OBJECT: Because the prognosis for patients with glioblastoma multiforme (GBM) remains poor, investigators have focused on developing new and more effective treatment modalities. Targeted toxins represent a new class of compounds composed of a potent protein toxin and a carrier ligand that will recognize cell surface antigens located on target tissue. A recombinant fusion protein was created that contains the translocation and catalytic portions of diphtheria toxin that are responsible for cell entry and killing, respectively, fused to the noninternalizing aminoterminal fragment portion of human plasminogen activator. This diptheria toxin-uPA fusion protein (DTAT) has the advantage over other fusion proteins of targeting malignant glioma cells and the endothelial cells of the neovasculature that express the urokinase-type plasminogen activator receptor (uPAR). Another protein, DTAT13, was synthesized to target uPAR on the neovasculature and the uPAR and interleukin-13 receptor-expressing GBM cells. The authors describe the in vitro and in vivo efficacy of DTAT and DTAT13 against GBM. METHODS: The in vitro cytotoxicity of DTAT and DTAT13 was measured using cell proliferation assays. In vivo studies were performed in which DTAT, DTAT13, or a control protein was injected directly into GBM flank tumors in athymic nude mice. Tumor volume was assessed over time and analyzed using the Student t-test. The systemic organ effects of DTAT and DTAT13 were examined functionally and histologically in tumor-free C57BL/6 mice. In vitro, DTAT and DTAT13 were found to be highly potent and selective against U118MG, U87MG, and U373MG GBM cell lines and human umbilical vein endothelial cells. In vivo, DTAT and DTAT13 both caused a statistically significant (p < 0.05) regression of U87MG GBM flank tumors when administered every other day at 10 mg/day for five doses. No tumor regression was seen in control animals. Both DTAT and DTAT13 had little effect on histological findings in the liver, kidney, spleen, and lungs. Serum analysis did not demonstrate an effect on blood urea nitrogen levels, but liver alanine aminotransferase levels rose to statistically significant (p = 0.046) but not life-threatening levels. Also, DTAT13 was less toxic than DTAT in studies of mortality rates. CONCLUSIONS: Both DTAT and DTAT13 might have potential for clinical application against GBM because of their ability to target both the tumor cells and neovasculature simultaneously with an absence of serious systemic side effects. The discovery that DTAT13 was less toxic than DTAT indicated that the bispecific fusion protein might target a broader subset of antigenetically diverse patients with tumors while reducing the systemic exposure to toxin that would be necessary if two agents were administered separately.