In 2010 we proposed, and in 2013 revised, a hypothetical model of the temporal evolution of biomarkers and clinical symptoms for individuals in the Alzheimer?s disease (AD) pathway. The biomarker model is based on a hypothesized cause and effect sequence which can be summarized as: amyloidosis (A), promotes tauopathy (T), which promotes neurodegeneration (N), which is the proximate cause of clinical symptoms (C). For simplicity, we use this ATNC notation: A? T ? N ? C. The current cycle of AG011378 was designed to test aspects of this model that were testable using imaging; however, a major missing element when the current cycle AG011378 began in 2013 was a method to measure fibrillar tau deposits with imaging. Tau PET has recently become available and, consequently for the first time, imaging biomarkers exist to ascertain three of the most important pathologic features of AD: amyloid, tau and neurodegeneration. An important insight gained from the current cycle of AG011378 was that modeling AD biomarkers on the assumption that AD is the only pathological process present in the general aging population is conceptually flawed. We argue that biomarker modeling within the general aging population should accommodate at least three broad groups based on currently available biomarkers: (1) individuals with no biomarker abnormalities whose future biomarker profiles are unknown; (2) individuals along the Alzheimer?s continuum; and (3) a heterogeneous group with primarily suspected non-AD pathologies (SNAP).
Our aims i n this renewal are based on modeling longitudinal amyloid PET (A), tau PET (T), and MRI (N) biomarkers. The overarching goals of the renewal are to empirically evaluate our hypothetical model of AD biomarkers and to create complementary imaging biomarker models for individuals who are not in the Alzheimer?s continuum.
Our specific aims are:
Aim 1. To model the sequence of biomarker and clinical transitions over time across multiple pathways.
Aim 1 employs hidden Markov models. 1a) To model the sequence of biomarker and clinical transitions in the Alzheimer?s continuum, 1b) To model the sequence of biomarker and clinical transitions in SNAP pathways.
Aim 2. To model continuous biomarker and clinical trajectories over time across multiple pathways.
Aim 2 analyses employ non-linear latent class mixed models. 2a) To model biomarker and clinical trajectories in the Alzheimer?s continuum, 2b) To model biomarker and clinical trajectories in SNAP pathways.
Aim 3. To formulate mechanistic inferences about biomarker and clinical changes over time across multiple pathways. 3a) inferences in the Alzheimer?s continuum, 3b) inferences in SNAP pathways.
Aim 4 : To establish temporal ordering of topographic spread of A, T, and N within each modality, across imaging modalities, and establish how these temporal orderings relate to cognitive impairment within (4a) the Alzheimer?s continuum and (4b) the SNAP pathways.
Aim 4 uses a conditional probability approach.
The current cycle of AG011378 was designed to test aspects of a hypothetical model of the temporal evolution of Alzheimer?s biomarkers that we proposed; however, a major missing element when the current cycle of AG011378 began was a method to measure fibrillar tau deposits with imaging. Tau PET has recently become available and consequently for the first time imaging biomarkers exist to ascertain three of the most important pathologic features of AD: amyloid, tau and neurodegeneration. Our aims in this renewal application are to empirically evaluate our hypothetical model of AD biomarkers and to create new imaging biomarker models for individuals who are not in the Alzheimer?s continuum using imaging measures of amyloid, tau, and neurodegeneration.
| Wang, Qi; Guo, Lei; Thompson, Paul M et al. (2018) The Added Value of Diffusion-Weighted MRI-Derived Structural Connectome in Evaluating Mild Cognitive Impairment: A Multi-Cohort Validation1. J Alzheimers Dis 64:149-169 |
