Development of Clinical Trial Simulation Tools for Alzheimer’s Disease through the Critical Path for Alzheimer’s Disease (CPAD) Consortium

Abstract

AbstractBackgroundOptimized clinical trial design remains a major challenge in drug development for Alzheimer’s Disease AD). To advance our understanding of the underlying pathophysiology of disease and help optimize clinical trial design, CPAD developed a clinical trial simulation (CTS) tool for the pre‐dementia stages of Alzheimer’s disease, using CDR‐SB as the primary endpoint. The tool is based on a disease progression model integrating baseline hippocampal volume, the effect of patient drop‐out and relevant sources of variability. The CTS was enhanced using data from contemporary clinical trials, adding fluid and imaging biomarkers of amyloid (Aβ) and tau pathology. Also included are dropout models, a placebo effect model, as well as drug‐effect models. Positron Emission Tomography (PET) measures of tau deposition, leveraging ongoing collaborative efforts to develop a standardized quantification scale across different tau tracers, will also be incorporated, to develop the first disease progression models utilizing a harmonized tau‐PET scale.MethodData were harmonized and both linear and non‐linear mixed effects models were developed for multiple endpoints, including demographics, time from and to diagnosis, genetic status (APOE4). Monte Carlo simulations were performed to compare the statistical power by sample size in trials with and without enrichment using relevant covariates. A longitudinal mixed‐effects model was fit with longitudinal tau and neurodegeneration outcomes. Cognition and biomarkers were explored in a multivariate model.ResultAs of October 2022, CPAD’s clinical trial repository contains over 7,000 unique participants that have baseline CDR‐SB measurement, 4,000+ participants with baseline Aβ measures (fluid or PET), and 2,000+ participants with baseline measures of amyloid, tau (e.g., cerebrospinal fluid p‐tau and/or tau‐PET), and neurodegeneration (e.g., MRI or neurofilament light). With these data, we modeled longitudinal trajectories of cognitive decline using baseline levels of amyloid, tau, and neurodegeneration. Future work will add models of longitudinal biomarker trajectories.ConclusionThe disease progression models will serve as the basis for a clinical trial simulation tool to facilitate design of contemporary clinical trials based on a quantitative understanding of the biological progression of key pathological features, amyloid and tau deposition, of AD.