Decreased fratricide of immune cells expressing NKG2D-based receptors

Abstract

This application relates to the field of immunotherapy, and more particularly to the production of cells for adoptive cell therapy. Provided herein are methods for preventing and / or reducing fratricide during the production of such cells, particularly cells expressing chimeric NKG2D receptors. Also provided are cells and compositions containing cells in which fratricide is prevented and / or reduced. [Selection diagram] None

Description

[Technical Field]

[0001] This application relates to the field of immunotherapy, and more particularly to the production of cells for adoptive cell therapy. Provided herein are methods for preventing and / or reducing fratricide during the production of such cells, particularly cells expressing chimeric NKG2D receptors. Also provided are cells and compositions containing cells in which fratricide is prevented and / or reduced. [Background technology]

[0002] Our growing understanding of the immune system has led to the development of a number of immune-focused therapies that produce objective clinical responses in patients with advanced cancer. One of these approaches is chimeric antigen receptor (CAR) T cells, in which a patient's T cells are genetically modified to express a tumor-targeting CAR and then transferred back into the patient in large numbers (1). This adoptive cell therapy has achieved a level of validation through objective clinical responses in patients with hematologic malignancies and is being further explored for the treatment of a broader range of cancer indications (2-7). The CAR concept has also expanded beyond CAR-bearing T cells, which have been explored against a variety of cell types, including natural killer cells (8).

[0003] CARs are modular protein receptors that broadly consist of an extracellular spacer domain and a target-binding domain linked to a structural domain containing a transmembrane region fused to an intracellular signaling domain. Upon ligand binding, downstream signaling initiated by the CAR activates T cell effector functions, thereby promoting direct tumor cell killing and immune-mobilizing cytokine production.

[0004] A problem in CART production is self-killing (or fratricide). This phenomenon occurs when the CAR target is expressed on a T cell population. T cell fratricide is well understood as a mechanism for maintaining T cell homeostasis (13). However, in therapeutic settings, T cell fratricide hinders the ability to generate the CART cell population desired for clinical use. This is particularly relevant in situations where the target itself is selected for T cell lineage specificity, such as CD7 (14) or CD5 (15), to enable targeting of T cell leukemia. However, this problem is not limited to CART cell therapy. T cells armed with TCRs that have high affinity for the T cell receptor specific for survivin (BIRC5) undergo fratricide upon expression of the target antigen (16,17). Similarly, NK cells expressing ligands for NK receptors have also been shown to undergo fratricide (24). In conclusion, we provide an approach to presumptively target any tumor cells by conferring predefined target specificity to immune cells (e.g., T cells or NK cells) using artificial cell surface chimeric antigen receptor (CAR) constructs. The success of this approach depends heavily on the profile of the target antigen itself, as most known tumor antigens are not tumor-specific and may be expressed on non-neoplastic cells. Under certain circumstances, the target antigen can be constitutively or transiently expressed on immune cells, i.e., CAR-modified cells may undergo self-killing or fratricidal processes.

[0005] Here, we focus on CAR cells engineered with specificity for stress ligands targeted by the natural killer group 2 member D (NKG2D) protein. CAR constructs are generated by fusing the full-length NKG2D sequence with the cytoplasmic domain of CD3ξ, and upon ligand binding by the NKG2D extracellular domain, T cell activation occurs via the CD3ξ domain (18). NKG2D has eight known ligands, including major histocompatibility complex class I-related genes A and B (MICA and MICB) and the UL16-binding protein (ULBP) family (ULBP1-6) (19). Expression of these NKG2D ligands is known to be induced under "stress" conditions, such as cell injury, infection, oxidative stress, heat stress, or malignant transformation. A variety of human tumors express NKG2D ligands, highlighting the attractiveness of NKG2D as a receptor to be utilized in the CAR context (19,20). Indeed, murine T cells bearing a murine NKG2D CAR effectively eradicated a variety of established hematologic and solid tumors in syngeneic model systems, demonstrating proof of concept for this approach ( 18 ).

[0006] Human T cells bearing the human NKG2D CAR (also known as NKR-2) can function as effector cells against a variety of tumor cells in vitro and can challenge established human tumors in the NSG mouse model. (21) However, upscaling the production of NKR-2 cells to increase cell numbers for clinical use has proven highly problematic due to T cell fratricide, which has been identified as a cause of transient expression of NKG2D ligands by activated T cells.

[0007] Thus, CARs composed of fusions of the NKG2D protein with CD3ξ (NKR-2) confer broad specificity for NKG2D ligands to T cells. However, T cells transiently express these ligands during activation, leading to fratricide in NKR-2 T cells, which essentially hampers the ability to utilize NKG2D as a therapeutic. Furthermore, for therapeutic use in a clinical setting, significant upscaling and cryopreservation are required to accommodate the necessary dosing schedule. Cryopreservation and freeze-thaw cycles are known to stress cells, which increases the expression of stress-inducible proteins, such as NKG2D ligands. Indeed, when T cells expressing chimeric NKG2D receptors were subjected to both upscaling and cryopreservation, a deterioration in cell yield was observed, likely due to self-killing or fratricide.

[0008] Therefore, preventing or reducing fratricide in these cells would be beneficial, potentially increasing cell yields, reducing production costs, and enhancing the therapeutic efficacy of these cells. From a practical and commercial perspective, this would allow for the upscaling and cryopreservation required for clinical use. [Prior art documents] [Non-patent literature]

[0009] [Non-Patent Document 1] Fesnak AD, June CH, Levine BL (2016) Engineered T cells: the promise and challenges of cancer immunotherapy. Nat Rev Cancer. 16: 566-81. doi: 10.1038 / nrc.2016.97 [Non-patent document 2] Brenner MK (2017) Next Steps in the CAR Journey of a Thousand Miles. Mol Ther. 25: 2226-7. doi: 10.1016 / j.ymthe.2017.09.013 [Non-Patent Document 3] Davila ML, Brentjens RJ (2016) CD19-Targeted CAR T cells as novel cancer immunotherapy for relapsed or refractory B-cell acute lymphoblastic leukemia. Clin Adv Hematol Oncol. 14: 802-8. [Non-Patent Document 4] Kochenderfer JN, Somerville RPT, Lu T et al. (2017) Long-Duration Complete Remissions of Diffuse Large B Cell Lymphoma after Anti-CD19 Chimeric Antigen Receptor T Cell Therapy. Mol Ther. 25: 2245-53. doi: 10.1016 / j.ymthe.2017.07.004 [Non-Patent Document 5] Park JH, Geyer MB, Brentjens RJ (2016) CD19-targeted CAR T-cell therapeutics for hematologic malignancies: interpreting clinical outcomes to date. Blood. 127: 3312-20. doi: 10.1182 / blood-2016-02-629063 [Non-Patent Document 6] Rossig C (2017) CAR T cell immunotherapy in hematology and beyond. Clin Immunol doi: 10.1016 / j. dim. (There seems to be a typo here, should it be 'clinimmunol'?) 2017.09.016

Non-licensed Document 7

Non-licensed literature 9

Non-licensed literature 10

Non-licensed Document 19

Non-licensed Document 20

Non-licensed Document 21

Non-licensed Document 22

Non-licensed Document 23

Direct Environment 24

Direct Entries 25

Direct Entries 26

Direct Accounts 27

[0010] Engineered immune cells, particularly chimeric antigen receptor (CAR) T cells, expressing a fusion of the NKG2D protein with CD3ξ (NKG2D-CART cells), acquire specificity for stress-inducible ligands expressed on hematologic and solid tumors. However, these stress ligands are also transiently expressed by activated immune cells or T cells. This suggests that NKG2D-based immune cells may undergo self-killing (fratricide) during cell manufacturing or freeze-thaw cycles before infusion into patients. [Means for solving the problem]

[0011] It is an object of the present invention to provide a method for reducing and / or preventing fratricide during the production of immune cells expressing chimeric NKG2D receptors, comprising functional inhibition of NKG2D signaling during the production process of said cells. It is also an object of the present invention to provide a method for reducing and / or preventing fratricide during the freezing and / or thawing of frozen immune cells expressing chimeric NKG2D receptors, comprising functional inhibition of NKG2D signaling during the freezing and / or thawing process of said cells.

[0012] Targeted inhibition of NKG2D expression, inhibition of NKG2D ligand expression or enzyme function, particularly PI3K function, can resolve target-driven CART fratricide. It is particularly envisioned that functional inhibition of NKG2D signaling can be achieved by one or more of the following: - permanent or transient inhibition of one or more NKG2D ligands of said immune cells; - transient inhibition of said chimeric NKG2D receptor; - transient inhibition of downstream signaling of said chimeric NKG2D receptor.

[0013] According to embodiments in which permanent inhibition of one or more NKG2D ligands is envisioned, this can be achieved in particular by gene knockdown, and for this purpose, gene editing technologies can be used, including but not limited to Crispr / Cas, TALEN, ZFN, meganucleases, MegaTAL nucleases.

[0014] In certain embodiments, transient inhibition of downstream signaling is envisaged, which is transient inhibition of PI3K signaling. In more specific embodiments, PI3K signaling can be inhibited using various PI3K inhibitors. One example of such an inhibitor is LY294002. Other suitable inhibitors include idelalisib (Cal-101).

[0015] According to certain embodiments, inhibition of receptor or ligand function can be achieved using inhibitory RNA (such as shRNA or siRNA) or antibodies against the NKG2D receptor or one or more of its ligands.

[0016] According to more specific embodiments, functional inhibition is achieved at the receptor level and is carried out via inhibitory RNA or antibodies against the NKG2D receptor. According to one aspect, antibodies against the NKG2D receptor are used. In a specific embodiment of this aspect, the antibody against the NKG2D receptor is an antibody that binds to the receptor without activating the chimeric receptor (i.e., a blocking antibody or antagonist antibody). According to a most specific embodiment, the antibody against the NKG2D receptor is the commercially available 1D11 antibody (named after the clone from which it was isolated).

[0017] According to another specific embodiment, functional inhibition is achieved at the ligand level, via inhibitory RNA or antibodies against one or more NKG2D ligands. The ligands are particularly envisioned to be one or both of MICA and MICB. According to one aspect, shRNA against one or more NKG2D ligands is used.

[0018] According to certain embodiments, the immune cells produced are cells for adoptive cell transfer, such as T cells, NK cells, NKT cells, stem cells or iPSCs.

[0019] In a further embodiment, cells are provided that are less prone to fratricide when manufactured for adoptive transfer. According to this embodiment, there is provided a genetically engineered immune cell comprising a nucleic acid molecule encoding a chimeric NKG2D receptor and at least one of the following: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - one or more shRNAs against a chimeric NKG2D receptor and / or one or more NKG2D ligands.

[0020] In these immune cells, the NKG2D ligand is selected from MICA, MICB, ULBP1, ULBP2, ULBP3, ULBP4, ULBP5 and ULBP6. Particularly contemplated ligands are MICA and / or MICB.

[0021] According to more specific embodiments, there is provided a genetically engineered immune cell comprising a nucleic acid molecule encoding a chimeric NKG2D receptor and at least one of the following: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - One or more shRNAs against one or more NKG2D ligands.

[0022] According to a further aspect, there is provided a composition comprising an immune cell comprising a nucleic acid molecule encoding a chimeric NKG2D receptor, a) the cells - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - further comprising one or more shRNAs against a chimeric NKG2D receptor and / or one or more NKG2D ligands; and / or b) the composition comprises: - with one or more antibodies against a chimeric NKG2D receptor and / or one or more NKG2D ligands; - an inhibitor of downstream signaling of the chimeric NKG2D receptor, in particular a PI3K inhibitor.

[0023] According to certain embodiments of this aspect, there is provided a composition comprising an immune cell comprising a nucleic acid molecule encoding a chimeric NKG2D receptor, The composition comprises: - one or more antibodies against a chimeric NKG2D receptor and / or one or more NKG2D ligands; and / or - Contains an inhibitor of downstream signaling of the chimeric NKG2D receptor, specifically a PI3K inhibitor.

[0024] According to still further particular embodiments, there is provided a composition comprising immune cells comprising a nucleic acid molecule encoding a chimeric NKG2D receptor, The composition comprises: - one or more antibodies against chimeric NKG2D receptors; and / or - Contains a PI3K inhibitor, specifically LY294002 or idelalisib.

[0025] According to yet a further aspect, the genetically engineered immune cells or compositions described herein are provided for use as pharmaceuticals, and are particularly suitable for the treatment of cancer.

[0026] This is equivalent to stating that there is provided a method of treating cancer, the method comprising administering to a subject in need thereof genetically engineered immune cells comprising a nucleic acid molecule encoding a chimeric NKG2D receptor and at least one of the following: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - one or more shRNAs against a chimeric NKG2D receptor and / or one or more NKG2D ligands.

[0027] Similarly, there is provided a method of treating cancer comprising administering to a subject in need thereof a composition comprising immune cells comprising a nucleic acid molecule encoding a chimeric NKG2D receptor; a) the cells - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - further comprising one or more shRNAs against a chimeric NKG2D receptor and / or one or more NKG2D ligands, and / or b) the composition comprises: with one or more antibodies against a chimeric NKG2D receptor and / or one or more NKG2D ligands; Methods for treating cancer involving inhibitors of downstream signaling of chimeric NKG2D receptors, specifically PI3K inhibitors, are provided.

[0028] The treatment method may be autologous (where the subject receives cells from their own body) or allogeneic (where the immune cells are derived from a donor other than the subject). [Brief explanation of the drawings]

[0029] [Figure 1X] Flow cytometric characterization of PBMC, tCD19, and NKG2D-CART cells in a representative NKR-2 T cell process. Mononuclear cells purified from blood using density gradients were analyzed before (PBMC) and after (tCD19 and NKR-2 T cells) the NKR-2 T cell process. After acquisition, cells were gated on SSC / FSC to obtain lymphocytes. Lymphocytes were then gated on CD3. All CD3-positive cells were then displayed as follows: (A) CD4 / CD8 distribution. (B) Surface expression of NKG2D (CD314). [Figure 1Y](C) Memory phenotypes. CD62L and CD45RA were used to distinguish naive (CD62L+CD45RA+), central memory (CD62L+CD45RA-), effector memory (CD62L-CD45RA-), and CD45RA-expressing terminally differentiated T cells (CD62L-CD45RA+). (D) Exhaustion phenotypes stained for CD223 (Lag-3) and CD279 (PD-1). One representative donor out of three is shown. [Figure 2] CD314 mean fluorescence intensity (MFI). A-B) MFI of CD314 in CD3+CD4+ or CD3+CD8+ T cells at harvest of untransduced, MockCD19-transduced, or NKR2-transduced T cells. Mean values ​​are shown for n = 9 for untransduced T cells, n = 11 for MockCD19-transduced T cells, and n = 16 for NKR2-transduced T cells. Significance was assessed using a two-tailed unpaired t-test with Welch's correction (***p = 0.0003. ****p < 0.0001). C) Mean fluorescence intensity values ​​for untransduced, MockCD19-transduced, or NKR2-transduced T cells from at least nine different donors. MFI and standard errors are shown for CD4 and CD8 T cells. [Figure 3X]NKR-2 T cells recognize chronic myeloid leukemia (K562) and pancreatic cancer (PANC-1) tumors and exhibit fratricidal effects due to NKG2DL expression. (A) T cells from healthy donors were transduced with tCD19 vector (tCD19 T cells) or NKR-2 T cell vector (NKG2D-CART cells) and cocultured with adapted cell lines or cultured alone (-). After overnight coculture, IFN-γ secretion (ng / ml) was quantified by ELISA. Each data point represents the mean value of replicate wells from independent experiments. Figure shows (N=3). (B) NKR-2 T cells exhibit cytolytic activity. PANC-1 cells were cocultured with thawed NKG2D-CART cells at a 1:1 E:T ratio. After 20 hours, Alamar Blue index was determined. Cytolysis rates were determined by comparing absorbance with untreated cancer cells (PANC-1). Data are shown as mean ± standard deviation for N = 3 independent T cell donors (PANC-1 cells alone, NKR-2 T cells: co-culture of NKR-2 T cells with PANC-1 cells, tCD19: co-culture of control tCD19 T cells with PANC-1 cells). (C) Transduced T cells were cultured in complete x-vivo medium (100 IU / mL IL-2) for 4 days. T cells were analyzed 96 hours after seeding to analyze proliferation rates relative to initial cell seeding density. Data are shown as mean ± standard deviation for N = 3 independent T cell donors. [Figure 3Y] (D) Representative coculture experiment (out of three). GFP-positive T cells were cultured alone (Mock-GFP) or cocultured with NKR-2 T cells from the same donor (Mock-GFP + NKR-2 T cells). Flow cytometry analysis of GFP positivity was performed at the start of incubation (T = 0 h) or 24 h after incubation (T = 24 h). [Figure 3Z](E–F) PBMCs were activated with 40 ng / mL anti-CD3 and 100 IU / mL IL-2 and cultured for a total of 8 days according to the manufacturer's standard protocol. Cell samples were collected every 2 days and analyzed for NKG2D ligand expression by either RNA (E) or cell surface protein (F). Statistical significance was assessed using a two-tailed unpaired t-test. p<0.05 was considered significant (*), and p<0.01 was considered significant (**). For both qPCR and flow cytometry comparisons, a two-tailed paired t-test was used. [Figure 4]NKR-2 T cells exhibit NKG2D-mediated fratricide, which can be inhibited by NKG2D-blocking antibodies or PI3K inhibitors. (A) MFI of NKG2D expression on NKR-2 T cells treated with or without increasing concentrations of LY294002. Data represent the mean ± standard deviation from N=3 independent donors. (B) Transduced T cells were cultured in complete x-vivo (100 IU / mL IL-2) supplemented with or without increasing concentrations of LY294002 for proliferation. T cells were analyzed 96 hours after seeding, and proliferation rates were analyzed relative to the initial cell seeding density. Data represent the mean ± standard deviation from N=3 independent donors. (C) Cell viability after cryopreservation. NKR-2 T cells were generated with increasing concentrations of LY294002. After generation, cells were harvested, washed, and formulated for cryopreservation. After cryopreservation, cells were thawed using a water bath and resuspended in Plasmalyte / human serum albumin (HSA) 5%. Cell viability was assessed immediately after thawing (TOh) or after 6 hours in Plasmalyte / HSA 5% at 4°C (T6h) (N=715, 3). (D) NKR-2 T cells were generated using increasing concentrations of LY294002. After generation, cells were harvested, washed, transferred to Plasmalyte / HSA 1% (50 × 10 NKR-2 T cells / ml) and stored at 4°C for 48 hours. After 48 hours, cell viability was assessed using trypan blue staining (N=3) and normalized to the cell number at the time of cryopreservation. (E) tCD19 T cells were co-cultured with NKR-2 T cells from the same donor in the presence of increasing concentrations of blocking Ab (0-10 μg / ml). After 44 hours of incubation, CD19 positivity and viability were analyzed by flow cytometry. Data were normalized to the CD19 positivity of mock-cultured NKR-2 T cells (n=3). (F) Thawed NKR-2 T cells were cocultured at a 1:1 ratio with either PANC-1 or K562 cells in the presence of CD314-blocking Ab, isotype control, or no Ab. After 24 hours of incubation, supernatants were collected and IFN-γ assays were performed (n=3).(G) Thawed NKR-2 T cells were cultured for 24 hours in the presence of isotype control or CD314-blocking Ab (or no Ab), and IFN-γ levels were measured. Data are shown as mean ± standard deviation from N = 3 independent donors. Statistical significance was assessed using a two-tailed unpaired t-test. p < 0.05 was considered significant (*), and p < 0.01 was considered significant (**). [Figure 5] MFI of NKG2D expression on NKR-2 T cells treated with or without increasing concentrations of LY294002. [Figure 6] Proliferation rate of tCD19 cells in the presence or absence of PI3K inhibitors. [Figure 7] NKR-2 T cells generated with a PI3K inhibitor produce higher amounts of IFN-γ (A) and exhibit an increased memory phenotype (B). [Figure 8] Proliferation rate of NKR-2 T cells with and without antibody blockade. [Figure 9]Adaptation of NKR-2 T cells to Ab blockade restores the CD4 / CD8 ratio. (A) Cytolytic activity kinetics, a representative killing assay among three. Thawed control tCD19 T cells or NKR-2 T cells treated with the PI3K inhibitor LY294002 or blocking Ab were cultured in the presence of NucLight-positive PANC-1 cells. PANC-1 viability was assessed every 2 hours using an IncuCyte S3 device. (B) CD4 / CD8 distribution at harvest. During the expansion phase, NKR-2 T cells were cultured with either 5 μM LY294002 or 5 μg / mL blocking Ab for 96 hours, harvested, and CD4 and CD8 populations were measured by flow cytometry. Data represent the mean ± standard deviation of N=4 independent T cell donors relative to the control tCD19 ratio. (C) CD4 / CD8 distribution after delayed addition of blocking Ab and harvest. During the expansion phase (days 4–8), NKR-2 T cells were treated with 5 μM blocking Abs either immediately (day 4) or 48 hours later (day 6). At the time of harvest (day 8), T cells were harvested and analyzed by flow cytometry for CD4 and CD8 populations. Data represent the mean ± standard deviation (SD) of N = 3 independent T cell donors relative to the CD4 / CD8 ratio of control tCD19 cells. (D) Comparison of proliferation rates. NKR-2 T cells were cultured for 8 or 10 days in the presence of LY or blocking Abs (added on days 4 or 6). T cells were analyzed at harvest for proliferation rates relative to the initial cell seeding density. (E) Differentiation potential assay by IFN-γ secretion. Thawed NKR-2 T cells treated with the above two methods were cocultured with PANC-1 cells. After 44 hours of coculture, IFN-γ secretion was measured by ELISA. Data represent the mean ± standard deviation (SD) of N = 4 independent T cell donors. (F) Cytolytic activity kinetics, a representative killing assay among the three. Thawed control tCD19 T cells or NKR-2 T cells treated with PI3Ki or blocking antibodies were cultured in the presence of NucLight-positive PANC-1 cells. PANC-1 cell viability was assessed every 2 hours using an IncuCyte S3 instrument. Statistical significance was assessed using a two-tailed unpaired t-test. p<0.05 was considered significant (*), p<0.01 (**), and p<0.001 (***). [Figure 10] Relationship between idelalisib (Cal101) and NKG2D-blocking antibodies. Comparison of cell proliferation (A), viability (B), and differentiation potential measured by IFN-γ secretion after co-culture with K562 cells in cells cultured in the presence of a blocking NKG2D antibody or 5 μM CAL101. [Figure 11] Expression of NKG2D ligands on the surface of CD4+ (A) and CD8+ (B) T cells. [Figure 12] Co-expression of MICA / B targeting shRNA reduces fratricide. [Figure 13] The reduction in fratricide increases cancer cell killing. DETAILED DESCRIPTION OF THE INVENTION

[0030] definition The present invention will be described based on specific embodiments and with reference to certain drawings. However, the present invention is not limited to these embodiments, but only by the claims. Reference signs in the claims should not be construed as limiting the scope of the present invention. The drawings are merely schematic and not limiting. For illustrative purposes, the dimensions of elements in the drawings may be exaggerated and not precisely depicted. The term "comprising" as used in this specification and the scope of the present invention does not exclude other elements or processes. A noun without a specified quantity includes the plural of that noun unless otherwise specified. The phrase "consisting essentially of" means that the recited elements are necessarily included, excluding elements that materially affect the basic and novel characteristics of the recited elements, and that other elements may optionally be included. The phrase "consisting of" means that all elements other than the recited elements are excluded. Embodiments defined by each of these terms are within the scope of the present invention.

[0031] Furthermore, terms such as "first," "second," and "third" used in the specification and claims are used to distinguish between similar elements and do not necessarily denote a sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances, and that the embodiments of the invention described herein can be practiced in other orders than those described or illustrated herein.

[0032] The following terms or definitions are provided solely to aid in the understanding of the invention.

[0033] Unless otherwise defined herein, all terms used herein have the same meaning as would be understood by one skilled in the art of the present invention. For definitions and terms of the art, practitioners are specifically referred to the following documents:

[0034] Green and Sambrook, Molecular Cloning: A Laboratory Manual, 4th Edition, Cold Spring Harbor Laboratory Press, New York (2012)

[0035] Ausubel et al, Current Protocols in Molecular Biology (up to Supplement 114), John Wiley&Sons, New York (2016)

[0036] The definitions provided herein should not be construed less broadly than understood by one of ordinary skill in the art.

[0037] As used herein, the term "fratricide" refers to the killing of cells by genetically identical cells, most specifically immune cells.

[0038] As used herein, the term "reducing and / or preventing fratricide" refers to a reduction in the occurrence of fratricide in a cell population compared to an appropriate control cell population (typically, but not necessarily, a population of identical cells in which NKG2D ligand inhibition does not occur). Reduction is expressed as a percentage reduction compared to the control cell population, e.g., a 10%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 90%, or even 100% reduction in fratricide. Reduction of fratricide can also be assessed by an increase in final cell yield or number (because fewer cells are killed, more cells survive and proliferate). Thus, for example, an increase in cell yield of 10%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 90%, 100%, or even 100% or more. Importantly, a decrease in fratricide, as defined herein, is associated with increased antigen-specific cytokine production (e.g., interferon-γ secretion) and / or, in the case of T cells, an increased frequency of T cells in the memory phenotype (e.g., CD62L + / CD45RA). While these measures may not directly correlate with absolute cell yield, if antigen-specific cytokine production increases, this means there will be more therapeutically active cells at the end of the manufacturing process, and less fratricide in the therapeutic cells. The same is true for an increase in T memory cell frequency. Thus, a reduction in fratricide can be assessed by, for example, a 10%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 90%, 100%, or even 100% or greater increase in antigen-specific cytokine production or a 10%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 90%, 100%, or even 100% or greater increase in memory T cell frequency.

[0039] Measures to reduce fratricide can be taken before the fratricide process begins (which is generally preferred as it is more effective) or during fratricide killing. Prevention of fratricide refers to measures taken before the fratricide process begins. According to certain embodiments, absolute prevention of fratricide means that NKG2D-induced fratricide does not occur (ideally, no fratricide occurs at all), i.e., a 100% reduction in fratricide.

[0040] As used herein, the term "immune cells" refers to cells that are part of the immune system (which may be either the adaptive or innate immune system). Particularly contemplated immune cells include white blood cells (leukocytes), including lymphocytes, monocytes, macrophages, and dendritic cells. Particularly contemplated lymphocytes include T cells, NK cells, and B cells, with T cells being most particularly contemplated. As used herein, immune cells are typically immune cells produced for adoptive cell transfer (either autologous or allogeneic). Note that in the context of adoptive transfer, immune cells are typically primary cells (i.e., cells that have been isolated directly from human or animal tissue and then not cultured or cultured for a short period of time), rather than cell lines (i.e., cells that have been continuously passaged over an extended period of time and have acquired homozygous genotypic and phenotypic characteristics). According to certain embodiments, the immune cells are primary cells. According to another particular embodiment, the immune cells are not derived from a cell line.

[0041] As used herein, the term "chimeric NKG2D receptor" refers to a non-naturally occurring receptor having specificity for an NKG2D ligand. It is chimeric because the binding moiety is fused to one or more different moieties (including at least one signaling moiety), at least one of which is of a different origin (e.g., a different protein) from the binding moiety. Particularly contemplated examples of chimeric NKG2D receptors include NKG2D CARs, i.e., chimeric antigen receptors having a binding moiety derived from an NKG2D receptor. Such NKG2D CARs are disclosed, for example, in WO2006 / 036445 and WO2014 / 117121. The definition of chimeric NKG2D receptors herein also includes CARs having a binding moiety that recognizes one or more NKG2D ligands that are not derived from an NKG2D receptor, for example, an antibody or antibody-like moiety (e.g., scFv, VHH, sdAb, etc.) against one or more NKG2D ligands. This is then typically fused to a signaling moiety that transduces a signal in immune cells, specifically T cells (such as the CD3 zeta chain or the Fc epsilon receptor gamma chain).

[0042] As used herein, the term "functional inhibition of NKG2D signaling" refers to interfering with the function of an NKG2D gene product (i.e., the product of a chimeric NKG2D receptor gene) at either the DNA level (by inhibiting the formation of the NKG2D gene product or one or more of its ligands, i.e., by preventing or interfering with transcription), the RNA level (by neutralizing or destabilizing mRNA to prevent or interfere with translation—the mRNA may be that of the chimeric NKG2D receptor and / or one or more of its ligands), or the protein level (by neutralizing or inhibiting the chimeric NKG2D protein and / or one or more of its ligands). Neutralization at the protein level can be achieved at the cell surface (e.g., by inhibiting receptor-ligand interaction) or before the protein is expressed on the cell surface (e.g., by retaining the protein in intracellular organelles). Typically, the ultimate functional effect of inhibiting NKG2D-induced signaling is inhibition of immune cell activation via signals generated by the chimeric NKG2D receptor, although this can be achieved indirectly (e.g., at the DNA level, i.e., by inhibiting one or more ligands).

[0043] Functional inhibition of NKG2D signaling does not necessarily mean complete elimination of the NKG2D-induced signal, although this is also expected. It is known that inhibition, particularly with antisense RNA and siRNA, as well as with antibodies, is often partial rather than complete. However, reducing functional NKG2D gene product or NKG2D ligand levels may have beneficial effects even if complete inhibition is not achieved—particularly because fratricide is typically cell density-dependent, and reduced availability of functional ligand or receptor may result in a reduction in cell density. (Note that reducing the actual cell density using culture dilution to generate an appropriate product given the large number of cells required for future clinical trials is not feasible.)

[0044] Thus, according to certain embodiments, inhibition results in a 10%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 90%, or up to 100% reduction in functional chimeric NKG2D receptor gene product or one or more NKG2D ligands. Methods for measuring the levels of functional NKG2D receptor gene product or ligand are known to those of skill in the art, and these can be measured before and after addition of an inhibitor to assess the reduction in the level of functional gene product.

[0045] As used herein, the term "transient inhibition" means that the inhibition is temporary (or temporarily regulated) and that the function of the chimeric NKG2D receptor is restored at a later time. Typically, NKG2D-induced signaling is inhibited during the manufacturing of the immune cells (i.e., before they are used as a pharmaceutical), but signaling capacity is restored once the cells are administered to a patient, because signaling through NKG2D is important for therapeutic efficacy.

[0046] As used herein, the term "NKG2D ligand" refers to the human genes MICA (Gene ID: 100507436), MICB (Gene ID: 4277), ULBP1 (Gene ID: 80329), ULBP2 (Gene ID: 80328), ULBP3 (Gene ID: 79465), ULBP4 or RAET1E (Gene ID: 135250), ULBP5 or RAET1G (Gene ID: 353091), ULBP6 or RAET1L (Gene ID: 154064), and their gene products (or related homologs when cells of other species are used).

[0047] This application is the first to demonstrate that fratricide of immune cells expressing a chimeric NKG2D receptor can be prevented or reduced by inhibiting the function of this receptor (by inhibiting either the receptor or one or more of its ligands, or both). This reduction in fratricide could improve cell yield and reduce the cost of the therapy, thereby facilitating its use in clinical settings. The inhibition may be transient (during the in vitro production process of the immune cells, but not at the time of administration to the patient—because the chimeric NKG2D receptor is required for the therapeutic immune cell function of the cells) or permanent (because if only the immune cell's ligand is inhibited, this does not interfere with the therapeutic effect).

[0048] It is therefore an object of the present invention to provide methods for reducing and / or preventing fratricide during the production of immune cells expressing chimeric NKG2D receptors, comprising functional inhibition of NKG2D signaling during the production process of said cells. These production methods may be performed in vitro or ex vivo. Also provided is a method for reducing and / or preventing fratricide during freeze-thaw cycling of immune cells expressing chimeric NKG2D receptors, comprising functional inhibition of NKG2D signaling during said freeze-thaw cycling. These freeze-thaw methods may also be performed in vitro or ex vivo.

[0049] It should be noted that, although NKG2D is the most well-studied receptor involved in the recognition of inducible autoantigens, it is not the only receptor in this family that recognizes stress-induced ligands (or inducible autoantigens or markers of abnormal self, all used interchangeably herein). Other receptors capable of binding to inducible autoantigens are natural cytotoxicity receptors (NCRs), such as NKG2C, NKG2E, NKG2F, NKG2H (all CD94 molecules, like NKG2D), or NKp46, NKp30, and NKp44, and it is contemplated that the above methods and compositions can be used mutatis mutandis with such chimeric receptors. Thus, whenever NKG2D is used herein, it also applies to NKG2C, NKG2E, NKG2F, NKG2H, NKp46, NKp30, and NKp44. It should be noted that the ligands for NKG2C, NKG2E, NKG2F and NKG2H are non-classical MHC glycoproteins class I (HLA-E in humans).

[0050] These methods are in vitro methods (as they correspond to cell production). Functional inhibition of NKG2D signaling is achieved by one or more of the following: -Permanent or transient inhibition of one or more NKG2D ligands on immune cells; -Transient inhibition of chimeric NKG2D receptors; -Transient inhibition of downstream signaling of chimeric NKG2D receptors.

[0051] Inhibition can occur in several ways. Some examples include contact inhibition (competitive or non-competitive), inhibition by interfering with ligand or receptor expression, interference with ligand or receptor localization (e.g., preventing migration to the cell surface), inhibition by binding to or interfering with the interaction of the ligand or receptor, and inhibition of downstream signaling.

[0052] Permanent inhibition of one or more NKG2D ligands is typically achieved by gene knockdown. Indeed, gene editing has proven to be a viable method for specifically eliminating expression of target antigens in genetically engineered T cells (14). However, given the potential expression of eight different ligands, gene editing techniques to eliminate all of these polymorphic targets are challenging, making it particularly contemplated that only one or a few ligands need to be permanently inactivated. If many ligands are expressed, an alternative approach could be to adapt them to fratricide control. The inventors have found that, of the eight NKG2D ligands, MICA and MICB are primarily expressed on the cell surface of CD4+ and CD8+ human T cells. Therefore, inhibition of MICA and MICB is specifically contemplated in certain embodiments.

[0053] Generally, functional inhibition can be achieved at three levels. First, at the DNA level, for example, by deleting or disrupting a gene (typically an NKG2D ligand gene) in the immune cell or by preventing transcription from occurring (in either case, preventing the synthesis of the gene product). Second, at the RNA level, for example, by preventing efficient translation from occurring (this is done via destabilizing the mRNA, so that it is degraded before transcription can translate) or by hybridizing to the mRNA. Third, at the protein level, for example, by binding to the protein, inhibiting its function, retaining the protein in a different cellular location, and / or marking the protein for degradation.

[0054] When inhibition is achieved at DNA level, it can be achieved by knocking out or destroying gene using gene therapy.This is typically a permanent inhibition, so the inhibition of NKG2D ligand in immune cells is particularly envisioned.As used herein, " knocking out " refers to gene knockdown, that is, gene can be knocked out by mutation such as point mutation, insertion, deletion, frameshift or missense mutation by techniques known in the art, including but not limited to retroviral gene transplantation.Another way to knock out gene is to use genetically engineered nuclease.Such genetically engineered nuclease includes but is not limited to meganuclease, zinc finger nuclease, TALEN, megaTAL and CRISPR nuclease.

[0055] Meganucleases, commonly found in microbial species, have the unique property of possessing very long recognition sequences (>14 bp) for generating site-specific double-strand breaks in nucleic acids. This naturally makes meganucleases highly specific for their target sequences, and hybrid meganuclease variants that recognize unique sequences can be obtained through mutagenesis and high-throughput screening. In contrast to meganucleases, the concept behind ZFN and TALEN technologies is based on nonspecific DNA-cleaving enzymes, which can then be linked to specific DNA sequence-recognizing peptides such as zinc finger and transcription activator-like effector (TALE) domains. Zinc finger nucleases (ZFNs) are artificial restriction enzymes generated by fusing a zinc finger DNA-binding domain to a DNA-cleavage domain. The zinc finger domain can be engineered to target desired DNA sequences, allowing zinc finger nucleases to target unique sequences within complex genomes. By exploiting endogenous DNA repair mechanisms, these reagents can be used to precisely modify the genomes of higher organisms. TALENs function similarly to zinc fingers, but are transcription activator-like effectors (TALEs) for DNA recognition. TALEs are repeating units with a one-to-one recognition ratio between amino acids and their corresponding nucleotide pairs. Because TALEs occur in a repeating pattern, different combinations can be tried to create a wide variety of sequence specificities.

[0056] MegaTALs are derived from the combination of two different classes of DNA-targeting enzymes. Meganucleases (also known as homing endonucleases) are single peptide chains that combine the benefits of both DNA recognition and nuclease functions in the same domain. However, meganuclease target recognition is difficult to modify, and they often exhibit less specificity and on-target cleavage efficiency than other genome-targeting endonucleases. Transcription activator-like (TAL) effectors are DNA recognition proteins linked to separate DNA endonuclease domains to achieve targeted DNA double-strand cleavage. In contrast to meganucleases, TALs are easily engineered to target specific DNA sequences. Current platforms rely on a pair of TAL effectors, each conjugated to a nonspecific DNA cleavage domain. DNA cleavage occurs only when both TAL effectors bind to their respective sequences and the two endonuclease domains dimerize to cleave the DNA. However, TAL effector nucleases can exhibit off-target activity, are much larger than meganucleases, and require the delivery of two separate proteins. A megaTAL is a combination of a TAL effector and a meganuclease.

[0057] CRISPR / Cas (Clustered Regularly Interspaced Short Palindromic Repeats / CRISPR associated protein) is a genome editing technology that uses a modified version of a prokaryotic defense mechanism to enable persistent genetic alterations in living organisms. By delivering a Cas (typically Cas9) nuclease complexed with a synthetic guide RNA (gRNA) into a cell, the cell's genome can be cut at desired locations, allowing for the removal of existing genes and / or the addition of new genes.

[0058] Permanent inhibition by gene knockdown of one or more NKG2D ligand genes is typically performed in immune cells in which chimeric NKG2D receptors are also present. One or more NKG2D ligand genes can refer to the inhibition of any combination of MICA, MICB, ULBP1, ULBP2, ULBP3, ULBP4, ULBP5, and ULBP6, and therefore can refer to the knockout of 1, 2, 3, 4, 5, 6, 7, or 8 genes.

[0059] In addition to permanent inhibition of NKG2D ligands, functional inhibition of NKG2D signaling can also be achieved by transient inhibition, which may involve inhibition of one or more NKG2D ligands on immune cells, or inhibition of chimeric NKG2D receptors or downstream signaling.

[0060] The time frame for transient inhibition typically coincides with the time frame for manufacturing. The production of immune cells expressing chimeric NKG2D receptors involves several steps, and protocols vary, but essentially always include a transduction step (introducing the chimeric NKG2D receptor into isolated immune cells), an expansion step (cultivating and expanding the cells), and a harvesting step (isolating and reformulating or concentrating the cells before administering to patients or for (cryo)storage). Transient inhibition specifically refers to inhibition during the expansion step (because fratricide is most likely to occur during this step), but transient inhibition may also occur throughout the entire manufacturing process (from or before the transduction step to or after the harvesting step). Typically, if external inhibitors (e.g., antibodies) are used, they are removed during the harvesting / reformulation process. Therefore, the method may include an activation step to remove the used inhibitor. However, in other instances, the inhibitor may be transient in nature (e.g., due to a short half-life) or may be under the control of an inducible promoter that is not active after the manufacturing process, and no activation step may be required to terminate the inhibition of functional NKG2D signaling. In a typical setting, transient inhibition entails inhibition from approximately the transduction step to the administration / infusion step, although shorter or longer time frames may also be envisioned.

[0061] One form of transient inhibition is by transient gene inactivation.Transient gene inactivation can be achieved, for example, through the expression of antisense RNA in immune cells or by administering antisense RNA to said cells.Antisense constructs can be delivered, for example, as expression plasmids, and when transcribed in cells, produce RNA that is complementary to at least a specific part of target mRNA (here, NKG2D ligand mRNA or chimeric NKG2D receptor mRNA).

[0062] A more rapid method for inhibiting gene expression is based on the use of shorter antisense oligomers composed of DNA or other synthetic structural types, such as phosphorothioates, 2'-O-alkylribonucleotide chimeras, locked nucleic acids (LNA), peptide nucleic acids (PNA), or morpholinos. With the exception of RNA oligomers, PNAs, and morpholinos, all other antisense oligomers act in eukaryotic cells via the mechanism of RNase H-mediated target cleavage. PNAs and morpholinos bind to complementary DNA and RNA targets with high affinity and specificity, acting through simple steric blocking and rendering them completely resistant to nuclease attack. "Antisense oligomer" refers to an antisense molecule or anti-gene agent comprising an oligomer at least about 10 nucleotides in length. In embodiments, the antisense oligomer comprises at least 15, 18, 20, 25, 30, 35, 40, or 50 nucleotides. The antisense approach involves the design of oligonucleotides (either DNA or RNA, or derivatives thereof) complementary to the mRNA encoded by the FMR1 polynucleotide sequence. Antisense RNA can be introduced into cells and inhibit the translation of complementary mRNA by base pairing with the complementary mRNA and physically interfering with the translational machinery. Therefore, this effect is stoichiometric. Absolute complementarity is preferred, but not required. As used herein, a sequence "complementary" to a portion of RNA refers to a sequence that has sufficient complementarity to be able to hybridize with the RNA and form a stable duplex. In the case of a double-stranded antisense polynucleotide sequence, a single strand of the double-stranded DNA may be tested, or triplex formation may be assayed. The ability to hybridize depends on both the degree of complementarity and the length of the antisense polynucleotide sequence. Generally, the longer the hybridizing polynucleotide sequence, the more base mismatches it may contain with the RNA, resulting in the formation of a stable duplex (or triplex, as the case may be). One skilled in the art can ascertain a tolerable degree of mismatch by using standard procedures to determine the melting temperature of the hybridized complex.Oligomers complementary to the 5' end of a message (e.g., the 5' untranslated region (UTR) through the AUG translation initiation codon) should be most effective at inhibiting translation. However, it has recently been shown that sequences complementary to the 3' UTR of mRNAs are also effective at inhibiting mRNA translation (Wagner, R. (1994) Nature 372, 333-335). Therefore, oligomers complementary to either the 5' UTR or 3' UTR, i.e., the non-coding region of a target gene, may be used in an antisense approach to inhibit translation of the endogenous mRNA encoded by the target gene. Oligomers complementary to the 5' UTR of the mRNA naturally contain the complement of the AUG initiation codon. Antisense oligomers complementary to mRNA coding regions are less efficient translation inhibitors, but can be used in the present invention. When designed to hybridize to either the 5', 3', or non-coding regions of the mRNA, antisense oligomers should be at least 10 nucleotides in length, preferably 15 to about 50 nucleotides in length. In certain embodiments, oligomers are at least 15, 18, 20, 25, 30, 35, 40, or 50 nucleotides in length. Related methods use ribozymes instead of antisense RNA. Ribozymes are catalytic RNA molecules with enzyme-like cleavage properties that can be engineered to target specific RNA sequences. Successful targeted gene inactivation, including transient and tissue-specific gene inactivation, using ribozymes has been reported in mice, zebrafish, and fruit flies. RNA interference (RNAi) is a form of post-transcriptional gene silencing. Reduced RNA interference was first observed and described in Caenorhabditis elegans, where exogenous double-stranded RNA (dsRNA) was shown to specifically and potently disrupt the activity of genes containing homologous sequences through a mechanism that induces rapid degradation of the target RNA.Several reports have described the same catalytic phenomenon in other organisms, including plants (Arabidopsis thaliana), protozoa (Trypanosoma brucei), invertebrates (Drosophila melanogaster), and vertebrate species (zebrafish and Xenopus laevis), including experiments demonstrating spatial and / or temporal control of gene inactivation. The mediators of sequence-specific messenger RNA degradation are small interfering RNAs (siRNAs), generated by RNase III cleavage from longer dsRNAs. siRNAs are typically between 20 and 25 nucleotides in length (Elbashir et al. (2001) Nature 411, 494–498). siRNAs typically contain a sense RNA strand and a complementary antisense RNA strand annealed together by standard Watson-Crick base-pairing interactions (hereafter referred to as "base pairing"). The sense strand contains a nucleic acid sequence identical to the target sequence contained in the target mRNA. The sense and antisense strands of the siRNAs of the present invention can comprise two complementary single-stranded RNA molecules, or can comprise a single molecule in which the two complementary portions are base-paired and covalently linked by a single-stranded "hairpin" region (often referred to as "shRNA"). The term "isolated" means altered or removed from its natural state through human intervention. For example, an siRNA naturally present in a living animal would be "non-isolated," whereas a synthetic siRNA or an siRNA partially or completely separated from its natural coexisting materials would be "isolated." Isolated siRNAs can exist in substantially purified form or in a non-native environment, such as a cell into which the siRNA has been delivered.

[0063] The siRNAs of the present invention can include partially purified RNA, substantially pure RNA, synthetic RNA, or recombinantly produced RNA, as well as modified RNA that differs from naturally occurring RNA by the addition, deletion, substitution, and / or modification of one or more nucleotides. Such modifications can include, for example, the addition of non-nucleotide material to the termini of the siRNA or to one or more internal nucleotides of the siRNA, including modifications that render the siRNA resistant to nuclease digestion.

[0064] One or both strands of the siRNA of the present invention can also contain a 3' overhang. A "3' overhang" refers to at least one unpaired nucleotide extending from the 3' end of an RNA strand. Thus, in one embodiment, the siRNA of the present invention contains at least one 3' overhang that is 1 to about 6 nucleotides in length (including ribonucleotides or deoxynucleotides), preferably 1 to about 5 nucleotides in length, more preferably 1 to about 4 nucleotides in length, and particularly preferably about 1 to about 4 nucleotides in length.

[0065] In the embodiment where both strands of siRNA molecule comprise 3' overhang, the length of the overhang can be the same or different for each strand.In the most preferred embodiment, 3' overhang is present in both strands of siRNA, and its length is 2 nucleotides.To enhance the stability of the siRNA of the present invention, 3' overhang can also be stabilized against degradation.In one embodiment, the overhang is stabilized by including purine nucleotides such as adenosine nucleotides or guanosine nucleotides.

[0066] Alternatively, substitution of pyrimidine nucleotides with modified analogs, such as substitution of uridine nucleotides with 2'-deoxythymidine in the 3' overhang, is tolerated and does not affect the efficiency of RNAi degradation. Specifically, the absence of a 2'-hydroxyl in 2'-deoxythymidine significantly enhances the nuclease resistance of the 3' overhang in tissue culture medium.

[0067] siRNA can be obtained using several techniques known to those skilled in the art.For example, siRNA can be chemically synthesized or recombinantly produced by methods known in the art.Preferably, the siRNA of the present invention is chemically synthesized using appropriately protected ribonucleoside phosphoramidites and a conventional DNA / RNA synthesizer.siRNA can be synthesized as two separate complementary RNA molecules or a single RNA molecule with two complementary regions.Commercial suppliers of synthetic RNA molecules or synthesis reagents include:

[0068] Proligo (Hamburg, Germany), Dharmacon Research (Lafayette, Colorado, USA), Pierce Chemical (part of Perbio Science, Rockford, Illinois, USA), Glen Research (Sterling, Virginia, USA), ChemGenes (Ashland, Massachusetts, USA) and Cruachem (Glasgow, UK).

[0069] Alternatively, siRNA can be expressed from recombinant circular or linear DNA plasmids using an appropriate promoter. Suitable promoters for expressing siRNA of the present invention from a plasmid include, for example, U6 or HI RNA pol III promoter sequences and cytomegalovirus promoters. The selection of other suitable promoters is within the skill of those in the art. The recombinant plasmids of the present invention can also contain inducible or regulatable promoters for expressing siRNA in specific tissues or specific intracellular environments. The siRNA expressed from the recombinant plasmid can be isolated from cultured cell expression systems using standard techniques or expressed in cells, such as breast tissue or neurons.

[0070] The siRNA of the present invention can also be expressed intracellularly from a recombinant viral vector. The recombinant viral vector contains a sequence encoding the siRNA of the present invention and a suitable promoter for expressing the siRNA sequence. Suitable promoters include, for example, the U6 or HI RNA pol III promoter sequence and the cytomegalovirus promoter. The selection of other suitable promoters is within the skill of those in the art. The recombinant viral vector of the present invention can also contain an inducible or regulatable promoter for expressing the siRNA in tumor-localized tissues.

[0071] As used herein, an "effective amount" of siRNA is an amount sufficient to cause RNAi-mediated degradation of the target mRNA or to reduce NKG2D signaling. RNAi-mediated degradation of the target mRNA can be detected by measuring the amount of target mRNA or protein in the cells of a subject using standard techniques for isolating and quantifying mRNA or protein, as described above.

[0072] Morpholino antisense oligonucleotides in zebrafish and frogs have been shown to overcome limitations of RNase H-competent antisense oligonucleotides, including numerous nonspecific effects due to nontarget-specific cleavage of other mRNA molecules resulting from the less stringent requirement for RNase H. Morpholino oligomers therefore represent an important new class of antisense molecules. The oligomers of the present invention can be synthesized using standard methods known in the art. For example, phosphorothioate oligomers can be synthesized as described in Stein et al. (1988) Nucleic Acids Res. 16, 3209-3021. Methylphosphonate oligomers can be prepared using controlled pore glass polymer supports (Sarin et al. (1988) Proceedings of the National Academy of Sciences of the United States of America 85, 7448-7451). Morpholino oligomers can be synthesized as described in U.S. Patent Nos. 5,217,866 and 5,185,444 by Summerton and Weller.

[0073] Inhibition, particularly transient inhibition, can also be achieved at the protein level by inhibitors, typical examples of which are antibodies against chimeric NKG2D receptors or antibodies against one or more NKG2D ligands.

[0074] The term "antibody" refers to an antibody characterized by its specificity for an NKG2D receptor, NKG2D ligand, or a functional derivative thereof, and is preferably a monoclonal antibody; or an antigen-binding fragment thereof in the form of F(ab'), F(ab), or single-chain Fv, or a recombinant antibody derived therefrom. These antibodies of the present invention (including specific polyclonal antisera prepared against the target protein and their functional derivatives) are free of cross-reactivity with other proteins. Monoclonal antibodies of the present invention can be produced, for example, by classical methods from splenocytes of animals, specifically mice or rats, immunized against the target protein or its functional derivative, and cells of myeloma cell lines, and then selected by hybridomas for their ability to produce monoclonal antibodies that recognize the target protein or its functional derivative originally used to immunize the animal. Monoclonal antibodies according to this embodiment of the present invention may also be humanized mouse monoclonal antibodies generated using recombinant DNA technology, diverging from mouse and / or human genomic DNA sequences encoding the heavy and light chains or cDNA clones encoding the heavy and light chains. Alternatively, the monoclonal antibodies according to this embodiment of the present invention may be human monoclonal antibodies. Such human monoclonal antibodies may be prepared, for example, by human peripheral blood lymphocyte (PBL) repopulation of severe combined immunodeficiency (SCID) mice as described in PCT / EP99 / 03605, or by using transgenic non-human animals capable of producing human antibodies as described in U.S. Patent No. 5,545,806. Fragments derived from these monoclonal antibodies, such as Fab, F(ab)'2, and scFv ("single-chain variable fragment"), also form part of the present invention, provided they retain their original binding properties. Such fragments are typically generated by enzymatic digestion of the antibody with, for example, papain, pepsin, or other proteases. Those skilled in the art will recognize that monoclonal antibodies or their fragments can be modified for various purposes. The antibodies involved in the present invention can be labeled with an appropriate label, such as an enzymatic, fluorescent, or radioactive label.In a particular embodiment, the antibody against a target protein or a functional fragment thereof is derived from a camel. Camelid antibodies are fully described in WO94 / 25591, WO94 / 04678 and WO97 / 49805.

[0075] Other inhibitors of NKG2D signaling at the protein level include, but are not limited to, peptide inhibitors of NKG2D ligands, peptide inhibitors of chimeric receptors, peptide aptamers of NKG2D ligands (Tomai et al., J Biol Chem. 2006) inhibitors, peptide aptamers of NKG2D ligands of chimeric NKG2D receptors, and protein interfering agents or Pept-lns™, which are described in WO2007 / 071789 and WO2012 / 123419, the contents of which are incorporated herein by reference.

[0076] Another method for protein-level inhibition is to interfere with secretory transport, preventing receptors and / or ligands from reaching the plasma membrane. Typically, this is a temporary form of inhibition, and the receptors can return to their normal cellular location upon receiving the appropriate signal. One example of a method based on this principle is the RUSH (Retention with Selective Hooks) system (Boncompain et al., Nature Methods 2012 and WO2010142785), which specifically targets the transient inhibition of chimeric NKG2D receptors.

[0077] Small molecule inhibitors (eg, small organic molecules) and other drug candidates can be obtained, for example, from combinatorial and natural product libraries.

[0078] For transient inhibition of functional NKG2D signaling, inhibition of downstream signaling via chimeric NKG2D receptors is particularly contemplated. It has been demonstrated herein that a significant portion of the observed fratricide is mediated via the NKG2D-induced PI3K signaling pathway, which is the primary signaling pathway of the NKG2D / DAP10 complex. Therefore, inhibition of PI3K signaling effectively inhibits the function of NKG2D signaling, since it reduces the functional effect of ligand-receptor binding. Thus, transient inhibition of downstream signaling can be achieved by transient inhibition of PI3K signaling. Examples of inhibitors include commercially available PI3K inhibitors. One particularly contemplated inhibitor is the broad-spectrum PI3K inhibitor LY294002. Another particularly contemplated inhibitor is Cal101 (idelalisib). Other examples include copanlisib, taselisib, buparlisib, duvelisib, alpelisib, and umbralisib.

[0079] The methods described herein are applicable during the production of NKG2D-expressing immune cells. Typically, immune cell production occurs when cells are prepared or cultured for adoptive transfer. This can be autologous adoptive transfer (a subject receives their own cells that have been modified and / or expanded) or allogeneic adoptive transfer (a subject receives cells from a different individual).

[0080] Many different types of immune cells are used in adoptive therapy and are contemplated for use in the methods described herein. Immune cells include, but are not limited to, T cells, NK cells, NKT cells, lymphocytes, stem cells, and iPSCs. While the latter two are not immune cells, they can be used in adoptive cell transfer for immunotherapy (see, e.g., Jiang et al., Cell Mol Immunol 2014; Themeli et al., Cell Stem Cell 2015). Typically, production begins with stem cells or iPSCs (or may begin with a dedifferentiation step from immune cells to iPSCs), but production necessarily involves a differentiation step into immune cells prior to administration. Because the methods of the present invention relate to the production process (i.e., steps prior to administration), stem cells and iPSCs used in the production of immune cells for adoptive transfer are considered immune cells herein. According to certain embodiments, the stem cells contemplated in the methods do not involve the destruction of a human embryo.

[0081] T cells and NK cells are particularly contemplated as cells for use in the methods of the present invention.

[0082] In a further aspect, genetically engineered immune cells are provided in which fratricide is reduced and / or prevented, characterized by functional inhibition of NKG2D signaling in the cells, e.g., via knockout of an NKG2D ligand, permanent or transient inhibition of an NKG2D ligand, or transient inhibition of a chimeric receptor (e.g., transient expression of an inhibitor, transient expression of an inhibitor, or transient inhibition by temporary retention of the receptor or ligand).

[0083] To this end, these cells comprise a nucleic acid molecule encoding a chimeric NKG2D receptor and at least one of the following: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - one or more inhibitors of a chimeric NKG2D receptor and / or one or more NKG2D ligands; - A binding tag fused to the chimeric NKG2D receptor and / or one or more NKG2D ligands.

[0084] Methods such as the RUSH (Retention by Selective Hooks) system (Boncompain et al, Nature Methods 2012 and WO2010142785) can use binding tags (e.g., streptavidin), and transient inhibition of chimeric NKG2D receptors is specifically envisioned.

[0085] Specifically contemplated are cells comprising a nucleic acid molecule encoding a chimeric NKG2D receptor and at least one of the following: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - one or more inhibitors of a chimeric NKG2D receptor and / or one or more NKG2D ligands.

[0086] Endogenous gene inactivation can be achieved using, for example, genome editing, engineered nucleases (CRISPR / Cas, TALEN, zinc finger nucleases, meganucleases, MegaTAL nucleases, etc.), or other suitable methods (including, but not limited to, Cre / Lox or Flp / FRT-based systems), as described above. While not a requirement, in most cases, the inactivated endogenous gene will be permanently inactive (i.e., there is no obvious reversal of inactivation). For this reason, this method is particularly suitable for inactivating ligands (not required for immune cells to exert immunotherapeutic effects) but is less suitable for inactivating the NKG2D receptor (since receptor function is required for immunotherapy).

[0087] Cells containing inhibitors typically contain a plasmid encoding the inhibitor. This is the most convenient way to ensure that the inhibitor is contained in the cells. For this reason, it is particularly envisioned that the inhibitor can be expressed from a plasmid. This can be done with antibodies or peptides, but most particularly envisioned are nucleic acid inhibitors, for example, RNA interference techniques (such as siRNA or shRNA). The inhibitor (such as an RNA inhibitor) can be directed against a chimeric NKG2D receptor and / or one or more NKG2D ligands. These NKG2D ligands are selected from the group consisting of MICA, MICB, ULBP1, ULBP2, ULBP3, ULBP4, ULBP5, and ULBP6.

[0088] The cells can be provided as is or as a composition. The composition can be provided such that NKG2D signaling is functionally inhibited by adding an (external) inhibitor to the cells—either internalized by the cells or performing its inhibitory function extracellularly. Thus, according to a further embodiment, the composition comprises immune cells expressing a chimeric NKG2D receptor and, further, an inhibitor. The cells can contain one or more endogenous genes encoding inactivating NKG2D ligands and / or one or more inhibitors for the chimeric NKG2D receptor and / or one or more NKG2D ligands encoded as nucleic acids in the cells and / or binding tags fused to the chimeric NKG2D receptor and / or one or more NKG2D ligands (the "internal inhibitors" detailed above).

[0089] Additionally or alternatively, the composition may include one or more inhibitors of the chimeric NKG2D receptor and / or one or more NKG2D ligands (not encoded as nucleic acid in the cell) and / or inhibitors of downstream signaling of the chimeric NKG2D receptor ("external inhibitors").

[0090] Antibodies against the chimeric NKG2D receptor and / or one or more NKG2D ligands are particularly contemplated as inhibitors of these proteins. PI3K inhibitors are particularly contemplated as inhibitors of downstream signaling of the chimeric NKG2D receptor.

[0091] In either case, the inhibitor may be contained intracellularly. Alternatively, the composition may comprise the immune cells and the inhibitor as separate components (i.e., extracellular). However, although the composition can be provided as separate components, the inhibitor may be taken up by the immune cells. For example, inhibitors of downstream signaling, such as PI3K inhibitors, are often small molecules that are easily taken up by cells. Antibodies may or may not be taken up by cells, but because the interaction between the NKG2D receptor and its ligand occurs extracellularly, cellular uptake is not a prerequisite for inhibition (e.g., competitive inhibitors may function extracellularly).

[0092] According to a further aspect, the genetically engineered immune cells or compositions described herein are provided for use as pharmaceuticals. According to yet a further aspect, the genetically engineered immune cells or compositions described herein are provided for use in treating diseases characterized by NKG2D ligand expression. It is well documented that NKG2D ligands, such as MICA and MICB, or members of the RAET1 / ULBP family, are inducible autoantigens, i.e., cellular ligands expressed under abnormal conditions or conditions of cellular stress, most particularly in stressed (e.g., inflamed), transformed, or infected cells. Thus, the cells or compositions are provided for use in treating a disease selected from inflammatory diseases, cancer, or infectious diseases (e.g., viral, bacterial, or fungal infections). Life-threatening diseases are particularly contemplated, as cell therapy is quite expensive. Thus, most particularly, the cells and compositions described herein are provided for use in cancer treatment. In principle, all cancers can be treated. These include, but are not limited to, bladder cancer, brain cancer, breast cancer, cervical cancer, colon cancer, esophageal cancer, glioblastoma, head and neck cancer, kidney cancer, leukemia, liver cancer, lung cancer, lymphoma, melanoma, mesothelioma, multiple myeloma, ovarian cancer, pancreatic cancer, prostate cancer, sarcoma, gastric cancer, and thyroid cancer. Most particularly contemplated cancers include leukemia (including AML), multiple myeloma, bladder cancer, breast cancer, colon cancer, ovarian cancer, and pancreatic cancer. These seven cancers typically have high NKG2D ligand expression.

[0093] "The cells and compositions are provided for use in therapy" has the same meaning as "a method of treating a disease is provided, comprising the step of administering these cells or compositions to a subject in need thereof." Thus, methods of treating inflammatory diseases, cancer, or infectious diseases are provided, comprising administering the cells.

[0094] Particularly contemplated is a method of treating cancer in a subject in need thereof, comprising administering to the subject genetically engineered immune cells, wherein the immune cells contain a nucleic acid molecule encoding a non-native chimeric NKG2D receptor and at least one of the following: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - one or more inhibitors of a chimeric NKG2D receptor and / or one or more NKG2D ligands; - A binding tag fused to the chimeric NKG2D receptor and / or one or more NKG2D ligands.

[0095] Even more particularly contemplated is a method of treating cancer in a subject in need thereof, comprising administering to the subject genetically engineered immune cells, said immune cells comprising a nucleic acid molecule encoding a non-native chimeric NKG2D receptor and at least one of the following: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - one or more inhibitors of a chimeric NKG2D receptor and / or one or more NKG2D ligands.

[0096] Also provided is a method of treating cancer in a subject in need thereof, comprising administering to the subject a composition comprising immune cells that express a chimeric NKG2D receptor and one or more of the following: one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; and / or - one or more inhibitors of a chimeric NKG2D receptor and / or one or more NKG2D ligands encoded as nucleic acids in the cell, and / or - a binding tag fused to a chimeric NKG2D receptor and / or one or more NKG2D ligands; and / or - one or more inhibitors of a chimeric NKG2D receptor (not encoded as a nucleic acid in a cell) and / or one or more NKG2D ligands; and / or -Inhibitors of downstream signaling of chimeric NKG2D receptors.

[0097] The immune cells may be autologous to the subject to whom they are administered, or may be allogeneic, ie, derived from another subject.

[0098] While the cells and methods of the present invention have been described in terms of specific embodiments, specific configurations, and materials and / or molecules, various changes and modifications in form and detail may be made without departing from the scope and spirit of the invention. The following examples are presented to further illustrate specific embodiments and are not intended to limit the scope of the present application. The present application is limited only by the scope of the claims.

[0099] Example preface Given the large number of NKG2D ligands that can be expressed in T cells, gene editing to eliminate all ligand expression was not considered the most efficient option. Therefore, alternatives to fratricide control have been explored to facilitate the delivery of NKG2D-targeted CART cell therapies. Two different approaches have been explored, using signaling inhibitors or antibody-based approaches. Both resulted in inhibition of fratricide, although to different degrees.

[0100] The inclusion of a phosphoinositol-3-kinase inhibitor (LY294002) blunted the fratricide effect, providing a general means of generating NKR-2CART cells. The use of a PI3K inhibitor further enhanced NKR-2-driven differentiation potential and shifted the cells toward a memory phenotype. A target-specific approach involving antibody blockade of the CAR itself induced a further improvement in NKR-2CART cell yield, along with a reduction in differentiation potential and a shift in the CD4 / CD8 ratio. These factors could be successfully biased in vitro to enhance differentiation potential and alter cell phenotype by delaying the addition of blocking antibodies. Despite differences in approach, inhibitor- or antibody-based approaches generated NKR-2CART cells with very similar phenotypes and in vivo activities. Finally, although it was not feasible to block all NKG2D ligands, transient (shRNA) or permanent (Crispr / Cas) inhibition of the two most important ligands generated NKR-2CART cells with very similar phenotypes.

[0101] These results demonstrate that target-driven fratricide can be overcome by using a different approach that allows the development of T cell therapies where self-expression of target ligands is a limiting factor.

[0102] material and method Antibodies and flow cytometry Following standard protocols, fluorescently labeled CD3 (BD, 345766), CD4 (BD, 345809), CD8 (BD, 345772), CD314 (BD, 558071), CD45RA (BD, 550855), CD62L (BD, 555544), CD279 (eBioscience, 12-2799-42), CD19 (BD, 345791), CD223 (eBioscience, 25-2239-41), MICA / B (R&D Systems, FAB13001G-100), MICB (R&D Systems, FAB1599G), ULBP1 (R&D Systems, FAB1380C), ULBP2 / 5 / 6 (R&D Systems, FAB1298A), and ULBP3 (R&D Systems, FAB13001G-100) were used. Cells were stained with antibodies against ULBP4 (R&D Systems, FAB1517P), ULBP4 (R&D Systems, FAB6285A), and the corresponding isotypes. Briefly, cells were harvested and resuspended in DPBS (Life Technologies, A1285801) buffer supplemented with 5% human serum albumin (Octapharma, 68209-633-02) and 0.01% NaN3 (Sigma, S2002). Cells were incubated with antibodies for 30 minutes at 4°C, washed with PBS, and analyzed using a Guava easyCyte 6HT cytometer (Millipore). All antibodies were titrated before use. Viable cells were selected based on FSC / SSC. Unlabeled and isotype controls were included in all cases. Analysis was performed using FlowJo v10.

[0103] Plasmid and vector production The chimeric NKG2D (chNKG2D) construct was generated as previously described (Zhang, Barber, & Sentman, 2006) and cloned into the Mo-MLV-based oncoretroviral vector SFG between the Ncol and Xhol restriction sites. The pSFG GFP plasmid and pSFG htCD19.1 (encoding a truncated form of human CD19 (tCD19)) were gifts from Celdara Medical LLC (Lebanon, NH, USA). GP2-293 packaging cells were transiently transfected with the VSV-G envelope plasmid. PG13 cells were spin-transduced with the retroviral suspension generated in PG2-293 cells to obtain a stable producer cell line. The vector particles used to transduce human T lymphocytes were harvested from PG13 stable producer cells after the cultures reached confluence. Vector titers were measured using the Retro-X™ qRT-PCR titration kit (Life Technologies, CL 631453).

[0104] NKG2D-CART cell generation Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood of healthy donors (ImmuneHealth, CHU, Tivoli) by Ficoll density gradient (VWR, 17-5442-03) according to standard procedures. Briefly, whole blood was diluted three times with DPBS and carefully added to the Ficoll layer in a 50 ml tube. The tube was centrifuged at 500 g, and the interphase was carefully removed. PBMCs were then washed three times with DPBS, collected, and subsequently activated in X-Vivo15 medium (Westburg, BE02-061Q) containing 5% human serum (Access Biologicals, 515-HI) and supplemented with 40 ng / ml OKT3 (Miltenyi, 170-076-124) and 100 IU / ml IL-2 (Miltenyi, 170-076-146). The cells were incubated for 2 days in an incubator maintained at 37°C and 5% CO. Then, 24-well plates (1 × 106 Cells were harvested from 24-well plates (24 cells / well), transduced with different viral vectors, and incubated for 2 days. Cells were then harvested from the 24-well plates, washed with HBSS (Westburg, BE10-543F), transferred to G-Rex containers (Wilson Wolf, 80040S), and expanded for an additional 4 days in complete X-Vivo 15 containing serum and IL-2, or as described in the text. At the end of the expansion phase, cells were harvested and used as appropriate.

[0105] Cell lines and culture reagents Chronic myeloid leukemia cell line K562 and pancreatic cancer cell line PANC-1 were purchased from ATCC and stored in IMDM (Westburg, LO BE12-722F) or DMEM (Westburg, LO BE12-604F) containing 10% FBS (Gibco, 16140071) and 1% penicillin / streptomycin (ThermoFisher Scientific, 15140122), respectively, until use. PI3K inhibitor LY294002 was purchased from Selleck Chemicals (S1105). Inhibitory antibodies against NKG2D or the corresponding isotype were purchased from BioLegend (inhibitory antibody (UltraleafCD314) clone: ​​1D11, ImTec Diagnostics NV, 320814).

[0106] Cytolysis assay Adherent PANC-1 cells were cultured in a phenol red-free X-Vivo15 medium containing 5% human serum (HS) in a 1:1 ratio in flat-bottom 96-well plates with or without thawed NKR-2 T cells or tCD19-transduced cells for 20 hours. T cells were washed, and the remaining adherent PANC-1 cells were labeled with Alamar Blue (ThermoFisher Scientific, DAL1025) for 4 hours. Viable cells were measured at 530 nm fluorescence using a SpectraMax M2 (Molecular Devices), and relative cytolytic activity was calculated.

[0107] Cytokine release assay Fresh NKR-2 T cells and / or control tCD19 cells were incubated with K562 or PANC-1 cells at a 1:1 ratio in X-Vivo15 containing 5% HS. After 24 h of incubation, supernatants were collected and IFN-γ was measured by ELISA (R&D Systems, SIF50) according to the manufacturer's protocol. As a positive control, cells were stimulated with PMA (Sigma-Aldrich, P8139-5MG) and ionomycin (Sigma-Aldrich, I0634-1MG). To assess background levels of activation, cells were left unstimulated.

[0108] Antibody inhibition assay NKR-2 T cells were incubated with 1 μg / mL of NKG2D-blocking antibodies, isotype control, or no antibody for 24 hours, and then NKR-2 T cell-mediated IFN-γ secretion was measured. Similarly, NKR-2 cells were cocultured with cancer cells in the presence of antibodies, and cytokine secretion was measured.

[0109] RNA extraction and qPCR assay PBMCs were stimulated with 40 ng / mL OKT3 and IL-2 (100 IU / mL) for 2 days, transduced with 40 ng / mL OKT3 and IL-2 (100 IU / mL) for an additional 2 days, cultured, and then expanded in the presence of IL-2 (100 IU / mL) until day 8 or 10 as detailed in the text. Total RNA was isolated every 2 days using the RNeasy Mini Kit (Qiagen, 74104). Quantitative PCR reactions were performed using predesigned TaqMan gene expression assays for NKG2D ligands (Hs04187752_mH, Hs01026642_ml, Hs00607609_mH, Hs00906262_ml, Hs00741286_ml, Hs01584111_mH, Hs04194671_sl, Hs00360941_ml, ThermoFisher Scientific) and Light Cycler 480 RNA master mix (Roche, 04991885001). Relative expression was based on the housekeeping gene cyclophilin using in-house designed primers (5'-GACGGCGAGCCCTTGG-3' and 5'-GC ACG A AA ATTTT CTG CTGT CTT-3') and probe (5'TEX615-TCTCCTTTGAGCTGTTTGCAGACAAGGT-3'BHQ™). Results are presented as fold induction compared to day 0 (calculated as 2^-ΔΔCT). All gene expression assays were performed in cancer cell lines known to express the ligand.

[0110] Animal experiments All in vivo experiments were performed at Voxcan (Marcy I'Etoile, France). Briefly, NOD / scid IL2rgnull (NSG) mice were irradiated 24 hours before tumor injection (day 1). On day 0, 5 × 10 IgG per mouse were injected. 6 THP-l-luc-GFP cells were engrafted by IV injection of 200 ml of PBS containing THP-l-luc-GFP cells. On day 7, THP-l-luc-GFP-positive mice were divided into four treatment groups: (i) Controls received a single IV infusion of vehicle (200 ml HBSS). (ii) 10 × 10 6 Mock tCD19 received a single IV infusion of Mock tCD19 T cells (200 μL) (iii) 10 × 10 6 NKR-2LY received a single infusion of NKR-2-LY T cells (200 ml) (iv) 10 × 10 6 NKR-2-optimized Ab received one IV infusion of NKR-2-optimized Ab (200 ml).

[0111] Tumor progression was assessed by bioluminescence imaging on days 4, 8, 15, 22, 28, and 35. Similarly, each animal was weighed three times a week starting on day 6.

[0112] statistical analysis Statistical significance was assessed using unpaired t-tests, paired t-tests, or the nonparametric Mann-Whitney U test, where possible. A p<0.05 was considered statistically significant.

[0113] Example 1 NKR-2CART cells undergo fratricide, which drives the phenotype and proliferation of engineered T cell populations.

[0114] After transduction and in vitro culture, in the absence of a method to control fratricide, the NKR-2 T cell population exhibited a predominantly CD8+ T cell subset composition compared to T cells transduced with a control vector (truncated CD19 (tCD19), Figure 1A). NKG2D expression was not limited to NKR-2 T cells, but was also evident in control tCD19 T cells; however, NKG2D expression was not confined to NKR-2 T cells, and binding of endogenous NKG2D prevented the delivery of therapeutic responses to CAR cells (32). However, the relative cell surface expression of NKG2D was significantly increased in the NKR-2 T cell population, indicating T cell transduction with the CAR construct (Figure 1B). The mean fluorescence intensity of NKG2D (CD314) was significantly higher in the NKR-2 T cell population in both the CD4+ and CD8+ subsets, confirming CAR expression in both subsets (Figure 2A-C).

[0115] Interestingly, when compared with the tCD19 control T cell population, which exhibited an increased effector cell phenotype, the NKR-2 T cell population exhibited a decreased relative frequency of naive cells (defined by double positivity for CD45RA+ and CD62L+ cells) and an increased frequency of CD279 (PD-1) and CD223 (Lag-3) (Figure 1C-D). When cocultured with cancer cell lines, NKR-2 T cells exhibited high levels of target cell-induced IFN-γ secretion (Figure 3A) and cytolytic activity (Figure 3B), confirming the functionality of NKR-2 T cells. However, the yield / proliferation rate (Figure 3C) and viability (data not shown) of NKR-2 T cells during culture and at harvest were consistently decreased compared with the tCD19 control T cells.

[0116] It is known that NKG2D receptor binding can result in natural killer cell fratricide (23, 24), and the expression of NKG2D ligands on T cells during mitogenic activation has been documented (22). This raised the question of whether NKR-2 T cell fratricide might occur after transduction, resulting in low cell yield and viability, a skewed CD4 / CD8 ratio, and increased effector memory differentiation. To test this, we transduced donor-derived T cells with an eGFP-expressing vector and mixed them with NKR-2 T cells from the same donor to examine whether conventional T cell killing occurred. After 24 hours, there was a clear decrease in eGFP T cells in the NKR-2 T cell coculture, suggesting targeted killing of autologous T cells by NKR-2 T cells (Figure 3D).

[0117] To understand the expression profile of NKG2D ligands in activated T cells, we performed qPCR analysis to examine the dynamics of NKG2D ligand mRNA expression profiles using three healthy donors as a source of untransduced activated T cells (Figure 3E). Within 2 days of T cell activation, a rapid increase in MICA and ULBP2 mRNA levels was detected. However, within the next 2 days, ULBP2 mRNA remained at a high level, while MICA rapidly decreased to baseline. ULBP3 mRNA gradually increased over time, whereas there was no relative increase in MICB, ULBP5, ULBP1, or ULBP6 transcription. Conversely, ULBP4 mRNA increased slightly on day 4 but then decreased back to baseline levels. At the cell surface protein level, MICA was transiently present on day 2, and MICA / B showed an expression pattern similar to its corresponding mRNA and subsequently declined (Figure 3F). Unfortunately, we were unable to detect individual ULBP2 / 5 and 6 proteins due to unavailability of suitable antibodies. However, the immunoreactivity observed with antibodies recognizing all three family members was likely due to ULBP2, based on its mRNA expression profile, which increased only after day 2. ULBP3 was undetectable on the cell surface, although it was highly induced at the mRNA level. ULBP4+ cells reached a peak positive expression by day 6 and returned to baseline by day 8, lagging the pattern of mRNA expression by 2 days (Figure 3F). These observations were reflected in the parallel kinetics of mean fluorescence intensity (Table 1).

[0118] Table 1: Mean fluorescence intensity (MFI) of the ligands for all three different donors after the manufacturing process and their corresponding SD (samples analyzed on days 0, 2, 4, 6, and 8).

[0119] [Table 1]

[0120] Taken together, these data imply a T cell-regulated expression of NKG2D ligands after mitogen activation with MICA, ULBP4, and putatively ULBP2 predominating at the protein level, although the kinetics of expression differ.

[0121] Example 2 PI3K inhibition improves the survival of NKR-2 T cells during cryopreservation and drives increased NKR-2 antigen-specific cytokine production and an enhanced memory phenotype.

[0122] Upon ligand binding, NKG2D and its associated DAP10, like CD28, initiate signal transduction via the PI3K pathway (25, 26). Therefore, we wondered whether inhibition of PI3K signaling could prevent NKR-2-mediated fratricide in T cell cultures. To this end, we added increasing concentrations of LY294002 (a broadly defined PI3K inhibitor) to the transduction and expansion stages of NKR-2 production.

[0123] The addition of LY294002 yielded several observations. First, cell surface expression of NKG2D on NKR-2 T cells was dose-dependently reduced, reaching the level of control tCD19 T cells at 10 μM (Figures 4A and 5). This reduction was further reversible, as NKG2 expression increased to the level of untreated NKR-2 T cells (data not shown) upon removal of LY from the culture. However, the inhibitor did not improve cell yield, suggesting that fratricidation during culture was incomplete or that the PI3K inhibitor adversely affected proliferation (Figure 3B). To assess whether LY294002 had an antiproliferative effect, control tCD19 T cells were treated with the PI3K inhibitor during culture. A clear decrease in the proliferative capacity of these control T cells was observed compared to untreated tCD19 T cells (Figure 6).

[0124] As expected, NKR-2 T cells generated with PI3K inhibitors showed increased cell viability both after cryopreservation and when stored at 4°C for 48 h (Figures 3C and 3D). NKR-2 T cells generated with PI3K inhibitors produced large amounts of IFN-γ in a LY294002-dependent manner (Figure 7A). Finally, NKR-2 T cells cultured with LY294002 also upregulated CD62L T cells, consistent with previously published studies using inhibitors to modulate T cell memory phenotypes (27, 28). + The CD45RA / CD45RA phenotype appeared to be increased (Figure 7B).

[0125] Overall, the addition of a PI3K inhibitor had a beneficial effect on the viability of NKR-2 T cells, which may be attractive for therapeutic use.

[0126] Example 3 Antibody-mediated NKG2D blockade prevents NKR-2CART cell fratricide.

[0127] Initial experiments involving the inclusion of an anti-NKG2D antibody (clone 1D11) in NKR-2 T cell cultures revealed that the NKR-2 T cell yield at the end of culture was comparable to that of control T cells (proliferation factor 2.6 for NKR-2 T cells, compared with 13.8 for antibody-blocked NKR-2 T cells) (Figure 8). This suggests that antibody blockade has the potential to prevent NKG2D target-driven fratricide. Dose titration experiments demonstrated that antibody concentrations of 2.5 μg / mL or higher protected tCD19 T cells from NKR-2 T cell target killing (Figure 4E). The anti-NKG2D antibody also effectively blocked IFN-γ release during target cell engagement (Figure 4F). This confirmed the specificity of NKR-2 T cells. The effective blockade of fratricide using anti-NKG2D antibodies was further supported by the fact that the release of IFN-γ observed during the production of NKR-2 T cells, likely due to T cell fratricide, was significantly reduced by the addition of blocking antibodies (Figure 4G). Because murine blocking antibodies potentially cause toxicity through antibody-dependent cellular cytotoxicity (ADCC), extensive washing steps were performed after harvesting. IgG ELISA and flow cytometry experiments revealed that no contaminating antibodies were detectable in the supernatant or on the cell surface after harvesting (data not shown). To assess ADCC, NK cells were co-cultured with autologous NKR-2 cells in the presence of 5 μg / mL of Ab without evidence of ADCC (data not shown). Together, these data demonstrate that the addition of anti-NKG2D blocking antibodies suppresses NKR-2 T cell CAR-driven fratricide.

[0128] Example 4 Antibody-mediated blockade of NKG2D and PI3K inhibition of NKG2D signaling are functionally equivalent in producing NKR-2 expressing cells.

[0129] Adaptation of the Ab blocking process during in vitro NKR-2 cell growth allowed for increased yields with comparable activity in vitro and in vivo.

[0130] Comparison of NKR-2 T cells generated with antibodies and those generated with PI3K inhibitors showed different cytolytic kinetics, suggesting changes in T cell characteristics following the Ab process (Figure 9A). The main difference observed between these two processes was the CD4 / CD8 ratio. Addition of a PI3K inhibitor consistently skewed the CD8 population by day 8. Interestingly, blockade of NKR-2 T cells rescued the CD4+ population, suggesting that the observed skewed CD4 / CD8 ratio is dependent on fratricide (Figure 9B). The most plausible explanations for this difference in ratio include a relative increase in the ratio due to CD4 T cell proliferation or elimination of CD4 T cells by CD8 T cells.

[0131] We hypothesized that the enhanced lytic activity of NKR-2 T cells generated using a PI3K inhibitor might be due to a decreased CD4 / CD8 ratio. To address this, we performed a blocking antibody process, adding the blocking antibody on day 6 after transduction rather than immediately (day 4). This change generated NKR-2 T cells with CD4 / CD8 ratios similar to those generated using a PI3K inhibitor (Figure 9C), while maintaining a proliferation rate comparable to that of control T cells (Figure 9D). Subsequently, despite minor differences between the two processes in certain parameters, such as expression of the activation marker CD25 and memory phenotype (data not shown), functional cytokine secretion and cytolytic activity of NKR-2 T cells against target cancer cells were comparable between the two processes (Figure 9E, Figure 9F).

[0132] Preliminary in vivo experiments using NOD SCID gamma mice administered NKR-2 cells generated by the LY and Ab processes demonstrated similar antitumor activity in an established acute myeloid leukemia (THP-1) tumor model (visualized by bioluminescence 8 days after injection: tCD19: 5.76E10 + / - 4.46E10; NKR-2LY: 7.15E08 + / - 1.01E09; NKR-2Ab: 6.43E08 + / - 1.25E08; data not shown). One-way ANOVA with Tukey's post-hoc test revealed significant differences between LY and tCD19 control cells (p = 0.02) and Ab-producing NKR-2 compared with tCD19 (p = 0.03). No differences were observed between the LY and Ab groups (p = 0.95). In addition, after 24 hours, similar engraftment was observed in both the NKR-2LY and Ab groups (LY group: 1.838±1.07%; Ab group: 1.792±0.56%; data not shown), indicating that no significant difference could be detected between the two groups in short-term engraftment.

[0133] Taken together, these combined data demonstrate that NKR-2 T cells generated by the adaptive blocking antibody process exhibit similar short-term engraftment and differentiation potential to cells generated using PI3K inhibitors.

[0134] The effect on NKR2 production is mediated by PI3K and not by a specific inhibitor.

[0135] To further confirm whether the effects of LY294002 are indeed related to its PI3K inhibitor activity, we tested several other PI3K inhibitors, including wortmannin and CAL-101 (idelalisib). Figure 10 shows representative data for CAL-101. CAL-101 behaves similarly to blocking antibodies in terms of proliferation rate (Figure 10A) and cell viability (B). As with other PI3K inhibitors, cells appear to produce more interferon (Figure 10C), which may contribute to their enhanced differentiation potential. We also tested downstream inhibitors of the PI3K pathway, such as the glycogen synthase kinase 3 beta inhibitor TWS119 and the mTOR inhibitor rapamycin. Similar results were obtained (data not shown), with increased proliferation compared to cells without NKG2D signaling inhibition, but PI3K inhibition appears to result in cells with the most desirable characteristics. This is likely due, at least in part, to the toxicity of rapamycin.

[0136] Example 5 Inhibition of NKG2D ligands also leads to improved cell yield and cytolytic activity.

[0137] NKG2D is known to bind to eight different stress-inducible ligands (NKG2DLs), which are widely present in tumors but rarely present in healthy tissues. We aimed to identify key NKG2DLs expressed on T cells upon activation. PBMCs were activated with OKT3 and anti-CD3 antibody on day 0. Every other day, we assessed the expression of eight NKG2DLs on the surface of CD4+ and CD8+ T cells (Figure 11).

[0138] Upon activation, MICA / B and MICB were upregulated on the cell surface of CD4 and CD8 T cells. Expression peaked between days 2 and 4 after activation, then declined until day 10. ULBP1 and ULBP2 were expressed at low levels, but ULBP2 was restricted to CD4+ T cells (Figure 11). There was little evidence of expression of other ligands on T cells.

[0139] Parallel studies identified MICA and MICB as the primary stimulators of NKG2D CARs (data not shown) and demonstrated that MICA and MICB are the primary NKG2DLs responsible for T cell fratricide.

[0140] Next, we investigated the feasibility of specifically targeting both MICA and MICB with a single shRNA due to their high sequence similarity. Primary T cells were transduced with different shRNAs and MICA and MICB protein expression was assessed. This screening identified two shRNAs that reduced cell surface expression of MICA and MICB (data not shown). Next, we engineered a single retroviral vector encoding an NKG2D CAR and co-expressing a candidate shRNA. We used cell proliferation rates to assess the level of fratricide in T cells engineered with an NKG2D-based CAR or shRNA co-expressing T cells (Figure 12). Engineering a single retroviral vector encoding an NKG2D CAR and shRNA generated T cells with significantly reduced in vitro fratricide compared to cells without shRNA (Figure 12), and the proliferation rate of NKG2D CART cells increased, approaching that of control T cells. We then evaluated the in vitro antitumor efficacy of NKG2D-based CART cells with and without shRNAs targeting MICA / B. Cells without shRNA showed specific killing of AML HL60 cells at different target-to-effector (E:T) ratios. However, coexpression of MICA / B shRNA #2 or #4 improved cancer cell killing, especially at low E:T ratios (Figure 13A). After 24 hours of coculture, shRNA expression improved T cell recovery, so the improved target cell killing was likely due to reduced fratricide (Figure 13B).

[0141] In conclusion, knockdown of NKG2D ligands in T cells showed the same enhanced production results as observed with NKG2D or PI3K inhibition. Similar data were obtained when MICA and MICB were permanently inactivated using Crispr / Cas (data not shown). A notable difference is that shRNA inhibition is temporary (e.g., only during the manufacturing process), whereas gene knockout (here, using Crispr / Cas) is permanent, which may or may not be desirable depending on the situation.

[0142] Consideration The recent approval of CD19C ART cell therapy for B-cell acute lymphoblastic leukemia (bALL) and diffuse large B-cell lymphoma (DLBCL) provides strong clinical validation of this approach, and CD19 + The development of CAR-based T cell therapy beyond B malignancies is gaining momentum. Target selection is crucial for the success of this therapy. Until now, identifying tumor-specific antigens has been difficult. Recent bioinformatics studies of acute myeloid leukemia (AML) combining proteomic and genomic approaches have revealed that tumor-specific cell surface antigens do not exist, suggesting that antibody-based CAR-based T cell therapy targeting AML may require a complex combinatorial targeting strategy (29). Consequently, most target antigens tested to date are tumor-associated antigens whose expression can occur on normal, healthy cells. Numerous examples include CD19 in B cell malignancies (3, 4), CD123 in AML (30), and CD7 in various solid tumors (14) and CEA (31). However, problems arise when the target antigen is persistently or transiently expressed on T cells. CAR-engineered T cells are likely to subsequently target themselves and other cells in culture, leading to T cell fratricide and effectively reducing or eliminating cell yield.

[0143] Current gene editing offers a clinically relevant method to prevent the expression of specific proteins, thereby enabling the expansion of CART cells that might otherwise undergo fratricidal transformation, such as CD7-specific CART cells (14). However, the multitarget specificity of NKG2D-based CARs means that gene editing to eliminate eight different proteins in early T cells, along with effectively expressing CAR constructs, is seemingly feasible but challenging for clinical application. Therefore, here we present an alternative strategy to enable the generation and delivery of NKG2D-focused CART cell therapeutics that avoids the fratricidal transformation that occurs during cell culture.

[0144] These examples provide a general approach to regulating fratricide by reducing NKG2D expression on the cell surface with PI3K inhibitors. To our knowledge, this is the first reported observation of this nature. The mechanism underlying the reduction of NKG2D on the cell surface after PI3K inhibitor treatment is currently unknown. What is known is that NKG2D localization on the cell surface is mediated by its binding to DAP10 (Upsahw et al. 2006). One hypothesis for the reduction of NKG2D on the cell surface is that DAP10, which is required for NKG2D-DAP10 binding, is affected by PI3K inhibitor treatment (e.g., by reducing RNA levels, inhibiting transcription, or inhibiting post-translational modifications such as glycosylation) (Park YP et al. 2011, Blood). This ultimately prevents the expression of the NKG2D-DAP10 complex on the cell surface, leading to the inhibition of fratricide.However, PI3K inhibition is also associated with decreased cell proliferation. (Aagaard-Tillery KM, Jelinek DF. Phosphatidylinositol 3-kinase activation in normal human B lymphocytes. J Immunol. 1996;156:4543-4554. 11. Fruman DA, Snapper SB, Yballe CM, et al. Impaired B cell development and proliferation in absence of phosphoinositide 3-kinase p85alpha. Science. 1999;283:393-397. 12. Shi J, Cinek T, Truitt KE, Imboden JB. Wortmannin, a phosphatidylinositol 3-kinase inhibitor, blocks antigen-mediated, but not CD3 monoclonal antibody-induced, activation of murine CD4 T cells. J Immunol. 1997;158:4688-4695. 13. Truitt KE, Shi J, Gibson S, Segal LG, Mills GB, Imboden JB. CD28 delivers costimulatory signals independently of its association with phosphatidylinositol 3-kinase. J Immunol. 1995;155:4702-4710). Thus, this inhibitor approach provides an effective solution for controlling T cell fratricide in more limited use situations where relatively low doses of CART cells are required, but where large numbers of cells are needed.

[0145] Another approach to inhibiting fratricide during NKR-2 T cell production was to use specific blocking antibodies during the expansion phase. This allowed for the control of T cell fratricide and for T cell proliferation to levels comparable to control tCD19 T cells. This method is highly dependent on the antibody used, as it requires blocking fratricide without inducing CAR activation itself (as indicated by a significantly reduced level of cytokine production during culture in the absence of target antigen). The addition of specific blocking antibodies provides a solution that allows for the large-scale expansion of NKR-2 T cells.

[0146] A major concern regarding the blocking antibody process was that expanded NKR-2CAR cells would likely become decorated with antibodies, thereby causing rapid elimination of the cells upon infusion via an antibody-dependent elimination mechanism. However, analysis clearly demonstrated that NKR-2 cells were not coated with anti-CD314 antibodies upon re-expression, suggesting that antibody binding likely resulted in the loss of NKG2D and NKR-2CAR from the cell surface. Endogenous NKG2D has been shown to undergo rapid internalization upon binding to target ligands as a mechanism for controlling NK cell activation. The observations made here suggest that in the case of NKR-2, CAR also appears to be internalized. From the perspective of adoptive T cell therapy, this is beneficial because it eliminates the need to develop specific processes to remove bound antibodies.

[0147] In addition to the observed differences between processes and the effects of PI3K inhibitors, these preclinical data are the first to confirm that NKR-2 is an effective treatment in a mouse model of AML. Furthermore, these results are consistent with other studies in which the absence of preconditioning and multiple injections of NKR-2 were the benchmarks for tumor eradication and long-term survival in treated mice. (Zhang et al., Cancer Res. 2007;67(22):11029-36; Barber et al., Gene Ther. 2011;18(5):509-16)

[0148] Furthermore, while it was not thought feasible or practical to knockdown or knockout all NKG2D ligands, similar improvements in manufacturing yield were achieved when the two most common ligands in T cells were inhibited via shRNA or CRISPR, demonstrating that even partial inhibition of NKG2D signaling can already improve fratricide.

[0149] In summary, this application discloses that T cell fratricide can be managed by general methods, such as PI3K inhibition (e.g., acting on downstream signaling), or receptor-specific approaches, such as blocking antibodies or receptor ligand depletion. Specifically, the PI3K inhibitor approach and the blocking antibody approach can be used to generate immune cell products with reduced fratricide, and each approach offers the potential advantage of being able to generate T cell products where other means of target depletion in T cell populations, such as gene editing, are difficult or not currently feasible or desirable.

Claims

1. 1. A method for reducing and / or preventing fratricide during the production of T cells expressing a chimeric NKG2D receptor, comprising functional inhibition of NKG2D signaling during the production process of said cells, The functional inhibition of the NKG2D signaling is - permanent or transient inhibition of one or more NKG2D ligands of said T cells; - transient inhibition of said chimeric NKG2D receptor; - transient inhibition of downstream signaling of said chimeric NKG2D receptor; The method is achieved by one or more of the following.

2. 10. The method of claim 1, wherein the permanent inhibition of one or more NKG2D ligands is achieved by gene knockdown.

3. The method of claim 1 or 2, wherein the transient inhibition of downstream signaling is transient inhibition of PI3K signaling.

4. 4. The method of claim 3, wherein the PI3K signaling is inhibited using a PI3K inhibitor selected from LY294002 and idelalisib.

5. The method of any one of claims 1 to 4, wherein said functional inhibition is achieved using shRNA or antibodies against the NKG2D receptor or against one or more of its ligands.

6. 1. A genetically engineered T cell comprising a nucleic acid molecule encoding a chimeric NKG2D receptor and at least one of the following, wherein the T cell has reduced and / or prevented fratricide: - one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; - one or more shRNAs against one or more NKG2D ligands.

7. 7. The genetically engineered T cell of claim 6, wherein the NKG2D ligand is selected from MICA, MICB, ULBP1, ULBP2, ULBP3, ULBP4, ULBP5 and ULBP6.

8. 1. A composition comprising T cells containing a nucleic acid molecule encoding a chimeric NKG2D receptor, wherein the T cells have reduced and / or prevented fratricide, The cells one or more endogenous genes encoding NKG2D ligands that have been genetically engineered to be inactive; or - one or more shRNAs against one or more NKG2D ligands; Including, and / or the composition comprises: - a composition comprising an inhibitor of downstream signaling of said chimeric NKG2D receptor, or a PI3K inhibitor.

9. The method for producing T cells according to claim 6, comprising: A method for producing T cells containing a nucleic acid molecule encoding a chimeric NKG2D receptor, comprising adding an inhibitor of downstream signaling of said chimeric NKG2D receptor or a PI3K inhibitor to the transduction and expansion steps.

Images