41Immune checkpoint therapy has produced remarkable improvements in the outcome for 42 certain cancers. To broaden the clinical impact of checkpoint targeting, we devised a 43 strategy that couples targeting of the cytokine-inducible SH2-containing (CIS) protein, a 44 key negative regulator of interleukin (IL)-15 signaling, with chimeric antigen receptor 45 (CAR) engineering of natural killer (NK) cells. This combined strategy boosted NK cell 46 effector function through enhancing the Akt/mTORC1 axis and c-MYC signaling, 47 resulting in increased aerobic glycolysis. When tested in a lymphoma mouse model, this 48 combined approach improved NK cell anti-tumor activity more than either alteration 49 alone, eradicating lymphoma xenografts without signs of any measurable toxicity. We 50 conclude that combining CIS checkpoint deletion with CAR engineering promotes the 51 metabolic fitness of NK cells in an otherwise suppressive tumor microenvironment. This 52 approach, together with the prolonged survival afforded by CAR modification, represents 53 a promising milestone in the development of the next generation of NK cells for cancer 54 immunotherapy. 55 56 57 Introduction 58
59Adoptive cellular therapy using autologous T cells transduced with a chimeric antigen 60 receptor (CAR) has proved to be a powerful approach in the treatment of human cancers, 61 especially B cell leukemias and lymphomas. 1,2 However, ongoing efforts to consolidate 62 and extend these gains face a number of obstacles: (i) uncoupling cytotoxicity against 63 tumor cells from systemic toxicity, (ii) finding ways to reduce target antigen negative 64 relapses, (iii) overcoming the inhibitory effects of checkpoint molecules in the infused 65 immune effector cells, and (iv) developing universal off-the-shelf cell therapy products 66 that avoid the logistic hurdles of generating autologous products, as well as several of the 67 pitfalls of allogeneic T-cell therapy, such as graft-versus-host disease (GvHD). 3 68 69 Natural killer (NK) cells are attractive candidates for the next wave of effective cancer 70 immunotherapies. They mediate potent cytotoxicity against a range of tumor cells 4 and, 71 unlike T cells, lack the capacity to induce GvHD in the allogeneic setting. 5 Moreover, 72 their ready availability in high numbers from various sources, such as umbilical cord 73 blood (CB), boosts their potential as an off-the-shelf product for widespread clinical 74 scalability. 6,7 One of the most intriguing recent advances in the development of NK cell-75 based immunotherapy was the demonstration that genetic modification of these cells to 76 express a CAR can enhance their effector function. 8 This led to the realization that one 77 might overcome some of the limitations of NK cell immunotherapy in cancer, such as the 78 lack of antigen specificity and poor persistence, by exploiting current genetic engineering 79 tools. In our experience, transducing NK cells with a retroviral vector encoding a CD19-80 4 specific CAR, the interleukin (IL)-15 cytokine and ...