Abstract Body: Background: Ischemic stroke (IS) is a major global cause of death and disability, especially affecting women, who have a 20% higher lifetime risk and poorer recovery. Despite its prevalence, reliable biomarkers that predict IS risk remain limited, impeding early detection and prevention efforts.

Methods: We examined proteomic data from 450 incident IS cases, with a median time-to-event of 5.5 yrs (IQR: 2.8 yrs), and 450 matched controls in a nested case-control study within women with a median age of 65.8 (IQR: 10.1 yrs) in the Nurses’ Health Study. We used the Olink Explore HT platform to measure 5,440 proteins in plasma. Conditional logistic regression models were fitted for each protein with IS as the outcome, adjusting for established cardiometabolic, lifestyle, and hormonal IS risk factors. The false discovery rate was controlled using the q value method. Weighted co-expression network analysis (WCNA) was applied to the subset of proteins nominally associated with stroke (p value < 0.05) to identify modules of highly correlated proteins. Functional gene set enrichment analysis (FGSEA) was conducted using eight established protein sets from MSigDB to characterize biological pathways represented within these modules.

Results: Fourteen proteins were significantly associated with IS risk (q value < 0.05). A total of 274 proteins had a p value < 0.05 and were used in WCNA analyses. Four distinct protein clusters were associated with IS. The cluster most positively correlated with IS risk (r = 0.16) was characterized by proteins related to aging and inflammation. Its hub protein, follistatin-like 3, has been linked to vascular remodeling and endothelial dysfunction. Conversely, the cluster with the strongest negative association with IS risk (r = -0.14) had a metabolic and anti-inflammatory profile. Its hub protein, dentin matrix acidic phosphoprotein 1, regulates phosphate and mineral metabolism, inhibits vascular calcification, and maintains metabolic homeostasis. FGSEA revealed neutrophil degranulation as the top pathway positively associated with IS, highlighting the role of immune activation.

Conclusions: This large-scale proteomic analysis identified novel circulating proteins and networks associated with IS risk in women. These findings highlight biologically distinct metabolic pathways that may inform the development of targeted interventions to lower stroke risk. Subsequent work will validate these findings in external cohorts like the UK Biobank.