“Bessel Functions and Temporal Orderings of Biomarkers in Preclinical Alzheimer’s Disease”
Abstract: Multivariate cross-sectional and longitudinal data across multiple modalities arise in studies of many chronic diseases such as Alzheimer Disease (AD), which are intended to measure different aspects of the underlying disease processes. An important scientific question is to locate the biomarkers that may exert the earliest changes in the disease processes, likely many years prior to the onset of clinical symptoms. The answer to this question is crucial as it may help identify the target for earliest possible prevention or therapeutic intervention. The statistical inferences on the ordering of biomarker changes, however, remain elusive. We develop a statistical methodology to address these very questions by jointly modeling multiple biomarkers, and define their orderings of changes by using the estimated rates of changes. Specifically, we choose a multivariate random intercept and random slope model, and propose a confidence interval estimate to and a statistical test for the ordering of changes across multiple biomarkers whose sampling distribution depends on the modified Bessel functions of the second type. Finally, we demonstrate the proposed methodology by applying it to the database of a biomarker study of AD including cerebrospinal fluid (CSF) and neuroimaging biomarkers, and infer on the likely orderings of the biomarker changes during the preclinical stage of AD.
Division of Biostatistics seminars
For inquiries contact Chengjie Xiong.