MESENCHYMAL STEM CELLS TO ENHANCE ANTI-TUMOR ACTIVITY OF IMMUNOTHERAPY

Abstract

Mesenchymal stem cells (MSCs) for use in the treatment of a tumor, wherein said treatment comprises the combined administration of said mesenchymal stem cells with an anti-tumor immunotherapy, and wherein said MSCs do not comprise exogenous nucleic acids that encode immune response-stimulating cytokines. In a preferred embodiment the invention relates to the use of said MSCs in the treatment of a tumor and/or malignant disease, wherein the anti-tumor immunotherapy for combined administration comprises the administration of a cellular immunotherapy, preferably chimeric antigen receptor (CAR) T cells, wherein said T cell receptor binds specifically to a tumor-associated antigen, and the mesenchymal stem cells are not genetically modified.

Claims

1. A method of supporting T cell proliferation and activation in a subject treated with CD4.sup.+ T cells in an anti-tumor treatment, comprising administering to the subject having a tumor genetically-unmodified CD34.sup.- mesenchymal stem cells (MSCs) derived from bone marrow combined with the CD4.sup.+ T cells therapy, wherein said MSCs do not carry anti-cancer agents and wherein the genetically-unmodified CD34.sup.- MSCs derived from bone marrow increase the percentage of the CD4.sup.+ T cells that are activated and proliferating, wherein the CD4.sup.+ T cells for combined administration express a chimeric antigen receptor (CAR), wherein said CAR binds specifically to a tumor-associated antigen, thereby treating the tumor.

2. The method according to claim 1, wherein the CD34.sup.-mesenchymal stem cells and/or CD4.sup.+ T cells for combined administration are autologous to the subject of medical treatment.

3. The method according to claim 1, wherein the CD34.sup.-mesenchymal stem cells and/or CD4.sup.+ T cells for combined administration are allogeneic to the subject of medical treatment.

4. The method according to claim 1, further comprising administering a cytokine or chemokine to the subject.

5. The method according to claim 1, further comprising administering a small molecule with anti-tumor immune-stimulating properties to the subject.

Description

FIGURES

[0144] The following figures are presented in order to describe particular embodiments of the invention, by demonstrating a practical implementation of the invention, without being limiting to the scope of the invention or the concepts described herein.

Short Description of the Figures

[0145] FIG. 1: #946 CAR expression on CEA-specific human CAR-T cells by two colour FACS.

[0146] FIG. 2: Wild-type MSC enhance the anti-tumor activity of CAR-T cell.

[0147] FIG. 3: Mesenchymal stem cells do not suppress CAR-T cell activation.

[0148] FIG. 4: MSC support proliferation and survival of CAR-T cells in activation dependent manner.

[0149] FIG. 5: Survival of activated CAR T cells in presence of MSC.

[0150] FIG. 6: Native MSC increase NK-cell dependent killing of tumor cells.

[0151] FIG. 7: Native MSC support CD4+ T cell proliferation and activation in vitro.

[0152] FIG. 8: Genetically engineered MSC which express the transgene alpha-1 anti-trypsin activate T cell proliferation in a comparable manner as native MSC.

DETAILED DESCRIPTION OF THE FIGURES

[0153] FIG. 1: #946 CAR expression on CEA-specific human CAR-T cells by two colour FACS.

[0154] Human peripheral T cells were isolated from the blood of a healthy donor and transduced by vector #946 encoding the sequence of a CAR consisting of an anti-Carcinoembryonic antigen (CEA) single chain Fv-binding domain, a human IgG1 derived hinge-CH2-CH3 spacer domain and a chimeric transmembrane and intracellular CD28-CD3zeta domain for combined CD28-CD3zeta signaling with specificity for carcinoembryonic antigen (CEA). 12 hrs upon end of transduction cells were tested for #946 CAR expression by two colour FACS using a PE-labeled polyclonal goat-anti-human IgG Fc antibody and the FITC-labeled murine anti-CD3 human monoclonal antibody UCHT1.

[0155] FIG. 2: Wild-type MSC enhance the anti-tumor activity of CAR-T cell

[0156] CEA is highly expressed on the surface of a majority of pancreatic adenocarcinomas. CEA expression is elevated in nearly 90% of cancers arising from endodermal tissue, including the gastrointestinal tract, pulmonary tissues, and breast. Anti-CEA-CAR grafted T cells and non-modified T cells w/o CAR for control (each 2×10(4)/well) were co-cultivated for 48 h with CEA+ LS174T or CEA- Colo320 tumor cells (each 2.5×10(4)/well) and wt or IL7/IL21 secreting MSC (each 3×10(3)/well) in 96 well round bottom plates. Viability of tumor cells was determined by a tetrazolium salt based XTT-assay. Cytolysis [%] was determined by 100-viability [%]. Values represent mean of triplicates +/- standard deviation (SD).

[0157] FIG. 3: Mesenchymal stem cells do not suppress CAR-T cell activation

[0158] Anti-CEA-CAR grafted T cells and non-modified T cells w/o CAR for control (each 2×10(4)/well) were co-cultivated for 48 h with CEA+ LS174T or CEA- Colo320 tumor cells (each 2.5×10(4)/well) and wt or IL7/IL21 secreting MSC (each 3×10(3)/well) in 96 well round bottom plates. Supernatants were harvested and IFN-gamma content was determined by ELISA. Values represent mean of experimental triplicates +/- standard deviation.

[0159] FIG. 4: MSC support proliferation and survival of CAR-T cells in activation dependent manner.

[0160] Modulation of CAR T cell proliferation upon CAR-stimulation by solid phase bound anti-idiotypic antibody in presence of MSC. Anti-CEA-CAR T cells (#946-CAR) were CFSE labeled and co-cultivated (5×10(4)/well) with non-modified or IL7/IL21 secreting MSC (5000/well), respectively, in 96-well plates that were coated with solid phase bound anti-idiotypic mAb BW2064 (4 .Math.g/ml) or PBS for control. After 5 days cells were recovered, CAR T cells stained with an anti-human IgG-PE antibody and analyzed by flow cytometry. CFSE dilution and the number of proliferating cells were determined. Values represent mean of triplicates +/- standard deviation (SD). Significant differences were determined by Student’s T test.

[0161] FIG. 5: Survival of activated CAR T cells in presence of MSC.

[0162] Anti-CEA-CAR T cells (#946-CAR) were CFSE labeled and co-cultivated (5×10(4)/well) with non-modified or IL7/IL21 secreting MSC (5000/well), respectively, in 96-well plates that were coated with solid phase bound anti-idiotypic mAb BW2064 (4 .Math.g/ml) or PBS for control. After 6 days cells were recovered, stained with an anti-human IgG-PE antibody to identify CAR T cells and 7-AAD for discrimination of life and dead cells. The cells were analyzed by flow cytometry. The percentage of living CAR-positive T cells relative to all T-cells was determined. Values represent mean of triplicates +/- standard deviation (SD).

[0163] FIG. 6: Native MSC increase NK-cell dependent killing of tumor cells.

[0164] Human native MSC were seeded in different concentrations (20 000/10 000/5000/1000/500 cells per 12-well) and incubated overnight in RPMI medium with 10% Fetal Calf Serum and 1% Glutamin. Human NK cells were isolated by negative MACS sorting frozen PBMC’s using the “NK Cell isolation kit” (Miltenyi, 130-092-657) according to the instructions of the manufacturer. The NK cell purity is assessed by FACS measurement using a CD56 specific antibody staining (Miltenyi, 130-100-683) according to the instructions of the manufacturer.

[0165] The purified NK cells (3×10^5 per sample) are co-cultured overnight with the plated native MSC or alone. On the next day an NK cytotoxicity assay is performed. The target tumor cell line K562 (1×10^6 cells) was stained with 10 .Math.M Calcein and incubated for 30 min at 37° C. (1 mM, C3099, Lot 1687887, Life Technologies). Afterwards the target cell line was incubated with the NK cells with an effector-target (E:T). 5000 NK and 5000 target cells were seeded into a well of 96 well-plate. The plate was incubated at 37° C. in 5% CO2 for 4 h. Supernatant was collected and the release of calcein as read-out for killing was measured in Fluorometer (excitation filter: 485 ± 9 nm; bandpass filter: 530 ± 9 nm). Killing was normalized relative to calcein-stained K562 that were lysed with triton X (Sigma Aldrich). Incubation of NK cells with MSC increases their killing activity in a dose dependent manner.

[0166] FIG. 7: Native MSC support CD4+ T cell proliferation and activation in vitro.

[0167] Human CD4+ cells were isolated from frozen PBMC using the “CD4 T cell isolation kit” (Miltenyi, 130-096-533) according to the instructions of the manufacturer. The CD4 cells were seeded into a 24-well plate (4×10^5 per well in 600 .Math.l) in RPMI1640 medium + 10% FCS + 1% Glutamax. CD4 cells were stimulated with anti-CD3 and anti-CD28 antibodies (Becton-Dickinson, 1 .Math.g/ml aCd3 5 .Math.g/ml aCd28). MSC were seeded in the well at a ratio (MSC:PBMC) of 1:16 or 1:80. The cultures were incubated for 3 days at 37° C. Supernatants were collected to detect IFNg, Proliferation was detected by FACS-analysis. A) The percentage of activated and proliferating CD4 cells was increased in the presence of MSC as seen in the FCS/SSC FACS analysis. Duplicates are displayed. B) The release of IFNg as sign of CD4-T cell activation was increased by the presence of MSC in a dose dependent manner.

[0168] FIG. 8: Genetically engineered MSC which express the transgene alpha-1 anti-trypsin activate T cell proliferation in a comparable manner as native MSC.

[0169] 4×10^5 PBMC per well (stained with 5 mM CFSE, molecular probes, CellTrace Cell Proliferation Kits, C34554, according to the instructions of the manufacturer) were cocultured with MSC in the indicated ratios in a 24-well plate (200 .Math.l RPMI-1640 medium + 10%FCS and 1% Glutamax). PMBC were stimulated with stimulated with anti-CD3 and anti-CD28 antibodies (Becton-Dickinson, 1 .Math.g/ml aCd3 5 .Math.g/ml aCd28). Cells were incubated for 3 days at 37° C. Proliferation of lymphocytes was analysed by FACS. Calcein signal intensity was used to determine the number of doublings that had occurred. The analysis indicated the coculture of PBMC with (A) native and (B) genetically modified MSC increased the proliferation of PBMC. In each case a larger fraction of the lymphocytes had undergone 1 or more doublings compared to the sample without any MSCs. Furthermore, the percentage of PBMC that had not undergone any doublings was reduced in the presence of MSC.

EXAMPLES

[0170] The following examples are presented in order to describe particular and in some cases preferred embodiments of the invention, by demonstrating a practical implementation of the invention, without being limiting to the scope of the invention or the concepts described herein.

Preparation of CEA-Specific Human CAR-T Cells

[0171] The #946 Chimeric antigen receptor consists of an anti-Carcinoembryonic antigen (CEA) single chain Fv-binding domain, a human IgG1 derived hinge-CH2-CH3 spacer domain and a chimeric transmembrane and intracellular CD28-CD3zeta domain for combined CD28-CD3zeta signaling. The CD28 signaling domain was furthermore mutated to abrogate IL2 secretion. The gamma-retroviral vector coding for the #946 CAR was produced according to SOP-VectProd using a Galv envelope. In summary vector particle production was done transiently on the human embryonic kidney cell line 293T after Lipofectamin (R) mediated DNA transfection. Vector particles were pseudotyped with Galv. No vector titer was determined. Supernatants were filtered through 0.45 .Math.m membrane and loaded on poly-D-lysin (PDL) coated culture plates by centrifugation for 30 min at 1500 xg.

[0172] Transduction of human blood lymphocytes was carried out accordingly. Human T cells were isolated from the blood of a healthy donor and separated by Ficoll density centrifugation.

[0173] Mononuclear cells were seeded at a density of 5×10.sup.6 cells/ml in RPMI 1640 + 10% FCS (v/v) and activated by 100 ng/ml anti-CD3 mAb OKT3 and 400 U/ml IL2. After 96 hrs cells were harvested, resuspended in Xvivo15 medium + 200 U/ml IL2 at a density of 5 ×10.sup.6 cells/ml and transduced on virus loaded plates by centrifugation for 90 min at 1500 x g in fresh supernatant of transfected 293T cells. The cells were incubated for 12 hrs. Cells were recovered and tested for #946 immunoreceptor expression (FIG. 1).

Preparation of Human Mesenchymal Stem Cells

[0174] The MSC were prepared according to the apceth process, for example as described using the cell culture medium disclosed in US8557578 B2, which is hereby incorporated in its entirety by reference. The cells were isolated from bone marrow from healthy human donors by plastic adherence and are cultured in growth medium.

[0175] Human MSCs are isolated from bone marrow by plastic adherence and are cultured in growth medium e.g. FBS containing DMEM as described by Pittinger, M.F. (2008) Mesenchymal stem cells from adult bone marrow, In D.J. Prockop, D.G. Phinney, B.A. Bunnell, Methods in Molecular Biology 449, Mesenchymal stem cells, Totowa: Humana Press), or in the culture medium as described in US8557578 B2.

Generation of Vectors for the Expression of Cytokines and Chemokine

[0176] The transgene expression cassettes consisting a promoter and a gene (e.g. cDNA) for an immunostimulatory factor or factor supporting immune response are constructed using standard cloning techniques as described in Julia Lodge, Peter Lund, Steve Minchin (2007) Gene Cloning, New York: Tylor and Francis Group. The promoters may be constitutive promoters like the CMV promoter or the PGK promoter or inducible promoters like Tie2, RANTES or the HSP70 promoter.

[0177] Examples for genes encoding immunostimulatory factors or factors supporting immune responses are IL-2, IL-7, IL-15, IL-21, IL-12, IFN gamma, IFN beta, SDF-1, CCL23, CCL19, CCL1, CCL2, CCL17, CCL22 and/or CCL4 (Strengell et al., M, The Journal of Immunology,2003, 170: 5464 -5469; Borish et al., J Allergy Clin Immunol. 2003 Feb;111 (2 Suppl): S460-7). The gene may or may not be fused with tag-sequences (e.g. marker proteins / peptides like the hemagglutinin-tag or the HIS-tag) to allow easy detection of expression later on (Hinrik Garoff, 1985, Annual Review of Cell Biology, Vol. 1: 403-445).

[0178] The transgene is then inserted into a suitable vector system (e.g. lentiviral or gamma-retroviral vector) by standard cloning techniques. A suitable vector is for example described by Baum (EP 1757703 A2). The vector may or may not include a second transgene cassette consisting of a promoter and a selectable marker gene (cell surface marker or resistance gene, for example the pac gene to confer puromycin resistance) to allow enrichment of genetically modified cells later in the process (David P. Clark, Nanette J. Pazdernik, 2009, Biotechnology: Applying the Genetic Revolution, London: Elsevier).

Genetic Modification of Mesenchymal Stem Cell

[0179] The transduction is performed with modifications as described by Murray et al., 1999 Human Gene Therapy. 10(11): 1743-1752 and Davis et al., 2004 Biophysical Journal Volume 86 1234-1242. In detail:

[0180] 6-well cell culture plates (e.g. Corning) are coated with Poly-L-Lysine (PLL) (e.g. Sigma-Aldrich, P4707-50ML): The PLL solution (0.01%) is diluted to final concentration between 0.0001% and 0.001% with PBS. 2 ml of the diluted PLL are used for each well. The plate is incubated at least for 2 h at room temperature. After incubation, the plates are washed carefully with PBS.

[0181] Viral vector supernatant in a final volume of 2 ml is added to each PLL-coated well. The number of particles should between 2×10e3 and 1×10e6 per well, which will result in multiplicity of infection of 0.25 and 10. The loaded plate is centrifuged for 2000 x g, 30 min, 4° C. Afterwards the supernatant is discarded and 1×10e5 mesenchymal stem cells are seeded per well. The plates are incubated at 37 ° with 5% CO2 for further use.

Combinatorial Treatment of a Mouse Tumor Model With MSCs and Antigen Specific CAR-T Cells Demonstrates Synergistic Effect

[0182] Immune-competent mice that express the CEA transgene (CEAtg) in the intestinal and pulmonary tracts (Chmielewski et al.; Gastroenterology. 2012 Oct;143(4):1095-107.e2. doi: 10.1053/j.gastro.2012.06.037. Epub 2012 Jun 27.) are given intrapancreatic injections of Panc02 CEA(+) cells (express CEA and click beetle luciferase) and tumors are grown for 10 days. Mice are then given single intravenous injections of T cells engineered to express a chimeric antigen receptor (CAR) with specificity for CEA either alone, or together with wt MSCs (not genetically modified), or with MSCs overexpressing and secreting IL7 and IL21. WT MSCs or MSCs expressing IL7 and IL21 are also injected alone without co-injection of CAR T cells. Injection of the anti-CEA CAR T cells alone slightly reduces the size of pancreatic tumors, whereas injection of WT MSCs alone has no effect on the tumor. MSCs expressing IL7 and IL21 has only slight anti-tumor effects. However, combinatorial injection of anti-CEA CAR T cells with MSCs (WT or IL7/IL21 expressing MSCs) reduces tumor size below the limit of detection in all mice and produces long-term tumor eradication.

[0183] The examples are supplemented by the figures, in which the surprising effect of the MSCs described herein without genetic modification is demonstrated with respect to enhancing the anti-tumor effect of CAR-Ts.

Combinatorial Treatment of a Mouse Tumor Model With MSCs and NK Cells Demonstrates Synergistic Effect

[0184] Immune-deficient mice are subjected to intracranial injection of Glioblastoma cell line U87 expressing luciferase and tumors are grown for 3 days. Mice are then subjected to single intravenous injections of isolated human NK-cells. In addition, mice receive intracranial injections with wt MSCs (not genetically modified), or with MSCs overexpressing and secreting IL7 and IL21. WT MSCs or MSCs expressing IL7 and IL21 are also injected alone, without co-injection of NK cells. Injection of the NK cells alone slightly reduces the size of pancreatic tumors, whereas injection of WT MSCs alone has no effect on the tumor. MSCs expressing IL7 and IL21 has only slight anti-tumor effects. However, combinatorial injection of NK cells with MSCs (WT or IL7/IL21 expressing MSCs) reduces tumor size below the limit of detection in all mice and produces long-term tumor eradication (FIG. 6).

Combinatorial Treatment of a Mouse Tumor Model With MSCs and Adoptive T Cell Transfer Demonstrates Synergistic Effect

[0185] Immune-deficient mice are subjected to intrahepatic injections of HUH7 liver carcinoma cells expressing luciferase and tumors are grown for 14 days. Mice are then subjected to single intravenous injections of isolated human T cells or together with wt MSCs (not genetically modified), or with MSCs overexpressing and secreting IL7 and IL21. WT MSCs or MSCs expressing IL7 and IL21 are also injected alone without co-injection of T cells. Injection of the T cells alone slightly reduces the size of pancreatic tumors, whereas injection of WT MSCs alone has no effect on the tumor. MSCs expressing IL7 and IL21 has only slight anti-tumor effects. However, combinatorial injection of NK cells with MSCs (WT or IL7/IL21 expressing MSCs) reduces tumor size below the limit of detection in all mice and produces long-term tumor eradication.

Combinatorial Treatment of a Mouse Tumor Model With MSCs and the Check Point Inhibitor (Anti-PD-L1 Antibody) Therapy Demonstrates Synergistic Effect

[0186] The experiment can be performed as described by Deng et al. (Deng et al. (2014) Irradiation and anti-PD-L1 treatment synergistically promote antitumor immunity in mice, J Clin Invest. 2014;124(2):687-695) with modifications:

[0187] Six- to 8-week old BALB/c mice are purchased from Harlan. All mice are maintained under specific pathogen-free conditions. The cell line MC38 is a colon adenocarcinoma cell line. MC38 tumor cells (1 × 106) are injected s.c. into the flanks of mice. MC38 are allowed to grow for about 8 days. For the PD-L1 blockade experiment, 100 .Math.g anti-PD-L1 antibody (clone 1 0F.9G2; Bio-XCell) is administered i.p. to mice every 3 days for a total of four times. In addition, mice receive MSC treatment alone (up to 3 injections with doses between 1×10^3 - 1×10^6 MSC per mouse) or in combination with anti-PD-L1. Anti-PD-L1 treatment leads to a decrease in tumor volume. Combination of anti-PD-L1 and MSC leads essentially to eradication of the tumor, while treatment with MSC alone only leads to slight decrease in tumor volume.

Combinatorial Treatment of a Mouse Tumor Model With MSCs and IDO Inhibitor Therapy Demonstrates Synergistic Effect

[0188] The experiment can be performed as described by Nakamura et al.( Nakamura et al.(2015) Effects of indoleamine 2,3-dioxygenase inhibitor in non-Hodgkin lymphoma model mice.lnt J Hematol.102(3):327-34) with modifications:

[0189] Female BALB/c mice aged 6-8 weeks are obtained from Charles River. A20, a BALB/c B-cell lymphoma cell line originally derived from a spontaneous reticulum cell neoplasm can be obtained from American Type Culture Collection (Manassas, VA, USA). The cell line is cultured in complete RPMI 1640 medium with 10% FBS, 2 mM 1-glu-tamine, 100 U/mL penicillin, 100 U/mL streptomycin, and 50 .Math.M 2-ME at 37° C. in a humidified 5% CO2 incubator.

[0190] To establish tumors, 1 × 10e6 A20 cells are resuspended in 200 .Math. L of PBS and s.c. injected into the lower back of naïve syngeneic BALB/c mice. Seven days later treatment is initiated. The IDO inhibitor 1-methyl-d-tryptophan (d-1 MT) (Sigma-Aldrich, St. Louis, MO, USA) is dissolved in sterile water (5 mg/mL) and adjusted to pH 9.9 with NaOH. The mice were fed with d-1 MT-supplemented water. In addition, mice received injections of MSCs alone (up to 3 injections with doses between 1×10^3 - 1×10^6 MSC per mouse) or in combination with d-1 MT -supplemented water. Tumor growth is monitored two times per week by caliper measurement. The tumor volume are calculated. The mice are killed by cervical dislocation on day 28. Compared to untreated control mice d-1 MT treatment results in reduced tumor growth. Treatment with MSC alone leads to slight tumor reduction. Combination of the IDO inhibitor with MSC leads essentially to tumor eradication in the mice.

Native MSC Support CD4+ T Cell Proliferation and Activation in Vitro

[0191] As shown in FIG. 7, the percentage of activated and proliferating CD4 cells is increased in the presence of MSC after co-culturing in vitro, as seen in the FCS/SSC FACS analysis. Furthermore, the release of IFNg as sign of CD4-T cell activation was increased by the presence of MSC in a dose dependent manner, indicating the ability of MSCs to activate CD4 T cells.

Genetically Engineered Msc Which Express the Transgene Alpha-1 Anti-Trypsin (Which Do Not Expressing Immune-Stimulating Cytokines From the Exogenous Nucleic Acid) Activate T Cell Proliferation In A Comparable Manner As Native Msc

[0192] As shown in FIG. 8, the co-culture of PBMC with either (A) native or (B) genetically modified MSC lead to an increase in the proliferation of PBMC. In each case a larger fraction of the lymphocytes had undergone 1 or more doublings compared to the sample without any MSCs. This provides support that genetically modified MSCs, without exogenous immune-stimulatory cytokines, are also capable of stimulating immune cell responses. Furthermore, the percentage of PBMC that had not undergone any doublings was reduced in the presence of MSC.