Normal Cognition Research Articles

Co‐morbidity of sensory loss (hearing, vision, olfaction) in persons with or at risk for dementia: Implications for brain structure and function

AbstractBackgroundHearing, vision, and olfactory loss have each independently been associated with cognitive decline and increased risk for dementia. Research has examined their individual relationships with brain structure and function, but these predictors have not been examined together in older adults who vary along a spectrum of risk for dementia. We examined groups who range from relatively low risk (those with normal cognition and no cognitive complaints (NC)), to those with higher risk, namely individuals with subjective reports of cognitive decline (SCD) but normal cognition and those with mild cognitive impairment (MCI).MethodWe used data from the Comprehensive Assessment of Neurodegeneration and Dementia (COMPASS‐ND) study. Hearing loss was assessed with a pure‐tone screening protocol and a speech‐in‐noise test, vision was assessed with the MNRead for visual acuity and the MARS test for contrast sensitivity, and olfaction was assessed with the Brief Smell Identification Test (BSIT). We examined the frequency of sensory deficits in 60 NC (mean age = 69 years; mean education = 16 years), 55 participants with SCD (age = 70; education = 17), 102 with MCI (age = 72; education = 16), and 47 with Alzheimer’s disease (AD, age = 75; education = 15). Participants were matched on age and education, except for AD participants who were older.ResultsThe frequency of having at least one impairment in at least one sensory domain increased across groups (55% NC, 62% SCD, 75% MCI, 96% AD). A deficit in olfaction was the most common single deficit. Sensory deficits became increasing more co‐morbid across the risk spectrum. Hearing loss and olfaction loss were each associated with lower hippocampal volume and hearing loss was associated with alterations in functional brain connectivity in the default mode network.ConclusionThese analyses indicate that sensory loss is highly frequent and co‐morbid in persons with or at risk for dementia. These findings have implications for cognition, brain function, functional activities, and care delivery for persons with or at risk for dementia.

Examining the measurement equivalence of Alzheimer’s Disease Centre Neuropsychological Test Battery (Version 3) across two distinct cultures

AbstractBackgroundNeuropsychological assessments are important in clinical diagnostic process for dementia as it help early detection of cognitive difficulties. Prior to using neuropsychological test batteries across different populations, it is critical to standardize the battery to ensure consistency in the assessment process, as well as valid and comparable results across different settings. The Alzheimer’s Disease Centre Neuropsychological Test Battery (Version 3) (ADCNTB3) is a widely used battery that had shown sensitivity in detecting cognitive decline in MCI and early dementia, with minimal test repetition effects (Boccardi et. al., 2022). While norms are available in America, harmonization efforts are currently in place to adapt ADCNTB3 for European populations.MethodThis study examined whether ADCNTB3 could be adapted to Singapore (multi‐ethic Asian population), a culture distinctly different from America, and whether ADCNTB3 demonstrates measurement equivalence when adapted to the Singapore population. Following the qualitative process of cultural adaptation, the ADCNTB3 was administered to 266 community‐dwelling older persons with normal cognition in Singapore. Scores from these participants were then compared with 1064 random samples with normal cognition from National Alzheimer’s Coordinating Centre (NACC) to evaluate whether the ADCNTB3 has equivalent scores across the two distinct populations. Score‐difference between the 2 populations was computed using multiple linear regression (adjusted for age, education, gender, family history of dementia, cardiovascular conditions, psychiatric and neurological history, smoking, untreated hearing loss, obesity, MoCA total score, CDR sum of boxes, NPI‐Q total score) and equivalence scores are considered to be present if the 90% confidence interval of the adjusted score‐difference fall within the equivalence margin. A priori, we defined the equivalence margin as 0.5 Standard Deviation of tests from ADCNTB3, as this often defines the minimally important clinical significance (Sloan et al., 2005).ResultOther than working and delayed memory tasks, all other tests from the ADCNTB3 are not directly equivalent. Notably, Asians performed better in visuospatial task, but consistently performed poorer in most other tasks.ConclusionOther than working and delayed memory tasks, all other tests from the ADCNTB3 are not directly equivalent. Notably, Asians performed better in visuospatial task, but consistently performed poorer in most other tasks.

FC44: Prevalence of mental health outcomes and the impact of sex and mood or cognitive comorbidity in older adults during the COVID-19 pandemic

Introduction:Older adults are at greater risk for developing severe illness from SARS-COV2 infection and may be more vulnerable to negative mental health outcomes as a result of public health guidelines that increase social isolation. We assessed mental health outcomes in older adults with normal cognition (NC), past history of major depressive disorder (i.e., remitted; rMDD), or mild cognitive impairment (MCI) to determine the prevalence of depression, anxiety, general stress, and post-traumatic stress during the COVID-19 pandemic and the impact of diagnosis and sex.Methods:The sample included 108 older adults (37 males, mean age=72.1 years): 71 older adults with normal cognition (NC) based on normal neuropsychological test performance and no psychiatric history, 21 rMDD participants based on DSM5 criteria, and 16 MCI participants based on NIA-AA criteria. Participants completed self-report measures of depression [Patient Health Questionnaire-9 (PHQ-9)], anxiety [Patient-Reported Outcomes Measurement Information System (PROMIS)], general stress [Perceived Stress Scale (PSS)] and post-traumatic stress [Impact of Events Scale Revised (IES-R)] through video- or teleconferencing. Prevalence rates of clinically significant psychiatric symptoms were expressed as the percentage of participants with total scores that exceed the normal cut-offs. Separate MANOVAs were used to examine the effects of diagnosis and sex. Non-normally distributed data (PHQ-9 and PROMIS total scores) were rank-transformed.Results:Approximately 1/3rd of participants endorsed clinically significant symptoms based on scores exceeding the cut-off for normal: 33.7% on PHQ-9, 31.3% on PROMIS-Anxiety, 35.5% on PSS, 38.3% on IES-R. rMDD participants scored higher on all measures compared to NC participants (p’s < .005) while MCI participants scored higher on the PSS compared to NC (p=.035). Women scored higher on all measures compared to men.Conclusions:These rates of approximately 1/3rd reporting clinically significant symptoms of depression, anxiety, general stress, and post-traumatic stress are higher than those described in population surveys of older adults but are comparable to prevalence rates of psychiatric symptoms in the general adult population. The effects of diagnosis and sex indicate that older adults with previous depression or current MCI, as well as women overall, are particularly vulnerable to developing clinically significant psychiatric

Synaptic Marker NPTX2 Predicts Risk of MCI Clinical Symptom Onset

AbstractBackgroundNeuronal pentraxin 2 (NPTX2) is a synaptic protein involved in the homeostatic regulation of cortical network dynamics and synaptic plasticity. Prior cross‐sectional studies suggest that NPTX2 levels decline across the spectrum of Alzheimer’s disease (AD), and that NPTX2 levels can distinguish individuals with normal cognition vs. mild cognitive impairment (MCI) and AD dementia. Less is known about how levels of NPTX2 among individuals with normal cognition relate to future clinical outcomes. To address this, we examined whether baseline levels of NPTX2 predicted time to onset of clinical symptoms of MCI, both alone and in combination with AD biomarkers, among individuals who were cognitively normal at baseline and have undergone 16.3y of follow‐up, on average.MethodCerebrospinal fluid (CSF) was collected at baseline from 269 cognitively normal BIOCARD Study participants (M baseline age = 57.7y), including 77 (29%) who subsequently progressed to MCI or dementia (Table 1). Baseline NPTX2 levels were measured from three highly correlated peptides using quantitative parallel reaction monitoring mass spectrometry. NPTX2 peptide levels were standardized and averaged to create an NPTX2 composite score. Baseline levels of Aβ42/Aβ40, p‐tau181, and t‐tau were measured using Lumipulse automated electrochemiluminescence assays.ResultIn Cox regression models, lower NPTX2 levels were significantly associated with earlier time to MCI clinical symptom onset (HR = 0.76, 95% CI (0.601, 0.962), p = 0.02; Figure 1a). After accounting for CSF AD biomarkers (i.e., Aβ42/Aβ40 and tau), the association between NPTX2 levels and MCI symptom onset appeared stronger (Table 2; Figure 1b). Importantly, NPTX2 did not interact with AD biomarker levels (all p > 0.11) or APOE‐ε4 genetic status (p = 0.71) in relation to MCI symptom onset.ConclusionCSF levels of the synaptic marker NPTX2 predict time to progress to MCI among individuals with normal cognition, suggesting that NPTX2 may be an early prognostic biomarker of synaptic dysfunction. Moreover, this association appears to be additive and independent of AD biomarker levels. Future work is underway to determine if NPTX2 provides resilience to cognitive impairment in the presence of AD pathology, and if there is a differential relationship with markers of amyloid vs. tau.

Association of individual and multiple comorbid neuropathologies with cognitive decline in the oldest‐old: Results from The 90+ Study

AbstractBackgroundAge‐related neuropathologies are common in the oldest‐old and associated with cognitive decline. In this study, we examined the associations of individual and multiple comorbid neuropathologies with cognitive decline in participants who had normal cognition in at least one in‐person visit during their time in The 90+ Study.MethodOur study included 251 individuals from The 90+ Study‐ a longitudinal study on aging, who had at least 2 in‐person visits, had normal cognition in at least one of them and had autopsy data. We used longitudinal Mini Mental State Examination (MMSE) scores to assess cognitive trajectories. For the first part of our analysis, we examined individual neuropathologies that were evaluated as: Alzheimer’s Disease neuropathologic change (ADNC) (intermediate/high), presence of Cerebral amyloid angiopathy (CAA), Microinfarcts (2+), atherosclerosis, arteriolosclerosis, Lewy body disease (LBD), Hippocampal sclerosis (HS), Limbic‐predominant age‐related TDP‐43 encephalopathy neuropathological change (LATE‐NC), and Age‐related tau astrogliopathy (ARTAG). For the second part of our analysis, we examined multiple comorbid neuropathologies. For this analysis we recoded each neuropathology as: 0 for absent/negligible levels, 0.4 for intermediate, or 1 for severe. These recoded scores were added to create four summary indices: all pathology index (total of all individual neuropathologies), vascular index (CAA, Microinfarcts, atherosclerosis, and arteriolosclerosis), neurodegenerative index (ADNC, LB, HS, and LATE‐NC) and non‐AD neurodegenerative index (LB, HS, and LATE‐NC). We used linear mixed effects models to examine the associations of individual neuropathologies and the pathology indices with cognitive decline.ResultAt baseline, participants had a mean age of 93 years and a mean MMSE score of 27.6. Mean follow up time was 5.1 years (Table1). Presence of LBD (ß = ‐0.20; p = 0.07), and HS (ß = ‐0.24; p = 0.06) trends toward to an association with faster cognitive decline (Table2). All pathology index (ß = ‐0.10; p = 0.01), neurodegenerative index (ß = ‐0.17; p = 0.003) and non‐AD neurodegenerative index (ß = ‐0.16; p = 0.02) were significantly associated with faster cognitive decline (Table3).ConclusionIn individuals who maintain normal cognition until age 90+ and then start declining, the rate of cognitive decline is related to the burden of neuropathologies, particularly neurodegenerative neuropathologies. Therefore, multiple comorbid neuropathologies need to be targeted for preventing cognitive decline in the oldest‐old.

Are there baseline structural MRI differences between individuals who progress along the Alzheimer’s disease continuum and those who remain cognitively stable? A systematic review

AbstractBackgroundEarly identification of individuals that will progress along the continuum from normal cognition to mild cognitive impairment (MCI) and to Alzheimer’s disease (AD) is imperative for timely prevention and intervention measures. Structural magnetic resonance imaging (MRI) is a non‐invasive technique commonly used to examine neurodegeneration. The current review summarizes the literature on structural brain differences between individuals who experience conversion to MCI from normal cognition or AD from MCI and those who remain stable.MethodMedline and PsycINFO databases were searched for studies that compared baseline structural MRI of individuals who converted from normal cognition to MCI, or from MCI to AD, to those with normal cognition or MCI who remained cognitively stable. Inclusion criteria included studies with baseline structural MRI, cognitive follow‐up to assess conversion/stability, and comparison between converters and non‐converters. Two authors reviewed each abstract and a third review resolved any conflicts. The same process was used for full‐text articles. Information regarding sample size, acquisition and analysis and results were extracted and all studies were reviewed for quality using the AXIS tool.ResultA total of 1670 studies were imported for screening in Covidence and 440 duplicates were removed. Following abstract screening, 293 studies were eligible for full‐text review. Many of the identified studies took a region of interest approach focused on hippocampal volume and found significant differences between converters and non‐converters. Many studies used multi‐site, multimodal databases (e.g., Alzheimer’s Disease Neuroimaging Initiative) to retrospectively examine converters and non‐converters. Many more studies looked at individuals who converted from MCI to AD than normal cognition to MCI. Further analyses of the full text review and findings will be presented.ConclusionCharacterization of baseline differences in brain structure between those who progress on the AD continuum and those who remain stable is crucial for early diagnosis and treatment. The current review indicates that such differences are evident on structural MRI. The medial temporal lobes have been identified as a region of interest. However, it will be important for future research to use a whole brain approach and for further examination of individuals who convert from normal cognition to MCI.

P172: A preliminary study for potential protective role of anti-oxidative stress markers for cognitive impairment: glutathione and glutathione reductase.

Objective:We aimed to study the relationship between glutathione (GSH), a key molecule of the anti-oxidant defense system in the blood, and glutathione reductase (GR), which reduces oxidized GSSG to GSH and maintains redox balance, with the prevalence of Alzheimer’s dementia and cognitive decline.Methods:20 with normal cognition and 20 with Alzheimer's dementia who completed the 3rd f/u clinical evaluation over 6 years were selected by matching age and gender. Plasma glutathione (GSH) and glutathione reductase (GR) concentrations were independent variables. Clinical diagnosis and neurocognitive test scores were used as dependent variables indicating cognitive status.Results:The higher the GR, the greater the possibility of normal cognition rather than Alzheimer's dementia. Also, the higher the GR, the higher the neurocognitive score. However, this association was not significant in GSH in any way. After 6 years, the conversion rate from normal cognition to cognitive impairment was significantly higher in the lower 50th percentile of the GR group than in the upper 50th percentile.Conclusion:According to the result of this study, the higher the GR, the lower the prevalence of Alzheimer's dementia and incidence of cognitive impairment, and the higher the cognitive outcome. Therefore, GR can be regarded as a protective biomarker for Alzheimer’s dementia and cognitive decline.

I‐CONECT Intervention Effects on Weekly Time Spent Outside of Home and Social Contacts among Socially Isolated Older Adults with Normal Cognition and Mild Cognitive Impairment

AbstractBackgroundSocial isolation has been linked to an increased risk of cognitive decline and incident dementia. Socially isolated individuals tend to lack access to cognitively stimulating activities. This may adversely impact their cognitive health. The Internet‐Based Conversational Engagement Clinical Trial (I‐CONECT) was designed to deliver online conversation sessions to socially isolated older adults to slow cognitive decline and ultimately delay the onset of dementia. The current study examined the intervention effects on participants’ weekly time spent out‐of‐home and their social interaction with family and friends as clinically meaningful functional outcomes.MethodIntervention group engaged in 30‐minutes video chat 4 times/week and 10‐minute weekly phone calls, while control group received only 10‐minute weekly phone calls over 48 weeks. In each call, we asked participants to report time spent out‐of‐home and whether they contacted family or friends during the week (yes/no). Linear mixed‐effect models were run for time spent out‐of‐home, and mixed‐effect models with a logistic link were performed for contact with family and friends. The intervention effect was modeled by including an interaction term of time (in weeks) and group assignments. We ran analyses separately for participants with normal cognition (NC) and mild cognitive impairments (MCI). All models controlled for age, sex, race, education, and a pre vs. post COVID‐19 pandemic indicator.ResultAs weeks passed, participants (N = 155, mean age 81, 72.3% female, 17.53% African American, 51.95% MCI) in both groups reported spending more time out of home, indicated by a significant time main effect (B = , p<.001). Among NC participants, the intervention group had a steeper increase in their reported time spent out‐of‐home compared to the control group (B = , p = .016). Among MCI participants, the intervention group had an increased likelihood of contacting friends over time (OR = 1.013, p = .001), while the intervention effect on contact with family was not significant for either the NC or MCI group.ConclusionPreviously we reported the I‐CONECT intervention positively affected cognitive functions. The current results suggest the translational effects of I‐CONECT to functional outcomes beyond cognitive test scores, activating weekly activities among socially isolated older adults. The intervention provides a remote social interaction platform that brings cognitive stimulation to at‐risk populations.

A modified Mediterranean‐ketogenic diet favorably modulates the metabolic profiles of mild cognitively impaired individuals

AbstractBackgroundImpaired glucose utilization is central to the pathogenesis of Alzheimer’s disease. In a hypometabolic state, the brain depends on alternative sources of energy, including amino acids, fatty acids, and ketone bodies (KBs). As KBs can cross the blood‐brain barrier, it is considered that a ketogenic diet could augment the high energetic demand of the central nervous system (CNS). A ketogenic diet is already a part of clinical care in epilepsy a CNS disorder. However, the holistic benefits of this diet for Alzheimer’s remain to be determined.MethodsA randomized crossover pilot trial was conducted where participants consumed either the Modified Mediterranean‐Ketogenic Diet (MMKD) or the American Heart Association Diet (AHAD) for 6 weeks, followed by a 6‐week washout, after which the respective other diet was consumed for 6 weeks. The cohort included both mild cognitively impaired (MCI) and cognitively normal (CN) individuals. Serum and cerebrospinal fluid (CSF) samples of these individuals were profiled using two metabolic platforms: MS‐based and NMR‐based metabolomics. Diet‐related changes in levels of metabolites were determined using linear‐mixed effect models with age and sex as confounders.ResultsWe identified post‐diet changes in the two diet groups (AHAD and MMKD), two cognitive groups (MCI and CN), and two fluids (serum and CSF). AHAD was associated with a decrease in serum levels of only a single metabolite, glutamine. MMKD led to widespread lipidomic and metabolic changes in the serum of mild cognitively impaired individuals. Remarkably, MMKD was associated with the reversal of AD‐related metabolic alterations identified in large cohorts like ADNI, AIBL, ROS/MAP, and signature for incident dementia in large epidemiological studies. The post‐MMKD changes include correcting metabolic dysregulation in branched‐chain amino acids, triglycerides, and plasmalogens a class of lipids implicated in membrane structure and function. Peripheral‐central metabolic correlations will be shared.ConclusionOur findings highlight the potential of dietary interventions, specifically the MMKD, in improving the cognitive and metabolic state of individuals with MCI. Further investigation is needed to capture the entirety of MMKD‐related modulation in metabolic profiles and the resulting benefit for AD.

Cell‐Based Genetic Subtypes Predict Neuroimaging Biomarkers and Clinical Conversion to Alzheimer’s Disease

AbstractBackgroundThe complex etiology and slow progression of AD have complicated efforts to establish robust patient subtypes resulting in high attrition rates among late‐stage clinical trials. We developed a novel procedure to stratify at risk individuals to develop AD using network‐based polygenic risk scores from brain cell‐type specific transcriptome signatures.MethodWe previously computed cell‐based polygenic risk scores (cbPRSs) in the Alzheimer’s Disease Neuroimaging Initiative (Sahelijo et al. 2022). Recently, we computed cbPRSs in 8,481 Framingham Heart Study (FHS) participants using the same genetic markers used in ADNI. We defined low‐ and high‐risk genetic subtypes using the first and the fourth quartiles of the cbPRS distributions, respectively. We conducted association tests with domain specific cognitive tests and neuroimaging biomarkers using subtype risk status as a binary outcome in cognitively normal (CN), mild cognitive impairment (MCI), AD, and all subjects from ADNI. We compared conversion rates from CN and MCI to AD between subtypes using a Cox proportional hazard regression in ADNI and FHS. Using brain‐wide gray matter volume maps with voxel‐based morphometry we compared brain atrophy patterns between subtypes in ADNI corrected for multiple comparisons using Standard FDR at q = 0.05.ResultGenetic subtypes for astrocytes (Ast‐M2) and oligodendrocytes (Oli‐M45) showed significant association with memory and executive function impairment, decreased hippocampal volume, increased amyloid deposition, and decreased glucose metabolism using all subjects from ADNI (P<1.39E‐3). These associations for memory impairment, global amyloid deposition, and decreased glucose metabolism remained significant only in MCI subjects (best P<7.23E‐09). Conversion rates for Ast‐M2 and Oli‐M45 genetic subtypes were significantly different in both ADNI (P<8.67E‐5) and FHS (P<3.90E‐9). These differential conversion rates for Ast‐M2 and Oli‐M45 in FHS were not fully explained by presence of APOE ɛ4 allele (P<0.05 in ɛ4 non‐carriers). Voxel‐wise whole brain analysis identified significant atrophy in bilateral hippocampi, entorhinal cortex, and amygdala for Ast‐M2 (P = 0.0007) and Oli‐M45 (P = 0.001).ConclusionWe demonstrate that polygenic risk scores informed by cell type specific brain transcriptomic networks can stratify at risk subjects with AD related changes in imaging biomarkers and higher chance of progression to AD. These findings provide insight into early detection of cell‐type specific at‐risk individuals for AD.

AD and ADRD Differences in Luminex and Lumipulse ATN Categories

AbstractBackgroundCerebrospinal fluid (CSF) biomarkers for amyloid (A), tau (T), and neurodegeneration (N) can be utilized to define pre‐mortem pathological status (ATN research framework). However, potential differences across methods present challenges when comparing studies. Our objective was to determine the ATN status in an AD/ADRD cohort using two different methods for measuring CSF AD‐related biomarkers; Fujirebio Lumipulse G1200 and the Luminex 200. The hypothesis was that the proportion of individuals in a given ATN category would differ between these two methods.MethodParticipants were recruited from the Cleveland Alzheimer’s Disease Research Center (CADRC), and Cleveland Clinic Lou Ruvo Center for Brain Health (CBH) and included cognitively normal (CN), mild cognitive impairment (MCI), AD, dementia with Lewy Bodies (DLB), and PD patient groups. Aβ42/40 (A), p‐Tau181 (T) and t‐Tau (N) were quantitated using both Lumipulse and Luminex systems. Receiver Operating Characteristic (ROC) curve was utilized to identify optimal cutpoints for the three biomarkers (A, T, and N) based on the CN and AD cohorts. Youden index was calculated to identify the strength of these biomarkers.ResultCorrelation analyses between the two platforms, indicated that there was a significant correlation for Aβ42/40 (A), p‐Tau181 (T) and t‐Tau (N) among most patient groups, especially for Aβ42/40. The cut‐points were 0.0562 (A), 65 pg/mL (T) and 345 pg/mL (N) for Lumipulse, and 0.1600 (A), 106.77 pg/mL (T) and 399.41 pg/mL (N) for Luminex. Out of the overall cohort of 283 individuals, over half had identical ATN status for both Lumipulse and Luminex.ConclusionThere was a strong correlation for CSF AD‐related biomarkers between the two platforms. However, there was a lack of complete ATN category overlap between platforms suggesting that across study comparisons using ATN categories derived from different platforms (e.g. Luminex and Lumipulse) should be approached with caution. In light of different techniques that are available, it is important to understand the difference between them to inform standardization and data harmonization efforts. A definite and effective method to quantitate AD related biomarkers holds the promise of better differential diagnosis in individuals with AD and ADRD.

Bio‐Hermes: A study to assess the relationship of blood and digital biomarkers with Aβ PET scans in older persons with normal cognition, MCI or mild AD

AbstractBackgroundRecently developed blood and digital biomarkers may enable the identification of persons with amyloid deposits in the brain. While these biomarkers are promising, most haven’t been evaluated in populations like those screened for clinical treatment trials or in populations with racial and ethnic diversity similar to the US population. The Global Alzheimer’s Platform Foundation® (GAP) completed its biomarker study (Bio‐Hermes) comparing results of blood and digital biomarker tests with brain amyloid PET scans and traditional cognitive tests. The trial enrolled 1,002 participants in three cohorts: Cognitively Normal (CN), Mild Cognitive Impairment (MCI), and Mild AD.MethodWithin 18 months, 17 sites enrolled 1,002 participants (60‐85 y/o) in clinically defined cohorts of: CN (N = 417), MCI (N = 312), and Mild AD (N = 273). Traditional and digital cognitive testing, amyloid PET imaging, biospecimen collection, and speech analytics were performed. A subset of participants received a retinal exam. Blood biomarkers included, Aß 42/40, p‐Tau181/217/231, NfL, GFAP, full genome sequencing, including APOE status, and proteomics. Digital biomarkers included memory recall, executive functions, and drawing‐based and speech elicitation tasks. Blood biomarkers’ relationship to amyloid PET was measured using Spearman’s rank correlation. A ROC curve analysis will assess the sensitivity and specificity of each biomarker and combination of biomarkers compared with amyloid positivity from PET.ResultStatistical models to be competed in April 2023 for presentation.24% of participants were from underrepresented populations (URP) either African American, Hispanic or other. The CN cohort was 61% female, with 21% amyloid PET Positive; the MCI cohort was 54% female with 34% amyloid PET Positive; the Mild AD cohort was 51% female with 62% amyloid PET Positive. Penalized multiple regression will be used to predict amyloid level (continuous) and amyloid positivity status. Candidate predictors include cohorts, blood‐based biomarkers, cognitive assessments (rating scales and digital), retinal measurements, and speech recognition digital biomarkers.ConclusionBio‐Hermes generated a unique, well‐characterized, diverse sample set that determined the utility of several promising biomarkers. Validation of these biomarkers will expedite AD clinical trial enrollment by quickly identifying appropriate participants while reducing the variability and burden of screen failures from high‐cost brain scan procedures.

Sex‐associated differences in plasma and cerebrospinal fluid biomarkers in Early‐onset Alzheimer’s disease

AbstractBackgroundFemale sex is associated with greater atrophy, amyloid and tau burden in Early‐onset Alzheimer’s Disease (EOAD) in the Longitudinal EOAD Study (LEADS). APOE‐ε4 non‐carrier‐status was found to be a further predictor of EOAD pathology. We expanded the analyses by examining the impact of sex and APOE‐ε4 on plasma and cerebrospinal fluid (CSF) biomarkers of AD. Plasma markers included: neurofilament light chain (NfL), plasma glial fibrillary acidic protein (GFAP), Aβ 42/40, and pTau231. CSF markers included: Aβ 42/40, neurogranin, tTau, pTau181, synaptosomal‐associated protein‐25 (SNAP25), YKL‐40, and visinin‐like protein‐1 (VILIP1).MethodWe included 201 amyloid‐positive EOAD, 64 amyloid‐negative Early‐onset cognitively impaired (EOnonAD), and 86 Cognitively Normal (CN) LEADS participants with plasma biomarker data. Of these 100 EOAD, 35 EOnonAD, and 38 CN participants also had CSF data. Participants were stratified by sex and APOE‐ε4 genotype. Demographics (age, education, APOE‐ε4, MMSE) and biomarker differences were compared using ANOVA within each diagnostic group. ANCOVAs were run to control for the effects of age and education on biomarkers.ResultCompared to men, EOAD women showed greater levels of plasma NfL (p = .03), and GFAP (p<.001) while Aβ 42/40, and pTau231 were comparable. In CSF, EOAD women showed higher levels of neurogranin (p = .01), tTau (p = .01), pTau181 (p = .01) and VILIP1 (p = .02). EOnonAD women showed significantly greater plasma GFAP levels (p = .02). In the CN cohort, women had greater CSF SNAP25 levels (p = .03) while men showed a trend for higher plasma pTau231 levels (p = .05). APOE‐ε4 carrier‐status was not associated with differences in levels of plasma or CSF biomarkers in either sex.ConclusionAs highly selective and specific AD fluid biomarkers emerge, understanding sex‐based differences in CSF and plasma is imperative. Female sex is associated with higher levels of neurodegeneration (NfL, tTau, pTau181, neurogranin, VILIP1) in EOAD and astrogliosis (GFAP) in both EOAD and EOnonAD.This data supports our previous findings that greater pathology burden is associated with female sex. This paves the way for further examination of sex and APOE‐ε4 genotype‐based differences in imaging and fluid biomarkers, their associations, and utility in early diagnosis.

Clinical Manifestations.

Mild behavioral impairment (MBI) identifies a high-risk group for Alzheimer's disease and related dementias (ADRD), leveraging the risk associated with later-life emergent and persistent neuropsychiatric symptoms (NPS). MBI may act as a complementary behavioral analog to later-life emergent cognitive symptoms (mild cognitive impairment, MCI). These constructs are not mutually exclusive and can co-occur. We characterized the relationship between MBI, grey matter volume in the hippocampus and entorhinal cortex, and incident cognitive decline in older adults with normal cognition (NC) and MCI. Seven-hundred-forty-two participants were included from the National Alzheimer's Coordinating Center Uniform Dataset. The Normative Morphometry Image Statistics tool was used to generate estimates of grey matter volume for the bilateral hippocampi and entorhinal cortices, normalized by age, sex, total intracranial volume, and image quality. NPS were evaluated using the informant-rated Neuropsychiatric Inventory Questionnaire, from which MBI scores were derived based on a published algorithm. Associations between MBI status and the regions-of-interest were modelled separately in NC and MCI using multivariable linear regressions, adjusting for education and clinical-MRI visit interval time. Associations between MBI status and incident cognitive decline were modelled using Cox proportional hazards regressions. Participant characteristics are summarized in Table 1. As per Figures 1 and 2, NC participants with MBI had lower grey matter volume in the bilateral hippocampi (B = -0.40, 95%CI:[-0.66, -0.15], p = .004), and a greater hazard of incident MCI or dementia, than those without MBI (HR = 3.34, 95%CI:[2.04, 5.48), p<.001). NC MBI+ participants also had lower grey matter volume in the bilateral entorhinal cortices compared to NC MBI- participants, although not statistically significant (B = -0.13, 95%CI:[-0.35, -0.08], p = .22). In MCI, participants with MBI had lower grey matter volume bilaterally in both the hippocampus (B = -0.46, 95%CI:[-0.80, -0.12], p = .01) and entorhinal cortex (B = -0.42, 95%CI:[-0.76, -0.12], p = .009), and had greater progression-rates to dementia (HR = 4.13, 95%CI:[2.45, 6.94), p<.001), than those without MBI. Our findings suggest that evaluating dementia-free older adults for MBI status in conjunction with cognitive status identifies a group further along the ADRD continuum than those without MBI, who have greater baseline atrophy in key ADRD regions and greater progression rates to later cognitive stages.

Which Convolutional Neural Network Architecture is the Best at Detecting Alzheimer’s Disease: A head‐to‐head Comparison Using Relevance Maps

AbstractBackgroundAlzheimer’s disease (AD) is a neurodegenerative disorder which causes gradual and irreversible damage to the brain. These patterns of brain damage can be detected using T1‐weighted MRI scans by modern neuroimaging methods. Recent developments in convolutional neural networks (CNN) have achieved promising results in various computer‐vision tasks. Feature attribution methods such as layer‐wise relevance propagation (LRP) allow tracing back the information flow in CNNs to derive relevance heatmaps, which approximate the contribution of the input image regions on the model decision. We addressed the open question, which of the most common CNN architectures is best suited for medical image detection, i.e. AD classification based on MRI data.MethodFour CNN architectures are widely used in the literature: AlexNet, VGG, ResNet, and DenseNet. We adapted these CNN architectures to be used with 3D brain MRI data and trained the models on a heterogeneous dataset with N&gt;2200 from four large studies. We applied tenfold cross‐validation and additionally evaluated the results on an independent test dataset.ResultThe more complex CNN architectures DenseNet and ResNet with ‘skip connections’, provided the best results for both group separation tasks of ‐ AD vs Cognitively Normal (CN), and Mild Cognitive Impairment (MCI) vs CN, although the overall differences in accuracy did not reach statistical significance. From the mean relevance maps obtained using the LRP relevance propagation method (see Fig. 1.), we found that DenseNet focused primarily on the medial temporal lobe and posterior cingulate cortex atrophy. ResNet’s mean relevance maps were more noisy and heterogeneous.ConclusionDenseNet provided the most focused relevance maps, best matching apriori expectations of brain regions contributing to the detection of AD. Its ‘dense connections’ between the layers enabled a highly efficient information flow at various scales. Our evaluation setup also demonstrated the value of a holistic evaluation of models, where relevance maps are used in combination with classical performance metrics.

The Validity of the Thai Electronic Version of the Montreal Cognitive Assessment (MoCA)

AbstractElectronic cognitive assessment tools are essential in screening and managing patients’ disease progression in clinical practices. Neurocognitive disorders, particularly mild cognitive impairment (MCI) and Alzheimer’s Dementia (AD), can be assessed using the Montreal Cognitive Assessment (MoCA) [1, 2]. A digital version of the MoCA could help reduce the burden on healthcare providers, increase accessibility, and maximize its utility, especially in countries that lack medical staff such as Thailand. This pilot study aimed to establish the concurrent validity between the electronic version of the MoCA‐Thai (eMoCA‐Thai), developed by the authors of this paper, to the original paper‐and‐pencil version (MoCA‐Thai) in adults who are cognitively normal (CN), have MCI, and have AD. 54 adults between the ages of 60 to 90 attending a dementia clinic were administered both the eMoCA‐Thai and the standard MoCA‐Thai one week apart. The primary outcome measures include the total scores and all subscale scores of the two test versions. Correlations and differences between scores were analyzed using concordance correlation coefficients (CCC) and various parametric t‐tests. Results show that there is high correlation for total MoCA scores with a CCC of 0.919 and a mean difference of ‐0.204 (‐6.311, 5.904). All the cognitive subdomain scores had moderate to high CCC of more than 0.4. The differences in average total score (22.81±7.49 for the MoCA‐Thai and 23.02±7.99 for the eMoCA‐Thai) were not statistically significant (p = 0.633), indicating no differences between the MoCA scores of both versions. When comparing total MoCA scores between participant groups, both CN and MCI had acceptable correlations (CCC &gt;0.2) with a mean difference of 0.355 (‐4.720, 5.429) and ‐1.643 (9.001, 5.715), respectively, whilst AD had excellent correlation (CC &gt;0.8) with a mean difference of 0.111 (‐6.496, 6.717). Subdomain scores between participant groups were not statistically significantly different except for Delayed Recall for all participants combined (p&lt;0.05). In conclusion, there is adequate concurrent validity between the MoCA‐Thai and eMoCA‐Thai, making it a useful tool to help improve workflow and increase accessibility.

Sex‐associated differences in plasma and cerebrospinal fluid biomarkers in Early‐onset Alzheimer’s disease

AbstractBackgroundFemale sex is associated with greater atrophy, amyloid and tau burden in Early‐onset Alzheimer’s Disease (EOAD) in the Longitudinal EOAD Study (LEADS). APOE‐e4 non‐carrier‐status was found to be a further predictor of EOAD pathology. We expanded the analyses by examining the impact of sex and APOE‐e4 on plasma and cerebrospinal fluid (CSF) biomarkers of AD. Plasma markers included: neurofilament light chain (NfL), plasma glial fibrillary acidic protein (GFAP), Aß 42/40, and pTau231. CSF markers included: Aß 42/40, neurogranin, tTau, pTau181, synaptosomal‐associated protein‐25 (SNAP25), YKL‐40, and visinin‐like protein‐1 (VILIP1).MethodWe included 201 amyloid‐positive EOAD, 64 amyloid‐negative Early‐onset cognitively impaired (EOnonAD), and 86 Cognitively Normal (CN) LEADS participants with plasma biomarker data. Of these 100 EOAD, 35 EOnonAD, and 38 CN participants also had CSF data. Participants were stratified by sex and APOE‐e4 genotype. Demographics (age, education, APOE‐e4, MMSE) and biomarker differences were compared using ANOVA within each diagnostic group. ANCOVAs were run to control for the effects of age and education on biomarkers.ResultCompared to men, EOAD women showed greater levels of plasma NfL (p = .03), and GFAP (p&lt;.001) while Aß 42/40, and pTau231 were comparable. In CSF, EOAD women showed higher levels of neurogranin (p = .01), tTau (p = .01), pTau181 (p = .01) and VILIP1 (p = .02). EOnonAD women showed significantly greater plasma GFAP levels (p = .02). In the CN cohort, women had greater CSF SNAP25 levels (p = .03) while men showed a trend for higher plasma pTau231 levels (p = .05). APOE‐e4 carrier‐status was not associated with differences in levels of plasma or CSF biomarkers in either sex.ConclusionAs highly selective and specific AD fluid biomarkers emerge, understanding sex‐based differences in CSF and plasma is imperative. Female sex is associated with higher levels of neurodegeneration (NfL, tTau, pTau181, neurogranin, VILIP1) in EOAD and astrogliosis (GFAP) in both EOAD and EOnonAD. This data supports our previous findings that greater pathology burden is associated with female sex. This paves the way for further examination of sex and APOE‐e4 genotype‐based differences in imaging and fluid biomarkers, their associations, and utility in early diagnosis.

Topological Network Analysis of Beta‐Amyloid and Tau in Alzheimer’s Disease Using PET Imaging Data

AbstractBackgroundAlzheimer’s disease (AD) is characterized by the presence of beta‐amyloid (Aβ) and tau in the brain. Understanding the interplay between these proteins and their spatial distribution may provide insights into disease progression. This study aims to investigate the topological features of Aβ and tau networks in AD, mild cognitive impairment (MCI), and normal cognition (NC) groups using PET images from the ADNI dataset.MethodPartial correlation matrices between the standardized uptake value ratio of 68 regions of interest (ROI’s) were used to generate networks for the AD, MCI, and NC groups, with nodes representing ROI’s and edges representing correlation coefficients. Topological features, including H0 features (connected components), H1 features (loops), and landscape functions of H1 features, were calculated for each sample’s persistent homology. For each group, 100 bootstrapping samples were used to generate the 95% confidence intervals for landscape functions. The connected components were further studied by representing the samples as hierarchical tree structures. In the tree space, the first few generated connections, representing the most consistent pairs of regions positive for Aβ and tau, were compared across groups. Clustering performances were compared using Wasserstein distance of H0 features, Wasserstein distance of H1 features, landscape function distance, and geodesic distance in tree space respectively.ResultThe geodesic distance in tree space showed the best clustering accuracy in both Aβ and tau analyses. Compared with Aβ, H1 features and the corresponding landscape functions performed better for tau. Unique connections in AD and MCI groups were identified, with AD showing more unique connections than MCI compared to NC samples. Dominant unique connections in the AD group included left‐right isthmus cingulate and left fusiform‐left inferior temporal for Aβ, and left‐right paracentral and left cuneus‐left pericalcarine for tau. These connections indicate region pairs with related concentrations of Aβ and tau in AD.ConclusionTopological network analysis can reveal complex underlying structures in PET images and identify consistent patterns in ROI’s associated with Aβ and tau localization. Tau aggregation exhibits more complex behavior than Aβ and is more strongly associated with AD. The identified unique connections in the AD group may serve as potential biomarkers for disease progression.

Which Convolutional Neural Network Architecture is the Best at Detecting Alzheimer’s Disease: A head‐to‐head Comparison Using Relevance Maps

AbstractBackgroundAlzheimer’s disease (AD) is a neurodegenerative disorder which causes gradual and irreversible damage to the brain. These patterns of brain damage can be detected using T1‐weighted MRI scans by modern neuroimaging methods. Recent developments in convolutional neural networks (CNN) have achieved promising results in various computer‐vision tasks. Feature attribution methods such as layer‐wise relevance propagation (LRP) allow tracing back the information flow in CNNs to derive relevance heatmaps, which approximate the contribution of the input image regions on the model decision. We addressed the open question, which of the most common CNN architectures is best suited for medical image detection, i.e. AD classification based on MRI data.MethodFour CNN architectures are widely used in the literature: AlexNet, VGG, ResNet, and DenseNet. We adapted these CNN architectures to be used with 3D brain MRI data and trained the models on a heterogeneous dataset with N&gt;2200 from four large studies. We applied tenfold cross‐validation and additionally evaluated the results on an independent test dataset.ResultThe more complex CNN architectures DenseNet and ResNet with ‘skip connections’, provided the best results for both group separation tasks of ‐ AD vs Cognitively Normal (CN), and Mild Cognitive Impairment (MCI) vs CN, although the overall differences in accuracy did not reach statistical significance. From the mean relevance maps obtained using the LRP relevance propagation method (see Fig. 1.), we found that DenseNet focused primarily on the medial temporal lobe and posterior cingulate cortex atrophy. ResNet’s mean relevance maps were more noisy and heterogeneous.ConclusionDenseNet provided the most focused relevance maps, best matching apriori expectations of brain regions contributing to the detection of AD. Its ‘dense connections’ between the layers enabled a highly efficient information flow at various scales. Our evaluation setup also demonstrated the value of a holistic evaluation of models, where relevance maps are used in combination with classical performance metrics.

Correlations of p‐tau217, p‐tau181 and tau levels between CSF and plasma that are measured by fully automated immunoassay platform

AbstractBackgroundCerebrospinal fluid (CSF)‐based ATN classification (A: β‐amyloid (Aβ) deposition, T: pathogenic tau accumulation, N: neurodegeneration) is being widely accepted for the characterization of Alzheimer’s disease (AD). However, for the implementation of ATN classification in clinical practice, blood‐based biomarker is preferable in terms of specimen collection ease. Therefore, we have developed the blood‐based biomarker assays that correlate well with CSF levels. We have previously demonstrated that the plasma tau and phosphorylated tau at threonine‐181 (p‐tau181) immunoassays can significantly differentiate between AD, mild cognitive impairment (MCI), and cognitively normal (CN) groups. Here, we evaluated the correlation of biomarker levels between CSF and plasma utilizing the biomarker panel with addition of phosphorylated tau at threonine‐217 (p‐tau217) immunoassay.MethodWe have measured p‐tau217, p‐tau181 and tau using commercially available specimens on a fully automated immunoassay platform, HISCLTM analyzer. The platform can achieve high degree of sensitivity, requires only 30 µl of sample and 17 minutes to complete the measurement per assay. The bio‐analytical performance for the immunoassays was evaluated by measuring lower limit of quantification (LLOQ). The developed immunoassays were used to measure the paired CSF and plasma specimens from AD (n = 10), MCI (n = 35) and CN (n = 9) groups.ResultThe quantitative sensitivity for p‐tau217, p‐tau181 and tau were sufficient to measure plasma specimens in all groups. The plasma biomarker levels were significantly different between the AD and CN groups for all the biomarkers. We have found good correlations of biomarker levels between CSF and plasma for p‐tau217 (rs = 0.594, p &lt; 0.01), p‐tau181 (rs = 0.453, p &lt; 0.01) and tau (rs = 0.326, p &lt; 0.05).ConclusionOur results indicated the possibility of detecting plasma p‐tau217, p‐tau181 and tau levels which may reflect the CSF levels. In addition to these assays, we have also developed plasma Ab and neurofilament light chain assays that had demonstrated good correlation with CSF levels. Thus, all the plasma assays developed on the HISCL analyzer could assess ATN classification. The blood‐based biomarker panel using our fully automated assay will accelerate research in the dementia field and may help in appropriate diagnosis of patients with dementia in routine clinical practice.