ERBB2/HER2-SPECIFIC NATURAL KILLER CELLS FOR ADOPTIVE IMMUNOTHERAPY OF GLIOBLASTOMA

BACKGROUND: While EGFRvIII appears a logical target for immunotherapy, only a subpopulation of tumor cells express EGFRvIII and immune escape has been demonstrated. ErbB2 is overexpressed in a substantial proportion of glioblastomas and has been successfully utilized in immunotherapies. Natural killer (NK) cells are the first line of defense against viral infections and malignant cells. The continuously growing cytotoxic cell line NK-92 holds promise for cancer immunotherapy. Safety of infusion of high doses of NK-92 was established in previous phase I clinical trials utilizing irradiated cells to prevent permanent engraftment. METHODS: To provide NK-92 cells with pre-determined tumor-cell specificity, we generated a lentiviral second generation chimeric antigen receptor (CAR) construct (5.28.z) employing the ErbB2 (HER2)-specific scFv(FRP5) antibody fragment for target cell recognition, and human CD28-CD3 ζ as a composite signaling moiety. An ErbB2-specific single cell clone (NK-92/5.28.z) was isolated, which showed high and selective cytotoxicity towards ErbB2-expressing tumor cells of various origins in vitro. We evaluated the cytotoxicity of NK-93/5.28.z cells against a panel of glioblastoma cell lines and primary glioblastoma cultures with different levels of Erb2 expression in vitro and in vivo. RESULTS: To provide NK-92 cells with pre-determined tumor-cell specificity, we generated a lentiviral second generation chimeric antigen receptor (CAR) construct (5.28.z) employing the ErbB2 (HER2)-specific scFv(FRP5) antibody fragment for target cell recognition, and human CD28-CD3 ζ as a composite signaling moiety. An ErbB2-specific single cell clone (NK-92/5.28.z) was isolated, which showed high and selective cytotoxicity towards ErbB2-expressing tumor cells of various origins in vitro. We evaluated the cytotoxicity of NK-93/5.28.z cells against a panel of glioblastoma cell lines and primary glioblastoma cultures with different levels of Erb2 expression in vitro and in vivo. CONCLUSIONS: Adoptive immunotherapy with application of ErbB2-specific NK-92/5.28.z cells may be a promising new immunotherapy approach for ErB2 positive glioblastoma. A phase I trial for glioblastoma patients with local injections of NK-92/5.28.z cells is in preparation. SECONDARY CATEGORY: Preclinical Experimental Therapeutics.