ABSTRACT The parent R01 project, funded in 2016, aims to address the unmet medical need of providing individually tailored therapy for multiple sclerosis (MS), given the growing expansion of the approved MS disease- modifying treatments (DMTs). Specifically, the parent project aims to produce an analytic approach capable of identifying MS disease activity in relation to treatment history using EHR data and integrate with genomics profile to develop a predictive model of therapeutic response to commonly prescribed DMT. Late-onset Alzheimer’s disease (AD) is the most common cause of dementia and neurological disability in the aging population. People with AD experience variable trajectories of cognitive and functional decline, resembling the variable trajectories of neurological decline in people with MS. While the choices of DMTs for AD are few and while more promising options are only beginning to emerge, real-world data such as EHR data are gaining importance in the drug approval process. For the supplemental project, we propose to deploy the analytical approaches that we developed for MS to ascertain individualized disease trajectory and treatment response to existing drugs in people with late-onset AD, using EHR data and linked registry data. We will build on existing collaboration with Dr. Tianxi Cai (Harvard, co-I on the parent R01) who is a leading expert on EHR data analysis and new collaboration with Dr. Howard Aizenstein (co-I, University of Pittsburgh) who provide domain expertise in AD and cognitive aging as well as Dr. Jonathan Silverstein (co-I, University of Pittsburgh) who will provide critical support for the EHR data generation. For data source, we will leverage the growing EHR data warehouse at the University of Pittsburgh Medical Center, which is linked to a well-established AD cohort research registry. Extending the parent R01 project, we will test the hypothesis that meaningful phenotypes of AD disease trajectory can be extracted from EHR data to inform treatment response in the supplemental project. We will accomplish two supplemental aims: (1) leverage EHR data to ascertain the individualized trajectory of neurological impairment in AD; (2) predict response to existing AD drugs using EHR data. The objective of the supplemental project is to identify individualized AD disease trajectory in relation to treatment history using EHR data and enable future assessment of AD drug efficacy using real-world data and pharmacogenomics.