Transforming growth factor-β (TGF-β) controls tissue homeostasis and mediates the repair response to tissue injury. While tumors espape from TGF-β’s homeostatic function, many metastatic cancers coopt the tissue repair function to enhance their invasive/metastatic phenotype. These effects are due to an altered responsiveness of the tumor cells themselves (tumor cell autonomous effects) or to actions of tumor-associated TGF-β on the supporting host cell infrastructure. This discovery has resulted in great enthusiasm for developing TGF-β antagonists (TβA) for the treatment of metastatic cancer. Proof of concept has been provided by preclinical studies utilizing TGF-β neutralizing antibodies, TGF-β antisense molecules, soluble TGF-β receptors (TβR), and selective and potent chemical inhibitors of the TβR kinases. In vivo, TβA appears to impact on both cell autonomous and host cell effects of TGF-β. However, their antitumor activity has been modest in magnitude and limited to specific models, suggesting that only select tumors may be clinically susceptible to treatment with TβA. Moreover, as the oncogenic role of TGF-β signaling appears to come into play at a relatively late stage of tumor progression, blocking this pathway will likely have to be combined with inhibitors of oncogenes that drive tumor growth.