Plasma proteomics model predicts cancer-associated thrombosis with 0.84 C-statistic, identifying IL-17-driven endothelial activation

Venous thromboembolism (VTE) remains a significant cause of morbidity and mortality in patients with cancer. Current risk stratification tools, such as the Khorana score, often fail to reliably predict VTE development, leading to suboptimal prophylactic strategies and increased patient risk. There is a critical need for more accurate predictive models that can identify high-risk individuals and uncover underlying mechanistic pathways. This study addresses this gap by leveraging high-throughput plasma proteomics to develop a superior predictive model and explore novel therapeutic targets for thrombo-inflammatory disease.

Researchers conducted a high-throughput proteomic analysis of 1105 plasma proteins from peripheral blood samples of patients with newly diagnosed lung or gastric cancer. These patients were prospectively monitored for VTE development. Using a Bayesian probabilistic machine learning approach, a predictive model was developed incorporating 11 protein biomarkers and 5 clinical parameters (age, sex, history of VTE, body mass index, and hemoglobin). The model's predictive power was then orthogonally validated in an external placebo cohort from a phase 3 trial. Further mechanistic studies involved CD200R1-deficient mice, assessing prothrombotic markers and inflammation, and administering anti-IL-17A antibodies to normalize thrombin-antithrombin complexes.

The developed plasma proteomics model significantly outperformed the standard Khorana prediction score, achieving a c-statistic of 0.84 (0.79 to 0.90) compared to 0.36 (0.27 to 0.45). Reduced plasma concentrations of CD200 receptor 1 (CD200R1), an immune checkpoint receptor, strongly contributed to the model and correlated with higher D-dimer concentrations and increased thrombosis risk. Mechanistic investigation in CD200R1-deficient mice revealed a prothrombotic state, characterized by elevated thrombin-antithrombin complexes, increased interleukin-17A (IL-17A), and endothelial inflammation. > Administration of anti-IL-17A antibodies to CD200R1-deficient mice normalized thrombin-antithrombin complexes in vivo, highlighting a potential therapeutic intervention. A meta-analysis of human COVID-19 studies further supported this, showing reduced pulmonary thromboembolism in patients receiving anti-IL-17A antibodies.