INTRODUCTION: The Health and Aging Brain Study–Health Disparities (HABS‐HD) is an ongoing prospective study aimed at understanding brain health and aging. The current work provides a description of the cohort baseline characteristics and outlines the study methodology. METHODS: We analyzed available data from n = 1066 non‐Hispanic Black (NHB), n = 1425 Hispanic, and n = 1349 non‐Hispanic White (NHW) partners who were actively enrolled in HABS‐HD. Descriptive statistics are presented for each racial/ethnic group across demographic, medical, and diagnostic characteristics. Differences in select amyloid (A), tau (T), and neurodegenerative (N) biomarkers spanning proteomic and neuroimaging were examined along with groupwise differences for cognitive test performance and select social determinants of health (SDoH) factors. RESULTS: The characteristics of the cohort revealed significant groupwise differences in age, education, sex, and cognitive diagnosis. Higher rates of cognitive impairment (mild cognitive impairment [MCI] and dementia) were found in NHB and Hispanic compared to NHW partners, despite the latter group being older. There were also higher rates of hypertension and diabetes among NHB and Hispanic compared to NHW partners. Differences were also found across many plasma (A/T[N]) biomarkers and select neuroimaging measures, including meta‐region of interest and white matter hyperintensities. Positron emission tomography amyloid positivity rates (but not tau positivity) were found to differ, with higher rates observed among NHW (15% amyloid positivity) compared to NHB (3%) partners. Groups also differed by select SDoH factors, including household income, insurance, having a health‐care provider, and Area Deprivation Index. All cognitive measures also revealed groupwise differences. DISCUSSION: The baseline cohort characteristics for HABS‐HD reveal significant group differences spanning demographic, medical, cognitive, and biological factors (including A/T[N] biomarkers), which are all critical to understand as they relate to aging and age‐related diseases. Highlights: The Health and Aging Brain Study–Health Disparities (HABS‐HD) cohort baseline characteristics and study methodology were provided. Cohort characteristic differences were found across demographic factors, including age, education, sex, and cognitive diagnosis across the representative groups. Biomarkers for amyloid/tau/neurodegeneration spanning blood and neuroimaging were shown to differ as well across cohort groups. Positivity rates for amyloid positron emission tomography were lower for non‐Hispanic Black partners in the cohort compared to non‐Hispanic White partners.

Background: The need for diagnostic biomarkers of cognitive decline is particularly important among aging adults with Down syndrome (DS). Growing empirical support has identified the utility of plasma derived biomarkers among neurotypical adults with mild cognitive impairment (MCI) and Alzheimer’s disease (AD); however, the application of such biomarkers has been limited among the DS population. Objective: This study aimed to investigate the cross-sectional diagnostic performance of plasma neurofilament light chain (Nf-L) and total-tau, individually and in combination among a cohort of DS adults. Methods: Plasma samples were analyzed from n = 305 (n = 225 cognitively stable (CS); n = 44 MCI-DS; n = 36 DS–AD) participants enrolled in the Alzheimer’s Biomarker Consortium —Down Syndrome. Results: In distinguishing DS-AD participants from CS, Nf-L alone produced an AUC of 90%, total-tau alone reached 74%, and combined reached an AUC of 86%. When age and gender were included, AUC increased to 93%. Higher values of Nf-L, total-tau, and age were all shown to be associated with increased risk for DS-AD. When distinguishing MCI-DS participants from CS, Nf-L alone produced an AUC of 65%, while total-tau alone reached 56%. A combined model with Nf-L, total-tau, age, and gender produced an AUC of 87%. Both higher values in age and total-tau were found to increase risk for MCI-DS; Nf-L levels were not associated with increased risk for MCI-DS. Conclusion: Advanced assay techniques make total-tau and particularly Nf-L useful biomarkers of both AD pathology and clinical status in DS and have the potential to serve as outcome measures in clinical trials for future disease-modifying drugs.

INTRODUCTION: Alzheimer's disease (AD) biomarkers of Amyloid(A), Tau(T), and Neurodegeneration(N) have been increasingly studied to fill the gap in our understanding of racial and ethnic differences. This study aimed to examine the relationship between plasma AT(N) biomarkers and (1) AT(N) neuroimaging biomarkers, (2) demographics, (3) medical comorbidities, and (4) cognitive diagnosis. METHODS: Data were analyzed from n = 764 non‐Hispanic Black (NHB), n = 1230 Hispanic, and n = 1232 non‐Hispanic White (NHW) participants. Plasma AT(N) biomarkers were derived using single molecule array (SIMOA) technology on an HD‐X imager and included amyloid beta (Aβ)42/40, total tau, ptau181, and neurofilament light chain (NfL). Clinical reads of positron emission tomography (PET) amyloid and tau positivity were used to examine the link between AT(N) plasma and neuroimaging biomarkers. Generalized linear models were conducted to examine the relationship between plasma AT(N) biomarkers and select demographic, diagnostic, and medical comorbidities (hypertension, diabetes, dyslipidemia, chronic kidney disease). RESULTS: Differences in the AT(N) biomarkers were found across racial/ethnic groups. Plasma Aβ42/40 was found to be associated with PET amyloid positivity only among NHW participants, while plasma NfL was found to correlate with Meta‐ROI among NHB and Hispanic participants. Ptau181 was associated with PET amyloid positivity among NHB and NHW participants and well as PET tau positivity among the latter group and Hispanic participants. Diabetes was related to increased plasma AT(N) biomarkers among NHB and Hispanic participants. CKD was associated with increased AT(N) biomarkers for all race/ethnic groups with the exception of Aβ42/40. While Aβ42/40, total tau, ptau181, and NfL were found to be related to a dementia diagnosis among NHW participants, only ptau181 and NfL were found to be related to this same diagnostic category among NHB and Hispanic participants. DISCUSSION: Our findings indicate differential relationships between comorbidities (demographic, medical, diagnostic) across NHB, Hispanic, and NHW participants. This work expands our knowledge regarding the associations of plasma biomarkers to AD pathology in diverse populations. Highlights: Differences in AT(N) plasma biomarkers were found in a diverse community cohort. While plasma Aβ42/40 was associated with PET amyloid positivity among non‐Hispanic white participants, this did not apply to non‐Hispanic Black or Hispanic participants. Medical comorbidity of diabetes and chronic kidney disease was related to increased plasma AT(N) biomarkers among the ethnically diverse segment of the cohort. Plasma AT(N) biomarkers were more so related to a diagnosis of dementia for non‐Hispanic white as compared to Hispanic or non‐Hispanic Black participants. Across racial/ethnic groups, the plasma biomarkers of neurodegeneration (NfL) and ptau181 were related to a diagnosis of dementia.

Introduction Previously generated serum and plasma proteomic profiles were examined among adults with Down syndrome (DS) to determine whether these profiles could discriminate those with mild cognitive impairment (MCI‐DS) and Alzheimer's disease (DS‐AD) from those cognitively stable (CS). Methods Data were analyzed on n = 305 (n = 225 CS; n = 44 MCI‐DS; n = 36 DS‐AD) enrolled in the Alzheimer's Biomarker Consortium–Down Syndrome (ABC–DS). Results Distinguishing MCI‐DS from CS, the serum profile produced an area under the curve (AUC) = 0.95 (sensitivity [SN] = 0.91; specificity [SP] = 0.99) and an AUC = 0.98 (SN = 0.96; SP = 0.97) for plasma when using an optimized cut‐off score. Distinguishing DS‐AD from CS, the serum profile produced an AUC = 0.93 (SN = 0.81; SP = 0.99) and an AUC = 0.95 (SN = 0.86; SP = 1.0) for plasma when using an optimized cut‐off score. AUC remained unchanged to slightly improved when age and sex were included. Eotaxin3, interleukin (IL)‐10, C‐reactive protein, IL‐18, serum amyloid A , and FABP3 correlated fractions at r² > = 0.90. Discussion Proteomic profiles showed excellent detection accuracy for MCI‐DS and DS‐AD.

Blood‐based biomarkers continue to be explored for disease detection, monitoring of progression, and therapeutic outcomes as the diagnostic determination of Alzheimer's Disease in Down Syndrome (DS‐AD) remains challenging in clinical settings. This perspective highlights the current status of this effort. Overall, amyloid (A), tau (T), and neurodegeneration (AT[N]) blood‐based biomarkers have been shown to increase with disease pathology for individuals with DS. Phosphorylated tau biomarkers (p‐tau217, p‐tau181) have been consistently shown to track disease progression for DS‐AD and are likely good candidates for use in clinical settings. Biomarkers of inflammation (glial fibrillary acidic protein) also show promise; however, additional work is needed. Findings from stability work of blood‐based biomarkers conducted among non‐DS also support the potential longitudinal utility of biomarkers such as neurofilament light chain and p‐tau181 in DS. Gaps in our knowledge are highlighted, and a potential role for sex differences in biomarker outcomes is noted, along with recommendations for determining the appropriate context of use when translating biomarkers into clinical applications. Highlights: An overview of blood‐based biomarkers for Alzheimer's disease (AD) was provided for consideration of their utility among individuals with Down syndrome when looking toward potential clinical applications. Longitudinal stability of many blood biomarkers and improvement in detection sensitivity make blood such as plasma a viable source for exploring AD pathology. Variability in reviewed findings regarding the application of blood biomarkers highlights the importance of understanding and defining the appropriate context of use, particularly when translating them into clinical practice.

Accurate detection is still a challenge in machine learning (ML) for Alzheimer’s disease (AD). Class imbalance in imbalanced AD data is another big challenge for machine-learning algorithms working under the assumption that the data are evenly distributed within classes. Here, we present a hyperparameter tuning workflow with high-performance computing (HPC) for imbalanced data related to prevalent mild cognitive impairment (MCI) and AD in the Health and Aging Brain Study-Health Disparities (HABS-HD) project. We applied a single-node multicore parallel mode to hyperparameter tuning of gamma, cost, and class weight using a support vector machine (SVM) model with 10 times repeated fivefold cross-validation. We executed the hyperparameter tuning workflow with R’s bigmemory, foreach, and doParallel packages on Texas Advanced Computing Center (TACC)’s Lonestar6 system. The computational time was dramatically reduced by up to 98.2% for the high-performance SVM hyperparameter tuning model, and the performance of cross-validation was also improved (the positive predictive value and the negative predictive value at base rate 12% were, respectively, 16.42% and 92.72%). Our results show that a single-node multicore parallel structure and high-performance SVM hyperparameter tuning model can deliver efficient and fast computation and achieve outstanding agility, simplicity, and productivity for imbalanced data in AD applications.