Leveraging speech and artificial intelligence to screen for early Alzheimer's disease and amyloid beta positivity.

Early detection of Alzheimer's disease is required to identify patients suitable for disease-modifying medications and to improve access to non-pharmacological preventative interventions. Prior research shows detectable changes in speech in Alzheimer's dementia and its clinical precursors. The current study assesses whether a fully automated speech-based artificial intelligence system can detect cognitive impairment and amyloid beta positivity, which characterize early stages of Alzheimer's disease. Two hundred participants (age 54-85, mean 70.6; 114 female, 86 male) from sister studies in the UK (NCT04828122) and the USA (NCT04928976), completed the same assessments and were combined in the current analyses. Participants were recruited from prior clinical trials where amyloid beta status (97 amyloid positive, 103 amyloid negative, as established via PET or CSF test) and clinical diagnostic status was known (94 cognitively unimpaired, 106 with mild cognitive impairment or mild Alzheimer's disease). The automatic story recall task was administered during supervised in-person or telemedicine assessments, where participants were asked to recall stories immediately and after a brief delay. An artificial intelligence text-pair evaluation model produced vector-based outputs from the original story text and recorded and transcribed participant recalls, quantifying differences between them. Vector-based representations were fed into logistic regression models, trained with tournament leave-pair-out cross-validation analysis to predict amyloid beta status (primary endpoint), mild cognitive impairment and amyloid beta status in diagnostic subgroups (secondary endpoints). Predictions were assessed by the area under the receiver operating characteristic curve for the test result in comparison with reference standards (diagnostic and amyloid status). Simulation analysis evaluated two potential benefits of speech-based screening: (i) mild cognitive impairment screening in primary care compared with the Mini-Mental State Exam, and (ii) pre-screening prior to PET scanning when identifying an amyloid positive sample. Speech-based screening predicted amyloid beta positivity (area under the curve = 0.77) and mild cognitive impairment or mild Alzheimer's disease (area under the curve = 0.83) in the full sample, and predicted amyloid beta in subsamples (mild cognitive impairment or mild Alzheimer's disease: area under the curve = 0.82; cognitively unimpaired: area under the curve = 0.71). Simulation analyses indicated that in primary care, speech-based screening could modestly improve detection of mild cognitive impairment (+8.5%), while reducing false positives (-59.1%). Furthermore, speech-based amyloid pre-screening was estimated to reduce the number of PET scans required by 35.3% and 35.5% in individuals with mild cognitive impairment and cognitively unimpaired individuals, respectively. Speech-based assessment offers accessible and scalable screening for mild cognitive impairment and amyloid beta positivity.

Protocol of the Cognitive Health in Ageing Register: Investigational, Observational and Trial Studies in Dementia Research (CHARIOT): Prospective Readiness cOhort (PRO) SubStudy.

INTRODUCTION: The Cognitive Health in Ageing Register: Investigational, Observational and Trial Studies in Dementia Research (CHARIOT): Prospective Readiness cOhort (PRO) SubStudy (CPSS), sponsored by Janssen Pharmaceutical Research & Development LLC, is an Alzheimer's disease (AD) biomarker enriched observational study that began 3 July 2015 CPSS aims to identify and validate determinants of AD, alongside cognitive, functional and biological changes in older adults with or without detectable evidence of AD pathology at baseline. METHODS AND ANALYSIS: CPSS is a dual-site longitudinal cohort (3.5 years) assessed quarterly. Cognitively normal participants (60-85 years) were recruited across Greater London and Edinburgh. Participants are classified as high, medium (amnestic or non-amnestic) or low risk for developing mild cognitive impairment-Alzheimer's disease based on their Repeatable Battery for the Assessment of Neuropsychological Status performance at screening. Additional AD-related assessments include: a novel cognitive composite, the Global Preclinical Alzheimer's Cognitive Composite, brain MRI and positron emission tomography and cerebrospinal fluid analysis. Lifestyle, other cognitive and functional data, as well as biosamples (blood, urine, and saliva) are collected. Primarily, study analyses will evaluate longitudinal change in cognitive and functional outcomes. Annual interim analyses for descriptive data occur throughout the course of the study, although inferential statistics are conducted as required. ETHICS AND DISSEMINATION: CPSS received ethical approvals from the London-Central Research Ethics Committee (15/LO/0711) and the Administration of Radioactive Substances Advisory Committee (RPC 630/3764/33110) The study is at the forefront of global AD prevention efforts, with frequent and robust sampling of the well-characterised cohort, allowing for detection of incipient pathophysiological, cognitive and functional changes that could inform therapeutic strategies to prevent and/or delay cognitive impairment and dementia. Dissemination of results will target the scientific community, research participants, volunteer community, public, industry, regulatory authorities and policymakers. On study completion, and following a predetermined embargo period, CPSS data are planned to be made accessible for analysis to facilitate further research into the determinants of AD pathology, onset of symptomatology and progression. TRIAL REGISTRATION NUMBER: The CHARIOT:PRO SubStudy is registered with clinicaltrials.gov (NCT02114372). Notices of protocol modifications will be made available through this trial registry.

Measuring cognition and function in the preclinical stage of Alzheimer's disease.

The Alzheimer's Association's Research Roundtable met in November 2016 to explore how best to measure changes in cognition and function in the preclinical stage of Alzheimer's disease. This review will cover the tools and instruments currently available to identify populations for prevention trials, and measure subtle disease progression in the earliest stages of Alzheimer's disease, and will include discussions of suitable cognitive, behavioral, functional, composite, and biological endpoints for prevention trials. Current prevention trials are reviewed including TOMMOROW, Alzheimer's Prevention Initiative Autosomal Dominant Alzheimer's Disease Trial, the Alzheimer's Prevention Initiative Generation Study, and the Anti-Amyloid Treatment in Asymptomatic Alzheimer's to compare current approaches and tools that are being developed.

Clinically Meaningful Outcomes in Early Alzheimer Disease: A Consortia-Driven Approach to Identifying What Matters to Patients.

BACKGROUND: Numerous statistically derived composite measures have recently been proposed as clinical outcome assessments (COAs) for clinical trials in the early stages of Alzheimer disease. Critical Path Institute's Coalition Against Major Diseases (CAMD) advanced a proposed statistically derived composite measure to regulatory agencies with the goal of qualifying it as a COA for pre-dementia trials. In response to FDA's requirement to demonstrate that proposed COAs are meaningful to patients, this project aimed to identify the most important cognition-related concerns patients and informants report early in the disease and determine how this information maps to what is assessed by several statistically derived composite measures. METHODS: Leveraging qualitative research completed by Critical Path Institute's Patient-Reported Outcome Consortium, CAMD utilized a summary report that included frequency grids of reported concerns of amnestic mild cognitive impairment patients and their informants, as well as the narrative transcripts from focus groups. Transcripts were reviewed and analyzed to identify which cognitive domains the patient- and informant-reported concerns mapped onto. The results were then compared to see how well these cognitive domains were represented in various statistically derived composite measures. RESULTS: The patient- and informant-reported concerns primarily mapped to the cognitive domains of episodic memory and, secondarily, orientation and language. Depending on the specified composite, there were varying levels of alignment between their subcomponents and these cognitive domains. CONCLUSION: Through secondary analyses of existing qualitative data, this study examined several statistically derived composite measures and found that they generally capture cognitive domains that reflect aspects of day-to-day functioning that patients and informants consider meaningful.

Recommended cognitive outcomes in preclinical Alzheimer's disease: Consensus statement from the European Prevention of Alzheimer's Dementia project.

The Horizon 2020/IMI European Prevention of Alzheimer's Dementia (EPAD) project will undertake large-scale proof-of-concept trials in predementia Alzheimer's disease (AD). Within EPAD, the monitoring of cognitive trajectories in the preclinical period will constitute a central outcome measure; however, there are currently no clear guidelines as to how this should be achieved as most measures have been developed for the period around dementia diagnosis. The EPAD Scientific Advisory Group for Clinical and Cognitive Outcomes identified appropriate cognitive measures based on a literature search covering both cognitive correlates of preclinical brain changes from imaging studies and cognitive changes observed over time in nondementia population cohorts developing incident dementia. These measures were evaluated according to the following criteria: validity, coherence with biomarker changes, psychometric properties, cross-cultural suitability, availability of alternative forms, and normative data limited practice effects. The resulting consensus statement provides recommendations for both future drug trials and research into preclinical Alzheimer's disease.

Detecting cognitive changes in preclinical Alzheimer's disease: A review of its feasibility.

Significant progress has been made in characterizing the biological changes occurring in preclinical Alzheimer's disease (AD). Cognitive dysfunction has been viewed, however, as a late-stage phenomenon, despite increasing evidence that changes may be detected in the decades preceding dementia. In the absence of comprehensive evidence-based guidelines for preclinical cognitive assessment, longitudinal cohort and neuroimaging studies have been reviewed to determine the temporal order and brain biomarker correlates of specific cognitive functions. Episodic memory decline was observed to be the most salient cognitive function, correlating with high levels of amyloid deposition and hypoconnectivity across large-scale brain networks. Prospective studies point to early decline in both episodic and semantic memory processing as well as executive functions in the predementia period. The cognitive tests have, however, been principally those used to diagnose dementia. New procedures are required which target more finely the medial temporal lobe subregions first affected by clinically silent AD pathology.

Novel Statistically-Derived Composite Measures for Assessing the Efficacy of Disease-Modifying Therapies in Prodromal Alzheimer's Disease Trials: An AIBL Study.

BACKGROUND: There is a growing consensus that disease-modifying therapies must be given at the prodromal or preclinical stages of Alzheimer's disease (AD) to be effective. A major unmet need is to develop and validate sensitive measures to track disease progression in these populations. OBJECTIVE: To generate novel statistically-derived composites from standard scores, which have increased sensitivity in the assessment of change from baseline in prodromal AD. METHODS: An empirically based method was employed to generate domain specific, global, and cognitive-functional novel composites. The novel composites were compared and contrasted with each other, as well as standard scores for their ability to track change from baseline. The longitudinal characteristics and power to detect decline of the measures were evaluated. Data from participants in the Australian Imaging, Biomarkers and Lifestyle (AIBL) Study characterized as mild cognitively impaired with high neocortical amyloid-ß burden were utilized for the study. RESULTS: The best performing standard scores were CDR Sum-of-Boxes and MMSE. The statistically-derived novel composites performed better than the standard scores from which they were derived. The domain-specific composites generally did not perform as well as the global composites or the cognitive-functional composites. CONCLUSION: A systematic method was employed to generate novel statistically-derived composite measures from standard scores. Composites comprised of measures including function and multiple cognitive domains appeared to best capture change from baseline. These composites may be useful to assess progression or lack thereof in prodromal AD. However, the results should be replicated and validated using an independent clinical sample before implementation in a clinical trial.

The NAB shape learning subtest as a predictor of lateralized seizure onset.

This study presents the first empirical evaluation of the predictive value of the Neuropsychological Assessment Battery Shape Learning (NAB-SL) subtest in a sample of patients with unilateral temporal lobe epilepsy. Stimulus characteristics of the NAB-SL may improve predictive ability over other commonly used visual memory tests. Forty-nine patients with unilateral temporal lobe epilepsy were compared on measures of non-verbal and verbal memory (NAB-SL and Wechsler Memory Scale-III subtests). Univariate and forward conditional logistic regressions identified predictive values for each memory test individually and in combination. The NAB-SL delayed memory demonstrated consistently stronger predictive power over visual reproduction at the univariate and multivariate levels. The NAB-SL was a good predictor (80% range) of lateralized seizure onset when combined with a verbal memory measure. These preliminary results provide support for the use of the NAB-SL in preoperative epilepsy evaluations as a predictor of non-dominant temporal lobe dysfunction. Potential benefits of this test are discussed.

Frequency of abnormal scores on the Neuropsychological Assessment Battery Screening Module (S-NAB) in a mixed neurological sample.

The Neuropsychological Assessment Battery (NAB; Stern & White, 2003; White & Stern, 2003) is a comprehensive, modular battery of tests comprised of the following six modules: (a) Screening, (b) Attention, (c) Language, (d) Memory, (e) Spatial, and (f) Executive Functions. The Screening Module is an abbreviated version of the full NAB. The purpose of this descriptive study was to present index and primary test score information for the Screening Module in a mixed sample of patients with known neurological conditions. Participants were 37 outpatients with clear evidence of neurological damage or disease. Performance decrements were found on the Attention Index, most notably on the Numbers and Letters tests. Decrements were also found on the Executive Functions Index, most notably on the Word Generation test. Somewhat surprisingly, patients performed well across most of the individual test scores. This mixed clinical sample showed less neuropsychological compromise than the clinical samples presented in the NAB manual.

The influence of cognitive reserve on neuropsychological functioning following coronary artery bypass grafting (CABG).

Neuropsychological impairment is common, yet variable, after coronary artery bypass grafting (CABG). Similar variability has been observed in other CNS-related diseases. Empirical findings in Alzheimer's disease and HIV, among other areas, suggest cognitive reserve (CR) may mediate the cognitive impact of these diseases. The present study examined whether CR mediates neuropsychological outcome after CABG. Participants were 42 (N=42) individuals who underwent elective, normothermic CABG. Each was placed in high (n=22) or low (n=20) CR groups based on estimated premorbid intelligence and occupational attainment. All were administered neuropsychological tests preoperatively and at discharge. The total incidence of neuropsychological decline (66.7%) was not significantly different between CR groups. However, on working memory and executive function tests, specifically, the high CR group demonstrated greater post-operative decline compared to the low CR group. These data are considered in the context of a threshold model of CR theory.

Preliminary examination of cognitive reserve theory in closed head injury.

This investigation was designed to provide preliminary support for cognitive reserve theory in closed head injury (CHI), and demonstrate the effectiveness of using the Oklahoma premorbid intelligence estimate (OPIE) in research and clinical activities. Out of a possible 124 consecutive referrals, 26 patients (N=26) who underwent neuropsychological assessment following brain injury met study inclusion/exclusion criteria. Participants were included if they had exited post-traumatic amnesia (PTA), demonstrated uncompromised upper extremity use, displayed adequate verbal communication, and were judged capable of completing a full neuropsychological evaluation. Participants were divided into a closed head injury-negative premorbid history (CHI-) or closed head injury-positive premorbid history (CHI+) group based upon premorbid variables (e.g., history of alcoholism). Groups did not differ in terms of demographic variables or premorbid IQ. Despite having less severe head injuries, the CHI+ group had a greater pre-post difference for PIQ, and a significantly larger VIQ/PIQ discrepancy than the CHI- group. In conclusion, these findings suggest that the CHI+ group had diminished cognitive reserve secondary to the aggregate effects of premorbid insult, which resulted in greater cognitive decline following an additional stressor (i.e., CHI) than what might otherwise be expected from the head injury alone.