up. Primary endpoint was the effect of rivastigmine on functional brain changes during visual attention (face and location matching) tasks. There were five in the rivastigmine arm and two in the placebo arm. The face matching task showed higher activation of visual areas after three months of treatment but no differences compared to baseline at six months. The location matching task showed a higher activation along the dorsal visual pathway at both three and six months follow ups. Treatment with rivastigmine demonstrates a significant effect on brain activation of the dorsal visual pathway during a location matching task in patients with aMCI. Our data support the potential use of task fMRI to map specific treatment effects of cholinergic drugs during prodromal stages of Alzheimer's disease (AD).

Predictors of cognitive decline and treatment response in a clinical trial on suspected prodromal Alzheimer's disease

We determined the value of hippocampus (Hp) and basal forebrain (BF) volumes for predicting cognitive decline and treatment response in a double-blind, randomized, placebo-controlled phase 4 trial at 28 academic centers (France) in patients with amnestic mild cognitive impairment (MCI) receiving Donepezil 10 mg daily or placebo over 12 months, and 6 months open label follow-up. Outcome measures were the rates of global and domain specific cognitive decline as non-primary efficacy endpoint. The intention-to-treat (ITT) sample analyzed comprised 215 cases. Baseline Hp volume was a significant predictor of rates of change in global cognitive function in linear mixed effects models. This effect was independent of treatment. BF volume was not associated with rates of global or domain specific cognitive decline. Rates of delayed free recall decline were higher in MCI cases treated with donepezil compared to placebo. Only Hp, but not BF volume was a useful predictor of cognitive decline in suspected prodromal AD patients. Both Hp and BF volumes were poor predictors of treatment response, questioning previous approaches on predicting treatment response without placebo control.

TRIAL REGISTRATION

clinicalTrials.gov Identifier NCT00403520.

Benefits and risks of add-on therapies for Alzheimer's disease

Despite three decades of intensive research, the efforts of scientific society and industry and the expenditures, numerous attempts to develop effective treatments for Alzheimer's disease have failed. Currently, approved and widely used medications to treat cognitive deficits in Alzheimer's disease are symptomatic only and show at best modest efficacy. In this context, the need to develop a successful, disease-modifying treatment is loudly expressed. One way to achieve this goal is the use of add-on therapies or various combinations of existing ‘conventional’ drugs. Results of several clinical studies and post hoc analyses of combination therapy with all cholinesterase inhibitors and memantine are published. Moreover, there is a need for studies on long-term efficacy of combination therapy in Alzheimer's.

Evaluation of Cross-Protocol Stability of a Fully Automated Brain Multi-Atlas Parcellation Tool

Brain parcellation tools based on multiple-atlas algorithms have recently emerged as a promising method with which to accurately define brain structures. When dealing with data from various sources, it is crucial that these tools are robust for many different imaging protocols. In this study, we tested the robustness of a multiple-atlas, likelihood fusion algorithm using Alzheimer’s Disease Neuroimaging Initiative (ADNI) data with six different protocols, comprising three manufacturers and two magnetic field strengths. The entire brain was parceled into five different levels of granularity. In each level, which defines a set of brain structures, ranging from eight to 286 regions, we evaluated the variability of brain volumes related to the protocol, age, and diagnosis (healthy or Alzheimer’s disease). Our results indicated that, with proper pre-processing steps, the impact of different protocols is minor compared to biological effects, such as age and pathology. A precise knowledge of the sources of data variation enables sufficient statistical power and ensures the reliability of an anatomical analysis when using this automated brain parcellation tool on datasets from various imaging protocols, such as clinical databases.