Chimeric antigen receptor constructs and uses thereof

Abstract

The present invention relates to a polynucleotide molecule comprising a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen-binding region which binds to CD19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPγ)-related protein (SGRP). The present invention relates further to a chimeric antigen receptor (CAR)-T cell expressing the polynucleotide molecule and methods of treating cancer using the polynucleotide molecule and / or the chimeric antigen receptor (CAR)-T cell expressing the polynucleotide molecule.

Description

[0001] Chimeric antigen receptor constructs and uses thereof

[0002] The field of the invention

[0003] The present invention relates to a polynucleotide molecule comprising a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SZRPy)-related protein (SGRP). The present invention relates further to a chimeric antigen receptor (CAR)-T cell expressing the polynucleotide molecule and methods of treating cancer using the polynucleotide molecule and / or the chimeric antigen receptor (CAR)-T cell expressing the polynucleotide molecule.

[0004] Background of the invention

[0005] Glioblastoma (GBM) is an aggressive malignant primary brain tumor resistant to the current standard of care (SOC). The hard-to-access localization and infiltrative nature of GBM often make complete surgical resection unachievable. Moreover, chemo- and radiotherapy regimens are not curative, invariably leading to recurrent disease. The survival time for patients with GBM is around 15 months, underscoring the need for a significant breakthrough in effective medical treatment1 3.

[0006] Chimeric antigen receptor (CAR) T cell-based immunotherapies have had remarkable outcomes in the clinical treatment of hematological malignancies, yet the development of effective CAR T cell therapies against solid tumors remains challenging4’5. A critical limitation of CAR T cells is the scarcity of known tumor-specific surface antigens and their heterogeneous expression profiles in GBM6. One of the most well-studied target antigens in GBM is EGFRvIII, a tumorspecific, mutated form of EGFR expressed in approximately 40% of GBM cases7,8. Although strictly expressed on tumor cells, EGFRvIII mutations arise with concomitant EGFR amplification during clonal evolution events in GBM development, resulting in EGFRvIII- mosaic tumors8,9. A clinical trial of anti-EGFRvIII CAR T cells against recurrent GBM (NCT01454596) showed safety and transient efficacy but failed to produce long-lasting treatment responses due to adaptive resistance and antigen escape10. Conversely, targeting more homogeneously-expressed GBM-associated antigens is controversial due to the potential risk of on-target / off-tumor cross-reactions leading to toxicity11,12.

[0007] Immune checkpoint inhibitors have recently shown promising responses against solid tumors13. However, the highly immunosuppressive tumor microenvironment (iTME) of GBM severely limits the efficacy of immune checkpoint blockade (ICB)14. Thus, understanding the complex context-dependent interactions of GBM with the surrounding iTME is crucial for the effective targeting of these tumors by immunotherapeutic approaches15.

[0008] The most important and numerous immune cells that populate GBM are pro-inflammatory and proliferative brain-resident microglia, peripheral monocyte-derived macrophages, polymorphonuclear myeloid-derived suppressor cells and Tregs16,17. As a major immune cell population in the GBM-iTME, glioma-associated macrophages and microglia (GAMs) substantially contribute to GBM progression11. Microglia are professional phagocytes of the brain. They play an important role in the brain's innate immune surveillance and strongly influence the outcome and response to pathological states through the release of cytokines, chemokines, and growth factors18,19. The phagocytic activity of both microglia and macrophages is regulated through, among others, the CD47-SIRPa phagocytosis axis, whereby SIRPa, expressed on the surface of microglia and macrophages, interacts with the ubiquitously- expressed CD47 transmembrane protein, thereby inhibiting phagocytosis20,21. Therefore, CD47 is an innate immune checkpoint co-opted by tumor cells as a ‘don't eat me’ signal, which results in immune evasion by tumor cells through reduced recognition by phagocytic cells11,22. Blockade of CD47 has been shown to rescue GAM phagocytic function in GBM-bearing mice leading to a strong antitumoral response in vivo23 25. However, clinical studies of systemic monotherapy with CD47 blockade have only recently begun to assess efficacy against solid tumors, showing promising clinical activity26,27. The published data from these trials suggest overall safety and considerable activity but also report low bioavailability within the tumor as well as treatment-associated toxicity28.

[0009] Hence the current treatment options for GBM are far from being satisfactory and there is still a high medical need to provide efficient treatments to patients.

[0010] Summary of the invention

[0011] The present invention relates to a polynucleotide molecule comprising a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPY)-related protein (SGRP).

[0012] The inventors of the present invention have developed a new combinatorial approach of intratumoral (i.t.) CAR T cell therapy and GAM modulation for an additive elimination of GBM with a fourth-generation CAR design, whereby anti-EGFRvIII CAR T cells constitutively release a soluble SIRPy-related protein (SGRP) with high affinity to CD4729. It has been found by the inventors of the present invention that the combination of tumortargeting CAR T cells with SGRP dramatically improved the elimination of EGFRvIII-mosaic GBM in vivo in orthotopic GBM and peripheral lymphoma xenograft mouse models.

[0013] The inventors herewith provide the present invention in its following aspects.

[0014] In one aspect, the present invention provides a polynucleotide molecule comprising a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPY)-related protein (SGRP).

[0015] In a further aspect, the present invention provides an amino sequence comprising a chimeric antigen receptor (CAR) or a fragment thereof comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); wherein the nucleotide sequence encoding the chimeric antigen receptor (CAR) is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPY)-related protein (SGRP). In a further aspect, the present invention provides an amino sequence comprising a chimeric antigen receptor (CAR) or a fragment thereof comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPy)-related protein (SGRP).

[0016] In a further aspect, the present invention provides a construct comprising the polynucleotide molecule described herein.

[0017] In a further aspect, the present invention provides a chimeric antigen receptor (CAR)-T cell comprising T cells expressing the polynucleotide molecule and / or the construct described herein.

[0018] In a further aspect, the present invention provides a method of treating cancer in a subject suffering from an epidermal growth factor receptor (EGFR)-associated cancer or suffering from a CD19-associated cancer, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell described herein.

[0019] In a further aspect, the present invention provides a method of treating a solid cancer in a subject, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell described herein.

[0020] In a further aspect, the present invention provides a method of treating cancer in a subject, wherein the cancer is located in the central nervous system and wherein the method comprises administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII).

[0021] In a further aspect, the present invention provides a method of treating cancer in a subject, wherein the cancer is selected from the group consisting of a breast cancer, lung cancer, melanoma, lymphoma, acute lymphocytic leukemia (ALL), and non-Hodgkin’s lymphoma (NHL), preferably lymphoma, and wherein the method comprises administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell described herein, whererin the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to CD 19.

[0022] In a further aspect, the present invention provides a method of preparing a population of activated T cells comprising a polynucleotide molecule described herein or a construct described herein, the method comprising: (i) contacting in vitro one or more T cells that have been modified to comprise a polynucleotide molecule described herein or a construct described herein with a stimulus that induces expansion of the T cells to provide an expanded T cell population; and (ii) activating in vitro the T cells to produce an activated T cell population.

[0023] In a further aspect, the present invention provides a nucleotide sequence encoding a signal regulatory protein gamma (SIRPyj-related protein (SGRP) comprising the sequence as shown in SEQ ID NO: 5.

[0024] In a further aspect, the present invention provides an amino sequence comprising a signal regulatory protein gamma (SIRPy)-related protein (SGRP) comprising the sequence as shown in SEQ ID NO: 6.

[0025] Brief description of the figures

[0026] Fig. 1: Proposed mechanism of action of conventional aEGFRvIII CAR T cell monotherapy (left panel) and aEGFRvIII- SGRP CAR T cell combination therapy whereby SGRP-mediated CD47 blockade induces phagocytic modulation of GAMs (right panel) in the context of EGFRvIII-heterogenous GBM and its iTME.

[0027] Fig- 2 : Outline of the SGRP engineering strategy including specific AA substitutions to the endogenous human SIRPy-Vl sequence and addition of an N-terminal IL-2 signal sequence (IL2sig) leading to constitutive SGRP secretion. Full CAR sequences are listed in Table 2. Fig. 3: AlphaFold30’31-generated in silico modeling displaying predicted protein-protein interactions of SGRP, hSIRPa-Vl and mSIRPa-Vl with hCD47. Amino acid sequence information regarding hCD47, IL2sig, hSIRPa-Vl, hSIRPy-Vl, IL2sig-SGRP, mSIRPa-Vl and SGRP are listed in Table 1.

[0028] Fig. 4: Maps of polycistronic lentiviral constructs encoding mCherry (mC)-labeled aCD19 CAR or aEGFRvIII CAR under the control of EF1A promoter + / - SGRP secretion.

[0029] Fig- 5 : Overview of lentiviral vector maps used in the study. Nucleotide sequence information for a CD19 (FMC63.BBz), aEGFRvIII (3C10.BBz), aEGFRvIII- SGRP and aEGFRvIII- SGRP-His CAR constructs are listed in Table 2.

[0030] Fig. 6: Representative T cell lentiviral transduction efficiency for the different CAR lentiviral vectors used, based on the percentage of mCherry+T cells assessed by flow cytometry 4 days after transduction (gated on live single cells). Mock-transduced T cells served as controls. Fig- 7 : Post-sort enrichment of CAR T cells after sorting for mCherry+cells. These were subsequently expanded and used for downstream experiments.

[0031] Fig. 8: Representative plots of CAR surface expression on aCD19 CAR and aEGFRvIII CAR T cells and assessed by FC using streptavidin (SA)-FITC labeling of biotinylated (bt)-CD19 (top plots) or bt-EGFRvIII (bottom plots) bound to the respective CAR; n = 2 HDs.

[0032] Fig. 9: aCD19 or aEGFRvIII CAR T cells were exposed to wt-EGFR using biotinylated (bt) EGFR recombinant protein. No binding of CARs to bt-EGFR by flow-cytometric analysis of streptavidin (SA)-FITC could be detected.

[0033] Fig. 10: TATA-box binding protein (TBP)-normalized expression of mCherry and SGRP detected by real-time quantitative PCR (RT-qPCR) in aEGFRvIII CAR or aEGFRvIII- SGRP CAR T cells, showing mCherry expression in CARs transduced with either construct and SGRP expression specifically in aEGFRvIII- SGRP CARs; n = 4 HDs. Statistical differences were assessed by unpaired t tests with Welch’s correction.

[0034] Fig. 11: Differentially secreted proteins in CAR T cell-conditioned media, detected by LC- MS, of resting aEGFRvIII- SGRP CAR vs aEGFRvIII CAR, highlighting the presence of SGRP exclusively in aEGFRvIII- SGRP CAR T cell-conditioned media; n = 2 HDs. Mean SGRP expression: -log l Ot|Vall'71og2lold-clia,lgc= 8.60 / 6.65.

[0035] Fig- 12: Volcano plots of healthy-donor (HD) specific differential supernatant secretome analysis by LC-MS of aEGFRvIII- SGRP CAR vs aEGFRvIII CAR T cells. Log2 (Fold change) indicates the mean expression level for each protein. Each dot represents one protein. The -loglO (qValue) represents the adjusted level of significance for each protein. SGRP is highly enriched in aEGFRvIII-SGRP CAR T cell-conditioned media from both healthy donors assessed. The raw expression data is available in Table S3.

[0036] Fig. 13: Representation of SGRP sequence coverage in LC-MS. SGRP was identified by 7 exclusive unique peptides and 8 exclusive unique spectra covering 47% of the 119 AA sequence.

[0037] Fig. 14: Flow cytometric representation of surface target expression of tumor cell lines U251vIII, U251, U87, BS153, Raji and control neural stem cell line NSC197. Note: all subsequent in vitro and in vivo experiments encompassing EGFRvIIF U251 and U87 were performed using sorted cells on the EGFRvIIF population.

[0038] Fig. 15: Cell culture microphotographs overlaying brightfield and fluorescent detection of nuclear (n) EGFP-reporter expression after lentiviral, puromycin-selective transduction of target cells BS153, U251 and U251vIII for later time-lapse co-culture experiments; Scale bar: 250 pm.

[0039] Fig. 16: Assessment of CAR T cell on-target killing capacity by co-culture time-lapse of nEGFP+U251vIII with mCherry+target-specific (aEGFRvIII CAR + / - SGRP) or nonspecific (aCD19 CAR + / - SGRP) at a 1 : 1 E:T ratio for 72 h.

[0040] Fig. 17: 72 h timelapse co-culture experiment of endogenously EGFRvIIF BS153 cells with aCD19 or aEGFRvIII CARs, respectively. Tumor confluence (cells / well) is plotted over time. Curves represent the mean of duplicate measurements. Differences between each co-culture compared to the ‘tumor alone’ control condition were analyzed using one-way ANOVA and Dunnett’s multiple comparisons tests.

[0041] Fig. 18: Tumor confluence (Green nuclei per well) in co-cultures of EGFRvIIF U251 with either aCD19, aCD19-SGRP, aEGFRvIII or aEGFRvIII- SGRP CAR T cells in 3 different effectortarget ratios at 24, 48 and 72 h timepoints. Differences between each co-culture compared to the ‘tumor alone’ control condition were analyzed using one-way ANOVA and Dunnett’s multiple comparisons tests.

[0042] Fig. 19: Dose-dependent CAR T cell killing capacity in a co-culture with U251vIII at defined time points. Straight horizontal lines represent the mean confluence in control wells with U251vIII only.

[0043] Fig. 20: Tumor confluence (Green nuclei per well) in co-cultures of EGFRvIIF U251 with either aCD19, aCD19-SGRP, aEGFRvIII or aEGFRvIII- SGRP CAR T cells in 3 different effector-target ratios at 24, 48 and 72 h timepoints. Fig. 21: FC assessment of CD4 / CD8 T cell phenotypes of aCD19, aCD19-SGRP, aEGFRvIII and aEGFRvIII- S GRP CAR T cells used in this study (gated on live singlets) at the time of experimental use. One representative healthy donor CAR T cell batch is shown.

[0044] Fig. 22: Representative histograms displaying the percentage of CAR T cell degranulation (CD107a expression) in co-cultures with U251vIII, BS153, U251 or U87 GBM cell lines for 24 h (gated on live mCherry+singlets). Conditions were performed in duplicates with n = 2 HDs and the experiment was repeated once

[0045] Fig. 23: fFNy release detected by ELISA in supernatants of CAR T cells co-cultured with GBM cell lines expressing endogenous EGFRvIII (BS153), overexpressing EGFRvIII (U251vIII) or EGFRvIIF (U251 and U87). Differences in fFNy release between EGFRvIII and EGFRvIII' co-cultures with each CAR were determined using Two-way ANOVA and Sidak’s multiple comparisons tests. Conditions were performed in triplicates with n = 2 HDs and the experiment was repeated once.

[0046] Fig. 24: Schematic (top panel) of the experimental setup of an SGRP / aCD47 blocking assay on CD47+target tumor cells where (1) BS153 were treated with aEGFRvIII- S GRP CAR conditioned-medium or aCD47, (2) exposed to bt-SIRPa which competitively bound to available CD47, and (3) CD47-bound bt-SIRPa was assessed by SA-FITC and MFIs calculated; Representative dot-plots (bottom panel) depicting CD47-blocking capacity determined by FC-based detection of SA-FITC coupled to GBM-bound bt-SIRPa. Conditions were performed in triplicates and the experiment was repeated once.

[0047] Fig. 25: FC representation of surface CD47 expression of aEGFRvIII CAR and aEGFRvIII- SGRP CAR T cells from 3 HDs. Unstained T cells served as a negative control.

[0048] Fig. 26: FC assessment of phagocytosis and macrophage polarization / effector function in cocultures of EGFRvIII-mosaic tumor cells with donor-matched macrophages and CAR T cells from 4 HDs. Heatmap showing the MFI of markers and the fractions of phagocytosed U87 and U251vIII cells. Gated on CD1 lb+cells. All differences between conventional CARs and SGRP-secreting CARs were not statistically significant using one-way ANOVA and Tukey’s multiple comparisons tests.

[0049] Fig. 27: Experimental setup of the EGFRvIII xenograft GBM tumor model and subsequent monotherapeutic CAR and antibody treatment schemes. 7 and 14 d after orthotopic tumor implantation with U251vIII-NLuc tumor cells, animals were treated with either intratumoral (i.t.) CAR T cells or antibodies followed by BLi and scoring until the humane endpoint was reached. Fig. 28: Overview of experimental groups / therapeutic conditions and treatment dosages

[0050] Fig. 29: Kaplan-Meier plot of overall survival (in days). Log-rank tests were used to compare selected treatment / control groups.

[0051] Fig. 30: Tumor progression in EGFRvIII xenografts was monitored for each individual mouse using BLi time course.

[0052] Fig. 31: Experimental treatment and monitoring schedule for EGFRvIII mosaic model. Animals were treated twice - at 7 and 14 days - after intracranial tumor implantation using the same stereotactic coordinates, and routinely monitored for clinical signs, weekly dual bioluminescence imaging (BLi) and morbidity / survival assessment. Plasma for cytokine analysis was collected on day 15 - 24 h after the second intratumoral (i.t.) treatment. Anti- CD47 therapy was prolonged for 4 additional intraperitoneal (i.p.) injections on days 19, 22, 26 and 29. Animals were euthanized upon reaching the humane endpoint. Upon reaching 90 d tumor-free survival, 5 aEGFRvIII-SGRP CAR-treated animals were tumor-rechallenged in the contralateral hemisphere using the same stereotactic coordinates.

[0053] Fig. 32: Experimental setup of orthotopic xenograft experiments in NSG mice encompassing co-implanted EGFRvIII U251vIII and EGFRvIII' U87 GBM cell lines mimicking tumor heterogeneity, and therapeutic / control cohorts including local CAR T cell or antibody monotherapies or combinations with local SGRP or local and systemic CD47 blockade.

[0054] Fig. 33: Kaplan-Meier plot of overall survival (in days). Log-rank tests were used to compare selected treatment / control groups. Survival and bioluminescence data were pooled from 3 independent experiments.

[0055] Fig. 34: Kaplan-Meier plot of tumor-free survival (in weeks), combining survival assessment with BLi monitoring scores. Log-rank tests were used to compare selected treatment / control groups. Survival and bioluminescence data were pooled from 3 independent experiments. Fig. 35: Tumor progression in EGFRvIII-mosaic xenografts was monitored for each individual mouse using differential BLi time course imaging with either FFz or D-luciferin substrates (in weeks); U251vIII NLuc-reporter BLi curves are depicted here.

[0056] Fig. 36: Tumor progression in EGFRvIII-mosaic xenografts was monitored for each individual mouse using differential BLi time course imaging with either FFz or D-luciferin substrates (in weeks); U87 Luc2-reporter BLi curves are depicted here.

[0057] Fig. 37: Cumulative differential monitoring in weeks by BLI for both grafted mosaic EGFRvIII and EGFRvIII' tumors per experimental condition as outlined in Fig. 32 Growth of U251vIII+tumors was measured by luminescence elicited by FFz, whereas growth of U87vlir tumors was detected by D-luciferin luminescence. Curves end whenever the humane endpoint was reached.

[0058] Fig. 38: Representative overlay images of dual BLi studies at week 7 of aEGFRvIII CAR, aEGFRvIII CAR + aCD47 and aEGFRvIII- S GRP CAR treated animals highlighting the suppression of EGFRvIIF tumors in aEGFRvIII- S GRP CAR treated animals.

[0059] Fig. 39: Quantification of BLi signal intensities (mean photon counts) as a surrogate for tumor burden for both EGFRvIII and EGFRvIIF tumors from Fig.38 at 7 weeks after tumor implantation. Each dot represents an individual mouse. Statistical comparisons were performed with one-way ANOVA with multiple comparisons corrections.

[0060] Fig. 40: Outline of in vivo experiment comparing the effect of aCD19-SGRP CAR and aEGFRvIII- S GRP CAR on the GBM iTME in the context of EGFRvIIF or EGFRvIII-mosaic intracerebral tumors.

[0061] Fig. 41: Heatmap of scaled median cluster-defining cell lineage marker expression on 10 immune cell populations indicated on the left y-axis. Cluster frequency is indicated on the right y-axis.

[0062] Fig. 42: Frequency of immune cell clusters in the tumor-injected brain hemispheres of U87 or U87+U25 IvIII tumor-engrafted animals treated with aCD19-SGRP CAR or aEGFRvIII- SGRP CAR (n = 5 per group).

[0063] Fig. 43: Kaplan-Meier plot of overall survival of aEGFRvIII- S GRP CAR-cured, tumor- rechallenged animals (in days). A Log-rank test was used to compare the rechallenge group to the historic Vehicle control group.

[0064] Fig. 44: Dual BLi monitoring of initially cured, mosaic tumor rechallenged mice in weeks after tumor rechallenge.

[0065] Fig. 45: IHC micrograph of a brain section from an aEGFRvIII- S GRP CAR-treated mouse that remained tumor-free until day 90 post tumor implantation, showing DAPI-stained cell nuclei (blue) and human CD3-stained grafted CAR T cells (brown); Left: Overview of brain section; Scale bar: 1 mm; Right: Close-up of the region defined by the white insert on the overview; White arrowheads indicate the location of CAR T cells; Scale bar: 50 pm.

[0066] Fig. 46: Frequency of mCherry-labeled aEGFRvIII- S GRP CAR T cells gated on single live cells (Zombie Aqua') in the brains of animals that remained tumor-free until day 90 post tumor implantation; Three brain regions were dissected and processed separately into singlecell suspensions: tumor-implanted (t.i.) hemisphere, contralateral (c.l.) hemisphere, and meninges; Each dot represents one animal; Statistical comparisons were performed using a one-way ANOVA with Tukey’s multiple comparison tests.

[0067] Fig. 47: Clustered heatmap of targeted post-therapy plasma proteomics showing the relative expression of proteins between all treatment groups; the analysis included samples from complementary datasets which were bridged and normalized (from lighter to darker grey: dataset 1, dataset 2, bridge samples). Individual protein expressions across the datasets are depicted on the y-axis, each cell represents the Z-score of all the measurements in this row. Data are clustered according to Euclidean distance. The barplot on the right shows the proportion of measurements under the limit of detection for each protein in each dataset. Fig. 48: Non-metric multidimensional scaling (NMDS) plot of 92 examined human soluble proteins in mouse plasma determined by proximity extension assay. Each individual mouse is represented by one data point, with symbols and colors differentiating healthy controls (Healthy, white; n = 6), mock-injected (Vehicle, black; n = 6), mouse IgGl -treated (Isotype, yellow; n = 6), anti-human CD47-treated (aCD47, orange; n = 6), aCD19 CAR-treated (red; n = 6), aCD19-SGRP CAR-treated (dark red; n = 6), aEGFRvIII CAR-treated (blue; n = 6), aEGFRvIII CAR + aCD47-treated (green; n = 6) or aEGFRvIII- S GRP CAR-treated; (dark blue; n = 6). All samples were collected approximately 24 h after the last local treatment. The ellipses represent the 95% confidence interval of each treatment group. Source data are provided in Table S7.

[0068] Fig. 49: Volcano plot comparing aEGFRvIII- S GRP CAR and aEGFRvIII CAR treatment groups, showing significant enrichment of immune markers CCL3, IL13, and reduction of CD27. Significant differences in protein expression are represented by symbols: adj. P < 0.001, star; adj. P < 0.05, pentagon; P < 0.05, triangle; P > 0.05 or fold change (FC) < 0.5, dot.

[0069] Fig. 50: Box plots showing the normalized protein expression of significant innate immune surrogate markers in plasma (CD27, CCL3, IL13, ARG1, ILIA and IFNy). Each data point represents one mouse. Statistics were calculated using two-sided Mann-Whitney-U tests for the comparisons of interest (aEGFRvIII CAR vs aEGFRvIII CAR + aCD47, aEGFRvIII CAR vs aEGFRvIII- S GRP CAR, aEGFRvIII CAR vs aCD19 CAR, aCD19-SGRP CAR vs aEGFRvIII- S GRP CAR, aCD19 CAR vs aCD19-SGRP CAR, aEGFRvIII CAR + aCD47 vs aEGFRvIII- S GRP CAR, aCD47 vs aEGFRvIII CAR + aCD47) with Benjamini-Hochberg correction. Only significant statistics are shown. Fig. 51: Experimental treatment and monitoring schedule of the CCL3 blockade in vivo cohort. Animals were treated i.t. twice - at 7 and 14 days - after i.c. implantation of EGFRvIII U251vIII and EGFRvIII' U87 GBM cell lines. Tumor and CAR T cells were injected using the same stereotactic coordinates. Systemic aCCL3 therapy was administered 3x per week for 5 weeks, starting on day 8. Animals were euthanized upon reaching the humane endpoint or at day 90. CAR T cell dose: 5 x 105cells delivered i.t.; Antibody dose: 50 ng delivered i.p.

[0070] Fig. 52: Kaplan-Meier plot of overall survival (in days). Log-rank tests were used to compare the indicated treatment / control groups.

[0071] Fig. 53: Overview of experimental groups / therapeutic conditions and treatment dosages for brain multiplex IF.

[0072] Fig. 54: H&E-stained sections of representative tumor-burdened brains at day 21 post tumor implantation and 7 days after the second treatment dose. Frontal sections of cerebrum from different experimental groups: (1) mock-injected, Vehicle, (2) mouse IgGl (MOPC-21), Isotype, (3) anti-human CD47 (B6.H12), aCD47, (4) aCD19 CAR, (5) aEGFRvIII CAR, (6) aEGFRvIII CAR + aCD47, and (7) aEGFRvIII- S GRP CAR; Scale bars: 1000 pm.

[0073] Fig. 55: DAPI nuclear-stained stitched assemblies of brain sections of experimental groups used for subsequent immunofluorescence multiplexing; Scale bars: 1000 pm.

[0074] Fig. 56: Histomorphological brain tumor size assessment at day 21; n = 6 brains per condition, n = 5 brains in aEGFRvIII CAR + aCD47 group. All comparisons were nonsignificant using one-way ANOVA with Dunnett’s multiple comparison tests.

[0075] Fig. 57: Brain collection intermediate post-therapeutic time points for conventional immunohistochemistry per experimental condition. Representative IHC images at d 13 for aEGFRvIII- S GRP CAR treatment and d 21 for all other groups showing CD3+human T cells within the tumor core and adjacent brain (human CD3 IHC). Left column', overview of tumor- burdened brain sections, scale bar: 1 mm; right column', close-up according to the inserts at the tumor-brain interface. Scale bar: 100 pm. Bar graphs: Quantification of CD3+T cells in the tumor rim and tumor core.

[0076] Fig. 58: Pie charts displaying relative comparisons of the percentage of marker-positive cells / all cells within the whole tumor or tumor cores in experimental conditions. Two slides per condition were assessed creating a ratio between overall positive / negative cells. No tumors were detected in any of the aEGFRvIII- S GRP CAR-treated brains analyzed. Fig. 59: Brain collection intermediate post-therapeutic time points for conventional immunohistochemistry per experimental condition. Representative IHC images at d 13 for aEGFRvIII-SGRP CAR treatment and d 21 for all other groups showing CD68 stained brain sections per therapeutic condition. Left column', overview of tumor-burdened brain sections, scale bar: 1 mm; right column', close-up according to the inserts at the tumor-brain interface. Scale bar: 100 pm. Percentage of CD68+cells within the tumor rim (defined as 100 pm expansion around the tumor). Percentage of CD68+cells within the tumor core (delineated based on nuclear stain (or DAPI) density).

[0077] Fig. 60: Assessment of spleen to bodyweight ratio on day 21 post tumor implantation, after two treatments on days 7 and 14. Comparisons of all treatment groups to vehicle were nonsignificant using a one-way ANOVA with Dunnett’s multiple comparisons tests.

[0078] Fig. 61: Weekly weight monitoring of animals in the EGFRvIII-mosaic GBM survival experiment, showing aEGFRvIII CAR, aEGFRvIII CAR + aCD47, and aEGFRvIII-SGRP CAR groups. Dashed vertical lines mark the treatments in weeks 1 and 2 post tumor implantation.

[0079] Fig. 62: Box plots with individual data points showing the normalized protein expression (NPX) of IL6 in plasma the day after the second treatment (day 15 post tumor implantation) determined by proximity extension assay. Each data point represents one mouse (n = 6 per group). Statistics were calculated for the comparisons of conventional CAR to its SGRP- secreting CAR counterpart using two-sided Mann-Whitney-U tests with Benjamini -Hochberg correction; All comparisons were non-significant.

[0080] Fig. 63: Mouse CRP in plasma collected the day after the first treatment (day 8 post tumor implantation), detected by ELISA. Only the aEGFRvIII CAR + aCD47 comparison to vehicle was significant (P = 0.0138) using one-way ANOVA with Dunnett’s multiple comparisons tests.

[0081] Fig. 64: Longitudinal monitoring of hematological parameters: erythrocyte count (RBC), platelet count (PLT) and neutrophil count (NEUT) for up to 20 days after a single treatment. The scale was normalized to Z-score to allow direct comparisons between time points. A mixed effects model was applied to each variable using the lme4 package in R. Post hoc tests for significant effects were conducted using estimated marginal means (emmeans package) to compare conditions within each time point. Significant differences between conditions were found only in neutrophil counts at day 3, 13, and 20. Fig. 65: Luxol fast blue-stained micrographs of mouse brain sections per therapeutic condition at day 7 post treatment (single dose) showing myelin density; Left column: Overview of representative brain sections per condition; White arrowheads indicate the magnified regions of the corpus callosum on the overview of each condition; Scale bars: 1 mm; Right column: Close-up of the regions defined by the inserts per condition; Scale bar: 100 pm.

[0082] Fig. 66: Quantification of myelin-covered brain area on day 7 post therapy; n = 5 brains per condition, n = 6 brains in the aEGFRvIII CAR group. All comparisons were non-significant using one-way ANOVA with Dunnett’s multiple comparison tests, a-g, Mice in all experiments were injected i.c. with a tumor mix of 2.5 x 104U87 and 2.5 x 104U251vIII cells; Each treatment dose consisted of 5 x 105CAR T cells and / or 5 pg antibody injected i.t. Fig. 67: Overview of titrated / optimized antibody panel and conjugated fhiorophores employed in spectral flow cytometry.

[0083] Fig. 68: Left panel: pseudocolor plots of a representative sample gating strategy to yield live (Zombie NIR'), CD45+or CD45' singlets. Right panel: Scatter bar plot depicting influx of overall increased CD45+cells after CAR treatments (n = 6 tumor dissociations per experimental group; each dot represents measurements from an individual animal). Statistics: One-way ANOVA with Tukey’s multiple comparison test * P < 0.05. The gating strategy is detailed in Fig. 71

[0084] Fig. 69: Left panel: Representative pseudocolor plots visualizing mCherry+CAR T cells and mTagBFP2+tumor cells per experimental conditions. Right panel: quantification of CAR T cells and tumor cells per condition. Statistics: One-way ANOVA with Tukey’s multiple comparison test. * P < 0.05.

[0085] Fig. 70: Pseudocolor plots displaying the 4 most prominent CD45+myeloid subsets in the TME of Vehicle or CAR-treated animals, differentiated by expression of P2RY12 and F4 / 80. Bottom plot: Backgating of the 4 populations termed ‘activated microglia’ (dark green, P2RY12intF4 / 80hi), ‘microglia’ (light green, P2RY12hiF4 / 80int), ‘moMacs’ (orange, P2RY12loF4 / 80hi), and ‘granulocytes’ (red, P2RY12'F4 / 80').

[0086] Fig. 71: Gating strategy to yield CD45+or CD45' live singlets. Further myeloid subdivision on CD45+ cells was performed by identifying 4 populations with differential expression of P2RY12 and F4 / 80, respectively, named monocyte-derived cells (MdCs), activated microglia, resting microglia, and neutrophils. Human CAR T cells (moCD45‘) and BFP2+tumor cells were assessed in the CD45' gate. Fig. 72: Representative heatmap overlay of expression values of CD11c (top), CD86 (center) and MHCII (bottom) per experimental condition on the 4 different subsets from Fig.70 .The color bar insert represents MFI values.

[0087] Fig. 73: Quantification of mTagBFP2 signal intensities within different phagocytic subsets. Top panel: Staggered histograms of the median fluorescent intensities of mTagBFP2 for all Vehicle (light blue), aEGFRvIII CAR (dark blue) and aEGFRvIII-SGRP CAR (red) -treated animals. Bottom panel: Scatter plots of mTagBFP2 MFI values per condition in activated microglia, microglia and macrophages. Each dot represents measurements from an individual brain (n = 6 per condition). Statistics: One-way ANOVA with Tukey’s multiple comparison test. Adj. P values * P < 0.05, ** P = 0.0013, *** P < 0.005, **** P < 0.0001.

[0088] Fig. 74: MFI of TNF and MHCII on the activated microglia subpopulation. Statistics: Oneway ANOVA with Tukey’s multiple comparison test. Adj. P values listed.

[0089] Fig. 75: TSNE (top) and UMAP (bottom) plots depicting the initial 15 clusters within the CD45+pre-gated population.

[0090] Fig. 76: Clustered heatmap of scaled median marker expression per cluster. The number of cells per cluster is indicated on the right y-axis.

[0091] Fig. 77: Histograms of marker expression per cluster.

[0092] Fig. 78: Frequency of clusters in aEGFRvIII CAR-treated animals (E, n = 5 animals), and aEGFRvIII- S GRP CAR-treated animals (ES, n = 5 animals).

[0093] Fig. 79: TSNE plot after merging and annotation of initial clustering. 8 clusters (microglia, activated microglia, MoMac, MoDC, monocytes, pDC, neutrophils, and unknowns) were identified.

[0094] Fig. 80: UMAP (plots after merging and annotation of initial clustering. 8 clusters (microglia, activated microglia, MoMac, MoDC, monocytes, pDC, neutrophils, and unknowns) were identified.

[0095] Fig. 81: Individual TSNEs per experimental condition and animal (E = aEGFRvIII CAR, ES = aEGFRvIII- S GRP CAR treatments).

[0096] Fig. 82: Stacked bar plots representing cluster frequencies per condition.

[0097] Fig. 83: Representative heatmap overlay of expression values of CD11c (top), CD86 (center) and MHCII (bottom) per experimental condition on the 4 different subsets identified by conventional gating and flow cytometric analysis. The color bar insert represents MFI values. Fig. 84: Overlay of P2RY12 / F4 / 80 (upper panel) and P2RY12 / CD11c on the TSNE plot from Fig. 79 displaying differential contribution of these markers to the microglia subclusters. Fig. 85: TSNE plot of microglia-specific sub clustering, resulting in 3 microglia populations: resting microglia, MHCII-high microglia and activated microglia. Right upper panel: merged TSNE plot for aEGFRvIII CAR-treated animals; right lower panel: TSNE plot for aEGFRvIII-SGRP CAR treated animals.

[0098] Fig. 86: Histograms displaying individual marker expression per microglia subcluster.

[0099] Fig. 87: Stacked bar plots displaying the frequency distribution of microglia subpopulations per individual experimental animal.

[0100] Fig. 88: Median expression heatmap displaying XCR1 expression levels in microglia subpopulation per experimental condition / individual animal.

[0101] Fig. 89: Overview of median expression heatmaps per marker assessed per microglia subpopulations.

[0102] Fig. 90: EGFRvIII-expression analysis by quantitative real-time PCR (qPCR) in U87, U251 and U251vIII cell lines. RFU: relative fluorescent units.

[0103] Fig. 91: Top panel'. Experimental setup of pharmacoscopy experiment. Batched, frozen single-cell suspensions from GBM patients (tumor center) were thawed and plated at equal numbers into 384 well plates. Beforehand, EGFRvIII status was determined on RNA extracts of matching samples. Single-cell suspensions were co-cultured with aEGFRvIII, aEGFRvIII- SGRP, aCD19 or aCD19-SGRP CAR T cells for 48 h, fixed, stained and imaged via confocal microscopy. Center panel'. Exemplary immunofluorescence readouts of co-cultures of CAR T cells with single-cell suspensions derived from 5 GBM patients; Scale: 100 pm. Bottom panel'. Left: Mean EGFRvIII tumor cell count displayed as fold change relative to the aCD19 CAR control. Middle: Mean Nestin+tumor cell count displayed as fold change relative to the aCD19 CAR control. Right: Mean CD14+tumor cell count displayed as fold change relative to the aCD19 CAR control. Statistics: One-way ANOVA with multiple comparisons correction.

[0104] Fig. 92: Results of EGFRvIII screening by qPCR of patient-derived GBM EGFRvIII positive samples.

[0105] Fig. 93: Results of EGFRvIII screening by qPCR of patient-derived GBM EGFRvIII negative samples.

[0106] Fig. 94: Boxplots summarizing expression of EGFRvIII (upper panel) and EGFR wt (lower panel) in the 2 cohorts.

[0107] Fig. 95: Experimental setup and timeline of interventions of peripheral CD19+lymphoma model treated with systemic CAR T cell infusions. Three days after tumor implantation in the right flank, mice were treated i.v. with CAR T cells, followed by 3 times weekly tumor volume assessment and clinical scoring. Mice were sacrificed upon reaching the humane endpoint.

[0108] Fig. 96: Overview of experimental groups / therapeutic conditions and treatment dosages.

[0109] Fig. 97: Kaplan-Meier plot of overall survival (in days). Log-rank tests were used to compare indicated treatment / control groups.

[0110] Fig. 98: Tumor volume measurements in mm3of individual animals over time (in days post tumor implantation).

[0111] Detailed description of the invention

[0112] As outlined above, the present invention relates to a polynucleotide molecule comprising a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SZRPy)-related protein (SGRP). The present invention relates further to a chimeric antigen receptor (CAR)-T cell expressing the polynucleotide molecule and methods of treating cancer using the polynucleotide molecule and / or the chimeric antigen receptor (CAR)-T cell expressing the polynucleotide molecule.

[0113] For the purposes of interpreting this specification, the following definitions will apply and whenever appropriate, terms used in the singular will also include the plural and vice versa. It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0114] Features, integers, characteristics, compounds described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and / or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. The invention is not restricted to the details of any foregoing embodiments.

[0115] The term “comprise” and variations thereof, such as, “comprises” and “comprising” is generally used in the sense of include, that is, as “including, but not limited to”, that is to say permitting the presence of one or more features or components.

[0116] The singular forms “a,” “an,” and “the” include plural referents unless the content clearly dictates otherwise.

[0117] The term "about" refers to a range of values ± 10% of a specified value. For example, the phrase "about 200" includes ± 10% of 200, or from 180 to 220.

[0118] The term “fragment thereof’ or “fragment” in relation to a chimeric antigen receptor (CAR) refer to functionally active fragments of the CAR, preferably to functionally active fragments of the CAR which are capable of exercising the same physiological function as the CAR.

[0119] Thus, in a first aspect the present invention provides a polynucleotide molecule comprising a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SZRPy)-related protein (SGRP).

[0120] In one embodiment the polynucleotide molecule further comprises a nucleotide sequence encoding a self-cleaving peptide.

[0121] In a preferred embodiment the polynucleotide molecule comprises in the following order from the 5' to the 3' end: a) a promoter; b) a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen binding region which binds to CD 19; wherein the nucleotide sequence encoding the chimeric antigen receptor (CAR) or a fragment thereof is operably linked to the promoter of a); c) a nucleotide sequence encoding a self-cleaving peptide; and d) a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPy)-related protein (SGRP).

[0122] In one embodiment the heterologous signal peptide fused to a signal regulatory protein gamma (SIRPy)-related protein (SGRP) is a signal peptide selected from the group consisting of interleukin 2 (IL-2) signal peptide, interleukin 4 (IL-4) signal peptide, interleukin 9 (IL-9) signal peptide and interferon gamma (IFNy) signal peptide, preferably selected from the group consisting of human IL-2 signal peptide, human IL-4 signal peptide, human IL-9 signal peptide and human IFNy signal peptide.

[0123] In a preferred embodiment the heterologous signal peptide fused to a signal regulatory protein gamma (SIRPy)-related protein (SGRP) is a human interleukin 2 (IL-2) signal peptide.

[0124] In a further preferred embodiment the heterologous signal peptide is fused to the N-terminal region of the signal regulatory protein gamma (SIRPy)-related protein (SGRP).

[0125] In one embodiment the promoter which is operably linked to the nucleotide sequence encoding the chimeric antigen receptor (CAR) or a fragment thereof is the elongation factor 1 alpha (EFl A) promoter or the elongation factor 1 alpha short (EFS) promoter, preferably the EFl A promoter.

[0126] In a more preferred embodiment the chimeric antigen receptor (CAR) or a fragment thereof comprises i) an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) an extracellular domain comprising an antigen binding region which binds to CD 19; wherein the CAR or a fragment thereof further comprises a CD8a leader, CD8a hinge and transmembrane domains, a TNF receptor superfamily member 9 costimulatory domain and a CD3(^ signaling domain. In an even more preferred embodiment the chimeric antigen receptor (CAR) or a fragment thereof comprises i) an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) an extracellular domain comprising an antigen binding region which binds to CD 19; wherein the the CAR or a fragment thereof further comprises a CD8a leader, CD8a hinge and transmembrane domains, a 4-1BB costimulatory domain and a CD3(^ signaling domain.

[0127] In one embodiment i) the extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII) is a single-chain variable fragment (scFv); or ii) the extracellular domain comprising an antigen binding region which binds to CD 19 is a single-chain variable fragment (scFv).

[0128] In one embodiment the self-cleaving peptide is a T2A peptide or a P2A peptide, preferably a T2A peptide.

[0129] In a preferred embodiment the polynucleotide molecule comprises the sequence as shown in SEQ ID NO: 1:

[0130] GGCTCCGGTGCCCGTCAGTGGGCAGAGCGCACATCGCCCACAGTCCCCGAGAAGT TGGGGGGAGGGGTCGGCAATTGAACCGGTGCCTAGAGAAGGTGGCGCGGGGTAA ACTGGGAAAGTGATGTCGTGTACTGGCTCCGCCTTTTTCCCGAGGGTGGGGGAGA ACCGTATATAAGTGCAGTAGTCGCCGTGAACGTTCTTTTTCGCAACGGGTTTGCCG CCAGAACACAGGTAAGTGCCGTGTGTGGTTCCCGCGGGCCTGGCCTCTTTACGGG TTATGGCCCTTGCGTGCCTTGAATTACTTCCACCTGGCTGCAGTACGTGATTCTTG ATCCCGAGCTTCGGGTTGGAAGTGGGTGGGAGAGTTCGAGGCCTTGCGCTTAAGG AGCCCCTTCGCCTCGTGCTTGAGTTGAGGCCTGGCCTGGGCGCTGGGGCCGCCGC GTGCGAATCTGGTGGCACCTTCGCGCCTGTCTCGCTGCTTTCGATAAGTCTCTAGC CATTTAAAATTTTTGATGACCTGCTGCGACGCTTTTTTTCTGGCAAGATAGTCTTG TAAATGCGGGCCAAGATCTGCACACTGGTATTTCGGTTTTTGGGGCCGCGGGCGG CGACGGGGCCCGTGCGTCCCAGCGCACATGTTCGGCGAGGCGGGGCCTGCGAGC GCGGCCACCGAGAATCGGACGGGGGTAGTCTCAAGCTGGCCGGCCTGCTCTGGTG CCTGGTCTCGCGCCGCCGTGTATCGCCCCGCCCTGGGCGGCAAGGCTGGCCCGGT CGGCACCAGTTGCGTGAGCGGAAAGATGGCCGCTTCCCGGCCCTGCTGCAGGGA GCTCAAAATGGAGGACGCGGCGCTCGGGAGAGCGGGCGGGTGAGTCACCCACAC

[0131] AAAGGAAAAGGGCCTTTCCGTCCTCAGCCGTCGCTTCATGTGACTCCACGGAGTA

[0132] CCGGGCGCCGTCCAGGCACCTCGATTAGTTCTCGAGCTTTTGGAGTACGTCGTCTT

[0133] TAGGTTGGGGGGAGGGGTTTTATGCGATGGAGTTTCCCCACACTGAGTGGGTGGA

[0134] GACTGAAGTTAGGCCAGCTTGGCACTTGATGTAATTCTCCTTGGAATTTGCCCTTT

[0135] TTGAGTTTGGATCTTGGTTCATTCTCAAGCCTCAGACAGTGGTTCAAAGTTTTTTT

[0136] CTTCCATTTCAGGTGTCGTGACAAGTTTGTACAAAAAAGCAGGCTGCCACCATGG

[0137] CCTTACCAGTGACCGCCTTGCTCCTGCCGCTGGCCTTGCTGCTCCACGCCGCCAGG

[0138] CCGGGATCCGAGATTCAGCTGCAGCAATCTGGGGCAGAACTTGTGAAGCCAGGG

[0139] GCCTCAGTCAAGCTGTCCTGCACAGGTTCTGGCTTCAACATTGAAGACTACTATAT

[0140] TCACTGGGTGAAGCAGAGGACTGAACAGGGCCTGGAATGGATTGGAAGGATTGA

[0141] TCCTGAGAATGATGAAACTAAATATGGCCCAATATTCCAGGGCAGGGCCACTATA

[0142] ACAGCAGACACATCCTCCAACACAGTCTACCTGCAACTCAGCAGCCTGACATCTG

[0143] AGGACACTGCCGTCTATTACTGTGCCTTTCGCGGTGGAGTCTACTGGGGGCCAGG

[0144] AACCACTCTCACAGTCTCCTCAGGAGGTGGTGGTTCCGGTGGTGGTGGTTCCGGA

[0145] GGTGGTGGTTCACATATGGATGTTGTGATGACCCAGTCTCCACTCACTCTATCGGT

[0146] TGCCATTGGACAATCAGCCTCCATCTCTTGCAAGTCAAGTCAGAGCCTCTTAGAT

[0147] AGTGATGGAAAGACATATTTGAATTGGTTGTTACAGAGGCCAGGCCAGTCTCCAA

[0148] AGCGCCTAATCTCTCTGGTGTCTAAACTGGACTCTGGAGTCCCTGACAGGTTCACT

[0149] GGCAGTGGATCAGGGACAGATTTCACACTGAGAATCAGCAGAGTGGAGGCTGAG

[0150] GATTTGGGAATTTATTATTGCTGGCAAGGTACACATTTTCCTGGGACGTTCGGTGG

[0151] AGGGACCAAGCTGGAGATAAAAGCTAGCACCACGACGCCAGCGCCGCGACCACC

[0152] AACACCGGCGCCCACCATCGCGTCGCAGCCCCTGTCCCTGCGCCCAGAGGCGTGC

[0153] CGGCCAGCGGCGGGGGGCGCAGTGCACACGAGGGGGCTGGACTTCGCCTGTGAT

[0154] ATCTACATCTGGGCGCCCTTGGCCGGGACTTGTGGGGTCCTTCTCCTGTCACTGGT

[0155] TATCACCCTTTACTGCAAACGGGGCAGAAAGAAACTCCTGTATATATTCAAACAA

[0156] CCATTTATGAGACCAGTACAAACTACTCAAGAGGAAGATGGCTGTAGCTGCCGAT

[0157] TTCCAGAAGAAGAAGAAGGAGGATGTGAACTGAGAGTGAAGTTCAGCAGGAGCG

[0158] CAGACGCCCCCGCGTACAAGCAGGGCCAGAACCAGCTCTATAACGAGCTCAATCT

[0159] AGGACGAAGAGAGGAGTACGATGTTTTGGACAAGAGACGTGGCCGGGACCCTGA

[0160] GATGGGGGGAAAGCCGAGAAGGAAGAACCCTCAGGAAGGCCTGTACAATGAACT

[0161] GCAGAAAGATAAGATGGCGGAGGCCTACAGTGAGATTGGGATGAAAGGCGAGCG

[0162] CCGGAGGGGCAAGGGGCACGATGGCCTTTACCAGGGTCTCAGTACAGCCACCAA GGACACCTACGACGCCCTTCACATGCAGGCCCTGCCCCCTCGCGGAAGCGGAGAG

[0163] GGCAGGGGAAGTCTTCTAACATGCGGGGACGTGGAGGAAAATCCCGGCCCCATG

[0164] TACAGGATGCAACTCCTGTCTTGCATTGCACTAAGTCTTGCACTTGTCACGAATTC

[0165] GGAGGAGGAGCTGCAGATCATCCAGCCAGAGAAGCTGCTGCTGGTGACCGTGGG

[0166] CAAGACGGCCACTCTTCATTGTACCATCACCTCTTTGTTCCCCGTGGGTCCCATCC

[0167] AGTGGTTCCGCGGGGTCGGACCCGGGCGGGTGCTCATCTACAACCAGAAGGACG

[0168] GCCACTTTCCTAGGGTCACCACAGTAAGCGACGGCACCAAGCGCAACAATATGG

[0169] ATTTCAGCATTCGCATCTCTTCCATTACTCCGGCGGACGTGGGCACCTATTACTGC

[0170] GTTAAATTTCGTAAGGGATCCCCGGAAGACGTGGAGTTCAAATCCGGCCCTGGTA CGGAGATGGCTCTGGGCGCCAAGCCCTCG.

[0171] In a further preferred embodiment the polynucleotide molecule comprises the sequence as shown in SEQ ID NO: 2:

[0172] GGCTCCGGTGCCCGTCAGTGGGCAGAGCGCACATCGCCCACAGTCCCCGAGAAGT

[0173] TGGGGGGAGGGGTCGGCAATTGAACCGGTGCCTAGAGAAGGTGGCGCGGGGTAA

[0174] ACTGGGAAAGTGATGTCGTGTACTGGCTCCGCCTTTTTCCCGAGGGTGGGGGAGA

[0175] ACCGTATATAAGTGCAGTAGTCGCCGTGAACGTTCTTTTTCGCAACGGGTTTGCCG

[0176] CCAGAACACAGGTAAGTGCCGTGTGTGGTTCCCGCGGGCCTGGCCTCTTTACGGG

[0177] TTATGGCCCTTGCGTGCCTTGAATTACTTCCACCTGGCTGCAGTACGTGATTCTTG

[0178] ATCCCGAGCTTCGGGTTGGAAGTGGGTGGGAGAGTTCGAGGCCTTGCGCTTAAGG

[0179] AGCCCCTTCGCCTCGTGCTTGAGTTGAGGCCTGGCCTGGGCGCTGGGGCCGCCGC

[0180] GTGCGAATCTGGTGGCACCTTCGCGCCTGTCTCGCTGCTTTCGATAAGTCTCTAGC

[0181] CATTTAAAATTTTTGATGACCTGCTGCGACGCTTTTTTTCTGGCAAGATAGTCTTG

[0182] TAAATGCGGGCCAAGATCTGCACACTGGTATTTCGGTTTTTGGGGCCGCGGGCGG

[0183] CGACGGGGCCCGTGCGTCCCAGCGCACATGTTCGGCGAGGCGGGGCCTGCGAGC

[0184] GCGGCCACCGAGAATCGGACGGGGGTAGTCTCAAGCTGGCCGGCCTGCTCTGGTG

[0185] CCTGGTCTCGCGCCGCCGTGTATCGCCCCGCCCTGGGCGGCAAGGCTGGCCCGGT

[0186] CGGCACCAGTTGCGTGAGCGGAAAGATGGCCGCTTCCCGGCCCTGCTGCAGGGA

[0187] GCTCAAAATGGAGGACGCGGCGCTCGGGAGAGCGGGCGGGTGAGTCACCCACAC

[0188] AAAGGAAAAGGGCCTTTCCGTCCTCAGCCGTCGCTTCATGTGACTCCACGGAGTA

[0189] CCGGGCGCCGTCCAGGCACCTCGATTAGTTCTCGAGCTTTTGGAGTACGTCGTCTT

[0190] TAGGTTGGGGGGAGGGGTTTTATGCGATGGAGTTTCCCCACACTGAGTGGGTGGA

[0191] GACTGAAGTTAGGCCAGCTTGGCACTTGATGTAATTCTCCTTGGAATTTGCCCTTT TTGAGTTTGGATCTTGGTTCATTCTCAAGCCTCAGACAGTGGTTCAAAGTTTTTTT

[0192] CTTCCATTTCAGGTGTCGTGACAAGTTTGTACAAAAAAGCAGGCTGCCACCATGG

[0193] CCTTACCAGTGACCGCCTTGCTCCTGCCGCTGGCCTTGCTGCTCCACGCCGCCAGG

[0194] CCGGGATCCGACATCCAGATGACACAGACTACATCCTCCCTGTCTGCCTCTCTGG

[0195] GAGACAGAGTCACCATCAGTTGCAGGGCAAGTCAGGACATTAGTAAATATTTAA

[0196] ATTGGTATCAGCAGAAACCAGATGGAACTGTTAAACTCCTGATCTACCATACATC

[0197] AAGATTACACTCAGGAGTCCCATCAAGGTTCAGTGGCAGTGGGTCTGGAACAGAT

[0198] TATTCTCTCACCATTAGCAACCTGGAGCAAGAAGATATTGCCACTTACTTTTGCCA

[0199] ACAGGGTAATACGCTTCCGTACACGTTCGGAGGGGGGACTAAGTTGGAAATAAC

[0200] AGGCTCCACCTCTGGATCCGGCAAGCCCGGATCTGGCGAGGGATCCACCAAGGG

[0201] CGAGGTGAAACTGCAGGAGTCAGGACCTGGCCTGGTGGCGCCCTCACAGAGCCT

[0202] GTCCGTCACATGCACTGTCTCAGGGGTCTCATTACCCGACTATGGTGTAAGCTGG

[0203] ATTCGCCAGCCTCCACGAAAGGGTCTGGAGTGGCTGGGAGTAATATGGGGTAGTG

[0204] AAACCACATACTATAATTCAGCTCTCAAATCCAGACTGACCATCATCAAGGACAA

[0205] CTCCAAGAGCCAAGTTTTCTTAAAAATGAACAGTCTGCAAACTGATGACACAGCC

[0206] ATTTACTACTGTGCCAAACATTATTACTACGGTGGTAGCTATGCTATGGACTACTG

[0207] GGGTCAAGGAACCTCAGTCACCGTCTCCTCAGCGGCCGCAGCTAGCACCACGACG

[0208] CCAGCGCCGCGACCACCAACACCGGCGCCCACCATCGCGTCGCAGCCCCTGTCCC

[0209] TGCGCCCAGAGGCGTGCCGGCCAGCGGCGGGGGGCGCAGTGCACACGAGGGGGC

[0210] TGGACTTCGCCTGTGATATCTACATCTGGGCGCCCTTGGCCGGGACTTGTGGGGTC

[0211] CTTCTCCTGTCACTGGTTATCACCCTTTACTGCAAACGGGGCAGAAAGAAACTCCT

[0212] GTATATATTCAAACAACCATTTATGAGACCAGTACAAACTACTCAAGAGGAAGAT

[0213] GGCTGTAGCTGCCGATTTCCAGAAGAAGAAGAAGGAGGATGTGAACTGAGAGTG

[0214] AAGTTCAGCAGGAGCGCAGACGCCCCCGCGTACAAGCAGGGCCAGAACCAGCTC

[0215] TATAACGAGCTCAATCTAGGACGAAGAGAGGAGTACGATGTTTTGGACAAGAGA

[0216] CGTGGCCGGGACCCTGAGATGGGGGGAAAGCCGAGAAGGAAGAACCCTCAGGA

[0217] AGGCCTGTACAATGAACTGCAGAAAGATAAGATGGCGGAGGCCTACAGTGAGAT

[0218] TGGGATGAAAGGCGAGCGCCGGAGGGGCAAGGGGCACGATGGCCTTTACCAGGG

[0219] TCTCAGTACAGCCACCAAGGACACCTACGACGCCCTTCACATGCAGGCCCTGCCC

[0220] CCTCGCGGAAGCGGAGAGGGCAGGGGAAGTCTTCTAACATGCGGGGACGTGGAG

[0221] GAAAATCCCGGCCCCATGTACAGGATGCAACTCCTGTCTTGCATTGCACTAAGTC

[0222] TTGCACTTGTCACGAATTCGGAGGAGGAGCTGCAGATCATCCAGCCAGAGAAGCT

[0223] GCTGCTGGTGACCGTGGGCAAGACGGCCACTCTTCATTGTACCATCACCTCTTTGT TCCCCGTGGGTCCCATCCAGTGGTTCCGCGGGGTCGGACCCGGGCGGGTGCTCAT CTACAACCAGAAGGACGGCCACTTTCCTAGGGTCACCACAGTAAGCGACGGCAC CAAGCGCAACAATATGGATTTCAGCATTCGCATCTCTTCCATTACTCCGGCGGAC GTGGGCACCTATTACTGCGTTAAATTTCGTAAGGGATCCCCGGAAGACGTGGAGT TCAAATCCGGCCCTGGTACGGAGATGGCTCTGGGCGCCAAGCCCTCG.

[0224] In one embodiment the polynucleotide molecule further comprises a peptide tag, preferably a tag selected from the group consisting of CD34, FLAG and MYC.

[0225] In a further aspect, the present invention provides an amino sequence comprising a chimeric antigen receptor (CAR) or a fragment thereof comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); wherein the nucleotide sequence encoding the chimeric antigen receptor (CAR) is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPy)-related protein (SGRP), wheren the amino acid comprises preferably the sequence as shown in SEQ ID NO: 3:

[0226] MALPVTALLLPLALLLHAARPGSEIQLQQSGAELVKPGASVKLSCTGSGFNIEDYYIH WVKQRTEQGLEWIGRIDPENDETKYGPIFQGRATITADTSSNTVYLQLSSLTSEDTAV YYCAFRGGVYWGPGTTLTVSSGGGGSGGGGSGGGGSHMDVVMTQSPLTLSVAIGQS ASISCKSSQSLLDSDGKTYLNWLLQRPGQSPKRLISLVSKLDSGVPDRFTGSGSGTDFT LRISRVEAEDLGIYYCWQGTHFPGTFGGGTKLEIKASTTTPAPRPPTPAPTIASQPLSLR PEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKLLYIFK QPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYKQGQNQLYNELNLG RREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMKGERRR GKGHDGLYQGLSTATKDTYDALHMQALPPRGSGEGRGSLLTCGDVEENPGPMYRM QLLSCIALSLALVTNSEEELQIIQPEKLLLVTVGKTATLHCTITSLFPVGPIQWFRGVGP GRVLIYNQKDGHFPRVTTVSDGTKRNNMDFSIRISSITPADVGTYYCVKFRKGSPEDV EFKSGPGTEMALGAKPS.

[0227] In a further aspect, the present invention provides an amino sequence comprising a chimeric antigen receptor (CAR) or a fragment thereof comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPy)-related protein (SGRP), wheren the amino acid comprises preferably the sequence as shown in SEQ ID NO: 4:

[0228] MALPVTALLLPLALLLHAARPGSDIQMTQTTSSLSASLGDRVTISCRASQDISKYLNW YQQKPDGTVKLLIYHTSRLHSGVPSRFSGSGSGTDYSLTISNLEQEDIATYFCQQGNTL PYTFGGGTKLEITGSTSGSGKPGSGEGSTKGEVKLQESGPGLVAPSQSLSVTCTVSGVS LPDYGVSWIRQPPRKGLEWLGVIWGSETTYYNSALKSRLTIIKDNSKSQVFLKMNSLQ TDDTAIYYCAKHYYYGGSYAMDYWGQGTSVTVSSAAAASTTTPAPRPPTPAPTIASQ PLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCKRGRKKL LYIFKQPFMRPVQTTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYKQGQNQLYN ELNLGRREEYDVLDKRRGRDPEMGGKPRRKNPQEGLYNELQKDKMAEAYSEIGMK GERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRGSGEGRGSLLTCGDVEENPGP MYRMQLLSCIALSLALVTNSEEELQIIQPEKLLLVTVGKTATLHCTITSLFPVGPIQWF RGVGPGRVLIYNQKDGHFPRVTT VSDGTKRNNMDF SIRIS SITPAD VGT YYCVKFRKG SPED VEFKSGPGTEMALGAKPS .

[0229] In a further aspect, the present invention provides a construct comprising the polynucleotide molecule as described herein. In one embodiment the construct is comprised by a viral expression vector, preferably by a lentiviral, retroviral or adenoviral expression vector or by a non-viral mammalian expression system, preferably by a PiggyBac, Sleeping Beauty or Tol2 expression system.

[0230] In a further aspect, the present invention provides a chimeric antigen receptor (CAR)-T cell comprising T cells expressing the polynucleotide molecule as described herein and / or the construct as described herein.

[0231] In one embodiment the T cells are present in a therapeutically effective amount for the prevention and / or treatment of a cancer entity expressing epidermal growth factor receptor variant III (EGFRvIII) and / or CD 19.

[0232] In one embodiment at least a portion of the T cells are activated and produce one or more cytokines. In one embodiment the one or more cytokines are selected from the group consisting of interferon gamma (IFNy), TNF, CCL3, IL- 13 and IL-1A.

[0233] In one embodiment at least a portion of the T cells express one or more surface markers selected from the group consisting of CD25, CD69 and CD 107a.

[0234] In one embodiment the chimeric antigen receptor (CAR)-T cell further comprises a pharmaceutically acceptable excipient, carrier, and / or diluent which supports maintenance of the T cells.

[0235] In a further aspect, the present invention provides a method of treating cancer in a subject suffering from an epidermal growth factor receptor (EGFR)-associated cancer or suffering from a CD19-associated cancer, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein.

[0236] Also provided is a chimeric antigen receptor (CAR)-T cell as described herein, for use in a method for the treatment of cancer in a subject suffering from an epidermal growth factor receptor (EGFR)-associated cancer or suffering from a CD19-associated cancer, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein.

[0237] Also provided is the use of a chimeric antigen receptor (CAR)-T cell as described herein for the manufacture of a medicament for the treatment of cancer in a subject suffering from an epidermal growth factor receptor (EGFR)-associated cancer or suffering from a CD 19- associated cancer.

[0238] Also provided is the use of a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein for the treatment of cancer in a subject suffering from an epidermal growth factor receptor (EGFR)-associated cancer or suffering from a CD 19- associated cancer. In a further aspect, the present invention provides a method of treating a solid cancer in a subject, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein.

[0239] Also provided is a chimeric antigen receptor (CAR)-T cell as described herein, for use in a method for the treatment of a solid cancer in a subject, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein.

[0240] Also provided is the use of a chimeric antigen receptor (CAR)-T cell as described herein for the manufacture of a medicament for the treatment of a solid cancer in a subject.

[0241] Also provided is the use of a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein for the treatment of a solid cancer in a subject comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein.

[0242] In a further aspect, the present invention provides a method of treating cancer in a subject, wherein the cancer is located in the central nervous system and wherein the method comprises administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII).

[0243] Also provided is a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII), for use in a method for the treatment of cancer in a subject, wherein the cancer is located in the central nervous system, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein. Also provided is the use of a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII), for the manufacture of a medicament for the treatment of cancer in a subject, wherein the cancer is located in the central nervous system,.

[0244] Also provided is the use of a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII), for the treatment cancer in a subject, wherein the cancer is located in the central nervous system, comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein.

[0245] In one embodiment the cancer or solid cancer is selected from the group consisting of glioma, glioblastoma (GBM), medulloblastoma, ependymoma or diffuse intrinsic pontine glioma (DIPG), and is preferably GBM.

[0246] In a further aspect, the present invention provides a method of treating cancer in a subject, wherein the cancer is selected from the group consisting of a breast cancer, lung cancer, melanoma, lymphoma, acute lymphocytic leukemia (ALL), and non-Hodgkin’s lymphoma (NHL), preferably lymphoma, and wherein the method comprises administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to CD 19.

[0247] Also provided is a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof comprising an extracellular domain comprising an antigen binding region which binds to CD 19, for use in a method for the treatment of cancer in a subject, wherein the cancer is selected from the group consisting of a breast cancer, lung cancer, melanoma, lymphoma, acute lymphocytic leukemia (ALL), and non-Hodgkin’s lymphoma (NHL), preferably lymphoma.

[0248] Also provided is the use of a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof, comprising an extracellular domain comprising an antigen binding region which binds to CD 19, for the manufacture of a medicament for the treatment of cancer in a subject, wherein the cancer is selected from the group consisting of a breast cancer, lung cancer, melanoma, lymphoma, acute lymphocytic leukemia (ALL), and non-Hodgkin’s lymphoma (NHL), preferably lymphoma.

[0249] Also provided is the use of a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell as described herein, wherein the CAR or a fragment thereof of the CAR-T cell comprises a CAR or a fragment thereof, comprising an extracellular domain comprising an antigen binding region which binds to CD 19, for the treatment of cancer in a subject, wherein the cancer is selected from the group consisting of a breast cancer, lung cancer, melanoma, lymphoma, acute lymphocytic leukemia (ALL), and non-Hodgkin’s lymphoma (NHL), preferably lymphoma.

[0250] In one embodiment the cancer is a target antigen-mosaic tumor or a target antigen- homogeneous tumor, preferably a target antigen-mosaic tumor.

[0251] In a further aspect, the present invention provides a method of preparing a population of activated T cells comprising a polynucleotide molecule as described herein or a construct as described herein, the method comprising: (i) contacting in vitro one or more T cells that have been modified to comprise a polynucleotide molecule as described herein or a construct as described herein with a stimulus that induces expansion of the T cells to provide an expanded T cell population; and (ii) activating in vitro the T cells to produce an activated T cell population. In one embodiment, the T cells are the chimeric antigen receptor (CAR)-T cell as described herein. In a further aspect, the present invention provides a nucleotide sequence encoding a signal regulatory protein gamma (SZRPy)-related protein (SGRP) comprising the sequence as shown in SEQ ID NO: 5:

[0252] TACAGGATGCAACTCCTGTCTTGCATTGCACTAAGTCTTGCACTTGTCACGAATTC

[0253] GGAGGAGGAGCTGCAGATCATCCAGCCAGAGAAGCTGCTGCTGGTGACCGTGGG CAAGACGGCCACTCTTCATTGTACCATCACCTCTTTGTTCCCCGTGGGTCCCATCC AGTGGTTCCGCGGGGTCGGACCCGGGCGGGTGCTCATCTACAACCAGAAGGACG

[0254] GCCACTTTCCTAGGGTCACCACAGTAAGCGACGGCACCAAGCGCAACAATATGG ATTTCAGCATTCGCATCTCTTCCATTACTCCGGCGGACGTGGGCACCTATTACTGC GTTAAATTTCGTAAGGGATCCCCGGAAGACGTGGAGTTCAAATCCGGCCCTGGTA CGGAGATGGCTCTGGGCGCCAAGCCCTCG.

[0255] In a further aspect, the present invention provides a amino sequence comprising a signal regulatory protein gamma (SIRPY)-related protein (SGRP) comprising the sequence as shown in SEQ ID NO: 6:

[0256] MYRMQLLSCIALSLALVTNSEEELQIIQPEKLLLVTVGKTATLHCTITSLFPVGPIQWF RGVGPGRVLIYNQKDGHFPRVTT VSDGTKRNNMDF SIRIS SITPAD VGT YYCVKFRKG SPEDVEFKSGPGTEMALGAKPS.

[0257] In a further aspect, the present invention provides a nucleotide sequence encoding a signal regulatory protein gamma (SIRPY)-related protein (SGRP) comprising the sequence as shown in SEQ ID NO: 18:

[0258] ATGGAGGAGGAGCTGCAGATCATCCAGCCAGAGAAGCTGCTGCTGGTGACCGTG GGCAAGACGGCCACTCTTCATTGTACCATCACCTCTTTGTTCCCCGTGGGTCCCAT CCAGTGGTTCCGCGGGGTCGGACCCGGGCGGGTGCTCATCTACAACCAGAAGGA

[0259] CGGCCACTTTCCTAGGGTCACCACAGTAAGCGACGGCACCAAGCGCAACAATATG GATTTCAGCATTCGCATCTCTTCCATTACTCCGGCGGACGTGGGCACCTATTACTG CGTTAAATTTCGTAAGGGATCCCCGGAAGACGTGGAGTTCAAATCCGGCCCTGGT ACGGAGATGGCTCTGGGCGCCAAGCCCTCG. Examples

[0260] Materials and Methods

[0261] SGRP engineering for production and release by CAR T cells

[0262] To generate SGRP, 9 amino acid (AA) substitutions were made to the endogenous human SIRPy binding domain (hSIRPy-Vl) AA sequence, as previously described by Ring and colleagues29. A human IL-2 signal peptide (IL2sig) sequence was added to the N-terminus of SGRP to allow its secretion by T cells to the extracellular space. The SGRP AA sequence was optimized for production by human T cells using GenSmart Codon Optimization (GenScript Biotech, USA) and reverse-translated into a nucleotide (nt) sequence. All AA and nt sequences are listed in Table 1 below.

[0263] CD47-SIRP / SGRP protein interaction modeling

[0264] Protein structures of SGRP, human SIRPa binding domain (hSIRPa-Vl), murine SIRPa binding domain (mSIRPa-Vl) and human CD47 (hCD47) were predicted by AlphaFold30’31, using the source code, trained weights and inference script available under an open-source license: https: / / github.com / deepmind / alphafold. AA sequences of SGRP, hSIRPa-Vl, mSIRPa-Vl and hCD47 were entered and folded using the multimer model. Sequences were superimposed against a genetic database to generate multiple sequence alignment statistics. Predictions ran on 'relax mode without GPU' and generated 3D interaction models for SGRP and hCD47, hSIRPa-Vl and hCD47, or mSIRPa-Vl and hCD47. Predicted local distance difference tests (pLDDT) were calculated to evaluate local distance differences of all atoms in each model and validate stereochemical plausibility (data not shown). All AA and nt sequences are listed in Table 1 below:

[0265] Table 1

[0266] Table 1 - continued

[0267] CAR construct and lentiviral expression vector design

[0268] For human T cell transduction, replication-defective lentiviruses were produced using a second- generation lentiviral system with transfer plasmids encoding a 3C10.BBz32’33(anti-EGFRvIII) or FMC63.BBz34,35(anti-CD19) CAR, an mCherry fluorescence reporter protein and, in some iterations, SGRP29,36. The CAR structure consisted of a CD8a leader, a single-chain variable fragment (scFv), CD8a hinge and transmembrane domains, a 4- IBB costimulatory domain and a CD3(^ signaling domain. Transgene expression was driven by the EF1A promoter and polyprotein sequences were cleaved by T2A or P2A peptides. All sequences were assembled with Vector Design Studio (VectorBuilder, USA). The vectors included an ampicillin resistance gene for positive selection of transformed E. coli. The lentiviral plasmids were purchased as bacterial glycerol stocks from VectorBuilder (VectorBuilder, USA). CAR vector maps are illustrated in Fig. 5 and CAR vector sequences are listed in Table 2 below: Table !

[0269] Table 2 - continued

[0270] Table 2 - continued Lentivirus production

[0271] Plasmid DNA was extracted with a QIAprep Spin Miniprep Kit (#27104, QIAGEN, Netherlands) from bacterial cultures grown overnight in the presence of 100 pg / mL ampicillin (#A5354, Sigma- Aldrich, USA). Lentiviral particles were generated by co-delivery of a transfer plasmid, a plasmid encoding a VSV-G envelope (pMD2.G; #12259, Addgene, USA) and an empty backbone plasmid (psPAX2; #12260, Addgene, USA) into HEK293T cells. Cells were maintained in DMEM (#11995065, Gibco, USA) supplemented with 10% inactivated fetal bovine serum (FBS; #P30-3302, PAN-Biotech, Germany), 1% pen strep (#15140-122, Gibco, USA) and 2 mM GlutaMAX-I (#35050-038, Gibco, USA). They were cultured as adherent monolayers, at 37°C, in a 5% CO2 atmosphere and regularly sub-cultured when reaching approximately 70-80% confluence. For the transfections, 5 x io5HEK293T cells were seeded per well in a 6-well plate and rested for 24 h. Growth media were replaced with antibiotic-free growth media, DNA plasmids were complexed with polyethyleneimine (PEI; #408727, Sigma- Aldrich, USA) for 15 min and added dropwise to the cells, followed by a 48 h incubation. The viral supernatants were collected and cleared from cells and debris by centrifugation at 500 x g for 10 min, followed by filtration through a 0.45 pm polyethersulfone filter (#SLHPM33RS, MilliporeSigma, USA). Virus particles were precipitated with Lenti-X Concentrator (#631232, Takara Bio, Japan), suspended in phosphate buffer saline (PBS; #D8537, Sigma- Aldrich, USA), quantified with Lenti-X GoStix Plus (#631280, Takara Bio, Japan), aliquoted and stored at - 80°C.

[0272] Healthy-donor T cell and monocyte isolation

[0273] Peripheral blood leukocytes from healthy donors (HDs) were obtained from the Blood Donation Center of the University Hospital Basel, Switzerland, after informed consent was obtained from all participants before blood collection. Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll Paque-PLUS (# GE17-1440-02, Cytiva, Germany) and density centrifugation. After up to 2 rounds of ACK-lysis (#A10492-01, Gibco, USA) for removal of erythrocytes, PBMC were washed with PBS. CD3+T cells were magnetically separated by negative selection with a Human Pan T cell isolation Kit (#130-096-535, Miltenyi Biotec, Germany) and stored long-term in Bambanker serum- free cell freezing medium (#BB01, GC Lymphotec, Japan) in liquid nitrogen (LN2). CD14+cells used in phagocytosis assays were magnetically separated by positive selection using Human CD14 MicroBeads (#130-050-201, Miltenyi Biotec, Germany) and stored in Bambanker serum-free cell freezing medium in LN2.

[0274] CAR T cell production

[0275] HD T cells were thawed, washed with PBS and rested in X-VIVO 15 (#BE02-060F, Lonza, Switzerland) at a density of 1 x 106cells per mL at 37°C, in a 5% CO2 atmosphere. After 24 h, the T cells were activated in X-VIVO 15 containing 150 U / mL of human IL-2 (#Ro 23-6019, Roche, Switzerland), 10 ng / mL of recombinant IL-7 (#200-07, PeproTech, USA), 10 ng / mL of recombinant IL-15 (#200-15, PeproTech, USA), 20 ng / mL of recombinant IL-21 (#200-21, PeproTech, USA) and Dynabeads human T-activator CD3 / CD28 (aCD3 / CD28; #1113 ID, Gibco, USA) in a 1 : 1 cell-bead ratio. After 48 h, aCD3 / CD28 beads were magnetically removed and T cells were resuspended in X-VIVO 15 with 5 pg / mL polybrene (#TR-1003-G, Sigma- Aldrich, USA) at a density of 3 * 106cells per mL. Lentiviral suspensions were added to the T cells at different multiplicity of infection (MOI) ratios and spinfected at 2500 rpm, at 30°C for 90 min. After spinfection, the T cells were washed and maintained at a density of 1 x 106cells per mL in X-VIVO 15 containing 500 U / mL IL-2 for 5-7 days. After this post-transduction cell expansion, T cells were sorted for mCherry expression using a BD FACSMelody Cell Sorter (BD Biosciences, USA). After sorting, the CAR T cell cultures were expanded in X-VIVO 15 containing 500 U / mL IL-2 and kept at a density of 1 x 106cells per mL by adjusting the cell density every 2-3 days based on automated cell counting with a LUNA-FL Dual Fluorescence Cell Counter (Logos Biosystems, South Korea) for 5-10 days until used in downstream assays.

[0276] CAR-biotinylated target protein binding assay

[0277] CAR T cell viability and count were assessed by Trypan blue exclusion. Cells were washed with PBS and seeded into a 96-well plate at a density of 2 x io5live cells per well. Cells were immediately stained with a Zombie NIR Viability kit (#423106, BioLegend, USA) diluted 1 :5000 in PBS for 20 min in the dark at RT. After viability staining, the cells were washed with autoMACS Running Buffer (#130-091-221, Miltenyi Biotec, Germany) and then resuspended in 100 pL per well of 10 pg / mL dilutions of biotinylated CD 19, EGFR or EGFRvIII proteins (#CD9-H82E9, #EGR-H82E3 and #EGR-H82E0, ACROBiosystems, USA) for 1 h in the dark at 4°C. After CAR-target exposure, the cells were washed with autoMACS Running Buffer and stained with 100 pL per well of FITC Streptavidin (#405202, BioLegend, USA) diluted 1 :50 in autoMACS Running Buffer, for 1 h in the dark at 4°C. Afterward, the cells were washed 3 times with autoMACS Running Buffer and resuspended in 100 pL of autoMACS Running Buffer. Samples were acquired with a CytoFLEX Flow Cytometer (Beckman Coulter, USA) and data were analyzed using FlowJo vlO Software (BD, USA).

[0278] CAR T cell secretome analysis

[0279] Supernatant isolation

[0280] Expanded aEGFRvIII CAR and aEGFRvIII-SGRP CAR cultures were rested in X VIVO medium without additional supplements for 24 h. The following day, cell viability and count were assessed by Trypan blue exclusion, after which cells were washed in PBS, resuspended in RPMI at a density of 1 x 106live cells per mL, and incubated for another 24 h at 37°C. Afterward, cultures were centrifuged at 300 x g for 5 min, and supernatants were collected, passed through 0.22 pm filters, and stored at -20°C.

[0281] Prot gestion

[0282] Supernatant samples were TCA-precipitated following the procedure: one volume of TCA was added to every 4 volumes of sample, mixed by vortexing, and incubated for 10 min at 4°C followed by precipitate collection by centrifugation for 5 min at 23000 x g. The supernatant was discarded, and the pellet was washed two times with acetone precooled to -20°C. The washed pellets were incubated in the open tube for 2 min at RT to allow residual acetone to evaporate. The pellets were resuspended in 2M guanidine hydrochloride (GUA), 10 mM TCEP, and 100 mM Ammonium Bicarbonate (AmBIC; pH = 8.5) by sonication. The samples were incubated for 10 min at 95°C, let to cool down to RT, followed by the addition of chloroacetamide in a final concentration of 15 mM. After an incubation of 30 min at 37°C, the samples were diluted with 100 mM AmBIC to achieve a final concentration of 0.5 M GUA. Sequencing-grade modified trypsin (1 / 50, w / w; #V5280, Promega, USA) was added, and the proteins were digested for 12 h at 37°C shaking at 300 rpm. Digests were acidified (pH < 3) using TFA and desalted using C18 spin columns (#74-4101, Harvard Apparatus, USA) according to the manufacturer’s instructions. Peptides were dried under vacuum and stored at - 20°C. tide mixture Io AH ;

[0283] A total of 1 pg of peptides were subjected to LC-MS / MS analysis using a Q Exactive Plus Mass Spectrometer (ThermoFisher Scientific, USA) fitted with an EASY-nLC 1000 (ThermoFisher Scientific, USA) and a custom-made column heater set to 60°C. Peptides were resolved using RP-HPLC columns (75 pm * 30 cm) packed in-house with Cl 8 resin (ReproSil-Pur C18-AQ, 1.9 pm resin; Dr. Maisch, Germany) at a flow rate of 0.2 pL per min. The following gradient was used for peptide separation: from 5% B to 10% B over 5 min to 35% B over 40 min to 50% B over 15 min to 95% B over 2 min followed by 18 min at 95% B. Buffer A was 0.1% formic acid in water and buffer B was 80% acetonitrile, 0.1% formic acid in water. ition

[0284] The mass spectrometer was operated in DDA mode with a total cycle time of approximately 1 sec. For MSI, 3e6 ions were accumulated in the Orbitrap over a maximum time of 100 ms and scanned at a resolution of 70000 FWHM at 200 m / z. Each MSI scan was followed by high- collision-dissociation (HCD) of the 10 most abundant precursor ions with dynamic exclusion set to 45 sec. MS2 scans were acquired at a target setting of le5 ions, a maximum accumulation time of 100 ms, and a resolution of 35000 FWHM at 200 m / z. Singly charged ions, ions with charge state > 6 and ions with unassigned charge state were excluded from triggering MS2 events. The normalized collision energy was set to 27%; the mass isolation window was set to 1.4 m / z, and one microscope was acquired for each spectrum.

[0285] Data preprocessing and analysis

[0286] Following the recommendation of Jin et al.—, we performed importation of the LCMS data using the random forest approach implemented in the missForest R library—. After imputation, the data was normalized using the MSnbase1^ R library (version 2.23.0). These analyses were conducted using R4.2.1. Raw data were imported into Progenesis QI v2.0 software (Nonlinear Dynamics, UK), which extracted peptide precursor ion intensities across all samples with the default parameters. The generated mgf file was searched using MASCOT against a human database (consisting of 41094 forward and reverse protein sequences downloaded from Uniprot in April 2020), a manually entered recombinant SGRP AA sequence as well as 392 commonly observed contaminants using the following search criteria: full tryptic specificity was required (cleavage after lysine or arginine residues, unless followed by proline); 3 missed cleavages were allowed; carbamidomethylation (C) was set as fixed modification; oxidation (M) and acetyl (Protein N-term) were applied as variable modifications; mass tolerance of 10 ppm (precursor) and 0.6 Da (fragments). The database search results were filtered using the ion score to set the false discovery rate (FDR) to 1% on the peptide and protein level, respectively, based on the number of reverse protein sequence hits in the dataset. Quantitative analysis results from label- free quantification were processed using the SafeQuant R package v.2.3.2 (https: / / github.com / eahrne / SafeQuant / ) (Ahrne, E., Molzahn, L., Glatter, T., & Schmidt, A. (2013). Critical assessment of proteome-wide label-free absolute abundance estimation strategies. Proteomics. Journal of Proteome Research https: / / www.ncbi.nlm.nih.gov / pubmed / 23794183) to obtain peptide relative abundances. This analysis included global data normalization by equalizing the total peak / reporter areas across all LC-MS runs, data imputation using the knn algorithm, summation of peak areas per protein and LC-MS / MS run, followed by calculation of peptide abundance ratios. Only isoform- specific peptide ion signals were considered for quantification. To meet additional assumptions (normality and homoscedasticity) underlying the use of linear regression models and t-tests, MS-intensity signals were transformed from the linear to the log scale. The summarized peptide expression values were used to test for differentially abundant peptides between conditions. Here, empirical Bayes-moderated t-tests were applied, as implemented in the R / Bioconductor limma package (http: / / bioconductor.org / packages / release / bioc / html / limma.html)—. The resulting per protein and condition comparison p-values were adjusted for multiple testing using the Benjamini -Hochberg method. Differential expression analysis compared aEGFRvIII-SGRP CAR and aEGFRvIII CAR samples from 2 HDs.

[0287] Cell lines and cell culture

[0288] BS153, U87, U251 and U251vIII are human glioma cell lines. BS153 cells were maintained in DMEM (#10938025, Gibco, USA) supplemented with 10% inactivated FBS, 1% pen strep, 2 mM GlutaMAX-I and 1 mM sodium pyruvate (#S8636, Sigma- Aldrich, USA). U87, U251 and U251vIII cells were maintained in MEM (#M4655, Sigma- Aldrich, USA) supplemented with 10% inactivated FBS, 1% pen strep, IX MEM NEAA (#11140-035, Gibco, USA), 2 mM GlutaMAX-I and 1 mM sodium pyruvate. Raji is a lymphoma cell line cultured in DMEM supplemented with 10% inactivated FBS, IX MEM NEAA and 1% pen strep. GBM cells were cultured as adherent monolayers whereas Raji cells were cultured in suspension. Cells were maintained at 37°C, in a 5% CO2 atmosphere and regularly sub-cultured when reaching approximately 70-80% confluence. All cell lines were routinely tested for mycoplasma contamination using a MycoAlert PLUS Mycoplasma Detection Kit (#LT07-710, Lonza, Switzerland).

[0289] GBM cell line lentiviral transduction Parental EGFRvIILnegative U87 and U251 cell lines were transduced with a pmp71 lentiviral vector encoding a full-length EGFRvIII, to generate stable EGFRvIII-expressing U87vIII and U251vIII cell lines, respectively. For in vitro cytotoxicity assays, BS153, U251 and U251vIII cells were transduced with an Incucyte NucLight Green lentivirus (#4624, Sartorius, Germany) to express a nuclear-restricted EGFP (nEGFP) fluorescence viability reporter protein. For in vivo tumor monitoring by bioluminescence imaging (BLi) and fluorescence labeling of tumor cells in downstream assays, U87 cells were transduced with a Luc2-mTagBFP2 lentivirus and U251vIII cells were transduced with an iRFP713-NLuc lentivirus. GBM cell line vector sequences are listed in Table 3 below:

[0290] Table 3

[0291]

[0292] For lentiviral transduction, tumor cells were seeded at 1 x 105cells per well of a 24-well plate and rested for 24 h. Growth media were replaced with antibiotic-free growth media containing 8 pg / mL polybrene. Lentiviral suspensions were added to the cells at different MOIs and incubated for 6 h at 37°C. Afterward, transduction media were replaced with fresh growth media and the cells were expanded for 1-2 weeks. Cells expressing the relevant surface receptor or fluorescence protein were sorted using a BD FACSMelody or a BD FACSAria SORP Cell Sorter (BD Biosciences, USA). Luc2 and NLuc expression was confirmed by exposing cells seeded in the wells of a flat-bottom white 96-well plate to 1 volume of 15 mg / mL D-luciferin (#LUCNA-1G, Goldbio, USA) or 1 volume of 0.5 mg / mL Nano-Gio In Vivo Substrate, fluorofurimazine (FFz; #CS320501, Promega, USA), respectively. Cells were imaged after a 10 min incubation protected from the light and imaged with a Fusion FX System (Vilber, France).

[0293] Cell surface marker expression analysis

[0294] The expression of cell surface markers was determined by FC. Briefly, single-cell suspensions were counted by Trypan blue exclusion and seeded in 96-well plates at a density of 2 * 105cells per well. Cells were washed with PBS and resuspended in 100 pL of antibody staining solution. Unless stated otherwise, a viability staining step was performed before (Zombie NIR Viability kit diluted 1 :5000) or after (DAPI Solution (#564907, BD Pharmigen, USA) diluted to 0.5X or DRAQ7 (#424001, BioLegend, USA) diluted 1 : 1000). All antibodies, viability stains and respective dilutions are listed in Table 4 below. Table 4 CAR T cell cytotoxicity assay

[0295] Killing assays were performed using an Incucyte S3 Live-Cell Analysis System (Sartorius, Germany). nEGFP-labeled target cells were seeded in flat-bottom, clear, 96-well plates at a density of 1 x 104cells per well and incubated for 24 h to allow the formation of a cell monolayer. mCherry-labeled CAR T cells were added at various effector-target (E:T) ratios and co-cultures were followed for 72 h. Brightfield and fluorescence images were recorded every 4 h with a 10X objective. Target cell viability kinetics were analyzed via time-lapse videos generated with Incucyte Software (Sartorius, Germany) and quantified using GraphPad Prism vlO Software (GraphPad, USA). Target cells incubated in medium alone were used to determine the baseline viability kinetics of each target cell line. Specific target cell lysis was calculated for each 4 h timepoint as green object area or count, depending on the experiment. All conditions were performed as duplicates or triplicates.

[0296] CAR T cell degranulation assay

[0297] T cell degranulation in co-cultures with GBM cells was assessed by FC. Briefly, tumor cells were washed with PBS, dissociated and counted by Trypan blue exclusion. GBM cells were seeded in a flat-bottom 96-well plate at a density of 1 x 104cells per well in 100 pL of growth medium and incubated for 24 h to allow the formation of a cell monolayer. Afterward, media in the wells was discarded, replaced by 100 pL of CAR T cell or mock-transduced T cell suspensions in GBM growth medium in 1 : 1 E:T and incubated for 24 h. At 24 h of co-culture, suspension cells were gently mixed in the supernatant and collected into a round-bottom 96- well plate. Cells were washed with PBS and stained with a BB700 anti-CD107a (clone H4A3, #566558, BD Biosciences, USA) for 20 min in the dark at 4°C. After surface staining, the cells were washed 3 times with autoMACS Running Buffer and resuspended in 100 pL per well of 0.5X DAPI diluted in autoMACS Running Buffer. Samples were acquired with a CytoFLEX Flow Cytometer and data were analyzed with FlowJo vlO Software (BD, USA). Positively stained cells were differentiated from the background using unstained controls. All conditions were performed as triplicates.

[0298] GBM-CAR T cell co-culture supernatant ELISA

[0299] Flat-bottom F96 MAXISORP NUNC-IMMUNO plates (#439454, Thermo Scientific, USA) were coated with 50 pL per well of Purified anti-human IFNy Antibody (clone MD-1, #507502, BioLegend, USA) diluted 1 :200 in IX PBS (#5460-0023, BioConcept, Switzerland) and incubated overnight at 4°C. The following day, wells were washed 3 times with PBS-T buffer (PBS with 0.05% Tween 20) and blocked for unspecific binding with 100 pL per well of 1% SureBlock solution (#SB232010, LubioScience, Switzerland) for 1 h at RT. After the blocking, plates were washed 3 times with PBS-T, then 50 pL of 24 h co-culture supernatants or serial dilutions of Recombinant Human IFNy standard (#300-02, Peprotech, USA) were added to the wells and incubated for 2 h at RT. Wells were washed 3 times before adding 50 pL per well of Biotin anti-human IFN-y Antibody (clone 4S.B3, #502504, BioLegend, USA) diluted 1 :400 in 1% SureBlock solution and incubated 1 h at RT. Wells were again washed 3 times before adding 50 pL per well of HRP Streptavidin (#405210, BioLegend, USA) diluted 1 :2000 in 1% SureBlock solution and incubated 1 h at RT. Wells were washed 3 times and 100 pL per well of SIGMAFAST OPD tablet (#P9187, Sigma- Aldrich, USA) solution was added. The chromogenic reactions were stopped by adding 50 pL per well of 10% sulfuric acid (H2SO4) and absorbance values were measured by a Synergy Hl Hybrid microplate reader (BioTek, USA). Absolute ZFNy concentrations in test samples were interpolated from a standard curve and represented using GraphPad Prism vlO Software.

[0300] On-cell CD47 blocking assay

[0301] BS153 viability and count were assessed by Trypan blue exclusion, after which cells were seeded in flat-bottom 96-well plates at a density of 3 * 105cells per well. Cells were then treated for 30 min at 4°C with 50 pL per well of 10 pg / mL of InVivoMAb anti-human CD47 (clone B6.H12, #BE0019-l, Bio X Cell, USA) or InVivoMAb mouse IgGl isotype control (clone MOPC-21, #BE0083, Bio X Cell, USA) or conditioned-media from 24 h-rested, antigen-naive aEGFRvIII or aEGFRvIII-SGRP CAR T cells seeded at a density of 1 x 106cells per mL of unsupplemented RPMI. Pre-treated tumor cells were washed with PBS and incubated for 30 min at 4°C with biotinylated SIRPa (bt-SIRPa; #CDA-H82F2, ACROBiosystems, USA), which bound to the unblocked CD47 on BS153 cells. Finally, APC Streptavidin (#405207, BioLegend, USA) staining was performed, followed by 3 washes with autoMACS Running Buffer and resuspension in 100 pL per well of 0.5X DAPI in autoMACS Running Buffer. Samples were acquired by a CytoFLEX Flow Cytometer and data were analysed with FlowJo vlO Software. CD14+monocyte differentiation and culture

[0302] Primary monocytes were thawed, washed in PBS and plated at 0.8 x 106cells per mL in RPMI 1640 with IX GlutaMAX (#61870-036, Gibco, USA), 10% FBS (#P30-3302, PAN-Biotech, Germany), 5% human serum (#H4522, Sigma- Aldrich, USA) and 25 ng / mL M-CSF (#300-25, PeproTech, USA), and incubated at 37°C in a 5% CO2 atmosphere. After 48 h, the medium was replaced with human serum-free growth media and refreshed every 2-3 days for a maximum of 14 days until phagocytosis assays.

[0303] Phagocytosis assay

[0304] Macrophages were detached from culture dishes using cell scrapers after a 15-minute incubation at 37°C in TrypLE Express (#12604-021, Gibco, USA). Cells were washed in PBS, counted, seeded at 5 x 104cells per well in a 96-well flat-bottom plate (#353072, Coming, USA), and incubated for 48 h to allow attachment. CAR T cells were counted and diluted to 1 x 106cells per mL in X VIVO 15. Tumor cells were washed in PBS, dissociated, and counted. All cell counts were performed by Trypan Blue exclusion. U251vIII cells were diluted to 1 x 106cells per mL in IMDM. U87 cells were stained with 62.5 x 10 nM CellTracker Green (#C2925, Thermo Scientific, USA) diluted in IMDM (#12440-053, Gibco, USA) for 30 min at 37°C at a density of 1 x 106cells per mL. After two washes with PBS, stained cells were counted and adjusted to 1 x 106cells per mL in IMDM. U87 and U251vIII cells were then mixed in a 1 :1 ratio. The tumor cell mixture (1 x 105cells) and CAR T cells (5 x 104cells) were added to each well containing macrophages. The contents were resuspended in IMDM and incubated at 37 °C for 3 h. After incubation, cells were detached using TrypLE, transferred to a 96-well U-bottom plate, and washed twice in PBS. Viability staining was carried out by incubating cells with a Zombie UV Viability kit (#423107, BioLegend, USA) for 20 min at 4°C in the dark. Fc-block was performed by incubating cells in a dilution of Human TruStain FcX (#101320, BioLegend, USA) for 10 min at 4°C. Antibody mastermixes (full-stains and FMOs) were prepared freshly in autoMACS Running Buffer. Cells were stained with surface marker antibody mastermixes for 25 min at 4°C in the dark and then washed twice in autoMACS Running Buffer. Cells were fixed by incubation for 20 min at RT using a Cyto-Fast Fix / Perm Buffer set (#426803, BioLegend, USA). After antibody staining, samples were washed twice with autoMACS Running Buffer, resuspended in 200 pL MACS, and acquired on a Cytek Aurora 5-Laser Spectral Analyzer (Cytek Biosciences, USA), using standard, daily quality-controlled, Cytek- Assay-Settings. All conditions were performed with donor-matched CAR T cells and macrophages from 3 HDs.

[0305] Animal experimentation

[0306] All animal handling, surveillance and experimentation were performed according to the guidelines of the Swiss Federal Veterinary Office (SFVO) and the Cantonal Veterinary Office (CVO) of Basel-Stadt, Switzerland. GBM model experiments were executed under license #2929_31795 and lymphoma model experiments under licenses #2929_31795 and #3176_35274. Animals were maintained at the local animal facility in pathogen- free, ventilated HEPA-filtered cages under stable housing conditions of 45-65% humidity, a temperature of 21- 25°C and a gradual light cycle from 7 am to 5 pm. Animals were provided standard food and water without restrictions.

[0307] Mouse strains and humane endpoint assessment

[0308] NOD.Cg-PrkdcscidI12rg / SzJ (NSG) mice with identifier RRID:IMSR_JAX:005557 were obtained from in-house breedings or externally (Janvier Labs, France) under protocols approved by the SFVO and CVO of Basel-Stadt. Co-housed animals were assigned to treatment or control groups using a randomized approach and euthanized upon reaching the humane endpoint, including significant reduction of locomotion, significant weight loss and mild-to-severe neurologic symptoms. Tumor cell implantation was set as day 0 and the survival time was set as the day of euthanasia.

[0309] GBM mouse models and survival assessment

[0310] All experiments were performed on NSG mice of the male sex, aged 7-12 weeks at the time of tumor implantation. To assess the efficacy of anti-EGFRvIII CAR T cell and anti-CD47 antibody monotherapies, mice were injected intracranially (i.c.) with 5 x 104U251vIII-NLuc cells. To test our proposed combination therapy, mice were injected i.c. with a total of 5 x io4GBM cells consisting of 2.5 x 104U87-Luc2 and 2.5 x 104U251-NLuc, resulting in EGFRvIII- mosaic tumors. In both models, the animals were anesthetized in an induction chamber with 2.0 ± 0.5% isoflurane in an O2 atmosphere immediately before the tumor injections. Anesthesia on the stereotactic frame (Neurostar, Germany) was maintained at 2.0 ± 0.5% isoflurane delivered through a nose / mouth adaptor. General analgesia (Buprenorphine; Bupaq-P, Streuli Tiergesundheit, Switzerland) was given subcutaneously (s.c.) immediately before surgery at 0.05 mg per kg and local analgesia was applied s.c. under the scalp. Eye gel (Lacrinorm, Bausch+Lomb Swiss AG, Switzerland) was applied to prevent drying of the eyes during surgery. The scalp was briefly swabbed with povidone-iodide solution, and a midline incision was made. The scalp was briefly swabbed with povidone-iodide solution and a midline incision was made. A burr hole was manually drilled 2 mm lateral from the cranial midline and 1 mm posterior of the bregma suture using a microdrill. A digitally-controlled injection was performed with the Stereodrive Software (Neurostar, Germany) using a 10 pL Microliter Syringe (#80300, Hamilton, USA). The syringe was lowered into the burr hole to a depth of 3 mm below the surface of the dura and retracted by 0.5 mm to form a small reservoir in the cortex. A volume of 4 pL of single-cell suspension was injected at 1 pL per min. The needle was left in place for at least 1 min and carefully retracted by 0.5 mm every 30 sec. After injection, the incision was sutured (#D7585, Ethicon, USA). Buprenorphine analgesia (Bupaq-P, Streuli Tiergesundheit, Switzerland) was given intraperitoneally (i.p.) at 0.05 mg per kg immediately post-op. The following treatments and controls were administered intratumorally (i.t.) in a volume of 4 pL on days 7 and 14: Vehicle (PBS), antibody Isotype (InVivoMAb mouse IgGl isotype control, clone MOPC-21; 5 pg), aCD47 (InVivoMAb anti-human CD47, clone B6.H12; 5 pg), aCD19 CAR (5 x 105cells), aCD19-SGRP CAR (5 x 105cells), aEGFRvIII CAR (5 x 105cells), aEGRFvIII CAR + aCD47 (5 x 105cells and 5 pg, respectively) or aEGFRvIII- S GRP CAR (5 x 105cells). In EGFRvIII-mosaic survival experiments, aCD47 and aEGFRvIII CAR + aCD47 treatment groups received additional doses of anti-CD47 (100 pg) administered i.p. in a volume of 100 pL on days 19, 22, 26 and 29. In the EGFRvIII-mosaic, CCL3 blockade survival experiment, CAR T cells were applied as described above and 50 ng of antibody per mouse (aCCL3 or isotype) were administered i.p. in a volume of 100 pL on days 8, 10, 13, 15, 17, 20, 22, 24, 27, 29, 31, 34, 36, 38, and 41. Mice were monitored for clinical signs until day 90, upon which all remaining survivors were either euthanized or assigned to a tumor rechallenge experiment. Animals assigned to a rechallenge experiment were injected with the same mix of tumo r cells using the same procedure described above. Tumor-rechallenged animals received no treatment and a historic Vehicle group was used as a control. Kaplan-Meier survival comparison was performed using a Log-rank (Mantel-Cox) test with GraphPad Prism vlO Software. In vivo GBM bioluminescence imaging

[0311] GBM engraftment and growth were monitored by bioluminescence imaging (BLi). Mice implanted i.c. with EGFRvIII-mosaic tumors were subjected to dual BLi with specific substrates to separately monitor the growth of the EGFRvIII and EGFRvIIF tumor fractions. The animals’ clinical scores and luminescence images were taken weekly, after i.p. injection of 150 mg kg'1of D-luciferin or intravenous (i.v.) injection of 0.325 pmol of fluorofurimazine per mouse. Subsequently, mice were anesthetized by isoflurane inhalation and imaged on a Newton 7.0 instrument (Vilber, France). Photon flux over a defined ROI (brain surface) was quantified with NEWTON v7 Software (Vilber, France).

[0312] Olin k proteomics of mouse plasma

[0313] Peripheral blood samples were collected approximately 24 h after each treatment dose (days 8 and 15 after tumor implantation). Unanestethized mice were briefly put under a heat lamp and then placed in a cylindrical restrainer. A small puncture was made on the tail to allow the dripping of approximately 100 pL of blood into lithium heparin-coated Microvette 100 capillary blood collection tubes (#20.1282.100, Sarstedt, Germany) and kept at RT. Blood samples were centrifuged at 4500 rpm for 15 min at RT and the top layer of plasma was transferred into sterile 0.5 pL Eppendorf tubes. Samples were immediately frozen at -80°C until analysis. 15 samples were analyzed by proximity extension assay technology using a standard Olink Target 96 Immuno-Oncology panel (Olink Holding, Sweden). Samples were analyzed across 2 plates, 16 samples from the first run were included in the second in order to perform a reference sample normalization. This bridging procedure consisted in calculating the median of the paired normalized protein expression levels (NPX) differences per protein between the overlapping samples to determine the adjustment factors to be applied between the 2 datasets. For the overlapping samples, the mean value was kept. A cyclic loess normalization was applied. Since data below the limit of detection may be non-linear, the differential expression for the contrasts of interest was assessed by a Mann- Whitney U test with Benjamini -Hochberg correction.

[0314] Lymphoma mouse model and survival assessment

[0315] To test the efficacy of the CAR constructs in a subcutaneous (s.c.) tumor model, NSG mice of the female sex, aged 8-12 weeks were anesthetized with isoflurane and injected with 5 x 105 CD19-expressing Raji cells s.c. in the right flank. Raji cells were suspended in BD Matrigel Basement Membrane Matrix High Concentration (#354248, BD Biosciences) diluted 1 : 1 in phenol red-free DMEM without additives in a total volume of 100 pL. Three days after tumor inoculation, each mouse received 8 x io5CAR T cells suspended in 200 pL of PBS or PBS alone, i.v. via the tail vein. Tumor-bearing mice injected with untargeted CAR T cells or PBS were used as controls. Tumor size was measured 3 times per week using a caliper. Animals were sacrificed before reaching a tumor volume of 1500 mm3or when reaching an exclusion criterion (ulceration, severe weight loss, severe infection or bite wounds). Tumor volume was calculated according to the following formula: Tumor volume (mm3) = (d2x D) / 2 with D and d being the longest and shortest tumor parameter in mm, respectively.

[0316] Mouse tissue collection and processing

[0317] For spectral FC analyses, mice were euthanized by CO2 suffocation, and brain regions were immediately harvested into ice-cold HBSS. Depending on the experiment, the tumor-injected hemisphere, contralateral hemisphere or brain meninges were carefully dissected and manually minced using razor blades and enzymatically dissociated at 37°C for 30 min with 1 mg / mL collagenase type IV (#LS004188, Worthington Biochemical Corporation, USA) and 250 U / mL DNase 1 (#10104159001, Roche, Switzerland) in a buffer containing HBSS with Ca2+ / Mg2+(#14205-050, Gibco, USA), 1% MEM NEAA, 1 mM sodium pyruvate, 44 mM sodium bicarbonate (#25080-060, Gibco, USA), 25 mM HEPES (#H0887, Gibco, USA), 1% GlutaMAX-I and 1% antibiotic-antimycotic (#15240062, Gibco, USA). Extraction of the meninges was perfomed as described earlier (Roussel-Queval A, Rebejac J, Eme-Scolan E, Paroutaud LA, Rua R. Flow cytometry and immunohistochemistry of the mouse dural meninges for immunological and virological assessments. STAR Protoc. 2023 Mar 17;4(1 ): 102119. doi: 10.1016 / j.xpro.2023.102119. Epub 2023 Feb 14. PMID: 36853673; PMCID: PMC9958090.). The resulting cell suspensions were filtered through a 70 pm strainer and centrifuged in a density gradient using debris removal solution (#130-109-398, Miltenyi Biotec, Germany) according to the manufacturer's protocol to remove myelin and cell debris. Erythrocytes were removed using ACK lysing solution (#A1049201, ThermoFisher Scientific, USA), and cell suspensions were washed with PBS and kept on ice until spectral flow staining. For IHC, quantification of brain tumor area, IF brain analysis, and spleen size measurement, mice were anesthetized i.p. with a mix of 80 mg per kg of ketamine (Ketanarkon, Streuli Tiergesundheit, Switzerland) and 16 mg per kg of xylazine (Rompun, Elanco, USA), and transcardially perfused with ice-cold PBS. Brains and spleens were dissected and immediately fixed in formalin at 4°C for 72 h before paraffin embedding.

[0318] Mouse brain immunohistochemistry

[0319] Stainings

[0320] Formalin-fixed brains were embedded in paraffin and 5 pm sections were made using a microtome. Slides were stained according to the standard H&E protocol using the automated Gemini AS Slide Stainer (ThermoFisher Scientific, USA) and covered with Permount Mounting Medium (#SP15-100, ThermoFisher Scientific, USA). For CD3 and CD68 IHC stainings, FFPE brain sections were stained with a Ventana DISCOVERY ULTRA automated slide preparation system (Ventana Medical Systems Inc., USA) using a Histofine Simple Stain MAX PO anti-rabbit (#414142F, Nicherei Biosciences Inc., Japan) as a detection reagent. For myelin staining, brain sections were deparaffinized and rehydrated before incubation in Luxol solution (#1B 389, Medite, Switzerland) at 60 °C for 2 h. After cooling, sections were washed with 96% ethanol and flowing tap water for 5 min, and then distilled water. Differentiation was performed with 0.1% lithium carbonate solution (#62470, Sigma- Aldrich, USA) for 10 sec. After washing the sections in distilled water, nuclei were stained using 1% cresyl violet (#1.05235, Merck, Germany) in 96% ethanol for 5 sec until the background became colorless. Sections were then dehydrated in 100% ethanol, Xylol (#253-VI53TE, Biosystems, Switzerland), and embedded (#41-4012-00, HistoLab, Sweden). ition

[0321] Slides were acquired using a Nanozoomer S60 digital slide scanner (Hamamatsu, Japan) with a 40x objective.

[0322] Histological brain tumor sii asurement

[0323] To calculate tumor size, a pixel classifier was trained to detect tumors in mouse brain histological sections in QuPath37. The Random Trees Classifier was used with the following parameters: Pixel size 3.53 pm; Channels: red, green and blue; Features: Gaussian and Laplacian of Gaussian; Scales: 4.0 and 8.0; no normalization; classification was done based on example annotations to distinguish brain tissue, tumor and background (ignore*). Results were visually verified. tification

[0324] Cells were segmented based on Hematoxylin staining with the StarDist2D38plugin within QuPath37using the following parameters: Probability threshold: 0.2; Pixel size: 0.2; Cell expansion: 2.0; using a pre-trained model ‘he heavy augment.pb’. Cells were classified for CD3 and CD68 positivity based on mean DAB staining in the nucleus with a threshold of 0.12 and 0.2, respectively.

[0325] Immunofluorescence multiplex staining and imaging

[0326] The protocol was adapted from Gut and colleagues39. FFPE brains were sectioned in 5 pm-thick sections using a microtome. Slides were deparaffinized 3 times for 5 min with RotiHistol (Roth, #6640) and 30 s with 100% EtOH, 95% EtOH, 70% EtOH, 50% EtOH and dH2O. For antigen retrieval, we used pre-heated citrate buffer (Merck, #C9999) in a microwave for 20 min, followed by a washing step with PBS for 10 min. Tissue sections were permeabilized in PBS with 0.2% Tween-20 (Sigma- Aldrich, #P5927) for 2 x 15 min. Thereafter, the slides were washed 2 x 5 min with PBS followed by blocking with 10% donkey serum (Jackson ImmunoResearch, #017-000-121) in PBS and incubation for 1 h in a dark humid chamber. Afterward, sections were stained in different cycles overnight at 4 °C with the following antibodies: anti-EGVRvIII (1 :1000, Rabbit, ThermoFisher, #RM419), anti-Ki-67 (1:500, Rat, ThermoFisher, #14-5698-82), anti-Iba-1 (1 :500, Goat, Novus Biological, #NB 100-1028), anti- CD69 (1 :500, R&D systems, #AF2359-SP), anti-TMEM119 (1 :500, Rabbit, Abeam, #ab209064), anti-CD3 (1 : 100, Rat, BioRad, #MCA1477), anti-CD206 (1 :200, Rabbit, BioConcept, #24595s). Subsequently, slides were washed for 3 x 5 min with PBS, and stained with fluorescent secondary antibody (Donkey-anti-rabbit 1 :300, Jackson ImmunoResearch, #711-605-152, donkey anti-rat 1 : 1000, Invitrogen, #2474993; donkey anti-goat 1 :300, Jackson Immunoresearch, #705-545-147) and DAPI (1 :1000), and incubated for 1 h in a humid chamber at RT. Slides were washed 3 x 5 min with PBS and mounted with imaging buffer at pH 7.4 (700 mM N-Acetyl-Cysteine (NAC, ThermoScientific, #160280250) in ddH2O + 20% HEPES (Sigma, #H0887)). After each cycle, slides were treated with 3 x 15 min elution buffer containing 0.5 M L-Glycine (Roth, #3790.2), 1.2 M Urea (Sigma, #U5378), 3 M Guanidium chloride (Sigma, #G3272), and freshly added 70 mM TCEP-HC1 (Sigma, #C4706) followed by ddJEO, and re-stained. ition

[0327] Images were acquired using a Nikon Eclipse Ni upright microscope equipped with 395, 470, 561 and 640 nm lasers with a Prior PL-200 robotic slide loader equipped with Microscan MiniHawk (Omron Microscan Systems, Inc.) and the Photometries Prior 95B camera. The acquisition was done using JOBS automation in NIS-Elements software (version 5.11.00). 4x Plan Apo NA 0.2 (Nikon) objective was used to make a slide overview and GA3 (general analysis3) with Otsu threshold was used to detect the tissue. The focus surface was created by performing software autofocus every 2 points based on the DAPI signal with Plan Apo X 20x NA 0.8 and the tissue was scanned with the same objective. Slides were scanned after each staining cycle, with blank (=unstained) cycles used to evaluate autofluorescence. The exposure time of each channel was constant for all cycles and corresponding blank channels (GFP, Cy3 and Cy5) were subtracted from the signal cycle.

[0328] Preprocessing

[0329] Fiji40was used for image preprocessing. Single tiles were stitched using Grid / Collection Stitching41. Stitched tissues were registered based on consecutive cycles based on DAPI (4',6- diamidino-2-phenylindole) channel using MultiStackReg plugin42and the same correction was propagated on the remaining channels. Images were saved as a pyramidal file using the Kheops43plugin, available under an open-source license: https : / / github . com / BIOP / ij p- kheops / releases.

[0330] Cell segmentation an fication

[0331] Cells were segmented based on DAPI staining with the StarDist2D38plugin within QuPath37using the following parameters: Probability threshold: 0.5; Pixel size: 0.5; Cell expansion: 2.0; using a pre-trained model ‘dsb2018_heavy_augment.pb’. Object classification in QuPath was used to define cell positivity for CD3, CD206, EGFRvIII and Ibal. All mouse brain IHC and IF images were prepared using OMERO.web app (www. openmi croscopy. org / omero / figure / ). Post-therapy toxicity monitoring

[0332] Spleen asurement

[0333] The coronal oblique length of FFPE spleen cross-sections was measured and plotted as the ratio of spleen length to mouse bodyweight at the spleen collection time point.

[0334] Mouse weight monitoring

[0335] Animals were weighed weekly, starting from the tumor-implantation day until endpoint.

[0336] Plasma

[0337] Plasma IL6 was assessed from the Olink proteomics dataset described in the “Olink proteomics of mouse plasma” section. Plasma mouse CRP levels were investigated with a Mouse C- Reactive Protein ELISA kit (#41-CRPMS-E01, ALPCO Diagnostics, USA), following the manufacturer’s instructions. The chromogenic reactions were stopped by adding 50 pL per well of 10% sulfuric acid (H2SO4), and absorbance values were measured by a Synergy Hl Hybrid microplate reader (BioTek, USA). Absolute CRP concentrations in test samples were interpolated from a standard curve and represented using GraphPad Prism vlO Software.

[0338] Automated hematolo alysis

[0339] Longitudinal analysis of hematological parameters was performed after a single CAR treatment using 20 pL of mouse blood collected from the tail vein into 1.5 mL Eppendorf tubes prefilled with 120 pL of 1 mM EDTA diluted in PBS. Blood samples from multiple time points (1, 3, 6, 13, and 20 days after treatment) were immediately acquired on a XN-1000 benchtop hemocytometer (Sysmex, Japan). Values were normalized and scaled to allow multiple time point comparisons.

[0340] Mye antification

[0341] Brain sample collection and processing and the myelin staining procedure were described above in the “Mouse brain immunohistochemistry” section. Images were color-deconvolved (Ruifrok AC, Johnston DA. Quantification of histochemical staining by color deconvolution. Anal Quant Cytol Histol. 2001 Aug;23(4):291-9. PMID: 11531144.) to separate luxol and cresyl violet staining using QuPath v0.5.0 (Bankhead P, Loughrey MB, Fernandez JA, Dombrowski Y, McArt DG, Dunne PD, McQuaid S, Gray RT, Murray LJ, Coleman HG, James JA, Salto-Tellez M, Hamilton PW. QuPath: Open source software for digital pathology image analysis. Sci Rep. 2017 Dec 4;7(1): 16878. doi: 10.1038 / s41598-017-17204-5. PMID: 29203879; PMCID: PMC5715110.). The Random Trees Classifier was trained to detect brain tissue and myelin area with the following parameters: Pixel size: 7.07 pm; Channels: luxol and cresyl violet; Features: Gaussian, structure tensor eigenvalue max, and hessian determinant"; Scales: 1.0 and 4.0; no normalization. The positive area of myelin (luxol staining) was quantified within brain tissue as an area above the threshold. Results were visually verified. The relative myelin area per sample was plotted using GraphPad Prism vlO Software.

[0342] Spectral flow cytometry

[0343] Freshly dissociated cells were resuspended in 400 pL PBS. 200 pL per well was distributed into a 96-well plate (200 pL for FS, 100 pL for US and 100 pL for FMO). Primary mouse and human cells were isolated as described above and stained with Zombie fixable viability dye for 20 min at RT (#423102, BioLegend, USA) and pre-incubated with anti-mouse Fc block at 10 mg mL-1 (#101320, BioLegend, USA). Surface markers were stained with appropriate antibodies for 20 min at 4°C. For intracellular staining, cells were fixed and permeabilized using Cyto-Fast Fix / Perm Buffer Set (#426803, BioLegend, USA), prior to staining for 20 min at RT in the dark. For intracellular cytokine staining, cells were fixed, permeabilized and washed using True-Nuclear Transcription Factor Buffer Set (#424401, BioLegend, USA) according to the manufacturer’s protocol. Followed by staining with intracellular antibodies for 30 min at RT in the dark. After the respective staining protocol, cells were washed twice and resuspended in FACS buffer. For FC, either a Cytoflex S (Beckman Coulter) flow cytometer or an Aurora spectral flow cytometer was used for cell acquisition and FlowJo software (v.10.8.1, TreeStar) for data analysis. Cell sorting was performed using a BD FACSAria III or BD FACSMelody (BD Bioscience). We performed compensation using Ultracomp beads Compensation Beads (#01-2222-42, Invitrogen, USA), which were stained with appropriate antibodies and analyzed on the same voltage and settings. Gates were drawn by using Fluorescent Minus One (FMO) controls. Either the percentage of cell population of interest or Median Fluorescent Intensity (MFI) is reported. Panel information: Table 5 below.

[0344] Table 5

[0345] FlowSOM analysis

[0346] Data were manually pre-gated to remove the debris and select for CD45+, live, single cells using FlowJo vlO Software. The analysis was subsequently performed in R (version 4.3.1). Data was transformed by asinh transformation using variance stabilizing cofactors for each channel (estParamFlowVS and transFlowVS functions from the FlowVS package). Pre-processing QC (using PeacoQC function from the PeacoQC package) was performed to remove outliers and unstable events (IT limit was set at 0.55 and MAD at 6). Clustering was performed using FlowSOM and ConsensusClusterPlus using the wrapper function cluster from the CATALYS package (xdim = 10, ydim = 10, maxK = 15, seed = 1234). The resulting clusters were manually annotated. Differential testing was performed using the diffcyt package (diffcyt-DA-edgeR for differential abundance and diffcyt-DS-limma for differential state).

[0347] Human GBM samples and patient information

[0348] Freshly resected primary tumor tissue samples from patients with a pathology-confirmed GBM diagnosis were obtained from the Neurosurgical Clinic of the University Hospital Basel, Switzerland, following the Swiss Human Research Act and institutional ethics commission (EKNZ 02019-02358). The study was conducted following the ethical principles of the Declaration of Helsinki, regulatory requirements, and the Code of Good Clinical Practice. All patients gave written informed consent for tumor biopsy collection and signed a declaration enabling the use of their biopsy specimens in scientific research (Req-2019-00553) with all identifying information removed. All patients were treatment-naive at the time of tissue collection. Clinical sample information is summarized in Table 6 below:

[0349] Table 6

[0350]

[0351] Human GBM tissue processing

[0352] GBM tissue samples were transported on ice and transferred to the laboratory for dissociation into single-cell suspensions within 2-3 h after surgical resection. Human brain tissue was mechanically minced using razor blades and enzymatically dissociated as described above (under “Spectral Flow Cytometry”) except that debris and myelin were removed by a 0.9 M sucrose (#84100, Sigma-Aldrich, USA) density gradient centrifugation. After ACK-lysis, the single-cell suspensions were washed with PBS, resuspended in Bambanker at an approximate density of 2 * 106live cells per mL and stored long-term in LN2.

[0353] Real-time quantitative PCR

[0354] The total RNA contents of approximately 1 x 106cells from human CAR T cells were extracted with TRIzol Reagent (#15596026, Invitrogen, USA) and an AllPrep DNA / RNA / Protein Mini kit (#80004, QIAGEN, USA). cDNA was synthesized with an iScript cDNA Synthesis kit (#1708891, Bio-Rad, USA) using 500 ng of input RNA in a 20 pL reaction. The reaction included priming for 5 min at 25°C, reverse transcription for 20 min at 46°C, and inactivation for 1 min at 95°C. RT-qPCR reactions were performed with a SsoFast EvaGreen supermix (#172-5201, Bio-Rad, USA) using 2 pL of input cDNA in a 20 pL reaction. The total RNA contents of approximately 2 * 106cells from culture-dissociated tumor cells or human GBM single-cell suspensions were extracted with TRIzol Reagent and a Direct-zol RNA Miniprep Plus kit (#R2070, Zymo Research, USA). cDNA was synthesized with a SuperScript VILO cDNA Synthesis kit (#11754-050, ThermoFisher Scientific, USA) using 20 ng of input RNA in a 20 pL reaction. The reaction included priming for 10 min at 25°C, reverse transcription for 60 min at 42°C, and inactivation for 5 min at 85°C. RT-qPCR reactions were performed with a SsoFast EvaGreen supermix using 2 pL of input cDNA in a 20 pL reaction. RT-qPCR primers were used at a final concentration of 1 pM and are listed in Table 7 below. Thermal cycling was performed using a CFX96 Touch Real-Time PCR Detection System (Bio-Rad). It included an initial step of enzyme activation for 30 sec at 95°C, followed by 40 cycles of denaturation for 5 sec at 95°C and annealing / extension for 5 sec at 60°C. Relative mCherry, SGRP, EGFR, and EGFRvIII expression was calculated with the AACt method, using TBP or GAPDH Ct to normalize signal expression.

[0355] Table 7

[0356] Pharmacoscopy

[0357] Frozen patient single-cell suspensions were thawed and resuspended in DMEM medium supplemented with 10% FBS, 25 mM HEPES and 1% pen strep and were seeded at 8 * 103cells per well in 50 pL per well into clear-bottom, tissue-culture treated, CellCarrier-384 Ultra Microplates (#50-209-8071, Perkin Elmer, USA). mCherry-labeled CAR T cells were added and co-cultured with the patient single cell suspensions for 48 h at 37°C, 5% CO2. Every CAR T cell co-incubation condition was tested with five technical replicate wells and six technical replicate wells for the PBS control. After the co-incubation period, the cells were fixed with 4% PF A, blocked with PBS containing 5% FBS, 0.1% Triton-X and 4 pg / mL DAPI (#422801, Biolegend, USA) for 1 h at RT and stained with two staining panels: (1) Alexa Fluor 488 antihuman CD3 (1 :300, #300415, Biolegend, USA) and Alexa Fluor 647 anti-human CD14 (1 :300, #325612, Biolegend, USA) and (2) primary antibodies mouse anti-human nestin (1 : 150, #656802, Biolegend, USA) and rabbit anti-human EGFRvIII (1 : 150, #MA5-36216, ThermoFisher Scientific, USA) overnight at 4°C. The wells were washed, and the cells that were stained with primary antibodies were incubated with the respective secondary antibodies, Alexa Fluor 488 goat anti-mouse IgG (1 :500, #A32723TR, Invitrogen, USA) and Alexa Fluor 647 goat anti-rabbit IgG (1 :500, #A32733, Invitrogen, USA) for 1 h at RT in the dark. The plates were imaged with an Opera Phenix automated spinning-disk confocal microscope at 20X magnification (Perkin Elmer, USA). Single cells were segmented based on nuclear DAPI staining with CellProfiler v2.2.0 Software. Downstream image analysis was performed with MATLAB R2021b. Marker-positive cell counts for each condition were identified based on a linear threshold of each channel, were averaged across each well and compared between CAR T cell co-incubation conditions.

[0358] Statistical analyses

[0359] Data analysis and visualization were performed using Excel version 16.14.1 (Microsoft) and Prism 8.4 (GraphPad). Graphs represent either group mean values ± s.d. (for in vitro experiments) or ± s.e.m. (for in vivo experiments) or individual values. For in vitro studies, statistical comparisons were made with either unpaired / -tests when comparing two groups or one-way ANOVA with multiple comparison corrections when comparing more than two groups. For in vivo studies, survival curves were compared with the log-rank test; tumor growth was compared with repeated-measures ANOVA and the Mann-Whitney test was used to compare two groups. P < 0.05 was considered statistically significant. Significance is shown with P or adjusted P values. In some Figures, P values are denoted with asterisks: * P < 0.05; ** < 0.01; *** < 0.001; and **** P < 0.0001.

[0360] Results drived blocker of the phagocytosis axis

[0361] Antigen escape is a significant mechanism of anti-EGFRvIII CAR T cell therapy resistance (O'Rourke DM, Nasrallah MP, Desai A, Melenhorst JJ, Mansfield K, Morrissette JJD, Martinez-Lage M, Brem S, Maloney E, Shen A, Isaacs R, Mohan S, Plesa G, Lacey SF, Navenot JM, Zheng Z, Levine BL, Okada H, June CH, Brogdon JL, Maus MV. A single dose of peripherally infused EGFRvIII-directed CAR T cells mediates antigen loss and induces adaptive resistance in patients with recurrent glioblastoma. Sci Transl Med. 2017 Jul 19;9(399):eaaa0984. doi: 10.1126 / scitranslmed.aaa0984. PMID: 28724573; PMCID: PMC5762203.) (Fig. 1). We propose a fourth-generation CAR design, whereby anti-EGFRvIII CAR T cells constitutively release a SIRPy-related protein (SGRP) with high affinity to CD47 (Fig. 1). The reported dissociation constant (KD) of SIRPa binding to human CD47 is 279 nM (Weiskopf K, Ring AM, Ho CC, Volkmer JP, Levin AM, Volkmer AK, Ozkan E, Femhoff NB, van de Rijn M, Weissman IL, Garcia KC. Engineered SIRPa variants as immunotherapeutic adjuvants to anticancer antibodies. Science. 2013 Jul 5;341(6141):88-91. doi: 10.1126 / science.1238856. Epub 2013 May 30. PMID: 23722425; PMCID: PMC3810306.). A SIRPa analog, CV1, recently used in CAR constructs for peripheral tumors, has a KD of 11.1 pM (Dacek MM, Kurtz KG, Wallisch P, Pierre SA, Khayat S, Bourne CM, Gardner TJ, Vogt KC, Aquino N, Younes A, Scheinberg DA. Potentiating antibody-dependent killing of cancers with CAR T cells secreting CD47-SIRPa checkpoint blocker. Blood. 2023 Apr 20;141(16):2003-2015. doi: 10.1182 / blood.2022016101. PMID: 36696633; PMCID: PMC10163312. ). In comparison, SGRP used in the present study, has a reported KD of 92 pM (US20160340397A1), far outcompeting endogenous human phagocyte-expressed SIRPa and SIRPa analogons in the literature. To design a CAR T cell that concomitantly targets GBM cells via EGFRvIII recognition and reprograms GAMs by blocking a major regulatory pathway of phagocytosis, we armed anti-EGFRvIII-BBz CAR T cells (3C10.BBz33, termed aEGFRvIII CAR) with a secretable SGRP by adding an IL2 signal peptide to the SGRP sequence (Fig. 2), resulting in aEGFRvIII- SGRP CAR). Synthetic SGRP shares a strong identity with the endogenous SIRPy binding domain (SIRPy-Vl) amino acid (AA) sequence (Table 1) and binds to human CD47 in a similar manner as competitive human and murine SIRPa (Fig. 3). Specifically, we generated / Wd -driven polycistronic constructs encoding overall identical CAR structures targeting either EGFRvIII or control antigen CD 19 (FMC63.BBz35; termed aCD19 CAR), a mCherry (mC) fluorescent reporter, and a secretable SGRP. The resulting polyproteins were generated by flanking T2A or P2A self-cleaving peptide sequences (Fig. 4 and Table 2) and were incorporated into lentiviral expression vectors (Fig. 5). EFlA-driven polycistronic constructs encoding overall identical CAR structures targeting either EGFRvIII or control antigen CD 19 (FMC63.BBz35; termed aCD19 CAR), and a secretable SGRP and a secretable SGRP (without a mCherry (mC) fluorescent reporter) are shown in SEQ ID NO: 1 (nucleotide sequence of CD19 construct) and SEQ ID NO: 2 (nucleotide sequence of EGFRvIII construct) (Table 2). The corresponding amino acid sequences are shown in SEQ ID NO: 3 (nucleotide sequence of CD19 construct) and SEQ ID NO: 4 (nucleotide sequence of EGFRvIII construct).

[0362] CAR T cells were produced using healthy donor (HD) peripheral blood mononuclear cells (PBMCs), depleted for non-T cells and activated with a cocktail of IL2, IL7, IL15, IL21 and anti-CD3 / CD28 microbeads for 48 h44. Activated T cells were stably transduced with lentiviral supernatants and T cell cultures were expanded in the presence of IL2 for 5 to 7 days, subsequently enriched for mCherry-expression by cell sorting, and further expanded for downstream applications (Fig. 6, 7). As intended, aCD19 and aEGFRvIII CARs showed specific binding affinity to their respective targets, but not to mismatched targets (Fig. 8) or wild type (wt) EGFR proteins (Fig. 9).

[0363] We then confirmed the presence of similar amounts of mCherry mRNA in aEGFRvIII CAR and aEGFRvIII- S GRP CAR, and SGRP mRNA specifically in aEGFRvIII- S GRP CAR T cells from four T cell donors (Fig. 10). In conditioned media from 24 h-rested, antigen-naive, CAR T cell cultures from 2 donors, SGRP was the most differentially enriched protein in the secretome of aEGFRvIII- SGRP CAR compared to aEGFRvIII CAR, detected by LC-MS (Fig. 11, 12). The identified peptide sequences included SGRP-specific AA modifications, confirming specific SGRP detection rather than contamination with peptides of highly conserved SIRP-family proteins (Fig. 13).

[0364] CAR I cell effector function against GBM mediated CD47 blockade in vitro

[0365] To validate CAR and CAR-SGRP T cells, we profiled surface target expression in four GBM cell lines (U251, U87, U251vIII with transgenic (tg) overexpression of EGFRvIII, BS153 with endogenous EGFRvIII expression), one Burkitt’s lymphoma cell line (Raji) and one normal neural stem cell (NSC) line (NSC197). We used aCD19 CAR and aCD19-SGRP CAR as controls, and Raji as CD19+target cells throughout the study (Fig. 14, left plot). We confirmed the EGFRvIlF status of GBM cell lines by surface staining of U251vIII and BS153 (99.3% and 28.9%, respectively), while U251 and U87 were EGFRvIII-neutral (16.4% and 23.5%, respectively), exhibiting smaller fractions of cells EGFRvIII-expressing cells (Fig. 14, center plot). Prior to in vitro and in vivo experiments, U251vIII and BS153 were enriched for EGFRvIII by cell sorting, while U251 and U87 were enriched for their EGFRvIlF fractions ensuring EGFRvIII negativity during experimental validation. Furthermore, we also confirmed CD47 overexpression in all aforementioned tumor cell lines (U251vIII: 56.5%, U251: 51.4%, U87: 53.5%, BS153: 80.6% and Raji: 58.4%), compared to normal NSCs (Fig. 14, right plot). We then established co-cultures of mCherry+CAR or CAR-SGRP T cells with GBM cells expressing a nuclear-restricted EGFP (nEGFP) as a fluorescence viability reporter. A decline of the nEGFP signal in the co-cultures meant a reduction of tumor viability, as previously validated by puromycin selection (Fig. 15). Time lapse imaging revealed the specific cytotoxic capacity of aEGFRvIII CAR and aEGFRvIII- SGRP CAR against U251vIII cells (Fig. 16). aEGFRvIII CAR also showed similar efficacy against BS153 cells (Fig. 17). By contrast, no measurable off-target cytotoxicity was observed against U251, an EGFRvIII-negative GBM cell line (Fig. 18). At various time points and effector-target (E:T) ratios, targeted CAR T cells displayed a dose-dependent effect against EGFRvIlFgGBM cells (Fig. 19). Moreover, CAR T cell cytotoxic capacity remained unaffected by SGRP secretion in co-cultures with U251vIII (Fig. 19) or U251 (Fig. 20), as shown by the largely identical responses of aEGFRvIII CAR and aEGFRvIII- SGRP CAR in co-cultures with GBM cells in the absence of tumor-associated myeloid cells. Notably, we observed a natural enrichment of CD4+T cells in CAR T cell products with our expansion protocol (Fig. 21).

[0366] We further examined the degranulating and cytokine-producing capacity of CAR T cells in cocultures with GBM cell lines of varying EGFRvIII status. FC analysis of CAR T cells in 24 h co-cultures showed that targeted CAR T cells specifically expressed higher levels of CD 107a in response to EGFRvIlF GBM cell lines (Fig. 22). We also detected significantly higher concentrations of IFNy in co-cultures of targeted CAR T cells with EGFRvIlF, but not with EGFRvIlF GBM cell lines (Fig. 23).

[0367] To determine the CD47-blocking capacity of T cell-secreted SGRP compared to anti-human CD47 antibody (aCD47) in vitro, we pre-treated BS153 tumor cells with CAR T cell- conditioned media (aEGFRvIII- SGRP CAR vs aEGFRvIII CAR), or antibody dilutions (anti- CD47, clone B6.H12 vs. IgGl isotype, clone MOPC-21). In a second step, BS153 cells were exposed to biotinylated SIRPa (bt-SIRPa) that bound to the remaining available CD47 on tumor cells. Cell-bound bt-SIRPa was detected by labeling with fluorescence-conjugated streptavidin (SA). aCD47 treatment induced a potent blockade of the CD47-SIRPa interaction (Fig. 24, left dot-plot) whereas treatment with aEGFRvIII- SGRP CAR-conditioned medium slightly impaired this interaction (Fig. 24, right dot-plot). Since CD47 is ubiquitously expressed21and SGRP has a very high affinity to CD47, we hypothesize that SGRP may be partially captured in autocrine or paracrine loops by CAR T cells. Consequently, the amount of available SGRP collected from cell culture supernatants may be reduced, contributing to the apparent lower efficacy of SGRP-mediated CD47 blockade. Importantly, CD47 expression is required for CAR T and peripheral T cell survival, as shown by studies using CD47 mutant T cells (Beckett AN, Chockley P, Pruett-Miller SM, Nguyen P, Vogel P, Sheppard H, Krenciute G, Gottschalk S, DeRenzo C. CD47 expression is critical for CAR T-cell survival in vivo. J Immunother Cancer. 2023 Mar;l l(3):e005857. doi: 10.1136 / jitc-2022-005857. PMID: 36918226; PMCID: PMC10016274; Komori S, Saito Y, Nishimura T, Respatika D, Endoh H, Yoshida H, Sugihara R, lida-Norita R, Afroj T, Takai T, Oduori OS, Nitta E, Kotani T, Murata Y, Kaneko Y, Nitta R, Ohnishi H, Matozaki T. CD47 promotes peripheral T cell survival by preventing dendritic cell-mediated T cell necroptosis. Proc Natl Acad Sci U S A. 2023 Aug 15;120(33):e2304943120. doi: 10.1073 / pnas.2304943120. Epub 2023 Aug 7. PMID: 37549290; PMCID: PMC10440595.). Despite constitutive secretion of SGRP, CD47 surface expression remained unchanged on aEGFRvIII-SGRP CAR T cells (Fig. 25).

[0368] In an attempt to quantify the effects of CAR targeting and SGRP-mediated macrophage modulation in vitro, we performed a 3 h-phagocytosis assay combining a mosaic GBM model consisting of EGFRvIII' U87, EGFRvIII U251vIII cells with donor-matched macrophages and CAR T cells in a 1 : 1 : 1 : 1 ratio (Fig. 26). In this short-term phagocytosis assay, we assessed: (1) macrophage polarization, by human CD163, CD206, HLA-DR, and CD86; (2) phagocytic marker expression, by human CD209; (3) antigen presentation, by human SIGLEC-1, and (4) tumor cell phagocytosis. However, in this in vitro system, we found no significant differences between conditions involving conventional CARs and SGRP-secreting CARs.

[0369] Nevertheless, in vitro models overlook crucial cell interactions in the context of the GBM TME, namely the impact of other phagocytes, e.g., microglia and dendritic cells (DCs), on in vivo tumor clearance. Thus, understanding the effect of T cell-secreted SGRP requires a GAM- infiltrated in vivo GBM model.

[0370] Anti- combination therapy leads to significant survival bene > ' in . H .vigressi i mi i .U! nosar ' h xeru J , model

[0371] Many aEGFRvIII CAR preclinical studies show effective cytotoxicity in vitro and in vivo but fail to model the heterogeneity of human GBM45,46. Although preclinical aEGFRvIII CAR monotherapy has demonstrated efficacy in overall tumor control, this strategy translates poorly in the clinical setting10. To establish a baseline for combination therapies, we first tested the efficacy of aEGFRvIII CAR and aCD47 monotherapies in an EGFRvIII GBM xenograft model. For these experiments, U251vIII cells were enriched for EGFRvIII by cell sorting and orthotopically implanted in NSG mice. Tumor-bearing mice were treated with 2 doses, 7 and 14 days after tumor implantation. Clinical scores and tumor growth were monitored weekly by bioluminescence imaging until a maximum of 13 weeks (90 days) or until animals reached the humane endpoint (Fig. 27). The therapeutic scheme included i.t. treatments of mouse IgGl (MOPC-21; Isotype), anti-human CD47 (B6.H12; aCD47), aCD19 CAR and aEGFRvIII CAR (Fig. 28). Overall survival analysis (Fig. 29) showed that local aCD47 monotherapy failed to improve survival (median survival: 34.5 days) when compared to the Isotype antibody control group (median survival: 30.5 days). Conversely, aEGFRvIII CAR treatment (median survival: 65.5 days) led to 40% survival by day 90 after tumor implantation, a stark contrast to aCD19 CAR control (median survival: 29 days). Bioluminescence assessment confirmed an overall lower tumor burden in the aEGFRvIII CAR treatment group (Fig. 30).

[0372] To assess the efficacy of CAR-SGRP combination therapy against GBM, we established a preclinical EGFRvIII-mosaic GBM model with orthotopic tumor implantation and intracranial therapy administration in mice. First, we enriched GBM cell lines for their EGFRvIlF (BS153 and U251vIII) or EGFRvIlF (U87 and U251) cell subsets based on cell surface staining followed by cell sorting and then assessed their engraftment in NSG mice. We selected U251vIII and U87 for their consistent engraftment and fast progression to the onset of clinical signs. Thereafter, we co-injected U251vIII and U87 intracranially in a 1 : 1 ratio as single-cell suspensions and allowed them to settle for 7 days. All animals received 2 intratumoral (i.t.) treatment doses (5 x 105CAR T cells or 5 pg of antibody) 7 and 14 d after tumor implantation. For better equipoise in comparing antibody-based CD47 blockade (aCD47) with constitutive CAR-mediated SGRP release, animals receiving aCD47 as monotherapy or in combination with aEGFRvIII CARs (aEGFRvIII CAR + aCD47) received 4 additional i.p. antibody doses (100 pg of antibody) in the 2 weeks following the i.t. treatment regimen. To monitor U251vIII and U87 cell populations separately in vivo, we differentially labeled U251 vIII with NanoLuciferase (NLuc) and U87 with Luciferase2 (Luc2) bioluminescence reporters using lentiviral transgene delivery. The experiment timeline and therapeutic setup are illustrated in Fig. 31, 32. Overall survival analysis (Fig. 33) showed that untreated animals in the Vehicle control group had a median survival of 31.5 days. Surprisingly, combined locoregional / systemic aCD47 monotherapy failed to improve survival (median survival: 34 days) when compared to an Isotype antibody control group (median survival: 32 days). We reasoned that despite combined local / systemic administration, the aCD47 treatment regimen did not maintain a sufficient blockade of CD47 to levels that could induce persistent GAM phagocytic activity. Notably, aggressive tumor models in the NSG context are difficult to treat using even higher doses of aCD47 antibodies23. As expected, treatment with non-GBM-targeted aCD19 CARs resulted in a minor, non-significant survival benefit (median survival: 36 days). Similarly, survival after aCD19-SGRP CAR therapy (median survival: 40 days) suggested that peripherally-released SGRP without direct CAR-mediated targeting was insufficient to eliminate orthotopic xenograft tumors. As expected, aEGFRvIII CAR monotherapy was less efficacious in the EGFRvIII-mosaic tumor model compared to the homogenous EGFRvIII+tumor model. Nonetheless, aEGFRvIII CAR was able to eliminate GBM in 20% of treated animals (Fig. 33; tumor-free survival analysis in Fig. 34). However, the combination of aEGFRvIII CAR + aCD47 failed to improve survival beyond what we had observed with aEGFRvIII CAR monotherapy (20%). This pointed towards a potential elimination of grafted CD47+CAR T cells via phagocytosis or insufficient continuous dosing of local anti-CD47 antibodies resulting in failure of additive efficacy in this model. Notably, clone B6.H12 was used in the experiments because of the limited availability of clone Hu5F9-G447. In contrast, aEGFRvIII- S GRP CAR combination therapy was highly potent in this challenging GBM model. A near complete therapeutic response was observed with 94.7% overall survival (Fig. 33) and 63.2% tumor-free survival (Fig. 34).

[0373] Differential bioluminescence monitoring of either U251vIII-NLuc (Fig. 35) or U87-Luc2 (Fig. 36) confirmed that a majority of EGFRvIIF tumors were cleared after aEGFRvIII- S GRP CAR treatment. Bioluminescence plots of individual treatment groups are shown in Fig. 37. A direct bioluminescence imaging comparison and quantification of tumor burden of EGFRvIIF and EGFRvIIF tumor fractions in aEGFRvIII CAR, aEGFRvIII CAR + aCD47 and aEGFRvIII- SGRP CAR treatment groups on a representative time point (week 7 post tumor implantation) are shown in Fig. 38, 39.

[0374] To further elucidate the mechanism of aEGFRvIII- S GRP CAR-mediated tumor clearance, we performed an in vivo experiment with orthotopic EGFRvIIF (U87) or EGFRvIII-mosaic (U87+U251vIII) tumor implantation and treatment with aEGFRvIII- S GRP or aCD19-SGRP CAR (Fig. 40). Three days after the treatment, we collected the tumor-implanted brain hemisphere to characterize condition-specific immune cell abundance (cluster definition is shown in Fig. 41). Treatment with aEGFRvIII- SGRP CAR resulted in higher T cell frequencies in the tumor-implanted brain hemisphere compared to aCD19-SGRP CAR, regardless of tumor type (Fig. 42). Notably, aEGFRvIII- SGRP CAR treatment led to a higher influx of neutrophils, moMacs, and monocyte-derived DCs (moDCs) / conventional type 1 DCs (cDCls) and an increase in CD25hlCD4+CAR T cells exclusively in EGFRvIII-mosaic tumors (Fig. 42). This suggests that the benefit of SGRP-mediated CD47 blockade combined with CAR-mediated tumor killing relies on CAR target priming, leading to CAR T cell activation and persistence and enhanced myeloid cell infiltration to the brain.

[0375] Tumor rechallenge of aEGFRvIII- SGRP CAR-cured mice in the GBM cohort of Fig. 33, 34 resulted in prolonged survival compared to the historic vehicle control group (Fig. 43). Interestingly, tumor monitoring after rechallenge revealed preferential control of the EGFRvIII fraction (Fig. 44). Residual aEGFRvIII-SGRP CAR T cells in tumor-free brains 90 days post tumor implantation were few and mostly found at the brain periphery (Fig. 45, 46), suggesting a potential role for host innate immune cell priming, or lasting TME alterations from aEGFRvIII-SGRP CAR treatment, resulting in reduced tumor engraftment.

[0376] Anti-1 UP C \H,’ combination therapy induces a myeloid-driven inflammatory respont

[0377] To elucidate potential mechanisms of innate immune activation upon locoregional CAR T cell GBM treatments, we conducted a human immuno-oncology-targeted proteomic analysis of plasma samples collected 24 h after the second CAR treatment (day 15 post tumor implantation). Using an Olink Immuno-Oncology panel, we analyzed a total of 92 proteins on 2 complementary datasets encompassing 54 plasma samples from individual mice (Fig. 47 ). Plasma from sex- and age-matched healthy control mice was used to determine baseline protein expression. We first visualized the distribution of all assessed markers with a principal component analysis which showed that aEGFRvIII CAR- and aEGFRvIII-SGRP CAR-treated plasma segregated the furthest from vehicle controls (Fig. 48). To examine the peripheral immune response associated with the superior aEGFRvIII- S GRP CAR treatment, we performed differential expression analysis against aEGFRvIII CAR monotherapy (Fig. 49), which showed significant enrichment of CCL3 and IL13 in aEGFRvIII- S GRP CAR-treated plasma (adj. P < 0.001 and adj. P < 0.05, respectively). Conversely, CD27 was significantly increased in aEGFRvIII CAR-treated plasma (adj. P < 0.05). To identify additional immune markers associated with aEGFRvIII CAR + / - SGRP, we looked for differences between SGRP-secreting CARs (aEGFRvIII- SGRP CAR vs aCD19-SGRP CAR) or non-SGRP-secreting CARs (aEGFRvIII CAR vs aCD19 CAR) across all 92 markers (Fig. 50). This analysis revealed significant differences in the expression of CCL3, CXCL1, CXCL8, GZMA, TNFRSF21, TNFSF14 and VEGFA between SGRP-secreting CAR T cell treatments. Moreover, it showed significant differences in CD5, CXCL1, GZMA, IFNG, KLRD1, PGF and TNFSF14 expression between non-SGRP-secreting CAR T cell monotherapies. Collectively, these analyses identified CCL3 as the key plasma marker associated with aEGFRvIII- SGRP CAR treatment response out of all 92 markers studied. T cell-activation markers including GZMA and IFNG were found to be similarly highly expressed in aEGFRvIII CAR and aEGFRvIII- SGRP CAR plasma.

[0378] The myeloid chemoattractant CCL3 emerged as a significantly upregulated protein in the plasma of mice treated with aEGFRvIII- S GRP CAR, potentially shaping peripheral responses to local antitumor therapy, promoting antigen presentation in tumor-draining lymph nodes or inducing T cell proliferation and differentiation (Allen F, Rauhe P, Askew D, Tong AA, Nthale J, Eid S, Myers JT, Tong C, Huang AY. CCL3 Enhances Antitumor Immune Priming in the Lymph Node via fFNy with Dependency on Natural Killer Cells. Front Immunol. 2017 Oct 23;8: 1390. doi: 10.3389 / fimmu.2017.01390. PMID: 29109732; PMCID: PMC5660298.). To investigate whether CCL3 plays a crucial role in aEGFRvIII- S GRP CAR efficacy, we administered a CCL3 -neutralizing antibody i.p. with concomitant i.t. CAR T cell therapy (Fig. 51). Control groups included an isotype antibody and PBS-injected animals. Despite CCL3 blockade, we observed no reduction in the efficacy of aEGFRvIII- SGRP CAR T cells compared to control groups (Fig. 52). The slight reduction in survival across all treatment / control groups compared to aEGFRvIII- SGRP CAR treatment in Fig. 33 could be attributed to overall lower activity of the T cell batch or the significant increase in frequency and duration of animal interventions.

[0379] Anti-EGFRvlll-SGRP CAR T cells elicit an early rejection of EGFRvlll- mosaic GBM without overt signs of toxicity

[0380] To better characterize the remarkable response of GBM xenografts to aEGFRvIII- SGRP CAR combination therapy, we collected spleen and brain tissue at intermediate time points and performed a histological and immunofluorescence workup of GBM-brains from animals at an intermediate treatment time point (7 days after the 2ndtreatment dose, 21 days post tumor implantation; Fig. 53). Tumor size assessment in the 7 experimental conditions confirmed the persistence of intracerebral GBM xenografts in all but the aEGFRvIII- SGRP CAR treated animals (Fig. 54-56). Summed tumor areas calculated from brain histological sections of day 21 post tumor implantation showed the lowest tumor burden in anti-EGFRvIII and anti- EGFRvIII SGRP CAR T cell treatment groups (Fig. 56) . Only minute tumor remnants at this time point in aEGFRvIII- SGRP treated animals could be discovered, precluding us from a detailed multidimensional analysis of the iTME in this condition despite whole-brain sectioning and staining. However, assessment of CAR T cell persistence was possible in all conditions using anti-human CD3 IHC (Fig. 57).

[0381] Multiplexed immunofluorescence focussing on GAMs (murine Ibal, Tmeml l9, CD206), astrocytes (murine GFAP), CAR T cells (human CD3) and tumor surface / proliferation markers (human EGFRvIII, Ki67) confirmed the persistence of CAR T cells in aCD19 CAR, aEGFRvIII CAR and aEGFRvIII-CAR + aCD47 treated animals, albeit with a comparatively reduced number of CAR T cells in aCD19 CAR and aEGFRvIII-CAR + aCD47 tumor burdened brains (Fig. 58). This underscores the importance of the presence of tumor antigen in case of aCD19 CAR treatments, and hints at a potential CAR phagocytic effect in case of additional anti-CD47 treatment. The quantification of EGFRvIII surface expression confirmed selective / preferential eradication of EGFRvIII tumor cells after any aEGFRvIII-specific treatment, accompanied by the persistence of EGFRvIIF tumor cells. Analysis of GAM polarization markers revealed a reduction of CD206+cells in aCD47 treated animals in line with published data24, while overall macrophage influx by Ibal+cells tended to be higher in aEGFRvIII-CAR + aCD47 treated animals (Fig. 58). Alternatively, anti-mouse CD68 IHC of tumor-burdened brains showed a tendency towards a more pronounced proinflammatory macrophage activation in aEGFRvIII- SGRP -treated mice over the other CAR-treated conditions (Fig. 59).

[0382] Despite its efficacy, aEGFRvIII- S GRP CAR showed no signs of systemic toxicity, as evidenced by the absence of splenomegaly after two rounds of therapy (Fig. 60), and the normal weight fluctuation throughout the experiment (Fig. 61). We observed a sharp drop in the weight of a few animals after the first aEGFRvIII- S GRP CAR treatment, suggestive of acute inflammatory events that resolve on their own after the second treatment (Fig. 61, right plot). Moreover, we examined CAR hematotoxicity parameters (Rejeski K, Perez A, Sesques P, Hoster E, Berger C, Jentzsch L, Mougiakakos D, Frblich L, Ackermann J, Bucklein V, Blumenberg V, Schmidt C, Jallades L, Fehse B, Faul C, Karschnia P, Weigert O, Dreyling M, Locke FL, von Bergwelt- Baildon M, Mackensen A, Bethge W, Ayuk F, Bachy E, Salles G, Jain MD, Subklewe M. CAR- HEMATOTOX: a model for CAR T-cell-related hematologic toxicity in relapsed / refractory large B-cell lymphoma. Blood. 2021 Dec 16;138(24):2499-2513. doi: 10.1182 / blood.2020010543. PMID: 34166502; PMCID: PMC8893508.: plasma IL6 and CRP, erythrocyte (RBC), platelet (PLT), and neutrophil (NEUT) counts, but found no indication of increased inflammation, anemia, thrombocytopenia or neutropenia (Fig. 62-64). The increase in neutrophil counts in the vehicle group is likely linked to a higher tumor burden in these animals from day 13 on (Fig. 64, bottom plot). Locally, we did not detect differences in brain myelination after therapy, an additional indication of safety, particularly in a model of intracerebral CAR T cell infusion (Fig. 65, 66).

[0383] CAR treatments induce a proinflammatory GAM conversion, and SGRP improves GAM-mediated tumor phagocytosis

[0384] To further elucidate the differential effects of aEGFRvIII-SGRP CAR treatments on the iTME during treatment, we performed a focused spectral flow cytometry analysis of myeloid cells shortly after i.t. treatment with CAR T cells. We designed and optimized a panel consisting of 17 myeloid-specific surface markers and 3 intracellular markers including the detection of endogenous fluorescence by tumor cells and CAR T cells (Fig. 67). In comparison to vehicle treatment, both aEGFRvIII and EGFRvIII-SGRP triggered a rapid influx of CD45+cells (Fig. 68). We confirmed the presence of mCherry+CAR T cells and mTagBFP2+EGFRvIII+tumor cells in the CD45' gate, while specifically gating for murine CD45+cells, and both CAR treatments reduced tumor cell burden at this stage (Fig. 69). We then specifically elaborated on potential early time point differences between tumor-associated microglia and macrophage subsets in the host NSG mice. Using conventional analysis, two main microglia populations arose upon CAR T cell treatment: ‘resting’ microglia were characterized by P2YR12hlF4 / 80intwhile activated microglia were defined as P2YR12intF4 / 80hl. Other influxing myeloid cells consisted of P2YR12negF4 / 80intmonocyte-derived macrophages and P2YR12negF4 / 80negmyeloid cells, most of which were determined to be Ly6g+neutrophils (Fig. 70, 71). Activated microglia expressed CD 11c, MHCII, CD86 and Siglec-H, associated with a proinflammatory phenotype (Fig. 72, 73). We next assessed the intracellular signal intensity of eBFP as a surrogate for tumor cell phagocytosis. aEGFRvIII- SGRP treated tumor-associated activated microglia, microglia and macrophages displayed a significantly increased eBFP-signal intensity compared to aEGFRvIII CAR T cell suggesting SGRP-mediated phagocytosis of BFP+tumor cells in the early treatment phase (Fig. 73). Furthermore, aEGFRvIII- SGRP CAR-treated activated microglia displayed increased intracellular TNFa measurements by trend compared to EGFRvIII-CARs alone (Fig. 74).

[0385] We alternatively performed a multiparametric analysis combined with FlowSOM metaclustering of GAMs. Initial clustering identified 15 metaclusters (TSNEs / UMAP in Fig. 75) that were characterized by differential lineage marker expression (Fig. 76, 77). Cluster frequencies between aEGFRvIII-CAR and aEGFRvIII- SGRP CAR treatment conditions were not different (Fig. 78). Subsequently, these clusters were merged into 8 main subsets including microglia, activated microglia, monocyte-derived macrophages (MoMacs), monocyte-derived dendritic cells (MoDCs), monocytes, plasmacytoid dendritic cells (pDCs), neutrophils, and unknowns (TSNE plots and UMAP in Fig. 79, 80), which exhibited similar population frequencies among conditions (Fig. 81, 82). We next focussed on the overall most abundant phagocyte population, microglia, which mainly expressed P2RY12, F4 / 80 in varying levels as seen in the conventional analysis (Fig. 83). Activated microglia expressed high levels of CD11c (Fig. 84). Further subclustering specifically on microglia revealed 3 major populations (resting, MHCIIhlgh, and activated microglia, TSNE in Fig. 85), characterized by marker profiles outlined in Fig. 86). While population frequencies between aEGFRvIII and aEGFRvIII-SGRP CAR treated animals were not different (Fig. 87), differential abundance analysis among conditions showed significantly increased XCR1 -expression in activated microglia of aEGFRvIII- S GRP CAR treated brain tumors (expression heatmap in Fig. 88, individual heatmaps of all assessed markers in Fig. 89). XCR1 is described to be a specific classical dendritic cell marker involved in antigen presentation48, however, in our dataset, the cluster ‘activated microglia’ is characterized by P2RY12int, SiglecHhlgh, MHCIIhlgh, F480hlgh, which most probably classifies them as of microglial origin. aEGFRvIII CARs outperform aCD19 CARs in co-cultures with patient- derived single cell suspensions

[0386] To demonstrate efficacy of aEGFRvIII- S GRP in a translational setting, we co-cultured CAR T cells with EGFRvIlE GBM patient-derived single cell suspensions, and performed multidimensional pharmacoscopy analysis in 5 patients (Fig. 90). CAR T cells and patient- derived GBM cells were not matched in this setup, raising the potential of alloimmunological phenomena. EGFRvIII status was determined by a custom qPCR assay (Fig. 91). Readout included assessment of Nestin+tumor cells and EGFRvIlE tumor cells, as well as CD14+immune cell counts (Fig. 92). In this setting, a direct benefit of aEGFRvIII- S GRP over EGFRvIII-CARs was not appreciated. As expected, all EGFRvIII targeted CARs significantly reduced the number of EGFRvIlE cells compared to all CD19-targeted control CARs (Fig. 93). Moreover, aCD19-SGRP CARs induced moderate antigen-independent killing of both EGFRvIlE and Nestin+tumor cells, presumably due to a potential SGRP-mediated effect or constitutive secretion of fFNy (Fig. 94). The fact that no differential effect between aEGVRvIII and aEGFRvIII- S GRP was detected could be explained by the use of frozen single-cell suspensions with reduced overall viability and subsequent functional deficits in myeloid cell composition.

[0387] Anti-CD19-SGRP CAR T cells have superior efficacy over conventional aCD19 CAR T cells in a peripheral lymphoma xenograft model

[0388] To demonstrate a potential benefit of iTME-targeted secretion of SGRP in another solid tumor context, we further assessed the capacity of aCD19-SGRP CAR T cells against CD19+lymphoma xenografts. In contrast to locoregional injection as the preferred application route in brain tumors, we considered a systemic approach mirroring CAR T cell treatments currently performed against leukemia49. CD19+Raji cells were injected s.c. in the right flank of NSG mice followed by a single dose of systemic CAR T cell infusion 3 days after tumor implantation. Animals were monitored for survival analysis and volumetric tumor burden quantification (Fig. 95). The therapeutic setup consisted of targeted CAR T cells (aCD19 CAR or aCD19-SGRP CAR) or non-targeted CAR T cells (aEGFRvIII CAR or aEFGRvIII-SGRP CAR) serving as controls (Fig. 96). As expected, all aCD19 CAR-treated animals had a significant survival benefit compared to either Vehicle or non-targeted CAR controls (Fig. 97). Strikingly, aCD19- SGRP CAR treatment resulted in the longest survival benefit (20% overall survival) with two cured animals (undetectable tumor), a significantly superior response compared to conventional aCD19 CAR application (P = 0.012; Fig. 97, 98). Thus, the contribution of SGRP-mediated innate immune modulation might be of relevance and clinical applicability in other, noncerebral cancer entities, depending on efficient CAR T cell homing to the respective tumor site.

[0389] Conclusion

[0390] In conclusion, we showed significant synergy of aEGFRvIII- SGRP CARs against orthotopic and patient-derived glioblastoma models, credentialing this approach for a first-in-human clinical trial. Additionally, this work highlights the importance of paracrine, high-affinity CD47 blockade to relieve macrophage-mediated immunosuppression, which might not be achievable by systemic or episodic local application of anti-CD47 antibodies in the context of aggressive solid tumors. In summary, this work represents one of the first examples of GBM targeting by adoptive T cells with concurrent local paracrine GAM modulation in a single therapeutic modality. References

[0391] 1. Cloughesy, T. F., Cavenee, W. K. & Mischel, P. S. Glioblastoma: from molecular pathology to targeted treatment. Annu. Rev. Pathol. 9, 1-25 (2014).

[0392] 2. McFaline-Figueroa, J. R. & Lee, E. Q. Brain Tumors. Am. J. Med. 131, 874-882 (2018).

[0393] 3. Ostrom, Q. T. et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2012-2016. Neuro-OncoL 21, vl- vlOO (2019).

[0394] 4. Newick, K., O’Brien, S., Moon, E. & Albelda, S. M. CAR T Cell Therapy for Solid Tumors. Annu. Rev. Med. 68, 139-152 (2017).

[0395] 5. June, C. H. & Sadelain, M. Chimeric Antigen Receptor Therapy. N. Engl. J. Med. 379, 64-73 (2018).

[0396] 6. Hou, A. J., Chen, L. C. & Chen, Y. Y. Navigating CAR-T cells through the solidtumour microenvironment. Nat. Rev. Drug Discov. 20, 531-550 (2021).

[0397] 7. Gan, H. K., Cvrljevic, A. N. & Johns, T. G. The epidermal growth factor receptor variant III (EGFRvIII): where wild things are altered. FEBS J. 280, 5350-5370 (2013).

[0398] 8. Felsberg, J. et al. Epidermal Growth Factor Receptor Variant III (EGFRvIII) Positivity in EGFR- Amplified Glioblastomas: Prognostic Role and Comparison between Primary and Recurrent Tumors. Clin. Cancer Res. 23, 6846-6855 (2017).

[0399] 9. An, Z., Aksoy, O., Zheng, T., Fan, Q.-W. & Weiss, W. A. Epidermal growth factor receptor and EGFRvIII in glioblastoma: signaling pathways and targeted therapies. Oncogene 37, 1561-1575 (2018).

[0400] 10. O’Rourke, D. M. et al. A single dose of peripherally infused EGFRvIII-directed CAR T cells mediates antigen loss and induces adaptive resistance in patients with recurrent glioblastoma. Sci TranslMed 9, (2017).

[0401] 11. Martins, T. A. et al. Microglia-Centered Combinatorial Strategies Against Glioblastoma. Front. Immunol. 11, 571951 (2020).

[0402] 12. Luksik, A. S., Yazigi, E., Shah, P. & Jackson, C. M. CAR T Cell Therapy in Glioblastoma: Overcoming Challenges Related to Antigen Expression. Cancers 15, 1414 (2023).

[0403] 13. Chae, Y. K. et al. Current landscape and future of dual anti-CTLA4 and PD-1 / PD-L1 blockade immunotherapy in cancer; lessons learned from clinical trials with melanoma and non-small cell lung cancer (NSCLC). J. Immunother. Cancer 6, 39 (2018). 14. Arrieta, V. A. et al. Immune checkpoint blockade in glioblastoma: from tumor heterogeneity to personalized treatment. J. Clin. Invest. 133, (2023).

[0404] 15. Yeo, A. T. et al. Single-cell RNA sequencing reveals evolution of immune landscape during glioblastoma progression. Nat. Immunol. 23, 971-984 (2022).

[0405] 16. Gieryng, A., Pszczolkowska, D., Walentynowicz, K. A., Rajan, W. D. & Kaminska, B. Immune microenvironment of gliomas. Lab. Invest. 97, 498-518 (2017).

[0406] 17. Lakshmanachetty, S. & Mitra, S. S. Mapping the tumor-infiltrating immune cells during glioblastoma progression. Nat. Immunol. 23, 826-828 (2022).

[0407] 18. Hanisch, U.-K. & Kettenmann, H. Microglia: active sensor and versatile effector cells in the normal and pathologic brain. Nat. Neurosci. 10, 1387-1394 (2007).

[0408] 19. Wolf, S. A., Boddeke, H. W. G. M. & Kettenmann, H. Microglia in Physiology and Disease. Annu. Rev. Physiol. 79, 619-643 (2017).

[0409] 20. Brooke, G., Holbrook, J. D., Brown, M. H. & Barclay, A. N. Human Lymphocytes Interact Directly with CD47 through a Novel Member of the Signal Regulatory Protein (SIRP) Family 1. J. Immunol. 173, 2562-2570 (2004).

[0410] 21. Barclay, A. N. & van den Berg, T. K. The Interaction Between Signal Regulatory Protein Alpha (SIRPa) and CD47: Structure, Function, and Therapeutic Target. Annu. Rev. Immunol. 32, 25-50 (2014).

[0411] 22. Willingham, S. B. et al. The CD47-signal regulatory protein alpha (SIRPa) interaction is a therapeutic target for human solid tumors. Proc. Natl. Acad. Sci. 109, 6662-6667 (2012).

[0412] 23. Hutter, G. et al. Microglia are effector cells of CD47-SIRPa antiphagocytic axis disruption against glioblastoma. Proc. Natl. Acad. Sci. U. S. A. 116, 997-1006 (2019).

[0413] 24. Zhang, M. et al. Anti-CD47 Treatment Stimulates Phagocytosis of Glioblastoma by Ml and M2 Polarized Macrophages and Promotes Ml Polarized Macrophages In Vivo. PLoS One 11, e0153550 (2016).

[0414] 25. Gholamin, S. et al. Disrupting the CD47-SIRP alpha anti-phagocytic axis by a humanized anti-CD47 antibody is an efficacious treatment for malignant pediatric brain tumors. Sci. Transl. Med. 9, (2017).

[0415] 26. Sikic, B. I. et al. First-in-Human, First-in-Class Phase I Trial of the Anti-CD47 Antibody Hu5F9-G4 in Patients With Advanced Cancers. J. Clin. Oncol. 37, 946-953 (2019).

[0416] 27. Advani, R. et al. CD47 Blockade by Hu5F9-G4 and Rituximab in Non-Hodgkin’s Lymphoma. N. Engl. J. Med. 379, 1711-1721 (2018).

[0417] 28. Dizman, N. & Buchbinder, E. I. Cancer Therapy Targeting CD47 / SIRPa. Cancers 13, 6229 (2021).

[0418] 29. Ring, A. M., Maute, R. L. & Curtis, A. (71) Applicant: Ab Initio Biotherapeutics, Inc., Los. 72.

[0419] 30. Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583-589 (2021).

[0420] 31. Tunyasuvunakool, K. et al. Highly accurate protein structure prediction for the human proteome. Nature 596, 590-596 (2021).

[0421] 32. Nakayashiki, N. et al. Production of a single-chain variable fragment antibody recognizing type III mutant epidermal growth factor receptor. Jpn. J. Cancer Res. Gann 91, 1035-1043 (2000).

[0422] 33. Johnson, L. A. et al. Rational development and characterization of humanized anti- EGFR variant III chimeric antigen receptor T cells for glioblastoma. Sci. TransL Med. 7, 275ra22 (2015).

[0423] 34. Nicholson, I. C. et al. Construction and characterisation of a functional CD19 specific single chain Fv fragment for immunotherapy of B lineage leukaemia and lymphoma. Mol. Immunol. 34, 1157-1165 (1997).

[0424] 35. Milone, M. C. et al. Chimeric Receptors Containing CD137 Signal Transduction Domains Mediate Enhanced Survival of T Cells and Increased Antileukemic Efficacy In Vivo. Mol. Ther. J. Am. Soc. Gene Ther. 17, 1453-1464 (2009).

[0425] 36. Weiskopf, K. et al. Engineered SIRPa Variants as Immunotherapeutic Adjuvants to Anticancer Antibodies. Science 341, 88-91 (2013).

[0426] 37. Bankhead, P. et al. QuPath: Open source software for digital pathology image analysis. Sci. Rep. 7, 16878 (2017).

[0427] 38. Schmidt, U., Weigert, M., Broaddus, C. & Myers, G. Cell Detection with Star-Convex Polygons, in Medical Image Computing and Computer Assisted Intervention - MICCAI 2018 (eds. Frangi, A. F., Schnabel, J. A., Davatzikos, C., Alberola-Lopez, C. & Fichtinger, G.) 265-273 (Springer International Publishing, 2018). doi: 10.1007 / 978-3-030-00934-2_30.

[0428] 39. Gut, G., Herrmann, M. D. & Pelkmans, L. Multiplexed protein maps link subcellular organization to cellular states. Science 361, (2018).

[0429] 40. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676-682 (2012).

[0430] 41. Preibisch, S., Saalfeld, S. & Tomancak, P. Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics 25, 1463-1465 (2009). 42. Thevenaz, P., Ruttimann, U. E. & Unser, M. A pyramid approach to subpixel registration based on intensity. IEEE Trans. Image Process. PubL IEEE Signal Process. Soc.

[0431] 7, 27-41 (1998).

[0432] 43. Guiet, R., Burri, 0., Chiaruttini, N., Seitz, A. & Eglinger, J. Kheops.

[0433] 44. Du, L. etal. IL-21 Optimizes the CAR-T Cell Preparation Through Improving Lentivirus Mediated Transfection Efficiency of T Cells and Enhancing CAR-T Cell Cytotoxic Activities. Front. Mol. Biosci. 8, 675179 (2021).

[0434] 45. Miao, H. el al. EGFRvIII-Specific Chimeric Antigen Receptor T Cells Migrate to and Kill Tumor Deposits Infiltrating the Brain Parenchyma in an Invasive Xenograft Model of Glioblastoma. PLOS ONE 9, e94281 (2014).

[0435] 46. Abbott, R. C. etal. Novel high-affinity EGFRvIII-specific chimeric antigen receptor T cells effectively eliminate human glioblastoma. Clin. TransL Immunol. 10, el283 (2021).

[0436] 47. Gholamin, S. et al. Disrupting the CD47-SIRPa anti-phagocytic axis by a humanized anti-CD47 antibody is an efficacious treatment for malignant pediatric brain tumors. Sci

[0437] Tr ansi Med 9, (2017).

[0438] 48. Fossum, E. et al. Targeting Antigens to Different Receptors on Conventional Type 1 Dendritic Cells Impacts the Immune Response. J. Immunol. Baltim. Md 1950205, 661-673 (2020).

[0439] 49. Porter, D. L., Levine, B. L., Kalos, M., Bagg, A. & June, C. H. Chimeric Antigen Receptor-Modified T Cells in Chronic Lymphoid Leukemia. N. Engl. J. Med. 365, 725-733 (2011).

Claims

Claims1. A polynucleotide molecule comprising a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigenbinding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPY)-related protein (SGRP).

2. The polynucleotide molecule of claim 1, wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a self-cleaving peptide.

3. The polynucleotide molecule of claim 2, wherein the polynucleotide molecule comprises in the following order from the 5' to the 3' end: a) a promoter; b) a nucleotide sequence encoding a chimeric antigen receptor (CAR) or a fragment thereof i) comprising an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) comprising an extracellular domain comprising an antigen binding region which binds to CD 19; wherein the nucleotide sequence encoding the chimeric antigen receptor (CAR) or a fragment thereof is operably linked to the promoter of a); c) a nucleotide sequence encoding a self-cleaving peptide; and d) a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPY)-related protein (SGRP).

4. The polynucleotide molecule of any one of claims 1-3, wherein the heterologous signal peptide fused to a signal regulatory protein gamma (SIRPY)-related protein (SGRP) is a signal peptide selected from the group consisting of interleukin 2 (IL-2) signal peptide, interleukin 4 (IL-4) signal peptide, interleukin 9 (IL-9) signal peptide and interferon gamma (IFNy) signal peptide, preferably selected from the group consisting of humanIL-2 signal peptide, human IL-4 signal peptide, human IL-9 signal peptide and human IFNy signal peptide.

5. The polynucleotide molecule of any one of claims 1-3, wherein the heterologous signal peptide fused to a signal regulatory protein gamma (SIRPY)-related protein (SGRP) is a human interleukin 2 (IL-2) signal peptide.

6. The polynucleotide molecule of any one of claims 1-5, wherein the heterologous signal peptide is fused to the N-terminal region of the signal regulatory protein gamma (SIRPY)-related protein (SGRP).

7. The polynucleotide molecule of any one of claims 1-6, wherein the promoter which is operably linked to the nucleotide sequence encoding the chimeric antigen receptor (CAR) or a fragment thereof is the elongation factor 1 alpha (EFl A) promoter or the elongation factor 1 alpha short (EFS) promoter, preferably the EFl A promoter.

8. The polynucleotide molecule of any one of claims 1-7, wherein the chimeric antigen receptor (CAR) or a fragment thereof comprises i) an extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); or ii) an extracellular domain comprising an antigen binding region which binds to CD 19; wherein the CAR or a fragment thereof further comprises a CD8a leader, CD8a hinge and transmembrane domains, a TNF receptor superfamily member 9 (4-1BB) costimulatory domain and a CD3(^ signaling domain.

9. The polynucleotide molecule of claim 8, wherein i) the extracellular domain comprising an antigen binding region which binds to epidermal growth factor receptor variant III (EGFRvIII) is a single-chain variable fragment (scFv); or ii) the extracellular domain comprising an antigen binding region which binds to CD 19 is a single-chain variable fragment (scFv).

10. The polynucleotide molecule of any one of claims 1-9, comprising the sequence as shown in SEQ ID NO: 1 or the sequence as shown in SEQ ID NO: 2.

11. An amino sequence comprising a chimeric antigen receptor (CAR) or a fragment thereof comprising an extracellular domain comprising an antigen-binding region which binds to epidermal growth factor receptor variant III (EGFRvIII); wherein the nucleotide sequence encoding the chimeric antigen receptor (CAR) is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPyj-related protein (SGRP) comprising the sequence as shown in SEQ ID NO: 3.

12. An amino sequence comprising a chimeric antigen receptor (CAR) or a fragment thereof comprising an extracellular domain comprising an antigen-binding region which binds to CD 19; wherein the nucleotide sequence encoding the CAR is operably linked to a promoter and wherein the polynucleotide molecule further comprises a nucleotide sequence encoding a heterologous signal peptide fused to a signal regulatory protein gamma (SIRPy)-related protein (SGRP) comprising the sequence as shown in SEQ ID NO: 4.

13. A construct comprising the polynucleotide molecule of any one of claims 1-10.

14. A chimeric antigen receptor (CAR)-T cell comprising T cells expressing the polynucleotide molecule of any one of claims 1-10 and / or the construct of claim 13.

15. The chimeric antigen receptor (CAR)-T cell of claim 14, for use in a method for the treatment of cancer in a subject suffering from an epidermal growth factor receptor (EGFR)-associated cancer or suffering from a CD19-associated cancer, the method comprising administering to the subject a therapeutically effective amount of a chimeric antigen receptor (CAR)-T cell of claim 14.

Images