CXCR3 overexpression in CAR-NK cells to elicit migration / homing to tumor microenvironment

Abstract

Nucleic acid constructs encoding CXCR3 and CAR and cells containing the same for use in the treatment of cancer are disclosed.

Description

[0001] Government licensing rights

[0002] This invention was made with government support under Grant No. R01CA190394 awarded by the National Institutes of Health. The government has certain rights in this invention.

[0003] CROSS-REFERENCE TO RELATED APPLICATIONS

[0004] This application claims the benefit of U.S. Provisional Application Serial No. 63 / 340,217, filed on May 10, 2022, under 35 U.S.C. §119(e), which is incorporated herein by reference in its entirety.

[0005] Sequence Listing

[0006] This application contains a sequence listing that has been submitted electronically in XML format and is incorporated herein by reference in its entirety. The XML copy, created on March 13, 2023, is named 52095_765001WO_ST.xml and is 78KB in size.

[0007] BACKGROUND OF THE DISCLOSURE

[0008] Activation of the antiviral stimulator of interferon genes (STING) pathway promotes anti-tumor immunity. However, STING agonists have not yet achieved clinical success.

[0009] Activation of innate anti-tumor immunity, including the STING pathway, can overcome barriers to therapeutic response, such as immune rejection and exhaustion. Clinical development of STING agonists has focused on stimulating myeloid cells into CD8-positive T cells, which reject syngeneic tumors transplanted into mice (Corrales et al., Cell Rep. 11: 1018-30 (2015); Sivick et al., Cell Rep. 25: 3074-3085 (2018); Amouzegar et al., Cancers (Basel) 13: 2695 (2021)). However, STING activation induces T cell stress, cell cycle arrest, and death (Cerboni et al., J. Exp. Med. 214: 1769-1785 (2017); Larkin et al., J. Immunol. 199: 397-402 (2017); Gulen et al., Nat. Commun. 8: 427 (2017)), which limits its clinical activity. Human tumors also undergo months to years of immune editing, making cross-species extrapolation of STING-induced T cell killing mechanisms fragile (O'Donnell et al., Nat. Rev. Clin. Oncol. 16: 151-167 (2019)). Despite these limitations, recent mouse studies have explored the complex interplay of STING signaling in the tumor immune microenvironment (TIME), identified novel effector mechanisms including NK cells, and juxtaposed the importance of STING activity in immune cells with that in tumor cells (Sivick et al., Cell Rep. 25:3074-3085 (2018); Marcus et al., Immunity 49:754-763 (2018); Nicolai et al., Sci Immunol 5:eaaz2738 (2020); Chen et al., Nature 533:493-498 (2016); Sen et al., Cancer Discov. 9:646-661 (2019)).

[0010] Chimeric antigen receptor (CAR)-expressing cells have shown significant efficacy and improved patient outcomes and have been approved by the FDA for the treatment of liquid tumors. In contrast, CAR T and CAR NK cell therapies are less effective against solid tumors for a number of reasons, including the presence of an immunosuppressive TME, vascular barriers, lack of chemokine gradients, and dysregulated immune cell trafficking. Therefore, there is an urgent need for cell therapies that are effective in treating solid tumors.

[0011] Overview of the Disclosure

[0012] In one aspect, the present disclosure provides a nucleic acid construct comprising a first nucleic acid comprising a promoter operably linked to a nucleic acid encoding a CXC motif chemokine receptor 3 (CXCR3) and a second nucleic acid encoding a chimeric antigen receptor (CAR), wherein the CAR comprises a ligand binding domain comprising a single-chain antibody fragment that binds to an antigen on a tumor cell, a transmembrane domain, and an intracellular domain comprising a signaling domain.

[0013] In another aspect, the present disclosure provides a vector containing (eg, having integrated or cloned therein) a nucleic acid construct.

[0014] On the other hand, the present disclosure provides a genetically engineered immune cell comprising one or more vectors containing a nucleic acid encoding CXCR3 and a nucleic acid encoding CAR. In some embodiments, the genetically engineered immune cell is a NK cell.

[0015] In another aspect, the present disclosure provides a pharmaceutical composition comprising an effective number of genetically engineered immune cells expressing a vector and a pharmaceutically acceptable carrier.

[0016] In yet another aspect, the present disclosure provides a method of treating cancer. The method requires administering an effective amount of a pharmaceutical composition to a subject in need thereof. In some embodiments, the method further requires administering an effective amount of a STING agonist to the subject prior to, substantially simultaneously with, or after administration of the pharmaceutical composition.

[0017] The working examples disclosed herein show that NK cells resist STING-mediated cytotoxicity, and that simultaneous contact with STING agonists enhances NK cell migration and killing, improving their therapeutic activity. This effect is further enhanced in genetically modified NK cells that overexpress CXCR3 and / or contain anti-mesothelin CARs. The working examples further show that malignant pleural mesothelioma cells robustly express STING, and that MPM cells respond to STING agonist treatment and ex vivo adoptive cell therapy.

[0018] BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1A-Figure 1D showed that STING was highly expressed in immune-exhausted MPM. Figure 1A is a set of dot plots and immunohistochemistry (IHC) micrographs. Figure 1B is a bar graph showing flow cytometry of immune cells from MPM specimens. Figure 1C and Figure 1D is a set of dot plots showing flow cytometry from freshly resected MPM specimens.

[0020] Figure 2A-2GSTING agonists were shown to promote anti-tumor immunity in MPM. Figure 2A is a schematic diagram illustrating the generation of patient-derived organotypic spheroids (PDOTS). Figure 2B This is a set of Hoechst / propidium iodide micrographs. Figure 2C is a set of dot plots showing a summary of the change in percent viable cell area. Figure 2D Waterfall plot of specimens from 35 patients treated with ADU-S100. Figure 2E is a waterfall plot showing sample #34 after treatment with ADU-S100. Figure 2F Waterfall plots of specimens from 13 patients treated with ADU-S100, TAK-676, or control. Figure 2G is a dot plot showing the live / dead percentage for sample #26.

[0021] Figure 3A-3D STING agonists were shown to activate tumor cells and fibroblasts. Figure 3A is a combined UMAP plot of extensive clustering from scRNA-sequencing MPM specimen #26. Figure 3B It is a combined UMAP plot of CXCR3 ligands (CXCL9 / CXCL10 / CXCL11) and mesothelin (MSLN). Figure 3C is a set of violin plots for selected ISG transcripts. Figure 3D is a set of UMAP plots and bar graphs from the combined samples, with contour plots showing the cell density in each individual sample superimposed on the bar graphs.

[0022] Figure 4A-4D STING agonists were shown to be toxic to T cells but not to NK cells. Figure 4A are dot plots showing immunoflow cytometry in MPM. Figure 4B is a set of bar graphs showing cell titer luminescence proliferation. Figure 4C is a set of bar graphs showing flow cytometry after 72 hours of treatment. Figure 4D is the western blot in TIL and NK cells.

[0023] Figure 5A-5D STING agonists were shown to enhance NK cell therapy. Figure 5A This is a schematic diagram of NK cell therapy. Figure 5B is a set of micrographs and dot plots showing representative Live / Dead IF and quantification of sample #37. Figure 5C is a set of bar graphs showing the quantification of the change in percent viable cell area. Figure 5D A set of representative live IF / dead IF micrographs from sample #32.

[0024] Figure 6A-6Dshowed that STING agonists enhanced adoptive NK cell migration and killing. Fig. 6A is a set of micrographs and dot plots showing superimposed IF and brightfield images. Figure 6B are dot plots showing quantification of tripartite NK cell migration. Figure 6C are micrographs and dot plots showing representative immunofluorescence modeling with NK cell migration. Fig.6D is a set of flow cytometry images showing Annexin V and Live / Dead staining.

[0025] Figure 7A-7E MPM STING and immune signatures are shown. Fig. 7A This is a dot plot of STING IHC. Figure 7B is a micrograph of STING IHC in normal pleura. Figure 7C is a set of micrographs of phospho-IRF3 (pIRF3) IHC. Fig.7D and Fig. 7E are bar graphs and dot plots showing flow cytometry from freshly resected MPM specimens.

[0026] Figure 8A-8F Shown is the expression and activation of STING in MPM cell lines. Fig. 8A are bar graphs and Western blots showing CXCL10 ELISA. Figure 8B is a set of violin plots showing mRNA expression data. Figure 8C is a dot plot showing 2'3' cGAMP ELISA from MPM cells. Fig.8D is a set of Western blots in MPM cell lines treated with ADU-S100. Fig. 8E is a Western blot of STING pathway components. Fig.8F is the CellTiter-Glo viability assay in MPM cell lines.

[0027] Figure 9A-9C Ex vivo STING agonist treatment of MPM tumors was shown with PDOTS. Fig. 9A is a set of bar graphs and dot plots showing Hoechst / propidium iodide cell area and percent live / dead. Fig. 9B are dot plots and line graphs showing CXCL10 ELISA from MPM explants. Fig. 9C is a dot plot showing summary of MPM PDOTS cell death.

[0028] Figures 10A-10E It was shown that STING agonists activated STING in tumor cells, leading to CD8 cell killing. Fig. 10Ais a set of dot plots showing the correlation between CD8 flow cytometry or CD8 immunofluorescence of MPM PDOTS and subsequent live / dead responses. Fig. 10B is a set of dot plots showing Hoechst / propidium iodide cell area and percent live / dead quantification. Fig. 10C are dot plots showing flow cytometry of S3 spheroid fragments. Fig. 10D is a set of pie charts showing MPM immuno-flow cytometry. Fig.10E is a set of IRF3 immunofluorescence micrographs after treatment with ADU-S100.

[0029] Figures 11A-11D Shown that scRNA-seq illustrates STING activation in tumor cells and fibroblasts. Fig.11A is a heatmap of a set of cluster-defined genes. Fig. 11B Violin plots of immune cells and fibroblasts. Fig. 11C It is a volcano plot of differentially expressed genes after 24 hr treatment with ADU-S100. Fig.11D A set of UMAP plots focusing on specific transcripts and markers of effector cell cluster 2.

[0030] Figure 12A-12C We show that scRNAseq suggests STING agonist toxicity in T cells. Fig. 12A Figure 2. Composite UMAP plot of extensive clustering of scRNA sequencing from MPM specimens. Fig. 12B It is a violin plot of selected NK cell activation transcripts / inhibition transcripts. Fig. 12C is a bar graph showing the scores of Treg transcript expression.

[0031] Figure 13A-13G STING agonists were shown to be toxic to T cells but not to NK cells. Fig.13A is a set of dot plots and line graphs showing CD3 / CD56 flow cytometry. Fig. 13B is a set of dot plots showing the activity of CellTiter-Glo. Fig. 13C is a set of dot plots showing CD4 / CD8 flow cytometry. Fig.13D is a set of dot plots showing the time course of CD4+ T cell cytotoxicity. Fig.13E is a set of dot plots showing the mean fluorescence intensity (MFI) of autophagolysosomal vacuoles by flow cytometry. Fig.13F is a Western blot from a stage II NSCLC sample. Figure 13G is a Western blot of bulk NK cells treated with ADU-S100.

[0032] Figure 14A-Figure 14Cshowed that STING agonists enhanced NK cell killing in MPM PDOT. Fig.14A is a set of bar graphs showing Hoechst / propidium iodide staining. Fig. 14B is a set of IF micrographs showing Hoechst / propidium iodide live / dead quantification. Fig. 14C is a set of IF micrographs and bar graphs showing quantification of live / dead cell area.

[0033] Figures 15A-15F STING agonists were shown to enhance NK cell migration and killing. Fig.15A is a set of four plots showing CXCL10 ELISA. Fig. 15B is a dot plot showing granzyme B ELISA. Fig. 15C is a set of line graphs showing flow cytometry of H2591MPM cells co-cultured with NK cells. Fig.15D is a set of micrographs and dot plots showing representative IF images of primary NK cells. Fig.15E is a set of micrographs of representative IF images of NK cell migration. Fig.15F is a set of schematic diagrams and IF micrographs used for the 3D migration assay.

[0034] Figure 16A-16C showed that STING agonists enhanced CAR-NK cell killing in MPM. Fig.16A is a schematic diagram showing NK cell isolation and transduction using anti-mesothelin (MSLN) CAR constructs. Fig. 16B is a set of flow cytometry images showing Annexin V and Live / Dead staining of H2591 MPM cells. Fig. 16C is a set of bar graphs and histograms showing the quantification of two NK cell donors.

[0035] Figure 17A-17B CXCR3 on the surface of NK cells is shown. Fig.17A are two flow cytometry graphs showing CXCR3 staining on primary NK cells and JURKAT CXCR3+ cells before and after hCXCL10 treatment. Fig. 17B is a line graph showing the mean fluorescence intensity (MFI) of CXCR3 staining on primary NK cells and JURKAT CXCR3+ cells.

[0036] Figure 18A-18B CXCR3 on the surface of NK cells is shown. Fig.18A are two flow cytometry graphs showing CXCR3 staining on NK92 cells and JURKAT CXCR3+ cells before and after hCXCL10 treatment. Fig.18Bis a line graph showing the mean fluorescence intensity (MFI) of CXCR3 staining on NK92 and JURKAT CXCR3+ cells.

[0037] Figures 19A-19D showed that CXCR3 was removed from the surface of NK cells. Fig.19A is a flow cytometry graph showing CXCR3 surface expression of primary NK cells stimulated with recombinant human CXCL10, CAR, and CAR-CXCR3 at 0 and 60 minutes after stimulation. Fig.19B is a bar graph of the data shown in FIG. 18C , with 0 min after stimulation represented as 0 and 60 min represented as 1. Fig.19C is a flow cytometry graph showing CXCR3 expression on cNK cells. Fig.19D is a flow cytometry graph showing CXCR3 staining on CIMLNK cells.

[0038] Figure 20A-20C cNK cell migration is shown. Fig. 20A is a schematic diagram showing the arrangement of a 3D model for evaluating immune cell interactions and trafficking using a microfluidic device. Fig. 20B is a bar graph showing quantification of triplicate experiments of NK cell migration into H226MPM cancer cell spheroids. Fig. 20C Two representative immunofluorescence (IF) images of primary CAR-expressing NK cells labeled with erythrocyte tracer with and without CXCR3 overexpression migrating toward H226MPM cancer cell spheroids after 3 days of culture (scale bar = 150 μm).

[0039] Figure 21A-21B cNK cell migration is shown. Fig.21A is a bar graph showing quantification of triplicate experiments of NK cell migration into H2591 cancer cell spheroids. Fig. 21B Are two representative IF images of primary CAR-expressing NK cells labeled with erythrocyte tracer migrating toward H2591 cancer cell spheroids with and without CXCR3 overexpression after 3 days in culture (scale bar = 150 μm).

[0040] Figure 22A-22B cNK cell migration is shown. Fig.22A is a bar graph showing quantification of triplicate experiments of NK cell migration with and without CXCR3 overexpression toward H226 cancer cell spheroids treated with and without ADU-S100. Fig. 22B yes Fig.22A Four micrographs of the quantified data.

[0041] Figure 23A-23D cNK cell migration is shown. Fig.23Ais a bar graph showing quantification of triplicate experiments of NK cell migration with and without CXCR3 overexpression toward H2591 cancer cell spheroids treated with and without ADU-S100. Fig. 23B yes Fig.23A Four micrographs of the quantified data. Fig.23C is a bar graph showing quantification of triplicate experiments of NK cell migration into H2591 cancer cell spheroids after 3 days of treatment with control (dH2O) or 50 mM ADU-S100. Fig.23D Representative IF images (scale bar = 150 μm) of primary CAR-expressing NK cells migrating toward H2591 cancer cell spheres with and without CXCR3 overexpression (upper panel) and with and without ADU-S100 treatment (lower panel) labeled with an erythrocyte tracer.

[0042] Figure 24A-Figure 24D Demonstrating CAR-NK cell migration and killing. Fig.24A is a bar graph showing the migration of CAR-NK cells toward H226 MPM cells with and without CXCR3 overexpression. Fig. 24B yes Fig.24A Two micrographs of the quantified data. Fig.24C is a bar graph showing the killing of H226 MPM cells by CAR-NK cells with and without CXCR3 overexpression. Fig.24D yes Fig. 24B Two micrographs of the quantified data.

[0043] Figure 25A-25B Showing CAR-NK cell migration. Fig.25A is a bar graph showing CAR-NK cell migration with and without ADU-S100 treatment. Fig.25B yes Fig.25A A set of micrographs of the data quantified in .

[0044] Figure 26A-26B CAR staining on NK cells from donor #27 is shown.

[0045] Figure 27A-27B CAR staining on NK cells from donor #28 is shown.

[0046] Figure 28A-28B Shown is CXCR3 staining on NK cells from donor #27.

[0047] Figure 29A-29B Shown is CXCR3 staining on NK cells from donor #28.

[0048] Figure 30A-Figure 30BTwo constructs used to generate genetically engineered immune cells are shown. Fig. 30A is a diagram showing a vector containing an anti-mesothelin CAR nucleic acid. Fig. 30B is a diagram showing a vector containing an anti-mesothelin CAR nucleic acid and a CXCR3 nucleic acid.

[0049] DETAILED DESCRIPTION OF THE DISCLOSURE

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the subject matter belongs. As used in the specification and appended claims, unless otherwise indicated, the following terms have the indicated meanings to facilitate understanding of the present disclosure.

[0051] As used in this description and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a composition" includes mixtures of two or more such compositions, reference to "an inhibitor" includes mixtures of two or more such inhibitors, reference to "a construct" includes use of more than one construct, and so forth.

[0052] Unless otherwise indicated, the term "about" is understood to be within the normal tolerance range in the art, such as within 2 standard deviations of the mean. "About" can be understood to be within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05% or 0.01% of the specified value. Unless the context clearly dictates otherwise, all numerical values ​​provided herein are modified by the term "about".

[0053] As used herein, the term "about" refers to a value that falls within 25%, 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1% or less in either direction (greater or less) of the reference value, unless otherwise specified or obvious from the context (unless such a number would exceed 100% of the possible value).

[0054] The transitional term "comprising," which is synonymous with "including," "containing," or "characterized by," is inclusive or open-ended and does not exclude other unrecited elements or method steps. In contrast, the transitional phrase "consisting of excludes any element, step, or ingredient not specified in the claim. The transitional phrase "consisting essentially of" limits the scope of the claim to the specified materials or steps of the disclosure "and those that do not materially affect the basic and novel characteristics of the claimed disclosure."

[0055] The terms "overexpression," "overexpressing," and "overexpressed" are used interchangeably herein and when referring to CXCR3 expression, refer to an increase of 30% or more in protein or messenger RNA compared to an appropriate control.

[0056] Nucleic acid construct

[0057] In one aspect, the present disclosure provides a nucleic acid construct, comprising a first nucleic acid containing a first promoter operably connected to a nucleic acid encoding a CXC motif chemokine receptor 3 (CXCR3) and a second nucleic acid encoding a chimeric antigen receptor (CAR), wherein CAR comprises a ligand binding domain, a transmembrane domain and an intracellular domain comprising a signal transduction domain of a single-chain antibody fragment containing antigen binding on tumor cells. In some embodiments, the second nucleic acid is operably connected to a second promoter, which may be the same or different from the first promoter. In some embodiments, the third nucleic acid encoding a self-cleavage peptide is arranged between the first nucleic acid and the second nucleic acid, and the first promoter drives the expression of CXCR3 nucleic acid, self-cleavage peptide and CAR nucleic acid.

[0058] The term "nucleic acid" as used herein refers to a polymer of nucleotides, each of which is an organic molecule consisting of a nucleoside (a nucleobase and a pentose) and a phosphate. The term nucleotide, unless otherwise specified or apparent from the context, includes a nucleoside having a ribose (i.e., a ribonucleotide forming ribonucleic acid, RNA) or a 2'-deoxyribose (i.e., a deoxyribonucleotide forming deoxyribonucleic acid, DNA). Nucleotides are used as monomer units of nucleic acid polymers or polynucleotides. The four nucleobases in DNA are guanine (G), adenine (A), cytosine (C) and thymine (T). The four nucleobases in RNA are guanine (G), adenine (A), cytosine (C) and uracil (U). Nucleic acids are linear chains of nucleotides (e.g., at least 3 nucleotides) chemically bonded by a series of ester bonds between the phosphate group of a nucleotide and the hydroxyl group of the sugar in the adjacent nucleotide (i.e., ribose or 2'-deoxyribose).

[0059] The term "promoter" as used herein refers to a nucleic acid that directly or indirectly regulates the transcription of a corresponding nucleic acid coding sequence that is operably connected thereto. A promoter can play a role in regulating transcription alone, or it can act in concert with one or more other regulatory sequences (e.g., enhancers or silencers, or regulatory elements that may be present in a vector). A promoter is located near the transcription start site of an open reading frame, on the same strand and upstream of the DNA (towards the 5' region of the sense strand). The length of a promoter is generally in the range of about 100 to 1000 base pairs.

[0060] As used herein, the term "operably linked" is understood to mean that the nucleic acid is located or positioned in the nucleic acid construct spatially relative to the promoter to drive the expression of the protein (eg, CXCR3) encoded by the nucleic acid.

[0061] CXCR3

[0062] CXCR3 is a chemokine receptor that induces cellular responses involved in immune cell trafficking. As shown in the working examples below, genetically engineered immune cells containing nucleic acids encoding CXCR3 have increased migration to the TME.

[0063] CXCR3 is a G protein-coupled receptor that binds three chemokines, called interferon-g-induced monokine (Mig / CXCL9), interferon-g-induced protein of 10 kDa (IP10 / CXCL10), and interferon-induced T cell alpha-chemoattractant (I-TAC / CXCL11). Binding of chemokines to CXCR3 induces cellular responses, including integrin activation, cytoskeletal changes, and chemotactic migration.

[0064] The amino acid sequence of a representative CXCR3 is provided in NCBI Accession No. NP_001495, Version NP_001495.1, which is incorporated herein by reference, and is listed in the Sequence Listing as SEQ ID NO: 1. The nucleic acid sequence encoding the CXCR3 protein (SEQ ID NO: 1) is provided in NCBI Accession No. NC_000023, Version NC_000023.11, which is incorporated herein by reference, and is listed in the Sequence Listing as SEQ ID NO: 2. The nucleic acid sequence encoding another representative CXCR3 is listed in the Sequence Listing as SEQ ID NO: 3.

[0065] CAR

[0066] CAR binds to antigens on the surface of cancer cells. CAR comprises a ligand binding domain (which comprises a single-chain antibody fragment that binds to a cancer (e.g., tumor cell) surface antigen), a transmembrane domain, and an intracellular domain comprising a signaling domain. In some embodiments, the ligand binding domain is an antibody fragment (e.g., scFv).

[0067] In some embodiments, CAR is specific for and binds to malignant pleural mesothelioma (MPM) antigens. In some of these embodiments, the MPM antigen is mesothelin. In some embodiments, the CAR ligand binding domain is derived from an anti-mesothelin antibody, an antibody fragment, or a derivative thereof. In some embodiments, the CAR ligand binding domain is derived from YP218, Amatu monoclonal antibody, RC88, 19C3, 3C10, or 7B1. The nucleic acid sequences of YP218 VH (SEQ ID NO: 4) and VL (SEQ ID NO: 5), Amelizumab VH (SEQ ID NO: 6) and VL (SEQ ID NO: 7), RC88 VH (SEQ ID NO: 8) and VL (SEQ ID NO: 9), 19C3 VH (SEQ ID NO: 10) and VL (SEQ ID NO: 11), 3C10 VH (SEQ ID NO: 12) and VL (SEQ ID NO: 13), and 7B1 VH (SEQ ID NO: 14) and VL (SEQ ID NO: 15) are listed in the sequence listing. The nucleic acid sequences of YP218 VH (SEQ ID NO: 16) and VL (SEQ ID NO: 17), Amelizumab VH (SEQ ID NO: 18) and VL (SEQ ID NO: 19), RC88 VH (SEQ ID NO: 20) and VL (SEQ ID NO: 21), 19C3 VH (SEQ ID NO: 22) and VL (SEQ ID NO: 23), 3C10 VH (SEQ ID NO: 24) and VL (SEQ ID NO: 25), and 7B1 VH (SEQ ID NO: 26) and VL (SEQ ID NO: 27) are listed in the sequence listing.

[0068] In some embodiments, the CAR ligand binding domain comprises a VH having an amino acid sequence of SEQ ID NO: 4. In some embodiments, the CAR ligand binding domain comprises a VL having an amino acid sequence of SEQ ID NO: 5.

[0069] Other anti-mesothelin antibodies and mesothelin-binding fragments thereof are known in the art. See, for example, U.S. Patents 7,081,518, 7,943,133, 8,460,660, 8,911,732, 9,272,002, 9,719,996, 10,022,452, 10,183,993, 10,793,641, and 10,919,975 and U.S. Patent Application Publications 2009 / 0047211, 2015 / 0252118, and 2022 / 0056147.

[0070] The transmembrane domain of CAR connects the CAR ligand binding domain to the intracellular domain. In some embodiments, the transmembrane domain is directly connected to the CAR ligand binding domain. In some embodiments, the transmembrane domain is derived from CD3α, CD3β, CD3γ, CD3ζ, CD3ε, CD4, CD5, CD8α, CD9, CD16, CD22, CD28, CD33, CD37, CD45, CD64, CD80, CD86, CD134, CD137 (4-1BB or TNF receptor superfamily member 9 (TNFRSF9)), CD154, FcεRIα, FcεRIβ, FcεRIγ, ICOS, KIR2DS2, MHC class I, MHC class II or NKG2D. The amino acid sequence of the representative transmembrane domain is listed in the sequence table with SEQ ID No: 14-18.

[0071] The amino acid sequence of the native transmembrane domain can be modified by amino acid substitution to avoid binding of these regions to the transmembrane domain of the same or different surface membrane proteins to minimize interactions with other members of the receptor complex. See, for example, U.S. Patent Application Publication 2021 / 0101954; Soudais et al., Nat. Genet. 3:77-81 (1993); Muller et al., Front. Immunol. 12:639818-13 (2021); and Elazar et al., elife 11:e75660-29 (2022).

[0072] In some embodiments, CAR also includes a hinge domain arranged between the ligand binding domain and the transmembrane domain. The hinge domain can provide flexibility in the optimal direction for allowing the ligand binding domain to obtain antigen binding, thereby enhancing the anti-tumor activity of genetically modified immune cells expressing CAR. In some embodiments, the hinge domain is derived from IgA, IgD, IgE, IgG or IgM. In some embodiments, the hinge domain is derived from CD3ζ, CD4, CD8α, CD28, IgG1, IgG2 or IgG4, and its representative amino acid sequences are listed in the sequence table as SEQ ID No: 19-25.

[0073] The intracellular domain of CAR includes a signal transduction domain that can realize intracellular signal transmission and immune cell function. The signal transduction domain may include a primary signal transduction domain and / or a costimulatory signal transduction domain. In some embodiments, the intracellular domain can transmit a signal close to the natural connection of a molecule or receptor complex (such as a TCR receptor complex) containing ITAM.

[0074] The intracellular domain includes a signal transduction domain that can realize intracellular signal transmission and immune cell function. The signal transduction domain may include a primary signal transduction domain and / or a costimulatory signal transduction domain. In some embodiments, the intracellular domain includes one or more phosphorylated intracellular motifs (ITAMs) that can transmit immune activation signals. In some embodiments, the intracellular domain can transmit signals close to the natural connection of molecules or receptor complexes (such as TCR receptor complexes) containing ITAMs.

[0075] In some embodiments, the signaling domain includes multiple, for example, 2 or 3 costimulatory signaling domains, for example, selected from 4-1BB, CD3ζ, CD28, CD27, ICOS and OX40. In some embodiments, the signaling domain may include a CD3ζ domain as the main signaling domain, and any one of the following pairs of costimulatory signaling domains from the extracellular to the intracellular direction: 4-1BB-CD27; CD27-4-1BB; 4-1BB-CD28; CD28-4-1BB; OX40-CD28; CD28-OX40; 4-1BB-CD3ζ; CD3ζ-4-1BB; CD28-CD3ζ; and CD3ζ-CD28. In some embodiments, the main signaling domain is derived from CD3ζ, CD27, CD28, CD40, KIR2DS2, MyD88 or OX40. In some embodiments, the costimulatory signaling domain is derived from 4-1BB (CD137; TNFRSF9), CD3γ, CD3δ, CD3ε, CD3ζ, CD4, CD5, CD8α, CD9, CD16, CD22, CD27, CD28, CD33, CD37, CD40, CD45, CD68, CD72, CD80, CD86, CD154, CLEC-1, DAP10 (hematopoietic cell signal transducer (HCST)), DAP12 (TYROBP), Dectin-1, FcαRI, FcγRI, FcγRII, FcγRIII, IL-2RB, ICOS, KIR2DS2, MyD88, OX40, and ZAP70. One or more of the amino acid sequences of representative signaling domains are listed in the sequence table as SEQ ID Nos: 26-43, respectively.

[0076] In some embodiments, the signaling domain is derived from CD3ζ, and the costimulatory domain is derived from 4-1BB. In some embodiments, the signaling domain is derived from CD3ζ, and the costimulatory domain is derived from CD28. In some embodiments, the signaling domain is derived from CD3ζ, and the costimulatory domain is derived from 4-1BB and CD28. The amino acid sequences of representative 4-1BB and CD28 are listed in SEQ ID NO:26 and SEQ ID NO:32, respectively, while other subtypes of CD28 are provided in the sequence table as SEQ ID No:44-46.

[0077] The expression of the first nucleic acid encoding CXCR3 and the expression of the second nucleic acid encoding CAR are controlled by one or more promoters, which may be natural or synthetic. In some embodiments, the third nucleic acid encoding a self-cleaving peptide or an internal ribosome entry site (IRES) is arranged between the first nucleic acid and the second nucleic acid. In these embodiments, the first nucleic acid and the second nucleic acid are controlled by the same promoter. In some embodiments, the second nucleic acid is controlled by a second promoter different from the first promoter.

[0078] In some embodiments, the first promoter is a strong promoter that overexpresses a nucleic acid to which it is operably linked. Overexpression can be achieved by providing a vector encoding a protein controlled by a constitutive promoter, or by removing a repressor, adding multiple gene copies to the cell, or upregulating an endogenous gene, etc. In some embodiments, one or both promoters are derived from elongation factor 1 alpha (EF-1 alpha), cytomegalovirus (CMV), β-actin, simian virus 40 (SV40) early promoter, human phosphoglycerate kinase (PGK), RPBSA (synthetic, from Sleeping Beauty) or CAG (synthetic, CMV early enhancer element, chicken β-actin and rabbit β-globin splice acceptor) promoter. When referring to a protein or nucleic acid, the term "derived from" as used herein refers to a protein or nucleic acid that is derived from another parent protein or nucleic acid. The sequence of the derived protein or nucleic acid can be identical to the parent sequence, can be a portion of the parent sequence, or can have at least one variant from the parent sequence. Variants can include amino acid and nucleotide substitutions, insertions or deletions. Thus, for example, an amino acid sequence derived from a parent sequence may be identical to a specific amino acid range of the parent, but does not include amino acids outside of that specific region.

[0079] In some embodiments, the promoter may have a core region located near the start of the nucleic acid coding sequence. In some embodiments, the promoter is modified relative to the natural promoter. One modification requires the removal of methylation sensitive sites (e.g., cytosine nucleotides followed by guanine nucleotides, or "CpG"). Another modification requires the addition of regulatory sequences that bind to DNA methylation-inhibiting transcription factors. In some embodiments, the expression vector includes an A / T-rich nuclear matrix interaction sequence, called a scaffold matrix attachment region (S / MAR), which can improve transformation efficiency and improve the stability of transgenic expression.

[0080] In some embodiments, the first promoter and the second promoter are derived from EF-1α. In some embodiments, the first promoter is derived from CMV and the second promoter is derived from EF-1α. The sequence of the EF-1α promoter is provided in NCBI Accession No. J04617.1. CMV promoter sequences from different CMV isolates are provided in NCBI Accession Nos. AY218848, AF477200, M64754, and AF286076. The sequence of the PGK promoter is provided in NCBI Accession No. NC_000023.11, ranging from 78104248 to 78129295. The sequence of the RPBSA promoter is provided in NCBI Accession No. MN811119.1. The sequence of the CAG promoter is provided in NCBI Accession No. MG763233.1.

[0081] In some embodiments, the nucleic acid construct comprises a self-cleaving polypeptide encoding nucleic acid located between the nucleic acid encoding CXCR3 and the nucleic acid encoding CAR. The nucleic acid sequences of representative self-cleaving polypeptides are listed in the sequence table as SEQ ID No 476-49.

[0082] In some embodiments, the nucleic acid construct comprises a selection marker to help separate, capture or detect. Selection markers generally require the addition of in-frame nucleic acids that will be translated into amino acids together with the protein to which they are attached. Representative examples of selection markers include enhanced green fluorescent protein (EGFP) (SEQ ID NO: 50), AU1 epitope (SEQ ID NO: 51), AU5 epitope (SEQ ID NO: 52), polyhistidine (SEQ ID NO: 53), FLAG epitope (SEQ ID NO: 54), FLAG His tag (SEQ ID NO: 55), histidine affinity tag (HAT) (SEQ ID NO: 56), herpes simplex virus (HSV) epitope (SEQ ID NO: 57), human influenza hemagglutinin (HA), glutathione S-transferase (GST), KT3 epitope, maltose binding protein (MBP), bacteriophage T7 epitope, myc tag. The amino acid sequences of representative selection markers are listed in the sequence table with SEQ ID No: 50-57.

[0083] Carrier

[0084] The nucleic acid encoding CXCR3 and CAR can be introduced into immune cells by the same or different vectors. Nucleic acid constructs are introduced into immune cells by suitable vectors. The vector is configured to contain additional regulatory elements necessary for realizing transport to immune cells and realizing nucleic acid expression after conversion. These elements include replication origin or promoter, poly-A tail sequence, selective marker, one or more suitable sites for inserting nucleic acid sequences, such as multiple cloning sites (MCS) and selective markers, and other optional regulatory elements.

[0085] In some embodiments, the vector is a viral vector, for example, a retroviral vector, a lentiviral vector, an adenoviral vector, a herpes virus vector, an adenovirus or an adeno-associated virus (AAV) vector. As used herein, the term "lentiviral vector" is intended to refer to infectious lentiviral particles. The lentivirinae or lentiviruses are enveloped retroviruses or a subfamily of retroviruses that differ from other viruses in virion structure, host range, and pathological effects. Infectious lentiviral particles are able to invade target host cells, including infection and transduction of non-dividing cells and immune cells. Slow virus features include, for example, infection or transduction of non-dividing host cells, including immune cells.

[0086] In some embodiments, the vector is a recombinant lentivirus comprising a recombinant genome comprising a lentiviral packaging psi sequence, an RNA nuclear export element, a transgene, a promoter and / or a sequence that facilitates RNA nuclear import between the LTR 5' and 3' lentiviral sequences, and a mutant integrase that prevents its genome from integrating into the host cell genome. For example, the construction of lentiviral vectors has been described in U.S. Patents 5,665,577, 5,981,276, 6,013,516, 7,090,837, 8,119,119 and 10,954,530.

[0087] In some embodiments, the vector is a non-integrating and non-replicating recombinant lentiviral vector. For example, the construction of lentiviral vectors has been described in U.S. Patents 5,665,577, 5,981,276, 6,013,516, 7,090,837, 8,119,119 and 10,954,530. The lentiviral vector includes a defective lentiviral genome, i.e., at least one of the lentiviral genes gag, pol and env has been inactivated or deleted.

[0088] Lentiviral vectors can also exhibit functions other than or different from those of natural lentiviruses. For example, lentiviral vectors can be modified to alter or reduce the characteristics of lentiviruses. Lentiviral vectors can also be modified to exhibit characteristics of one or more other retroviruses, retroviral vectors, host cells, or heterologous cells. For example, modifications can include pseudotyping, modifying the binding and / or fusion functions of envelope polypeptides, incorporating heterologous, chimeric, or multifunctional polypeptides into vectors, incorporating non-lentiviral genomes, or incorporating heterologous genes into lentiviral vector genomes.

[0089] The terms "pseudotype", "pseudotyping", "pseudotyped vector" and "pseudotyped vector particle" are used herein to refer to a vector that carries components (e.g., envelope or capsid) from more than one source. The source can be from a heterologous virus or a non-viral protein. The non-viral protein can include antibodies and antigen-binding fragments thereof. A representative pseudotyped vector is a vector that carries non-glycoprotein components from a first virus and an envelope glycoprotein from a second virus. Thus, the host range of the pseudotyped vector may be expanded or altered, depending on the type of cell surface receptor that binds to the glycoprotein from the second virus.

[0090] In some embodiments, the lentiviral vector is pseudotyped with baboon envelope (BaEV) glycoprotein (BaEV-gp). The amino acid sequence of a representative BaEV-gp is listed in the sequence listing as SEQ ID NO: 58. The nucleic acid sequence encoding BaEV-gp (SEQ ID NO: 58) is listed in SEQ ID NO: 59. Other BaEv pseudotyped lentiviral vectors are known in the art. See, e.g., Levy et al., J. Thromb. Haemost. 14: 2478-2492 (2016), Costa et al., Leukemia 31: 977-980 (2017), and Bari et al., Front. Immunol. 10: 2001 (2019). The nucleic acid sequence of a representative BaEV vector is listed in the sequence listing as SEQ ID NO: 60.

[0091] In some embodiments, the vector comprises a plx307-based nucleic acid construct. In some embodiments, the vector comprises a pHIV-based nucleic acid construct. The nucleic acid sequence of a representative vector comprising a pHIV-based nucleic acid construct (containing a nucleic acid encoding a CAR that binds mesothelin (pHIV-aMesoCAR-GFP) is listed in the sequence table as SEQ ID NO: 61. A representative vector comprising a pHIV-based nucleic acid construct encoding CXCR3 and a CAR that binds mesothelin (pHIV-aMesoCAR-CXCR3; as Fig. 30B The nucleic acid sequence of the present invention is listed as SEQ ID NO:62 in the sequence listing.

[0092] In some embodiments, the vector comprises a nucleic acid construct based on pCMV. The nucleic acid sequence of a representative vector comprising a nucleic acid construct based on pCMV (pCMV-dR8.91) is listed in the sequence table with SEQ ID NO: 63. In some embodiments, the vector comprises a nucleic acid construct based on pAdv. The nucleic acid sequence of a representative vector comprising a nucleic acid construct based on pAdv (pAdvAntage) is listed in the sequence table with SEQ ID NO: 64.

[0093] cell

[0094] One aspect of the present disclosure is a genetically engineered (or transformed) immune cell, which includes a vector containing a nucleic acid construct encoding CXCR3 and CAR. As used herein, "immune cell" refers to a cell of hematopoietic origin, which is functionally involved in the initiation and / or execution of innate and / or adaptive immune responses. Representative examples of immune cells include natural killer (NK) cells, T cells, macrophages, and dendritic cells. A combination of different genetically engineered immune cells can be used. In some embodiments, the genetically engineered immune cell is a NK cell. In some embodiments, the genetically engineered immune cell is from a NK cell line, a primary NK cell, a stem cell-derived NK cell, a cord blood-derived NK cell, a peripheral blood mononuclear cell (PBMC)-derived NK cell, a memory-like NK cell, or an induced memory-like NK cell. Suitable NK cell lines for use in the present method include NK-92, NKG, NKL, KHYG-1, YT, NK-YS, SNK-6, IMC-1, YTS, NKL cells and high affinity NK (haNK, a NK / T cell lymphoma cell line).

[0095] In some embodiments, the genetically engineered immune cells are memory-like NK cells.Memory-like NK cells can be produced by collecting NK cells from a subject, such as purifying from a peripheral blood sample, stimulating a suitable period of time (e.g., about 12 hours to 7 days) with cytokines (e.g., IL-12, IL-15, and IL-18), removing cytokines, and transducing to express CXCR3 and CAR. In some embodiments, the genetically engineered immune cells are cytokine-induced memory-like (CIML) NK cells.CIML NK cells can be produced by stimulating NK cells with one or more (but usually in combination) of IL-12, IL-15, and IL-18. See, e.g., Cooper et al., Proc. Natl Acad. Sci. USA 106: 1915-9 (2009); Ni et al., J. Exp. Med. 209: 2351-65 (2012); Keppel et al., J. Immunol. 190: 4754-62 (2013).

[0096] In some embodiments, the cell is a T cell. In some embodiments, the T cell is a naive T cell, a memory stem cell T cell, a central memory T cell, an effector memory T cell, a helper T cell, a CD4+T cell, a CD8+T cell, a CD8 / CD4+T cell, an αβT cell, a γδT cell, and a natural killer T (NKT) cell and a Th17 T cell. T cell separation and classification into T cell subsets are known in the art. See, for example, U.S. Patents 10,507,219, 11,135,245, and 11,242,376, and U.S. Patent Application Publications 2013 / 0060011, 2019 / 0276540, 2020 / 0347350, and 2021 / 0106622.

[0097] Methods for introducing vectors containing nucleic acid constructs into immune cells are known in the art. See, for example, U.S. Patents 7,399,633, 7,575,925, 10,072,062, 10,370,452, and 10,829,735, and U.S. Patent Application Publications 2019 / 0000880 and 2021 / 0407639.

[0098] In some embodiments, a lentiviral vector is transduced into an immune cell. In other embodiments, the method requires the use of a gamma retroviral vector. See, for example, U.S. Patents 9,669,049, 11,065,311, and 11,230,719. In some embodiments, the method requires the use of adenovirus, adeno-associated virus (AAV), dsRNA, ssDNA, or dsRNA to deliver a nucleic acid construct. See, for example, U.S. Patent 10,563,226, and U.S. Patent Application Publications 2019 / 0225991, 2020 / 0080108, and 2022 / 0186263.

[0099] Pharmaceutical composition

[0100] The pharmaceutical composition of the present disclosure includes an effective number of genetically modified immune cells and a pharmaceutically acceptable carrier. The term "effective number of genetically modified immune cells" used herein (indirectly including a corresponding amount of CXCR3 and CAR) refers to a genetically modified immune cell containing nucleic acids encoding CXCR3 and CAR in a quantity sufficient to provide the desired effect.

[0101] The composition can be provided in the form of a sterile liquid preparation, such as an isotonic aqueous solution, suspension, emulsion, dispersion or viscous composition, which can be buffered to a selected pH. Liquid carriers include aqueous or non-aqueous carriers. Representative examples of liquid carriers include saline, phosphate buffer, soluble proteins, dimethyl sulfoxide (DMSO), polyols (e.g., glycerol, propylene glycol, liquid polyethylene glycol, etc.) and suitable mixtures thereof. In some embodiments, the liquid carrier includes proteins dissolved or dispersed therein, representative examples of which include serum albumin (e.g., human serum albumin, recombinant human albumin), gelatin and casein. The compositions are generally isotonic, i.e., they have the same osmotic pressure as blood. Sodium chloride and isotonic electrolyte solutions (e.g., ) to achieve the desired isotonicity. Other excipients may be added according to the carrier and the genetically modified immune cells, for example, wetting agents, dispersants or emulsifiers, gelling agents and viscosity increasing agents, preservatives, etc. known in the art.

[0102] cancer

[0103] In some aspects, the present disclosure relates to treating cancer in a subject. The method requires that an effective number of genetically modified immune cells containing a nucleic acid construct be administered to a subject in need thereof, and the nucleic acid construct comprises a first nucleic acid encoding CXCR3 and a second nucleic acid encoding CAR (also referred to herein as "genetically modified immune cells"). The term "cancer" as used herein refers to a disease characterized by uncontrolled cell proliferation, reduced apoptosis, and the spread of abnormal cells that invade and destroy non-cancerous tissues. Cancer cells can be in the form of a tumor (i.e., a solid tumor), or can be present alone in a subject's body, also referred to as a liquid tumor. The term cancer includes cancers before becoming malignant and malignant cancers.

[0104] In some embodiments, cancer is a solid tumor. Since solid tumors form different types of tissues during tumor growth, solid tumors are highly heterogeneous. In some embodiments, solid tumors are sarcomas or carcinomas. In some embodiments, cancer is MPM. Some embodiments relate to methods for treating MPM by administering an effective amount of NK cells or their pharmaceutical compositions containing nucleic acid constructs with CXCR3 and CAR to a subject in need thereof. In some embodiments, the method also requires administering an effective amount of a STING agonist to the subject before, substantially simultaneously with, or after administering the NK cells or their pharmaceutical compositions.

[0105] In some embodiments, the cancer comprises hypermethylation of the promoter of cyclic GMP-AMP synthase (cGAS) or STING genes. In some of these embodiments, the cancer is bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), cervical squamous cell carcinoma and cervical adenocarcinoma (CESC), cervical precancerous lesions (CPL), colon adenocarcinoma (COAD), glioma (e.g., glioblastoma), head and neck squamous cell carcinoma (HNSC), renal clear cell carcinoma (KIRC), papillary renal cell carcinoma (KIRP), hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), melanoma, ovarian cancer, pancreatic adenocarcinoma (PAAD), prostate adenocarcinoma (PRAD), rectal adenocarcinoma (READ), sarcoma (SARC), cutaneous melanoma (SKCM), gastric adenocarcinoma (STAD), thyroid cancer (THCA) and endometrial carcinoma (UCEC). See Konno et al., Oncogene 37:2037-2051 (2018), deQueiroz et al., Mol. Cancer Res. 17:974-986 (2019), Huang et al., Front. Genet. 10:1-11 (2019), Falahat et al. al., Proc. Natl. Acad. Sci. USA 118: 1-9 (2021), Low et al., Cancer Cell 40: 439-440 (2022).

[0106] In some embodiments, the cancer has high basal STING expression, also referred to herein as STING + As used herein, the term "high basal" expression of a gene refers to elevated expression of the gene in a disease state compared to a reference non-disease state. In some embodiments, STING + The cancer is melanoma (e.g., malignant melanoma), gastric cancer, liver cancer (e.g., hepatocellular carcinoma (HCC)), lung cancer (e.g., non-small cell lung cancer (NSCLC)), bladder cancer, colorectal cancer, or breast cancer. Other cancers in which STING has been shown to play a role are known in the art and include leukemia (e.g., acute myeloid leukemia), lymphoma (e.g., malignant lymphoma), breast cancer, colorectal cancer, glioma, head and neck squamous cell carcinoma, lung cancer, melanoma, nasopharyngeal carcinoma, ovarian cancer, pancreatic cancer, prostate cancer, and tongue squamous cell carcinoma. See Zhu et al., Mol. Cancer 18(1):152(2019).

[0107] As used herein, the terms "treat," "treating," and "treatment" refer to any type of intervention, procedure, or administration of an active agent to a subject in need thereof, with the therapeutic purpose ("therapeutic effect") being to reverse, alleviate, ameliorate, inhibit, reduce, slow down, halt, stabilize, or prevent the onset, progression, development, severity, or recurrence of a symptom, complication, or condition, or biochemical marker associated with cancer.

[0108] As used herein, the term "subject" (or "patient") includes all members of the animal kingdom susceptible to or suffering from the cancer. Therefore, subjects "suffering from cancer" or "in need of" treatment according to the present disclosure broadly include subjects who have been diagnosed positive, including subjects with active disease who may have received one or more rounds of treatment before, and subjects who are not currently receiving treatment (e.g., in remission) but are still at risk of recurrence, and subjects who have not been diagnosed positive but are susceptible to cancer (e.g., due to past medical history and / or family history, or otherwise the presence of one or more risk factors, so that medical professionals may reasonably suspect that the subject is susceptible to cancer).

[0109] Application

[0110] The number of genetically engineered immune cells administered to a subject will vary widely, depending on the location, type, and severity of the cancer, as well as the age, weight, and physical condition of the individual being treated. The physician will ultimately determine the appropriate number and dose of cells to be used. Typically, the genetically engineered immune cells will be administered in a single dose. In some embodiments, the effective number of genetically engineered immune cells is about 1×10 per subject. 5 to about 1×10 10 In some embodiments, the effective number of genetically engineered immune cells is about 1×10 per kg of subject body weight. 5 About 6×10 8 between cells.

[0111] Compositions containing a therapeutically effective number of genetically engineered immune cells can be administered to a subject by any medically acceptable route to treat cancer. Genetically engineered immune cells are typically delivered intravenously, although they may also be introduced into other convenient locations (e.g., intratumorally administered to affected organs or tissues) or modes, as determined by the attending physician. Amplifiers and differentiation agents may be provided before, during, or after cell administration to increase differentiation, amplification, or persistence of genetically engineered immune cells (e.g., NK cells).

[0112] In some embodiments, the genetically engineered immune cells are administered as a single intravenous infusion over a period of time. Representative infusion times are 30 minutes, 60 minutes, and 90 minutes. In some embodiments, the infusion time is between 30 minutes and 60 minutes. In some embodiments, the first administration is injected into the patient within 90 minutes, and the subsequent administration is injected into the patient within 30 minutes.

[0113] Combination therapy

[0114] In some embodiments, the method includes administering a STING agonist in combination. The term "administered in combination" includes substantially simultaneous administration via the same or different dosage forms, or sequential administration, for example, as part of the same treatment regimen or via a continuous treatment regimen. The order and time intervals can be determined so that the therapies administered in combination can work together (e.g., synergistically provide a greater benefit than administration in other ways). For example, multiple therapeutic agents can be administered simultaneously or sequentially in any order at different time points; however, if not administered simultaneously, they can be administered in a sufficiently close time to provide the desired therapeutic effect, which can be in a synergistic manner.

[0115] In some embodiments, the genetically modified immune cells of the present disclosure are used in combination with a STING agonist. In some embodiments, the STING agonist is ADU-S100, TAK-676, BI-STING, BMS-986301, GSK532, DMXAA (ASA-404), GSK3745417, JNJ-4412, MK-1454, SB11285, 3'3'-cyclic AIMP, ALG-031048, E7766, JNJ-'6196, MK-2118, MSA-1, MSA-2, SNX281m SR-717, KAT676, TTI-10001, XMT-2056, CRD-5500, c-di-AMP, a synthetic cyclic dinucleotide (DCN) molecule, an analog thereof, or a combination thereof. In some embodiments, the STING agonist is ADU-S100 or TAK-676. In some embodiments, the STING agonist is delivered by intratumoral injection or systemic (i.e., intravenous). See Woodward et al., Science 328:1703-5 (2010), Motedayen Aval et al., J. Clin. Med. 9:3323 (2020) and U.S. Patents 11,285,131 and 11,312,772, and U.S. Patent Application Publications 2018 / 0028553, 2019 / 0328762, 2020 / 0330556, and 2021 / 0170043.

[0116] In some embodiments, the method comprises administering genetically engineered immune cells in combination with another anti-cancer agent, with or without a STING agonist. Representative examples of other anti-cancer agents are described below.

[0117] Anticancer agents that can be used in combination with the cells of the invention are known in the art. See, e.g., U.S. Pat. No. 9,101,622 (Section 5.2 thereof). For example, an "anticancer" agent can adversely affect a subject's cancer by killing cancer cells, inducing apoptosis in cancer cells, reducing the growth rate of cancer cells, reducing the incidence or number of metastases, reducing tumor size, inhibiting tumor growth, reducing the blood supply to a tumor or cancer cell, promoting an immune response to a cancer cell or tumor, preventing or inhibiting cancer progression, or increasing the life span of a subject with cancer.

[0118] In some embodiments, the genetically modified immune cells of the present disclosure, or genetically modified immune cells combined with a STING agonist, are used in combination with a type I IFN agonist. In some embodiments, the type I INF agonist is a recombinantly synthesized type I INF protein, such as interferon alfacon-1. Recombinant interferon α-2b (Intron ), interferon beta-1b Interferon α-2c (Berofor ), interferon α-n4 (Alferon ) or pegylated IFN, such as pegylated interferon beta-1a

[0119] In some embodiments, the genetically modified immune cells of the present disclosure are used in combination with a DNA methylation inhibitor. In some embodiments, the DNA methylation inhibitor is a DNA methyltransferase (DNMT) enzyme inhibitor. Representative DNMT inhibitors include azacitidine and decitabine (5-aza-2'-deoxycytidine) In some embodiments, the additional anticancer agent comprises an epigenetic therapy. In some embodiments, the epigenetic therapy azacitidine ( ), decitabine (5-aza-2'-deoxycytidine) Zebulin (pyrimidin-2-one β-D-ribofuranoside), guandecitabine, 5-fluoro-2'deoxycytidine, (-)-epigallocatechin gallate, curcumin, hydralazine, procainamide, RG-108 and SG-1027. See Nepali et al., J. Biomed. Sci. 28:27 (2021); Giri et al., Front. Pharmacol. 10:1-11 (2019).

[0120] Immunotherapy

[0121] In some embodiments, the additional anticancer agent includes an immunotherapy, for example, an immune checkpoint inhibitor. Representative examples of immune checkpoint molecules that can be targeted by additional therapies include PD-1, PDL1, CTLA4, KIR, TIGIT, TIM-3, LAG-3, BTLA, VISTA, CD47, and NKG2A. Examples of clinically available immune checkpoint inhibitors include durvalumab Atezolizumab and avelumab Examples of clinically available PD-1 inhibitors include nivolumab Pembrolizumab and cimiprilimab

[0122] Chemotherapy

[0123] Anticancer therapy also includes various combination therapies with chemotherapy and radiation-based treatments. Combination chemotherapy includes, for example, Hexamethylmelamine, docetaxel, Methotrexate, Cisplatin (CDDP), carboplatin, procarbazine, nitrogen mustard, cyclophosphamide, camptothecin, ifosfamide, melphalan, chlorambucil, busulfan, nitrosourea, actinomycin D, daunorubicin, doxorubicin, bleomycin, emtricitabine, mitomycin, etoposide (VP16), tamoxifen, raloxifene, estrogen receptor binding agents, Gemcitabine, Farnesyl protein transferase inhibitor, trans-platinum, 5-fluorouracil, vincristine, vinblastine and methotrexate, or any analog or derivative variant thereof and combinations thereof.

[0124] Radiation therapy

[0125] Anticancer therapy also includes radiation-based DNA damaging therapies. Combined radiation therapy includes what are commonly called gamma rays, X-rays, and / or the targeted delivery of radioisotopes to cancer cells, which causes extensive damage to DNA, DNA replication and repair, and chromosome assembly and maintenance. The dosage range for radioisotopes varies widely and depends on the half-life of the isotope, the strength and type of radiation emitted, and the uptake by the tumor cells, and will be determined by the attending physician.

[0126] These and other aspects of the present application will be further understood upon consideration of the following examples, which are intended to illustrate certain embodiments of the application but are not intended to limit the scope as defined by the claims. Example

[0127] Example 1: Materials and Methods

[0128] Patient samples. Formalin-fixed, paraffin-embedded (FFPE) tissue microarray slides from patients with SCLC, NSCLC, and thymoma were purchased from Biomax (LC245, LC817, LC2081, THY761). In addition, FFPE slides were collected from DFCI / BWH patients with SCLC (n=58), MPM (n=68), and benign pleura (n=9) according to Dana-Farber / Harvard Cancer Center protocols 02-180 and 98-063. Tumors from patients with MPM treated at DFCI / BWH from July 2018 to October 2021 were collected after surgery according to protocol 98-063. Figure 1A-Figure 1D and Figure 7A-7E The patient samples analyzed by flow cytometry in the Figures differ from the patient samples processed ex vivo in the remaining Figures. NK cells were detected in blood collected from patients with head and neck squamous cell carcinoma or oral hyperplastic verrucous leukoplakia according to protocols 17-255 and 18-387 ( Figure 1D ). Mann-Whitney test: *** p<0.001. Detailed information on flow cytometry antibodies is provided in Table 1. TIM-3 = T-cell immunoglobulin and mucin domain-containing protein 3; PD-1 = programmed cell death protein 1; LAG3 = lymphocyte activation gene 3; EMRA = effector memory re-expressing CD45 RA; EM = effector memory; CM = central memory. These studies were conducted in accordance with the Declaration of Helsinki and were approved by the DFCI and BWH Institutional Review Boards. Written informed consent was obtained from all patients with tumors who underwent the study.

[0129] Immunohistochemistry. STING and 25hosphoro-IRF3 immunohistochemistry (IHC) were performed on a Leica Bond III automated staining platform. The STING antibody (Cell Signaling Technology #13647, clone D2P2F) was run at a 1:50 dilution using the Leica Biosystems Refine Detection Kit with citrate antigen retrieval solution. The forhosphoro-IRF3 antibody (Cell Signaling Technology #29047, clone D6O1M) was run at a 1:100 dilution using the Leica Biosystems Refine Detection Kit with EDTA antigen retrieval solution. This was optimized by a series of dilutions and comparison of citrate and EDTA antigen retrieval solutions on an MPM cell line treated in vitro with 50 μM ADU-S100 for 24 hours prior to paraformaldehyde fixation and paraffin embedding ( Figure 7C ). STING IHC staining was quantified using QuPath software (version 0.2.3) (Bankhead et al., Sci. Rep. 7:16878 (2017)). Positive pixel detection analysis was used with the default settings of DAB staining, and positive pixels were detected and quantified in each of three randomly selected areas of each tumor and then averaged.

[0130] Flow cytometry immunoassay. Fresh tumors were mechanically and enzymatically dissociated in a dissociation buffer consisting of RPMI (Life Technologies) + 10% fetal bovine serum (FBS; HyClone), 100U / ml type IV collagenase (Life Technology), and 50μg / ml DNase I (Roche). The suspension was incubated at 37°C for 45 minutes and then further mechanically dissociated. Red blood cells were removed from the sample using red blood cell lysis buffer (Biolegend). The sample was precipitated and then resuspended in fresh RPMI + 10% FBS and filtered through a 40μm filter. The cells were incubated with live / dead Zombie NIR (Biolegend) at room temperature in the dark for 5 minutes. Before surface antibody staining, Fc receptors were blocked using human TruStain FcX blocking reagent (Biolegend). The cells were stained on ice in the dark for 15 minutes and rinsed twice with PBS + 2% FBS. Cells were analyzed on a BD LSRFortessa (BD Biosciences) equipped with FACSDiva software. Data were analyzed using FlowJo software version 10.5.3. Antibodies are listed by protein target, with clone, manufacturer and catalog number in parentheses: CD69 (FN50, BioLegend, 310904), CD16 (3G8, BioLegend, 302006), CD8 (RPA-T8, Thermo Fisher, BDB560662), CCR2 (K036C2, BioLegend, 357203), CD38 (HIT2, BioLegend, 303506), CD11c (3.9, BioLegend, 301605), CCR7 (150503, Thermo Fisher, BDB62381), CD56 (GDC56, BioLegend, 318348), LAG-3 (11C3C65, BioLegend, 369309), CD103 (B-Ly7, Thermo Fisher, BDB560662). Fisher, 25-1038-41), TIM-3 (F38-2E2, BioLegend, 345012), PD-L1 (29E.2A3, BioLegend, 329708), CD3 (UCHT1, BioLegend, 300424), PD-1 (EH12.2H7, BioLegend, 329920), HLA-DR (G46-6, Thermo Fisher, BDB562804), CD45RA (HL100, BioLegend, 304142), CD15 (SSA-1, BioLegend, 323028), CTLA-4 (BNI3, BioLegend, 369609), CD19 (HIB19, BioLegend, 302243) , CD45 (H130, BioLegend, 304050), CD4 (PRA-T4, BioLegend, 300554), CD14 (M5E2, BioLegend, 301840), mesothelin (REA1057, Miltenyi, 130-118-168), STING (D2P2F, Cell Signaling, 13647), pTBK1 (D52C2, Cell Signaling, 5483), TBK1 (polyclonal, Cell Signaling, 3013), pIRF3 (4D4G, Cell Signaling, 4947), IRF3 (D6I4C, Cell Signaling, 11904), pSTAT1 (58D6, Cell Signaling, 9167), STAT1 (polyclonal, Cell Signaling, 9172), IFNAR-1 (polyclonal, Thermal Fischer, PA5-79441), and β-actin (C4, Santa Cruz, sc-47778).

[0131] Patient-derived organotypic tumor spheroids (PDOTS). PDOTS were generated as previously described (Jenkins et al., Cancer Discov. 8: 196-215 (2018); Aref et al., Lab Chip 18: 3129-3143 (2018)). Briefly, fresh tumor specimens were minced in full culture medium (DMEM + 10% FBS from Thermo Fisher Scientific) + 100 U / mL type IV collagenase (Life Technologies) and 50 μg / mL DNase I (Roche) in a 15 mL falcon tube preheated to 37 ° C for about 20 minutes using sterile scissors and pipetting. The dissociated material was filtered over a 100 μm filter and a 40 μm filter to generate S1 (>100 μm), S2 (40 μm-100 μm), and S3 (<40 μm) spheroid fractions, which were subsequently stored in ultra-low attachment (ULA) tissue culture plates (Corning). The S1 fraction was treated with 50 μM ADU-S100 (Chemietek) for cytokine analysis and single-cell RNA sequencing. The S2 fraction was used for ex vivo culture by resuspending it in type I rat tail collagen (Corning) at a concentration of 2.8 mg / mL and then loading it onto the central gel area of ​​a 3-D microfluidic culture device (AIM Biotech) and incubating it at 37°C for 40 minutes in a humidity chamber to allow polymerization. Collagen hydrogels containing PDOTS were hydrated in medium with or without the specified treatment. TAK-676 was provided by Takeda and diluted in dH2O. Recombinant human interferon beta (100 ng / mL; R&D Systems) was used as a positive control for activation of the STAT1 pathway downstream of STING. CD8a was neutralized with 50 μg / mL of InVivoMAb antibody and IgG control (BE0092). CXCR3 was neutralized with 5 μg / mL of human CXCR3 antibody (R&D MAB160).

[0132] PDOTS immunofluorescence and live / dead quantification. Double labeling was performed by loading microfluidic devices with Nexcelom ViaStain acridine orange / propidium iodide (AO / PI) staining solution (Nexcelom, CS2-0106) or 10 μg / mL Hoechst 33342 solution (Thermo Fisher Scientific) and 1 μg / mL PI solution (Thermo Fisher Scientific). After incubation with dye (for AO / PI, incubate at room temperature in dark for 20 minutes or for Hoechst 33342 / PI, incubate at room temperature in dark for 45 minutes), images were captured using the 4x objective lens of a Nikon Eclipse 80i fluorescence microscope equipped with an automatic electric stage (Proscan), Z-stack (Prior) and Zyla5.5sCMOS camera (Andor). Image capture and analysis were performed using the NIS-Elements AR software package. Live and dead cell quantification was performed by measuring the total cell area of ​​each dye. For additional immunofluorescence studies, PDOTS were washed with PBS and blocked with FcR blocking reagent (Miltenyi) for 30 minutes at room temperature. Directly conjugated antibodies CD326 EpCAM-AlexaFluor647 (clone 9C4), CD45-AlexaFluor647 (HI30) (BioLegend) and mesothelin-PE (clone REA1057, Miltenyi) were diluted 1:50 in 10 μg / mL Hoechst 33342 (Thermo Fisher Scientific) in PBS and loaded into the microfluidic device and incubated for 1 hour at room temperature in the dark. Spheroids were washed twice with PBS containing 0.1% Tween20 and then washed with PBS. For viability assessment, the microfluidic device was loaded with 1:1,000 Calcein AM (Thermo Fisher Scientific) in PBS. For IRF3 IF, PDOTS were treated with dH2O control or 50 μM ADU-S100 for 3 hours, washed with PBS, fixed with 4% paraformaldehyde for 15 minutes, and permeabilized with 0.1% Triton-X for 10 minutes. Cell Signaling Antibody #11904 (clone D6I4C) was diluted 1:50 with PBS and incubated for 45 minutes, washed, and then incubated with 1:100 diluted FITC-conjugated anti-rabbit secondary antibody (Thermo Fisher Scientific) for 30 minutes. PDOTS were washed twice with PBS containing 0.1% Tween20 and counterstained with 1 μg / mL Hoechst 33342 solution.Images of live / dead double staining were captured using a 20x objective as described above.

[0133] Cytokine analysis. CXCL10 ELISA (R&D Systems DIP100) and granzyme B ELISA (R&D Systems DY008) were performed on conditioned medium collected from cell cultures according to the manufacturer's instructions. Cytokine analysis of conditioned medium after 3 days of culture of explants (S1) Fig. 9B ) was assayed using the MSD U-PLEX Viral Combo 1 assay (Hu: K15343K-2) according to the manufacturer's instructions.

[0134] 2'3'cGAMP ELISA. The Cayman Chemical 2'3'cGAMP ELISA kit was performed according to the manufacturer's instructions to detect the levels of 2'3'cGAMP in the supernatants of MPM cell lines. For these experiments, 3 × 10 5 to 5×10 5 Cells were plated in 6-well plates and transfected using X-tremeGENE HP DNA Transfection Reagent in combination with Opti-MEM Reduced Serum Medium and 1 μg poly(dG:dC) (Invivogen) and incubated for 30 minutes. 2'3' cGAMP (Invivogen) was used as a positive control.

[0135] Cell culture. MPM cell lines were cultured in RPMI-1640 (Thermo Fisher Scientific) supplemented with 10% FBS (Gemini Bio-products). H226, H28, MSTO-211H, H2452, and H2052 were purchased from ATCC. MS428 was provided by Richards Lab. H2461 and H2591 were provided by NIH. Dr. (Pass et al., Ann.Thorac.Surg.59:835-44 (1995)). JMN1B (Demetri et al., Blood 74:940-6 (1989)) and MS589 (Gordon et al., Am.J.Pathol.166:1827-40 (2005)) were obtained from BWH / DFCI and shared internally with permission. All experiments were performed before reaching 10 generations. Conditioned medium from each cell line was regularly tested for mycoplasma infection by PCR using the aforementioned primers (Kitajima et al., Cancer Discov.9:34-45 (2019)).

[0136] Immunoblotting. Cells were lysed in RIPA buffer containing 1x protease inhibitor (Roche 11-836-145-001) and phosphatase inhibitors (50 mmol / L NaF and 100 mmol / L Na3VO4). Immunoblotting was performed as previously described (Kitajima et al., Cancer Discov. 9:34-45 (2019)) using the antibodies listed in Table 1. Secondary antibodies were from LI-COR Biosciences: IRDye 680LT goat anti-mouse IgG (#926-68020) and IRDye 800CW goat anti-rabbit IgG (#926-32211). Blots were imaged using the LI-COR Odyssey system.

[0137] Dynamic single-cell RNA sequencing and data analysis. The previous protocol (Sehgal et al., J. Clin. Invest. 131: e135038 (2021)) was adapted to test S1 explants from MPM PDOTS. The sample tested (#26) showed 63% baseline viability and 18-hour cytokine release corresponding to treatment ( Fig. 10B ). After 24 h of treatment in ULA culture dishes, tumor spheroids were trypsinized for 5 min in a 37°C incubator to obtain single cell suspensions. Cells were loaded onto a 10x chromium instrument (10x Genomics) according to the manufacturer's instructions. ScRNA libraries were generated using a single cell 3' kit (10x Genomics) according to the user guide. The completed libraries were quality controlled using a bioanalyzer high sensitivity DNA kit (Agilent) and then sequenced using an Illumina NextSeq500 platform.

[0138] The raw sequencing reads were processed using the 10xGenomics CellRanger bioinformatics pipeline v6.0.1. The assembled matrix was then input into the standard workflow of the R package Seurat v4.0.4. Genes expressed in at least 3 cells were retained, and only cells expressing at least 2 genes were processed downstream. In addition, cells expressing more than 7000 genes and cells with more than 10% of UMIs mapped to mitochondrial genes were removed from the analysis. All samples were prepared and sequenced on the same platform. The filter matrix was log-normalized using global scaling in Seurat. UMI and mitochondrial transcript content were used as regression parameters. The normalized matrix was scaled and centered by gene, and then principal component analysis (PCA) was used to reduce the dimensionality of highly variable genes. After visual inspection of the PCA elbow plot, the top 10 PCs were selected for further analysis. The selected PCs were clustered using the shared nearest neighbor algorithm in Seurat and default parameters.

[0139] Uniform manifold approximation and projection (UMAP) maps were calculated and plotted using the DimPlot module of Seurat. Cluster differential expression analysis was performed using the FindMarkers command in Seurat using the Wilcoxon rank sum test without threshold. Contour maps superimposed on UMAP were generated using the R package ggplot2 (Wickham, Springer-Verlag New York (2016)).

[0140] Based on comparative analysis of previously published markers, cell types were determined (Han et al., Cell 172: 1091-1107 (2018); Muhl et al., Nat. Commun. 11: 3953 (2020); Correia et al., Proc. Natl. Acad. Sci. USA 115: E5980-E5989 (2018); Gueugnon et al., Am. J. Pathol. 178: 1033-42 (2011)), as well as marker genes identified in this study, which were used to remove genes commonly expressed in cell subpopulations. Collagen encoding genes were added to the fibroblast signature. A list of gene markers used for enrichment analysis is shown in Table 1.

[0141] Table 1: Gene markers for scRNAseq

[0142]

[0143] Isolation of tumor infiltrating lymphocytes. TILs were isolated from patient specimens according to IRB protocol 02-180, and PDOTS were filtered as described above. The S3 fraction was amplified in 24-well plates using RPMI-1640 containing L-glutamine, 1% penicillin-streptomycin solution, 1 mM sodium pyruvate, 0.0375% sodium bicarbonate, 50 nM mercaptoethanol, 10% human AB serum, and 6000 U / mL IL-2, and passaged every other day at 1:2 for 8 to 10 days. After amplification, they were frozen / stored in liquid nitrogen.

[0144] Amplification and transduction of primary T cells. After informed consent, peripheral blood mononuclear cells (PBMCs) were obtained from healthy donors and separated by using Ficoll density centrifugation. Activate isolated PBMCs in complete medium (RPMI-1640 supplemented with 10% FBS in the presence of 10ng / ml IL-2) with TransAct (1:100, Miltenyi). Two days after activation, in the presence of Lentiboost (1:100, Sirion Biotech), T cells were lentivirally transduced by rotating with BCMACAR virus (1% viral volume). The BCMACAR sequence has been previously described (Works et al., Mol. Cancer Ther. 18:2246-2257 (2019)) and cloned into pHAGE lentiviral vector (Addgene plasmid #24526), ​​and the generated plasmid was sequenced and verified. To package and produce lentiviral particles, 293 Lenti-X packaging cells (Takara) were seeded into 15 cm plates (8 × 10 6 Cells / plate) for 24h, then transfected with a plasmid encoding CAR (pHIV-aMeso-CAR), pMD.2G encoding VSV-G envelope, and packaging vector psPAX2 using PEI transfection reagent (Polysciences). Viral supernatant was harvested 24 hours and 48 hours after transfection, filtered through a 0.45 μm filter, concentrated by ultracentrifugation, and stored at -80 ° C before transduction. After transduction, T cells were amplified with cytokines IL-2 (10ng / ml), IL-7 (3ng / ml), and IL-15 (10ng / ml) in RPMI-1640 supplemented with 10% FBS, and their transduction efficiency was determined by FACS 3 days after transduction.

[0145] Amplification and transduction of primary NK cells. For experiments using untreated primary NK cells, CD56+CD3-NK cells were amplified from human PBMCs (Lonza) using the CellXVivo Human NK Cell Expansion Kit (R&D Systems). After 14 days of expansion, the cells were transformed into CTS OpTmizer T cell expansion medium supplemented with 5% human AB serum (Sigma Aldrich), 1% GlutaMAX, 1% HEPES, and 1% penicillin-streptomycin and cultured in the presence of IL-2 (PeproTech or Miltenyi; 200U / mL for flow cytometry experiments, 500U / mL for killing experiments including PDOTS). Unless otherwise stated, all NK cell culture reagents were purchased from Life Technologies.

[0146] For experiments with primary NK cells treated with transduction and controls, they were extracted from whole blood leukocyte apheresis using RosetteSep (StemCell technologies) and Ficoll-Paque density gradient centrifugation according to approved Crimson research protocol T0197. The purity of the isolated NK cells was checked and cultured for 2 days in the presence of IL-15 (1 ng / mL; Miltenyi) in RPMI (Gibco) supplemented with 10% heat-inactivated (HI)-FBS (Gibco), 1% penicillin-streptomycin, 2mM L-glutamine and HEPES. Subsequently, isolated NK cells were transduced or cultured in the presence of IL-2 (500 U / mL; Miltenyi) in the NK MAC culture medium (Mildenyi) supplemented with 5% human serum (Sigma) and 1% v / v penicillin-streptomycin (Gemini Bio-products) as follows.

[0147] CAR constructs. The CAR constructs were designed with an extracellular ScFv domain and a transmembrane fragment from the CD8 protein. This was followed by the traditional 4-1BB and CD3 co-stimulatory domains. The CAR gene was designed to incorporate an HA tag for analysis using flow cytometry. For gene constructs with CXCR3, the CAR gene was followed by a P2A self-cleaving peptide nucleic acid and CXCR3. To generate CAR or CAR-CXCR3 NK cells, primary NK cells were purified from peripheral blood and activated with IL-12, IL-15, and IL-18, which resulted in NK cell activation and differentiation, producing cytokine-induced memory-like cells (CIML). Conventional NK cells (cNK) were used as controls and maintained at a low dose of IL-15 (1 ng / mL). The CAR gene was transduced into cNK or CIML NK cells using our optimized baboon lentiviral system that achieved high transduction efficiency. The anti-mesothelin CAR (αMSLN) was constructed in a pHIV backbone, as Fig. 30A As shown, it has a mesothelin-specific ScFv derived from the YP218 antibody, followed by a transmembrane domain and co-stimulatory domains (4-1BB and CD3ζ). The construct also contains an EGFP fragment separated from the CAR fragment by self-cleavage of P2A ( Fig.16A ). The CAR gene construct was packaged into the BaEV pseudotype lentiviral system by transfecting HEK-293 cells with pCMV-BaEV, pCMV-Δ8.9 and pAdv plasmids. Viral particles were titrated using Jurkat cells. Assuming that the multiplicity of infection (MOI) of Jurkat cells was 1, the viral titer was calculated by transducing NK cells with an MOI of 10. Two days after extracting NK cells, Retronectin and vectofusin were used for transduction, followed by rotational infection with + / - active lentivirus (cNK control without virus), and then cultured in the presence of IL-2 (500U / mL; Miltenyi) in NK MAC medium (Miltenyi) supplemented with 5% human serum and 1% penicillin-streptomycin (Gemini Bio-products). Using APC human lectin (HA), the percentage of NK cells expressing CAR was determined by flow cytometric analysis of GFP and surface expression of ScFv.

[0148] Immunocytotoxicity assay.For flow cytometry immunocytotoxicity assay, primary NK cells and TIL were seeded in 96-well plates at 200,000 cells per well (single NK or TIL or 1:1 of each type of cell 100,000 million) alone or co-cultured, and then treated with 10 μM or 50 μM ADU-S100 (Chemietek) or dH2O control (with or without IL-2 (Miltenyi or PeproTech)) at a specified concentration for 72 hours. After treatment, samples were stained with anti-CD45 antibody, anti-CD3 antibody, anti-CD4 antibody, anti-CD8 antibody and anti-CD56 antibody (Table 1) and Zombie Green live / dead (Biolegend 423111) and analyzed by flow cytometry as described above. Data were analyzed using FlowJo software version 10.5.3. As an orthogonal measure of viability, CellTiter-Glo was performed on primary T cells and NK cells.

[0149] CellTiter-Glo luminescent cell viability assay. Cell viability was assessed using the CellTiter-Glo luminescent cell viability assay (Promega, G7571) according to the manufacturer's instructions. For untransduced primary T cells and BCMACAR T cells, 25,000 cells per well were seeded in 96-well plates and treated with ADU-S100 or dH2O as controls at specified concentrations for 24 hours. For NK cells, 25,000 cells were seeded per well and treated with ADU-S100 or dH2O as controls at specified concentrations for 24 hours. For MPM cell lines, 10,000 cells / well (MS428) or 12,500 cells / well (H2461, H2591) were seeded in 96-well plates and treated with ADU-S100 50 μM or culture medium as controls for specified times. All conditions were tested three times and read on a Tecan Infinite Mplex microplate reader.

[0150] Autophagy staining. Autophagy was assessed by vacuolar staining to identify autolysosomes using the CYTO-ID Autophagy Detection Kit 2.0 (Enzo ENZ-51031-0050) according to the manufacturer’s instructions. 5Isolated primary NK cells or TILs were incubated in T cell growth medium (TCGM) containing 500U / mL IL-2, which was renewed every time the medium was changed to ensure proper growth and selection. CLQ in the kit (Enzo 51005-CLQ) was used at the recommended initial dose of 10μM compared to the DMSO control. The medium was changed after 24h, and the cells were treated with CLQ+10μM ADU-S100 for another 24h. The culture medium was collected and flow cytometry was performed after staining with CYTO-ID Green detection reagent 2 according to the manufacturer's instructions.

[0151] NK cell killing assay. Target cells (MPM cell lines) were separated by trypsin digestion, labeled with CellTraceViolet (CTV, LifeTechnologies), and then seeded in 96-well plates at a cell density of 25,000 cells per well. Target cells were allowed to adhere for 12 to 16 hours, and then NK or αMSLN-CAR-NK cells were added with different effector cell to target cell ratios (E:T) (1:1, 2:1, 5:1 and 10:1) with or without ADU-S100 (50 μM). After 6 hours of co-culture, cells were harvested and incubated with apoptosis marker annexin V (PE) antibody and live / dead stain 7-AAD (Biolegend). Cells were analyzed on BD LSRFortessa (BD Biosciences) equipped with FACSDiva software. Data were analyzed using FlowJo software version 10.5.3. Apoptotic cells were assessed by gating on the CTV+ population and expressed as a percentage of live or dead (late apoptotic) cells. For each target MPM cell line, apoptotic cell analysis was performed using NK cells derived from up to 4 different healthy donors to incorporate baseline donor variability.

[0152] NK cell infiltration assay. Immune cell infiltration was assessed as previously described (Kitajima et al., Cancer Discov. 9:34-45 (2019); Mahadevan et al., Cancer Discov. 11:1952-1969 (2021)). Briefly, 5×10 5Mesothelioma cancer cell spheroids (H2591, H2461, H226) were generated from cells. In the last 6 hours of spheroid formation, H226 cells were treated with 50 μM ADU-S100 to establish a cytokine gradient. The sample was then precipitated and resuspended in type I rat tail collagen (Corning) at a concentration of 2.5 mg / mL after adding 10xPBS containing phenol red and adjusting the pH with NaOH. The pH was confirmed to be 7.0-7.5 using PANPEHA Whatman test paper (Sigma-Aldrich). Cells and collagen were kept on ice to prevent polymerization. The sphere-collagen suspension was then injected into the central gel area of ​​a 3D DAX-1 microfluidic cell culture chip (AIM Biotech). A microfluidic device was used as previously described (Aref et al., Lab Chip 18:3129-3143 (2018)), in which the central area contained a cell-collagen mixture (3×10 in 10 μL) in a 3D microenvironment. 4 H2591 cells and H2461 cells, 2×10 4 H226 cells) and flanked by 2 channels of culture medium. After injection, the collagen hydrogels containing cells were incubated in a humidity chamber at 37°C for 40 min and then hydrated with culture medium, and labeled primary NK cells (E:T ratio 2:1) were added to one of the side channels. Primary NK cells were labeled with Cell Tracer (Red) (Thermo Fisher Scientific) according to the manufacturer's instructions. After 96 h of incubation, cancer cell spheroids and infiltrating immune cells were stained for viability (incubated with 1 μg / mL propidium iodide solution for 20 min; Thermo Fisher Scientific). For experiments performed with a CXCR3 neutralizing antibody (R&D MAB160), NK cells were pretreated for 30 min before loading.

[0153] For quantification, images were captured on a Nikon Eclipse 80i fluorescence microscope equipped with a Z-stack (Prior) and a CoolSNAP CCD camera (Roper Scientific). Image capture and analysis were performed using the NIS-Elements AR software package. A complete image of the device was obtained by stitching multiple captures. Immune cell infiltration into the 3D tumor microenvironment was quantified by measuring the total cell area of ​​the cell tracer dye throughout the gel area. For CXCR3 neutralizing antibodies (R&D MAB160; Figure 6B ) experiments, the staining was quantified in the square area in the center of the channel to focus on the effect of CXCR3 ligands released by tumor cells. As with the above PDOTS, the percentage of dead cells was quantified.

[0154] 3D vascular model. To generate a tumor vascular model, H226 spheroids were mixed with collagen rat tail hydrogel (2.5 mg / ml) and injected into the central gel area of ​​the 3D microfluidic chamber (10 μL to 15 μL per microfluidic chamber). After incubation in a sterile humidity chamber at 37°C for 30 minutes, the side wall of one channel (culture medium channel) was coated with 150 μg / ml collagen PBS solution to allow ECs to better adhere to the channel. After 15 minutes, the channel was rinsed once with culture medium. To create 3D blood vessels, 3×10 6 25 μL cell suspension of human umbilical vein endothelial cells (HUVEC; C2519AS, Lonza) with 10 cells / ml was injected into the collagen-coated medium channel. The channel was rotated twice to create a fused hollow lumen 3D blood vessel. In order for the cells to attach to the medium-gel interface and form a monolayer, the chip was incubated with the cell side down for 15 minutes. Next, 50 μL of cell suspension was re-injected and the chip was flipped to cover the upper part of the 3D vascular channel. After incubation in a humidity chamber at 37°C for 90 minutes, the cell culture medium was gently added to both channels and further incubated to form a fused monolayer. After vascularization, NK cells (labeled with cell tracers) were added to the 3D blood vessels at an E:T ratio of 2:1. STING agonists (ADU-S100, TAK-676) were treated in the fluid channel opposite the vascular barrier. At 24 hours, NK cell migration + / - blood vessels was quantified. Image capture and analysis were performed using a fluorescent confocal microscope and processing software. The 3D vascular channel was rinsed in PBS and fixed with 4% PFA for 15 min at room temperature. The cell membrane was permeabilized with 0.1% Triton X-100 for 5 min at room temperature and washed twice with PBS. HUVEC cells were stained for F-actin with green phalloidin (Thermo Fisher Scientific A12379) and Hoechst 33342. Images were captured on a Nikon Eclipse 80i fluorescence microscope equipped with a Z-stack (Prior) and a CoolSNAP CCD camera (RoperScientific).

[0155] Immune cell migration assays were performed as previously described (Kitajima et al., Cancer Discov. 9:34-45 (2019)). Briefly, NCI-H226 cells were plated at 5 × 10 5 Cells were plated in 6-well plates at a density of 10 cells / mL and treated with 50 μM STING agonist (ADU-S100) for 24 h. 5spheroids were generated by 24 h in ultra-low attachment culture dishes and labeled with fluorescent dye (cell proliferation dye eFluor 450, Invitrogen, 65-0842) according to the manufacturer's instructions. The spheroids were pelleted and then resuspended in type I rat tail collagen (Corning) at a final concentration of 2.5 mg / mL after adding 10xPBS containing phenol red on ice. The pH of the resulting spheroid suspension was adjusted to 7.0–7.5 using NaOH and confirmed using PANPEHA Whatman test paper (Sigma-Aldrich). The spheroid-collagen suspension was then introduced into the central channel of a 3-D microfluidic cell culture chamber (AIM Biotech, design described previously (Aref et al., LabChip 18:3129-3143 (2018)). The collagen hydrogel containing cancer cell spheroids was incubated in a humidity chamber at 37°C for 40 min, and then a side channel located next to the central channel was perfused with RPMI-1640 medium containing NK cells at an effector cell to target cell ratio (E:T) of 2:1. Cancer cell spheroids and NK cells were co-cultured for 3 days, after which NK cells migrated into the collagen hydrogel and migration was observed by images captured on a Nikon Eclipse 80i fluorescence microscope equipped with a Z-stack (Prior) and CoolSNAP CCD camera (RoperScientific) and analyzed using the NIS-Elements AR software package. Immune cell infiltration into the central channel was quantified by measuring the total area occupied by Cell Tracker Red dye-positive cells located in the region of interest (ROI; 6ROI / microfluidic cell culture chamber).

[0156] Statistical analysis. Statistical significance was assessed using unpaired two-tailed Student's t-test for pairwise comparisons, one-sample t-test for expected values ​​of 0% change or 100% control, or one-way ANOVA followed by Tukey post hoc test. Kruskal Wallis global test followed by Dunn's multiple comparison post hoc test was used for nonparametric analysis of IHC scores and mRNA expression in MPM cell lines (data obtained from the Cancer Cell Line Encyclopedia of the Broad Institute). P values ​​less than 0.05 were considered significant. Asterisks used to indicate significance correspond to * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. Columns represent mean±SD. Asterisks appear only in pairs of items of interest in post hoc tests after one-way ANOVA. GraphPad Prism (version 9.2.0) was used for all statistical analyses.

[0157] Example 2: Malignant pleural mesothelioma (MPM) induces STING activation

[0158] Described herein is the discovery that malignant pleural mesothelioma strongly expresses tumor cell STING and responds to ex vivo STING agonist treatment. Dynamic single-cell RNA sequencing of explants treated with STING agonists revealed activation of the CXCR3 chemokine primarily in tumor cells and cancer-associated fibroblasts, as well as T cell cytotoxicity. In contrast, primary NK cells resisted STING agonist-induced cytotoxicity. STING agonists enhanced migration and killing of NK cells, and in particular anti-mesothelin chimeric antigen receptor (CAR)-NK cells, thereby improving therapeutic activity. These studies reveal the fundamental importance of using human tumor samples to evaluate innate and cellular immunotherapies. Functional analysis of mesothelioma tumor explants with elevated tumor cell STING expression supports the combination of NK and CAR-NK cell therapies, revealing different consequences of human STING agonist treatment.

[0159] The progress of studying human TIME using patient samples allows the development of immunotherapy by processing patient-derived organotypic tumor spheres (PDOTS) and tissue fragment explants in short-term culture (Jenkins et al., Cancer Discov. 8: 196-215 (2018); Voabil et al., Nat. Med. 27: 1250-1261 (2021)). So far, these platforms have focused on anti-PD (L) -1 immune checkpoints and can parallel patient responses, but they also provide hope for designing new cancer immunotherapies. Compared with patient-derived xenografts grown in humanized mouse models, these systems provide the ability to interrogate immune responses in natural human tumor immune environments and have the potential to study cell therapies without interference from the mouse microenvironment. In addition, insights gained from direct human TIME simulations may also help narrow the conversion differences of immunotherapies that are effective in syngeneic mouse models but fail in clinical trials.

[0160] Recently, multiple human cancer types have been shown to silence STING and downstream interferon responses to avoid immune detection (Ghoshet al., Cancer Cell 39:494-508 (2021); Kitajima et al., Cancer Discov. 9:34-45 (2019); Lau et al., Science 350:568-71 (2015)), clearly demonstrating the important role of tumor cell STING signaling in human cancer. Higher STING expression is also associated with better response to treatment across cancer types (Hayman et al., Nat. Commun. 12:2327 (2021); Zugazagoitia et al., Clin. Cancer Res. 26:4360-4368 (2020); Qi et al., Biosci. Rep. 40:BSR20202603 (2020); Chon et al., J. Cancer 10:4932-4938 (2019)). However, how STING agonists affect tumor cells and different cell types in human TIME has not been carefully studied and may provide information for the development of new therapeutic combinations including cell therapy. This article addresses this issue by conducting large-scale studies using PDOTS and developing methods for dynamic single-cell RNA sequencing in tumor explants, as well as dissecting STING agonist responses in inflammatory tissue types.

[0161] To identify human tumor tissue types with intact STING, immunohistochemical analysis was performed on 300 archival samples from different thoracic malignancies ( Figure 1A ). Among the malignancies evaluated, MPM expressed the highest levels of STING. MPM originates from the serosal layer of the lung and has a poor prognosis despite the recent incorporation of combined checkpoint immunotherapy into standard treatment (Janes et al., N. Engl. J. Med. 385: 1207-1218 (2021)). Compared with non-small cell lung cancer (NSCLC), thymoma, and especially small cell lung cancer (SCLC) (Mahadevan et al., Cancer Discov. 11: 1952-1969 (2021); et al., Nat. Med. 24:1143-1150 (2018)), MPM demonstrated that STING protein is almost universally overexpressed in tumor and stromal cells ( Figure 1A and Fig. 7A STING is also highly expressed in benign pleura, consistent with its baseline upregulation in mesothelial cell types ( Figure 7B). The MPM-derived cell line expressed high levels of STING protein, but neither the cell line nor the tumors exhibited baseline cGAS-STING pathway activation as measured by phosphorylated IRF3, CXCL10, and IFIT1 expression, and secreted 2'3'-cGAMP ( Figure 7C , Figure 8A-8D Flow cytometry-based immune profiling of a large number of resected MPM specimens further demonstrated a robust immune infiltrate in the majority of tumors, but with a signature of exhaustion of multiple immune cell subsets, including heterogeneous expression of the checkpoint proteins PD-1, TIM-3, and LAG-3 ( Figure 1B – Figure 1D and Figure 7D-Figure 7E ) (Awad et al., Cancer Immunol. Res. 4: 1038-1048 (2016)). T cell characterization revealed terminal differentiation consistent with exhaustion; monocyte / macrophage subtypes showed enrichment of intermediate cells; NK cell characterization showed reduced cytotoxic capacity (increased CD56 bright / CD16 low compared to circulating NK cells; Figure 1D ). Thus, MPM express high levels of STING and exhibit inflammatory but depleted TIME.

[0162] Despite the lack of baseline STING pathway activation, multiple MPM cell lines treated with the clinical STING agonist ADU-S100 (Corrales et al., Cell Rep. 11:1018-30 (2015); Amouzegar et al., Cancers (Basel) 13:2695 (2021)) exhibited potent pathway activation, secreting high levels of CXCL10 ( Figure 8A-8D The MPM cell lines that were least responsive to clinical STING agonists in vitro showed reduced expression of the IRF3 transcription factor ( Fig. 8A , Fig.8D and Fig. 8E ). Overall, however, the high levels of inactive basal STING and the significant induction of CXCL10 release by STING agonism suggest that STING signaling has elicited a response in MPM. Next, we analyzed STING agonism in human tumor specimens using a freshly resected MPM tumor explant model that retains the relevant TIME ( Figure 2A)(Jenkins et al., Cancer Discov. 8:196-215(2018)). After treatment, 40 μm to 100 μm (S2) PDOTS were suspended in collagen and treated for 6 days, and the response was evaluated by live / dead immunofluorescence and cytokine production (Jenkins et al., Cancer Discov. 8:196-215(2018)). After 6 days of treatment in specific samples, ADU-S100 induced CXCL10 release and strongly killed PDOTS ( Figure 2A-2B and Figure 9A-9B ). Figure 2B Cell area and live / dead percentage quantification for each staining are shown. t-test vs. dH2O control: **p<0.01, ****p<0.0001. Scale bar = 100 μm. Importantly, STING activation in cultured MPM cell lines did not cause cytotoxicity, suggesting that the role of TIME ( Fig.8F Overall, treatment of 35 patient specimens with ADU-S100 showed statistically significant cell death exceeding 20% ​​of control in 12 / 35 (34%) patient specimens (p<0.05), with 7 specimens approaching clinical response criteria with a cell area reduction of more than 30%, p<0.05 ( Figure 2C-Figure 2D and Fig. 9C ). Figure 2C Responses based on criteria of viable cell area (≥30% reduction) and ≥20% increase in cell death (p<0.05 between ADU-S100 and dH2O control by t-test of treated triplicate wells) are shown. Figure 2D The responses were shown to be reduced in viable cell area, epithelioid MPM (E), biphasic MPM (B), yes / no (Y / N) neoadjuvant therapy, and male / female (M / F). There was no significant trend toward higher responses in patients who received neoadjuvant chemotherapy (HR 1.4, 95% CI 0.23-7.27, chi-square p=0.72), and responses were not associated with histology (epithelioid and biphasic specimens were tested), age, or sex (Table 10). In addition, treatment with TAK-676, a next-generation systemic STING agonist in the second clinical phase, also showed promising in vitro activity, with responses observed in 4 / 13 (31%) patient specimens ( Figure 2F A potential correlation between tumor CD8 abundance and response to in vitro treatment with ADU-S100 was observed (R2 = 0.35, p < 0.05; Fig. 10A ), prompting testing the impact of CD8 neutralization. Indeed, anti-CD8 antibody treatment partially rescued ADU-S100 cytotoxicity in three patient specimens ( Figure 2E-2G and Figure 10A-10B), driving the development of higher-resolution methods to understand the impact of STING agonism on different cells during TIME.

[0163] Table 2: Patient Demographics of MPM Tumors Treated with Ex Vivo STING Agonists

[0164]

[0165]

[0166] Example 3: Dynamic scRNAseq of MPM explants.

[0167] After 24 hours of STING agonist treatment, we used an adapted approach for dynamic single-cell RNA sequencing (scRNAse) (Sehgal et al., J. Clin. Invest. 131:e135038 (2021)), focusing on samples with a modest dose-dependent killing response to ADU-S100 that was rescued by CD8 neutralization ( Figure 3A-3D , Fig. 10B , Figures 11A-11D and Figure 12A-12C ). Figure 3D The bar graphs in the figure show the scores for each cluster by treatment, normalized to the number of cells per sample. Flow cytometry analysis before treatment shows the mean percentage of T cells and the mean percentage of the monocyte / macrophage population above ( Figure 1B and Fig. 10C IRF3 immunofluorescence also showed nuclear translocation after ADU-S100 treatment, confirming effective STING activation in PDOTS ( Fig.10E ). Tumor fragments larger than 100 μm suspended in culture medium were used for short-term scRNAseq analysis, confirming that size filtration did not alter the leukocyte composition of each fraction (Jenkins et al., Cancer Discov. 8: 196-215 (2018)) Fig. 10D UMAP clustering validates broad representation of tumor cell, fibroblast, and immune cell populations ( Figure 3A , Figure 11A-11B Interestingly, ADU-S100 strongly induced the expression of CXCR3 ligands (CXCL9, CXCL10, CXCL11), mainly in the mesothelin (MSLN)-positive tumor cell subpopulation (Cluster 1) and cancer-associated fibroblasts (CAFs; Clusters 4 and 5), but not in the myeloid cell population (Clusters 7 and 8; Figure 3B – Figure 3C). This analysis also revealed that STING agonists potently and uniquely induced IL-33 expression in CAFs, whereas other ISGs such as IFIT1 showed broader expression across the cell population, confirming broad target engagement ( Figure 3C Differential expression analysis showed that the MPM cell subset with the highest expression of CXCR3 ligands in response to ADU-S100 (Cluster 1) also showed an increase in a large number of ISGs and downregulation of TGFB1 expression compared to other MPM cells (Cluster 0, Cluster 3) ( Fig. 11C , Table 3 –

[0168] Table 4). Multiple granzyme- and perforin-positive T cells and NK cells were identified in cluster 2 ( Fig.11D ), which is consistent with an effect of CD8 neutralization and a potential role for T cell and / or NK cell-mediated killing. However, analysis of the abundance of cells derived from each individual sample ( Figure 3D and Figure 12A-12C The contour plots and score plots in Figure 2 (in the figure) showed that after high-dose STING agonist treatment, the CD8-positive cell population (including Tregs) was reduced, in contrast to the increased STING agonist response in MPM cluster 1. Analysis of NK cell ligands in tumor cells showed that ADU-S100 dose-dependently increased the expression of HLA-A / B / C (inhibitory), plus increased NECTIN2 (CD112; activating), while reducing the expression of the major inhibitory ligand PVR (CD155) (Lupo et al., J. Hematol. Oncol. 13:76 (2020)) ( Fig. 12B ). These data reveal the prominent role of tumor cells and CAFs as targets for STING agonism, promoting the release of T cell and NK cell chemokines and changes in tumor / CAF cell states. However, these results are also consistent with reports that excessive STING activity may be toxic to T cells (Cerboni et al., J. Exp. Med. 214: 1769-1785 (2017); Larkin et al., J. Immunol. 199: 397-402 (2017); Gulen et al., Nat Commun. 8: 427 (2017)), suggesting that STING-mediated enhancement of tumor CXCR3 chemokine release may be offset by the cytotoxicity of immune effector cells.

[0169] Table 3: 40 genes with the greatest increase in expression after STING agonist treatment

[0170]

[0171] Table 4: 20 genes with the greatest reduction in expression after STING agonist treatment

[0172]

[0173] Example 4: STING agonists are toxic to human T cells.

[0174] To further explore this observation, STING-induced cytotoxicity against T cells (Cerboni et al., J. Exp. Med. 214: 1769-1785 (2017); Larkin et al., J. Immunol. 199: 397-402 (2017); Gulen et al., Nat. Commun. 8: 427 (2017)), as well as cytotoxicity in other immune cell types, was assessed using the model described herein. ADU-S100 treatment was cytotoxic to T cells in MPM tumor explants, as measured by flow cytometry, compared to downstream IFNβ exposure, increasing over time within 24 to 72 hours of STING agonist treatment ( Figure 4A and Fig.13A T cells purified from peripheral blood expressing or not expressing the B-cell maturation antigen chimeric antigen receptor (BCMACAR) also showed dose-dependent cytotoxicity after STING agonist treatment ( Figure 4B , Fig. 13B ), which may limit the recently proposed combination of STING agonists with CAR T cells (Xu et al., J. Exp. Med. 218: e20200844 (2021); Smith et al., J. Clin. Invest. 127: 2176-2191 (2017)).

[0175] Although NK T cells are also sensitive to STING agonists ( Figure 4A ), but human NK cells did not show significant cytotoxicity in response to STING agonist treatment, whether cultured in IL-2 or co-cultured with TILs ( Figure 4A-4C and Figures 13A-13D ). Figure 4C Flow cytometry is shown after 72 hours of treatment with 50μM ADU-S100, 10μM TAK-676, or dH2O control + / - 200U / mL IL-2, gated on live cells out of 10,000 total events expressing CD8 or CD56. Batch 3 primary NK cells were expanded from PBMCs and TILs from a 66-year-old male with stage I NSCLC. One-way ANOVA with corrected pairwise comparisons p<0.01: ** p<0.01, ***p<0.001. These findings apply to human tumors and expanded primary NK cells, even after 72 hours of high-dose ADU-S100 or TAK-676 exposure. NK cells rely primarily on metabolism through oxidative phosphorylation (Keppel et al., J. Immunol. 194:1954-62 (2015)), which requires sustained autophagic flux (Wang et al., Nat. Commun. 7:11023 (2016)), while T cells rely on glycolysis and tolerate defective autophagy (Clarke et al., Nat. Rev. Immunol. 19:170-183 (2019)). In fact, consistent with the increase in autophagic flux after ADU-S100 treatment ( Fig.13E ), the level of STING protein is low in NK cells, and STING is rapidly degraded within 3 to 6 hours of STING agonism ( Figure 4D and Figure 13F-13G In contrast, ADU-S100 phosphorylated and activated STING in T cells, but with minimal degradation ( Figure 4D and Fig.13F ). Autophagy-dependent STING recycling in NK cells was confirmed because treatment with chloroquine (CLQ), which blocks autophagic flux by inhibiting the fusion of autophagosomes with lysosomes (Mauthe et al., Autophagy 14:1435-1455 (2018)), prevented ADU-S100-induced STING degradation ( Figure 13G ). Thus, although STING agonist treatment elicited cytotoxicity against effector T cells, primary NK cells were largely unharmed.

[0176] Example 5: STING agonists enhance NK cell therapy.

[0177] NK cells are usually present in low numbers in MPM specimens ( Figure 1B ), and may also be inhibited by inhibitory signals on tumor cells such as MHC-1, which may increase after STING agonist treatment ( Fig. 12B ). We next investigated STING agonism in combination with adoptive transfer of either primary NK cells or engineered NK cells to determine whether combining tumor CXCR3 chemokine release with effector cell types resistant to STING agonist cytotoxicity represents a promising therapeutic strategy. While primary NK cells added alone to the treatment channel of the microfluidic device failed to kill MPM PDOTS, co-treatment with ADU-S100 using cells from 2 of 3 donors significantly enhanced the primary NK cell response ( Figure 5A and Fig.14A). Furthermore, ADU-S100-mediated enhancement of NK cell cytotoxicity in MPM PDOTS was abolished by CXCR3 neutralization ( Figure 5B ). To further overcome the potential inhibitory effects of MHC-I, the next focus is the anti-MSLN CAR strategy that is being clinically developed for MPM (J Janes et al., N. Engl. J. Med. 385: 1207-1218 (2021)), which uses NK cells instead of T cells as an alternative carrier of anti-MSLN CAR ( Fig.16A Indeed, using MSLN-positive MPM specimens, we found that anti-MSLN CAR NK cells significantly enhanced the activity of ADU-S100 in PDOTS samples at 6 days, which was the sample that had the lowest response to ADU-S100 treatment alone ( Fig. 14B ). In addition, in contrast to the rebound seen on day 10 after single-dose therapy, the combined addition of NK and especially CAR-NK cell therapy with ADU-S100 promoted profound growth inhibition of MLSN+PDOTS in culture over time (Fig. 5C; Fig. 14C ). These data confirm that sustained exposure to STING agonists is not toxic to NK cell therapy and suggest that it can enhance activity, especially when combined with anti-MSLN CAR NK cells.

[0178] To consider the role of tumor cells alone and further validate these findings, MPM cell lines that highly express STING and secrete CXCL10 over time during STING agonist treatment (H2591, H226, MS428) or cell lines that uniquely lack STING expression and are unresponsive to STING agonists (H2461; Fig. 8A , Fig.15A ) and compared the migration and killing of NK cells treated with - / +ADU-S100 in vitro ( Figure 6 , Figures 15A-15F , Figure 16A-16C ). STING agonism enhanced NK cell granzyme release only in co-culture with STING-expressing MPM cells ( Fig. 15B ) and tumor cell apoptosis ( Fig. 6A – Fig.6D , Fig. 15C , Fig. 16B – Fig. 16C These findings were observed at different E:T ratios ( Fig. 15C and Fig. 16B ), 2D and 3D culture ( Figure 6A-6D ) and vary with different NK cell donors, as in experiments using patient specimens ( Figure 5A-5D , Figure 6A-6D , Figure 14A-Figure 14C , Fig. 15C, Fig. 16C The STING-agonist-induced increase in NK cell migration toward tumor cells was rescued by treatment with a CXCR3-neutralizing antibody ( Figure 6B , Fig.15E ), similar to the NK cell cytotoxicity results in MPM patient specimens ( Figure 5B To simulate NK cell migration across the vascular barrier, human umbilical vein endothelial cells (HUVECs) were cultured in 3D to form blood vessels, and then physiological NK cell migration out of the vessels and across collagen to reach the MPM tumor cell line was assessed ( Fig.15F ADU-S100, and especially TAK-676, enhanced NK cell migration in the presence and absence of vascular barriers, with a predicted reduction in total migration through blood vessels ( Figure 6C ).

[0179] Finally, since PDOTS data suggest that mesothelin CAR construct expression can enhance adoptive NK cell therapy for MPM when combined with STING agonists ( Figure 5C ), so this treatment combination was evaluated in vitro to assess cytotoxicity. Anti-mesothelin CAR expression further enhanced NK cell killing in vitro and, in combination with ADU-S100 treatment, resulted in the death of most tumor cells ( Fig.6D and Figure 16A-16C Together, these data reveal that STING agonism in STING-positive human tumor models activates antiviral signaling programs, promotes CXCR3-ligand chemokine release from tumor cells, enhances NK cell recruitment and cytotoxicity, and may have potent combinatorial activity with NK cell therapy.

[0180] Evaluating human tumors in short-term cultures that preserve the tumor immune microenvironment can overcome some of the limitations of mouse models, patient-derived xenografts, and passaged organoids to potentially provide information for clinical trials of next-generation immunotherapy combinations (including cell therapies). This article describes dynamic single-cell RNA sequencing of human tumor explants treated with ADU-S100 to dissect the mechanism of action of clinical-stage STING agonists. STING agonism engages its targets in most cells of TIME, but mainly drives CXCR3 chemokine activation in tumor cells and cancer-associated fibroblasts, while causing T cell cytotoxicity. The passivation of effector T cell activity is an unexpected result, which may lead to the disappointing clinical activity of STING agonists in humans so far. However, these studies have shown that this shortcoming can be overcome by adding NK cell therapy (Myers et al., Nat. Rev. Clin. Oncol. 18: 85-100 (2021)), which benefits from STING agonists to enhance NK cell migration and killing.

[0181] More generally, available data from mouse models and clinical trials of injectable STING agonists support the complex interactions of STING activation in TIME. In fact, cell types other than CD8 T cells, such as monocytes and NK cells, may be involved in the rare clinical responses to STING agonists reported in patients (Harrington et al., Annals of Oncology 29: viii712 (2018); Meric-Bernstam et al., Journal of Clinical Oncology 37: 2507 (2019)). In addition, recent work in syngeneic mouse models has revealed an important role for NK cells in tumor control mediated by endogenous STING agonist ligand 2'3'-cGAMP (Marcus et al., Immunity 49: 754-763 (2018); Nicolai et al., Sci. Immunol. 5: eaaz2738 (2020)). These data demonstrate in vivo NK cell responses to mouse STING agonists in a manner that is otherwise difficult to mimic using humanized mouse xenografts. In a related paper describing the preclinical activity of TAK-676, we observed enhanced trafficking and activation of NK cells following systemic administration of TAK-676 in our mouse model. In our ex vivo model, TAK-676 treatment was also particularly effective in overcoming the human vascular barrier.

[0182] Adding to the complexity of injectable STING agonist trials is the potential threshold effect of cytokine release, whereby tumor cell STING activation spans from metastasis promotion (Chen et al., Nature 533:493-498 (2016); Bakhoum et al., Nature 553:467-472 (2018)) to immune rejection. In MPM, minimal baseline phosphorylation of downstream IRF3 is observed in patient specimens ( Figure 7C ), and the extracellular 2'3'-cGAMP released by the cell line is negligible ( Figure 8C ), suggesting that the contribution of basal cGAS-STING signaling to the observed immune exhaustion is likely low ( Figure 1A-Figure 1D ). But these data suggest that elevated STING expression in tumor cells in MPM is particularly susceptible to therapeutic STING agonism, especially when combined with NK cell therapy. This vulnerability may also extend to other tumor types with high basal STING expression, or STING-silenced tumors treated with epigenetic inhibitors (Falahat et al., Proc. Natl. Acad. Sci. USA 118: e2013598118 (2021)).

[0183] Clinical development of STING agonists is further limited by the narrow therapeutic window of injectable agents, which are quickly cleared (Harrington et al., Annals of Oncology 29: viii712 (2018); Meric-Bernstam et al., Journal of Clinical Oncology 37: 2507 (2019)). Although new sustained-release and systemic formulations of STING agonists may address some of these issues (Amouzegar et al., Cancers (Basel) 13: 2695 (2021)), the data disclosed herein suggest that sustained exposure may kill endogenous effector T cells and also limit combination with adoptively transferred transgenic TCR-T or CAR T cell therapy (Xu et al., J. Exp. Med. 218: e20200844 (2021); Smith et al., J. Clin. Invest. 127: 2176-2191 (2017)). In contrast, the findings described herein, that NK cells are resistant to sustained high-dose STING agonist exposure and are in fact activated and recruited to kill MPM cells, support this new immunobiology and provide a direct combination approach for clinical development of NK cell therapy. In addition, the benefits of adding STING agonists to NK cell therapy may not necessarily rely on CAR constructs, but rather allow for combination with a variety of emerging NK effector cells (Myers et al., Nat. Rev. Clin. Oncol. 18: 85-100 (2021)). Therapies that enhance natural NK cell activation may also be effective in combination with STING agonists. Interestingly, while single-cell RNA sequencing data showed upregulation of NK inhibitory MHC I, they also revealed specific regulation of CD112 and CD155, which can be combined to activate NK cells ( Fig. 12B ), especially when combined with adoptive NK cell therapy ( Figure 5A-5D and Figure 6A-6D ).

[0184] The timing and sequence of combined immunotherapy remain critical, as high-dose STING agonism (together with NK cell infusion) can prevent cytotoxicity to T cells and allow for later T cell cross-priming via NK to dendritic cell to T cell crosstalk, thereby enhancing anti-tumor immunity. Potent / specific TBK1 inhibitors described in Jenkins et al., Cancer Discov. 8:196-215 (2018) activate T cells and can be a combination therapy with the inventive therapies described herein.

[0185] Example 6: CXCR3 overexpression in CAR-NK cells triggers migration and homing to the tumor microenvironment.

[0186] After stimulation with 200 ng of recombinant human CXC motif chemokine ligand 10 (hCXCL10) at different time points, CXCR3 was degraded from the cell surface of primary NK cells and NK cell lines NK92 and JURKAT, both of which express CXCR3, as shown in Figure 17a- Fig.18B As shown, measured by flow cytometry and expressed as the median fluorescence intensity (MFI) of CXCR3 receptors. CXCR3 is also degraded from cNK cells expressing CAR and CAR-CXCR. Figure 18C-Figure 18D shows the surface expression of CXCR3 measured by flow cytometry of primary NK cells (cNK) expressing CAR, CAR-CXCR3 or controls stimulated with 200ng recombinant human CXCL10 at different time points (0 minutes and 60 minutes). After hCXCL10 treatment for 1 hour, CXCR3 was degraded from the surface of cNK NT, CAR-cNK and CAR-NK CXCR+ cells ( Fig.19C ). After 1 hour of hCXCL10 treatment, CXCR3 was also degraded from cytokine-induced memory-like (CIML) NK NT, CIMLCAR-NK, and CIML CAR-NK CXCR+ ( Fig.19D ). These data suggest that CXCR3 is lost from the cell surface following hCXCL10 treatment and that CXCR3-overexpressing cell lines (NK92, Jurkat) have higher CXCR3 surface expression and, therefore, lower CXCR3 receptor internalization relative to endogenous NK CXCR3 expression.

[0187] Immune cell migration assay was performed on control cNK cells and cNK cells overexpressing CXCR3. CXCR3 overexpression caused NK cells to migrate to H226 MPM cells ( Figure 20A-20B ) and H2591 MPM cells ( Figure 21A-21B )migrate.

[0188] Next, we tested the effects of CXCR3 and ADU-S100 (abbreviated as ADU) on NK cell migration. ADU-S100 increased the migration of cNK cells, but reduced the migration of cNK cells overexpressing CXCX3 to H226 MPM cells ( Figure 22A-22B ) and H2591 MPM cells ( Figure 22A-Figure 23B )migrate.

[0189] In addition, CXCR3 overexpression increased CAR-NK cell migration and cytotoxicity. Migration of CAR-NK control cells or CAR-NK cells overexpressing CXCR3 (CAR-NK CXCR+) toward H226MPM cells and killing of H226 cells were tested. Figure 24A-24B ) and killed more H226 cells than NK control cells ( Figure 24C-Figure 24D ).exist Fig.24D In the figure, cNK cells are labeled in red, all cells (live and dead) are labeled in blue with DAPI, and dead cells are labeled in yellow with Draq7; the scale bar represents 150 μm.

[0190] The STING agonist ADU-S100 enhances CAR-NK migration. The migration of CAR-NK control cells and CAR-NK CXCR+ cells was tested in the presence and absence of ADU-S100. ADU-S100 did not affect the migration of CAR-NK control cells toward H226 MPM cells; after ADU-S100 treatment, the migration of CAR-NK CXCR+ cells increased ( Figure 25A-25B ).

[0191] The expression of CAR was confirmed in cNK and CIML NK cells isolated and generated from two donors. Untransduced (abbreviated as UNT) cNK and CIML NK cells did not show any binding to anti-APC-HA antibody, whereas cNK and CIML NK cells transduced with an anti-mesothelin CAR construct (containing a human lectin (HA) tag; abbreviated as CAR) with or without CXCR3 overexpression showed increased binding to anti-APC-HA antibody (abbreviated as CAR-CXCR3) ( Figure 26A-Figure 27B For donor 27, the proportion of high CAR cNK cells was 23.9 ± 6.8% in cells expressing CAR-CXCR, compared with 44.8% in cells expressing CAR alone and 0.02% in non-transduced cells ( Fig.26A For donor 28, the proportion of high CAR cNK cells was 21.7 ± 1.8 in cells expressing CAR-CXCR, compared with 44% in cells expressing CAR alone and 0.06% in untransduced cells ( Fig.27A For donor 27, the proportion of high CAR CIML cells was 49.9 ± 8.3% in cells expressing CAR-CXCR, compared with 64.5% in cells expressing CAR alone and 0.01% in untransduced cells ( Fig.26B For donor 28, the proportion of CAR-high CIML cells was 39 ± 7.3% in cells expressing CAR-CXCR, compared with 53.7% in cells expressing CAR alone and 0.02% in non-transduced cells ( Fig.27B ).

[0192] Similar to the confirmation of CAR expression, flow cytometry was performed on cNK ( Fig.28A , Fig.29A) and CIML NK cells ( Fig.28B , Fig.29B ) were confirmed to be overexpressed in cNK and NK cells that were not transduced or transduced with anti-mesothelin CAR constructs together with or without CXCR3 overexpression constructs ( Figure 28A-Figure 29B For donor 27, the proportion of cNK cells with high CXCR3 increased to 57.6 ± 3.3% in cells expressing CAR-CXCR, compared to 36.7% in cells expressing CAR alone ( Fig.28A For donor 28, the proportion of cNK cells with high CXCR3 increased to 89.9 ± 0.3% in cells expressing CAR-CXCR, compared to 77.1% in cells expressing CAR alone ( Fig.29A For donor 27, the proportion of CIML cells with high CXCR3 increased to 87.3 ± 1.2% in cells expressing CAR-CXCR, compared to 64.3% in cells expressing CAR alone ( Fig.28B For donor 28, the proportion of CIML cells with high CXCR3 increased to 97.2 ± 0.4% in cells expressing CAR-CXCR compared to 89.8% in cells expressing CAR alone ( Fig.29B ).

[0193] More generally, translation of combination immunotherapies into the clinic remains challenging, and the use of patient-derived tumor samples to investigate innate / adaptive immune crosstalk and the impact of activating one pathway on the broader TIME may inform optimal approaches to enhance emerging cell therapies and overcome immune exhaustion.

[0194] All patent publications and non-patent publications are indicative of the level of skill of those skilled in the art to which the present disclosure pertains. All of these publications are herein incorporated by reference to the same extent as if each individual publication was specifically and individually indicated to be incorporated by reference.

[0195] Although the present disclosure has been described with reference to specific embodiments, it should be understood that these embodiments are merely illustrations of the principles and applications of the present disclosure. Therefore, it should be understood that various modifications may be made to these illustrative embodiments and other arrangements may be designed without departing from the spirit and scope of the present disclosure as defined by the appended claims.

Claims

1. A nucleic acid construct comprising: a first nucleic acid comprising a first promoter operably linked to a nucleic acid encoding CXC motif chemokine receptor 3 (CXCR3); and A second nucleic acid encoding a chimeric antigen receptor (CAR), wherein the CAR comprises a ligand binding domain comprising a single-chain antibody fragment that binds an antigen on a tumor cell, a transmembrane domain, and an intracellular domain comprising a signaling domain.

2. The nucleic acid construct of claim 1, wherein the second nucleic acid is operably linked to a second promoter, which may be the same as or different from the first promoter; or A third nucleic acid encoding a self-cleaving peptide is arranged between the first nucleic acid and the second nucleic acid, and the first promoter drives the expression of the first nucleic acid, the second nucleic acid and the third nucleic acid. The nucleic acid construct according to claim 1 , wherein the first promoter is configured to overexpress the CXCR3.

4. The nucleic acid construct according to claim 1, wherein the antigen on the tumor cell is a malignant pleural mesothelioma (MPM) antigen.

5. The nucleic acid construct of claim 4, wherein the MPM antigen is mesothelin. The nucleic acid construct according to claim 5 , wherein the ligand binding domain is derived from a portion of the anti-mesothelin ScFvYP218 antibody.

7. The nucleic acid construct of claim 1, wherein the signaling domain comprises a primary signaling domain, a co-stimulatory signaling domain, or both a primary signaling domain and a co-stimulatory signaling domain.

8. The nucleic acid construct of claim 7, wherein the signaling domain comprises a CD3ζ primary signaling domain and a 4-1BB co-stimulatory signaling domain, or a CD28 co-stimulatory signaling domain, or both a 4-1BB co-stimulatory signaling domain and a CD28 co-stimulatory signaling domain.

9. A vector comprising the nucleic acid construct according to claim 1.

10. The vector according to claim 9, wherein the vector is a lentiviral vector.

11. The vector according to claim 10, wherein the lentiviral vector is a baboon envelope pseudotyped lentiviral vector.

12. A genetically engineered immune cell comprising one or more vectors comprising the nucleic acid construct of claim 1.

13. The genetically engineered immune cell of claim 12, wherein the genetically engineered immune cell is a natural killer (NK) cell, a T cell, or a combination thereof.

14. A pharmaceutical composition comprising an effective amount of the genetically modified immune cell according to claim 12 and a pharmaceutically acceptable carrier.

15. A method of treating cancer in a subject, comprising: An effective amount of the pharmaceutical composition according to claim 14 is administered to a subject in need thereof.

16. The method of claim 15, further comprising administering to the subject an effective amount of a STING agonist prior to, substantially simultaneously with, or after administration of the pharmaceutical composition.

17. The method of claim 16, wherein the STING agonist comprises ADU-S100, TAK-676, BI-STING, BMS-986301, GSK532, JNJ-4412, MK-1454, SB11285, 3'3'-scylic AIMP, ALG-031048, E7766, JNJ-'6196, MK-2118, MSA-1, MSA-2, SNX281m SR-717, KAT676, TTI-10001, XMT-2056, CRD-5500, or a combination thereof.

18. The method of claim 17, wherein the STING agonist comprises ADU-S100 or TAK-676.

19. The method of claim 16, wherein the STING agonist is delivered by intratumoral or intravenous injection.

20. The method of claim 15, wherein the cancer is a solid tumor.

21. The method of claim 20, wherein the cancer is MPM, melanoma, gastric cancer, liver cancer, lung cancer, bladder cancer, colorectal cancer, or breast cancer.

22. The method of claim 21, wherein the cancer is MPM.

Images