Abstract 13089: Plasma Proteomics and Risk of Incident Ischemic Stroke: Meta-Analysis of the Atherosclerosis Risk in Communities (ARIC) Study and the Cardiovascular Health Study (CHS)

Background: Proteomic profiling can screen large numbers of proteins to discover novel biomarkers related to ischemic stroke. However, prior studies have used a targeted approach or assessed a limited number of proteins (<1300). Using data from the Atherosclerosis Risk in Communities (ARIC) study and the Cardiovascular Health Study (CHS), we examined the prospective association between 4954 plasma protein targets and incident ischemic stroke. Methods: Participants free of prevalent stroke at the time of proteomic profiling were included. Proteomics data were obtained from the 1990-92 exam in ARIC and the 1992-93 exam in CHS. Plasma proteins were quantified using the aptamer-based SOMAScan platform. Ischemic stroke events were ascertained and adjudicated through 2019 in ARIC and June 30, 2015 in CHS. Cox proportional hazards models adjusting for demographics, ischemic stroke risk factors, and estimated glomerular filtration rate were used. ARIC and CHS results were combined using fixed-effects meta-analysis. A Bonferroni-corrected p<1.01e-5 was considered significant. Results: Median follow-up time in ARIC and CHS were 25 and 12 years, respectively. Among 10,856 ARIC participants (mean age 57 years, 56% female, 23% Black), 890 had an incident ischemic stroke. Among 3,283 CHS participants (mean age 74 years, 62% female, 18% Black), 507 had an incident ischemic stroke. In the meta-analysis, 44 proteins were significantly associated with incident ischemic stroke, of which 32 have not been previously reported ( Figure ). Of the 32, amyloid-like protein and kit ligand were the most strongly associated with ischemic stroke; each SD increment in log 2 (amyloid-like protein) and log 2 (kit ligand) were associated with 20% and 19% lower risk of ischemic stroke. Conclusion: In a meta-analysis of two community-based cohorts, we identified 32 novel protein associations with incident ischemic stroke. Further research is needed to determine causal associations.