ABSTRACT Alzheimer’s disease (AD) is characterized by progressive loss of synapses and neurons along with amyloid and Tau pathologies. Although amyloid-β (Aβ)-reducing immunotherapy received full approval by the FDA for patients with mild cognitive impairment (MCI) or mild AD, developing future treatments against AD at all stages is an urgent priority. One of our recent 12 publications described our cohorts of patients with MCI, mild, moderate, or severe AD within the Department of Veterans Affairs (VA) Healthcare System. Using electronic health records (EHRs), we reported a slower progression to AD among veterans on prescribed vasodilators, including anti-hypertensive angiotensin converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs). We compared a spectrum of anti-hypertension, anti-diabetic or anti-hypercholesterolemia medications and reported a reduced risk of AD onset among vasodilator users (e.g., Valsartan, Carvedilol, Losartan) and an increased risk of AD onset among vasoconstrictor users (e.g., Propranolol). We also reported that ACEIs reduced phosphorylated Tau (pTau) in mouse brains. We further reported that Fasudil, a specific vasodilator and a Rho- associated coiled-coil kinase (ROCK) 1/2 inhibitor, decreased pTau in human neuro-spheroids. We therefore propose to perform data mining of 25 million veterans’ 76 billion medical records from the VA Corporate Data Warehouse and to test the clinical effectiveness of more potent vasodilators, such as a selective ROCK2 inhibitor Belumosudil, in delaying AD onset among veterans. We will also test if Belumosudil suppresses neurodegeneration in AD mouse models. Our access to VA EHR as well as our expertise in development and multidisciplinary analysis of Familial AD mouse models place us in a unique position to test the hypothesis that vasodilation alleviates cognitive impairment in AD patients as well as memory deficits and neurodegeneration in AD mouse models. In this competing continuation application, we propose to evaluate the effects of more potent vasodilators on cognitive impairment of AD patients and AD mouse models (Aim 1). We will perform big data analysis of EHR to collect vasodilators that are associated with delayed onset of AD, compared to the benchmark ARB drug Valsartan. We will also test whether vasodilators, such as Belumosudil, alleviate neurodegeneration and memory impairment in AD mouse models. We will also explore molecular mechanisms of vasodilators using single nucleus RNAseq and proteomic analyses of AD postmortem brains and mouse models (Aim 2). The significant impact of the current study is that these small molecules derived from big data mining and validated in AD mouse models may be fast-tracked to next generation safe therapies of AD.