Efficacy of an anti-CD5-ricin A chain immunoconjugate in an improved human peripheral blood lymphocyte--reconstituted severe combined immunodeficient mouse model.

Severe combined immunodeficient mice reconstituted with human peripheral blood lymphocytes (hu-PBL-SCID mice) were used to evaluate in vivo efficacy of a mouse IgG1 monoclonal antibody (mAb)-ricin toxin A chain immunoconjugate (H65-RTA) directed against the CD5 cell surface antigen present on human T-cells. Initial studies demonstrated that plasma levels of human soluble interleukin-2 receptor (sIL-2R) are predictive of human T-cell engraftment in spleens and blood of SCID mice transplanted with human PBL. Therefore, chimeric mice with detectable plasma levels of human sIL-2R were used in subsequent studies. Systemic injection of such mice with H65-RTA resulted in a significant depletion of human T-cells from spleens, blood and bone marrow, and a decrease in plasma levels of human sIL-2R as compared to vehicle-treated control animals. The effect of H65-RTA was dose-dependent, treatment schedule-dependent, and mAb-specific, as an isotype-, linker- and toxin-matched immunoconjugate of irrelevant binding specificity was not efficacious. Moreover, human T-cells remained depleted from SCID tissues for at least 10 days after cessation of H65-RTA treatment, indicating that the cells were killed by the immunoconjugate. Unconjugated H65 mAb and an H65-derived F(ab')2-RTA conjugate, but not unconjugated F(ab')2, were also efficacious, suggesting that the Fc portion of the mAb and the toxin moiety may both play a role in the mechanism of human T-cell depletion by H65-RTA in this model. Results indicate that the hu-PBL-SCID mouse model can be used to compare in vivo efficacy of immunosuppressive agents specifically directed against human T-cells.