A Platform for the Scalable Derivation of Genetically-Enhanced T and NK Lymphocytes from Naive Human Induced Pluripotent Stem Cells for Cancer Immunotherapy

Human induced pluripotent stem cell (hiPSC) technology enables the generation of a potentially unlimited source of therapeutically viable hematopoietic cells for the treatment of numerous hematological and non-hematological malignancies, and represents a highly promising approach for overcoming many of the challenges and limitations of patient-derived cancer immunotherapies. To advance the promise of hiPSC technology as an "off-the-shelf" source of hematopoietic cellular therapeutics, it is essential to be able to efficiently and reproducibly generate not only hematopoietic stem cells (HSCs) but also immune effector populations, including the diverse subsets of T and NK lymphoid cells, through a robust and scalable process. The in vitro derivation of HSCs and lymphocytes is complicated by the existence of at least two temporally and spatially distinct waves of blood cell formation during embryonic development: primitive and definitive. Primitive hematopoiesis initiates in the extraembryonic yolk sac and generates a transient and restricted hematopoietic repertoire consisting mainly of primitive erythroid and myeloid cells. Nascent HSCs only emerge later during the definitive wave from a specialized endothelial progenitor within the arterial vasculature termed hemogenic endothelium (HE). HE undergoes an endothelial-to-hematopoietic transition to give rise to HSCs, which then ultimately migrate to the bone marrow where they sustain multi-lineage hematopoiesis, including T and NK lymphoid cells, throughout adult life. Therefore the generation of HSCs and the formation of lymphoid effector cells from hiPSCs is dependent upon our ability to accurately recapitulate the intricate stages of early embryonic hematopoietic development towards the definitive program.

While a limited number of studies have described the directed differentiation of hiPSCs to definitive HE in vitro, a major hurdle in utilizing hiPSCs for therapeutic purposes has been the requirement to initially co-culture such cells with murine-derived stromal cells in the presence of ill-defined serum-containing media in order to maintain pluripotency and induce differentiation. In addition, these protocols have employed an intermediate strategy consisting of embryoid body (EB) formation, which is difficult to scale and hindered by lack of reproducibility. We have previously demonstrated that our proprietary platform for robust and rapid derivation of clonal hiPSC lines, which utilizes small molecule reprograming and single cell selection strategies, generates cells with properties indicative of the naïve, or ground state of pluripotency. In addition to maintaining a homogeneous population of hiPSCs, our platform enables the genetic engineering of such pluripotent cells, at a single cell level, in both nuclease-dependent and -independent strategies.

Here we describe a novel method for the generation of definitive HE from naïve hiPSCs in a scalable manner, void of an EB intermediate, under serum/feeder-free conditions. This platform represents a well-defined, small molecule-driven, staged protocol that can readily be translated to meet current good manufacturing practice (cGMP) requirements for the development of "off-the-shelf" hematopoietic cell-based immunotherapies. The derived HE population is definitive in nature as determine by Notch dependency and exhibits multi-lineage potential, as demonstrated through the formation of both T and NK lymphoid cells. HE generated by this protocol can be successfully cryopreserved and banked, serving as a highly-stable feedstock for subsequent derivation of various cell types for therapeutic use, including for T and NK cell-based immunotherapies. We have demonstrated that our proprietary, clinically-adaptable method for the large-scale production of definitive HE can efficiently give rise to a variety of lymphoid cell subsets. These derived lymphocytes, including NK cells, have been extensively characterized in vitro and in vivo, and we have demonstrated functionality through cytokine release and cellular cytotoxicity. Furthermore, through genetic modifications at the single cell hiPSC stage, tumor antigen-targeting and inducible caspase-mediated safety systems have been introduced into safe harbor loci to improve the specificity and safety profiles of hiPSC-derived T and NK cells for cancer immunotherapy applications.