Anti-CSPG4 chimeric antigen receptor-expressing phagocytes

Abstract

Disclosed herein are modified phagocytes comprising chimeric antigen receptors (CARs). Also disclosed herein are methods of treating cancer using the modified phagocytes and methods of making the modified phagocytes.

Description

ANTI-CSPG4 CHIMERIC ANTIGEN RECEPTOR-EXPRESSING PHAGOCYTESCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 599,974, filed November 16, 2023, and U.S. Provisional Patent Application No. 63 / 640,095, filed April 29, 2024, each of which is incorporated herein by reference in its entirety.REFERENCE TO SEQUENCE LISTING

[0002] This application was filed with a Sequence Listing XML in ST.26 XML format accordance with 37 C.F.R. § 1.831 and PCT Rule 13ter. The Sequence Listing XML file submitted in the USPTO Patent Center, “026389-0018-W001_sequence_listing_xml_14-NOV- 2024. xml,” was created on November 14, 2024, contains 18 sequences, has a file size of 32.0 kilobytes (32,768 bytes), and is incorporated by reference in its entirety into the specification.BACKGROUND

[0003] Cutaneous melanoma is a form of skin cancer that arises from melanocytes. According to the latest Surveillance, Epidemiology, and End Results (SEER) data, for patients who are diagnosed with distant metastatic melanoma, the 5-year survival is below 40%. In the last 10 years, immune checkpoint blockade (ICB) approaches have drastically improved the survival of patients with metastatic melanoma. However, despite this success, approximately 50% of patients with metastatic melanoma do not benefit from ICB, even with combinations of αCTLA-4 and αPD-1. Patients can exhibit both primary resistance and acquire resistance during treatment. In addition, ICB approaches have been used for glioblastoma, a fast-growing and aggressive brain tumor that originates in the brain or spinal cord. However, glioblastoma is often resistant to ICB therapies, and it is difficult for ICB therapies to cross the blood-brain barrier (BBB). Thus, it remains essential to explore additional treatment options, particularly for patients who do not respond to current treatment strategies.

[0004] Chimeric antigen receptor (CAR) T-cell therapy has demonstrated the groundbreaking potential to harness the immune system against cancer. While this revolutionary approach has reshaped the treatment of hematological malignancies, its promise has yet to be fully realized in the context of solid tumors. Several complexities specific to solid tumor biology present unique obstacles that have limited the success of CAR-T cell therapies. Tumor heterogeneity presents a formidable challenge for CAR-T cell therapy. Antigen heterogeneity in solid tumors means that CAR-T cells designed to target a single antigen may only eliminate cells with high antigenexpression, leaving behind populations with low or no expression, resulting in therapeutic resistance and tumor relapse. Solid tumors also have dense physical barriers of extracellular matrix that impede active T-cell infiltration and cultivate a highly immunosuppressive microenvironment. This environment is characterized by the presence of suppressive immune cells (such as regulatory T cells, myeloid-derived suppressor cells, and tumor-associated macrophages), inhibitory cytokines (such as TGF-beta), hypoxia, and metabolic factors that can disarm and exhaust CAR-T cells, ultimately limiting their anti-tumor activity. While efforts to further engineer CAR-T cells for improved activity against solid tumors are ongoing, an alternative approach is engineering other immune cells, like macrophages to overcome these challenges.

[0005] Unlike T-cells, macrophages comprise a large portion of the tumor mass in solid tumors. Macrophages are professional phagocytes, eating their targets and easily infiltrating barriers surrounding solid tumors. Macrophages can also reprogram other immune cells such as regulating the polarization of other macrophages and educating adaptive immune cells through antigen presentation. Given that macrophages phagocytose their tumor targets, the antigen presentation capabilities of macrophages suggest that macrophages will present multiple tumor antigens to T-cells, thus potentially bypassing challenges associated with tumor heterogeneity. The central role that macrophages play in the immune system suggests that macrophages are an ideal cell to use for adoptive cell therapies in cancer, as reprogramming the tumor immune response is likely to lead to a sustained and robust anti-tumor response.

[0006] Several groups have explored the potential of using chimeric antigen receptorexpressing macrophages (CAR-Ms) to treat various types of solid tumors such as ovarian cancer. These CAR-Ms have predominantly targeted a variety of canonical tumor antigens including, but not limited to, CD19, CD47, HER2, and EGFRvlll. These studies show that CAR-Ms targeting certain cancer types are specific and efficient, and CAR-Ms effectively reduce tumor burden in in vivo xenograft models. However, it is still unclear whether CAR-Ms will be an effective strategy to treat all solid tumors, and whether the ideal cancer antigen has been selected.

[0007] Thus, there is a need for identification of an ideal cancer antigen and development of a CAR-M against the cancer antigen for treatment of solid tumor cancer such as melanoma and glioblastoma.SUMMARY

[0008] In an aspect, the disclosure relates to a modified phagocyte comprising a chimeric antigen receptor (CAR), wherein the CAR comprises one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and anendodomain. In an embodiment, the endodomain comprises a phagocytic receptor or a fragment thereof. In another embodiment, the phagocytic receptor is multiple EGF like domains 10 (MegflO), high affinity immunoglobulin epsilon receptor subunit gamma (FcRy), or MER protooncogene tyrosine kinase (MerTK). In another embodiment, the CAR does not comprise an endodomain domain comprising a phagocytic receptor or a fragment thereof. In another embodiment, the endodomain comprises a His, Myc, luciferase, or fluorescent tag. In another embodiment, the ectodomain further comprises a spacer, a co-stimulation domain, a secreted antibody domain, a cytokine secretion domain, or a combination thereof. In another embodiment, the spacer comprises from about 90 nucleic acids to about 150 nucleic acids or from about 30 amino acids to about 50 amino acids. In another embodiment, the transmembrane domain is a CD8 transmembrane domain or a FcRy transmembrane domain. In another embodiment, the CSPG4 binding domain is encoded by SEQ ID NO: 1. In another embodiment, the transmembrane domain is encoded by SEQ ID NO: 3 or SEQ ID NO: 7. In another embodiment, the endodomain is encoded by SEQ ID NO: 9. In another embodiment, the CAR is encoded by a plasmid having SEQ ID NO: 13.

[0009] In a further aspect, the disclosure relates a to pharmaceutical composition comprising a modified phagocyte comprising a chimeric antigen receptor (CAR), wherein the CAR comprises one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain. In an embodiment, the composition further comprises an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof. In another embodiment, the composition further comprises a pharmaceutically acceptable carrier.

[0010] Another aspect of the disclosure provides a method of treating cancer in a subject in need thereof comprising administering to the subject a therapeutically effective amount of the pharmaceutical composition described herein. In an embodiment, the therapeutically effective amount of the pharmaceutical composition comprises from about 1 million to about 10 million of the modified phagocytes. In another embodiment, the method further comprises administering a therapeutically effective amount of an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof; or the composition further comprises an anti-CD47 antibody, an inhibitor of SIRPα, an anti-SIRPα antibody, or a combination thereof. In another embodiment, the therapeutically effective amount of the anti-CD47 antibody is from about 1 mg / kg to about 45 mg / kg. In another embodiment, the anti-CD47 antibody is coadministered with the pharmaceutical composition, administered before the pharmaceutical composition, administered after the pharmaceutical composition, or a combination thereof. Inanother embodiment, the pharmaceutical composition is administered to the subject locally, systemically, or a combination thereof. In another embodiment, the pharmaceutical composition is administered locally in the brain at time of tumor resection, intravenously, or intrathecally. In another embodiment, the anti-CD47 antibody, the inhibitor of SIRPα, an anti-SIRPα antibody, or a combination thereof is administered locally in the brain at time of tumor resection, intravenously, or intrathecally. In another embodiment, the cancer is a solid tumor. In another embodiment, the cancer is glioblastoma, melanoma, breast cancer, or any other CSPG4-expressing solid tumor.

[0011] Another aspect of the disclosure provides a method of modifying a phagocyte comprising: delivering to the phagocyte a plasmid comprising one or more nucleic acid sequences encoding one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain. In an embodiment, the plasmid is a viral vector. In another embodiment, the viral vector is a lentiviral vector or an adeno-associated viruses (AAV). In another embodiment, the plasmid has SEQ ID NO: 13.

[0012] The disclosure provides for other aspects and embodiments that will be apparent in light of the following detailed description and accompanying figures.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] FIGS. 1A-I show that CSPG4-targeting CAR-Ms phagocytose CSPG4-expressing melanoma cells in 2D cell culture. FIG. 1A is a graph showing CSPG4 transcript levels by singlecell RNA sequencing analysis of healthy (non-cycling) tissue vs. melanoma tumors, rank mean normalized. FIG. 1 B is a diagram showing a plasmid for a chimeric antigen receptor macrophage construct. FIG. 1C is a schematic showing chimeric antigen receptor macrophage constructs. FIG. 1 D is a schematic showing a method of generation of primary human CAR-Ms. FIG. 1 E is representative histograms showing phagocytic events after 24-hour coculture for CAR-MGFPand CSPG4aCSPG4<763 74\ FIG. 1 F is representative flow cytometry contour plots showing CAR-M- mediated phagocytosis for CAR-MGFPcontrol and CSPG4aCSPG4<763-74>. FIG. 1 G is a graph showing quantification of CAR-M-mediated phagocytosis from FIG. 1 D and FIG. 1 E for CAR-MGFPand CAR-M with 3 scFvs (225.28, SK5, 763.74). FIG. 1 H is representative single z-plane images of CAR-MαCSPG4(763 74)(green)-mediated phagocytosis of A375-Lck-mScarlet cells (magenta) after 24 hours of 2D coculture by microscopy. White dashed line outlines CAR-MαCSPG4cell boundary, yellow arrows indicate CAR-M phagocytosis. Scale bar is 20 microns. FIG. 11 is a graph showing quantification of CSPG4αCSPG4(763.74)phagocytosis compared to CAR-MGFPand CAR- MαcspG4(763.74)ΔFcRyflow cytometry. For all graphs, each dot on the graph is a biological replicate,each as a shade of gray. For FIG. 1 F and FIG. 1 H, mean ± SEM, 1-way ANOVA with Tukey’s multiple comparisons test.

[0014] FIGS. 2A-E show that CSPG4 expression is overexpressed and specific to melanoma. FIG. 2A is a graph showing a fraction of cells expressing CSPG4 in single-cell RNA sequencing analysis of healthy (non-cycling, gray) tissue vs. melanoma tumors (black) rank mean normalized from FIG. 1A. FIG. 2B is a violin plot of CSPG4 transcript levels by single-cell RNA sequencing analysis of malignant cells (arrow) in tumors compared to non-malignant cells within the tumor. Quartiles are demarcated on the violin plots. FIG. 2C is a graph showing CSPG4 transcript counts in normal cell types (Human Protein Atlas). FIG. 2D is a box plot of CSPG4 expression from the Melanoma-Riaz study on the Immuno-genomic atlas for immune checkpoint blockade-based cancer therapy by RECIST status (PD = Progressive disease, SD = stable disease, PR = partial response, CR = complete response) (1-way ANOVA with Tukey’s multiple comparisons test). FIG. 2E is a box plot of CSPG4 expression from the Melanoma-Riaz study on the Immuno- genomic atlas for immune checkpoint blockade-based cancer therapy by pre-treatment vs. post- aPD-1 / L1 treatment (unpaired t-test, non-significant comparisons are not indicated on the graph).

[0015] FIGS. 3A-F show a phagocytosis gating strategy and CAR-M transduction efficiencies. FIG. 3A is a schematic of CSPG4 scFvs showing predicted (SK5) and validated (225.28, 763.74) domain binding. FIG. 3B is a violin plot of primary macrophage transduction efficiency for each CAR-M construct (N: GFP = 12, 763.74 = 12, SK5 = 4, 225.28 = 3). FIG. 3C is representative histograms showing CSPG4 surface expression by flow cytometry of A375 cells compared to primary macrophages (n = 3 donors as MF #1-3). FIG. 3D is representative images of a gating strategy for quantifying CAR-M transduction and CAR-M-mediated phagocytosis of A375-Lck- mScarlet cells based on 0-hour coculture. FIG. 3E is a graph showing a comparison of CAR-MαCSPG4(763.74)and CAR-MGFPphagocytosis and transduction efficiency from flow cytometry experiments with line of best fit (slope - CAR-MGFP: 0.086 and CAR-MαCSPG4: -0.13) and correlation. FIG. 3F is representative single z-plane images of CAR-MαCSPG4(SK5)or CAR-MαCSPG4(225.28)(green) and A375-Lck-mScarlet cells (magenta). The white dashed line outlines CAR-MαCSPG4cell boundary, and the yellow arrows highlight engulfment events. Images were acquired with a 63X objective on an LSM 880 microscope. Scale bar is 10 microns.

[0016] FIGS. 4A-E show that melanoma particles are fully internalized within CSPG4- targeting CAR-Ms. FIG. 4A is representative X-Z plane images of engulfed Lck-mScarlet fragments inside of CAR-MαCSPG4cells shown in FIG 1G. Images were acquired with a 63X objective on an LSM 880 microscope. Scale bars on both x- and z- axes represent 5 microns. FIG. 4B is representative still images from timelapse recording of CAR-MaCSPG4(green)engulfment of Lck-mScarlet fragments (magenta). Yellow arrows highlight engulfment events. Scale bar is 50 microns. FIG. 4C is representative images of A375-Lck-mScarlet cells cocultured with CAR-MαCSPG4or CAR-MGFPfor 24 hours, images were taken at 20*, yellow arrows indicate CAR-M phagocytosis, scale bar is 40 microns. FIG. 4D is a violin plot, with quartiles, of quantification of Lck-mScarlet signal overlapping with GFP+ CAR-Ms in FIG. 4C from a maximum intensity projection (n = 3 biological replicates). Mean ± SEM, 1-way ANOVA with Tukey’s multiple comparisons test. FIG. 4E is a violin plot, with quartiles, of surface area measured for of internalized Lck-mScarlet fragments in CAR-MαCSPG4(N = 34 engulfments, median = 64.69 microns2) or unengulfed A375- Lck-mScarlet cells (N = 28 cells, median = 473.19 microns2).

[0017] FIGS. 5A-F show that CSPG4-targeting CAR-M phagocytosis is specific to CSPG4- expressing cells. FIG. 5A is a representative flow plot of CSPG4 surface expression on A375, 624-mel, and WM793 cells with flow cytometry. FIG. 5B is a graph showing quantification of CSPG4 surface expression in FIG. 5A normalized to A375 expression. FIG. 5C is a graph showing quantification of CAR-M-mediated phagocytosis (CAR-MGFPand CAR-MαCSPG4) of A375, 624-mel, and WM793 cells after 24 hours of coculture. Mean ± SEM, 2-way ANOVA with Sidak’s multiple comparisons test. FIG. 5D is a representative flow plot of CSPG4 surface expression on A375 cells expressing non-targeted shRNA or 2 different shCSPG4 RNAs. FIG. 5E is a graph showing quantification of CSPG4 surface expression in FIG. 5D normalized to shNT expression. FIG. 5F is a graph showing quantification of CAR-M-mediated phagocytosis (CAR-MGFPand CAR- MαCsPG4) of A375 cells treated with non-targeting shRNAs or 2 different shCSPG4 RNAs. For all graphs, each dot on the graph is a biological replicate, each as a shade of gray. Mean ± SEM, 2-way ANOVA with Tukey’s multiple comparisons test. Non-significant comparisons are not indicated on the graph.

[0018] FIGS. 6A-F show that CSPG4-targeting CAR-Ms primarily trogocytose target melanoma cells. FIG. 6A is a schematic of CAR-M coculture with melanoma cells and analysis of bite size by image-based flow cytometry (Imagestream®, Cytek® Amnis®). FIG. 6B is an image of representative gating of the size of internalized A375-Lck-mScarlet fragments in CAR-Ms, images were acquired with a 63X objective lens on the Imagestream®. FIG. 6C is a graph showing quantification of data in FIG. 6B - small (< 75 μm2) versus large (> 75 μm2) A375-Lck-mScarlet fragments inside CAR-Ms. n = 1 biological donor. FIG. 6D is representative images of “small” and “large” A375-Lck-mScarlet fragments (red) inside CAR-Ms (green). FIG. 6E is a graph showing quantification of percent CAR-M phagocytosis of A375-H2B-mCherry cells. Mean ± SEM, 1-way ANOVA with Tukey’s multiple comparison test. Non-significant comparisons are not indicated on the graph. FIG. 6F is a graph showing quantification of proportion of live, A375-H2B-mCherry cells remaining in the population by flow cytometry after 24 hours of coculture with CAR- MOCSPG4 compared to CAR-MGFP. Each dot on the graph is a biological replicate, each as a shade of gray. Mean ± SEM, 1-way ANOVA with Tukey’s multiple comparison test Non-significant comparisons are not indicated on the graph.

[0019] FIGS. 7A-G show Imagestream® gating strategy for phagocytic events. Imagestream® gating strategy for identification of internalized Lck-mScarlet+ melanoma cell fragments in GFP+ CAR-M cells (data in FIGS. 6A-F). FIG. 7A is an image of representative gating of the aspect ratio of all events. FIG. 7B is an image of representative gating of normalized frequency of cells. FIG. 7C is an image of representative gating of DARI in focus area. FIG. 7D is an image of representative gating of RFP and GFP stained cells live. FIG. 7E is an image of representative gating of normalized frequency of total GFP-positive (GFP+) cells. FIG. 7F is an image of representative gating of area intensity of RFP of non-saturated GFP cells. FIG. 7G is an image of representative gating of RFP minimum intensity.

[0020] FIGS. 8A-E show that combining CSPG4-targeting CAR-Ms with aCD47 leads to melanoma cell death in 3D. FIG. 8A is a schematic of CAR-M and melanoma cells co-forming 3D spheroids with and without aCD47. FIG. 8B is a graph showing quantification of percent CAR- M phagocytosis of melanoma cells in spheroids after 72 hours. FIG. 8C is a graph showing the proportion of remaining melanoma cells in spheroids after 72 hours by flow cytometry. FIG. 8D is representative images of replated spheroids showing “unengulfed” melanoma cells (single spherical nuclei, colocalization with H2B-mCherry, no engulfment by CAR-M); “engulfed + live” (melanoma engulfment by CAR-M, A375-H2B-mCherry nuclei intact and colocalizing with DAPI); or “engulfed + dead” (melanoma engulfment by CAR-M, A375-H2B-mCherry does not colocalize with DAPI). White dashed outline marks individual CAR-Ms. Green arrows indicate CAR-M nuclei. Magenta arrows indicate the A375-H2B-mCherry signal colocalizing with DAPI inside CAR-Ms. The asterisk indicates dispersed A375-H2B-mCherry signal not colocalizing with DAPI inside CAR-Ms. Scale bar indicates 10 microns. FIG. 8E is a graph showing quantification of ruptured nuclei as percent of total melanoma cells after 72 hours of coculture. CAR-MGFP / lgG(N = 104), CAR-MGFP / aCD47(N = 70), CAR-MαCSPG4AFcRY / lgG(N = 1 10), CAR-MαCSPG4AFcRY / aCD47(N = 47), CAR-MαCSPG4 / lgG(N = 65), CAR-MαCSPG4 / aCD47(N = 34). For FIGS. 8B-C and FIG. 8E, n = 3 biological replicates; each shade of gray indicates one biological replicate, with each dot indicating technical replicates. Mean ± SEM, 2-way ANOVA with Tukey’s multiple comparisons test. Nonsignificant comparisons are not indicated on the graphs.

[0021] FIGS. 9A-I show that aCD47 treatment increases CSPG4-CAR-M-mediated melanoma cell death in 3D, but not in 2D. FIG. 9A is a graph showing Imagestream®quantification of fully internalized A375 Lck-mScarlet puncta in GFP+ CAR-M cells in 3D spheroids after 24 hours; images were acquired with a 40X objective lens on an Imagestream®. FIG. 9B is a graph showing classification of melanoma cells (live or dead) after 3 days of coculture with CAR-Ms as co-formed spheroids from FIG. 8D. FIG. 9C is a graph showing quantification of unengulfed nuclei as percent of total melanoma cells after 72 hours of coculture in FIG. 9B. FIG. 9D is a graph showing quantification of engulfed nuclei as percent of total melanoma cells after 72 hours of coculture in FIG. 9B. FIG. 9E is representative flow cytometry contour plots of CAR- M phagocytosis of A375-Lck-mScarlet cells treated with IgG isotype control or 10 μg / mL aCD47 for 24 hours in 2D culture. FIG. 9F is a normalized to mode histogram of Lck-mScarlet fluorescence in CAR-Ms at 0 and 24 hours in CAR-MaCSPG4 with and without aCD47 in 2D culture. FIG. 9G is a graph showing quantification of CAR-M phagocytosis of A375-Lck-mScarlet cells treated with and without aCD47 in FIG. 9E and FIG. 9F. FIG. 9H is a graph showing quantification of CAR-Ms phagocytosis of A375-H2B-mCherry cells treated with IgG isotype control or 10 μg / mL aCD47 in 2D culture. FIG. 9I is a graph showing the proportion of target A375-H2B-mCherry cells remaining after 24 hours of coculture with CAR-Ms from FIG. 9H. Each dot is 1 biological replicate, as shades of gray. For FIGS. 9C, D, G, H, and I, mean ± SEM, 2- way AVOVA with Tukey’s multiple comparisons test. Non-significant comparisons are not indicated on the graphs.

[0022] FIGS. 10A-E show that combining CSPG4-targeting CAR-Ms with aCD47 inhibits melanoma spheroid growth in 3D. FIG. 10A is a schematic of pre-formed A375 melanoma spheroid, ± aCD47, and CAR-Ms for short-term adherence assay and long-term imaging. FIG. 10B is representative images of pre-formed A375-Lck-mScarlet spheroids (magenta) cultured with CAR-Ms (green) (following schematic in FIG. 10A) for 4-8 hours with aCD47 or IgG controls and embedded in Matrigel for imaging. Scale bar is 100 microns. FIG. 10C is a graph showing quantification of GFP+ CAR-Ms adhered to mScarlet+ melanoma spheroids in FIG. 10B. FIG. 10D is a representative plot of A375-H2B-mCherry spheroid growth in culture with CSPG4- targeting CAR-Ms or control CAR-Ms over 10 days, as measured by the total integrated intensity of mCherry signal over time. FIG. 10E is a graph showing quantification of data in FIG. 10D at day 10 across conditions (N = 17 spheroids across N = 3 biological replicates). Data is shown as Iog2(fold change from time zero). For graphs, each shade of gray indicates one biological replicate, with each dot indicating technical replicates. For FIG. 10C and FIG. 10E, mean ± SEM, 2-way ANOVA with Tukey’s multiple comparisons test. Non-significant comparisons are not indicated on the graph.

[0023] FIGS. 11A-B show that CSPG4 CAR-Ms infiltrate melanoma spheroids and phagocytose melanoma cells. FIG. 11A is representative still images with orthogonal x-z, and y- z slices of CAR-MαCSPG4(green) inside A375-Lck-mScarlet spheroids (magenta) after 72 hours of coculture. The spheroid boundary is marked by white dashed line. Scale bar is 100 microns. Images were acquired with a 10X objective. FIG. 11 B is representative images of maximum intensity projection of 3 slices indicating CAR-M (green) engulfment of mScarlet+ fragments. Two areas of the spheroid (boxes) are magnified to the right as Area 1 and Area 2, with CAR-Ms outlined in white hatched lines. Scale bar of main image is 100 microns; scale bar of inset is 10 microns.

[0024] FIGS. 12A-G show that combining CSPG4-targeting CAR-Ms with aCD47 approaches leads to whole-cell melanoma phagocytosis in 3D. FIG. 12A is representative images of pre- formed A375-H2B-mCherry spheroids (gray) at day 0 (before CAR-Ms), 8 hours after CAR-M addition, and 10 days after CAR-M addition. Scale bar is 200 microns. FIG. 12B is a graph showing quantification of A375 spheroid growth as measured by mCherry area at day 10 across conditions (N = 17 spheroids across N = 3 biological replicates). Data shown as log2(fold change from time zero). Mean ± SEM, 2-way ANOVA with Dunnett’s multiple comparisons test. FIG. 12C is representative images of pre-formed A375-H2B-mCherry spheroids (magenta) and CAR- Ms (green) on day 10; white line indicates line scan measurements in FIG. 12D. Scale bar is 200 microns. FIG. 12D is a graph showing line scan signal intensity from representative spheroid images in FIG. 12C. FIG. 12E is a graph showing quantification of mScarlet integrated density in A375-Lck-mScarlet tumor spheroids after 10 days. A375-Lck-mScarlet cells were either treated with shNT or shCSPG4 RNA. FIG. 12F is representative images of shNT and shCSPG4 #2 clones at day 0, day 3 (first media change), and day 10. Scale bar is 200 microns. FIG. 12G is a graph showing flow cytometry CAR-M phagocytosis quantification of A375-H2B-mCherry cells after 72 hours of 3D coculture with decreasing concentrations of aCD47. Mean ± SEM, 2-way ANOVA with Tukey’s multiple comparisons test. N = 4 biological replicates. Non-significant comparisons are not indicated on the graphs, with the exception of on FIG. 12G.

[0025] FIGS. 13A-C show that CSPG4-targeting CAR-Ms using the 225.28 scFv do not efficiently phagocytose murine melanoma cells. FIG. 13A is representative flow cytometry histograms of CSPG4 surface expression across a panel of murine melanoma cell lines. FIG. 13B is a diagram showing UniProt alignment of human vs. mouse CSPG4 amino acid sequences for epitopes identified for 763.74 and 225.28 scFvs. FIG. 13C is a graph showing quantification of cSPG4αCSPG4(22528)phagocytosis of YUMM1.7 murine melanoma cells compared to CAR-MGFP and CAR-MαCSPG4(763 74)by flow cytometry. 1-way ANOVA with Tukey’s multiple comparisons test, non-significant comparisons are not indicated on the graph.

[0026] FIGS. 14A-G show that aCSPG4 CAR-Ms inhibit melanoma growth in vivo. FIG. 14A is a diagram showing an exemplary experimental outline showing that A375-H2B-mCherry cells were injected in NRG mice, and then injected with CAR-MGFPafter 9 days. Tumors were dissociated and processed for flow cytometry on indicated days to determine presence of CAR- Ms. FIG. 14B is a graph showing quantification of the presence of CAR-MGFPin tumors over time in FIG. 14A. FIG. 14C is a diagram showing an exemplary experimental outline showing A375- H2B-mCherry same-day injections with CAR-MαCSPG4or CAR-MGFPin NRG mice (N = 4 mice per group). FIG. 14D is a graph showing tumor volume for FIG. 14C plotted over time. Arrows indicate CAR-M injections. FIG. 14E is a diagram showing an exemplary experimental outline showing CAR-MαCSPG4or CAR-MGFPinjections after A375-H2B-mCherry engraftment in NRG mice (N = 6 mice per group). FIG. 14F is flow cytometry plots showing percent bone marrow-derived macrophage transduction for CAR-MGFPand CAR-MαCSPG4. FIG. 14G is a graph showing tumor volume for FIG. 14E and FIG. 14F plotted over time. Arrows indicate CAR-M injections. For FIG. 14D and FIG. 14G, 1-way ANOVA with Tukey’s multiple comparisons test at terminal time point. *p < 0.05 versus indicated group.

[0027] FIGS. 15A-G show that CSPG4-targeting CAR-Ms do not affect mouse weight and do not persist in tumors 5 days after last peritumoral injection. FIG. 15A is a graph showing animal weights from FIG. 14D. FIG. 15B is a graph showing individual mouse tumor volumes from FIG. 14D. FIG. 15C is a graph showing grouped mouse tumor volumes including vehicle from FIG. 14D. FIG. 15D is a graph showing animal weights from FIG. 14G. FIG. 15E is a graph showing individual mouse tumor volumes from FIG. 14G. FIG. 15F is a graph showing grouped mouse tumor volumes including vehicle from FIG. 14G. FIG. 15G is representative 40X images of immunofluorescent staining for GFP and DAPI of tumors from FIG. 14G. As a positive control for GFP staining, a tumor isolated from a GFP+ mouse was used. For FIG. 15C and FIG. 15F, mean ± SEM. Arrows indicate CAR-M injections.

[0028] FIGS. 16A-E show that peri-tumoral injections of CSPG4-targeting CAR-Ms caused decreased melanoma tumor growth in vivo using both immune-compromised and immune- competent mice. FIG. 16A is a representation of CSPG4 surface expression (dark gray) on murine YUMM1.7 cells by flow cytometry. ISO = isotype control (light gray). FIG. 16B is a representative flow cytometry contour plot showing transduction efficiency of GFP+ CARs in murine bone marrow derived macrophages. FIG. 16C is a diagram showing an exemplary experimental outline. FIG. 16D is a graph showing tumor immune profiling in tumors from FIG.16E at day 10. Mean ± SEM, 2-way ANOVA. FIG. 16E is graph showing murine melanoma YUMM1.7 tumor growth in C57BL6 / J mice treated with murine CAR-Ms. n = 4 mice per condition.

[0029] FIGS. 17A-B show that concentration of anti-CD47 can be significantly decreased in vivo. FIG. 17A is a graph showing percent (%) phagocytosis of CSPG4-targeting CAR-Ms with decreasing concentrations of anti-CD47 (Hu5F9-G4) compared to control GFP-CAR-Ms in 3D spheroids, n = 5 biological replicates, mean ± SEM. 2-way ANOVA. Only significance is displayed. FIG. 17B is a graph showing quantification of A375M melanoma growth in NRG mice upon CAR-M treatment with and without 5mg / kg aCD47. Arrow points to tumors treated with aCD47. mean ± SEM. n = 8 mice per condition.

[0030] FIG. 18 is a graph showing YUMM1.7 tumor growth in C57BL6 / J mice after tumor resection and leaving a 2 mm x 2 mm tumor piece behind (arrow), n = 7 mice.

[0031] FIGS. 19A-F show that NG2-CAR-MS adhere to malignant glioma spheroids. FIG. 19A is a representative histogram demonstrating NG2 binding compared to isotype control. FIG. 19B is a graph showing quantification of surface NG2 expression across cell lines, n = 3 technical replicates. FIG. 19C is a representative histogram demonstrating anti-NG2 ScFv binding compared to isotype control. FIG. 19D is a graph showing quantification of surface anti-NG2 scFv across cell lines, n = 3 technical replicates. FIG. 19E is a schematic of an exemplary adherence assay and representative images that show CAR-Ms (green) adhering to spheroids (magenta). FIG. 19F is a graph showing quantification of adherent macrophages, n = 4 independent PBMC donors, non-parametric t test.

[0032] FIGS. 20A-F show that NG2-CAR-Ms infiltrate and phagocytose NG2-expressing glioblastoma cells in 3D cell culture. FIG. 20A is representative images of CAR-M infiltration (green) in U87 spheroids (magenta). FIG. 20B is a graph showing quantification of macrophage infiltration, n = 3 technical replicates; nonparametric t test. FIG. 20C is representative flow graph of eating events: CD45+ GFP+ macrophages that internalized RFP+ tumor cells. FIG. 20D is a graph showing quantification of eating events, n = 3 independent PBMC donors; nonparametric t test. FIG. 20E is a graph showing representative in vitro tumor spheroid growth over time, mean ± SEM, n = 10 technical replicates. FIG. 20F is a graph showing quantification of tumor spheroid area at 232 hours of co-culture. n = 3 independent PBMC donors; nonparametric t test.

[0033] FIGS. 21 A-E show that NG2-CAR-MS are successfully injected intracranially and stay localized at the injection site where the tumor is located. FIG. 21A is a schematic showing an exemplary intracranial injection strategy for tumor growth study and representative bioluminescence images. FIG. 21 B is a graph showing quantification of tumor volume over time as measured by normalized total flux (n = 3). Data was analyzed using nonparametric t testbetween experimental groups at final endpoint FIG. 21 C is a schematic showing an exemplary injection strategy for a survival study. FIG. 21 D is a graph of a Kaplan Meier curve showing survival probability over time (n = 8 for NG2, Control, Vehicle). FIG. 21 E is a graph showing localization of control CAR-Ms compared to NG2-CAR-Ms after total brain dissociation (n = 3).DETAILED DESCRIPTION

[0034] Described herein are modified macrophages comprising a chimeric antigen receptor (CAR), wherein the CAR may comprise one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain. Herein are CAR-Ms that recognize and phagocytose melanoma and glioblastoma cells. Numerous tumor-associated antigens enriched in melanomas have been identified, and CSPG4, also known as NG2 and HMW-MAA, was chosen. CSPG4 is an about 300 kDa transmembrane proteoglycan that regulates cancer cell migration, invasion, epithelial-mesenchymal transition, and proliferation. CSPG4 is frequently highly expressed in melanoma tumors and glioblastoma tumors, and expression in non-malignant cells is low.

[0035] Herein, it is demonstrated that CSPG4-targeting CAR-Ms phagocytose melanoma cells and glioblastoma cells in vitro and in vivo, and CSPG4-targeting CAR-Ms in combination with CD47 blocking antibodies efficiently inhibit melanoma spheroid growth in 3D.1. Definitions

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. The meaning and scope of the terms should be clear. In case of conflict, the present document, including definitions, take precedent over any dictionary or extrinsic definition. Further, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular. Preferred methods and materials are described below, although methods and materials similar or equivalent to those described herein can be used in practice or testing of the present invention. All publications, patent applications, patents and other references mentioned herein are incorporated by reference in their entirety. The materials, methods, and examples disclosed herein are illustrative only and not intended to be limiting.

[0037] The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms, or words that do not preclude the possibility of additional acts or structures. The singular forms “a,” “and,” and “the” include plural references unless the context clearly dictates otherwise. The presentdisclosure also contemplates other embodiments “comprising,” “consisting of,” and “consisting essentially of,” the embodiments or elements presented herein, whether explicitly set forth or not.

[0038] For the recitation of numeric ranges herein, each intervening number there between with the same degree of precision is explicitly contemplated. For example, for the range of 6-9, the numbers 7 and 8 are contemplated in addition to 6 and 9, and for the range 6.0-7.0, the number 6.0, 6.1 , 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9, and 7.0 are explicitly contemplated.

[0039] The term “about” or “approximately” as used herein as applied to one or more values of interest, refers to a value that is similar to a stated reference value, or within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, such as the limitations of the measurement system. In certain aspects, the term “about” refers to a range of values that fall within 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 than or less than) of the stated reference value unless otherwise stated or otherwise evident from the context (except where such number would exceed 100% of a possible value). Alternatively, “about” can mean within 3 or more than 3 standard deviations, per the practice in the art. Alternatively, such as with respect to biological systems or processes, the term “about” can mean within an order of magnitude, preferably within 5-fold, and more preferably within 2-fold, of a value.

[0040] “Activation,” as used herein, refers to the state of a cell, for example a monocyte, macrophage, or dendritic cell that has been sufficiently stimulated to induce detectable cellular proliferation or has been stimulated to exert its effector function. Activation can also be associated with induced cytokine production, cytokine secretion, phagocytosis, cell signaling (e.g., gene expression changes), target cell killing, metabolic changes, production of inflammatory mediators, proliferation, epigenetic reprogramming, phenotypic switching of macrophages (e.g., M1 polarization), suppression of pro-tumor or M2 macrophages, phenotypic switching of pro-tumor or M2 macrophages, and / or antigen processing and presentation.

[0041] “Activated monocytes / macrophages / dendritic cells” refers to, among other things, monocyte / macrophage / dendritic cells that are undergoing cell division or exerting effector function. The term “activated monocytes / macrophages / dendritic cells” refers to, among others thing, cells that are performing an effector function or exerting any activity not seen in the resting state, including phagocytosis, cytokine secretion, proliferation, gene expression changes, metabolic changes, production of inflammatory mediators, proliferation, epigenetic reprogramming, phenotypic switching of macrophages (e.g., M1 polarization), suppression of pro-tumor or M2 macrophages, phenotypic switching of pro-tumor or M2 macrophages, and other functions.

[0042] “Allogeneic” refers to any material derived from another subject of the same species. Allogeneic cells are genetically distinct and immunologically incompatible yet belong to the same species. Typically, “allogeneic” is used to define cells, such as CAR-comprising cells, that are transplanted from a donor to a recipient of the same species.

[0043] “Amino acid” as used herein refers to naturally occurring and non-natural synthetic amino acids, as well as amino acid analogs and amino acid mimetics that function in a manner similar to the naturally occurring amino acids. Naturally occurring amino acids are those encoded by the genetic code. Amino acids can be referred to herein by either their commonly known three- letter symbols or by the one-letter symbols recommended by the IUPAC-IUB Biochemical Nomenclature Commission. Amino acids include the side chain and polypeptide backbone portions.

[0044] As used herein, the term “antibody” refers to a polypeptide that includes canonical immunoglobulin sequence elements sufficient to confer specific binding to a particular target antigen. As is known in the art, intact antibodies as produced in nature are approximately 150 kD tetrameric agents comprising two identical heavy chain polypeptides (about 50 kD each) and two identical light chain polypeptides (about 25 kD each) that associate with each other into what is commonly referred to as a “Y-shaped” structure. Each heavy chain comprises at least four domains (each about 110 amino acids long) - an amino-terminal variable (VH) domain (located at the tips of the Y structure), followed by three constant domains: CH1, CH2, and the carboxy - terminal CH3 (located at the base of the Y’s stem). A short region, known as the “switch,” connects the heavy chain variable and constant regions. The “hinge” connects CH2 and CH3 domains to the rest of the antibody. Two disulfide bonds in this hinge region connect the two heavy chain polypeptides to one another in an intact antibody. Each light chain comprises two domains - an amino-terminal variable (VL) domain, followed by a carboxy- terminal constant (CL) domain, separated from one another by another “switch.” Intact antibody tetramers comprises two heavy chain-light chain dimers in which the heavy and light chains are linked to one another by a single disulfide bond; two other disulfide bonds connect the heavy chain hinge regions to one another, so that the dimers are connected to one another and a tetramer is formed. Naturally- produced antibodies are also glycosylated, typically on the CH2 domain. Each domain in a natural antibody has a structure characterized by an “immunoglobulin fold” formed from two beta sheets (e.g., 3-, 4-, or 5-stranded sheets) packed against each other in a compressed antiparallel beta barrel. Each variable domain contains three hypervariable loops known as “complementaritydetermining regions” (CDR1 , CDR2, and CDR3) and four somewhat invariant “framework” regions (FR1 , FR2, FR3, and FR4). When natural antibodies fold, the FR regions form the beta sheets that provide the structural framework for the domains, and the CDR loop regions from both the heavy and light chains are brought together in three-dimensional space so that they create a single hypervariable antigen binding site located at the tip of the Y structure. The Fc region of naturally occurring antibodies binds to elements of the complement system, and also to receptors on effector cells, including, for example, effector cells that mediate cytotoxicity. Affinity and / or other binding attributes of Fc regions for Fc receptors can be modulated through glycosylation or other modification. In some embodiments, any polypeptide or complex of polypeptides that includes sufficient immunoglobulin domain sequences as found in natural antibodies can be referred to and / or used as an “antibody,” whether such polypeptide is naturally produced (e.g., generated by an organism reacting to an antigen), or produced by recombinant engineering, chemical synthesis, or other artificial system or methodology. In some embodiments, an antibody is polyclonal. In some embodiments, an antibody is monoclonal. In some embodiments, an antibody has constant region sequences that are characteristic of mouse, rabbit, primate, or human antibodies. In some embodiments, antibody sequence elements are humanized, primatized, chimeric, etc, as is known in the art. Moreover, the term “antibody,” as used herein, can refer in appropriate embodiments (unless otherwise stated or clear from context) to any of the art-known or developed constructs or formats for utilizing antibody structural and functional features in alternative presentation. For example, in some embodiments, an antibody utilized in accordance with the present invention is in a format selected from, but not limited to, intact IgA, IgG, IgE or IgM antibodies; bi- or multi- specific antibodies; antibody fragments such as Fab fragments, Fab’ fragments, F(ab’)2 fragments, Fd’ fragments, Fd fragments, and isolated CDRs or sets thereof; single chain Fvs; polypeptide-Fc fusions; single domain antibodies; camelid antibodies; masked antibodies; Small Modular ImmunoPharmaceuticals (“SMIPs™”); single chain or Tandem diabodies (TandAb®); VHHs; Anticalins®; Nanobodies® minibodies; BiTE®s; ankyrin repeat proteins or DARPINs®; Avimers®; DARTs; TCR-like antibodies;, Adnectins®; Affilins®; Trans-bodies®; Affibodies®; TrimerX®; MicroProteins; Fynomers®, Centyrins®; and KALBITOR®s. An antibody may lack a covalent modification (e.g., attachment of a glycan) that it would have if produced naturally. An antibody may contain a covalent modification (e.g., attachment of a glycan, a payload (e.g., a detectable moiety, a therapeutic moiety, a catalytic moiety, etc.), or other pendant group (e.g., poly-ethylene glycol, etc.)).

[0045] As used herein, the term “antibody fragment” refers to a portion of an intact antibody and refers to the antigenic determining variable regions of an intact antibody. Examples ofantibody fragments include, but are not limited to, Fab, Fab', F(ab')2, and Fv fragments, linear antibodies, scFv antibodies, and multispecific antibodies formed from antibody fragments and human and humanized versions thereof.

[0046] As used herein, the term “antibody heavy chain” refers to the larger of the two types of polypeptide chains present in all antibody molecules in their naturally occurring conformations.

[0047] As used herein, the term “antibody light chain” refers to the smaller of the two types of polypeptide chains present in all antibody molecules in their naturally occurring conformations.

[0048] As used herein, the term “antigen” or “Ag” refers to a molecule that is capable of provoking an immune response. This immune response may involve either antibody production, the activation of specific immunologically-competent cells, or both. A skilled artisan will understand that any macromolecule, including virtually all proteins or peptides, can serve as an antigen. Furthermore, antigens can be derived from recombinant or genomic DNA.

[0049] As used herein, the term “anti-tumor effect” refers to a biological effect which can be manifested by a decrease in tumor volume, a decrease in the number of tumor cells, a decrease in the number of metastases, an increase in life expectancy, or amelioration of various physiological symptoms associated with the cancerous condition. An “anti-tumor effect” can also be manifested by the ability of the peptides, polynucleotides, cells, and antibodies described herein in prevention of the occurrence of a tumor in the first place.

[0050] “Autologous” refers to any material derived from a subject and re-introduced to the same subject.

[0051] “Binding region” as used herein refers to the region within a target region that is recognized and bound by a chimeric antigen receptor (CAR) described herein.

[0052] The terms “cancer,” “cancer cell,” “tumor,” and “tumor cell” are used interchangeably herein and refer generally to a group of diseases characterized by uncontrolled, abnormal growth of cells (e.g., a neoplasia). In some forms of cancer, the cancer cells can spread locally or through the bloodstream and lymphatic system to other parts of the body (“metastatic cancer”). In some forms of cancer, the cancer cells may be localized, such as a solid tumor. “Cancer” refers to all types of cancer or neoplasm or malignant tumors found in animals, including carcinoma, adenoma, melanoma, sarcoma, lymphoma, leukemia, blastoma, glioma, astrocytoma, mesothelioma, or a germ cell tumor. Cancer may include cancer of, for example, the colon, rectum, stomach, bladder, cervix, uterus, skin, epithelium, muscle, kidney, liver, lymph, bone, blood, ovary, prostate, lung, brain, head and neck, and / or breast. Cancer may include melanoma and / or glioblastoma.

[0053] The term “chimeric antigen receptor” or “CAR,” as used herein, refers to an artificial cell surface receptor that is engineered to be expressed on an immune effector cell and specifically targets a cell and / or binds an antigen. CARs may be used, for example, as a therapy with adoptive cell transfer. For example, monocytes, macrophages, and / or dendritic cells are removed from a patient (e.g., from blood, tumor or ascites fluid) and modified so that they express a receptor specific to a particular form of antigen.

[0054] “Coding sequence” or “encoding nucleic acid” as used herein means the nucleic acids (RNA or DNA molecule) that comprise a nucleotide sequence which encodes a protein. The coding sequence can further include initiation and termination signals operably linked to regulatory elements including a promoter and polyadenylation signal capable of directing expression in the cells of an individual or mammal to which the nucleic acid is administered. The coding sequence may be codon optimized.

[0055] “Complement” or “complementary” as used herein means a nucleic acid can mean Watson-Crick (e.g., A-T / U and C-G) or Hoogsteen base pairing between nucleotides or nucleotide analogs of nucleic acid molecules. “Complementarity” refers to a property shared between two nucleic acid sequences, such that when they are aligned antiparallel to each other, the nucleotide bases at each position will be complementary.

[0056] The term “contacting” as used herein refers to bringing a disclosed composition, CAR- M, or fragment thereof, together with an intended target (such as, e.g., a cell or population of cells, a receptor, an antigen, or other biological entity) in such a manner that the disclosed composition, CAR-M, or fragment thereof can affect the activity of the intended target (e.g., receptor, cell, population of cells, etc.), either directly (i.e., by interacting with the target itself), or indirectly (i.e., by interacting with another molecule, cell, co-factor, factor, or protein on which the activity of the target is dependent). In an aspect, a cell or population of cells, such as cancer cells, can be contacted with a composition, CAR-M, or fragment thereof.

[0057] The terms “control,” “reference level,” and “reference” are used herein interchangeably. The reference level may be a predetermined value or range, which is employed as a benchmark against which to assess the measured result. “Control group” as used herein refers to a group of control subjects. The predetermined level may be a cutoff value from a control group. The predetermined level may be an average from a control group. Cutoff values (or predetermined cutoff values) may be determined by Adaptive Index Model (AIM) methodology. Cutoff values (or predetermined cutoff values) may be determined by a receiver operating curve (ROC) analysis from biological samples of the patient group. ROC analysis, as generally known in the biological arts, is a determination of the ability of a test to discriminate one condition fromanother, e.g., to identify a patient having cancer. A description of ROC analysis is provided in P.J. Heagerty et al. (Biometrics 2000, 56, 337-44), the disclosure of which is hereby incorporated by reference in its entirety. Alternatively, cutoff values may be determined by a quartile analysis of biological samples of a patient group. For example, a cutoff value may be determined by selecting a value that corresponds to any value in the 25th-75th percentile range, preferably a value that corresponds to the 25th percentile, the 50th percentile or the 75th percentile, and more preferably the 75th percentile. Such statistical analyses may be performed using any method known in the art and can be implemented through any number of commercially available software packages (e.g., from Analyse-it Software Ltd., Leeds, UK; StataCorp LP, College Station, TX; SAS Institute Inc., Cary, NC.). The healthy or normal levels or ranges for a target or for a protein activity may be defined in accordance with standard practice. A control may be a subject or cell without a CAR-M as detailed herein. A control may be a subject, or a sample therefrom, whose disease state is known. The subject, or sample therefrom, may be healthy, diseased, diseased prior to treatment, diseased during treatment, or diseased after treatment, or a combination thereof.

[0058] As used herein, the term “cytotoxic” or “cytotoxicity” refers to killing or damaging cells. In one embodiment, cytotoxicity of the metabolically enhanced cells is improved, e.g., increased cytolytic activity of macrophages.

[0059] As used herein, the term “diagnosed” means having been subjected to a physical examination by a person of skill, for example, a physician, and found to have a condition that can be treated by the CAR-Ms, compositions, or methods disclosed herein. For example, “diagnosed with a cancer” means having been subjected to a physical examination by a person of skill, for example, a physician, and found to have a condition that can be diagnosed or can be treated by CAR-Ms, compositions, or methods disclosed herein that can prevent or inhibit malignant cell growth and / or induce apoptosis in a population of cells, such as cancer cells. Such a diagnosis can be in reference to a disorder, such as cancer, and the like, as discussed herein.

[0060] As used herein, “ectodomain” refers to the extracellular domain of a membrane protein that extends into the space outside of a cell. They are often the parts of proteins that initiate contact with surfaces, which leads to signal transduction.

[0061] As used herein, “effector function” or “effector activity” refers to a specific activity carried out by an immune cell in response to stimulation of the immune cell. For example, an effector function of macrophages to engulf and digest cellular debris, foreign substances, microbes, cancer cells, and other unhealthy cells by phagocytosis.

[0062] An “endodomain” as used herein refers to the intracellular part of a transmembrane protein, specifically the signaling domain within the cell cytoplasm. For example, regarding CARs, an endodomain is the part of the receptor that is designed to trigger specific intracellular signaling pathways once the CAR binds to its target antigen on a cancer cell.

[0063] As used herein “endogenous” refers to any material from or produced inside a particular organism, cell, tissue, or system.

[0064] As used herein, the term “exogenous” refers to any material introduced from or produced outside a particular organism, cell, tissue, or system. For example, an ectodomain may be exogenous.

[0065] As used herein, the term “expand” refers to increasing in number, as in an increase in the number of monocytes / macrophages. In one embodiment, monocytes, macrophages, or dendritic cells that are expanded ex vivo increase in number relative to the number originally present in the culture. In another embodiment, monocytes, macrophages, or dendritic cells that are expanded ex vivo increase in number relative to other cell types in the culture. The term “ex vivo,” as used herein, refers to cells that have been removed from a living organism, (e.g., a human) and propagated outside the organism (e.g., in a culture dish, test tube, or bioreactor).

[0066] As used herein, the term “expression” of a nucleic acid sequence refers to generation of any gene product from a nucleic acid sequence. In some embodiments, a gene product can be a transcript. In some embodiments, a gene product can be a polypeptide. In some embodiments, expression of a nucleic acid sequence involves one or more of the following: (1) production of an RNA template from a DNA sequence (e.g, by transcription); (2) processing of an RNA transcript (e.g., by splicing, editing, 5’ cap formation, and / or 3’ end formation); (3) translation of an RNA into a polypeptide or protein; and / or (4) post-translational modification of a polypeptide or protein.

[0067] “Functional” and “full-functional” as used herein describes protein that has biological activity. A “functional gene” refers to a gene transcribed to mRNA, which is translated to a functional protein.

[0068] “Fusion protein” as used herein refers to a chimeric protein created through the joining of two or more genes that originally coded for separate proteins. The translation of the fusion gene results in a single polypeptide with functional properties derived from each of the original proteins.

[0069] “Genetic construct” as used herein refers to the DNA or RNA molecules that comprise a polynucleotide that encodes a protein. The coding sequence includes initiation and termination signals operably linked to regulatory elements including a promoter and polyadenylation signalcapable of directing expression in the cells of the individual to whom the nucleic acid molecule is administered. As used herein, the term “expressible form” refers to gene constructs that contain the necessary regulatory elements operably linked to a coding sequence that encodes a protein such that when present in the cell of the individual, the coding sequence will be expressed.

[0070] The term “heterologous” as used herein refers to nucleic acid comprising two or more subsequences that are not found in the same relationship to each other in nature. For instance, a nucleic acid that is recombinantly produced typically has two or more sequences from unrelated genes synthetically arranged to make a new functional nucleic acid, for example, a promoter from one source and a coding region from another source. The two nucleic acids are thus heterologous to each other in this context. When added to a cell, the recombinant nucleic acids would also be heterologous to the endogenous genes of the cell. Thus, in a chromosome, a heterologous nucleic acid would include a non-native (non-naturally occurring) nucleic acid that has integrated into the chromosome, or a non-native (non-naturally occurring) extrachromosomal nucleic acid. Similarly, a heterologous protein indicates that the protein comprises two or more subsequences that are not found in the same relationship to each other in nature (for example, a “fusion protein,” where the two subsequences are encoded by a single nucleic acid sequence).

[0071] “Identical” or “identity” as used herein in the context of two or more polynucleotide or polypeptide sequences means that the sequences have a specified percentage of residues that are the same over a specified region. The percentage may be calculated by optimally aligning the two sequences, comparing the two sequences over the specified region, determining the number of positions at which the identical residue occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the specified region, and multiplying the result by 100 to yield the percentage of sequence identity. In cases where the two sequences are of different lengths or the alignment produces one or more staggered ends and the specified region of comparison includes only a single sequence, the residues of single sequence are included in the denominator but not the numerator of the calculation. When comparing DNA and RNA, thymine (T) and uracil (U) may be considered equivalent. Identity may be performed manually or by using a computer sequence algorithm such as BLAST or BLAST 2.0.

[0072] As used herein, the term “immune cell,” refers to a cell that is involved in an immune response, e.g., promotion of an immune response. Examples of immune cells include, but are not limited to, macrophages, monocytes, dendritic cells, neutrophils, eosinophils, mast cells, platelets, large granular lymphocytes, Langerhans' cells, natural killer (NK) cells, T- lymphocytes, or B-lymphocytes. A source of immune cells (e.g., macrophages, monocytes, or dendritic cells)can be obtained from a subject Examples of subjects include humans, monkeys, chimpanzees, dogs, cats, mice, rats, and transgenic species thereof.

[0073] As used herein the term “immune response” refers to a cellular response to an antigen that occurs when lymphocytes identify antigenic molecules as foreign and induce the formation of antibodies and / or activate lymphocytes to remove the antigen.

[0074] As used herein, the term “immunoglobulin” or “Ig,” refers to a class of proteins that function as antibodies. Antibodies expressed by B cells are sometimes referred to as a BCR (B cell receptor) or antigen receptor. The five members included in this class of proteins are IgA, IgG, IgM, IgD, and IgE. IgA is the primary antibody that is present in body secretions, such as saliva, tears, breast milk, gastrointestinal secretions, and mucus secretions of the respiratory and genitourinary tracts. IgG is the most common circulating antibody. IgM is the main immunoglobulin produced in the primary immune response in most subjects. It is the most efficient immunoglobulin in agglutination, complement fixation, and other antibody responses, and is important in defense against bacteria and viruses. IgD is an immunoglobulin that has no known antibody function, but may serve as an antigen receptor. IgE is an immunoglobulin that mediates immediate hypersensitivity by causing release of mediators from mast cells and basophils upon exposure to allergen.

[0075] As used herein, the term “isolated” refers to something altered or removed from the natural state. For example, a nucleic acid or a peptide naturally present in a living animal is not “isolated,” but the same nucleic acid or peptide partially or completely separated from the coexisting materials of its natural state is “isolated.” An isolated nucleic acid or protein can exist in substantially purified form, or can exist in a non-native environment such as, for example, a host cell.

[0076] As used herein, the term “lentivirus” refers to a genus of the Retroviridae family. Lentiviruses are unique among the retroviruses in being able to infect non dividing cells; they can deliver a significant amount of genetic information into the DNA of a host cell, so they are one of the most efficient methods of a gene delivery vector. HIV, SIV, and FIV are all examples of lentiviruses. Vectors derived from lentiviruses offer the means to achieve significant levels of gene transfer in vivo.

[0077] As used herein, “macrophage” refers to an immune cell that is specialized for detection, phagocytosis, and destruction of target cells, such as pathogens or tumor cells. Macrophages are potent effectors of the innate immune system and are capable of at least three distinct anti-tumor functions: phagocytosis of dead and dying cells, microorganisms, cancer cells,cellular debris, or other foreign substances; cytotoxicity against tumor cells; and presentation of tumor antigens to orchestrate an adaptive anti-tumor immune response.

[0078] “Normal” as used herein refers to cells that can reproduce when and where they need to, stick together in the right place in the body, and self-destruct when they become damaged or too old. “Normal” as used herein also refers to wildtype animals that are healthy (no diseases or disorders) and are unmodified (e.g., not genetically modified).

[0079] “Nucleic acid” or “oligonucleotide” or “polynucleotide” as used herein means at least two nucleotides covalently linked together. The depiction of a single strand also defines the sequence of the complementary strand. Thus, a polynucleotide also encompasses the complementary strand of a depicted single strand. Many variants of a polynucleotide may be used for the same purpose as a given polynucleotide. Thus, a polynucleotide also encompasses substantially identical polynucleotides and complements thereof. A single strand provides a probe that may hybridize to a target sequence under stringent hybridization conditions. Thus, a polynucleotide also encompasses a probe that hybridizes under stringent hybridization conditions. Polynucleotides may be single stranded or double stranded or may contain portions of both double stranded and single stranded sequence. The polynucleotide can be nucleic acid, natural or synthetic, DNA, genomic DNA, cDNA, RNA, or a hybrid, where the polynucleotide can contain combinations of deoxyribo- and ribo-nucleotides, and combinations of bases including, for example, uracil, adenine, thymine, cytosine, guanine, inosine, xanthine hypoxanthine, isocytosine, and isoguanine. Polynucleotides can be obtained by chemical synthesis methods or by recombinant methods.

[0080] “Operably linked” as used herein means that expression of a gene is under the control of a promoter with which it is spatially connected. A promoter may be positioned 5' (upstream) or 3' (downstream) of a gene under its control. The distance between the promoter and a gene may be approximately the same as the distance between that promoter and the gene it controls in the gene from which the promoter is derived. As is known in the art, variation in this distance may be accommodated without loss of promoter function. Nucleic acid or amino acid sequences are “operably linked” (or “operatively linked”) when placed into a functional relationship with one another. For instance, a promoter or enhancer is operably linked to a coding sequence if it regulates, or contributes to the modulation of, the transcription of the coding sequence. Operably linked DNA sequences are typically contiguous, and operably linked amino acid sequences are typically contiguous and in the same reading frame. However, since enhancers generally function when separated from the promoter by up to several kilobases or more and intronic sequences may be of variable lengths, some polynucleotide elements may be operably linked but notcontiguous. Similarly, certain amino acid sequences that are non-contiguous in a primary polypeptide sequence may nonetheless be operably linked due to, for example folding of a polypeptide chain. With respect to fusion polypeptides, the terms “operatively linked” and “operably linked” can refer to the fact that each of the components performs the same function in linkage to the other component as it would if it were not so linked.

[0081] As used herein, “phagocyte” refers to a type of immune cell that actively engulfs and destroys foreign particles, bacteria, dead cells, and / or debris by a process called phagocytosis, essentially “eating” the harmful substances to protect the body. Examples of phagocytes include, but are not limited to, macrophages, neutrophils, and monocytes, which are all types of white blood cells.

[0082] A “peptide” or “polypeptide” is a linked sequence of two or more amino acids linked by peptide bonds. The polypeptide can be natural, synthetic, or a modification or combination of natural and synthetic. Peptides and polypeptides include proteins such as binding proteins, receptors, and antibodies. The terms “polypeptide,” “protein,” and “peptide” are used interchangeably herein. “Primary structure” refers to the amino acid sequence of a particular peptide. “Secondary structure” refers to locally ordered, three dimensional structures within a polypeptide. These structures are commonly known as domains, for example, enzymatic domains, extracellular domains, transmembrane domains, pore domains, and cytoplasmic tail domains. “Domains” are portions of a polypeptide that form a compact unit of the polypeptide and are typically 15 to 350 amino acids long. Exemplary domains include domains with enzymatic activity or ligand binding activity. Typical domains are made up of sections of lesser organization such as stretches of beta-sheet and alpha-helices. “Tertiary structure” refers to the complete three-dimensional structure of a polypeptide monomer. “Quaternary structure” refers to the three- dimensional structure formed by the noncovalent association of independent tertiary units. A “motif” is a portion of a polypeptide sequence and includes at least two amino acids. A motif may be 2 to 20, 2 to 15, or 2 to 10 amino acids in length. For example, a motif includes 3, 4, 5, 6, or 7 sequential amino acids. A domain may be comprised of a series of the same type of motif.

[0083] “Promoter” as used herein means a synthetic or naturally derived molecule which is capable of conferring, activating, or enhancing expression of a nucleic acid in a cell. A promoter may comprise one or more specific transcriptional regulatory sequences to further enhance expression and / or to alter the spatial expression and / or temporal expression of same. A promoter may also comprise distal enhancer or repressor elements, which may be located as much as several thousand base pairs from the start site of transcription. A promoter may be derived from sources including viral, bacterial, fungal, plants, insects, and animals. A promoter may regulatethe expression of a gene component constitutively, or differentially with respect to cell, the tissue or organ in which expression occurs or, with respect to the developmental stage at which expression occurs, or in response to external stimuli such as physiological stresses, pathogens, metal ions, or inducing agents. Representative examples of promoters include the bacteriophage T7 promoter, bacteriophage T3 promoter, SP6 promoter, lac operator-promoter, tac promoter, SV40 late promoter, SV40 early promoter, RSV-LTR promoter, CMV IE promoter, SV40 early promoter or SV40 late promoter, human U6 (hU6) promoter, spleen focus-forming virus (SFFV), and CMV IE promoter.

[0084] The term “recombinant” when used with reference to, for example, a cell, nucleic acid, protein, or vector, indicates that the cell, nucleic acid, protein, or vector, has been modified by the introduction of a heterologous nucleic acid or protein or the alteration of a native nucleic acid or protein, or that the cell is derived from a cell so modified. Thus, for example, recombinant cells express genes that are not found within the native (naturally occurring) form of the cell or express a second copy of a native gene that is otherwise normally or abnormally expressed, under expressed, or not expressed at all.

[0085] “Sample” or “test sample” as used herein can mean any sample in which the presence and / or level of a target is to be detected or determined or any sample comprising a CAR-M or component thereof as detailed herein. Samples may include liquids, solutions, emulsions, or suspensions. Samples may include a medical sample. Samples may include any biological fluid or tissue, such as blood, whole blood, fractions of blood such as plasma and serum, muscle, interstitial fluid, sweat, saliva, urine, tears, synovial fluid, bone marrow, cerebrospinal fluid, nasal secretions, sputum, amniotic fluid, bronchoalveolar lavage fluid, gastric lavage, emesis, fecal matter, lung tissue, peripheral blood mononuclear cells, total white blood cells, lymph node cells, spleen cells, tonsil cells, cancer cells, tumor cells, bile, digestive fluid, skin, or combinations thereof. In some embodiments, the sample comprises an aliquot. In other embodiments, the sample comprises a biological fluid. Samples can be obtained by any means known in the art. The sample can be used directly as obtained from a patient or can be pre-treated, such as by filtration, distillation, extraction, concentration, centrifugation, inactivation of interfering components, addition of reagents, and the like, to modify the character of the sample in some manner as discussed herein or otherwise as is known in the art.

[0086] As used herein, the term “signal transduction pathway” refers to the biochemical relationship between a variety of signal transduction molecules that play a role in the transmission of a signal from one portion of a cell to another portion of a cell. The phrase “cell surface receptor”includes molecules and complexes of molecules capable of receiving a signal and transmitting signal across the plasma membrane of a cell.

[0087] As used herein, the term “stimulation,” refers to a primary response induced by binding of a stimulatory molecule (e.g., an FcR complex, a TLR complex, or a TCR / CD3 complex) with its cognate ligand thereby mediating a signal transduction event, such as, but not limited to, signal transduction via Fc receptor machinery or via a synthetic CAR. Stimulation can mediate altered expression of certain molecules, such as downregulation of TGF-beta, and / or reorganization of cytoskeletal structures, and the like. As used herein, the term “stimulatory molecule,” refers to a molecule of a monocyte, macrophage, or dendritic cell that specifically binds with a cognate stimulatory ligand present on an antigen presenting cell.

[0088] As used herein, the term “stimulatory ligand,” refers to a ligand that when present on an antigen presenting cell (e.g., an APC, a macrophage, a dendritic cell, a B-cell, and the like) or tumor cell can specifically bind with a cognate binding partner (referred to herein as a “stimulatory molecule”) on a monocyte, macrophage, or dendritic cell thereby mediating a response by the immune cell, including, but not limited to, activation, initiation of an immune response, proliferation, and the like. Stimulatory ligands are well-known in the art and encompass, inter alia, Toll-like receptor (TLR) ligand, an anti-toll-like receptor antibody, an agonist, and an antibody for a monocyte / macrophage receptor. In addition, cytokines, such as interferon-gamma, are potent stimulants of macrophages.

[0089] “Subject” and “patient” as used herein interchangeably refers to any vertebrate, including, but not limited to, a mammal that wants or is in need of the herein described compositions or methods. The subject may be a human or a non-human. The subject may be a vertebrate. The subject may be a mammal. The mammal may be a primate or a non-primate. The mammal can be a non-primate such as, for example, cow, pig, camel, llama, hedgehog, anteater, platypus, elephant, alpaca, horse, goat, rabbit, sheep, hamsters, guinea pig, cat, dog, rat, and mouse. The mammal can be a primate such as a human. The mammal can be a non- human primate such as, for example, monkey, cynomolgous monkey, rhesus monkey, chimpanzee, gorilla, orangutan, and gibbon. The subject may be of any age or stage of development, such as, for example, an adult, an adolescent, or an infant. The subject may be male. The subject may be female. In some embodiments, the subject has cancer. The subject may be undergoing other forms of treatment.

[0090] “Substantially identical” can mean that a first and second amino acid or polynucleotide sequence are at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% identical over a region of 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 , 22,23, 24, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, or 1100 amino acids or nucleotides, respectively.

[0091] As used herein, the term “synthetic antibody” refers to an antibody that is generated using recombinant DNA technology, such as, for example, an antibody expressed by a bacteriophage. The term should also be construed to mean an antibody which has been generated by the synthesis of a DNA molecule encoding the antibody and which DNA molecule expresses an antibody protein, or an amino acid sequence specifying the antibody, wherein the DNA or amino acid sequence has been obtained using synthetic DNA or amino acid sequence technology which is available and well known in the art.

[0092] “T cells” are a type of white blood cell of the immune system and play a central role in the adaptive immune response. T cells express a T-cell receptor (TCR) on their cell surface. The T cell receptor (TCR) of a T cell is able to interact with immunogenic peptides (epitopes) bound to major histocompatibility complex (MHC) molecules and presented on the surface of target cells. Specific binding of the TCR triggers a signal cascade inside the T cell leading to proliferation and differentiation into a maturated effector T cell. T cells may differentiate into different types of T cells. T cells may include, for example, CD8+ T cells (“killer T cells” or “cytotoxic T cells) and CD4+ T cells (“helper T cells”). CD8+ T cells and CD4+ T cells may further differentiate into other types of T cells including, for example, regulatory T cells (“suppressor T cells”) and memory T cells. An antigen-naive T cell expands and differentiates into a memory T cell after encountering the cognate antigen within the context of a major histocompatibility complex (MHC) molecule on the surface of an antigen presenting cell. Memory T cells may be CD8+ or CD4+. Memory T cells are long-lived and can quickly expand to large numbers of effector T cells upon re-exposure to their cognate antigen. Tissue-resident memory T cells (TRMcells) are a subset of a long-lived memory T cells that occupy epithelial, mucosal, and other tissues such as skin, mucosa, lung, brain, pancreas, and gastrointestinal tract, without recirculating. TRMcells may be transcriptionally, phenotypically, and functionally different from central memory (TCM) and effector memory (TEM) T cells that recirculate between blood, the T cell zones of secondary lymphoid organ, lymph tissues, and nonlymphoid tissues. TRMcells can develop from circulating effector memory T cell precursors in response to an antigen. TRMcells may be CD103+. TRMcells may provide superior protection against infection in extra-lymphoid tissues.

[0093] As used herein, a “targeting moiety” can be specific to a recognition molecule on the surface of a cell or a population of cells, such as, for example CAR-Ms. In an aspect of the disclosed methods, a targeting moiety can include, but is not limited to: a monoclonal antibody, a polyclonal antibody, full-length antibody, a chimeric antibody, Fab’, Fab, F(ab)2, F(ab’)2, a singledomain antibody (DAB), Fv, a single chain Fv (scFv), a minibody, a diabody, a triabody, hybrid fragments, a phage display antibody, a ribosome display antibody, a peptide, a peptide ligand, a hormone, a growth factor, or a cytokine. In some embodiments, a targeting moiety may be a scFv on a CAR.

[0094] “Treatment” or “treating” or “treatment” when referring to protection of a subject from a disease, means suppressing, repressing, reversing, alleviating, ameliorating, or inhibiting the progress of disease, or completely eliminating a disease. A treatment may be either performed in an acute or chronic way. The term also refers to reducing the severity of a disease or symptoms associated with such disease prior to affliction with the disease. Preventing the disease involves administering a composition of the present invention to a subject prior to onset of the disease. Suppressing the disease involves administering a composition of the present invention to a subject after induction of the disease but before its clinical appearance. Repressing or ameliorating the disease involves administering a composition of the present invention to a subject after clinical appearance of the disease.

[0095] “Variant” used herein with respect to a polynucleotide means (i) a portion or fragment of a referenced nucleotide sequence; (ii) the complement of a referenced nucleotide sequence or portion thereof; (iii) a nucleic acid that is substantially identical to a referenced nucleic acid or the complement thereof; or (iv) a nucleic acid that hybridizes under stringent conditions to the referenced nucleic acid, complement thereof, or a sequences substantially identical thereto.

[0096] “Variant” with respect to a peptide or polypeptide that differs in amino acid sequence by the insertion, deletion, or conservative substitution of amino acids, but retain at least one biological activity. Variant may also mean a protein with an amino acid sequence that is substantially identical to a referenced protein with an amino acid sequence that retains at least one biological activity. Representative examples of “biological activity” include the ability to be bound by a specific antibody or polypeptide or to promote an immune response. Variant can mean a functional fragment thereof. Variant can also mean multiple copies of a polypeptide. The multiple copies can be in tandem or separated by a linker. A conservative substitution of an amino acid, for example, replacing an amino acid with a different amino acid of similar properties (for example, hydrophilicity, degree and distribution of charged regions) is recognized in the art as typically involving a minor change. These minor changes may be identified, in part, by considering the hydropathic index of amino acids, as understood in the art (Kyte et al., J. Mol. Biol. 1982, 157, 105-132). The hydropathic index of an amino acid is based on a consideration of its hydrophobicity and charge. It is known in the art that amino acids of similar hydropathic indexes may be substituted and still retain protein function. In one aspect, amino acids having hydropathicindexes of ±2 are substituted. The hydrophilicity of amino acids may also be used to reveal substitutions that would result in proteins retaining biological function. A consideration of the hydrophilicity of amino acids in the context of a peptide permits calculation of the greatest local average hydrophilicity of that peptide. Substitutions may be performed with amino acids having hydrophilicity values within ±2 of each other. Both the hydrophobicity index and the hydrophilicity value of amino acids are influenced by the particular side chain of that amino acid. Consistent with that observation, amino acid substitutions that are compatible with biological function are understood to depend on the relative similarity of the amino acids, and particularly the side chains of those amino acids, as revealed by the hydrophobicity, hydrophilicity, charge, size, and other properties.

[0097] “Vector” or “expression vector” as used herein means a nucleic acid sequence containing an origin of replication. A vector may be a viral vector, bacteriophage, bacterial artificial chromosome, or yeast artificial chromosome. A vector may be a DNA or RNA vector. A vector may be a self-replicating extrachromosomal vector, and preferably, is a DNA plasmid. For example, the vector may encode a CAR described herein.

[0098] As used herein, the term “xenogeneic” refers to a graft (e.g., a population of cells) derived from an animal of a different species.2. Chimeric Antigen Receptors (CARs)

[0099] Provided herein are CARs that comprise one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain. The CAR may be encoded by a plasmid having SEQ ID NO: 13. The CAR may be the amino acid sequence of SEQ ID NO: 16, or a variant thereof.[000100] CARs may be expressed with specificity to an antigen, for example, a tumor associated antigen. The tumor associated antigen may be CSPG4, also known as neuron-glial antigen 2 (NG2). The CSPG4 binding domain may be encoded by SEQ ID NO: 1 or a variant thereof. The CSPG4 binding domain may be the amino acid sequence of SEQ ID NO: 2 or a variant thereof. The CSPG4 binding domain may comprise a monoclonal antibody, a polyclonal antibody, a synthetic antibody, a human antibody, a humanized antibody, a non-human antibody, or any fragment thereof, for example an scFv. a. Ectodomain[000101] A CAR may comprise an ectodomain operably linked to another domain of the CAR, such as a transmembrane domain or an endodomain, for expression in an immune cell. A nucleic acid encoding an ectodomain may be operably linked to a nucleic acid encoding a transmembranedomain and the nucleic acid encoding the transmembrane domain may be operably linked to a nucleic acid encoding an endodomain.[000102] The ectodomain may comprise a spacer, a co-stimulation domain, a secreted antibody domain, a cytokine secretion domain, or a combination thereof. A CAR may comprise a spacer domain or hinge between an ectodomain and a transmembrane domain. A CAR may comprise a spacer domain or hinge between an endodomain and a transmembrane domain. As used herein, the term “spacer” or “spacer domain” or “hinge” or “hinge domain” refers to any oligo- or polypeptide that functions to link a transmembrane domain to either an ectodomain or to an endodomain in a polypeptide chain. The spacer or hinge may comprise from about 90 nucleic acids to about 150 nucleic acids, about 100 nucleic acids to about 150 nucleic acids, about 110 nucleic acids to about 150 nucleic acids, about 120 nucleic acids to about 150 nucleic acids, about130 nucleic acids to about 150 nucleic acids, about 140 nucleic acids to about 150 nucleic acids, about 90 nucleic acids to about 140 nucleic acids, about 90 nucleic acids to about 130 nucleic acids, about 90 nucleic acids to about 120 nucleic acids, about 90 nucleic acids to about 110 nucleic acids, or about 90 nucleic acids to about 100 nucleic acids in length. The spacer or hinge may be encoded by SEQ ID NO: 17, or a variant thereof. The spacer domain or hinge may comprise up to 300 amino acids, 10 to 100 amino acids, or 25 to 50 amino acids. The spacer or hinge may comprise from about 30 amino acids to about 50 amino acids, about 32 amino acids to about 50 amino acids, about 34 amino acids to about 50 amino acids, about 36 amino acids to about 50 amino acids, about 38 amino acids to about 50 amino acids, about 40 amino acids to about 50 amino acids, about 42 amino acids to about 50 amino acids, about 44 amino acids to about 50 amino acids, about 46 amino acids to about 50 amino acids, about 48 amino acids to about 50 amino acids, about 30 amino acids to about 48 amino acids, about 30 amino acids to about 46 amino acids, about 30 amino acids to about 44 amino acids, about 30 amino acids to about 42 amino acids, about 30 amino acids to about 40 amino acids, about 30 amino acids to about 38 amino acids, about 30 amino acids to about 36 amino acids, about 30 amino acids to about 34 amino acids, or about 30 amino acids to about 32 amino acids in length. The spacer or hinge may be the amino acid sequence of SEQ ID NO: 18, or a variant thereof.[000103] A CAR may comprise one or more extracellular spacer or hinge domains as described herein. An extracellular hinge domain may comprise a human extracellular hinge domain. An extracellular hinge domain may be a domain that is endogenous to a particular immune cell type (e.g., a modified phagocyte as provided herein). An extracellular hinge domain may be a domain that is not endogenous to a particular immune cell type (e.g., a modified phagocyte as provided herein). One or more extracellular hinge domains may comprise a CD8 extracellular hingedomain or an lgG4 or a CD28 extracellular hinge domain. One or more extracellular hinge domains may comprise a CD8 extracellular hinge domain. An extracellular hinge domain may optimize the physicochemical parameters of a CAR, e.g., optimal size relative to tumor antigen (e.g., allowing for exclusion of inhibitory molecules), optimal flexibility, optimal protein folding, optimal protein stability, optimal binding, optimal homodimerization, and / or lack of homodimerization.[000104] A co-stimulation domain may comprise a domain such as, but not limited to, 4-1 BB (also known as CD137 or TNF receptor superfamily member 9 (TNFRSF9)).[000105] A secreted antibody domain may comprise, for example, a domain secreting a CD47 antibody.[000106] A cytokine secretion domain may comprise, for example, an Interferon gamma secretion domain.[000107] A short oligopeptide or polypeptide linker, from about 2 amino acids to about 10 amino acids in length, may form a linkage between a transmembrane domain and endodomain of a CAR. An example of a linker includes a glycine-serine doublet. b. Transmembrane Domain[000108] A CAR may comprise a transmembrane domain, for example, that connects an ectodomain to an endodomain. A transmembrane domain may be naturally associated with one or more other domain(s) of a CAR. A transmembrane domain can be modified to avoid binding to transmembrane domains of other surface membrane proteins, in order to minimize interactions with other members of a receptor complex. A transmembrane domain may be derived either from a naturally-occurring or from a synthetic source. A transmembrane domain may be derived from a naturally-occurring membrane-bound or transmembrane protein. A transmembrane domain may comprise a human transmembrane domain. A transmembrane domain may be a domain that is endogenous to a particular immune cell type (e.g., a modified phagocyte as provided herein). A transmembrane domain may be a domain that is not endogenous to a particular immune cell type (e.g., a modified phagocyte as provided herein). A transmembrane domain may comprise a CD28, CD8, CD8a, CD64, CD32a, CD32c, CD16a, TRL1 , TLR2, TLR3, TRL4, TLR5, TLR6, TLR7, TLR8, TLR9, ALK, AXL, DDR2, EGFR, EphAI, INSR, cMET, MUSK, PDGFR, PTK7, RET, ROR1, ROS1 , RYK, TIE2, TRK, VEGFR, CD40, CD19, CD20, 41 BB, CD28, 0X40, GITR, TREM-1 , TREM-2, DAP 12, MR, ICOS, MyD88, CD3-zeta, FcRg, V / l / LxYxxL / V, SIRPα, CD45, Siglec-10, PD1 , SHP-1, SHP-2, KIR-2DL, KIR-3DL, NKG2A, CD170, CD33, BTLA, CD32b, SIRPb, CD22, PIR-B, LILRBI, CD36, or Syk transmembrane domain. The transmembrane domain may be a CD8 transmembrane domain or a FcRy transmembrane domain. Thetransmembrane domain may be encoded by SEQ ID NO: 3, SEQ ID NO: 5, or SEQ ID NO: 7, or a variant thereof. The transmembrane domain may be the amino acid sequence of SEQ ID NO: 4, SEQ ID NO: 6, or SEQ ID NO: 8, or a variant thereof. c. Endodomain[000109] A CAR may comprise one or more endodomains. An endodomain may comprise a human intracellular domain, or portion thereof. An endodomain may be a domain that is endogenous to a particular immune cell type (e.g., a modified phagocyte as provided herein). An endodomain may be a domain that is not endogenous to a particular immune cell type (e.g., a modified phagocyte as provided herein). An endodomain and / or other cytoplasmic domain of a CAR may be responsible for activation of the cell in which the CAR is expressed (e.g., an immune cell). An endodomain of a CAR may be responsible for signal activation and / or transduction in an immune cell comprising the CAR. An endodomain of a CAR may include at least one domain responsible for signal activation and / or transduction. An endodomain may comprise at least one of a co-stimulatory molecule and a signaling domain. An endodomain of a CAR may comprise dual signaling domains. An endodomain of a CAR may comprise more than two signaling domains.[000110] An endodomain may comprise a cytoplasmic portion of a surface receptor. An endodomain may comprise a co-stimulatory molecule. An endodomain may comprise a molecule that acts to initiate signal transduction in an immune cell. An endodomain of a CAR may include any portion of one or more co-stimulatory molecules, such as at least one signaling domain from CD3, Fc epsilon Rl gamma chain, any derivative or variant thereof, any synthetic sequence thereof that has the same functional capability, and any combination thereof.[000111] The endodomain of the CAR may comprise a phagocytic receptor or a fragment thereof. The phagocytic receptor may be multiple EGF like domains 10 (MegflO), high affinity immunoglobulin epsilon receptor subunit gamma (FcRy), or MER proto-oncogene tyrosine kinase (MerTK). The endodomain may comprise a His, Myc, luciferase, or fluorescent tag. The fluorescent tag may be any known in the art, such as, but not limited to, green fluorescent protein (GFP), yellow fluorescent protein (YFP), red fluorescent protein (RFP), blue fluorescent protein (BFP), and cyan fluorescent protein (CYP). The endodomain may be encoded by SEQ ID NO: 9 or SEQ ID NO: 11, or a variant thereof. The endodomain may be the amino acid sequence of SEQ ID NO: 10 or SEQ ID NO: 12, or a variant thereof. In some embodiments, the CAR does not comprise an endodomain domain comprising a phagocytic receptor or a fragment thereof.3. Genetic Constructs[000112] The CAR may be encoded by or comprised within a genetic construct. The genetic construct, such as a plasmid or expression vector, may comprise a nucleic acid that encodes the CAR or a fragment thereof. The genetic construct encoding the CAR may comprise an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain. The CSPG4 binding domain may be encoded by SEQ ID NO: 1 or a variant thereof. The transmembrane domain may be encoded by SEQ ID NO: 3, SEQ ID NO: 5, or SEQ ID NO: 7, or a variant thereof. The endodomain may be encoded by SEQ ID NO: 9, or SEQ ID NO: 11 , or a variant thereof. The CAR may be encoded by a plasmid having SEQ ID NO: 13.[000113] In some embodiments, a genetic construct may comprise two or more nucleic acid sequences encoding a CAR, e.g., at least one CAR described herein, and a second CAR, e.g., a different CAR described herein. In some embodiments, two or more nucleic acid sequences encoding a CAR and a second CAR may be encoded by a single genetic construct, e.g., in same frame and as a single polypeptide chain. In some embodiments, two or more CARs may be separated by one or more cleavage peptide sites (e.g., an auto-cleavage site or a substrate for an intracellular protease). In certain embodiments, a cleavage peptide may comprise a porcine teschovirus-1 (P2A) peptide, Thosea asigna virus (T2A) peptide, equine rhinitis A virus (E2A) peptide, foot-and-mouth disease virus (F2A) peptide, or a variant thereof.[000114] Genetic constructs may include polynucleotides such as vectors and plasmids. The genetic construct may be a linear minichromosome including centromere, telomeres, or plasmids or cosmids. The vector may be an expression vectors or system to produce protein by routine techniques and readily available starting materials including Sambrook et al., Molecular Cloning and Laboratory Manual, Second Ed., Cold Spring Harbor (1989), which is incorporated fully by reference. The construct may be recombinant. The genetic construct may be part of a genome of a recombinant viral vector, including recombinant lentivirus, recombinant adenovirus, and recombinant adenovirus associated virus. The genetic construct may comprise regulatory elements for gene expression of the coding sequences of the nucleic acid. The regulatory elements may be a promoter, an enhancer, an initiation codon, a stop codon, or a polyadenylation signal.[000115] Genetic constructs encoding at least one CAR described herein or a fragment thereof can be a DNA molecule, an RNA molecule, or a combination thereof. In some embodiments, a genetic construct may comprise or may be a messenger RNA (mRNA) transcript encoding at least one CAR described herein or a fragment thereof. In some embodiments, a genetic construct maycomprise or may be a DNA construct encoding at least one CAR described herein or a fragment thereof.[000116] All or a fragment of a CAR described herein may be encoded by a codon optimized genetic construct, such as for expression in a cell (e.g., a mammalian cell). A variety of codon optimization methods are known in the art, e.g., as disclosed in US Patent Nos. 5,786,464 and 6,114,148, each of which is hereby incorporated by reference in its entirety.[000117] A vector may comprise a genetic construct encoding at least one CAR as described herein or a fragment thereof. In some embodiments, a vector may comprise a plasmid, viral vector, phagemid, retrotransposon (e.g., piggyback or sleeping beauty), site directed insertion vector (e.g., CRISPR / Cas systems (e.g., CRISPR / Cas systems comprising one or more of Cas9, Casl2a, or C2c2), Zn finger nucleases, or TALEN for insertion of a template donor DNA comprising a genetic construct encoding at least one CAR as described herein), suicide expression vector, or any other vector known in the art. Vectors can be suitable for replication and integration in eukaryotes. Vectors can include expression vectors, replication vectors, probe generation vectors, and sequencing vectors.[000118] A vector can comprise an origin of replication, a promoter sequence (e.g., a constitutive or inducible promoter), and / or convenient restriction endonuclease sites (e.g., WO 01 / 96584; WO 01 / 29058; and U.S. Pat. No. 6,326,193, each of which are hereby incorporated by reference in their entirety). A vector can also include a signal sequence to facilitate secretion, a polyadenylation signal, and transcription terminator (e.g., a Bovine Growth Hormone (BGH) polyadenylation signal), an element allowing episomal replication and replication in prokaryotes (e.g., SV40 origin and / or ColEI), elements to allow selection (e.g., an ampicillin resistance gene and / or zeocin marker), and / or reporter genes (e.g., luciferase, betagalactosidase, chloramphenicol acetyl transferase, secreted alkaline phosphatase, or green fluorescent protein).[000119] Expression of genetic constructs as described herein may be achieved by operably linking a nucleic acid encoding a CAR polypeptide or fragment thereof to a promoter in an expression vector. Exemplary promoters (e.g., constitutive promoters) include, but are not limited to, an elongation factor-la promoter (EF-la) promoter, immediate early cytomegalovirus (CMV) promoter, ubiquitin C promoter, phosphoglycerokinase (PGK) promoter, simian virus 40 (SV40) early promoter, mouse mammary tumor virus (MMTV) promoter, human immunodeficiency virus (HIV) long terminal repeat (LTR) promoter, Moloney murine leukemia virus (MoMuLV) promoter, an avian leukemia virus promoter, an Epstein-Barr virus immediate early promoter, a Rous sarcoma virus promoter, an actin promoter, a myosin promoter, a hemoglobin promoter, or acreatine kinase promoter. Examples of inducible promoters include, but are not limited to a metallothionine promoter, a glucocorticoid promoter, a progesterone promoter, and a tetracycline promoter. A vector can also comprise additional promoter elements, e.g., enhancers, to regulate the frequency of transcriptional initiation.[000120] In some embodiments, a vector comprising a genetic construct encoding at least one CAR as described herein or a fragment thereof may comprise or may be a viral vector. Viral vector technology is well known in the art and is described (e.g., in Sambrook et al., 2012, MOLECULAR CLONING: A LABORATORY MANUAL, volumes 1-4, Cold Spring Harbor Press, NY). Examples of viral vectors include, but are not limited to, adenoviral vectors, adeno- associated viral vectors, or retroviral vectors (e.g., a lentiviral vector or a gammaretroviral vector). In some embodiments, a vector may be a lentiviral vector (e.g., as described in US Patent No. 9,149,519 or International Publication No. WO 2017 / 044487, each of which is hereby incorporated by reference in its entirety). In some embodiments, a viral vector may be an adenoviral vector. In some embodiments, a viral vector may be an adeno-associated virus (AAV) vector. AAV systems are generally well known in the art (e.g., Kelleher and Vos, Biotechniques, 17(6): 1110-17 (1994); Cotten et ah, P.N.A.S. U.S.A., 89(13):6094-98 (1992); Curiel, Nat lmmun, 13(2-3): 141-64 (1994); Muzyczka, Curr Top Microbiol Immunol, 158:97-129 (1992); and Asokan A, et ah, Mol. Then, 20(4):699-708 (2012)). Methods for generating and using recombinant AAV (rAAV) vectors are described, for example, in U.S. Pat. Nos. 5,139,941 and 4,797,368, each of which is hereby incorporated by reference in its entirety.[000121] In some embodiments, a genetic construct may comprise at least one nucleic acid sequence encoding a CAR, e.g., at least one CAR described herein, and at least one nucleic acid encoding at least one gene co-expressed with a CAR, e.g., a cytokine (e.g., TNF, IL-12, IFN, GM- CSF, G-CSF, M-CSF, and / or IL-1) or a stimulatory ligand (e.g., CD7, B7-1 (CD80), B7-2 (CD86), PD-L1 , PD-L2, 4-1 BBL, OX40L, ICOS-L, ICAM, CD30L, CD40, CD70, CD83, HLA-G, MICA, MICB, HVEM, lymphotoxin beta receptor, 3 / TR6, ILT3, ILT4, HVEM, an agonist or antibody that binds Toll ligand receptor, and / or a B7-H3 ligand.4. Modified Phagocytes[000122] Provided herein are modified phagocytes (e.g., macrophages, monocytes, or dendritic cells) comprising at least one chimeric antigen receptor (CAR) as described herein. A modified phagocyte may be generated by expressing a CAR therein. A modified phagocyte may comprise a CAR that comprises one or more of: an ectodomain comprising a chondroitin sulfateproteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain. The modified phagocyte may comprise a plasmid encoding the CAR described herein.[000123] A phagocyte may be activated, e.g., a phagocyte exhibits increased cytokine production, chemokine production, phagocytosis, cell signaling, target cell killing, and / or antigen presentation relative to an inactive cell. An activated phagocyte may exhibit changes in gene expression, e.g., an induction of pro-inflammatory gene expression (e.g., one or more of TNF, IL- 12, IFN, GM-CSF, G-CSF, M-CSF, and IL-1) relative to an inactive cell. Activated phagocytes may be undergoing cell division.[000124] The modified phagocyte may be a macrophage (e.g., a CAR-M). A macrophage may be an undifferentiated or MO macrophage. A macrophage may comprise or express one or more of CD14, CD16, CD64, CD68, CD71 , and CCR5. Exposure to various stimuli can induce MO macrophages to polarize into several distinct populations, which may be identified by macrophage phenotype markers, cytokine production, and / or chemokine secretion. The macrophage may be a M1 macrophage. A macrophage may express one or more markers of M1 macrophages (e.g., CD86, CD80, MHC II, IL-IR, TLR2, TLR4, iNOS, SOCS3, CD83, PD-L1 , CD69, MHO I, CD64, CD32, CD16, I LIR, a I FIT family member, and / or an ISG family member). A macrophage may be an M2 macrophage (e.g., an M2a, M2b, M2c, or M2d macrophage). An M2a macrophage can be induced by IL-4, IL-13, and / or fungal infection. An M2b macrophage can be induced by IL-1 R ligands, an immune complex, and / or LPS. An M2c macrophage can be induced by IL-10 and / or TGFp. An M2d macrophage can be induced by IL-6 and / or adenosine. A macrophage may express one or more markers of M2 macrophages (e.g., CD206, CD163, and / or CD209). A macrophage may comprise at least one upregulated M1 marker and / or at least one downregulated M2 marker. The at least one M1 marker (e.g, HLA DR, CD86, CD80, PD-L1 , CD83, CD69, MHC I, CD64, CD32, CD16, IL1 R, a IFIT family member, and / or an ISG family member) may be upregulated in a macrophage. The at least one M2 marker (e.g., CD206, CD163, and / or CD209) may be downregulated in a macrophage.[000125] A phagocyte comprising or expressing at least one CAR described herein may exhibit increased phagocytosis relative to a phagocyte without a CAR as described herein. A phagocyte comprising or expressing at least one CAR described herein may exhibit increased cytotoxicity against a tumor cell relative to a phagocyte without a CAR as described herein. A phagocyte comprising or expressing at least one CAR described herein may exhibit increased tumor antigen presentation (e.g., post-phagocytosis presentation) and / or increased antigen processing relative to a phagocyte without a CAR as described herein. A phagocyte comprising or expressing at least one CAR described herein may exhibit increased tumor killing (e.g., by phagocytosis, lysis,apoptosis, or production of tumor killing cytokines (e.g., TNFa) relative to a phagocyte without a CAR as described herein.[000126] The phagocytes (e.g., macrophages, monocytes, or dendritic cells) may be obtained (e.g., isolated) from a subject. Phagocytes may be autologous or sourced from allogeneic or universal donors. Phagocytes can be obtained from a number of sources including peripheral blood mononuclear cells, bone marrow, lymph node tissue, spleen tissue, umbilical cord, tumors, and / or induced pluripotent stem cells, such as embryonic stem cells (ESCs). The phagocytes can be enriched. Following enrichment, differentiation of immune cells can include stimulation with GM-CSF. The phagocytes may be treated with at least one pro-inflammatory agent that promotes a M1 phenotype of the phagocytes (e.g., a switch from M2 to M1 phenotype).[000127] In some embodiments, targeted effector activity of a phagocyte may be enhanced by inhibition of CD47 or signal regulatory protein a (SIRPα). CD47 may be inhibited by treating a cell with an anti-CD47 antibody or by any method known to those skilled in the art. For example, an anti-CD47 antibody may be magrolimab, SRF231 , IMM0306, AK117, gentulizumab, and the like. SIRPα may be inhibited by treating a cell with an anti-SIRPα antibody (such as Bl 765063, BR105, KWAR23, 1 H9, and the like), an inhibitor of SIRPα (such as SMC18, SP5, and the like), or by any method known to those skilled in the art.5. Pharmaceutical Compositions[000128] Further provided herein are pharmaceutical compositions comprising the above- described genetic constructs or modified phagocytes. In some embodiments, the pharmaceutical composition may comprise about 10 μg / mL to about 200 μg / mL, about 20 μg / mL to about 200 μg / mL, about 30 μg / mL to about 200 μg / mL, about 40 μg / mL to about 200 μg / mL, about 50 μg / mL to about 200 μg / mL, about 60 μg / mL to about 200 μg / mL, about 70 μg / mL to about 200 μg / mL, about 80 μg / mL to about 200 μg / mL, about 90 μg / mL to about 200 μg / mL, about 100 μg / mL to about 200 μg / mL, about 110 μg / mL to about 200 μg / mL, about 120 μg / mL to about 200 μg / mL, about 130 μg / mL to about 200 μg / mL, about 140 μg / mL to about 200 μg / mL, about 150 μg / mL to about 200 μg / mL, about 160 μg / mL to about 200 μg / mL, about 170 μg / mL to about 200 μg / mL, about 180 μg / mL to about 200 μg / mL, about 190 μg / mL to about 200 μg / mL, about 10 μg / mL to about 190 μg / mL, about 10 μg / mL to about 180 μg / mL, about 10 μg / mL to about 170 μg / mL, about 10 μg / mL to about 160 μg / mL, about 10 μg / mL to about 150 μg / mL, about 10 μg / mL to about 140 pg / m L, about 10 pg / m L to about 130 pg / m L, about 10 pg / m L to about 120 pg / m L, about 10 μg / mL to about 110 μg / mL, about 10 μg / mL to about 100 μg / mL, about 10 μg / mL to about 90 μg / mL, about 10 μg / mL to about 80 μg / mL, about 10 μg / mL to about 70 μg / mL, about 10 μg / mL to about60 μg / mL, about 10 μg / mL to about 50 μg / mL, about 10 μg / mL to about 40 μg / mL, about 10 μg / mL to about 30 μg / mL, or about 10 μg / mL to about 20 μg / mL of mRNA encoding the CAR described herein. mRNA encoding the CAR described herein may be delivered using, for example, nanoparticles. In some embodiments, the pharmaceutical composition may comprise from about 1 million to about 10 million, about 2 million to about 10 million, about 3 million to about 10 million, about 4 million to about 10 million, about 5 million to about 10 million, about 6 million to about 10 million, about 7 million to about 10 million, about 8 million to about 10 million, about 9 million to about 10 million, about 1 million to about 9 million, about 1 million to about 8 million, about 1 million to about 7 million, about 1 million to about 6 million, about 1 million to about 5 million, about 1 million to about 4 million, about 1 million to about 3 million, or about 1 million to about 2 million of modified phagocytes described herein. In an embodiment, a pharmaceutical composition may comprise a modified macrophage comprising a CAR comprising one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain described herein. The composition may further comprise from about 1 mg / kg to about 45 mg / kg, about 5 mg / kg to about 45 mg / kg, about 10 mg / kg to about 45 mg / kg, about 15 mg / kg to about 45 mg / kg, about 20 mg / kg to about 45 mg / kg, about 25 mg / kg to about 45 mg / kg, about 30 mg / kg to about 45 mg / kg, about 35 mg / kg to about 45 mg / kg, about 40 mg / kg to about 45 mg / kg, about 1 mg / kg to about 40 mg / kg, about 1 mg / kg to about 35 mg / kg, about 1 mg / kg to about 30 mg / kg, about 1 mg / kg to about 25 mg / kg, about 1 mg / kg to about 20 mg / kg, about 1 mg / kg to about 15 mg / kg, about 1 mg / kg to about 10 mg / kg, or about 1 mg / kg to about 5 mg / kg of an anti-CD47 antibody.[000129] The modified phagocytes or genetic constructs as detailed herein, or at least one component thereof, may be formulated into pharmaceutical compositions in accordance with standard techniques well known to those skilled in the pharmaceutical art. The pharmaceutical compositions can be formulated according to the mode of administration to be used. In cases where pharmaceutical compositions are injectable pharmaceutical compositions, they are sterile, pyrogen free, and particulate free. An isotonic formulation is preferably used. Generally, additives for isotonicity may include sodium chloride, dextrose, mannitol, sorbitol and lactose. In some cases, isotonic solutions such as phosphate buffered saline are preferred. Stabilizers include gelatin and albumin. In some embodiments, a vasoconstriction agent is added to the formulation. [000130] The composition may further comprise a pharmaceutically acceptable excipient. The pharmaceutically acceptable excipient may be functional molecules as vehicles, adjuvants, carriers, or diluents. The term “pharmaceutically acceptable carrier,” may be a non-toxic, inert solid, semi-solid or liquid filler, diluent, encapsulating material or formulation auxiliary of any type.Pharmaceutically acceptable carriers include, for example, diluents, lubricants, binders, disintegrants, colorants, flavors, sweeteners, antioxidants, preservatives, glidants, solvents, suspending agents, wetting agents, surfactants, emollients, propellants, humectants, powders, pH adjusting agents, and combinations thereof. The pharmaceutically acceptable excipient may be a transfection facilitating agent, which may include surface active agents, such as immune- stimulating complexes (ISCOMS), Freunds incomplete adjuvant, LPS analog including monophosphoryl lipid A, muramyl peptides, quinone analogs, vesicles such as squalene and squalene, hyaluronic acid, lipids, liposomes, calcium ions, viral proteins, polyanions, polycations, or nanoparticles, or other known transfection facilitating agents. The transfection facilitating agent may be a polyanion, polycation, including poly-L-glutamate (LGS), or lipid. In some embodiments, the pharmaceutically acceptable carrier may be Matrigel® which contains high levels of type IV collagen as well as laminin and enactin / nidogen or any hydrogel known in the art.6. Administration[000131] The genetic constructs as detailed herein or compositions comprising the same, or at least one component thereof, may be administered or delivered to a cell. Methods of introducing a nucleic acid into a host cell are known in the art, and any known method can be used to introduce a nucleic acid (e.g., an expression construct) into a cell. Suitable methods include, for example, viral or bacteriophage infection, transfection, conjugation, protoplast fusion, polycation or lipid:nucleic acid conjugates, lipofection, electroporation, nucleofection, immunoliposomes, calcium phosphate precipitation, polyethyleneimine (PEI)-mediated transfection, DEAE-dextran mediated transfection, liposome-mediated transfection, particle gun technology, calcium phosphate precipitation, direct micro injection, nanoparticle-mediated nucleic acid delivery, and the like. In some embodiments, the composition may be delivered by mRNA delivery and ribonucleoprotein (RNP) complex delivery. The genetic construct, or composition comprising the same, may be electroporated using BioRad Gene Pulser Xcell or Amaxa Nucleofector lib devices or other electroporation device. Several different buffers may be used, including BioRad electroporation solution, Sigma phosphate-buffered saline product #D8537 (PBS), Invitrogen OptiMEM I (OM), or Amaxa Nucleofector solution V (N.V.). Transfections may include a transfection reagent, such as Lipofectamine 2000.[000132] The modified phagocytes or genetic constructs as detailed herein, or at least one component thereof, or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody or a combination thereof, or the pharmaceutical compositions comprising the same, may be administered to a subject. Such compositions can be administered in dosagesand by techniques well known to those skilled in the medical arts taking into consideration such factors as the age, sex, weight, and condition of the particular subject, and the route of administration. The presently disclosed modified phagocytes, or at least one component thereof, genetic constructs, or compositions comprising the same, or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be administered to a subject by different routes including orally, parenterally, sublingually, transdermally, rectally, transmucosally, topically, intranasal, intravaginal, via inhalation, via buccal administration, intrapleurally, intravenous, intraarterial, intraperitoneal, subcutaneous, intradermally, epidermally, intramuscular, intranasal, intrathecal, intracranial, and intraarticular or combinations thereof. In certain embodiments, the modified phagocytes, genetic construct, or composition comprising the same, or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be administered to a subject locally at a tumor site, intravenously, or a combination thereof. The genetic constructs or compositions comprising the same, or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be delivered to a subject by several technologies including DNA injection (also referred to as DNA vaccination) with and without in vivo electroporation, liposome mediated, nanoparticle facilitated, recombinant vectors such as recombinant lentivirus, recombinant adenovirus, and recombinant adenovirus associated virus. The composition or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be injected into the brain or other component of the central nervous system. The composition or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be injected into the skin. The composition or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be injected into tumor lesions. The composition or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be administered locally in the brain at time of tumor resection.[000133] For veterinary use, the modified phagocytes, genetic constructs, or compositions comprising the same, or an anti-CD47 antibody may be administered as a suitably acceptable formulation in accordance with normal veterinary practice. The veterinarian may readily determine the dosing regimen and route of administration that is most appropriate for a particular animal.[000134] The modified phagocytes, genetic constructs, or compositions comprising the same, or an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPαantibody, or a combination thereof may be administered by traditional syringes, needleless injection devices, “microprojectile bombardment gone guns,” or other physical methods such as electroporation (“EP”), “hydrodynamic method,” or ultrasound.7. Kits[000135] Provided herein is a kit, which may be used to treat cancer or prevent the recurrence of cancer. The kit comprises genetic constructs or a composition comprising the same, for generating modified phagocytes that express a CAR described herein, and instructions for using said composition. In an embodiment, the kit comprises at least one CAR comprising or encoded by a polynucleotide sequence of SEQ ID NO: 13, a complement thereof, a variant thereof, or fragment thereof, and instructions for using the CAR.[000136] Instructions included in kits may be affixed to packaging material or may be included as a package insert. While the instructions are typically written on printed materials, they are not limited to such. Any medium capable of storing such instructions and communicating them to an end user is contemplated by this disclosure. Such media include, but are not limited to, electronic storage media (e.g., magnetic discs, tapes, cartridges, chips), optical media (e.g., CD ROM), and the like. As used herein, the term “instructions” may include the address of an internet site that provides the instructions.[000137] The genetic constructs or a composition comprising thereof for treating cancer or preventing the recurrence of cancer may include a modified lentiviral or AAV vector that includes a CAR as described above, that specifically binds to CSPG4 on tumor cells. The modified phagocytes, as described above, may be included in the kit to specifically bind and target a particular tumor cell, for example, tumor cells that express CSPG4.8. Methods a. Methods of Treating Cancer[000138] Provided herein are methods of treating cancer in a subject in need thereof. Also provided herein are methods of preventing recurrent cancer in a subject in need thereof. The methods may include administering to the subject a therapeutically effective amount of the pharmaceutical composition described herein. From about 1 million to about 10 million, about 2 million to about 10 million, about 3 million to about 10 million, about 4 million to about 10 million, about 5 million to about 10 million, about 6 million to about 10 million, about 7 million to about 10 million, about 8 million to about 10 million, about 9 million to about 10 million, about 1 million to about 9 million, about 1 million to about 8 million, about 1 million to about 7 million, about 1 millionto about 6 million, about 1 million to about 5 million, about 1 million to about 4 million, about 1 million to about 3 million, or about 1 million to about 2 million of modified phagocytes described herein may be administered to the subject. The therapeutically effective amount of the pharmaceutical composition described herein may be administered to a subject at the time of surgical removal of a malignant tumor to prevent recurrence of the cancer.[000139] The method may further comprise administering a therapeutically effective amount of an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof, or a composition that comprises an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof, as described herein. From about 1 mg / kg to about 45 mg / kg, about 5 mg / kg to about 45 mg / kg, about 10 mg / kg to about 45 mg / kg, about 15 mg / kg to about 45 mg / kg, about 20 mg / kg to about 45 mg / kg, about 25 mg / kg to about 45 mg / kg, about 30 mg / kg to about 45 mg / kg, about 35 mg / kg to about 45 mg / kg, about 40 mg / kg to about 45 mg / kg, about 1 mg / kg to about 40 mg / kg, about 1 mg / kg to about 35 mg / kg, about 1 mg / kg to about 30 mg / kg, about 1 mg / kg to about 25 mg / kg, about 1 mg / kg to about 20 mg / kg, about 1 mg / kg to about 15 mg / kg, about 1 mg / kg to about 10 mg / kg, or about 1 mg / kg to about 5 mg / kg of the anti-CD47 antibody may be administered to the subject.[000140] The anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti- SIRPα antibody, or a combination thereof may be co-administered with the pharmaceutical composition, administered before the pharmaceutical composition, administered after the pharmaceutical composition, or a combination thereof. The pharmaceutical composition, anti- CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be administered to the subject multiple times, for example, 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, or more times. When the pharmaceutical composition, anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof may be administered to the subject is administered to the subject multiple times, the subsequent doses may be administered 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, or more seconds; 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, or more minutes; 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, or more hours; 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more days; 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, or more weeks; 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, or more months; or 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, or more years apart.[000141] The cancer may be a solid tumor. The cancer may be any cancer that expresses CSPG4. The cancer may be glioblastoma, melanoma, breast cancer, or any other CSPG4- expressing solid tumor. In some embodiments, a cancer may be brain tumor, e.g., a glioblastoma, a gliosarcoma, or a recurrent brain tumor. In some embodiments, a cancer is a skin cancer, e.g.,a melanoma (e.g., a stage ll-IV melanoma, an HLA-A2 positive melanoma, an unresectable melanoma, or a metastatic melanoma), or a Merkel cell carcinoma. In some embodiments, a cancer is a breast cancer, e.g., a breast cancer that does not express one, two or all of estrogen receptor, progesterone receptor, or Her2 / neu, e.g., a triple negative breast cancer. b. Methods of Modifying a Phagocyte[000142] Provided herein are methods of modifying a phagocyte such as a macrophage. The methods may include: delivering to the phagocyte a plasmid comprising one or more nucleic acid sequences encoding one or more of: an ectodomain comprising a CSPG4 binding domain; a transmembrane domain; and an endodomain described herein.EXAMPLES[000143] The foregoing may be better understood by reference to the following examples, which are presented for purposes of illustration and are not intended to limit the scope of the invention. The present disclosure has multiple aspects and embodiments, illustrated by the appended non- limiting examples.Example 1Materials and Methods[000144] Cell culture. A375, YUMM 1.7, B16F10 cell lines were obtained from the ATCC. Shorttandem repeat confirmed 624-mel and WM793 cell lines were obtained from the Judson-Torres lab at the University of Utah. YUMM1.1, YUMM3.2, and YUMM5.2 cells were generously provided by Matthew Williams’ lab at the University of Utah. A375 cells were cultured in DMEM (THERMO FISHER® #11965118; Thermo Fisher, Waltham, MA) supplemented with 10% FBS. 624-mel and WM793 cells were cultured in RPMI (THERMO FISHER® #11875119; Thermo Fisher, Waltham, MA) supplemented with 10% fetal bovine serum (FBS) (SIGMA ALDRICH® #F4135; Sigma Aldrich, St. Louis, MO) and 1 % Glutamax (THERMO FISHER® #35050061 ; Thermo Fisher, Waltham, MA). YUMM1.1 , YUMM1.7, YUMM3.2, and YUMM5.2 cells were cultured in DMEM / F12 (THERMO FISHER® #11330057; Thermo Fisher, Waltham, MA) supplemented with 10% FBS and 1 % MEM non-essential amino acids (THERMO FISHER #11140050; Thermo Fisher, Waltham, MA). Cell lines were routinely tested for mycoplasma detection using a PCR detection kit (ATCC®, #30-1012K; ATCC, Manassas, VA). Cells were used for <20 passages.[000145] Lentiviral particle production. pCMV-VSV-G, psPAX2, and transgene plasmids were transfected into HEK293FT cells to generate lentivirus media as previously described (Greiner etal., Curr Protoc 2022; 2(8): e514) and concentrated using Lenti-X concentrator (TAKARA BIO® #631232; Takara Bio, Kusatsu, Shiga, Japan) according to the manufacturers protocol.[000146] Generation of fluorescent tumor cell lines. Lentivirus containing pLKO-Lck-mScarlet or pl_enti6-H2B-mCherry was generated from HEK293FT cells as described in ‘Lentiviral particle production’ above. To generate shCSPG4 cell lines, lentivirus containing plasmids from either non-target short hairpin RNA (Sigma #SHC0020) or shCSPG4 (gene target NM_001897) (Sigma #TRCN0000422139 or Sigma #TRCN 0000437747), psPAX2, and pCMV-VSV-G were generated from HEK293FT cells as described in ‘Lentiviral particle production’ above. All short hairpin RNA constructs were cloned into the pLKO.1 backbone. Melanoma cells expressing Lck-mScarlet were transduced using concentrated lentivirus containing the short hairpin RNA.[000147] Human primary cell isolation and CAR generation. Anonymous donor healthy human blood in either leukoreduction filters or apheresis cones were obtained from Associated Regional and University Pathologists, Inc (ARUP, Salt Lake City, UT). Leukocytes were recovered from blood, CD14+ monocyte isolated by adhesion, transduced, and cultured as previously described (Greiner et al., Curr Protoc 2022; 2(8): e514).[000148] Mouse primary cell isolation and CAR generation. Bone marrow monocyte-derived macrophages (BMDMs) were isolated from C57BL / 6 mice femurs as previously described (Weischenfeldt and Porse, CSH Protoc 2008; 2008:pdb.prot5080). BMDMs were cultured in RPMI supplemented with 10% FBS, 1 % Penicillin-Streptomycin, 1 % Glutamax, 1% Sodium Pyruvate (THERMO FISHER® #11360070; Thermo Fisher, Waltham, MA), 0.1% 2- Mercaptoethanol and 50 ng / mL recombinant human M-CSF (PeproTech® #300-25; Thermo Fisher, Waltham, MA). BMDMs were transduced with concentrated lentivirus on day 3 with subsequent media changes every 2 days.[000149] 2D phagocytosis flow assay. Cells were detached, counted, and CAR-Ms were plated with melanoma cells for 24 hours. Cells were detached and processed for flow cytometry. CAR- M phagocytosis percentages were determined by gating live cells, singlets gates, CD11 B+ macrophages, GFP+ for transduced macrophages, and setting an Lck-mScarlet+ / GFP+ gate for phagocytosis based on 0-hour coculture (cells pooled together just prior to flow). See FIG. 3D for full gating strategy.[000150] 2D phagocytosis imaging assay. A375-Lck-mScarlet cells and CAR-Ms were plated together as described in ‘2D phagocytosis flow assay’ above and cultured for 24 hours. Images were acquired on a Zeiss LSM 880 Airyscan® microscope using either the LSM acquisition mode or Airyscan® FAST acquisition mode (and subject to deconvolution using Zen software (Carl Zeiss, Oberkochen, Germany)) with ‘auto’ settings. Maximum intensity projections of images ora central z-plane were analyzed in FIJI (Schindelin et al., Nat Methods 2012; 9(7): 676-82) (version 2.14.0) such that an automated mask was generated around GFP+ macrophages. Lck- mScarlet signal intensity was measured inside the GFP+ mask. Lck-mScarlet signal surface area was manually masked on maximum intensity projection images (FIG. 4E). For images showing x-z, or y-z representations (FIG. 4A, FIG. 11A), images were resliced in FIJI, and then scaled on the z-axis to improve visibility.[000151] 2D timelapse overnight imaging. A375-Lck-mScarlet cells and CAR-Ms were plated together as described in ‘2D phagocytosis imaging assay’ above and cultured for 1 hour to allow the cells to settle. Images were acquired using Airyscan® FAST acquisition mode every 10 minutes for 18 hours and maintained at 37°C and at 5% CO2 with an on-stage incubator.[000152] 3D Phagocytosis How assay. CAR-Ms were plated with Lck-mScarlet or H2B-mCherry transduced melanoma cells for 24 or 72 hours to generate spheroids with 10 μg / mL of IgG control antibody or aCD47. Four spheroids were pooled for each technical replicate, then pelleted, and the supernatant removed. Cells were processed for flow cytometry as described above in ‘2D phagocytosis flow assay’.[000153] Nuclei status assay. Spheroids were prepared as described above in ‘3D phagocytosis flow assay’. Spheroids were then dissociated into single cells by pipetting and plated on glass bottom imaging dishes for 6 hours for cells to adhere. Cells were fixed, then permeabilized and stained with DAPI. Images were acquired on a Zeiss® LSM 880 microscope using LSM acquisition mode. H2B-mCherry cells were manually counted for either unengulfed (i.e., A375-H2B-mCherry cell not surrounded by GFP+ macrophages), engulfed and live (i.e., A375-H2B-mCherry surrounded by GFP+ macrophage in X, Y, and Z plane, with H2B-mCherry colocalizing with DAPI), or engulfed and dead (i.e., A375-H2B-mCherry surrounded by GFP+ macrophage in X, Y, and Z plane, with H2B-mCherry signal dispersed and not colocalizing with DAPI).[000154] 3D spheroid growth assay. A375-H2B-mCherry cells or YUMM1.7-H2B-mCherry cells were plated for 72 hours to generate spheroids. CAR-Ms or control CAR-Ms were then added to the wells along with 10 μg / mL of IgG control antibody or aCD47 for 10 days. Media was changed every 3 days by removing 100 pL of media and replacing it with fresh media and treatments. Spheroids were imaged every 8 hours using an Incucyte® SX5 analysis system, and H2B- mCherry total object integrated intensity and largest object area were quantified using the spheroid analysis module following spectral unmixing using Incucyte® software (version 2020C Rev1).[000155] CAR-M adherence assay. A375-Lck-mScarlet cells were plated as described above in ‘3D spheroid growth assay’. CAR-M were added to the wells along with 10 μg / mL of IgG control antibody or aCD47 for 8 hours. The spheroids were then physically removed from the well with a wide-bore pipette tip, mixed with Matrigel, and then added to imaging dishes. Images were then acquired of the spheroids with on a Zeiss® LSM 880 microscope using Airyscan® FAST acquisition mode. Images were subject to deconvolution using Zen software with ‘auto’ settings and then processed in FIJI to generate a mask of the Lck-mScarlet+ spheroid. The number of GFP+ cells that were attached to the spheroid was then quantified.[000156] CSPG4 expression flow assay. Cells were plated and incubated for 72 hours. Cells were then scraped into fresh media and then pelleted at 300G for 5 minutes. Cells were stained with aCSPG4 antibodies or isotype control antibodies and secondary antibodies on ice, protected from light. Cells were then washed with PBS, then resuspended in an appropriate volume of flow buffer and analyzed for fluorescence expression on the Fortessa. Cells were analyzed for live cells, singlets as described in ‘2D Phagocytosis Flow Assay’ and then gated for CSPG4+ cells against isotype-stained control cells.[000157] Image cytometry. Cells were plated as described above in ‘2D phagocytosis flow assay’ (FIG. 6B and FIG. 6C) or ‘3D phagocytosis flow assay’ (FIG. 9A). After 24 hours of coculture, cells were prepared for flow as described above in ‘2D phagocytosis flow assay’. Cells were stained with DAPI for viability gating. Imaging cytometry was performed on an Imagestream® Mk II (Amnis, Seattle, WA). Cells were gated by Area / Aspect ratio, in-focus, DAPI for viability, total GFP+ events, non-saturated GFP+ events. Internalization events were detected by setting a minimum RFP+ intensity and applying a GFP signal adaptive erode to detect an RFP+ signal within a GFP+ signal. Internalized phagocytosis events were classified by size based on the area intensity of the RFP+ signal (about 75-microns).[000158] Xenograft experiments. Animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Utah and conducted with assistance from the Preclinical Research Shared Resource at Huntsman Cancer Institute Research. 250K A375- H2B-mCherry cells were injected with Matrigel® subcutaneously into 6-8 week-old female NOD.Cg-Rag1tm1Momll2rgtm1Wjl / SzJ (NRG) mice. The subsequent injections were performed peri-tumorally with vehicle and CAR-Ms only in PBS on days indicated. For same-day injections only, at first injection, vehicle or CAR-Ms were injected with the tumor cells in Matrigel®. Tumor volume measurements were taken with calipers as indicated on graphs until experimental endpoints.[000159] Tumor digestion for flow cytometry. Tumors were mechanically dissociated with scalpels, then digested with RPMI (+.032% Aqueous Collagenase D, +.008% DNAse) for 40 minutes at 37°C on a shaker, inverting every 10 minutes. Digested tumor was then strained over 70-micron filters, washed in PBS, and then cells were counted for flow cytometry.[000160] Tumor immunostaining and imaging. Tumors were harvested and fixed in formalin prior to embedding in paraffin. Samples were deparaffinized and rehydrated using Citrisolv® (#89426-268, VWR, Radnor, PA), followed by rinses in sequential dilutions (100%, 95%, 80%, 70%) of ethanol, and then ddH2O and PBS. Antigen retrieval was done in Tris-EDTA (pH 9) buffer overnight at 60°C, and then washed 2x with ddH2O and PBS. Samples were blocked 1 hour, followed by primary chicken anti-GFP (ABCAM® #13970; Abeam, Cambridge, United Kingdom) (1 :500) for 1 hour, all at room temperature. Samples were washed in TBS-T, and then stained for 1 hourwith Goat Alexa fluor® 488 anti-Chicken (1 :500) (#103-545-155, Jackson Immuno, West Grove, PA) at room temperature. Autofluorescence was then quenched using 0.1 % Sudan Black in 70% ethanol for 10 minutes at room temperature, and then stained with DAPI prior to mounting and imaging. Images were acquired with a Zeiss® LSM880 microscope using LSM acquisition mode. E0771 tumors grown in mice expressing GFP-tagged mitochondria in macrophages were used (mice carrying both of the following transgenes: B6.Cg-Gt(ROSA)26Sortm1 (CAG- EGFP)Brsy / J (#032290, Jackson Laboratory, Bar Harbor, ME) and a B6.129P2-Lyz2tm1 (cre)lfo / J (#004781 , Jackson Laboratory, Bar Harbor, ME)) to set a threshold for positive GFP expression. [000161] Image analysis. All images were acquired with a Zeiss® LSM 880 microscope or an Incucyte® SX5 system. For images acquired with the LSM 880, selected z-planes were used to generate maximum intensity projections using Zen software. Linear adjustments to brightness and contrast were done with FIJI. Representative images were cropped and assembled with Adobe® Photoshop® (version 24.4.1) and Illustrator® (version 27.5).[000162] Bioinformatic analysis. CSPG4 normalized expression values from fresh resected healthy human skin and metastatic melanoma specimens were obtained from previously published single-cell RNA-sequencing (scRNAseq) data sets GSE151091 and the Single Cell portal (portals.broadinstitute.org / single_cell / study / melanoma-immunotherapy-resistance) (Belote et al., Nat Cell Biol 2021 ; 23(9): 1035-47; Jerby-Arnon et al., Cell 2018; 175(4): 984-97). Both studies were analyzed using SMART-seq® 2 based pipelines and then compared using rank mean normalization. Only non-cycling healthy human skin cells were included in the analysis. Immunotherapy response data was obtained as in Riaz et al (Riaz et al., Cell 2017; 171(4): 934- 49), via the Immuno-genomic atlas for immune checkpoint blockade-based cancer immunotherapy (bioinfo.vanderbilt.edu / database / Cancer-lmmu / ) (Yang et al., Cancer Res 2021 ;82(4): 539-42). Data visualization and statistical analyses were performed in GraphPad® Prism® (version 9.4.1), python (version 3.7.4), and R (version 4.3.2).[000163] Statistical analysis. All statistical analysis was done using GraphPad® Prism® and presented as mean values ± standard error of the mean (SEM). Outliers greater than two standard deviations were removed from analysis. The figure legend indicates the statistical test used and the number of biological and technical replicates. To determine the number of biological replicates, it was sought to detecta 50% difference in phagocytosis between conditions, assuming a 10% standard deviation. Thus, a minimum of 3 biological replicates was required for the studies herein. Flow cytometry data were analyzed using FlowJo® software (version 10.10.0). Because animal experiments were not performed across both male and female mice, sex was not addressed as a biological variable in these studies.Example 2CSPG4-targeting CAR-Ms phagocytose CSPG4-expressing cancer cells[000164] CSPG4 was first characterized as overexpressed in melanoma; however, CSPG4 upregulation in numerous other cancers including breast and glioblastoma has since been described. CSPG4 has been reported to promote tumor growth and survival through numerous signaling pathways associated with its extracellular domains. Transcriptional analysis suggests that CSPG4 mRNA expression is low in healthy tissue, with undetectable protein expression levels by immunohistochemistry. Published single-cell RNA-sequencing data from fresh healthy human skin and metastatic melanomas were analyzed (Belote et al., Nat Cell Biol 2021 ; 23(9): 1035-47; Jerby-Arnon et al., Cell 2018; 175(4): 984-97; Tirosh et al., Science 2016; 352(6282): 189-96). These analyses revealed significantly higher CSPG4 transcript abundance in a majority of malignant tumor cells from patient samples compared to healthy skin cells, including melanocytes (FIG. 1A; FIG. 2A). Importantly, non-malignant cells, both in the tumor microenvironment (FIG. 2B) and across broader cell types (FIG. 2C), generally presented low CSPG4 transcript counts, with notable exceptions being Sertoli cells, oligodendrocyte precursors cells, and a population of smooth muscle cells. However, immunohistochemistry approaches do not reveal high levels of CSPG4 protein expression in oligodendrocyte precursor cells in adult brain, and the blood-testis barrier represent one of the tightest blood-tissue barriers in the body, protecting Sertoli cells. In pre-clinical studies of rats treated with a cross-reactive rat IgE antibody against CSPG4, rats did not exhibit gross abnormalities or long-term toxicity, suggesting that CSPG4-targeting strategies do not yield significant off-target effects.[000165] Next, the correlation between ICB therapy outcomes and CSPG4 expression was determined. Since ICBs targeting the PD-1 / PDL-1 axis is the most common immunotherapy in melanoma patients, the expression in aPD-1 non-responders was analyzed. It was found that patients who did not respond to aPD-1 treatment correlated with higher CSPG4 expression (FIG. 2D). Furthermore, aPD-1 treatment itself did not change CSPG4 expression (FIG. 2E). These results suggest that CSPG4 is a good target for melanoma patients who did not respond well to ICB treatment.[000166] To target CSPG4-expressing cells, CAR constructs were engineered incorporating one of three distinct single-chain variable fragments (scFv) exhibiting affinity for CSPG4: 225.28, SK5, and 763.74, each of which recognizes different extracellular domains (FIG. 3A). Previous CAR designs were followed (Morrissey et al., Elife 2018; 7) using an FcRy signaling domain, and this domain was combined with CSPG4 scFvs. As controls, CARs lacking the intracellular FcRy phagocytic signaling domain were designed, as well as a CAR expressing GFP alone, to control for activation associated with transduction (FIG. 1C). To generate CAR-Ms, human primary blood monocyte-derived macrophages were used, which were isolated from healthy blood donors, transduced monocytes with CSPG4-targeting CARs, and then differentiated the monocytes into macrophages (FIG. 1 D). Using Applicant’s previously established protocols (Greiner et al., Curr Protoc 2022; 2(8): e514), CSPG4-CAR expressing macrophages were generated with high transduction efficiencies (FIG. 3B), and thus primary macrophage-derived CAR-Ms were used for all subsequent studies.[000167] The human metastatic melanoma cell line A375 was used to test whether CSPG4- targeting CAR-Ms phagocytose melanoma cells. A375 cells exhibit high CSPG4 expression, compared with primary macrophages from multiple donors as a negative control (FIG. 3C). A375 cells were engineered to express a fluorescent membrane marker ( Lek- m Scarlet) to visualize the melanoma cells. Then, A375-Lck-mScarlet cells were cocultured with each of the three different CSPG4-targeting CAR-Ms. Using flow cytometry to quantify CAR-M-mediated phagocytosis as the percentage of CAR-Ms with A375-Lck-mScarlet fragments (FIG. 3D), it was found that all CAR-MαCSPG4(225-28 / SK5 / 763-74)phagocytosed melanoma cells at higher rates than the control CAR- MGFP (FIGS. 1 E-G). These differences in phagocytosis rates were not due to the level of CAR transduction efficiency, as the percent CAR transduction efficiency in macrophages did not affect the percent of CAR-M-mediated phagocytosis (FIG. 3E). To confirm these flow cytometric events were bona fide CAR-M-mediated phagocytosis, rather than A375 membrane fragments adhering to the exterior of CAR-Ms, CAR-M and A375 cocultures were imaged with high-resolution microscopy. After 24 hours of coculture, it was observed that there were Lck-mScarlet punctaeinside of CAR-MαCSPG4(225.28 / SK5'763.74)cells compared to CAR-MGFP(FIG. 1 H; FIG. 3F). CAR- l\ / lacspG4(763.74) (herein referred to as CAR-MαCSPG4) were used for the remainder of the studies herein due to the consistency of phagocytosis, availability of published data regarding its binding region on CSPG4, and its current use in phase I CAR-T cell clinical trials (NCT06096038). To better understand CAR-M-mediated phagocytosis of melanoma cells, timelapse recording and 3D visualization approaches were used. Using 3D reconstruction, it was confirmed that the Lck- mScarlet punctae were fully internalized in the CAR-MαCSPG4(FIG. 4A). Live timelapse recordings reveal that CAR-MαCSPG4began phagocytosing melanoma cells within the first 4 hours of coculture, with most events being trogocytosis (cell nibbling), rather than whole cell phagocytosis (FIG. 4B).[000168] Next, it was tested whether the intracellular FcRy region was critical for CAR-M- mediated phagocytosis. Consistent with previous studies with CD19-targeting CAR-Ms, removal of the FcRy domain (aCSGP4AFcRy) resulted in a statistically significant reduction in the percent phagocytosis by flow cytometry compared to the full CSPG4-targeting CAR-M (FIG. 11). Interestingly however, CAR-MαCSGP4AFcRvexhibited higher phagocytosis rates than control CAR- MGFP(FIG. 11). It was hypothesized that the physical interaction between the CSPG4-scFv and CSPG4 on the surface of melanoma cells is sufficient to promote CAR-M-mediated phagocytosis, likely via the maintenance of endogenous FcRy signaling. Live cell imaging of CAR-MαCSGP4AFcRv cocultures with A375-Lck-mScarlet cells also show higher rates of phagocytosis of melanoma cells compared to CAR-MGFP(FIG. 4C and FIG. 4D). Taken together, these results suggest that CSPG4-targeting CAR-Ms efficiently phagocytose metastatic melanoma cells.[000169] Next, it was sought to determine the specificity of CSPG4-CAR-M-mediated phagocytosis. Using human melanoma cell lines with differing CSPG4 surface expression, the level of cancer cell phagocytosis by CAR-MαCSGP4and CAR-MGFPwas quantified. It was found that A375 cells and WM793 melanoma cells exhibited high levels of CSPG4 expression on the surface of the cells, and 624-mel melanoma exhibited low CSPG4 expression (FIG. 5A and FIG. 5B). When each cell line was cocultured with control CAR-MGFP, it was observed that there were no differences in CAR-M-mediated phagocytosis. However, when each of these cell lines was cocultured with CAR-MαCSPG4, it was observed that there were high rates of phagocytosis with A375 and WM793 cells, but not with 624-mel cells (FIG. 5C). To further confirm specificity, CSPG4 expression was knocked down in A375 cells. A375 surface expression was knocked down with two different CSPG4 shRNAs to ~40% compared to non-targeting controls (FIG. 5D and FIG. 5E) and it was found that the percent of CAR-MαCSPG4phagocytosis was significantly reduced when CSPG4 expression was decreased on A375 cells compared to non-targetingcontrols (FIG. 5F). These results suggest that CAR-MαCSPG4phagocytosis is specific to CSPG4- expressing cells.Example 3CSPG4-targeting CAR- Ms alone do not cause reduced melanoma cell survival[000170] It was next determined whether CAR-MαCSPG4phagocytosis affected melanoma cell survival. First, the amount of melanoma cell trogocytosis was quantified since nibbling of target cells often does not result in cell death and this rate was compared to whole cell phagocytosis. The median area of Lck-mScarlet melanoma cell fragments internalized in CAR-MαCSPG4cells was significantly lower (65 μm2) than the median area of a whole, unengulfed Lck-mScarlet melanoma cell (473 μm2) (FIG. 4E), suggesting that trogocytosis events were predominantly being imaged and not whole cell phagocytosis events. Next, image-based flow cytometry was used, where both complete internalization could be masked, and a relative size of the internalized fragment across thousands of cells could be quantified (FIG. 6A and FIG. 6B). Based on the engulfment size measurements (FIG. 4E), phagocytic events were stratified as CAR-Ms with large melanoma fragments (> 75 μm2) versus CAR-Ms with small melanoma fragments (< 75 μm2). It was observed that both had a higher percent of overall phagocytosis, as well as a higher proportion of CAR-Ms with large melanoma fragments in CAR-MαCSPG4compared to control CAR-MGFP(FIG. 6C and FIG. 6D; FIGS. 7A-G). To specifically quantify whole-cell phagocytosis, A375 cells were engineered to express a fluorescently-tagged histone H2B in the nucleus (H2B-mCherry), allowing for measurement of nuclear engulfment as a read-out for whole-cell phagocytosis. Consistent with the image-based flow cytometry observations, this analysis revealed increased levels of whole-cell phagocytosis by CAR-MαCSPG4versus CAR-MαCSGP4AFcRv or CAR-MGFPcontrols (FIG. 6E). However, the overall level of whole-cell phagocytosis was low, at less than 2%. Then, melanoma cell survival was analyzed as the proportion of melanoma cells remaining in the coculture population and it was observed that there were no significant changes in the overall percentage of melanoma cells in the coculture population across all conditions (FIG. 6F). These results suggest that although CAR-MαCSPG4exhibited higher rates of whole cell melanoma phagocytosis compared to control CAR-Ms, these phagocytic events were not sufficient to reduce melanoma cell survival.Example 4Combining CSPG4-targeting CAR-Ms with aCD47 inhibits melanoma growth in 3D[000171] Puzzled by the lack of change in melanoma cell survival in the system above, the effects of CAR-Ms were tested in a more physiologically relevant 3D environment as macrophages in suspension compared to those adhered on tissue culture dishes exhibit a much greater capacity for phagocytosis. Many cancer cells, including melanoma, overexpress CD47 on their surface, a “don’t eat me” signal. Melanoma expressed CD47 interacts with SIRPα on macrophages, protecting the melanoma cells from phagocytosis. Blocking CD47 has been shown to inhibit A375 tumor growth in mice. Therefore, it was hypothesized that elevated expression of CD47 in A375 cells inhibits CAR-MαCSPG4-mediated whole cell phagocytosis. First, CAR- M / melanoma spheroids were cultured (FIG. 8A) using A375 cells expressing Lck-mScarlet and phagocytosis was quantified with image-based flow cytometry after 24 hours. In 3D cultures with CAR-MαCSPG4, a high level of melanoma phagocytosis to almost 50% at 24 hours, with a higher proportion of CAR-Ms with large, internalized melanoma fragments were observed when Magrolimab (Hu5F9-G4), a humanized lgG4 monoclonal antibody that blocks the CD47 / SIRPα axis, was added to the cultures compared to IgG controls (FIG. 9A). Control CAR-MGFPdid not show altered phagocytosis of melanoma cells in the presence of aCD47 (FIG. 9A). These results suggest that in 3D cultures, CAR-MαCSPG4and aCD47 work cooperatively to enable phagocytosis. To test between trogocytosis and whole-cell phagocytosis, melanoma cells with fluorescently tagged nuclei (H2B-mCherry) were used. It was observed that there was a significant increase in the percent of CAR-M-mediated whole cell phagocytosis by CAR-MαCSPG4compared to control CAR-MGFP, and that this whole cell phagocytosis was further increased to ~35% with aCD47 (FIG. 8B). Excitingly, it was found that the proportion of melanoma cells in the final spheroid was significantly reduced in the presence of CAR-MαCSPG4with aCD47 (FIG. 8C). To determine whether the lack of melanoma cells was due to CAR-M-mediated melanoma cell death, the 3D spheroid experiments were performed and then the spheroids were dissociated to visualize the cells with high-resolution microscopy. Dead cells exhibit ruptured and fragmented nuclei and histones become separated from DNA during apoptosis; thus, the number of internalized melanoma cells with ruptured H2B-mCherry nuclei was quantified as a measure of cell death. It was observed that a population of A375-H2B-mCherry cells that were not phagocytosed by CAR- Ms (“unengulfed”); a population of A375-H2B-mCherry cells that were engulfed by CAR-Ms, but the nuclei remained intact with the H2B-mCherry signal colocalizing with DAPI (“engulfed+live”); and a population of A375-H2B-mCherry cells that were engulfed by CAR-Ms with their nuclei broken down and H2B-mCherry no longer localized with DNA, indicated cell death (“engulfed+dead”). It was found that a significantly higher proportion of A375 cells exhibited the“engulfed+dead” phenotype in cultures with CAR-MαCSPG4and aCD47, compared to all other conditions (FIG. 8D and FIG. 8E; FIGS. 9B-D).[000172] Interestingly, the reduction in melanoma cells and robust CAR-M phagocytosis with aCD47 was only observed in 3D culture conditions and not observed in 2D cultures (FIGS. 9E-I). These results suggest that the combination of CAR-MαCSPG4with aCD47 in 3D leads to high levels of melanoma cell death.[000173] To better mimic tumor growth conditions, the 3D approaches were altered and melanoma spheroids were allowed to form prior to addition of CAR-Ms (FIG. 10A). After 4-8 hours of CAR-M addition, the spheroids were removed from the U-bottom low-attachment plates and embedded in Matrigel to visualize CAR-M interactions with the melanoma spheroid. It was found that CAR-MαCSPG4remain adhered to the melanoma spheroids when spheroids were transferred to Matrigel, whereas the CAR-MGFPdid not (FIG. 10B and FIG. 10C). Adding aCD47 to the culture did not affect CAR-M attachment to the spheroid. Using high resolution imaging, it was further observed that CAR-MαCSPG4infiltrated the spheroid (FIG. 11 A) and phagocytosed melanoma cells within the spheroid (FIG. 11 B).[000174] Next, it was tested how CAR-Ms affect melanoma spheroid growth over time. Melanoma spheroids were grown for 72 hours, CAR-Ms were added, and then spheroids were allowed to continue to grow for 10 days in the U-well low-attachment plates (FIG. 10A). Spheroid growth was monitored over time and a synergistic effect was observed in which combining CAR- MαCSPG4wjth aCD47 resulted in robust inhibition of tumor spheroid growth (FIG. 10D and FIG. 10E; FIG. 12A and FIG. 12B). Notably, CAR-MαCSPG4infiltrated the melanoma spheroid; whereas, CAR-MGFPlocalized primarily to the exterior on the melanoma spheroid (FIG. 12C and FIG. 12D), suggesting increased interactions between spheroids and CSPG4-targeting CAR-Ms versus controls. Next, it was sought to determine whether this inhibition of spheroid growth was specific for cells expressing CSPG4; however, consistent with published literature, the A375 cells expressing the shCSPG4 clones failed to form viable spheroids and disassemble during media changes, thus precluding analysis (FIG. 12E and FIG. 12F). Taken together, these results suggest that combining CSPG4-targeting CAR-Ms with aCD47 approaches result in robust inhibition of melanoma growth in 3D by increased phagocytosis.[000175] The dramatic synergistic effects of combining CSPG4-targeting CAR-Ms with aCD47 led to the question whether the concentration of aCD47 could be reduced, and still result in a robust CAR-M-mediated phagocytosis. Reducing the aCD47 concentration, while still showing efficient CAR-M-mediated phagocytosis of cancer cells, may reduce off-target effects when used in patients given that almost all healthy cells express some level of CD47. 10 μg / mL of aCD47was initially used as used in previous work (Gholamin et al., Sci Transl Med 2017; 9(381 ); Zhang et al., PLoS One 2016; 11 (4): e0153550) and a serial dilution series was performed. Combining aCD47 with control CAR-MGFPshowed no significant increase in melanoma phagocytosis until 10 μg / mL of aCD47 (FIG. 12G; white bar at 104compared to IgG control). At this aCD47 concentration, it was also observed that a significant increase in CAR-MαCSPG4-mediated phagocytosis compared to CAR-MGFPcontrol (FIG. 12G; black bar vs white bar at 104). This level of CAR-MαCSPG4-mediated phagocytosis did not decrease until the concentration of aCD47 was reduced to 100 ng / mL aCD47 (FIG. 12G; comparing black bars at 104to 102). These results suggest that significantly decreased concentrations of aCD47 still enhance CSPG4-CAR-M- mediated phagocytosis of melanoma cells, but not with CAR-MGFP.Example 5CSPG4-targeting CAR-Ms inhibit melanoma growth in vivo[000176] Given the above in vitro findings, it was sought to use a syngeneic system using murine melanoma cells and murine bone marrow-derived macrophages. Expression of CSPG4 was measured in several mouse melanoma cell lines by flow cytometry and high CSPG4 expression was observed (FIG. 13A) in the YUMM1.7 and YUMM1.1 cell lines (Meeth et al., Pigment Cell Melanoma Res 2016; 29(5): 590-7). Therefore, YUMM1.7 cells expressing H2B- mCherry were generated. Then, it was determined which CSPG4 scFv to use for these murine studies. The 763.74 CSPG4 scFV has no demonstrated reactivity with the murine CSPG4 homolog, likely due to a two amino acid difference between human and mouse CSPG4 in the binding epitope (FIG. 13B). Conversely, the 225.28 binding epitope only has a one amino acid difference between human and mouse CSPG4 and has previously demonstrated some crossspecies reactivity. Thus, it was first tested whether CAR-Ms generated with the 225.28 scFv phagocytosed murine YUMM1.7-H2B-mCherry melanoma cells in 3D (FIG. 13C). CAR- MαcspG4(225.28)exhibited a low rate of phagocytosis, with only ~4% of CAR-MαCSPG4(22528> phagocytosing YUMM1.7-H2B-mCherry cells (FIG. 13C), a rate similarly observed in CAR- MαcspG4(763.74)anc|control CAR-MGFPcells. These results suggest that CAR-MαCSPG4(225-28)do not effectively target murine CSPG4, and thus precluded the ability to test the function of CAR- MαcspG4(225.28) jnsyngeneic mouse models. Therefore, human CAR-Ms and human melanoma cells in immune-compromised models were tested.[000177] It was determined whether CAR-MαCSPG4(76374)inhibits A375 human melanoma growth in immune-compromised NRG mice. CAR-Ms were injected peri-tumorally, rather than systemically, to maximize on-target effects, while minimizing potential off-target effects in othertissues associated with intravenous administration. First, it was determined how long CAR-Ms are detectable in the tumor to determine the treatment regimen. It was reasoned that the CAR- Ms did not need to stay in the tumor for the entirety of the experiment, but CAR-Ms must be present in the tumor long enough to phagocytose CSPG4-positive melanoma cells and potentially reprogram the tumor environment for a sustained response. Thus, 400K A375 cells were injected into the right flank of 6-8 week-old female NRG mice, tumors were given time to become palpable, and then 1M CAR-Ms were peri-tumorally injected on day 9. Then, mice were euthanized on days 12, 14, 16, and 19 to determine whether CAR-Ms were present in the tumor by flow cytometry (FIG. 14A). It was found that CAR-Ms were present at 1 % of the total live cells in the tumor after 3 days post-injection (day 12), and that this number dropped to almost zero starting at 5 days post-injection (day 14) (FIG. 14B). Thus, a two-stage CAR-M injection strategy was used, 3-5 days apart.[000178] First, subcutaneous same-day injections of A375-H2B-mCherry cells and CAR-Ms into the flanks of mice were tested. 400K A375-H2B-mCherry cells and 1 M CAR-MαCSPG4, CAR-MGFP, or vehicle control were injected into the mice and then the mice were re-injected with only CAR- Ms or vehicle control again 5 days later (FIG. 14C). It was found that there were no differences in animal weight across conditions, suggesting no gross off-target toxicity (FIG. 15A). It was found that treating mice with CAR-MαCSPG4inhibited the growth of the tumor compared to CAR- MGFPcontrols, although these differences were not statistically significant (FIG. 14D, FIG. 15B and FIG. 15C). As expected, treatment with control CAR-MGFPalso inhibited melanoma growth compared to vehicle controls (FIG. 15B and FIG. 15C), suggesting that the addition of any exogenous macrophages alone affects tumor growth. Together these results suggest that CSPG4-targeting CAR-Ms may reduce melanoma growth in vivo.[000179] Next, CAR-M treatment was modeled following tumor growth, and thus moved beyond same-day injection studies. A375-H2B-mCherry cells were injected into NRG mice, and first the tumors were allowed to engraft for 5 days (FIG. 14E). Then, CAR-MαCSPG4or controls were peritumorally injected (FIG. 14F) on day 5 and again on day 9. Again, it was found that there were no differences in animal weight across conditions (FIG. 15D). Excitingly, it was found that CAR-MαCSPG4-treated animals exhibited a significant reduction in tumor growth compared to CAR- MGFP-treated animals (FIG. 14G; FIG. 15E and FIG. 15F), and that this inhibition of tumor growth began at day 22 and was maintained through day 27, well after CAR-Ms are no longer detected in the tumor.[000180] Immunofluorescence of the tumors at day 27 confirmed that majority of CAR-Ms were no longer present at the tumor (FIG. 15G), suggesting that the sustained inhibition of melanomagrowth is likely mediated by reprogramming of the myeloid cells present in the tumor microenvironment. These results show that first, CSPG4-targeting CAR-Ms effectively inhibit melanoma growth in vivo, and second, the inhibition of tumor growth occurs in the absence of adaptive immune cells, suggesting the innate immune system is sufficient for this response.Example 6[000181] Current therapies for treating advanced stages of melanoma have improved patient survival, but primary resistance is still a relatively common occurrence, as well as recurrence via acquired resistance. Pro-tumorigenic macrophages constitute a large portion of the tumor microenvironment in melanoma. Since CAR-T cells have been unsuccessful at infiltrating solid tumors, it was hypothesized herein that CAR-Ms may effectively target melanoma tumors. In this study, it was unveiled that a potential treatment strategy for melanoma is by CSPG4-targeting CAR-Ms. The findings herein demonstrate that CSPG4-targeting CAR-Ms efficiently phagocytose metastatic melanoma cells. When CSPG4-targeting CAR-Ms were combined with CD47 blocking antibodies, thereby blocking a key tumor cell “don’t eat me” signal, it was observed herein that there was enhanced phagocytosis of melanoma cells and robust inhibition of melanoma spheroid growth in 3D. Importantly, it was shown herein that CSPG4-targeting CAR- Ms control tumor growth in vivo. Herein, it was demonstrated that in a xenograft model, with mice lacking adaptive immune cells, CSPG4-targeting CAR-Ms exhibited sustained inhibition of melanoma growth. These data support a growing body of work presenting CSPG4 as a potential therapeutic target, and that CAR-Ms may be effective at clearing cancer cells that overexpress CSPG4, particularly when combined with aCD47.[000182] The findings herein raise important questions regarding how long CAR-Ms must persist at the tumor to maintain tumor control. If CAR-M-mediated phagocytosis of cancer cells is the primary mechanism for anti-tumor activity, then both sustained persistence and delivery of high CAR-M numbers is critical. Alternatively, if tumor immune reprogramming by CAR-Ms is the primary mechanism for anti-tumor activity, transient CAR-M persistence at the tumor may be sufficient for tumor control. Both mechanisms likely contribute to anti-tumor activity, and further work is required to parse out the contributions of each of these mechanisms. The data herein show that CAR-Ms are not detectable in the melanoma tumor after approximately 5 days. Despite the short-term CAR-M persistence in the tumor, as in this study, CAR-Ms targeting multiple tumor antigens across a variety of models maintain tumor control. These results and the results herein suggest that CAR-M-mediated phagocytosis of cancer cells is not sufficient to prevent tumor growth alone, and likely requires contributions from other stromal cells. Provided herein isevidence that CSPG4-targeting CAR-Ms inhibit melanoma growth in immune-compromised animals over 18 days from the last CAR-M injection, well after CAR-Ms are no longer detected in the tumor. These results suggest that innate immune cells are sufficient to inhibit melanoma growth in animals treated with CSPG4-targeted CAR-Ms, likely by local reprogramming of the tumor-associated macrophages into anti-tumor macrophages, thereby increasing phagocytosis of melanoma cells by endogenous macrophages, and / or by decreasing immunosuppressive signals at the tumor and promoting general anti-tumor responses. It is expected that while shown herein that innate immune cells inhibit melanoma growth, CAR-Ms likely also regulate the adaptive immune response. Future CAR-M studies must perform systematic immune profiling at multiple time points over the growth of the tumor, dissecting each immune cell subset, their activation patterns, and exhaustion markers to provide a more in-depth understanding of the dynamics driving immune cell reprogramming in the tumor during CAR-M therapies. Furthermore, the number of CAR-Ms required, the injection frequency, as well as whether local versus systemic CAR-M injections provide better anti-tumor responses need to be systematically tested and compared.[000183] Teasing apart the requirement for CAR-M persistence at the tumor and CAR-M- mediated immune reprogramming will guide the development of next-generation CAR-Ms and combination therapies. It is hypothesized herein that increasing CAR-M-mediated phagocytosis of cancer cells is a critical aspect of anti-tumor activity for three reasons: 1 ) increase cancer cell cytotoxicity directly through phagocytosis; 2) facilitate the repolarization of tumor-associated macrophages into anti-tumor macrophages; 3) increase antigen presentation to activate the adaptive immune arm. Thus, efforts herein were focused on improving CAR-M-mediated phagocytosis of melanoma cells by engineering the CSPG4-targeting chimeric antigen receptor to contain the intracellular FcRy phagocytic signaling domain and combining the CSPG4-targeting CAR-Ms with CD47 blocking antibodies. Future approaches to improve phagocytosis include using activation domains more specific to the engulfment machinery such as the Rac2E62K mutation and targeting additional “don’t eat me” signals. If sustained CAR-M persistence at the tumor is deemed to be critical for tumor growth control, then efforts to improve CAR-M survival such as overexpression of cytokines like M-CSF may be needed. Another approach to improve CAR-M phagocytosis activity is to directly polarize CAR-Ms to a pro-inflammatory phenotype prior to administration. This approach would have the benefit of immediately releasing cytokines to dampen the immunosuppressive environment of the tumor; however, it remains unknown whether this approach increases the potential for dangerous side-effects such as cytokine release syndrome. Alternative approaches include combining CAR-Ms with ICB treatments. Due to theantigen presentation capabilities of macrophages, the strength of CAR-Ms may lie in their ability to phagocytose cancer cell targets and display additional tumor antigens to T-cells as a mechanism to overcome tumor heterogeneity. Combining CSPG4-CAR-Ms with ICB approaches may represent a powerful approach to tackle the long-standing challenge of tumor heterogeneity particularly in melanoma. Preliminary results suggest that combining CAR-Ms with ICB therapies can have an additive effect in reducing tumor burden, however if it is unclear how this combination affects tumor immune populations or ICB resistance. The work herein provides compelling evidence that CSPG4-targeting CAR-Ms successfully reduces primary melanoma growth across in vitro and in vivo melanoma models. The next critical steps involve determining whether local administration of CSPG4-targeting CAR-Ms at the time of melanoma tumor resection prevents local melanoma recurrence. Furthermore, future work is required to determine whether systemic versus local administration of CSPG4-targeting CAR-Ms prevent melanoma metastases.Example 7Combining CSPG4-CAR-Ms with anti-CD47 approaches leads to reduced melanoma growth and increased survival[000184] As described above, it was tested whether CSPG4-targeting CAR-Ms can prevent tumor growth in part by CAR-M-mediated phagocytosis of cancer cells, but also by reprogramming the tumor immune environment, which would lead to long-lasting anti-tumor responses. It was hypothesized herein that phagocytosis of melanoma cells by CSPG4-targeting CAR-Ms allow for increased cross-presentation to T-cells and reprogramming of tumor- associated macrophages. Thus, to test this hypothesis, immune-competent mouse models were used. A different CSPG4-targeting CAR-M was also used - switching to CSPG422528-CAR-Ms - as the 228.25 CSPG4-scFv is a murine clone, with cross-reactivity with human cells. A murine melanoma cell line, YUMM1.7, was also used which has high expression of CSPG4 on the cell surface (FIG. 16A). 2.5e5murine melanoma YUMM1.7-mScarlet cells subcutaneously into the right flank of 6-8-week-old C57BL / 6J male mice. Then, bone marrow-derived macrophages were transduced with CSPG422528-scFv with a transduction efficiency of >95% (FIG. 16B). Five days after YUMM1 .7 implantation, 1e6CSPG422528-CAR-Ms were injected peri-tumorally every 2 days for 6 days (3 injections total). Consistent with our xenotransplantation results above, it was found that injection of CSPG4-targeting CAR-Ms inhibited the growth of murine melanoma tumors compared to GFP-CAR-Ms and vehicle-injected mice (FIG. 16C and FIG. 16E). The mice were culled early in the experiment (at day 10) to profile the immune compartments in the tumor. First, differences in the proportion of general immune cell populations were surveyed by gating forlive / dead, and then CD45+ cells (changes in the overall CD45+ percentage across conditions were not observed), and then further analyzing CD3+ (T-cells), B220+ (B-cells), Ly6C+ (monocytes), CD11 c+ (dendritic cells), and CD11 b / F480+ (macrophages). Even at this early time point, differences in the proportion of immune cells in tumors injected with CSPG4-targeting CAR- Ms versus GFP control were observed (FIG. 16D), with a subtle, but significant, increase in the number of T-cells and a significant decrease in macrophages in CSGP4-CAR-M-injected tumors compared to control. This profiling was performed only 4 days after the last CAR-M injection and suggest the initiation of tumor immune reprogramming even at this early time point, a hypothesis that will be tested further. Together these results suggest a working model whereby CSPG4- targeting CAR-Ms efficiently target and inhibit growth of melanoma tumors, in part by reprogramming the tumor environment.Example 8Combining CSPG4-targeting CAR-M with anti-CD47 inhibits melanoma growth and increases survival in vivo[000185] While the inhibition of tumor growth by CSPG4-targeting CAR-Ms herein was significant, the tumors did still grow. Therefore, whether combining CSPG4-targeting CAR-Ms with anti-CD47 approaches will be tested to see if the treatments synergistically inhibit tumor growth and increase animal survival. To determine the concentration of CD47 blocking antibodies (Hu5F9-G4) to use for these experiments, 3D phagocytosis experiments were performed where the concentration of CD47 blocking antibodies were decreased in combination with CSPG4- targeting CAR-Ms. It was reasoned herein that decreasing the concentration of anti-CD47, while still showing efficient phagocytosis of cancer cells, may reduce off-target effects when used in patients. 10 μg / mL of CD47 blocking antibodies was the starting concentration and a serial dilution series was performed. Combining anti-CD47 with control GFP-CAR-Ms showed no significant increase in melanoma phagocytosis until 10 μg / mL of anti-CD47 (FIG. 17A, white bar at 104compared to IgG control). At this anti-CD47 concentration, it was also observed that there was a significant increase in CAR-M-mediated phagocytosis (FIG. 17A, black bar vs white bar at 104). This level of CSPG4-CAR-M-mediated phagocytosis did not decrease until the concentration of anti-CD47 was reduced to 10 ng / mL anti-CD47 (FIG. 17A, comparing black bars at 104to 101). These results suggest that the concentration of anti-CD47 can be significantly decreased, and robust CSPG4-CAR-M-mediated phagocytosis of melanoma cells can still be observed. Thus, in vivo experiments in NRG mice were initiated with lower anti-CD47 concentrations. When 1e6CAR-Ms were peritu morally injected, 5 mg / kg CD47 blocking antibodywas also intraperitoneally injected (which is less than half the concentration as used in previous in vivo studies, which ranges from 10-20 mg / kg). It was found that combining CSPG4-targeting CAR-Ms with anti-CD47 approaches further decreased melanoma growth (FIG. 17B, dark gray hatched line), but not more than anti-CD47 approaches alone (FIG. 17B, black hatched line, overlapping with dark gray hatched line). These results are interpreted as meaning that the concentration of anti-CD47 was still too high to determine whether combining anti-CD47 with CSPG4-targeting CAR-Ms additively or synergistically inhibit melanoma growth. Thus, these experiments will be continued by first performing dose-dependent anti-CD47 studies (0.5, 1.25, 2.5, and 5 mg / kg) to determine an anti-CD47 concentration that only modestly reduces tumor growth such that it can be tested whether combining anti-CD47 with CAR-Ms exhibit synergistic or additive effects on tumor growth. Tumor growth will be monitored over time using caliper measurements as performed previously, and mouse survival studies will be performed separately. These experiments will also be performed in both male and female mice to determine any sexspecific differences. As an orthogonal approach, similar studies will be performed in immune- competent C57BL6 / J mice and YUMM1.7 murine melanoma cells as described in FIGS. 16A-E. In these immune-competent mouse models, CAR-Ms generated from bone marrow-derived macrophages will be used and combined with mouse anti-CD47 approaches (BioXCell, Lebanon, NH; clone # MIAP410).Example 9Evaluation of the infiltration and spatial organization of CSPG4-CAR-Ms in tumors[000186] It was predicted herein that CAR-Ms will infiltrate solid tumors as macrophages are known to easily penetrate matrix barriers, and it was hypothesized herein that the presence of anti-CD47 may further enhance CAR-M infiltration. Thus, the extent of CSPG4-CAR-M infiltration into tumors using histological approaches will be quantified. Using both immune-compromised and immune-competent mouse models, tumors will be treated with peritumoral injections of CSPG4-CAR-M with and without intraperitoneal injections of anti-CD47, wait 2 days, and then the tumors will be processed for immunohistochemistry. Paraffin-embedded tumor sections will be stained with anti-RFP to quantify mScarlet+ cancer cells, and the number of tumor-infiltrated CSPG4-CAR-Ms (CD11 B+ or F480+, and GFP+) will be quantified. Macrophage infiltration in tumors injected with CSPG4-CAR-Ms versus control CAR-Ms, with and without anti-CD47, will be compared. These experiments will determine whether CSPG4-CAR-Ms efficiently infiltrate tumors.Example 10Testing the efficacy of CSPG4-CAR-Ms and anti-CD47 in patient-derived xenograft organoids in vitro and in vivo[000187] All of the preliminary data are generated with immortalized human and mouse metastatic melanoma cell lines. Thus, whether CSPG4-targeting CAR-Ms inhibit the growth of patient-derived melanoma cells will be tested. For the patient-derived lines, melanoma lines from patients diagnosed with late-stage metastatic melanoma and harboring a constitutively active BRAF mutation (for comparisons with the melanoma cells used herein) will be sought. It was also ensured that these patient-derived lines could be propagated ex vivo and express CSPG4 at the transcript level. From these parameters, three patient-derived melanoma lines were identified (TABLE 1), these patient-derived cells attach and can be propagated on cell culture dishes. Thus, CAR-M-mediated phagocytosis and inhibition of melanoma spheroid growth in 3D matrices will be performed. These cell lines will be implanted into the right flank of 6-8-week-old male and female NRG mice. Then, CSPG-CAR-Ms will be peritumorally injected with and without intraperitoneal injections of anti-CD47 and tumor growth and animal survival will be monitored. These experiments will determine whether combined CSPG4-CAR-M and anti-CD47 treatment prevents patient-derived melanoma growth in vivo.Example 11Determining whether CAR-M injections after tumor resection prevent tumor recurrence in vivo[000188] To better recapitulate strategies in the clinic whereby patients will first undergo tumor resection, followed by treatment, a tumor resection model will be employed. The presence of CSPG4-targeting CAR-Ms may allow for efficient clearance of remaining cancer cells not visible to the surgeon to prevent tumor recurrence. In this model, the tumors will be allowed to grow to 500 mm3, and then most of the primary tumor will be resected while leaving a 2 mm * 2 mm tumor piece behind to allow for tumor recurrence. The wound will be closed with survival surgery, and then 1e6CSPG4-CAR-Ms or control CAR-Ms will be immediately injected. A trial run of this experiment was performed in NRG mice with human A375M melanoma cells, and it was foundthat tumors could be easily removed, and a tumor piece left behind (FIG. 18, arrow indicates time of “partial tumor resection”). CAR-M injections could also easily be performed at the surgery site - CAR-M injections in 2 mice showed that the cells stayed within the surgery site and did not leak out through the staples. However, it was noted that there was a considerable amount of variability with tumor regrowth rate (FIG. 18, only non-CAR-M treated tumors shown). Thus, to achieve 95% confidence with a predicted minimum of 40% change in tumor regrowth rate upon CAR-M injections (as determined by previous in vivo CAR-M studies herein), this experiment will be repeated with 12 mice per condition. Tumor regrowth will be measured by measuring tumors every 2 days with calipers and compare CSPG4-CAR-M injected animals with control CAR-M injected animals with and without anti-CD47. These experiments will also be performed with immune-competent C57BL6 / J mice and YUMM1.7 cells with bone marrow-derived CAR-Ms and mouse anti-CD47.[000189] Alternatively, melanoma tumors will be treated with CAR-Ms and then, the tumor will be resected. Then, the mice will be challenged by injecting them again with melanoma cells to determine whether tumors grow. Additionally, in these experiments, the brain (of female and male mice) and testis (of male mice only) will be isolated to confirm no off-target effects of CSPG4- targeting CAR-Ms on oligodendrocyte precursor cells and Sertoli cells, in which higher levels of CSPG4 RNA was observed herein, although previous work indicates no detection of CSPG4 protein. The studies will also be moved toward melanoma brain metastases, and it will be tested whether CSPG4-targeting CAR-Ms eradicate focal melanoma metastases in the brain. Second, strategies for injecting CSPG4-targeting CAR-Ms intravenously to determine the effect on primary tumor growth as well as metastatic sites will be initiated.Example 12Sustained CSPG4-CAR-M-mediated inhibition of melanoma growth is via reprogramming of the tumor immune environment[000190] Data herein suggest that melanoma tumors injected with CSPG4-targeting CAR-Ms exhibit changes in the proportions of tumor immune cells compared to tumors injected with control CAR-Ms (FIG. 16D). The initial analysis was performed only 4 days after the last CAR-M injection and it suggests the onset of tumor immune reprogramming. Given that a significant inhibition in tumor growth in CSPG4-CAR-M treated tumors compared to controls was observed well after CAR-Ms were no longer detected in the tumor, whether tumor immune reprogramming leads to sustained inhibition of tumor growth will be further tested. Specifically, it will be determined whether CSPG4-targeting CAR-Ms repolarize tumor-resident macrophages to an anti-tumor likestate leading to additional melanoma cell phagocytosis and cell death. Also, it will be determined whether CSPG4-targeting CAR-Ms improve T-cell infiltration into tumors by cross presenting melanoma antigens to T-cells, thereby further inhibiting melanoma growth.[000191] Temporal dynamics of changes in the tumor immune landscape in CSPG4-CAR-M- injected melanomas will be determined. Given that B-cells and T-cells comprised the largest proportion of CD45+ cells in the tumor (FIG. 16D), and changes in the proportions of T-cells and B-cells were observed in the CSPG4-CAR-M-injected tumors compared to control tumors (FIG. 16D), flow cytometry-based immune profiling of CAR-M-treated tumors will be carried out at additional time points to determine the temporal dynamics of immune microenvironment remodeling and identify the time point with the largest changes in immune populations. Tumors injected with CSPG4-targeting CAR-Ms, control CAR-Ms, or vehicle at day 10 (as done previously herein in FIG. 16D), and days 14, 18, and 25 (at which time the cohort may have to be culled due to tumor size in the vehicle-injected controls) will be profiled. The flow cytometry panel will be used which will enable quantification of CD8 T-cells, effector and regulatory CD4 T-cells, B cells, monocytes, macrophages, and dendritic cells (as in FIG. 16D, plus CD4, CD8, Foxp3, and CD25). From the variability observed herein, at least n = 6 mice for each condition will be required. Based on data herein that show sustained inhibition of tumor growth after injection with CSPG4-targeting CAR-Ms compared to controls, it is predicted that a further increase in the proportion of T-cells — including both CD8 and CD4 T cells — and B-cells infiltrating the tumor over time in CSPG4-CAR- M treated mice compared to controls will be observed. Also, it is predicted herein that the overall numbers of macrophages will decrease, and that this decrease might reflect a reduction in tumorpromoting macrophages. These initial studies will allow for determination of the ideal time point to further dissect immune cell status. However, it is possible that dramatic changes will not be visualized in general immune cell infiltration between the conditions given that the overall number of a particular immune cell type might not change, but the subsets within that immune cell type might be dramatically altered. These subclassifications will be further determined, but if that this is the case, then the tumors will be profiled at the time when the largest change in tumor cell size between CSPG4-CAR-M treated tumors versus controls is first visualized, rather than focusing at the time point with the largest differences in immune populations.[000192] Whether CSPG4-targeting CAR-Ms polarize endogenous tumor-associated macrophages to an anti-tumor state will be determined. The data herein show that the proportion of macrophages decrease in melanoma tumors four days after the tumors were treated with CSPG4-CAR-Ms compared to control, and the experiments above will delineate the appropriate time point to make further comparisons (either via overall changes in amount of macrophageinfiltration or by tumor size). It is expected herein that macrophages in untreated melanoma tumors will be predominantly differentiated into a pro-tumor, immunosuppressive state, driving tumor progression. To test the hypothesis that CAR-Ms repolarize tumor-associated suppressive macrophages to a more anti-tumor state, flow cytometry will be used to quantify the expression of CD80, CD86, CD206, MHCII, CD163 on endogenous macrophages (defined as CD11 b+ F4 / 80+ GFP-; as CAR-Ms are GFP+). It is expected that the expression profile of tumor- associated macrophages to shift toward MHClIhi, CD86hi, CD80hi, CD206IO, CD163IO in CSPG4- CAR-M-injected mice versus controls. These results would suggest that CSPG4-targeting CAR- Ms re-educate pro-tumor macrophages to take on an anti-tumor state. To functionally validate this finding, F4 / 80+ macrophages will additionally be isolated from CAR-M-treated tumors (using magnetic beads) and their ability to suppress T-cell proliferation will be tested in an in vitro assay. Naive CD8 T-cells will be isolated from the spleens of healthy mice, loaded with CFSE, and cocultured with CD3 / CD28 beads and macrophages from either CSPG4-CAR-M-treated or control- CAR-M-treated mice. T-cell proliferation will be quantified after 72 hours, and the hypothesis that macrophages from CSPG4-CAR-M-treated tumors will exhibit less capacity to suppress T-cells than those from control mice will be tested. To test this hypothesis directly in a human system, orthogonal experiments will be performed in cell culture with primary human macrophages. First, naive human macrophages will be cocultured with A375M human melanoma cells for 5 days in 3D such that the macrophages differentiate into pro-tumor macrophages. Then, CSPG4-targeting CAR-Ms will be added and whether the non-CAR, pro-tumor macrophages are reprogrammed to an anti-tumor state after an additional 5 days of coculturing will be determined using flow cytometry analysis using the same markers above. Chemokine and cytokine production will be measured via ELISA. Herein, it was already determined that macrophage / cancer cell cocultures are viable for 10 days in 3D (FIGS. 10A-E). The rates of macrophage-dependent phagocytosis of melanoma cells and melanoma cell death will be quantified to determine whether repolarizing macrophages to an anti-tumor state induces additional phagocytosis of melanoma cells and subsequent cell death.[000193] Whether CSPG4-targeting CAR-Ms activate T-cells in melanoma tumors through antigen presentation will be determined. The results herein show increased proportions of T-cells in CSPG4-CAR-M-injected tumors compared to control (FIG. 16D). It is hypothesized that further increases in the proportion of T-cells in CSPG4-CAR-M-injected tumors over time compared to control tumors will be observed, and that this T-cell population will be comprised of mainly effector CD4 and CD8 T-cells. Whether a higher proportion of each effector CD4 and CD8 T-cells is activated in tumors injected with CSPG4-targeting CAR-Ms versus controls will be determined.Next, the effect of CAR-Ms specifically on CD8 T-cells will be analyzed. First, whether CSPG4- targeting CAR-Ms efficiently present class I melanoma antigens after phagocytosis will be determined; and then whether CSPG4-targeting CAR-Ms directly activate CD8 T-cells at a higher rate than control CAR-Ms will be determined. The presentation of class I antigens to CD8 T cells is of focus, future work can extend these studies to investigate presentation of class II antigens to CD4 T-cells, particularly if it found that effector CD4 T-cells comprise a major fraction of total tumor-infiltrating T cells or are increased by CSPG4-targeting CAR-Ms.[000194] T-cell activation in tumors treated with CSPG4-targeting CAR-Ms will be analyzed. T- cell activation states in melanoma tumors treated with CSPG4-targeting CAR-Ms versus controls will be determined by profiling tumors for CD69, CD25, CD44, and PD1. If T-cells are activated in CSPG4-CAR-M-treated tumors, it is expected that there will be an increase proportion of T- cells expressing these markers. T-cell exhaustion with PD1 , CTLA4, and LAG3will also be tested. As a complementary approach, functional activity of tumor-infiltrating T-cells will be assayed by measuring production of effector cytokines. Tumors from CSPG4-CAR-M-treated mice will be harvested, dissociated, stimulated ex vivo with PMA / lonomycin for 4 hours, and then production of I FNy, TN Fa, IL-2, IL-4, and IL-17A in CD4s and CD8s will be measured by flow cytometry. It is hypothesized that T-cells from CSPG4-CAR-M-treated mice will exhibit greater functional activity, namely, greater production of IFNy, TNFa, and IL-2 compared to mice treated with control CAR-Ms.[000195] Whether CSPG4-CAR-Ms cross-present melanoma antigen will be determined. It is hypothesized that CSPG4-targeting CAR-Ms will present more antigen because they are phagocytosing more melanoma cells than control CAR-Ms. To determine whether CAR-Ms cross-present antigen to T-cells, CAR-Ms will be cocultured with YUMM1 .7 cells expressing OVA. OVA-expressing tumors will be used in these assays because of the unique reagents that exist for measuring OVA antigen presentation, namely, clone 25D1.16 antibodies that enables detection of the OVA antigen SIINFEKL (SEQ ID NO: 14) loaded on MHCI (H-2Kb / SIINFEKL). After 5 days of coculture, the 25D1.16 antibody will be used to measure macrophage presentation of SIINKFEKL (SEQ ID NO: 14). It is hypothesized that CSPG4-targeting CAR-Ms will present the OVA antigen SIINFEKL (SEQ ID NO: 14) at a higher abundance than control GFP-CAR-Ms, which will be observed as a higher mean fluorescence intensity measured by flow cytometry. As a positive control for detection, bone marrow-derived dendritic cells will be pulsed with increasing concentrations of SIINFEKL (SEQ ID NO: 14) peptide.[000196] Whether CPSG4-CAR-Ms activate T-cells by cross-presenting melanoma antigen will be determined. The ability of CSPG4-targeting CAR-Ms to re-stimulate OVA-specific T-cellsharvested from YUMM1.7-OVA-bearing mice will be tested. After coculturing CSPG4-targeting or control CAR-Ms with YUMM1.7-0VA tumor cells in vitro to enable antigen uptake, antigen- loaded CSPG4-CAR-Ms will be combined with SHNFEKL-specific CD8 T-cells isolated from the spleen of YUMM1.7-OVA-bearing mice (using flow cytometry to isolate tetramer+ T cells), and the production of IFNy will be measured by ELISpot assays. It is hypothesized that a higher number of spots (i.e., greater IFNy production) will be observed by SIINFEKL (SEQ ID NO: 14)- specific T-cells cocultured with CSPG4-targeting CAR-Ms than control CAR-Ms, due to the superior antigen presentation of CSPG4-targeting CAR-Ms. These experiments could be done alternatively or in parallel using SIINFEKL (SEQ ID NO: 14)-specific OT-I CD8 T-cells.[000197] Recognizing that OVA is a strong, non-endogenous antigen, as an alternative approach the same experiment will be carried out using non-OVA-expressing YUMM1.7 tumor cells and the response to TRP2 will be focused on. TRP2 is a widely-expressed melanoma antigen that contains an H2-Kb-restricted peptide SVYDFFVWL (SEQ ID NO: 15). The abundance of TRP2-specific T-cells has not yet been quantified in YUMM1.7-bearing mice; however, if there is not a sufficient quantity of TRP2-specific T-cells in this model, B16-F10 will alternatively be used. B16-F10 is known to generate a significant TRP2-specific T-cell population. B16-F10 cells do not express high levels of CSPG4. Overexpressing CSPG4 promotes phagocytosis by CSPG4-targeting CAR-Ms. Thus, B16F-10 cells overexpressing CSPG4 will be used to ensure proper CSPG4-CAR-M targeting, and perform these experiments.[000198] To determine whether CD8 T-cell activation is via MHCI antigen presentation by CAR- Ms, bone marrow-derived monocytes will be isolated from [32-microglobulin mutant mice (JAX, B6.129P2-B2mtm1 Unc / DcrJ, #002087). Mice homozygous for the [32-microglobulin targeted mutation ([32Mmut) have little, if any, MHC class I expression. Thus, macrophages from these knockout mice will have phagocytosis intact, but CD8 antigen presentation is lost. CSPG4- targeting CAR-Ms will be generated from these [32Mmutbone marrow-derived monocytes, and the function of CSPG4-CAR-M- [32Mmutin inhibiting YUMM1.7 tumor growth in C57BL6 / J mice over time will be tested as in FIGS. 16A-E compared to CSPG4-targeting CAR-Ms generated in wildtype macrophages. It is expected that CSPG4-CAR-M-mediated inhibition of tumor growth will be significantly dampened in CSPG4-targeting CAR-Ms generated with [32Mmutmacrophages versus CSPG4-targeting CAR-Ms generated with wildtype macrophages, which would demonstrate that tumor control of CAR-Ms is partially dependent on MHCI antigen presentation. [000199] Whether combining CSPG4-targeting CAR-Ms with anti-CD47 approaches enhances tumor immune reprogramming will be determined. It is expected that increasing CAR-M-mediated phagocytosis of melanoma cells by using anti-CD47 approaches will further drive thereprograming of tumor immune cells, either by enhancing the activation of T-cells and reprogramming of tumor-associated macrophages into an anti-tumor-like state and / or by accelerating this reprogramming. There is also evidence that CD47 blockade increases T-cell- mediated destruction of tumors independently of macrophages. Thus, experiments will be performed in the presence and absence of anti-CD47 to determine the temporal changes tumor immune populations when CD47 function is inhibited. Activation states in both macrophage populations will be profiled (using the in vivo profiling strategies described herein) and T-cell populations (using T-cell priming strategies described herein) in CSPG4-targeting CAR-Ms versus control GFP-CAR-Ms with and without mouse CD47 blocking antibodies.[000200] Taken together, the experiments proposed in this Example will determine how combining CSPG4-targeting CAR-Ms with anti-CD47 approaches reprogram the tumor immune landscape for sustained inhibition of melanoma growth in vitro and in vivo.[000201] Tumor-associated macrophages may not be reprogrammed into an anti-tumor state. That may be an unlikely result, mostly because a difference in tumor size in xenograft models with immune-deficient NRG mice was observed herein, suggesting a myeloid cell role in inhibiting tumor size in the absence of adaptive immune cells. However, despite these mice maintaining their myeloid population, their dendritic cells and macrophages are not normal. Thus, if a lack of macrophage reprogramming is observed, examining dendritic cells and natural killer cells would be clear next steps. When using the |32Mmutmacrophages for CAR-M generation, no changes in T-cell activation compared to wildtype macrophages may be found. These results would be interesting as they would suggest that the inhibition of tumor growth could be mediated by CD4+ T-cells, activated via class II presentation by CAR-Ms. Given that the proportions of CD4+ versus CD8+ T cells in CSPG4-CAR-M-treated tumors will be quantified, the infiltration of each of these T-cell populations will be determined, which will further refine the hypotheses herein regarding CD4 versus CD8 T-cells. Central to performing any of these experiments is generating enough murine CAR-Ms for the immune-competent mouse experiments. Given the previous experiences herein with transducing murine monocytes at high efficiencies, it is expected that sufficient CAR- Ms will be generated for these studies. If there is difficulty transducing p2Mmutbone marrow derived macrophages, then strategies for knocking down p2M expression with shRNA approaches in bone marrow derived macrophages from wildtype C57BL6 / J mice may be used.Example 13CSPG4 / NG2-targeting CAR-Ms in glioblastoma[000202] CSPG4 is also known as NG2 (Neural-Glial Antigen 2), a proteoglycan upregulated in glioblastoma. FIGS. 19A-F, FIGS. 20A-F, and FIGS. 21A-E show that NG2-CAR-Ms infiltrate and phagocytose NG2-expressing glioblastoma cells, as well as kill glioblastoma cells. The following materials and method were used to generate the data in FIGS. 19A-F, FIGS. 20A-F, and FIGS. 21A-E.[000203] Glioblastoma Cell Culture. U87-MG / RFP luciferase stable cells were obtained from Cellomics Technology, Arbutus, Maryland, Glioma Stem Cell Lines (GSC) 262 and 8-11 were gifts from the Frederick Lang lab at MD Anderson Cancer Center. Direct from patient lines (PD) 101922 and 101821 were obtained from the Samuel Cheshier lab at University of Utah, Hunstman Cancer Institute. U87-MG / RFP luciferase were propagated and cultured in 2D in DMEM with 10% FBS. GSC 262, GSC 8-11 , PD 101922 and PD 101821 were cultured in suspension in Neural Stem Cell Media (50% Neural Basal Media supplemented / 50% DMEM / F12 supplemented with 1X MEM Non-essential amino acids, 1X GlutaMax-l, 1X Sodium Pyruvate, 1X Sodium Bicarbonate, 25 mM HEPES, 1X Penicillin-Streptomycin, 1X B27, 20 ng / mL EGF, 20 ng / mL bFGF, 10 ng / mL Heparin, 20 ng / mL PDGF-AA and 20 ng / mL PDGF-BB). Cell lines were routinely tested (6-month intervals) for mycoplasma detection using PCR detection kit (ATCC, Manassas, Virginia; #30-1012K). U87-MG / RFP luciferase, GSC 262, and GSC 8-11 were used for <20 passages. PD 101922 and 101821 were used for <5 passages.[000204] Glioblastoma Spheroid Formation. GSC 262, GSC 8-11, PD 101922, and PD 101821 were cultured in suspension in T25 or T75 flasks in Neural Stem Cell Media and allowed to form spheroids over the course of 7 days supplementing with fresh media every 2-3 days. After spheroids exponential growth plateaued, spheroids were mechanically dissociated to single cells via gentle trituration using a 21G needle. Single cells were transferred to new flasks at a density of 50K cells per 1 mL for propagation.[000205] U87-MG / RFP luciferase cells were lifted using TrypLE Express, counted, and plated in Neural Stem Cell Media in low-adhesion 96-well U-bottom plates at a density of 10K cells per well. Spheroids were allowed to form over the course of 72 hours prior to downstream assays.[000206] U87-MG / RFP Luciferase 3D CAR-M Spheroid Adherence Assay. After 72-hours of U87-MG / RFP luciferase spheroid formation, 50K control CAR-Ms or NG2-CAR-MS were directly added to the wells along with GM-CSF. The CAR-Ms and cancer cells were co-cultured for 8 hours. Spheroids were then gently washed in Neural Stem Cell Media to remove any nonadherent CAR-Ms and then plated on Matrigel coated glass-bottom plates. Images were acquired using a Plan-Apochromat 10X / 0.45 objective on an LSM 880 using Airyscan® FAST® acquisition mode. After Airyscan® processing, maximum intensity projections were created to visualize thespheroid surface. Images were analyzed using FIJI to quantify the percent of total spheroid surface area with adherent CAR-Ms.[000207] U87-MG / RFP Luciferase 3D CAR-M Spheroid Infiltration Assay. After 72-hours of U87-MG / RFP luciferase spheroid formation, 50K control CAR-Ms or NG2-CAR-Ms were directly added to the wells along with GM-CSF. The CAR-Ms and cancer cells were co-cultured for 72 hours. Spheroids were then gently washed in Neural Stem Cell Media to remove any nonadherent CAR-Ms and then plated on Matrigel coated glass-bottom plates. Images were acquired using a Plan-Apochromat 10X / 0.45 objective on an LSM 880 using Airyscan® FAST® acquisition mode in 100-micron step size to create a z-stack. After Airyscan® processing images were analyzed using FIJI to quantify the percent of total spheroid volume with infiltrating CAR-Ms.[000208] U87-MG / RFP Luciferase 3D Phagocytosis Flow Assay. After 72 hours of U87- MG / RFP luciferase spheroid formation, 10K control CAR-Ms or NG2-CAR-MS were directly added to the wells along with GM-CSF. CAR-Ms and cancer cells were allowed to co-culture for 24 hours. After 24 hours, spheroids and macrophages from 3 wells were pooled for each technical replicate. Cells were then pelleted and stained for flow cytometry.[000209] U87-MG / RFP Luciferase 3D Spheroid Growth Assay. After 72 hours of U87-MG / RFP luciferase spheroid formation, 50K control CAR-Ms or NG2-CAR-Ms were directly added to the wells along with GM-CSF and imaged every 8 hours for 184 hours using a 10X / 0.3 objective with an Incucyte® Sx5 analysis system. Media was changed every 3 days. RFP total object integrated intensity were quantified using the spheroid analysis module following spectral unmixing using the Incucyte® software (version 2020C Rev 1).[000210] Xenograft Experiments. Animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Utah. Mice underwent right frontal intracranial injections with either 100K U87-MG / RFP luciferase cells were injected alone or coinjections with 100K U87-MG / RFP luciferase cells and 300K CAR-Ms. Tumor volumes were monitored over 24 days using bioluminescence imaging on the MS platform. Mice were monitored for survival as determined by veterinary-defined endpoints.[000211] Brain and Tumor Isolation and Digestion for Flow Cytometry Analysis. Mice were deeply sedated using 10 pL / g of Ketamine. After confirming adequate sedation with toe pinch, the thoracic cavity was opened and transcardial perfusion of cold PBS was begun. After cardiac arrest, mice underwent cervical dislocation, and the brain was harvested in a sterile fashion. The right hemisphere, left hemisphere, and cerebellum were sharply dissected from each other and aliquoted separately for down-stream analysis. Brains were enzymatically and mechanicallydissociated. Flow cytometry was performed to evaluate for GFP+ CAR-Ms as well as RFP+ tumor cells based on location.***[000212] The foregoing description of the specific aspects will so fully reveal the general nature of the invention that others can, by applying knowledge within the skill of the art, readily modify and / or adapt for various applications such specific aspects, without undue experimentation, without departing from the general concept of the present disclosure. Therefore, such adaptations and modifications are intended to be within the meaning and range of equivalents of the disclosed aspects, based on the teaching and guidance presented herein. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by the skilled artisan in light of the teachings and guidance.[000213] The breadth and scope of the present disclosure should not be limited by any of the above-described exemplary aspects, but should be defined only in accordance with the following claims and their equivalents.[000214] All publications, patents, patent applications, and / or other documents cited in this application are incorporated by reference in their entirety for all purposes to the same extent as if each individual publication, patent, patent application, and / or other document were individually indicated to be incorporated by reference for all purposes.[000215] For reasons of completeness, various aspects of the invention are set out in the following numbered clauses:[000216] Clause 1. A modified phagocyte comprising a chimeric antigen receptor (CAR), wherein the CAR comprises one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain.[000217] Clause 2. The modified phagocyte of clause 1 , wherein the endodomain comprises a phagocytic receptor or a fragment thereof.[000218] Clause 3. The modified phagocyte of clause 2, wherein the phagocytic receptor is multiple EGF like domains 10 (Megf10), high affinity immunoglobulin epsilon receptor subunit gamma (FcRy), or MER proto-oncogene tyrosine kinase (MerTK).[000219] Clause 4. The modified phagocyte of clause 1 , wherein the CAR does not comprise an endodomain domain comprising a phagocytic receptor or a fragment thereof.[000220] Clause 5. The modified phagocyte of any one of clauses 1 -4, wherein the endodomain comprises a His, Myc, luciferase, or fluorescent tag.[000221] Clause 6. The modified phagocyte of any one of clauses 1-5, wherein the ectodomain further comprises a spacer, a co-stimulation domain, a secreted antibody domain, a cytokine secretion domain, or a combination thereof.[000222] Clause 7. The modified phagocyte of clause 5, wherein the spacer comprises from about 90 nucleic acids to about 150 nucleic acids or from about 30 amino acids to about 50 amino acids.[000223] Clause 8. The modified phagocyte of any one of clauses 1-7, wherein the transmembrane domain is a CD8 transmembrane domain or a FcRy transmembrane domain.[000224] Clause 9. The modified phagocyte of any one of clauses 1-8, wherein the CSPG4 binding domain is encoded by SEQ ID NO: 1.[000225] Clause 10. The modified phagocyte of any one of clauses 1-9, wherein the transmembrane domain is encoded by SEQ ID NO: 3 or SEQ ID NO: 7.[000226] Clause 11. The modified phagocyte of any one of clauses 1-10, wherein the endodomain is encoded by SEQ ID NO: 9.[000227] Clause 12. The modified phagocyte of any one of clauses 1-11 , wherein the CAR is encoded by a plasmid having SEQ ID NO: 13.[000228] Clause 13. A pharmaceutical composition comprising a modified phagocyte comprising a chimeric antigen receptor (CAR), wherein the CAR comprises one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain.[000229] Clause 14. The pharmaceutical composition of clause 13, wherein the composition further comprises an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof.[000230] Clause 15. The pharmaceutical composition of clause 13 or clause 14, wherein the composition further comprises a pharmaceutically acceptable carrier.[000231] Clause 16. A method of treating cancer in a subject in need thereof comprising administering to the subject a therapeutically effective amount of the pharmaceutical composition of any one of clauses 13-15.[000232] Clause 17. The method of clause 16, wherein the therapeutically effective amount of the pharmaceutical composition comprises from about 1 million to about 10 million of the modified phagocytes.[000233] Clause 18. The method of clause 16 or clause 17, wherein the method further comprises administering a therapeutically effective amount of an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof; or thecomposition further comprises an anti-CD47 antibody, an inhibitor of SIRPα, an anti-SIRPα antibody, or a combination thereof.[000234] Clause 19. The method of clause 18, wherein the therapeutically effective amount of the anti-CD47 antibody is from about 1 mg / kg to about 45 mg / kg.[000235] Clause 20. The method of clause 18 or clause 19, wherein the anti-CD47 antibody is co-administered with the pharmaceutical composition, administered before the pharmaceutical composition, administered after the pharmaceutical composition, or a combination thereof.[000236] Clause 21. The method of any one of clauses 16-20, wherein the pharmaceutical composition is administered to the subject locally, systemically, or a combination thereof.[000237] Clause 22. The method of clause 21 , wherein the pharmaceutical composition is administered locally in the brain at time of tumor resection, intravenously, or intrathecally.[000238] Clause 23. The method of any one of clauses 18-22, wherein the anti-CD47 antibody, the inhibitor of SIRPα, an anti-SIRPα antibody, or a combination thereof is administered locally in the brain at time of tumor resection, intravenously, or intrathecally.[000239] Clause 24. The method of any one of clauses 16-23, wherein the cancer is a solid tumor.[000240] Clause 25. The method of any one of clauses 16-24, wherein the cancer is glioblastoma, melanoma, breast cancer, or any other CSPG4-expressing solid tumor.[000241] Clause 26. A method of modifying a phagocyte comprising: delivering to the phagocyte a plasmid comprising one or more nucleic acid sequences encoding one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain.[000242] Clause 27. The method of clause 26, wherein the plasmid is a viral vector.[000243] Clause 28. The method of clause 27, wherein the viral vector is a lentiviral vector or an adeno-associated viruses (AAV).[000244] Clause 29. The method of any one of clauses 26-28, wherein the plasmid has SEQ ID NO: 13.SEQUENCES

Claims

CLAIMSWhat is claimed is:

1. A modified phagocyte comprising a chimeric antigen receptor (CAR), wherein the CAR comprises one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain.

2. The modified phagocyte of claim 1, wherein the endodomain comprises a phagocytic receptor or a fragment thereof.

3. The modified phagocyte of claim 2, wherein the phagocytic receptor is multiple EGF like domains 10 (MegflO), high affinity immunoglobulin epsilon receptor subunit gamma (FcRy), or MER proto-oncogene tyrosine kinase (MerTK).

4. The modified phagocyte of claim 1 , wherein the CAR does not comprise an endodomain domain comprising a phagocytic receptor or a fragment thereof.

5. The modified phagocyte of claim 1 , wherein the endodomain comprises a His, Myc, luciferase, or fluorescent tag.

6. The modified phagocyte of claim 1 , wherein the ectodomain further comprises a spacer, a co-stimulation domain, a secreted antibody domain, a cytokine secretion domain, or a combination thereof.

7. The modified phagocyte of claim 5, wherein the spacer comprises from about 90 nucleic acids to about 150 nucleic acids or from about 30 amino acids to about 50 amino acids.

8. The modified phagocyte of claim 1 , wherein the transmembrane domain is a CD8 transmembrane domain or a FcRy transmembrane domain.

9. The modified phagocyte of claim 1 , wherein the CSPG4 binding domain is encoded by SEQ ID NO: 1.

10. The modified phagocyte of claim 1 , wherein the transmembrane domain is encoded by SEQ ID NO: 3 or SEQ ID NO: 7.

11. The modified phagocyte of claim 1 , wherein the endodomain is encoded by SEQ ID NO: 9.

12. The modified phagocyte of claim 1 , wherein the CAR is encoded by a plasmid having SEQ ID NO: 13.

13. A pharmaceutical composition comprising a modified phagocyte comprising a chimeric antigen receptor (CAR), wherein the CAR comprises one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain.

14. The pharmaceutical composition of claim 13, wherein the composition further comprises an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof.

15. The pharmaceutical composition of claim 13, wherein the composition further comprises a pharmaceutically acceptable carrier.

16. A method of treating cancer in a subject in need thereof comprising administering to the subject a therapeutically effective amount of the pharmaceutical composition of claim 13.

17. The method of claim 16, wherein the therapeutically effective amount of the pharmaceutical composition comprises from about 1 million to about 10 million of the modified phagocytes.

18. The method of claim 16, wherein the method further comprises administering a therapeutically effective amount of an anti-CD47 antibody, an inhibitor of signal regulatory protein a (SIRPα), an anti-SIRPα antibody, or a combination thereof; orthe composition further comprises an anti-CD47 antibody, an inhibitor of SIRPα, an anti-SIRPα antibody, or a combination thereof.

19. The method of claim 18, wherein the therapeutically effective amount of the anti-CD47 antibody is from about 1 mg / kg to about 45 mg / kg.

20. The method of claim 18, wherein the anti-CD47 antibody is co-administered with the pharmaceutical composition, administered before the pharmaceutical composition, administered after the pharmaceutical composition, or a combination thereof.

21. The method of claim 16, wherein the pharmaceutical composition is administered to the subject locally, systemically, or a combination thereof.

22. The method of claim 21 , wherein the pharmaceutical composition is administered locally in the brain at time of tumor resection, intravenously, or intrathecal ly.

23. The method of claim 18, wherein the anti-CD47 antibody, the inhibitor of SIRPα, an anti- SIRPα antibody, or a combination thereof is administered locally in the brain at time of tumor resection, intravenously, or intrathecally.

24. The method of claim 16, wherein the cancer is a solid tumor.

25. The method of claim 16, wherein the cancer is glioblastoma, melanoma, breast cancer, or any other CSPG4-expressing solid tumor.

26. A method of modifying a phagocyte comprising: delivering to the phagocyte a plasmid comprising one or more nucleic acid sequences encoding one or more of: an ectodomain comprising a chondroitin sulfate proteoglycan 4 (CSPG4) binding domain; a transmembrane domain; and an endodomain.

27. The method of claim 26, wherein the plasmid is a viral vector.

28. The method of claim 27, wherein the viral vector is a lentiviral vector or an adeno- associated viruses (AAV).

29. The method of claim 26, wherein the plasmid has SEQ ID NO: 13.

Images