Two Preclinical Latent Scores to Predict Occurrence of DAT The cognitive decline associated with Alzheimer's disease (AD) occurs years prior to the clinical diagnosis. However, the emergence of the earliest cognitive and functional impairment and the precise duration of the preclinical progression remain poorly understood by clinicians. Better methods are therefore urgently needed to reliably detect the antecedent cognitive and functional changes before the onset of the dementia of Alzheimer type (DAT). Whereas the conventional scores of the standard cognitive and functional batteries are successful in discriminating fully expressed DAT from normal aging, their ability to track subtle preclinical disease progression is uncertain, although it is possible that many individual items from them may predict the symptomatic onset of AD. Using rich and high quality longitudinal data from Washington University (WU) Alzheimer's Disease Research Center (ADRC), Rush University (RU) Alzheimer's Disease Center (ADC), and the Einstein Aging Study (EAS) and the Bronx Aging Study (BAS) at Albert Einstein College of Medicine (AECOM), this project will first conduct longitudinal item analyses to determine whether and to what degree individual item scores from tests of the 4 cognitive and functional batteries are sensitive and informative to longitudinal preclinical changes and predictive to the development of DAT, and how these item changes are correlated with cognitive reserve proxies (e.g., education and cognitive activities), ApoE genotype, preclinical measures of biomarkers including cerebrospinal fluid (CSF) molecular biomarkers, MRI brain volumetric markers, amyloid neuroimaging with PIB, as well as neuropathological diagnoses. Second, informative items will be optimally integrated through Item Response Theory (IRT) to estimate the preclinical latent cognitive and functional constructs and assess the longitudinal growth pattern of these constructs as well as the precise duration of the preclinical AD. Third, we will compare the predictive power of DAT between the estimated preclinical latent cognitive and functional constructs and the conventional test scores. Finally, we will develop clinically useful score reports for the National Alzheimer's Coordinating Center (NACC) Uniform Data Set (UDS) to summarize the optimally estimated preclinical latent cognitive and functional constructs for tracking the antecedent longitudinal changes of AD. We will also provide optimally estimated design parameters (e.g., sample sizes) for the Alzheimer's Disease Cooperative Study (ADCS) to conduct future preventive and therapeutic trials on Mild Cognitive Impairment (MCI) when the estimated preclinical latent cognitive and functional constructs are used as primary efficacy endpoints. Our interdisciplinary team of investigators at the WU ADRC, RU ADC, and AECOM will demonstrate the degree of improved detection of preclinical longitudinal cognitive and functional changes of AD using the state-of-the-art longitudinal statistical methods, modern psychometric theory (i.e., IRT), and cutting-edge bioinformatics techniques.

Public Health Relevance

This project will focus on detecting the earliest possible signs of preclinical cognitive and functional changes of Alzheimer's disease. This project is significant because understanding very early cognitive and functional changes antecedent to the onset of DAT will allow therapeutic interventions to be administered well before dementia symptoms are fully developed and a clinical diagnosis is rendered. The knowledge obtained will greatly help develop early therapeutic treatments or preventions of the disease before it is too late.

National Institute of Health (NIH)
National Institute on Aging (NIA)
Research Project (R01)
Project #
Application #
Study Section
Clinical Neuroscience and Neurodegeneration Study Section (CNN)
Program Officer
Hsiao, John
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Washington University
Biostatistics & Other Math Sci
Schools of Medicine
Saint Louis
United States
Zip Code
Shaparin, Naum; White, Robert; Andreae, Michael et al. (2014) A longitudinal linear model of patient characteristics to predict failure to attend an inner-city chronic pain clinic. J Pain 15:704-11
Davidai, G; Cotton, D; Gorelick, P et al. (2014) Dipyridamole-induced headache and lower recurrence risk in secondary prevention of ischaemic stroke: a post hoc analysis. Eur J Neurol 21:1311-7
Aurora, S K; Dodick, D W; Diener, H-C et al. (2014) OnabotulinumtoxinA for chronic migraine: efficacy, safety, and tolerability in patients who received all five treatment cycles in the PREEMPT clinical program. Acta Neurol Scand 129:61-70
Xiong, Chengjie; Luo, Jingqin; Gao, Feng et al. (2014) Optimizing parameters in clinical trials with a randomized start or withdrawal design. Comput Stat Data Anal 69:101-113
Xiong, Chengjie; Weng, Hua; Bennett, David A et al. (2014) Subsets of a large cognitive battery better power clinical trials on early stage Alzheimer's disease. Neuroepidemiology 43:131-9
Lipton, Richard B; Buse, Dawn C; Hall, Charles B et al. (2014) Reduction in perceived stress as a migraine trigger: testing the "let-down headache" hypothesis. Neurology 82:1395-401
Lipton, Richard B; Serrano, Daniel; Pavlovic, Jelena M et al. (2014) Improving the classification of migraine subtypes: an empirical approach based on factor mixture models in the American Migraine Prevalence and Prevention (AMPP) Study. Headache 54:830-49
Dong, Tuochuan; Kang, Le; Hutson, Alan et al. (2014) Confidence interval estimation of the difference between two sensitivities to the early disease stage. Biom J 56:270-86
Attwood, Kristopher; Tian, Lili; Xiong, Chengjie (2014) Diagnostic thresholds with three ordinal groups. J Biopharm Stat 24:608-33
Haut, Sheryl R; Hall, Charles B; Borkowski, Thomas et al. (2013) Modeling seizure self-prediction: an e-diary study. Epilepsia 54:1960-7

Showing the most recent 10 out of 17 publications