Epithelial ovarian cancer cells spread by two major pathways. One is by exfoliation of tumor cells from the ovarian surface, with resulting implants on peritoneal surfaces such as omentum, diaphragm, and bowel serosa. The second pathway of spread of epithelial ovarian cancer is that of invasion into lymphatic channels, with involvement of the retroperitoneal lymph nodes. Invasion of tumor cells appears to result from a deregulation of the normal processes that govern physiologic controlled invasion. An example of physiologic invasion is that of trophoblastic implantation of the endometrium followed by invasion to access the maternal blood supply in the uterus. The normal invasive process is controled by a balance between protease activity and that of their inhibitors at the level of the individual cell. For the large majority of tumors, extracellular matrix degrading proteases have been shown to have an important role in tumor invasion and metastases. In epithelial ovarian cancer, the major focus has been on the activities of the serine protease urokinase plasminogen activator (uPA) (1 -4 ), and the matrix metalloproteinases (MMP) (5 ,6 ). The cysteine proteases, such as cathepsin B, and aspartic proteases, such as cathepsin D, also play a role in invasive-ness of ovarian cancer cells. Each class of proteases contributes to the process of invasion, and cooperates with each other to further the optimal degradation of the extracellular matrix by ovarian cancer cells (7 ). Urokinase activates plasminogen to plasmin, which activates both pro-uPA and latent MMPs, and cleaves the MMP inhibitor TIMP-2. Moreover, cathepsin B, which is expressed by ovarian cancer cells, appears to facilitate the action of uPA, most likely by activating pro-uPA (8 ).