Immunotoxins (IT) are hybrid molecules comprised of an antibody coupled chemically or genetically to a toxin or ribosome inactivating protein (RIP) such as saporin Kreitman, 1999;Frankel et al, 2000). The antibody component allows for the selective delivery of the toxin to the surface of cells bearing the target antigen. In order to kill the target cell it is an absolute requirement that the conjugate is internalized by receptormediated endocytosis and that subsequently the toxin component is routed to the appropriate intracellular compartment where translocation of the toxin component to the cytosol can take place. Once in the cytosol RIPs such as saporin catalytically inactivate 28S ribosomal RNA via a highly specific N-glycosidase enzymatic activity (Endo, 1988) that results in cessation of protein synthesis in the cell and leads to apoptotic cell death (Bergamaschi et al, 1996). Arguably, the most important factor that theoretically limits the therapeutic efficacy of immunotoxins that have no bystander effect and which therefore have to exert their cytotoxic influence directly on all cells in the tumour, is the heterogeneity of target antigen expression within the global tumour cell population. Thus, relatively small numbers of tumour cells down-regulated or negative for the target antigen would escape killing by any single immunotoxin and if these surviving cells possess growth potential this would lead to subsequent tumour re-growth. One way of overcoming this problem would be to simultaneously target against more than one target molecule on the tumour cell surface in the expectation that multiple antigen-negative tumour cells would occur with a much lower frequency than single antigen-negative cells. There would also be the added bonus that targeting against more than one antigen on the tumour cell surface would ensure delivery of greater and therefore more effective quantities of the cytotoxic agent carried by antibody to tumour cells that were multiple antigen positive. Other factors such as the cell signalling effects and recruitment of cytotoxic host effectors may also work in conjunction to achieve an improved therapeutic effect when combinations of antibody-based therapeutics are utilized. We have In contrast treatment of SCID-CEM mice with a combination of both ITs led not only to a significantly greater delay in time to leukaemia development but also in the numbers of animals remaining leukaemia free (60%). The native HB2 and OKT10 antibodies (both murine IgG 1 antibodies) exerted significant, though relatively weak therapeutic effects, probably mediated through an antibody-dependent cellular cytotoxicity (ADCC) mechanism. Moreover, there was no in vivo additivity of therapeutic effect when both antibodies were used in combination. Apparent, however, was that the combination of HB2-SAPORIN IT with OKT10 antibody led to an intermediate therapeutic effect that was significantly greater than that obtained when either was used alone but significantly less than that obtained when the two IT combination...