An explainable machine learning model of cognitive decline derived from speech.

INTRODUCTION: Traditional Alzheimer's disease (AD) and mild cognitive impairment (MCI) screening lacks the sensitivity and timeliness required to detect subtle indicators of cognitive decline. Multimodal artificial intelligence technologies using only speech data promise improved detection of neurodegenerative disorders. METHODS: Speech collected over the telephone from 91 older participants who were cognitively healthy (n = 29) or had diagnoses of AD (n = 30) or amnestic MCI (aMCI; n = 32) was analyzed with multimodal natural language and speech processing methods. An explainable ensemble decision tree classifier for the multiclass prediction of cognitive decline was created. RESULTS: This approach was 75% accurate overall-an improvement over traditional speech-based screening tools and a unimodal language-based model. We include a dashboard for the examination of the results, allowing for novel ways of interpreting such data. DISCUSSION: This work provides a foundation for a meaningful change in medicine as clinical translation, scalability, and user friendliness were core to the methodologies. Highlights: Remote assessments and artificial intelligence (AI) models allow greater access to cognitive decline screening.Speech impairments differ significantly between mild AD, amnestic mild cognitive impairment (aMCI), and healthy controls.AI predictions of cognitive decline are more accurate than experts and standard tools.The AI model was 75% accurate in classifying mild AD, aMCI, and healthy controls.

Influence of amyloid and diagnostic syndrome on non-traditional memory scores in early-onset Alzheimer's disease.

INTRODUCTION: The Rey Auditory Verbal Learning Test (RAVLT) is a useful neuropsychological test for describing episodic memory impairment in dementia. However, there is limited research on its utility in early-onset Alzheimer's disease (EOAD). We assess the influence of amyloid and diagnostic syndrome on several memory scores in EOAD. METHODS: We transcribed RAVLT recordings from 303 subjects in the Longitudinal Early-Onset Alzheimer's Disease Study. Subjects were grouped by amyloid status and syndrome. Primacy, recency, J-curve, duration, stopping time, and speed score were calculated and entered into linear mixed effects models as dependent variables. RESULTS: Compared with amyloid negative subjects, positive subjects exhibited effects on raw score, primacy, recency, and stopping time. Inter-syndromic differences were noted with raw score, primacy, recency, J-curve, and stopping time. DISCUSSION: RAVLT measures are sensitive to the effects of amyloid and syndrome in EOAD. Future work is needed to quantify the predictive value of these scores. HIGHLIGHTS: RAVLT patterns characterize various presentations of EOAD and EOnonAD Amyloid impacts raw score, primacy, recency, and stopping time Timing-based scores add value over traditional count-based scores.

A blood biomarker test for brain amyloid impacts the clinical evaluation of cognitive impairment.

OBJECTIVE: The objective of this study was to examine clinicians' patient selection and result interpretation of a clinically validated mass spectrometry test measuring amyloid beta and ApoE blood biomarkers combined with patient age (PrecivityAD® blood test) in symptomatic patients evaluated for Alzheimer's disease (AD) or other causes of cognitive decline. METHODS: The Quality Improvement and Clinical Utility PrecivityAD Clinician Survey (QUIP I, ClinicalTrials.gov Identifier: NCT05477056) was a prospective, single-arm cohort study among 366 patients evaluated by neurologists and other cognitive specialists. Participants underwent blood biomarker testing and received an amyloid probability score (APS), indicating the likelihood of a positive result on an amyloid positron emission tomography (PET) scan. The primary study outcomes were appropriateness of patient selection as well as result interpretation associated with PrecivityAD blood testing. RESULTS: A 95% (347/366) concordance rate was noted between clinicians' patient selection and the test's intended use criteria. In the final analysis including these 347 patients (median age 75 years, 56% women), prespecified test result categories incorporated 133 (38%) low APS, 162 (47%) high APS, and 52 (15%) intermediate APS patients. Clinicians' pretest and posttest AD diagnosis probability changed from 58% to 23% in low APS patients and 71% to 89% in high APS patients (p < 0.0001). Anti-AD drug therapy decreased by 46% in low APS patients (p < 0.0001) and increased by 57% in high APS patients (p < 0.0001). INTERPRETATION: These findings demonstrate the clinical utility of the PrecivityAD blood test in clinical care and may have added relevance as new AD therapies are introduced.

Sex and APOE ε4 carrier effects on atrophy, amyloid PET, and tau PET burden in early-onset Alzheimer's disease.

INTRODUCTION: We used sex and apolipoprotein E ε4 (APOE ε4) carrier status as predictors of pathologic burden in early-onset Alzheimer's disease (EOAD). METHODS: We included baseline data from 77 cognitively normal (CN), 230 EOAD, and 70 EO non-Alzheimer's disease (EOnonAD) participants from the Longitudinal Early-Onset Alzheimer's Disease Study (LEADS). We stratified each diagnostic group by males and females, then further subdivided each sex by APOE ε4 carrier status and compared imaging biomarkers in each stratification. Voxel-wise multiple linear regressions yielded statistical brain maps of gray matter density, amyloid, and tau PET burden. RESULTS: EOAD females had greater amyloid and tau PET burdens than males. EOAD female APOE ε4 non-carriers had greater amyloid PET burdens and greater gray matter atrophy than female ε4 carriers. EOnonAD female ε4 non-carriers also had greater gray matter atrophy than female ε4 carriers. DISCUSSION: The effects of sex and APOE ε4 must be considered when studying these populations. HIGHLIGHTS: Novel analysis examining the effects of biological sex and apolipoprotein E ε4 (APOE Îµ4) carrier status on neuroimaging biomarkers among early-onset Alzheimer's disease (EOAD), early-onset non-AD (EOnonAD), and cognitively normal (CN) participants. Female sex is associated with greater pathology burden in the EOAD cohort compared to male sex. The effect of APOE ε4 carrier status on pathology burden was the most impactful in females across all cohorts.

White matter hyperintensities are higher among early-onset Alzheimer's disease participants than their cognitively normal and early-onset nonAD peers: Longitudinal Early-onset Alzheimer's Disease Study (LEADS).

INTRODUCTION: We compared white matter hyperintensities (WMHs) in early-onset Alzheimer's disease (EOAD) with cognitively normal (CN) and early-onset amyloid-negative cognitively impaired (EOnonAD) groups in the Longitudinal Early-Onset Alzheimer's Disease Study. METHODS: We investigated the role of increased WMH in cognition and amyloid and tau burden. We compared WMH burden of 205 EOAD, 68 EOnonAD, and 89 CN participants in lobar regions using t-tests and analyses of covariance. Linear regression analyses were used to investigate the association between WMH and cognitive impairment and that between amyloid and tau burden. RESULTS: EOAD showed greater WMHs compared with CN and EOnonAD participants across all regions with no significant differences between CN and EOnonAD groups. Greater WMHs were associated with worse cognition. Tau burden was positively associated with WMH burden in the EOAD group. DISCUSSION: EOAD consistently showed higher WMH volumes. Overall, greater WMHs were associated with worse cognition and higher tau burden in EOAD. HIGHLIGHTS: This study represents a comprehensive characterization of WMHs in sporadic EOAD. WMH volumes are associated with tau burden from positron emission tomography (PET) in EOAD, suggesting WMHs are correlated with increasing burden of AD. Greater WMH volumes are associated with worse performance on global cognitive tests. EOAD participants have higher WMH volumes compared with CN and early-onset amyloid-negative cognitively impaired (EOnonAD) groups across all brain regions.

Cerebrospinal fluid biomarkers in the Longitudinal Early-onset Alzheimer's Disease Study.

INTRODUCTION: One goal of the Longitudinal Early Onset Alzheimer's Disease Study (LEADS) is to define the fluid biomarker characteristics of early-onset Alzheimer's disease (EOAD). METHODS: Cerebrospinal fluid (CSF) concentrations of Aß1-40, Aß1-42, total tau (tTau), pTau181, VILIP-1, SNAP-25, neurogranin (Ng), neurofilament light chain (NfL), and YKL-40 were measured by immunoassay in 165 LEADS participants. The associations of biomarker concentrations with diagnostic group and standard cognitive tests were evaluated. RESULTS: Biomarkers were correlated with one another. Levels of CSF Aß42/40, pTau181, tTau, SNAP-25, and Ng in EOAD differed significantly from cognitively normal and early-onset non-AD dementia; NfL, YKL-40, and VILIP-1 did not. Across groups, all biomarkers except SNAP-25 were correlated with cognition. Within the EOAD group, Aß42/40, NfL, Ng, and SNAP-25 were correlated with at least one cognitive measure. DISCUSSION: This study provides a comprehensive analysis of CSF biomarkers in sporadic EOAD that can inform EOAD clinical trial design.

Baseline neuropsychiatric symptoms and psychotropic medication use midway through data collection of the Longitudinal Early-Onset Alzheimer's Disease Study (LEADS) cohort.

INTRODUCTION: We examined neuropsychiatric symptoms (NPS) and psychotropic medication use in a large sample of individuals with early-onset Alzheimer's disease (EOAD; onset 40-64 years) at the midway point of data collection for the Longitudinal Early-onset Alzheimer's Disease Study (LEADS). METHODS: Baseline NPS (Neuropsychiatric Inventory - Questionnaire; Geriatric Depression Scale) and psychotropic medication use from 282 participants enrolled in LEADS were compared across diagnostic groups - amyloid-positive EOAD (n = 212) and amyloid negative early-onset non-Alzheimer's disease (EOnonAD; n = 70). RESULTS: Affective behaviors were the most common NPS in EOAD at similar frequencies to EOnonAD. Tension and impulse control behaviors were more common in EOnonAD. A minority of participants were using psychotropic medications, and use was higher in EOnonAD. DISCUSSION: Overall NPS burden and psychotropic medication use were higher in EOnonAD than EOAD participants. Future research will investigate moderators and etiological drivers of NPS, and NPS differences in EOAD versus late-onset AD.

Increasing access to cognitive screening in the elderly: Applying natural language processing methods to speech collected over the telephone.

Barriers to healthcare access are widespread in elderly populations, with a major consequence that older people are not benefiting from the latest technologies to diagnose disease. Recent advances in the automated analysis of speech show promising results in the identification of cognitive decline associated with Alzheimer's disease (AD), as well as its purported pre-clinical stage. We utilized automated methods to analyze speech recorded over the telephone in 91 community-dwelling older adults diagnosed with mild AD, amnestic mild cognitive impairment (aMCI) or cognitively healthy. We asked whether natural language processing (NLP) and machine learning could more accurately identify groups than traditional screening tools and be sensitive to subtle differences in speech between the groups. Despite variable recording quality, NLP methods differentiated the three groups with greater accuracy than two traditional dementia screeners and a clinician who read transcripts of their speech. Imperfect speech data collected via a telephone is of sufficient quality to be examined with the latest speech technologies. Critically, these data reveal significant differences in speech that closely match the clinical diagnoses of AD, aMCI and healthy control.

Nilotinib Effects on Safety, Tolerability, and Biomarkers in Alzheimer's Disease.

OBJECTIVE: Preclinical evidence with nilotinib, a US Food and Drug Administration (FDA)-approved drug for leukemia, indicates improvement in Alzheimer's disease phenotypes. We investigated whether nilotinib is safe, and detectable in cerebrospinal fluid, and alters biomarkers and clinical decline in Alzheimer's disease. METHODS: This single-center, phase 2, randomized, double-blind, placebo-controlled study investigated the safety, tolerability, and pharmacokinetics of nilotinib, and measured biomarkers in participants with mild to moderate dementia due to Alzheimer's disease. The diagnosis was supported by cerebrospinal fluid or amyloid positron emission tomography biomarkers. Nilotinib 150 mg versus matching placebo was taken orally once daily for 26 weeks followed by nilotinib 300 mg versus placebo for another 26 weeks. RESULTS: Of the 37 individuals enrolled, 27 were women and the mean (SD) age was 70.7 (6.48) years. Nilotinib was well-tolerated, although more adverse events, particularly mood swings, were noted with the 300 mg dose. In the nilotinib group, central nervous system (CNS) amyloid burden was significantly reduced in the frontal lobe compared to the placebo group. Cerebrospinal fluid Aß40 was reduced at 6 months and Aß42 was reduced at 12 months in the nilotinib group compared to the placebo. Hippocampal volume loss was attenuated (-27%) at 12 months and phospho-tau-181 was reduced at 6 months and 12 months in the nilotinib group. INTERPRETATION: Nilotinib is safe and achieves pharmacologically relevant cerebrospinal fluid concentrations. Biomarkers of disease were altered in response to nilotinib treatment. These data support a larger, longer, multicenter study to determine the safety and efficacy of nilotinib in Alzheimer's disease. ANN NEUROL 2020 ANN NEUROL 2020;88:183-194.

The BIOCARD Index: A Summary Measure to Predict Onset of Mild Cognitive Impairment.

BACKGROUND: Changes in neuropsychological testing, neuroimaging, and cerebrospinal fluid may precede mild cognitive impairment (MCI). However, these markers are not routinely performed in outpatient clinical visits. OBJECTIVE: To evaluate whether a simple clinical index, consisting of questions given to patients and their informants, could predict the onset of symptoms of MCI among cognitively normal individuals. MATERIALS AND METHODS: Two hundred twenty-two participants in the BIOCARD study received a detailed history, physical examination, and neuropsychological testing annually. An index was calculated by including questions about memory problems, depression, age, education, history of cerebrovascular disease risk factors, and brain injury, family history of dementia, and the Mini-Mental State examination score. Cox regression analyses were used to determine if this index score was related to diagnosis of MCI. RESULTS: The BIOCARD Index score mean for individuals who progressed to MCI was 20.3 (SD=2.9), whereas the score for individuals who remained normal was 24.8 (SD=2.3) (P<0.001) [hazard ratio, SE for subsequent diagnosis of MCI=0.75 (0.67 to 0.84); P<0.001]. CONCLUSIONS: Lower BIOCARD Index score predicted symptoms of MCI several years before the MCI diagnosis. The BIOCARD Index can be easily used in clinics to identify cognitively normal older individuals who are at risk for deterioration.

Amyloid-beta-induced ion flux in artificial lipid bilayers and neuronal cells: resolving a controversy.

Understanding the pathogenicity of amyloid-beta (Abeta) peptides constitutes a major goal in research on Alzheimer's disease (AD). One hypothesis entails that Abeta peptides induce uncontrolled, neurotoxic ion flux through cellular membranes. The exact biophysical mechanism of this ion flux is, however, a subject of an ongoing controversy which has attenuated progress toward understanding the importance of Abeta-induced ion flux in AD. The work presented here addresses two prevalent controversies regarding the nature of transmembrane ion flux induced by Alphabeta peptides. First, the results clarify that Alphabeta can induce stepwise ion flux across planar lipid bilayers as opposed to a gradual increase in transmembrane current; they show that the previously reported gradual thinning of membranes with concomitant increase in transmembrane current arises from residues of the solvent hexafluoroisopropanol, which is commonly used for the preparation of amyloid samples. Second, the results provide additional evidence suggesting that Abeta peptides can induce ion channel-like ion flux in cellular membranes that is independent from the postulated ability of Alphabeta to modulate intrinsic cellular ion channels or transporter proteins.