Project summary
Alzheimer’s disease (AD) and related dementias (ADRD) are heterogeneous neurodegenerative disease
devastating for patients, families, caregivers, and society. Patients demonstrate various progressive decline
patterns in different cognitive domains such as memory, language, executive function, visuospatial function,
personality and behaviors. Each progression trajectory is associated with specific genotypes, molecular marker
features, brain imaging patterns, risk factors, and needs in drug management and other cares. Managing ADRD
is challenging due to the heterogeneity and complexity in the disease itself, in the medical comorbidities, and in
social determinants of health (SDoH). Specifically, there is an urgent need in managing medicine safety for
patients who are 1) demonstrating different cognitive impairment patterns and cognitive declining rates, 2) at
different stages of the disease progression trajectories, 3) demonstrating different clinical, molecular, and genetic
markers, 4) with different comorbid conditions, and 5) showing health disparity related with socioeconomic status
and access to healthcare and other community resources. In the Parent R01 project recently funded by National
Library of Medicine (R01LM013771), we have been developing DEPOT (DisEase PrOgression Trajectory), a
generalizable clinical informatics system to reveal the heterogeneous health trajectories of complex chronic
diseases and identify adverse effects of drug-drug interactions (DDIs) in both the general population and
trajectory-specific subpopulations using longitudinal electronic health records (EHR), with chronic kidney disease
(CKD) as the disease model and acute kidney disease (AKI) as the drug adverse event. The Parent Project is
based on the IUSM longitudinal EHR collection (cohort size: 82million), which is composed of the Optum
Clinformatics™ claim data and the Indiana Network for Patient Care (INPC) Research Database. We propose
to extend the Parent Proposal and develop a tailored DEPOT system to address the urgent need in precision
drug management for ADRD patients. We hypothesize that there are different ADRD progression paths which
are: 1) driven by different pathogenic mechanisms, 2) susceptible to different nephrotoxic drugs and DDIs, and
3) identifiable by longitudinal EHR data. The goal of this work is to 1) establish EHR-based ADRD progression
trajectories and 2) learn actionable knowledge to prevent drug interaction induced AKI. The multi-specialty team
proposes to: Aim 1. Establish ADRD progression trajectories using graph artificial intelligence model, Aim 2.
Identify a precision medicine approach to protect against DDI-induced AKI with a special conscious on
patients with ADRD. The success of the proposed development of the DEPOT model for ADRD will generate
novel knowledge about ADRD health trajectories and nephrotoxic drug interactions, bridging gaps between rich
longitudinal EHR data and decision support for precision medicine in ADRD. This work will shift paradigms of big
data and complex disease research, enabling EHR data to become part of daily ADRD management.