The Critical Path for Alzheimer’s Disease (CPAD) Consortium: A platform for pre‐competitive data sharing, standardization, and analysis to support quantitative tools for AD drug development

Abstract

AbstractBackgroundThe primary objective of the Critical Path for Alzheimer’s Disease (CPAD) consortium, a pre‐competitive consortium of the Critical Path Institute, is to promote, support, and manage pre‐competitive data and knowledge sharing from Alzheimer disease (AD) drug development stakeholders. Since 2019, CPAD has embarked on a new data sharing initiative to acquire fluid and imaging biomarker‐rich data sources for development of novel tools to quantify Alzheimer’s Disease progression across the AD continuum.MethodPatient‐level data from contemporary Phase2/Phase3 AD clinical trials and observational studies is acquired through formal data contribution agreements within CPAD, curated in collaboration with data contributors, standardized to common structure and terminology, and input into a relational data model (RDM) tool to generate analysis subsets. Data standardization is being accomplished using Clinical Data Interchange Standards Consortium (CDISC) standards. The RDM was developed using the open‐source software R and converts CDISC structure to analysis ready files for quantitative analyses through user defined primary keys, i.e. unique identifying variables for extracting information. The RDM includes a front‐end data interrogator that provides users with versatile capabilities to visualize and tabulate the data.ResultAs of October 2020, CPAD’s database (https://codr.c‐path.org/) contains 41 studies, standardized to the AD CDISC standards, of which three datasets were used to generate analysis subsets. Four additional datasets were recently acquired and are being standardized. Execution of Data Contribution Agreements (DCAs) is in progress for two contemporary datasets rich in key fluid and imaging biomarkers. Nine additional prioritized datasets are targeted for acquisition. The RDM is currently being used to transform the CDISC standardized data into analysis ready subsets containing patient demographics (age, sex), fluid biomarker information (Aβ 40/42), and clinical outcome assessments (Clinical Dementia Rating Scale – Sum of Boxes).ConclusionThe data acquisition and analysis platform developed is stimulating pre‐competitive efforts to generate quantitative tools for advancing AD drug development. Previously developed tools, including a regulatory endorsed clinical trial simulation tool, help optimize trial design and advance drug development in AD.