Comparison of cerebrospinal fluid, plasma and neuroimaging biomarker utility in Alzheimer's disease.

Alzheimer's disease biomarkers are crucial to understanding disease pathophysiology, aiding accurate diagnosis and identifying target treatments. Although the number of biomarkers continues to grow, the relative utility and uniqueness of each is poorly understood as prior work has typically calculated serial pairwise relationships on only a handful of markers at a time. The present study assessed the cross-sectional relationships among 27 Alzheimer's disease biomarkers simultaneously and determined their ability to predict meaningful clinical outcomes using machine learning. Data were obtained from 527 community-dwelling volunteers enrolled in studies at the Charles F. and Joanne Knight Alzheimer Disease Research Center at Washington University in St Louis. We used hierarchical clustering to group 27 imaging, CSF and plasma measures of amyloid beta, tau [phosphorylated tau (p-tau), total tau t-tau)], neuronal injury and inflammation drawn from MRI, PET, mass-spectrometry assays and immunoassays. Neuropsychological and genetic measures were also included. Random forest-based feature selection identified the strongest predictors of amyloid PET positivity across the entire cohort. Models also predicted cognitive impairment across the entire cohort and in amyloid PET-positive individuals. Four clusters emerged reflecting: core Alzheimer's disease pathology (amyloid and tau), neurodegeneration, AT8 antibody-associated phosphorylated tau sites and neuronal dysfunction. In the entire cohort, CSF p-tau181/Aß40lumi and Aß42/Aß40lumi and mass spectrometry measurements for CSF pT217/T217, pT111/T111, pT231/T231 were the strongest predictors of amyloid PET status. Given their ability to denote individuals on an Alzheimer's disease pathological trajectory, these same markers (CSF pT217/T217, pT111/T111, p-tau/Aß40lumi and t-tau/Aß40lumi) were largely the best predictors of worse cognition in the entire cohort. When restricting analyses to amyloid-positive individuals, the strongest predictors of impaired cognition were tau PET, CSF t-tau/Aß40lumi, p-tau181/Aß40lumi, CSF pT217/217 and pT205/T205. Non-specific CSF measures of neuronal dysfunction and inflammation were poor predictors of amyloid PET and cognitive status. The current work utilized machine learning to understand the interrelationship structure and utility of a large number of biomarkers. The results demonstrate that, although the number of biomarkers has rapidly expanded, many are interrelated and few strongly predict clinical outcomes. Examining the entire corpus of available biomarkers simultaneously provides a meaningful framework to understand Alzheimer's disease pathobiological change as well as insight into which biomarkers may be most useful in Alzheimer's disease clinical practice and trials.

Cerebrospinal fluid neurofilament light chain is a marker of aging and white matter damage.

BACKGROUND: Cerebrospinal fluid (CSF) neurofilament light chain (NfL) reflects neuro-axonal damage and is increasingly used to evaluate disease progression across neurological conditions including Alzheimer disease (AD). However, it is unknown how NfL relates to specific types of brain tissue. We sought to determine whether CSF NfL is more strongly associated with total gray matter, white matter, or white matter hyperintensity (WMH) volume, and to quantify the relative importance of brain tissue volume, age, and AD marker status (i.e., APOE genotype, brain amyloidosis, tauopathy, and cognitive status) in predicting CSF NfL. METHODS: 419 participants (Clinical Dementia Rating [CDR] Scale > 0, N = 71) had CSF, magnetic resonance imaging (MRI), and neuropsychological data. A subset had amyloid positron emission tomography (PET) and tau PET. Pearson correlation analysis was used to determine the association between CSF NfL and age. Multiple regression was used to determine which brain volume (i.e., gray, white, or WMH volume) most strongly associated with CSF NfL. Stepwise regression and dominance analyses were used to determine the individual contributions and relative importance of brain volume, age, and AD marker status in predicting CSF NfL. RESULTS: CSF NfL increased with age (r = 0.59, p < 0.001). Elevated CSF NfL was associated with greater total WMH volume (p < 0.001), but not gray or white matter volume (p's > 0.05) when considered simultaneously. Age and WMH volume were consistently more important (i.e., have greater R2 values) than AD markers when predicting CSF NfL. CONCLUSIONS: CSF NfL is a non-specific marker of aging and white matter integrity with limited sensitivity to specific markers of AD. CSF NfL likely reflects processes associated with cerebrovascular disease.

Network dysfunction in cognitively normal APOE ε4 carriers is related to subclinical tau.

INTRODUCTION: Apolipoprotein E (APOE) ε4 allele status is associated with amyloid and tau-related pathological changes related to Alzheimer's disease (AD). However, it is unknown whether brain network changes are related to amyloid beta (Aß) and/or tau-related pathology in cognitively normal APOE ε4 carriers with subthreshold Aß accumulation. METHODS: Resting state functional connectivity measures of network integrity were evaluated in cognitively normal individuals (n = 121, mean age 76.6 ± 7.8 years, 15% APOE ε4 carriers, 65% female) with minimal Aß per cerebrospinal fluid (CSF) or amyloid positron emission tomography. RESULTS: APOE ε4 carriers had increased lateralized connections relative to callosal connections within the default-mode, memory, and salience networks (P = .02), with significant weighting on linear regression toward CSF total tau (P = .03) and CSF phosphorylated tau at codon 181 (P = .03), but not CSF Aß42 . DISCUSSION: Cognitively normal APOE ε4 carriers with subthreshold amyloid accumulation may have network reorganization associated with tau.

Resting state network mapping in individuals using deep learning.

Introduction: Resting state functional MRI (RS-fMRI) is currently used in numerous clinical and research settings. The localization of resting state networks (RSNs) has been utilized in applications ranging from group analysis of neurodegenerative diseases to individual network mapping for pre-surgical planning of tumor resections. Reproducibility of these results has been shown to require a substantial amount of high-quality data, which is not often available in clinical or research settings. Methods: In this work, we report voxelwise mapping of a standard set of RSNs using a novel deep 3D convolutional neural network (3DCNN). The 3DCNN was trained on publicly available functional MRI data acquired in n = 2010 healthy participants. After training, maps that represent the probability of a voxel belonging to a particular RSN were generated for each participant, and then used to calculate mean and standard deviation (STD) probability maps, which are made publicly available. Further, we compared our results to previously published resting state and task-based functional mappings. Results: Our results indicate this method can be applied in individual subjects and is highly resistant to both noisy data and fewer RS-fMRI time points than are typically acquired. Further, our results show core regions within each network that exhibit high average probability and low STD. Discussion: The 3DCNN algorithm can generate individual RSN localization maps, which are necessary for clinical applications. The similarity between 3DCNN mapping results and task-based fMRI responses supports the association of specific functional tasks with RSNs.

Modeling the Effects of HIV and Aging on Resting-State Networks Using Machine Learning.

BACKGROUND: The relationship between HIV infection, the functional organization of the brain, cognitive impairment, and aging remains poorly understood. Understanding disease progression over the life span is vital for the care of people living with HIV (PLWH). SETTING: Virologically suppressed PLWH (n = 297) on combination antiretroviral therapy and 1509 HIV-uninfected healthy controls were evaluated. PLWH were further classified as cognitively normal (CN) or cognitively impaired (CI) based on neuropsychological testing. METHODS: Feature selection identified resting-state networks (RSNs) that predicted HIV status and cognitive status within specific age bins (younger than 35 years, 35-55 years, and older than 55 years). Deep learning models generated voxelwise maps of RSNs to identify regional differences. RESULTS: Salience (SAL) and parietal memory networks (PMNs) differentiated individuals by HIV status. When comparing controls with PLWH CN, the PMN and SAL had the strongest predictive strength across all ages. When comparing controls with PLWH CI, the SAL, PMN, and frontal parietal network (FPN) were the best predictors. When comparing PLWH CN with PLWH CI, the SAL, FPN, basal ganglia, and ventral attention were the strongest predictors. Only minor variability in predictive strength was observed with aging. Anatomically, differences in RSN topology occurred primarily in the dorsal and rostral lateral prefrontal cortex, cingulate, and caudate. CONCLUSION: Machine learning identified RSNs that classified individuals by HIV status and cognitive status. The PMN and SAL were sensitive for discriminating HIV status, with involvement of FPN occurring with cognitive impairment. Minor differences in RSN predictive strength were observed by age. These results suggest that specific RSNs are affected by HIV, aging, and HIV-associated cognitive impairment.

Neuroimaging the Neuropathogenesis of HIV.

PURPOSE OF REVIEW: This review highlights neuroimaging studies of HIV conducted over the last 2 years and discusses how relevant findings further our knowledge of the neuropathology of HIV. Three major avenues of neuroimaging research are covered with a particular emphasis on inflammation, aging, and substance use in persons living with HIV (PLWH). RECENT FINDINGS: Neuroimaging has been a critical tool for understanding the neuropathological underpinnings observed in HIV. Recent studies comparing levels of neuroinflammation in PLWH and HIV-negative controls show inconsistent results but report an association between elevated neuroinflammation and poorer cognition in PLWH. Other recent neuroimaging studies suggest that older PLWH are at increased risk for brain and cognitive compromise compared to their younger counterparts. Finally, recent findings also suggest that the effects of HIV may be exacerbated by alcohol and drug abuse. These neuroimaging studies provide insight into the structural, functional, and molecular changes occurring in the brain due to HIV. HIV triggers a strong neuroimmune response and may lead to a cascade of events including increased chronic inflammation and cognitive decline. These outcomes are further exacerbated by age and age-related comorbidities, as well as lifestyle factors such as drug use/abuse.

Modeling autosomal dominant Alzheimer's disease with machine learning.

INTRODUCTION: Machine learning models were used to discover novel disease trajectories for autosomal dominant Alzheimer's disease. METHODS: Longitudinal structural magnetic resonance imaging, amyloid positron emission tomography (PET), and fluorodeoxyglucose PET were acquired in 131 mutation carriers and 74 non-carriers from the Dominantly Inherited Alzheimer Network; the groups were matched for age, education, sex, and apolipoprotein ε4 (APOE ε4). A deep neural network was trained to predict disease progression for each modality. Relief algorithms identified the strongest predictors of mutation status. RESULTS: The Relief algorithm identified the caudate, cingulate, and precuneus as the strongest predictors among all modalities. The model yielded accurate results for predicting future Pittsburgh compound B (R2  = 0.95), fluorodeoxyglucose (R2  = 0.93), and atrophy (R2  = 0.95) in mutation carriers compared to non-carriers. DISCUSSION: Results suggest a sigmoidal trajectory for amyloid, a biphasic response for metabolism, and a gradual decrease in volume, with disease progression primarily in subcortical, middle frontal, and posterior parietal regions.

Socioeconomic Status Mediates Racial Differences Seen Using the AT(N) Framework.

OBJECTIVES: African Americans are at greater risk for developing Alzheimer's disease (AD) dementia than non-Hispanic whites. In addition to biological considerations (eg, genetic influences and comorbid disorders), social and environmental factors may increase the risk of AD dementia. This paper (1) assesses neuroimaging biomarkers of amyloid (A), tau (T), and neurodegeneration (N) for potential racial differences and (2) considers mediating effects of socioeconomic status (SES) and measures of small vessel and cardiovascular disease on observed race differences. METHODS: Imaging measures of AT(N) (amyloid and tau positron emission tomography [PET]) structural magnetic resonance imaging (MRI), and resting state functional connectivity (rs-fc) were collected from African American (n = 131) and white (n = 685) cognitively normal participants age 45 years and older. Measures of small vessel and cardiovascular disease (white matter hyperintensities [WMHs] on MRI, blood pressure, and body mass index [BMI]) and area-based SES were included in mediation analyses. RESULTS: Compared to white participants, African American participants had greater neurodegeneration, as measured by decreased cortical volumes (Cohen's f2 = 0.05, p < 0.001). SES mediated the relationship between race and cortical volumes. There were no significant race effects for amyloid, tau, or rs-fc signature. INTERPRETATION: Modifiable factors, such as differences in social contexts and resources, particularly area-level SES, may contribute to observed racial differences in AD. Future studies should emphasize collection of relevant psychosocial factors in addition to the development of intentional diversity and inclusion efforts to improve the racial/ethnic and socioeconomic representativeness of AD studies. ANN NEUROL 2021;89:254-265.

Sex-related Differences in Tau Positron Emission Tomography (PET) and the Effects of Hormone Therapy (HT).

IMPORTANCE: Female sex is a major risk factor for late-onset Alzheimer disease (AD), and sex hormones have been implicated as a possible protective factor. Neuroimaging studies that evaluated the effects of sex hormones on brain integrity have primarily emphasized neurodegenerative measures rather than amyloid and tau burden. OBJECTIVE: We compared cortical amyloid and regional tau positron emission tomography (PET) deposition between cognitively normal males and females. We also compared preclinical AD pathology between females who have and have not used hormone therapy (HT). Finally, we compared the effects of amyloid and tau pathology on cognition, testing for both sex and HT effects. DESIGN, SETTING, AND PARTICIPANTS: We analyzed amyloid, tau, and cognition in a cognitively normal cross-sectional cohort of older individuals (n=148) followed at the Knight Alzheimer Disease Research Center. Amyloid and tau PET, medication history, and neuropsychological testing were obtained for each participant. RESULTS: Within cognitively normal individuals, there was no difference in amyloid burden by sex. Whether or not we controlled for amyloid burden, female participants had significantly higher tau PET levels than males in multiple regions, including the rostral middle frontal and superior and middle temporal regions. HT accounted for a small reduction in tau PET; however, males still had substantially lower tau PET compared with females. Amyloid PET and tau PET burden were negatively associated with cognitive performance, although increasing amyloid PET did not have a deleterious effect on cognitive performance for women with a history of HT. CONCLUSIONS AND RELEVANCE: Regional sex-related differences in tau PET burden may contribute to the disparities in AD prevalence between males and females. The observed decreases tau PET burden in HT users has important implications for clinical practice and trials and deserves future consideration in longitudinal studies.

Cerebrospinal fluid Aß42 moderates the relationship between brain functional network dynamics and cognitive intraindividual variability.

As Alzheimer's disease (AD) pathology accumulates, resting-state functional connectivity (rs-fc) within and between brain networks decreases, and fluctuations in cognitive performance known as intraindividual variability (IIV) increase. Here, we assessed the relationship between IIV and anticorrelations in rs-fc between the default mode network (DMN)-dorsal attention network (DAN) in cognitively normal older adults and symptomatic AD participants. We also evaluated the relationship between cerebrospinal fluid (CSF) biomarkers of AD (amyloid-beta [Aß42] and tau) and IIV-anticorrelation in rs-fc. We observed that cognitive IIV and anticorrelations between DMN × DAN were higher in individuals with AD compared with cognitively normal participants. As DMN × DAN relationship became more positive, cognitive IIV increased, indicating that stronger anticorrelations between networks support more consistent cognitive performance. Moderation analyses indicated that continuous CSF Aß42, but not CSF total tau, moderated the relationship between cognitive IIV and DMN × DAN, collectively demonstrating that greater amyloid burden and alterations in functional network dynamics are associated with cognitive changes seen in AD. These findings are valuable, as they suggest that amyloid affects cognitive functioning during the early stages of AD.