Adoptive transfer of haplo-donor NK cells can achieve remission in patients with relapsed or refractory acute myeloid leukemia (AML). This strategy is limited by NK cell heterogeneity, lack of specificity, and single dosed strategies. To overcome these barriers, we genetically modified induced pluripotent stem cells (iPSCs) to stably over-express NKG2C alone or along with DAP12. CD94/NKG2C is a heterodimeric receptor that binds to HLA-E and associates with DAP12, a protein containing an immunoreceptor tyrosine-based activating motif. Efficient expression of CD94/NKG2C on the cell surface requires the presence of DAP12, and charged amino acids in the transmembrane domains of DAP12 and NKG2C mediate this interaction. Physiologic NKG2C expression is limited to subsets of NK and CD8+ T cells that expand in response to cytomegalovirus (CMV) infection.
Engineered iPSCs were differentiated into CD34+ hematopoietic progenitor cells and subsequently differentiated into CD56+ NK cells (iNK cells) using a stepwise differentiation and expansion strategy. Surface NKG2C expression was low on unmodified iNK cells (10.7%) with higher levels of NKG2C on the surface of iNK cells derived from the NKG2C only line (65.1%). These results demonstrate that transgenic NKG2C can pair with endogenous CD94 and DAP12 to form stable structures on the cell surface. Importantly, surface NKG2C expression was highest on iNK cells derived from the NKG2C/DAP12 line (84.1%) with higher MFI, demonstrating the importance of engineering a line with both components (Figure 1A). To validate function via engagement of NKG2C on iNK cells, we used a redirected antibody dependent cellular cytotoxicity (ADCC) assay where P815 cells (which bind the Fc portion of antibodies) were coated with a monoclonal anti-NKG2C agonist antibody at a range of concentrations (0.01-10 mg/ml) and co-cultured with unmodified iNK cells, NKG2C iNK cells or NKG2C/DAP12 iNK cells. We found that at all concentrations tested, NKG2C/DAP12 iNK cells demonstrated the strongest functional response as measured by the frequencies of surface CD107a expression, interferon (IFN)-g production and tumor necrosis factor (TNF) production. Intermediate functional responses were observed for NKG2C iNK cells (Figure 1B).
Having demonstrated stable expression and intact signaling of NKG2C/DAP12 in iNK cells, we next created a Tri-specific Killer Engager (TriKE) molecule with single chain Fv fragments specific for NKG2C (on iNK cells) and CD33 (on AML cells), as well as IL-15 (to support iNK cell survival and proliferation). This NKG2C/CD33/IL-15 TriKE was tested in functional assays where NKG2C iNK cells or NKG2C/DAP12 iNK cells were used as effectors against CD33+ THP1 cells (an AML cell line shown to be refractory to NK cell cytotoxicity). Without the addition of NKG2C/CD33/IL-15 TriKE, the frequencies of both NKG2C iNK cells and NKG2C/DAP12 iNK cells producing IFN-g was relatively low (< 10%). Addition of NKG2C/IL-15/CD33 TriKE in the assay markedly increased the frequency of IFN-g production by both NKG2C iNK cells (19.1%) and NKG2C/DAP12 iNK cells (25.8%) (Figure 1C). Finally, we directly tested the ability of the NKG2C/IL-15/CD33 TriKE molecule to trigger target killing by engineered iNK cells monitored over a 24-hour period. THP1 target cells were dye-labeled and plated in wells with either NKG2C iNK cells or NKG2C/DAP12 iNK cells at a 5:1 effector-to-target ratio with or without NKG2C/CD33/IL-15 TriKE. We found that in the absence of TriKE, there was no killing of THP1 targets. In contrast, both NKG2C iNK cells and NKG2C/DAP12 iNK cells mediated robust THP1 killing with the addition of NKG2C/CD33/IL-15 TriKE, with NKG2C/DAP12 iNK cells exhibiting the strongest response (Figure 1D). Our results provide a proof-of-concept that iNK cells engineered to express NKG2C/DAP12 can be used in combination with a novel NKG2C/IL-15/CD33 TriKE molecule to effectively target cancer cells. Engaging NK cells through NKG2C, restricted to adaptive NK cells or genetically modified iPSC will be more specific than CD16, which will bind to CD16A on NK cells but also have off-target binding to CD16B on neutrophils. Because of the enormous expansion and engineering capacity of the iNK cell platform, we are in a unique position to create an "off-the-shelf" NK cellular therapy that is targeted, specific and efficacious.
Cichocki:Fate Therapeutics Inc.: Consultancy, Research Funding. Felices:GT Biopharma: Research Funding. Chu:Fate Therapeutics Inc.: Employment. Ge:Fate Therapeutics Inc.: Employment. Bjordahl:Fate Therapeutics Inc.: Employment. Kaufman:Fate Therapeutics: Consultancy, Research Funding. Malmberg:Fate Therapeutics Inc.: Consultancy, Research Funding. Valamehr:Fate Therapeutics Inc.: Employment.
Asterisk with author names denotes non-ASH members.