Computational Modeling of Pro-inflammatory Cytokine-Enhanced Blood Coagulation

The interplay between inflammation and coagulation is a central driver of thrombotic risk across various diseases. While mathematical models of blood coagulation are well established, there remains a critical gap in quantitative frameworks that capture inflammation-induced hypercoagulability. In this study, we develop a mathematical model that explicitly simulates the interaction between pro-inflammatory cytokines and the coagulation cascade. The model incorporates key mechanisms, including: (a) up-regulation of tissue factor by interleukin-1𝛽, interleukin-6, and tumor necrosis factor-α; (b) suppression of natural anticoagulants, namely, antithrombin III and tissue factor pathway inhibitor, by interleukin-6 and tumor necrosis factor-α; and (c) feedback amplification of pro-inflammatory cytokines by thrombin. By encoding the bidirectional feedback between inflammatory and coagulation pathways, the model captures essential features of inflammation-driven hypercoagulability and enables systematic quantification of how variability in inflammatory extent and duration result in heterogeneous thrombin generation (TG) dynamics. To evaluate its effectiveness, we integrate the model with TG assays and apply it to virtual patient cohorts representing 4 clinically distinct conditions: COVID-19, sickle cell disease, type 2 diabetes mellitus and hemophilia A. Model simulations predict that disease-specific inflammatory environments induce distinct shifts in TG dynamics. In COVID-19 and type 2 diabetes mellitus, elevated cytokine levels lead to shortened lag times and increased thrombin peak, whereas in sickle cell disease, shortened lag times are accompanied by a reduced thrombin peak. These effects are strongly modulated by both cytokine concentration and duration of exposure. These results demonstrate that the proposed computational model augments conventional TG assays by mechanistically linking inflammatory signaling to disease-specific coagulation responses. Collectively, the proposed computational framework extends conventional TG assays by considering the interplay between inflammation and coagulation, thereby providing a mechanistic exploratory platform for generating testable hypotheses regarding disease-specific thromboinflammatory regulation. This framework lays a foundation to support future clinical prediction or individualized therapeutic decision-making after validation using prospective patient-level longitudinal datasets.