PROJECT SUMMARY
Alzheimer’s disease (AD) is a progressive neurodegenerative condition with profound impacts on
memory and cognition that currently affects more than 6 million adults living in the United States.
Treatment options for Alzheimer’s remain limited, with high rates of failure in AD drug trials.
Drug repurposing to identify new uses for existing drugs offers several advantages over traditional drug
development, including higher success rates, lower costs, shorter timelines, and increased assurance
of safety. Furthermore, drugs with genetic support of efficacy have been found to be twice as successful
in clinical development.
This work leverages new developments in genetics and informatics to propose two independent
approaches to identify drug repurposing candidates for AD: (1) a virtual transcriptome approach
identifying drugs that can reverse gene expression changes observed in AD, and (2) a Mendelian
randomization approach identifying drugs acting on genes found to be causally associated with AD.
Identified repurposing candidates will be clinically validated using electronic health record (EHR) data
from Vanderbilt University Medical Center, as well as external datasets (clinical data from the NIH All
of Us Research Program Database or claims data from the Centers for Medicare & Medicaid Services).
In AIM 1, we will develop a virtual transcriptomic signature for AD using GWAS summary statistics, and
query this signature against large-scale drug perturbation databases. In AIM 2, we will use conditionally
independent genetic variants as instrumental variables to perform Mendelian randomization on AD
GWAS, and evaluate for causal relationships between actionable druggable genes and AD. Both aims
will involve further validation of repurposing candidates using real-world clinical data, which
has seldom been done before. Successful completion of this project will yield novel insights into high-
priority drug repurposing candidates for AD that can be further investigated in randomized clinical trials,
and will establish a high-throughput AD repurposing pipeline integrating genetic, -omics, and
EHR data that may be adapted to other diseases of interest.