Modulation of Natural Killer Cell Activity in the Setting of Oncolytic Virotherapy and with a Chimeric Antigen Receptor

Natural killer (NK) cells are innate lymphocytes that can rapidly eradicate tumor cells, especially those lacking MHC Class I molecules.  NK cells can also rapidly eradicate herpes virus-infected cells.  We designed an oncolytic herpes virus (oHSV) to selectively infect, replicate within, and lyse glioblastoma (GBM), a devastating brain tumor with a median survival of only 15 months following diagnosis. We have shown that the rapid influx of NK cells limits oHSV efficacy in GBM as they impede oHSV replication and spread [Alvarez-Breckenridge et al., Nat Med, 2012, 18(12):1827-34]. In the current study, we developed NK cell-based novel GBM therapies by decreasing the brain influx of NK cells to enhance the efficacy of oHSV, while arming NK cells in the brain with a chimeric antigen receptor (CAR) that targets both the wild-type EGFR and its mutant form EGFRvIII, two GBM tumor-associated antigens. We then investigated the synergistic effects between EGFR-CAR NK cells and oHSV.

Transforming growth factor (TGF)-β is a potent immunosuppressive cytokine of NK cells [Yu et al, Immunity, 2006, 24(5):575-90]. We first determined if oHSV efficacy for treatment of GBM would be augmented by inhibiting anti-oHSV activity of NK cells with TGF-β pre-treatment. In vitro, NK cells pre-treated with TGF-β displayed less cytolytic capacity against oHSV-infected GBM cell lines and patient-derived GBM stem-like cells.  In viral replication assays, co-culturing oHSV-infected GBM cells with NK cells pre-treated with TGF-β significantly increased virus titers. In an immunocompetent syngeneic GBM mouse model, administration of TGF-β to GBM-bearing mice prior to oHSV injection significantly inhibited intracranial infiltration and activation of NK cells (P < 0.05). In orthotopic human GBM xenograft mouse models and in syngeneic GBM mouse models, TGF-β treatment in vivo prior to oHSV therapy resulted in inhibition of NK cell infiltration, suppression of tumor growth and significantly prolonged survival of GBM-bearing mice (P < 0.05). Furthermore, depletion of NK cells incompletely blocked the positive effects of in vivo treatment of GBM with TGF-β on survival, suggesting that TGF-β may also directly act on other innate immune cells such as macrophages/microglia. These data demonstrate a single dose of TGF-β prior to oHSV administration enhances anti-tumor efficacy for GBM at least in part through the transient inhibition of the innate immune responses to oHSV infection.

We next investigated whether NK cell activity could be enhanced to more directly target brain tumors while sparing eradication of oHSV. We therefore infected both human NK-92 cells and primary human NK cells to express the second generation CAR targeting both EGFR and EGFRvIII that we designed.  Further, we asked if the treatment with EGFR-CAR NK cells plus oHSV could create a therapeutic synergy for the treatment to brain tumors. In vitro, compared with mock-transduced CAR-NK-cells, EGFR-CAR NK cells exhibited significantly higher cytotoxicity and IFN-γ production when co-cultured with tumor cells, for both NK-92 and primary NK cells (P < 0.01). Further, significantly higher cytolytic activity against tumor cells was obtained when CAR NK cells were combined with oHSV-1 infection of tumor cells, compared to either of the monotherapies alone (P < 0.05). In mice, to avoid oHSV clearance by the EGFR-CAR NK cells following the inoculation of the mouse with tumor cells, we administered these two agents sequentially; administering EGFR-CAR NK cells directly into the tumor first as a single injection of 2 × 10⁶ cells, followed by intracranial infection with 2 × 10⁵ plaque-forming units oHSV five days later, presumably after EGFR-CAR NK survival has diminished. Compared to vehicle controls, intracranial administration of either EGFR-CAR NK cells or oHSV blunted tumor growth. However, the combination of EGFR-CAR NK cells followed by oHSV infection resulted in significantly more efficient killing of tumor cells (P < 0.05) and significantly longer survival for tumor-bearing mice when compared to either monotherapy alone.  

Collectively, our studies demonstrate that in animal tumor models, we can combine novel NK cell and oHSV therapies to significantly improve survival.