Metastasis is the principal cause of death in breast cancer patients. New and improved treatments for eradicating micrometastases are needed. To this end, a novel alpha-emitting protein construct, 213Bi-labelled plasminogen activator inhibitor type-2 (PAI-2) (alpha-PAI-2), was evaluated in vitro. This construct exploits: (a) the overexpression of the cell-surface receptor bound urokinase plasminogen activator (uPA) in the metastatic spread of breast cancer cells; (b) the binding and inhibition of receptor-bound uPA by PAI-2; and (c) the high cytotoxicity of alpha radiation. High labeling efficiencies and stability of 213Bi bound to human recombinant PAI-2 conjugated with cyclic diethylenetriaminepentaacetic acid anhydride were achieved (greater than 90%). The uPA inhibitory activity of the chelated PAI-2 was maintained as determined by complex formation with uPA and by inhibition of uPA activity. Furthermore, the reactivity of alpha-PAI-2 was confirmed in a cell assay as this construct was highly cytotoxic to breast cancer cell lines that express active, receptor bound uPA. The specificity of alpha-PAI-2 targeting was shown using several controls. Firstly, an active uPA blocking agent that limits PAI-2 binding significantly improved cell survival by a factor greater than three. Secondly, a non-specific alpha-BSA construct had minimal cytotoxic effect. Moreover, alpha-PAI-2 was not cytotoxic to freshly isolated normal human leukocytes, confirming that cells which do not contain active, receptor bound uPA cannot be targeted by alpha-PAI-2. In conclusion, we have validated, in vitro, the potential of alpha-PAI-2 as a novel therapeutic agent for breast cancer.