Cognitive and functional performance and plasma biomarkers of early Alzheimer's disease in Down syndrome.

INTRODUCTION: People with Down syndrome (DS) have a 75% to 90% lifetime risk of Alzheimer's disease (AD). AD pathology begins a decade or more prior to onset of clinical AD dementia in people with DS. It is not clear if plasma biomarkers of AD pathology are correlated with early cognitive and functional impairments in DS, and if these biomarkers could be used to track the early stages of AD in DS or to inform inclusion criteria for clinical AD treatment trials. METHODS: This large cross-sectional cohort study investigated the associations between plasma biomarkers of amyloid beta (Aß)42/40, total tau, and neurofilament light chain (NfL) and cognitive (episodic memory, visual-motor integration, and visuospatial abilities) and functional (adaptive behavior) impairments in 260 adults with DS without dementia (aged 25-81 years). RESULTS: In general linear models lower plasma Aß42/40 was related to lower visuospatial ability, higher total tau was related to lower episodic memory, and higher NfL was related to lower visuospatial ability and lower episodic memory. DISCUSSION: Plasma biomarkers may have utility in tracking AD pathology associated with early stages of cognitive decline in adults with DS, although associations were modest. Highlights: Plasma Alzheimer's disease (AD) biomarkers correlate with cognition prior to dementia in Down syndrome.Lower plasma amyloid beta 42/40 was related to lower visuospatial abilities.Higher plasma total tau and neurofilament light chain were associated with lower cognitive performance.Plasma biomarkers show potential for tracking early stages of AD symptomology.

Neurofilament light chain concentration mediates the association between regional medial temporal lobe structure and memory in adults with Down syndrome.

INTRODUCTION: Virtually all people with Down syndrome (DS) develop neuropathology associated with Alzheimer's disease (AD). Atrophy of the hippocampus and entorhinal cortex (EC), as well as elevated plasma concentrations of neurofilament light chain (NfL) protein, are markers of neurodegeneration associated with late-onset AD. We hypothesized that hippocampus and EC gray matter loss and increased plasma NfL concentrations are associated with memory in adults with DS. METHODS: T1-weighted structural magnetic resonance imaging (MRI) data were collected from 101 participants with DS. Hippocampus and EC volume, as well as EC subregional cortical thickness, were derived. In a subset of participants, plasma NfL concentrations and modified Cued Recall Test scores were obtained. Partial correlation and mediation were used to test relationships between medial temporal lobe (MTL) atrophy, plasma NfL, and episodic memory. RESULTS: Hippocampus volume, left anterolateral EC (alEC) thickness, and plasma NfL were correlated with each other and were associated with memory. Plasma NfL mediated the relationship between left alEC thickness and memory as well as hippocampus volume and memory. DISCUSSION: The relationship between MTL gray matter and memory is mediated by plasma NfL levels, suggesting a link between neurodegenerative processes underlying axonal injury and frank gray matter loss in key structures supporting episodic memory in people with DS.

Cerebrovascular disease drives Alzheimer plasma biomarker concentrations in adults with Down syndrome.

Importance: By age 40 years over 90% of adults with Down syndrome (DS) have Alzheimer's disease (AD) pathology and most progress to dementia. Despite having few systemic vascular risk factors, individuals with DS have elevated cerebrovascular disease (CVD) markers that track with the clinical progression of AD, suggesting a role for CVD that is hypothesized to be mediated by inflammatory factors. Objective: To examine the pathways through which small vessel CVD contributes to AD-related pathophysiology and neurodegeneration in adults with DS. Design: Cross sectional analysis of neuroimaging, plasma, and clinical data. Setting: Participants were enrolled in Alzheimer's Biomarker Consortium - Down Syndrome (ABC-DS), a multisite study of AD in adults with DS. Participants: One hundred eighty-five participants (mean [SD] age=45.2 [9.3] years) with available MRI and plasma biomarker data were included. White matter hyperintensity (WMH) volumes were derived from T2-weighted FLAIR MRI scans and plasma biomarker concentrations of amyloid beta (Aß42/Aß40), phosphorylated tau (p-tau217), astrocytosis (glial fibrillary acidic protein, GFAP), and neurodegeneration (neurofilament light chain, NfL) were measured with ultrasensitive immunoassays. Main Outcomes and Measures: We examined the bivariate relationships of WMH, Aß42/Aß40, p-tau217, and GFAP with age-residualized NfL across AD diagnostic groups. A series of mediation and path analyses examined causal pathways linking WMH and AD pathophysiology to promote neurodegeneration in the total sample and groups stratified by clinical diagnosis. Results: There was a direct and indirect bidirectional effect through GFAP of WMH on p-tau217 concentration, which was associated with NfL concentration in the entire sample. Among cognitively stable participants, WMH was directly and indirectly, through GFAP, associated with p-tau217 concentration, and in those with MCI, there was a direct effect of WMH on p-tau217 and NfL concentrations. There were no associations of WMH with biomarker concentrations among those diagnosed with dementia. Conclusions and Relevance: The findings suggest that among individuals with DS, CVD promotes neurodegeneration by increasing astrocytosis and tau pathophysiology in the presymptomatic phases of AD. This work joins an emerging literature that implicates CVD and its interface with neuroinflammation as a core pathological feature of AD in adults with DS.

Bilingual neurocognitive resiliency, vulnerability, and Alzheimer's disease biomarker correlates in Latino older adults enrolled in the Health and Aging Brain Study - Health Disparities (HABS-HD).

INTRODUCTION: The effects of bilingualism on neuropsychological test performance in bilinguals with and without cognitive impairment are not well-understood and are relatively limited by small sample sizes of Latinos. METHODS: Using analysis of covariance (ANCOVA), we explored patterns of cognitive performance and impairment across a large sample of community-dwelling bilingual and monolingual Latino older adults with (n = 180) and without (n = 643) mild cognitive impairment (MCI) enrolled in HABS-HD. RESULTS: Bilinguals demonstrated cognitive resiliency in the form of significantly better performance on the Trail Making Test and Digit Symbol Substitution Test, observed across the cognitively unimpaired and MCI groups. In contrast, bilinguals demonstrated cognitive vulnerability in the form of significantly poorer performance and higher impairment rates on phonemic fluency in the MCI phase, only. Follow-up analyses revealed less balanced bilinguals demonstrated poorer performance and higher impairment rates on this measure, supported by lower levels of plasma Aß 42/40. DISCUSSION: Patterns of cognitive performance and impairment differ as a function of bilingualism. Bilingualism must be considered when evaluating cognitive and biomarker outcomes in Latino older adults. Highlights: Latino bilinguals perform better on measures of processing speed and coding.Latino bilinguals with MCI demonstrate cognitive vulnerability in verbal fluency.Less balanced bilinguals demonstrate greatest vulnerability anchored by Aß 42/40.

Comorbidities Incorporated to Improve Prediction for Prevalent Mild Cognitive Impairment and Alzheimer's Disease in the HABS-HD Study.

BACKGROUND: Blood biomarkers have the potential to transform Alzheimer's disease (AD) diagnosis and monitoring, yet their integration with common medical comorbidities remains insufficiently explored. OBJECTIVE: This study aims to enhance blood biomarkers' sensitivity, specificity, and predictive performance by incorporating comorbidities. We assess this integration's efficacy in diagnostic classification using machine learning, hypothesizing that it can identify a confident set of predictive features. METHODS: We analyzed data from 1,705 participants in the Health and Aging Brain Study-Health Disparities, including 116 AD patients, 261 with mild cognitive impairment, and 1,328 cognitively normal controls. Blood samples were assayed using electrochemiluminescence and single molecule array technology, alongside comorbidity data gathered through clinical interviews and medical records. We visually explored blood biomarker and comorbidity characteristics, developed a Feature Importance and SVM-based Leave-One-Out Recursive Feature Elimination (FI-SVM-RFE-LOO) method to optimize feature selection, and compared four models: Biomarker Only, Comorbidity Only, Biomarker and Comorbidity, and Feature-Selected Biomarker and Comorbidity. RESULTS: The combination model incorporating 17 blood biomarkers and 12 comorbidity variables outperformed single-modal models, with NPV12 at 92.78%, AUC at 67.59%, and Sensitivity at 65.70%. Feature selection led to 22 chosen features, resulting in the highest performance, with NPV12 at 93.76%, AUC at 69.22%, and Sensitivity at 70.69%. Additionally, interpretative machine learning highlighted factors contributing to improved prediction performance. CONCLUSIONS: In conclusion, combining feature-selected biomarkers and comorbidities enhances prediction performance, while feature selection optimizes their integration. These findings hold promise for understanding AD pathophysiology and advancing preventive treatments.

Precision Medicine for Preventing Alzheimer's Disease: Analysis of the ADAPT Study.

BACKGROUND: The Alzheimer's Disease Anti-inflammatory Prevention Trial (ADAPT) was the first-ever large-scale anti-inflammatory prevention trial targeting Alzheimer's disease. OBJECTIVE: The overall goal of this study was to evaluate predictive blood biomarker profiles that identified individuals most likely to be responders on NSAID treatment or placebo at 12 and 24 months. METHODS: Baseline (n = 193) and 12-month (n = 562) plasma samples were assayed. The predictive biomarker profile was generated using SVM analyses with response on treatment (yes/no) as the outcome variable. RESULTS: Baseline (AUC = 0.99) and 12-month (AUC = 0.99) predictive biomarker profiles were highly accurate in predicting response on Celecoxib arm at 12 and 24 months. The baseline (AUC = 0.95) and 12-month (AUC = 0.9) predictive biomarker profile predicting response on Naproxen were also highly accurate at 12 and 24 months. The baseline (AUC = 0.93) and 12-month (AUC = 0.99) predictive biomarker profile was also highly accurate in predicting response on placebo. As with our prior work, the profiles varied by treatment arm. CONCLUSIONS: The current results provide additional support for a precision medicine model for treating and preventing Alzheimer's disease.

Evaluation of Neighborhood-Level Disadvantage and Cognition in Mexican American and Non-Hispanic White Adults 50 Years and Older in the US.

Importance: Understanding how socioeconomic factors are associated with cognitive aging is important for addressing health disparities in Alzheimer disease. Objective: To examine the association of neighborhood disadvantage with cognition among a multiethnic cohort of older adults. Design, Setting, and Participants: In this cross-sectional study, data were collected between September 1, 2017, and May 31, 2022. Participants were from the Health and Aging Brain Study-Health Disparities, which is a community-based single-center study in the Dallas/Fort Worth area of Texas. A total of 1614 Mexican American and non-Hispanic White adults 50 years and older were included. Exposure: Neighborhood disadvantage for participants' current residence was measured by the validated Area Deprivation Index (ADI); ADI Texas state deciles were converted to quintiles, with quintile 1 representing the least disadvantaged area and quintile 5 the most disadvantaged area. Covariates included age, sex, and educational level. Main Outcomes and Measures: Performance on cognitive tests assessing memory, language, attention, processing speed, and executive functioning; measures included the Spanish-English Verbal Learning Test (SEVLT) Learning and Delayed Recall subscales; Wechsler Memory Scale, third edition (WMS-III) Digit Span Forward, Digit Span Backward, and Logical Memory 1 and 2 subscales; Trail Making Test (TMT) parts A and B; Digit Symbol Substitution Test (DSST); Letter Fluency; and Animal Naming. Raw scores were used for analyses. Associations between neighborhood disadvantage and neuropsychological performance were examined via demographically adjusted linear regression models stratified by ethnic group. Results: Among 1614 older adults (mean [SD] age, 66.3 [8.7] years; 980 women [60.7%]), 853 were Mexican American (mean [SD] age, 63.9 [7.9] years; 566 women [66.4%]), and 761 were non-Hispanic White (mean [SD] age, 69.1 [8.7] years; 414 women [54.4%]). Older Mexican American adults were more likely to reside in the most disadvantaged areas (ADI quintiles 3-5), with 280 individuals (32.8%) living in ADI quintile 5, whereas a large proportion of older non-Hispanic White adults resided in ADI quintile 1 (296 individuals [38.9%]). Mexican American individuals living in more disadvantaged areas had worse performance than those living in ADI quintile 1 on 7 of 11 cognitive tests, including SEVLT Learning (ADI quintile 5: ß = -2.50; 95% CI, -4.46 to -0.54), SEVLT Delayed Recall (eg, ADI quintile 3: ß = -1.11; 95% CI, -1.97 to -0.24), WMS-III Digit Span Forward (eg, ADI quintile 4: ß = -1.14; 95% CI, -1.60 to -0.67), TMT part A (ADI quintile 5: ß = 7.85; 95% CI, 1.28-14.42), TMT part B (eg, ADI quintile 5: ß = 31.5; 95% CI, 12.16-51.35), Letter Fluency (ADI quintile 4: ß = -2.91; 95% CI, -5.39 to -0.43), and DSST (eg, ADI quintile 5: ß = -4.45; 95% CI, -6.77 to -2.14). In contrast, only non-Hispanic White individuals living in ADI quintile 4 had worse performance than those living in ADI quintile 1 on 4 of 11 cognitive tests, including SEVLT Learning (ß = -2.35; 95% CI, -4.40 to -0.30), SEVLT Delayed Recall (ß = -0.95; 95% CI, -1.73 to -0.17), TMT part B (ß = 15.95; 95% CI, 2.47-29.44), and DSST (ß = -3.96; 95% CI, -6.49 to -1.43). Conclusions and Relevance: In this cross-sectional study, aging in a disadvantaged area was associated with worse cognitive functioning, particularly for older Mexican American adults. Future studies examining the implications of exposure to neighborhood disadvantage across the life span will be important for improving cognitive outcomes in diverse populations.

Plasma NT1-tau and Aß42 correlate with age and cognitive function in two large Down syndrome cohorts.

INTRODUCTION: People with Down syndrome (DS) often develop Alzheimer's disease (AD). Here, we asked whether ultrasensitive plasma immunoassays for a tau N-terminal fragment (NT1-tau) and Aß isoforms predict cognitive impairment. METHODS: Plasma NT1-tau, Aß37 , Aß40 , and Aß42 levels were measured in a longitudinal discovery cohort (N = 85 participants, 220 samples) and a cross-sectional validation cohort (N = 239). We developed linear models and predicted values in the validation cohort. RESULTS: Discovery cohort linear mixed models for NT1-tau, Aß42 , and Aß37:42 were significant for age; there was no main effect of time. In cross-sectional models, NT1-tau increased and Aß42 decreased with age. NT1-tau predicted cognitive and functional scores. The discovery cohort linear model for NT1-tau predicted levels in the validation cohort. DISCUSSION: NT1-tau correlates with age and worse cognition in DS. Further validation of NT1-tau and other plasma biomarkers of AD neuropathology in DS cohorts is important for clinical utility.

APOE and immunity: Research highlights.

INTRODUCTION: At the Alzheimer's Association's APOE and Immunity virtual conference, held in October 2021, leading neuroscience experts shared recent research advances on and inspiring insights into the various roles that both the apolipoprotein E gene (APOE) and facets of immunity play in neurodegenerative diseases, including Alzheimer's disease and other dementias. METHODS: The meeting brought together more than 1200 registered attendees from 62 different countries, representing the realms of academia and industry. RESULTS: During the 4-day meeting, presenters illuminated aspects of the cross-talk between APOE and immunity, with a focus on the roles of microglia, triggering receptor expressed on myeloid cells 2 (TREM2), and components of inflammation (e.g., tumor necrosis factor α [TNFα]). DISCUSSION: This manuscript emphasizes the importance of diversity in current and future research and presents an integrated view of innate immune functions in Alzheimer's disease as well as related promising directions in drug development.

Plasma NT1-tau and Aß 42 correlate with age and cognitive function in two large Down syndrome cohorts.

Introduction: People with Down syndrome (DS) often develop Alzheimer disease (AD). Here we asked whether ultrasensitive plasma immunoassays for a tau N-terminal fragment (NT1-tau) and Aß isoforms predict cognitive impairment. Methods: Plasma NT1-tau, Aß 37 , Aß 40 , and Aß 42 levels were measured in a longitudinal discovery cohort (N = 85 participants, 220 samples) and a cross-sectional validation cohort (N = 239). We developed linear models and predicted values in the validation cohort. Results: Linear mixed models for NT1-tau, Aß 42, and Aß 37:42 were significant for age, there was no main effect of time in the discovery cohort. In cross-sectional models, NT1-tau and Aß 42 increased with age. NT1-tau predicted DLD scores. The discovery cohort linear model for NT1-tau predicted NT1-tau levels in the validation cohort. Discussion: NT1-tau correlates with age and worse cognition in DS. Further validation of NT1-tau and other plasma biomarkers of AD neuropathology in DS cohorts is important for clinical utility.

Efficacy of fluralaner chewable tablets (Bravecto®) against Asian longhorned tick (Haemaphysalis longicornis) infestations of dogs.

BACKGROUND: The parthenogenic reproductive ability of Haemaphysalis longicornis, facilitating quick life cycle completion and rapid geographic spread and its pathogen vector potential make infestations a risk to human and canine health. Two 90-day studies were initiated to evaluate the efficacy of a single fluralaner administration for the treatment and prevention of H. longicornis infestations on dogs. METHODS: Dogs were randomly assigned (10 dogs/group) to either an untreated control group or a group treated once (Day 0) with 13.64% w/w fluralaner chewable tablets (Bravecto®) at the minimum label dose rate of 25 mg/kg. Each dog was infested with approximately 50 H. longicornis ticks on Days -9 or -6 and on Days -2, 28, 58 and 88. A different US tick isolate was used in each study. Tick counts were completed on Days -7 or -4, 2, 30, 60 and 90. The primary efficacy criterion was a 90% reduction in arithmetic mean tick counts between the treated and control groups. For between-group comparisons at any assessment, at least six control dogs were required to retain at least 25% of the infestation dose (13 live ticks). RESULTS: Pre-study infestations demonstrated susceptibility of all study dogs to challenge with H. longicornis. At each subsequent assessment in both studies, at least seven untreated control dogs retained ≥ 25% of the challenge, demonstrating adequate infestations for each efficacy calculation. On Days 2, 30, 60 and 90 the mean live tick infestation rate (number of ticks recovered from each dog/infesting challenge of each dog) of untreated control dogs ranged from 27.8 to 60.8%. No live ticks, free or attached, were found on any fluralaner-treated dog in either study. Between-group differences were statistically significant (P ≤ 0.0002) at each assessment. CONCLUSION: At the minimum recommended label dose rate of 25 mg/kg, fluralaner chewable tablets were 100% effective in eliminating H. longicornis ticks from dogs infested at the time of treatment. Complete efficacy against both US isolates of this tick was maintained through 90 days following a single treatment. Therefore, fluralaner is a treatment of choice for protecting dogs against this invasive tick species.

Efficacy of two topical fluralaner formulations (Bravecto®; Bravecto® Plus) against Asian longhorned tick (Haemaphysalis longicornis) infestations of cats.

BACKGROUND: The invasive tick species, Haemaphysalis longicornis, is becoming established in the USA, presenting a growing threat to dogs and cats. Two 90-day studies were initiated, the same protocol in each, to confirm the efficacy of a single application of two fluralaner formulations against H. longicornis infestations of cats. METHODS: Cats were randomized among three groups in a 1:1:1 ratio (10 cats/group). Group 1 cats were untreated controls; Group 2 cats were treated with a topical fluralaner formulation (Bravecto®); Group 3 cats received a topical formulation containing fluralaner and moxidectin (Bravecto® Plus). Treatments were administered once (Day 0) at the label dose rates. Each cat was infested with 50 H. longicornis ticks on Day 7 for study qualification and also infested with 50 ticks on Days 2, 28, 58 and 88. Tick counts were completed on Days 5, 2, 30, 60 and 90. The primary objective was based on percentage reductions in arithmetic mean tick counts. RESULTS: Pre-study infestations showed all study cats were susceptible to tick challenge. Except for Day 2 in one study, at least six control cats retained ≥ 25% of each challenge, demonstrating an adequate infestation for efficacy assessments. Across studies on Days 2, 30, 60 and 90, the mean live tick infestation rate (number of ticks recovered from each cat/infesting challenge to each cat) of Group 1 cats ranged from 25.0 to 69.6%. Efficacy of each formulation, based on live tick counts, was 100% on Day 2 and > 95 to 100% at each subsequent assessment. Between-group differences were statistically significant (P < 0.0001) for each treatment versus control comparison. CONCLUSION: At the label dose rate, both topical fluralaner formulations were 100% effective in eliminating H. longicornis ticks from cats infested at the time of treatment. Efficacy of > 95 to 100% was then maintained through 90 days following a single application. Fluralaner is therefore a treatment of choice for protecting cats against this invasive tick species.

Medical comorbidities and ethnicity impact plasma Alzheimer's disease biomarkers: Important considerations for clinical trials and practice.

INTRODUCTION: Despite the clinical implementation, there remain significant gaps in our knowledge regarding the impact of race/ethnicity or common medical comorbidity on plasma Alzheimer's disease (AD) biomarkers. METHODS: Plasma biomarkers of amyloid beta (Aß)40, Aß42 , total tau, and neurofilament light chain (NfL) were measured across cognitively normal Mexican Americans (n = 445) and non-Hispanic Whites (n = 520). RESULTS: Dyslipidemia was associated with elevated Aß40 (P = .01) and Aß42 (P = .001) while hypertension was associated with elevated Aß40 (P = .003), Aß42 (P < .001), and total tau (P = .002) levels. Diabetes was associated with higher Aß40 (P < .001), Aß42 (P < .001), total tau (P < .001), and NfL (P < .001) levels. Chronic kidney disease (CKD) was associated with elevations in Aß40 (P < .001), Aß42 (P < .001), total tau (P < .001), and NfL (P < .001) levels. Mexican Americans had significantly lower Aß40 (P < .001) and higher total tau (P = .005) levels. DISCUSSION: Plasma AD biomarkers vary significantly in association with common medical comorbidities as well as ethnicity. These findings are important for those using these biomarkers in clinical practice and clinical trials.

Plasma Biomarkers of Alzheimer's Disease Are Associated with Physical Functioning Outcomes Among Cognitively Normal Adults in the Multiethnic HABS-HD Cohort.

In this study, we examined the link between plasma Alzheimer's disease (AD) biomarkers and physical functioning outcomes within a community-dwelling, multiethnic cohort. Data from 1 328 cognitively unimpaired participants (n = 659 Mexican American and n = 669 non-Hispanic White) from the ongoing Health & Aging Brain Study-Health Disparities (HABS-HD) cohort were examined. Plasma AD biomarkers (amyloid beta [Aß]40, Aß42, total tau [t-tau], and neurofilament light chain [NfL]) were assayed using the ultra-sensitive Simoa platform. Physical functioning measures were the Timed Up and Go (TUG) and the Short Physical Performance Battery (SPPB). Cross-sectional linear regression analyses revealed that plasma Aß 40 (p < .001), Aß 42 (p = .003), and NfL (p < .001) were each significantly associated with TUG time in seconds. Plasma Aß 40 (p < .001), Aß 42 (p < .001), t-tau (p = .002), and NfL (p < .001) were each significantly associated with SPPB Total Score. Additional analyses demonstrate that the link between plasma AD biomarkers and physical functioning outcomes were strongest among Mexican Americans. Plasma AD biomarkers are receiving a great deal of attention in the literature and are now available clinically including use in clinical trials. The examination of AD biomarkers and physical functioning may allow for the development of risk profiles, which could stratify a person's risk for neurodegenerative diseases, such as AD, based on plasma AD biomarkers, physical functioning, ethnicity, or a combination of these measures prior to the onset of cognitive impairment.

Predicting overall survival in diffuse glioma from the presurgical connectome.

Diffuse gliomas are incurable brain tumors, yet there is significant heterogeneity in patient survival. Advanced computational techniques such as radiomics show potential for presurgical prediction of survival and other outcomes from neuroimaging. However, these techniques ignore non-lesioned brain features that could be essential for improving prediction accuracy. Gray matter covariance network (connectome) features were retrospectively identified from the T1-weighted MRIs of 305 adult patients diagnosed with diffuse glioma. These features were entered into a Cox proportional hazards model to predict overall survival with 10-folds cross-validation. The mean time-dependent area under the curve (AUC) of the connectome model was compared with the mean AUCs of clinical and radiomic models using a pairwise t-test with Bonferroni correction. One clinical model included only features that are known presurgery (clinical) and another included an advantaged set of features that are not typically known presurgery (clinical +). The median survival time for all patients was 134.2 months. The connectome model (AUC 0.88 ± 0.01) demonstrated superior performance (P < 0.001, corrected) compared to the clinical (AUC 0.61 ± 0.02), clinical + (AUC 0.79 ± 0.01) and radiomic models (AUC 0.75 ± 0.02). These findings indicate that the connectome is a feasible and reliable early biomarker for predicting survival in patients with diffuse glioma. Connectome and other whole-brain models could be valuable tools for precision medicine by informing patient risk stratification and treatment decision-making.

Hyperparameter Tuning with High Performance Computing Machine Learning for Imbalanced Alzheimer's Disease Data.

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.

Characterization of Mild Cognitive Impairment and Dementia among Community-Dwelling Mexican Americans and Non-Hispanic Whites.

BACKGROUND: Despite tremendous advancements in the field, our understanding of mild cognitive impairment (MCI) and Alzheimer's disease (AD) among Mexican Americans remains limited. OBJECTIVE: The aim of this study was to characterize MCI and dementia among Mexican Americans and non-Hispanic whites. METHODS: Baseline data were analyzed from n = 1,705 (n = 890 Mexican American; n = 815 non-Hispanic white) participants enrolled in the Health and Aging Brain Study-Health Disparities (HABS-HD). RESULTS: Among Mexican Americans, age (OR = 1.07), depression (OR = 1.09), and MRI-based neurodegeneration (OR = 0.01) were associated with dementia, but none of these factors were associated with MCI. Among non-Hispanic whites, male gender (OR = 0.33), neighborhood deprivation (OR = 1.34), depression (OR = 1.09), and MRI-based neurodegeneration (OR = 0.03) were associated with MCI, while depression (OR = 1.09) and APOEɛ4 genotype (OR = 4.38) were associated with dementia. CONCLUSION: Findings from this study revealed that the demographic, clinical, sociocultural and biomarker characteristics of MCI and dementia are different among Mexican Americans as compared to non-Hispanic whites.

Combination of Serum and Plasma Biomarkers Could Improve Prediction Performance for Alzheimer's Disease.

Alzheimer's disease (AD) can be predicted either by serum or plasma biomarkers, and a combination may increase predictive power, but due to the high complexity of machine learning, it may also incur overfitting problems. In this paper, we investigated whether combining serum and plasma biomarkers with feature selection could improve prediction performance for AD. 150 D patients and 150 normal controls (NCs) were enrolled for a serum test, and 100 patients and 100 NCs were enrolled for the plasma test. Among these, 79 ADs and 65 NCs had serum and plasma samples in common. A 10 times repeated 5-fold cross-validation model and a feature selection method were used to overcome the overfitting problem when serum and plasma biomarkers were combined. First, we tested to see if simply adding serum and plasma biomarkers improved prediction performance but also caused overfitting. Then we employed a feature selection algorithm we developed to overcome the overfitting problem. Lastly, we tested the prediction performance in a 10 times repeated 5-fold cross validation model for training and testing sets. We found that the combined biomarkers improved AD prediction but also caused overfitting. A further feature selection based on the combination of serum and plasma biomarkers solved the problem and produced an even higher prediction performance than either serum or plasma biomarkers on their own. The combined feature-selected serum-plasma biomarkers may have critical implications for understanding the pathophysiology of AD and for developing preventative treatments.

Characterizing Plasma Biomarkers of Alzheimer's in a Diverse Community-Based Cohort: A Cross-Sectional Study of the HAB-HD Cohort.

Background: Due to their low cost, less invasive nature, and ready availability, plasma biomarkers of Alzheimer's disease have been proposed as one-time screening tools for clinical trials and research. The impact of ethnoracial factors on these biomarkers has received little attention. The current cross-sectional study investigated the levels of Aß40, Aß42, total tau (t tau), and neurofilament light (NfL) across diagnoses for each of the three major ethnoracial groups in the United States in a community-based cohort of older adults. Methods: A total of 1,862 participants (852 Mexican Americans (MAs); 775 non-Hispanic Whites (NHWs), and 235 African Americans (AAs)) drawn from The Health & Aging Brain Study-Health Disparities (HABS-HD) study were included. Diagnoses were assigned using an algorithm (decision tree) verified by consensus review. Plasma samples were assayed using Simoa technology. Levels of each biomarker were compared for the three ethnoracial groups across cognitive diagnoses using ANOVA covarying sex and age. Results: Significant differences were found across the groups at each level of cognitive impairment. Cognitively unimpaired (CU) AA had significantly lower levels of each of the biomarkers than cognitively unimpaired MA or NHW and NHW had higher levels of Aß40, and NfL than the other two groups. MA had higher t tau than AA or NHW. Mild cognitive impairment (MCI) group NHW had the highest levels on all the biomarkers and AA had the lowest. NHW and MA have higher levels of Aß40, Aß42, and t tau there was no difference between the groups for Aß42. NHW had significantly higher levels of Aß40, t tau, and NfL than AA. AA had a higher Aß42/Aß40 ratio than either NHW or MA for CU MCI. Conclusions: The use of plasma biomarkers of cognitive decline is promising given their advantages over other biomarkers such as CSF and imaging but as the current research shows, ethnoracial differences must be considered to enhance accuracy and utility. Developing ethnoracial-specific cut points and establishing normative ranges by assay platform for each of the biomarkers are needed. Longitudinal research to assess changes in biomarkers during a cognitive decline is ongoing.