ARTEMIS: ARTificial blood vessels for Thrombosis, Endothelial Modelling, and artificial Intelligence Simulation

Cardiovascular disease is the largest cause of mortality worldwide, causing 32% of global deaths. Of these deaths, 85% are due to unwanted blood clots (thrombi) that form and block blood vessels leading to heart attacks and strokes. Therefore, targeting these unwanted thrombi with effective therapy is crucial to have a profound effect on patient wellbeing. Although a range of therapies exists to prevent or dissolve blood clots, their effectiveness can vary based on factors such as race, sex, and individual resistance to treatment, with significant side effects like bleeding. Currently, only about 20% of drugs that show promise in lab-based and animal models prove effective in clinical settings. This discrepancy is largely due to the limitations of existing preclinical models. To develop new and more effective drugs, it is essential to combine mathematical/artificial intelligence models, lab-based assays, and animal models that reliably predict clinical outcomes. Present lab-based models can replicate thrombus formation but lack key physiological features—such as the proper size, shape, and stiffness of blood vessels—needed to mimic the natural environment. While animal models offer physiological conditions, their high variability and poor correlation to human responses further reduce the predictive value of these studies. This in turn impacts the accuracy of mathematical and artificial intelligence systems that rely on such data. Our approach seeks to overcome these challenges by engineering Artificial Blood Vessels (ABVs) that closely mimic the biophysical properties of native vessels in both healthy and disease states. We will line the interior of these ABVs with vascular endothelial cells derived from arteries or veins and encase them with smooth muscle cells. This arrangement replicates the natural positioning of these critical cell types within the vascular bed, whether studying arterial or venous thrombus formation. Additionally, by introducing essential physiological factors such as the differing oxygen concentrations found in veins versus arteries, we aim to significantly enhance the replication of the natural blood vessel environment. By developing this more physiologically relevant model system, we expect to identify drug compounds that are more likely to succeed in clinical trials. In addition, our vascular model would be able to evaluate differences in patient cohort’s response to drugs and drug-potency both prior to and during clinical trials, thereby driving the best clinical outcomes for patients suffering from thrombosis. This will not only significantly reduce reliance on animal experiments, but also improve the assessment of how different patient cohorts respond to therapy. Ultimately, our work is designed to drive significant advancements in multidisciplinary science, drug development, service delivery and patient outcomes for those suffering from vascular disease