MODEL ORGANISM DEVELOPMENT AND EVALUATION FOR LATE-ONSET ALZHEIMER’S DISEASE: MODEL-AD

One of the major obstacles to developing therapies for Alzheimer's disease (AD) is the lack of predictive animal models to be used in preclinical trials. One reason for this may be that existing models are based on familial mutations, while the vast majority of the clinical population has non-familial late-onset AD (LOAD). The Model Organism Development and Evaluation for Late-onset AD (MODEL-AD) program has been established with the goals of: identifying novel genetic variants, genes and biomarkers from LOAD patient data; generating and validating mouse models with construct and face validity for LOAD; and developing a preclinical testing strategy to evaluate potential therapeutic agents in these novel models. The Bioinformatics Core has prioritized genetic risk variants for LOAD, and is working to consolidate information on human and animal model biomarkers in a publicly accessible interface. The Disease Modeling Project has engineered and is validating novel models (alongside existing models including 5XFAD, APP/PS1 and hTau) using human-relevant biomarkers, imaging methods, neuropathology and functional assays. Future models to be created will be prioritized by a Steering Committee and External Advisory Board with input from the scientific community encouraged. The Preclinical Testing Core has developed a scheme to optimally match test compounds to these new models and to evaluate test compounds through a tiered screening strategy that encompasses pharmacokinetics, target engagement, and functional outcome measures. APOE4 and Trem2 alleles are the strongest known genetic risk factors for LOAD, so we have created a novel model expressing both human APOE4 and the R47H allele of Trem2. This is being phenotypically characterized, and will serve as the standard background as we introduce additional LOAD genetic variants, including ABCA7, IL1RAP and CR1. The MODEL-AD program is generating and making available novel models of LOAD, along with related datasets. We aim to make up to 40 new models in the first five years. All new models will be made available for both academic and for-profit use from JAX, and all data will be shared via the SAGE-Synapse web portal. We seek input and collaborations from the basic research and pharma/biotech communities.