Engineered iPSC-Derived NK Cells Expressing Recombinant CD64 for Enhanced ADCC

Natural killer (NK) cells are innate cytotoxic lymphocytes. They target malignant cells via non-clonotypic receptors to induce natural cytotoxicity and also recognize tumor-bound antibodies to induce antibody-dependent cell-mediated cytotoxicity (ADCC). While ADCC by NK cells is a key mechanism of several clinically successful therapeutic monoclonal antibodies (mAbs), most patients exhibit or acquire resistance to mAb therapies. ADCC by human NK cells is exclusively mediated by the IgG Fc receptor, CD16A (FcγRIIIA). Studies have demonstrated that increasing the binding affinity between CD16A and therapeutic mAbs can augment their clinical efficacy. Given the exquisite specificity and diverse antigen detection of anti-tumor mAbs, we are interested in enhancing the ADCC potency of NK cell-based therapies for various malignancies.

CD64 is the only high affinity FcγR family member and binds to the same IgG isotypes as CD16A (IgG1 and IgG3) but with > 30-fold higher affinity. CD64 (FcγRI) is normally expressed by certain myeloid cells but not by NK cells. We generated a recombinant version of this receptor consisting of the extracellular region of CD64 and the transmembrane and intracellular regions of human CD16A, referred to as CD64/16A (figure 1A). An important feature of CD64/16A is that due to its high affinity state, soluble monomeric anti-tumor mAbs can be pre-adsorbed to engineered NK cells expressing the recombinant FcγR, and these pre-absorbed mAbs can be switched or mixed for universal tumor antigen targeting (figure 1B). The engineered NK cells used in our study were derived from genetically edited and clonally derived induced pluripotent stem cells (iPSCs) through a series of stepwise differentiation stages (figure 2). Engineered iPSC-derived NK (iNK) cells can be produced in a uniform and clinically scalable manner (figure 2).

In Figure 3, using an in vitro Delfia® ADCC assay, we show that iNK-CD64/16A cells mediated ADCC against SKOV3 cells, an ovarian adenocarcinoma cell line, in the presence of the anti-HER2 therapeutic mAb trastuzumab (Herceptin) or anti-EGFR1 therapeutic mAb cetuximab (Erbitux), when either added to the assay or pre-adsorbed to the iNK cells (figure 3). Considering the high affinity state of CD64, we examined the effects of free IgG in human serum on ADCC by iNK-CD64/16A cells. Using an IncuCyte® Live Cell Analysis System, ADCC was evaluated in the presence or absence of 5% human AB serum, in which free IgG was approximately 50-fold higher than the IgG saturation level of the CD64/16A receptors on iNK cells (data not shown). Despite the high levels of excess free IgG, iNK-CD64/16A cells mediated efficient ADCC when Herceptin was either added to the assay or pre-adsorbed to the cells (figure 4). ADCC assays were also performed with Raji cells, a Burkitt lymphoma cell line, as target cells and the therapeutic mAb rituximab (Rituxan). iNK-CD64/16A cells were added with or without pre-adsorbed Rituxan and the assay was performed in 10% AB serum. Again, iNK-CD64/16A cells mediated effective target cell killing in the presence of serum IgG (figure 5), demonstrating that saturating levels of free IgG did not prevent ADCC.

To determine if we can further optimize the function of recombinant CD64, we engineered CD64 with the transmembrane regions of CD16A or NKG2D and signaling/co-signaling domain from CD28, 2B4 (CD244), 4-1BB (CD137), and CD3ζ (figure 6). CD64/16A signals by non-covalent association with the immunoreceptor tyrosine-based activation motif (ITAM)-containing signaling adapters CD3ζ and FcRγ found in the cell membrane, whereas the other recombinant CD64 constructs use ITAM and non-ITAM regions to mediate their signaling. The various recombinant CD64 constructs were initially expressed in NK92 cells (lacks expression of endogenous FcγRs) (figure 7). Using the Delfia® ADCC assay system, we examined the function of each recombinant CD64 construct and found all combinations are able to effectively induce ADCC (figure 8). We are in the process of generating iNK cells with these constructs and testing their ability to kill hematologic and solid tumors in vitro and in vivo. Our goal is to utilize this docking approach to pre-absorb mAbs to iNK cells for adoptive cell therapy. The mAbs would thus provide tumor-targeting elements that could be exchanged as a means of preventing tumor cell escape by selectively and easily altering NK cell specificity for tumor antigens.