Development and Preclinical Characterization of an Fc-Optimized CD133 Antibody for Induction of NK Cell Reactivity Against Myeloid Leukemia

NK cells are cytotoxic lymphocytes that play a major role in anti-tumor immunity and largely contribute to the efficacy of allogenic stem cell transplantation (SCT) in leukemia. Another clinically important feature of NK cells is their ability to mediate antibody-dependent cellular cytotoxicity (ADCC) upon application of monoclonal antibodies (mAb) like Rituximab, a feature considered critical for the therapeutic success of antibody treament. Modifications of the human IgG1 Fc-part in anti-tumor antibodies lead to markedly improved capability to recruit Fc-receptor bearing effector cells as highlighted by the improved clinical efficacy of the Fc-engineered CD20 antibody Ofatumumab as compared to its unmodified counterpart Rituximab in CLL. So far, no immunotherapeutic antibodies are available for the treatment of myeloid leukemias. Here we report on the development and preclinical characterization of an Fc-optimized mAb directed towards CD133, which is expressed on a wide variety of malignant cell types. As a first step we evaluated binding of three different mouse anti-human CD133 mAbs (clones AC133, W6B3 and 293C3) to 20 primary AML and 6 primary CML samples in order to identify a clone with optimal binding characteristics. AC133 and W6B3 comparably bound to the leukemic cells in 11/20 AML and 5/6 CML samples. In contrast, binding of 293C3 was observed in 18/20 AML cases and 5/6 CML cases. Thus, 293C3 recognizes a different epitope than the other two antibody clones, which is expressed in a high proportion of myeloid leukemia cases. Accordingly, 293C3 was selected for generating chimeric mAbs with either a wildtype Fc part (293C3-WT) or a variant containing distinct modifications (S239D/I332E) to enhance its affinity to the activating Fc receptor CD16 (293C3-SDIE). The binding specificity of 293C3-WT and 293C3-SDIE was validated by FACS in analyses with CD133 transfectants and mock controls. When comparing 293C3-WT and 293C3-SDIE with regard to their immunostimulatory properties, we found that already 293C3-WT induced NK cell ADCC against primary leukemia cells as revealed by analyses of degranulation and target cell lysis. These effects were by far exceeded by treatment with 293C3-SDIE, confirming the functional relevance of the SDIE modification in its Fc part. Notably, treatment with 293C3-SDIE also enhanced the reactivity of NK cells against CD133-positive AML cells in an autologous setting. Considering the expression of CD133, among others, on healthy hematopoietic progenitor cells, we further performed colony forming unit assays with healthy bone marrow cells, which did not reveal any toxicity of 293C3-SDIE at the level of committed hematopoietic progenitor cells. Thus, 293C3-SDIE constitutes an attractive immunotherapeutic compound which we envisage in particular for the elimination of minimal residual disease in CD133 bearing leukemia, especially in the context of allogenic SCT.