Venous thromboembolism (VTE) represents a major cause of morbidity and mortality, yet its underlying biological mechanisms remain incompletely characterized. In a prospective multi-cohort proteomics analysis published in Circulation, baseline plasma protein profiles were evaluated in relation to incident noncancer VTE.

The discovery meta-analysis included participants from ARIC (Atherosclerosis Risk in Communities), CHS (Cardiovascular Health Study), and MESA (Multi-Ethnic Study of Atherosclerosis), with replication in HUNT (Trøndelag Health). Among 20,737 adults followed for 10 to 29 years, 1,371 incident noncancer VTE events were identified. Baseline plasma levels of approximately 5,000 to 7,000 proteins were quantified using high-throughput aptamer-based proteomics. Cox proportional hazards regression estimated associations between protein biomarkers and VTE risk. External validation was conducted in 39,097 participants from UK Biobank, among whom 783 incident noncancer VTE events occurred.

Twenty-three proteins met false discovery rate–adjusted significance and were replicated in HUNT; 15 were newly associated with VTE, and three (TAGLN, SVEP1, TIMP4) exceeded Bonferroni correction. Mendelian randomization supported a possible causal association for TIMD4. As an observational cohort analysis, findings reflect associations rather than confirmed causality.