Genetically Engineered Pluripotent Cell-Derived Natural Killer Cell Therapy Provides Enhanced Antibody Dependent Cellular Cytotoxicity Against Hematologic Malignancies and Solid Tumors in Combination with Monoclonal Antibody Therapy

Antibody-dependent cellular cytotoxicity (ADCC) is a key effector mechanism mediated by natural killer (NK) cells that has been shown to significantly contribute to the anti-tumor effect of therapeutic monoclonal antibodies (mAbs) including rituximab (anti-CD20), cetuximab (anti-EGFR) and trastuzumab (anti-ErbB2). ADCC is induced by the crosslinking of the FcγRIIIA (CD16a) receptor expressed on NK cells with cell surface-bound IgG antibodies. Upon NK cell activation, CD16a is proteolytically cleaved from the surface of NK cells, a process that decreases ADCC. There exist two allelic variants of CD16a, with the high-affinity CD16a 158V allele found in about 10% of the population. Additionally, we previously identified a single substitution mutation that renders CD16a resistant to activation-induced cleavage. Engineering of human induced pluripotent stem cells (hiPSCs) with a non-cleavable version of the high-affinity CD16a 158V variant, followed by differentiation of hiPSCs to NK cells, enables us to develop an off-the-shelf NK cell therapy with improved ADCC properties when combined with monoclonal antibody therapy for treatment of both hematologic malignancies and solid tumors. We evaluated cell surface phenotype, as well as the in vitro and in vivo tumor-killing function, of these high affinity, non-cleavable CD16a iNK (hnCD16-iNK) cells alone and in combination with anti-CD20 mAb against multiple B-cell leukemia and lymphoma targets. In contrast to the endogenous expression of CD16a on peripheral blood NK (PB-NK) cells, which is variable depending on the donor, the derivation of NK cells from a clonal hiPSC line engineered to express hnCD16a produced a uniform population of CD16+ NK cells (>95% CD16+ for hnCD16-iNK vs. 40-60% CD16+ PB-NK cells). These hnCD16-iNK cells are phenotypically mature, highly functional in vitro, and exhibit enhanced ADCC when combined with anti-CD20 mAb against multiple lymphoma targets, including Raji, Nalm6 and Jeko-1. Both direct and CD16-mediated activation of hnCD16-iNK cells result in production of the inflammatory cytokines tumor necrosis factor alpha (TNFα) and interferon gamma (IFNγ). Furthermore, in contrast to PB-NK cells, hnCD16-iNK cells proved to be highly resistant to activation-induced cleavage of CD16 in response to Raji cells coated with anti-CD20 (68±26% cleavage for PB-NK vs. 8.0±4.2% cleavage for hnCD16-iNK cells). To extend these findings to an in vivo model, we treated Raji-lymphoma bearing mice with either anti-CD20 mAb (rituximab) alone or in combination with either PB-NK cells or hnCD16-iNK cells. Combination treatment with hnCD16-iNK cells enhanced the efficacy of anti-CD20 compared to mAb treatment only, and provided improved tumor regression compared to mAb plus PB-NK (p<0.001; tumor burden measured six days after NK cell injection). Additionally, mice that received combination treatment with hnCD16-iNK cells had significantly better survival than mice treated with mAb alone (median survival rate of 39 days for rituximab treatment only group vs. >50 days for combination treatment with hnCD16-iNK). Similar results were seen with the combination of anti-HER2 antibody and hnCD16-iNK cells in an SKOV3 ovarian cancer mouse model. In summary, these data demonstrate that combination therapy hnCD16-iNK cells with monoclonal antibody therapy is highly functional against various tumor cell lines in vitro and in vivo, and justify further investigation into their clinical application to enhance ADCC and the anti-tumor effects of therapeutic mAbs.