There is now considerable evidence that the blood coagulation system plays an important role in the biology of malignant tumors. This evidence has been derived from a combination of clinical, biochemical, histological, and pharmacological observations that point to the possibility of favorably affecting the course of malignant disease with agents that interfere with blood coagulation pathways. For a number of years our laboratory has used experimental models of blood-borne metastasis to study the events that follow the introduction of procoagulant-bearing tumor cells into the circulating blood. This article summarizes our experience with these models, which suggests that intravascular coagulation is a necessary prelude to lung tumor formation and that interruption of coagulation pathways in various ways may be an effective antimetastatic strategy. We have shown that anticoagulation with commonly used agents such as unfractionated heparin and warfarin (Coumadin) prevent tumor formation by limiting the ability of tumor cells to be retained in the pulmonary microvasculature. Binding of fibrin-coated tumor cells to activated platelets is essential for this retention and, therefore, treatment with potent antiplatelet agents such as abciximab is also effective. The predominant tumor procoagulant is tissue factor (TF), and direct targeting of this protein with concanavalin A, monoclonal antibodies, and tissue factor pathway inhibitor (TFPI) has provided compelling evidence that TF is an important determinant of tumor seeding in these experimental models. Collectively, our data provide strong support for the concept that some form of anticoagulant therapy would be a useful adjunct to existing cancer treatments.