Repurposing drugs for the prevention of vascular dementia: Evidence from drug target Mendelian randomization

ABSTRACT Importance Vascular dementia (VaD) is a devastating cerebrovascular disease with no disease-modifying treatments available. Repurposing existing drugs for VaD risk factors could have an important clinical impact. Objective To determine whether lipid-lowering, anti-hypertensive, or anti-inflammatory drug targets affect the risk of vascular dementia using Mendelian randomization. Design Evidence suggests that higher cholesterol and blood pressure are associated with increased VaD risk, and inflammation is thought to play a key role in pathogenesis. Two-sample MR was conducted using cis- acting genetic variants in genes encoding each drug target, and data on five VaD-related outcomes. Instrument performance was assessed with positive controls (coronary artery disease, heart failure, stroke and rheumatoid arthritis). Setting Summary: -level genetic data Participants Publicly available genetic association data from large cohorts of European ancestry. To maximize the sample size for vascular dementia risk as an outcome, we conducted a meta-analysis of case-control data from FinnGen & MEGAVCID. Exposures Genetically proxied drug effects for 46 lipid-lowering (n=17), antihypertensive (n=18), and anti-inflammatory (n=11) targets. Main Outcomes and Measures Odds ratios/betas and 95% CIs for VaD outcomes (clinical diagnosis, white matter hyperintensity volume, fractional anisotropy, mean diffusivity and lacunar stroke diagnosis) were estimated per 1-unit change in the exposure. Results For VaD risk, N=7,009 cases and N=899,672 controls were used. Neuroimaging outcome datasets included a maximum of N=50,559 participants. Beta-1 adrenergic receptor (ADRB1) was the only target for which there was consistent, albeit modest, evidence of benefit for four out of the five outcomes (clinical diagnosis: OR= 0.90, 95%CI 0.80 to 1.01, white matter hyperintensities: β= -0.03, 95%CI -0.07 to 0.00, mean diffusivity: β= -0.18, 95%CI -0.37 to 0.00, lacunar stroke: OR= 0.91, 95%CI 0.80 to 1.03). Angiotensin-converting enzyme (ACE) inhibition was suggested to increased VaD risk (OR= 1.12, 95%CI 1.01 to 1.24). There was little evidence to suggest other targets affect the outcomes. Conclusions and relevance ARDB1 antagonism may be a promising repurposing candidate for VaD. Pharmacovigilance studies are required to further examine ACE inhibitors’ potential to increase VaD risk. There is little evidence to support repurposing of many licensed lipid-lowering, antihypertensive and anti-inflammatory drugs for VaD prevention or treatment. KEY POINTS Question Can existing drugs with potential neurovascular benefits (lipid-lowering, antihypertensive, and anti-inflammatory therapies) be repurposed for vascular dementia (VaD) treatment/prevention? Findings We used large genetic datasets for VaD and its neuroimaging features in drug target Mendelian randomization. ARDB1 antagonist exposure was associated with lower VaD risk and better neuroimaging phenotypes. Conversely, exposure to ACE inhibitors may increase VaD risk. There was little evidence for other drug targets. Meaning ADRB1 antagonists could be promising candidates for drug repurposing for VaD prevention and treatment. Pharmacovigilance research is warranted to confirm or refute a link between ACE inhibitor use and VaD risk.