We have continued to develop and improve recombinant immunotoxins for the treatment of cancer. Our major effort has been to bring the single chain immunotoxin, B3(Fv)PE38, into the clinic. A clinical trial was initiated in February 1995 and to date ten patients have been enrolled in this trial involving patients with colon cancer, breast cancer and other epithelial cancers. Improved forms of LMB-7 have been made by stabilizing the Fv interaction in a new approach in which a disulfide bond is genetically engineered into the framework regions to hold the Fv fragments together. These disulfide-linked immunotoxins are stable at 37 degrees for 14 days in human plasma. Recombinant immunotoxins containing dsFv domains show better antitumor activity in animal models than single chain immunotoxins without an increase in nonspecific toxicity. We are also developing smaller recombinant immunotoxins which should penetrate tumors better and be less immunogenic. We have also humanized the Fv portion of the B3 antibody as a first step in making less immunogenic recombinant immunotoxins. We have also used the dsFv approach to make a disulfide-linked T cell receptor which is extremely stable, binds specifically to MHC class II peptide complex and could be useful for structural studies. We have identified several B cell epitopes in PE and are constructing mutant molecules in which these epitopes have an altered sequence in order to make less immunogenic immunotoxins. We have made a dsFv fragment of the anti-Tac antibody, showed it can be radiolabeled efficiently, and will rapidly localize in IL2 receptor- bearing tumors growing in nude mice. This agent could be useful for the diagnosis and perhaps treatment of IL2 receptor-bearing malignancies. We are utilizing phage antibody display to try and improve existing antibodies and to develop new antibodies. We have modified the method so that disulfide-linked Fvs are displayed on the surface of phage and these are considerably more stable than single chain Fvs and should produce different types of antibodies. We have begun to investigate possible mechanisms of immunotoxin resistance and isolated several cDNAs which when expressed at high levels make MCF7 cells resistant to Pseudomonas toxin, diphtheria toxin and tumor necrosis alpha and beta. The mechanism by which these toxins act is under study.
