Natural killer (NK) cells distinguish tumor from healthy tissue via multiple mechanisms, including recognition of stress ligands and loss of MHC class I expression. For example, KIR mismatching enables allogenic NK cells to respond to MHC positive tumors in a similar manner to MHC negative tumors, making allogeneic NK cell therapy a promising approach for broad oncology indications. Accordingly, allogenic human HD-NK cells, including gene-modified cells, have demonstrated an impressive safety and efficacy profile when administered to patients with advanced hematologic malignancies. However, effector function of allogeneic NK cells can be diminished by the lack of functional persistence, as well as tumor-intrinsic immunosuppressive mechanisms, such as production of TGF-β.
To this end, we developed a next-generation allogeneic NK cell therapy using CRISPR-Cas12a gene editing to enhance NK cell function through knockout of the genes CISH and TGFBR2. Both single and simultaneous targeting (DKO) of TGFBR2 and CISH in NK cells using CRISPR-Cas12a produced in/dels at both targets in greater than 80% of NK cells, with greater than 90% of edited NK cells viable at 72 hours post-editing. Importantly, we find that DKO NK cells do not phosphorylate the SMAD2/3 protein downstream of the TGF-b receptor complex and demonstrate increased phosphorylation of pSTAT3 and pSTAT5 upon IL-15 stimulation, consistent with protein level knockout of TGFBR2 and CISH. To determine whether DKO NK cells exhibited superior function relative to control NK cells, we first measured the ability of DKO NK cells to kill Nalm6 cells (adult B cell ALL) relative to unedited control NK cells. We find that in the presence of exogenous TGF-b, DKO NK cells demonstrate improved cytotoxicity against Nalm6 tumor targets by delaying tumor re-growth in comparison to control NK cells.
To better characterize the ability of DKO NK cells to kill tumor cells, we developed an in vitro serial killing assay. In this long-duration assay, up to 30 days, control and DKO NK cells (grown in the presence of IL-15) were challenged every 48 hours with a new bolus of Nalm6 tumor targets. Both DKO and unedited NK cells control Nalm6 target cell growth for greater than 18 days (9 additions of new Nalm6 target cells), demonstrating a surprising ability for the same NK cells to serially kill new Nalm6 target cells for a prolonged period of time in vitro. We find that DKO NK cells produce higher levels of IFN-γ and TNF-α relative to control NK cells over the duration of the entire serial killing assay, suggesting that DKO NK cells can continue to produce these inflammatory cytokines even after serial killing.
As many tumors, including hematologic malignancies, have high concentrations of TGF-β in their microenvironments, we next tested the ability of DKO NK cells to control the growth of Nalm6 cells in our serial killing assay in the presence of TGF-b. 10ng/mL TGF-β was added at the start of the assay as well as at each addition of new Nalm6 target cells. We observed that control NK cells fail to restrict Nalm6 target cell growth beyond 4 days (after 1 addition of new Nalm6 target cells) whereas DKO NK cells control Nalm6 target cell growth for greater than 18 days (after 9 additions of new Nalm6 target cells). Similar to the serial killing assay without TGF-b, we find that DKO NK cells produce higher concentrations of IFN-γ and TNF-α relative to control NK cells over the duration of the entire serial killing assay. Broadening our repertoire of target cells beyond B cell malignancies is now in progress, including the AML-like cell lines HL-60 and THP-1, the multiple myeloma cell line RPMI 8226, and various solid tumor targets.
In summary, using CRISPR-Cas12a we demonstrated highly efficient gene editing of primary human NK cells at two unique targets designed to augment NK cell anti-tumor activity across a variety of malignancies. Most significantly, we demonstrate sustained anti-tumor serial-killing activity in the presence of the potent immunosuppressive cytokine TGF-β. Together, the increased overall effector function of CISH/TGFBR2 DKO primary human NK cells and their ability to serial kill, support their development as a potent allogeneic cell-based medicine for cancer. This potential medicine, termed EDIT-201, is being advanced to clinical study.
Disclosures
Borges: Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Wasko:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Nasser:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Donahue:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Pfautz:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Antony:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Leary:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Sexton:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Morgan:Editas Medicine: Current Employment, Current equity holder in publicly-traded company. Wong:Editas Medicine: Current Employment, Current equity holder in publicly-traded company.
The CD56bright CD62L+ NKG2A+ immature cell subset is dominantly expanded in human cytokine-induced memory-like NK cells