Ovarian cancer (OvCa) is the most fatal gynecological malignancy due to delayed clinical presentation, passive metastasis, frequent tumor recurrence, and lack of effective targeted therapies. OvCa tumors predominantly shed as single cells or spheroids and disseminate throughout the peritoneal cavity. Spheroids are considered critical metastatic units that contribute to disease recurrence and chemoresistance. Despite recent advances in treatment strategies, there has been little improvement in patient survival. There is an urgent need to develop a better understanding of molecular mechanisms that are fundamental for OvCa dissemination, in order to provide alternative therapeutic options for long-term remission. Matriptase, a membrane-anchored serine protease (MASP), and its substrate protease-activated receptor-2 (PAR-2) both exhibit elevated expression in OvCa compared to normal ovary tissues which worsens patient survival, suggesting a critical role for activation of this pathway in OvCa progression. Overactive matriptase, induced by an imbalanced ratio with its cognate inhibitor hepatocyte growth factor activator inhibitor-1 (HAI-1) has been implicated in disruption of barrier integrity and epithelial carcinogenesis, although the mechanism is not known. This study, demonstrates a functional role for an imbalanced matriptase:HAI-1 ratio in OvCa progression, and identifies a matriptase/PAR-2/PI3K/Akt/MMP-9/E-cadherin signaling axis that disrupts cell-cell interactions and promotes formation of pro-metastatic loose spheroids in vitro, in in vivo orthotopic xenograft models, and in patient-derived tumor cells. This matriptase/PAR-2 pro-metastatic signaling is in direct contrast to another MASP family member, testisin, suggesting a novel mechanism of biased agonism that orchestrates OvCa dissemination. Since over-activity of MASPs is associated with advanced OvCa, we have developed a MASP-activated pro-drug based on a re-engineered anthrax toxin (PAS:LF) that is selectively activated by zymogen-activating proteases on the tumor cell surface to inhibit tumor cell-survival pathways. This therapeutic strategy is radically different from anti-proliferative mechanisms of standard chemotherapeutic agents. We have demonstrated anti-tumor efficacy in a range of OvCa cells and spheroids in vitro, in various in vivo orthotopic xenograft models of OvCa dissemination, in patient ascites-derived tumor cells and in patient-derived xenograft models in vivo, with no off-target adverse effects. Clinical translation of these preclinical findings could establish PAS:LF as a promising treatment strategy to improve patient outcomes.