You are viewing the site in preview mode

Skip to main content
Download PDF

Efficacy of MET-targeting CAR T cells against glioblastoma patient-derived xenograft models

Journal of Translational Medicine volume 23, Article number: 460 (2025) Cite this article

Abstract

Background

Genetic alteration of the MET receptor tyrosine kinase frequently occurs in glioblastoma (GBM). Clinically, bevacizumab treatment results in MET signaling activation, leading to GBM recurrence with a more malignant phenotype. While MET has been a promising therapeutic target, MET inhibitors have not been successful in treating GBM patients. MET-directed chimeric antigen receptor (CAR) T cells hold the promise of targeting MET-positive GBM regardless of genetic alterations or kinase activity.

Methods

GBM patient-derived xenografts (PDX) harboring MET amplification (METamp) or PTPRZ-MET fusion (ZM) were propagated in vivo followed by glioma stem cell (GSC) isolation. Cell-based assays were used for comparing GSC survival in response to MET inhibitors and CAR T cells. Multi-panel cytokine release was analyzed to profile MET-CAR T cell activation during co-culture with GBM. Orthotopic tumor growth and real-time imaging were performed to evaluate MET-CAR T cell therapeutic efficacy in vivo.

Results

Although GBM are heterogeneous tumors, neuro-sphere cells isolated from METamp or ZM fusion PDX tumors showed universal cognate genetic MET alteration along with GSC markers such as SOX2 and nestin. Both METamp and ZM fusion tumors showed MET overexpression but only the METamp cells presented activated MET signaling which was vulnerable to MET inhibitors. In contrast, MET-CAR T cells specifically inhibited all MET-positive tumor growth regardless of MET activation status.

Conclusions

Whereas MET inhibitors are effective in MET-active tumors, MET-CAR T cells eradicate MET-positive GBM growth in an antigen-dependent manner, demonstrating a promising therapeutic approach for treating MET-positive GBM. MET overexpression, especially METamp and ZM fusion may be used to predefine the GBM patients for treating with MET-CAR T cell therapy.

Introduction

Glioblastoma (GBM) is the most common and malignant primary brain tumor without effective treatment [1]. Glioma stem cells (GSCs), a subset of poorly differentiated yet highly self-renewing glioma cells, have invasive and tumorigenic properties that are responsible for irradiation- and chemo-resistance [2,3,4,5]. As such, developing methods for targeting GSCs is the key to successful GBM treatment. Genetic alteration of receptor tyrosine kinase (RTK) family proteins frequently occurs in GBM patients, with MET as one of the most affected RTKs [6]. Clinically, MET amplification (METamp) and MET exon 7–8 deletion (METΔ7–8) are found in primary GBM [6, 7], while fusion of protein tyrosine phosphatase receptor type Z1 (PTPRZ1) and MET (ZM fusion) and MET exon 14 deletion (METΔ14) are more common in secondary GBM [8]. Overall, MET overexpression occurs in 30–45% of GBM patients and is responsible for short survival and therapeutic resistance [9,10,11]. While bevacizumab is an FDA-approved GBM treatment inhibiting vascular epithelial growth factor receptor 2 (VEGFR2), about 36.8% of GBM patients resistant to bevacizumab treatment had MET overexpression which is associated with a significantly shorter time-to-progression as compared to those with low or no MET expression [12, 13]. Among all patients, concurrent MET/VEGFR2 overexpression (18.7%) demonstrated the worst overall survival time at about 13 months [13]. Using transgenic mouse models, we showed that overexpressing human MET and its ligand hepatocyte growth factor (HGF) along with Tpr53 suppression was sufficient to initiate GBM development [14]. Others also reported that irradiation-induced DNA damage accompanied with loss of tumor suppressor genes triggers glioma formation driven by Met amplification as the most significant oncogenic event [15]. In both models, activated MET signaling resulted in prevalent GSC markers such as nestin and sox2, supporting that the HGF/MET axis transforms neuro stem cells (NSCs) into GSCs, leading to GBM development [14, 15]. In human GBM cell line models, MET pathway activation induces GSC reprogramming and maintains GBM stemness through the SOX2/OCT4 signaling network [16, 17]. Although MET has been a promising target for GBM treatment, clinical trials indicate that effective MET inhibitors require constitutive MET signaling activation for tumor growth, making patient stratification difficult [18]. MET inhibitors commonly induce resistance, largely due to crosstalk with other RTK members [19]. Thus, it is critical to develop alternative strategies for targeting the MET without leading to GBM therapeutic resistance and recurrence.

Recent progress in cancer immuno-therapy encourages the expansion of chimeric antigen receptor (CAR) T cell therapy in solid tumors. In GBM, several targets, such as IL13Rα and GD2, have been developed into clinical trials [20, 21]. Within the RTK family, epidermal growth factor receptor (EGFR) and its mutants including human epidermal growth factor receptor 2 (HER2) and EGFR variant III (EGFRvIII) [22, 23] are also being developed into major targets based on the prevalence of their overexpression in GBM. However, MET as a CAR T cell target for clinical development has not been evaluated. Of note, MET-overexpression, especially ZM fusion and METamp [24, 25], can be used to predefine the subsets of patients suitable for MET-CAR T cell therapy. Strategically, MET-CAR T cells function through antigen-dependent cytotoxic T cell activities rather than RTK activity, thus the treatment may overcome MET inhibitor-mediated tumor resistance and will be independent of previous treatments.

MetMab (onartuzumab) is a humanized monovalent anti-MET antibody that potently inhibited ligand-dependent MET activation and tumor growth in preclinical tumor models [26]. However, MetMab failed to improve therapeutic efficacy when combined with bevacizumab in a GBM phase II clinical trial, although a subgroup of patients with high expression of HGF may have benefited [27]. Based on MetMab structure, we have established novel MET-CAR T cells that specifically and potently inhibited MET-positive tumor growth in preclinical hepatocellular carcinoma mouse models [28]. Promising efficacy and safety of MET-CAR T cell therapy have been reported from phase 1 clinical trials for melanoma and breast carcinoma patients [29]. In this study, we characterized GBM- patient-derived xenograft (PDX) models bearing METamp and ZM fusion and evaluated their therapeutic response to MET- tyrosine kinase inhibitors (TKIs) and specific MET-CAR T cells.

Materials and methods

Cell lines, PDX models, and compounds

HEK293 cells and U87 cells were originally purchased from American Type Culture Collection (ATCC) and were grown in Dulbecco's Modified Eagle's Medium (DMEM) (Gibco) supplemented with 10% fetal bovine serum (FBS) and penicillin and streptomycin (Invitrogen). G159, G215 and G91 PDX models were obtained from the Mayo Clinic PDX National Source. V-4084 is a MET inhibitor provided by Vertex Pharmaceutics [30], cabozantinib and erlotinib were purchased through LC Laboratories. PD-901 was purchased from Selleck Chemicals. All compounds were dissolved in DMSO at 0.01 M and aliquots were stored at − 80 °C until use. Human whole blood from healthy donors was purchased from Physician Plasma Alliance, Grey, TN. Upon collection, peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll (GE Healthcare Bio-Science AB) gradient centrifugation, suspended in plasma with 10% DMSO and kept in liquid nitrogen until use.

Neuro-sphere generation and confocal analysis

Subcutaneous GBM-PDX tumors were excised and cells were isolated after accutase (Invitrogen) dissociation for neurospheres to grow in DMEM/F12 serum-free medium supplemented with B27 (Invitrogen), epidermal growth factor (EGF) (20 ng/mL, R&D System), basic fibroblast growth factor (bFGF) (20 ng/mL, R&D), and 1% penicillin and streptomycin (Invitrogen). For confocal imaging, G159 and G215 cells were seeded at 5000 cells/well in Lab Tek chamber slides (Thermo Fisher Scientific). After neurosphere formation, cells were washed with PBS, fixed with 4% paraformaldehyde, permeabilized in 0.5% Triton X-100 for 20 min, and blocked in 5% BSA for 1 h at room temperature. To perform immunofluorescence (IF) staining, cells were incubated with anti-glial fibrillary acidic protein (GFAP) antibody (Novus), anti-Sox2 antibody (Cell Signaling Technology) or anti-nestin antibody (Sigma-Aldrich) overnight at 4 °C. After washing with PBS, cells were incubated with goat anti-rabbit/Alexa568 antibody (1:500 in PBST, Invitrogen) for 50 min. The nuclei were stained with DAPI (1 µg/ml, Thermo Fisher Scientific). Cells treated with secondary antibody only and stained with DAPI were used to determine the background. Z-stack images of sphere cells were captured using a confocal microscope (Leica TCS SP8).

Fluorescent in situ hybridization (FISH) Analysis

The set of ZM fusion probes (7q31.32/7q31.2) was purchased from Empire Genomics (Catalog #PTPRZ1-MET-20-ORGR). The DNA probe set contains two probes which specifically hybridize to PTPRZ1 at chromosome 7q31.32 (red) and MET at 7q31.2 (green) and were used for both metaphase spreads and interphase nuclei. The Chr7 centromere probe was purchased from Abbott Molecular (Catalog #32–112007). The automated hybridization protocol was provided by Empire Genomics. FISH images were analyzed using Leica CytoVision system (Version 7.7, Leica IL, USA). For each cell line, at least 100 cells at interphase were analyzed.

Immunohistochemistry staining of GBM orthotopic tumors

At the time of necropsy, mouse brains with tumors were dissected, fixed with 10% neutral buffered formalin (Sigma-Aldrich), and embedded into paraffin blocks for H&E and IHC staining using standard clinical techniques as described previously [14]. For IHC staining, primary antibodies used are rabbit anti-human Met (DIC2) at 1:100 dilution (Cell Signaling), rabbit anti-nestin (1:100, Sigma-Aldrich) and rabbit anti-Sox2 (1:100, Novus). An ABC kit with biotinylated anti-rabbit IgG was used for secondary antibody detection (Vector Laboratories, Newark, CA). The Super Sensitiveâ„¢ one-step polymer-HRP IHC Detection System (BioGenex Laboratory, Fremont, CA) was used for diaminobenzidine (DAB) staining. Images were captured by a Nikon light microscope.

RT-PCR primers and sequencing analysis

Cells were seeded in 10-cm dishes for neurosphere growth. Total RNA was extracted from cells using TRIzol reagent (Invitrogen). Reverse transcription (RT) was done with the SuperScript IV reverse transcription kit (Invitrogen) and the cDNA was used for PCR analysis with the DreamTaq Green PCR Master Mix (2x) (ThermoFisher Scientific). Primers used for RT-PCR reactions are ZM fusion forward: 5′-CCGTCTGGAAATGCGAATCCTAAA-3′, reverse: 5′-CAGGCCCAGT CTTGTACTCAGCAA-3′ [8]; MET forward: 5ʹ-ACAGTGGCATGTCAACATCGCT, reverse: 5ʹ-GCTCGGTAGTCTACAGATTC-3ʹ; GAPDH forward: 5ʹ-GCACCACCAACTGCTTAGCA-3ʹ, reverse: 5ʹ-GTCTTCTGGGTGGCAGTGATG-3ʹ. For PCR, denaturation, annealing, and extension were done at 94 °C, 55 °C, and 72 °C, respectively, for 1 min each, for a total of 25 cycles followed by an extension period at 72 °C for 5 min. The PCR product was run on a 1.0% agarose gel. The ZM fusion PCR product was purified using a gel extraction kit (Invitrogen) for Sanger sequencing at Azenta Life Science.

Protein structure analysis

The 793 residue amino acid sequence of the extracellular domain (ECD) of the ZM fusion protein containing the PTPRZ1 carbonic anhydrase-like (CAH) domain (exon 1–8) through the MET SEMA domain (see Supplemental Fig. 1 C) was analyzed by the artificial intelligence (AI) protein structure prediction program AlphaFold-2 using the default settings [31, 32]. The highest ranked output structure models were analyzed in the BIOVIA Discovery Studio visualization software in which they were compared to the PDB structure files of the CAH domain of PTPRZ1 (PDB: 3JXF) and the SEMA domain of MET (PDB: 4 K3J & 7MO7) [26, 33,34,35].

Generation of MET-specific CAR T cells

The MET-CAR retroviral particles were produced via transient transfection of HEK293 cells as previously described [28]. To produce MET-CAR T cells, PBMCs isolated from healthy donors were stimulated with anti-CD3/CD28 antibodies (Miltenyl Biotec) in the presence of interleukin (IL)−7 (10 ng/ml, Miltenyl Biotec) and IL-15 (5 ng/ml, Miltenyl Biotec) to expand CD3+ T cells for transduction on plates coated with RetroNectin (Takara Bio). All MET-CAR constructs use truncated CD19 to serve as the surrogate marker for CAR expression. MET-CAR transduction efficacy was measured by flow cytometry (Acuri C6 Plus) using CD3-FITC and CD19-PE antibodies. Isotype control antibodies used are IgG1-FITC and IgG1-PE (BD Bioscience).

Cell survival analysis

To measure TKI-mediated GBM growth inhibition in vitro, G215 and G159 cells were seeded into a 96-well plate at 5 × 103 cells/well and grown overnight at 37 °C followed by treatment with various TKIs at the indicated concentrations. Triplicate wells were used for each concentration. After an additional 72 h, MTS reagent was added into each well and incubation continued for an additional 4 h at 37 °C following the manufacturer’s instructions (Promega). To measure MET-CAR T cell-mediated cytotoxicity, HEK293 and U87 cells were seeded in 96-well plates at 2 × 104 cells/well and grown at 37 °C for 24 h. For G91 and G251 sphere cells, poly-d lysine-coated plates were used for sphere cells to attach. On day 2, non-transformed (NT) or MET-CAR T cells derived from the same healthy donor were added into each well at various effector T cell:tumor cell (E:T) ratios for co-culture. Triplicates were used for each E:T ratio. After an additional 24 h, effector T cells were washed out using PBS. The viability of tumor cells was determined using the MTS assay (Promega). Because G159 sphere cells did not attach to the wells, G159 cells were transfected to overexpress GFP-Luciferase as a reporter gene to create G159GFP−Luc cells for MET-CAR T cell treatment. The viability of the G159GFP−Luc tumor cells was determined using a luciferase assay (Promega) without washing out the effector T cells.

Multiplex cytokine assay

To prepare conditioned medium, GBM cells were seeded in 24-well plates at 1 × 105 cells/well and grown overnight followed by co-culture with NT and MET-CAR T cells generated using PBMCs from the same healthy subjects (n = 4) at E:T = 2:1 ratio. After an additional 24 h, the medium was collected for analysis using the Human Procarta Plex 21-plex panel (Thermo Fisher Scientific) on a Bio-Plex MAGPIX system (Bio-Rad Laboratories, Hercules, CA) following the manufacturer’s instructions. The plate was read using a Bio-Plex MAGPIX instrument for the concentration of each cytokine produced by each type of MET-CAR T cell.

Signaling pathway analysis and western blot

To test MET signaling activation, cells were cultured in 10-cm dishes until 80% confluence followed by cell lysis for western blot analysis. We used antibodies against human Met (clone 25H2), phospho-Met (Y1234/1235), AKT, phospho-AKT (S473), p42/44 MAPK, phospho-p42/44 MAPK (T202/Y204) and GAPDH (Cell Signaling Technology). Secondary antibodies used were goat anti-rabbit IgG-HRP and goat anti-mouse IgG-HRP (Santa Cruz Biotechnology, Dallas, TX).

Therapeutic efficacy of MET-CAR T cells against GBM orthotopic tumor growth

B-NDG mice (Envigo) at 6–8 weeks of age were used for this study. GBM cells (5 × 105) were orthotopically injected into mouse brains to initiate intracranial (i.c.) tumor growth as previously described [30]. Seven days after surgery, mice bearing tumors were randomized into 2 groups (n = 10) based upon bioluminescence imaging (BLI) signal intensity to receive a one-time treatment of MET-CAR T cells (5 × 106 cells, i.c.) at the same coordinates. Intracranial tumor growth was measured by BLI once a week. All studies involving animals were approved by East Tennessee State University Institutional Animal Care and Use Committees.

Statistical analysis

GraphPad Prism 8 software (GraphPad Software, La Jolla, CA) was used for statistical analysis. In vitro GBM cell survival data were analyzed with the student’s t test (p < 0.05). To evaluate the effectiveness of in vivo MET-CAR T cell treatment, the average BLI signal intensity at each time point was analyzed with the student’s t test (p < 0.05). Survival time was analyzed with the Kaplan–Meier test (p < 0.05).

Results

Characterization of GBM-PDX models with MET alterations

Among the various MET alterations, METamp is found in about 4% of primary GBM and ZM fusion is found in about 15% of secondary GBM [8, 24, 28]. Both alterations lead to MET overexpression in tumors which associates to the responsiveness to MET inhibitors. Based on the Mayo Clinic datasets [36], we identified GBM-PDX G159 and G215 models to harbor METamp or ZM fusion, respectively, and regenerated orthotopic tumor growth for further analysis. Histologically, G159 tumor cells infiltrate surrounding normal brain tissue without a clear tumor margin (Fig. 1A a, c, e). Necrotic centers and mitotic cells also were observed as GBM hallmarks (Fig. 1A b, d). G215 orthotopic tumors also showed tumor cell invasion of adjacent normal tissue (Fig. 1B a, c), but is less extensive compared to G159. While G215 mitotic cells were identified (Fig. 1B b), areas of necrosis were not observed. With IHC staining, both G159 and G215 tumors show a high level of MET overexpression; however, only G159 tumors had strong phosphor-MET expression (Fig. 1A c–e vs. Figure 1B c, d), demonstrating an active MET signaling pathway. Glioma stemness is known to be the key to radio- and chemo-resistance, which leads to tumor recurrence as a more malignant phenotype. Both G159 and G215 tumors expressed GSC markers such as SOX2 and nestin (Fig. 1A f, g and B e, f), demonstrating that they have GSC properties.

Fig. 1
figure 1

Characterization of GBM-PDX orthotopic tumor growth. A Representative H&E images of G159 orthotopic tumor growth including invasive tumor growth (a), central necrosis (b), and IHC staining for MET (c-d), p-MET (e), SOX2 (f) and nestin (g) expression levels. Red arrows (d) indicate mitotic cells. B Pathological analysis of G215 orthotopic tumor growth using H&E (a,b) and IHC staining (c-f). Note that G215 tumors demonstrate a less invasive phenotype as compared with G159 tumors (A a, c). While tumors were positive for MET (c), SOX2 (e) and nestin expression (f), p-MET expression was negative (d)

Full size image

GBM are heterogeneous tumors driven by multiple key players. To determine whether METamp and ZM fusion directly attribute to glioma stemness and GSC formation, low-passage GBM neurosphere culture (< 5 passages) with G159 and G215 tumors was performed for FISH analysis to count for copy numbers of MET and PTPRZ1 genes in individual cells. With G159, at least 200 cells were counted with 100% of the cells showing METamp ranging from 40–120 copies (Fig. 2A). With G215, 100 cells were counted with 100% of the cells showing ZM fusion at chr7, ranging from 2–6 copies. Following the confocal imaging analysis, all G159 and G215 sphere cells expressed GSC markers GFAP, nestin, and SOX2 (Fig. 2B). These results suggest that both METamp and ZM fusion play an essential role in glioma stemness and GSC initiation in G159 and G215 PDX tumors. Taken together, both G159 and G215 recapitulate human GBM genetics and histology and, therefore, are good models for studying GBM resistance mechanisms as well as for testing therapeutic reagents.

Fig. 2
figure 2

GSCs isolation and characterization. A METamp and ZM fusion identified by FISH analysis. B Expression of GFAP, SOX2 and nestin in G159 and G215 neurosphere cells detected by confocal imaging. C Structural analysis of ZM fusion protein. Comparison of the ZM fusion protein structure predicted by AlphaFold2 to the SEMA domain structure of MET (PDB: 4 K3J) and the CAH domain of PTPRZ1 (PDB:3JXF). The PTPRZ1 portion of the ZM fusion protein is purple. The seven β-folds of the MET-SEMA domains are numbered 1–7. Note that yellow and red refer to the binding site to MetMab (β-folds 5–6) and HGF-β (β-folds 2, 3 and part of 4) chains, respectively. For PTPRZ1-CAH, blue and red refers to the β-folds and α-helix structures, respectively

Full size image

Previous study has identified multiple variants of ZM fusions composed of PTPRZ1 exon 1 through exon 2, 3 or 8, all of which are fused to the beginning of MET exon 2 [8]. To characterize the ZM fusion identified in G215 cells, we first performed the RT-PCR using the primers encompassing the 5’ start codon of PTPRZ1 to 3’ end of MET exon 2 which yielded a DNA fragment 1217 bp in size. Sanger sequencing further confirmed the product as a fused DNA of exon 1–8 of PTPRZ to exon 2 of MET (Supplemental Fig. 1A, B). We then performed AlphaFold 2 protein structural analysis [31, 32] of a 793 amino acid portion of the ZM fusion which included the CAH domain of PTPRZ1 (residues 37–310) fused to the MET-SEMA domain (residues 311–830) (see Supplemental Fig. 1 C). The single PSI (Plexins, Semaphorins, Integrins) domain and the four IPT (Ig-like, plexin, transcription factor) domains of MET’s ECD (residues 831–1264) were not included. Interestingly, the ZM fusion protein produced a loop between the CAH domain of PTPRZ1 and the SEMA domain of MET without changing each structure significantly (Fig. 2C). Because the MET-SEMA domain is the docking site of HGF binding, it is expected that the ZM fusion protein should remain accessible for HGF binding and stimulation (Supplemental Fig. 1D). This result also suggests the different MET activation status in G159 and G215 tumors. While METamp indicates a constitutive MET activation independent of ligand stimulation (Fig. 1A e), overexpression of ZM fusion by itself will not lead to MET activation (Fig. 1B d) but retains the capacity for a ligand-dependent activation when binding to HGF.

Differential response of METamp and ZM fusion to MET inhibitors

Our previous studies suggested that MET inhibitors require active MET signaling in tumors for growth suppression [9, 37]. To test, G159 and G215 cells were treated with MET inhibitors cabozantinib and V4084, MET downstream MEK inhibitor PD-901, or the EGFR inhibitor erlotinib as a control. Both cabozantinib and V-4084 significantly inhibited G159 cell proliferation at 0.1–10 μM (Fig. 3A), while PD-901 and erlotinib only inhibited at the concentration at 10 μM or above. When tested for MET signaling pathway, G159 cells demonstrated MET overexpression with constitutive MET signaling activation (p-MET/p-MAPK/p-Akt) which was significantly blocked by cabozantinib and V-4084 (Fig. 3B). In contrast, G215 cells also had MET overexpression (Fig. 1B c) but showed no response to any of these four inhibitors (Fig. 3A). However, HGF stimulated MET signaling activation in these cells which was inhibited by cabozantinib and V-4084 (Fig. 3B). Thus, it is MET activation but not overexpression of MET that indicates therapeutic response to MET inhibitors.

Fig. 3
figure 3

TKI-mediated inhibition of cell survival and signaling pathway in vitro. A Inhibition of G159 and G215 cell survival by cabozantinib, V-4084, PD-901, and erlotinib using the MTS assay. Survival (%) of each treatment = (Signal intensity of treated sample at OD490)/(signal intensity of untreated sample) × 100%. The short vertical bars indicate the standard deviation of the average of 2 independent assays with triplicates for each concentration. Compared with untreated sample (Ctl) and vehicle (DMSO) treatment. * p < 0.05. B Cabozantinib- and V-4084-mediated inhibition of MET downstream signaling pathway in G159 and G215 using Western blot analysis. Note that MET was only activated when G215 cells were grown in the presence of HGF

Full size image

MET-CAR T cells specifically kill MET-positive GBM cells in vitro

MetMab (onartuzumab) is an anti-MET monovalent antibody inhibiting tumor growth driven by HGF-dependent MET activation but not MET overexpression [26, 37]. Based on MetMab structure, we have generated 2 positive MET-CARs (MET-CAR.CD28.ζ and MET-CAR.4-1BB.ζ) using a MET-CAR without the TCR co-stimulatory domain (MET-CAR.Δ) for negative control (Fig. 4A). We also demonstrated that MET-CAR T cells specifically kill hepatocellular carcinoma (HCC) cells with MET overexpression regardless of downstream RTK signaling pathway activation [28]. To determine if the same principle applies to GBM, we tested NT and the 3 types of MET-CAR T cells for the killing activity against G159 and G215 cells using HEK293 and G91 cells for MET-negative controls; U87 cells were also included for the feature of HGF-autocrine mediated MET activation [37] (Fig. 4B). MET-CAR T cells were generated as previously described with overall T cell transduction efficacy ranging from 60–90% [28]. We show that both MET-CAR.CD28ζ and MET-CAR.4-1BBζ T cells potently killed U87, G159, and G215 cells in a dose-dependent manner (Fig. 4C, D) but had no effect on HEK293 T or G91 cells which continued to proliferate. This demonstrates a specific killing of MET-positive GBM cells, independent of genetic MET alteration or pathway activation. Of note, MetMab binds to MET-SEMA domain at β-barrel 5–6 [26], preventing its binding to HGF-α chain (Supplemental Fig. 1D). As the ZM fusion does not interrupt the SEMA structure (Fig. 2C), the MetMab binding domain remains intact, which also supports the concept of targeting ZM fusion with MET-CAR T cell therapy.

Fig. 4
figure 4

MET-CAR T cell killing activity against GBM cells in vitro. A MET-CAR structure was described previously [28]. In brief, the scFv domain of MetMab is linked with a TCR transmembrane (TM) domain (CD28 TM or CD8α TM). MET-CAR.CD28ζ and MET-CAR.4-1BBζ vectors use a CD3ζ domain (CD28ζ or 4-1BBζ, respectively) as the TCR signaling module. MET-CAR.Δ vector is a negative control that does not have the CD3ζ domain. All vectors have a 2A sequence linked to CD19 as an expression marker for measuring T cell transduction efficacy using flow cytometry, B MET expression in HEK293 and GBM cells measured by flow cytometry. C MET-CAR T cell killing activity against HEK293, G91, G215 and U87 cells with the MTS assay in vitro. Survival (%) of each treatment = (Signal intensity at OD490 of treated sample)/(signal intensity of tumor cell only) × 100%. The short vertical bars indicate the standard deviation of the average of MET-CAR T cells derived from 3 independent healthy donors with triplicates for each concentration. D MET-CAR T cell killing activity against G159 cells with the luciferase assay in vitro. Survival (%) of each treatment = (luminescent signal intensity of treated sample)/(signal intensity of tumor cell only) × 100%. The short vertical bars indicate the standard deviation of the average of MET-CAR T cells derived from 3 independent healthy donors with triplicates for each concentration

Full size image

Multi-panel analysis of cytokine release by MET-CAR T cells in response to GBM stimulation

We have reported that MET-positive HCC tumor cells stimulated a panel of cytokine release in both MET-CAR.CD28ζ and MET-CAR.41BB.ζ T cells [28]. To determine the cytokine release profile of MET-CAR T cells in response to GBM stimulation, we co-cultured NT and 3 types of MET-CAR T cells with G91, G159, G215 and U87 cells for 24 h at an E:T ratio = 2:1 and assayed for the production of 21 cytokines (Fig. 5). IL-2 and IFNγ are the most commonly used cytokines indicating CAR T cell activity; both were significantly up-regulated in MET-CD28ζ and MET-CAR.41BBζ T cells after co-culture with the 3 MET-positive GBM cells (G159, G215 and U87). MET-negative G91 cells failed to up-regulate any of the cytokines in any type of the MET-CAR T cells, which demonstrate a specific phenotype of targeting MET-positive GBM. Other cytokines up-regulated in activated MET-CD28ζ and MET-CAR.41BBζ T cells also include GM-CSF, Granzyme A, MIP-1a, MIP-1b, MCP-1, TNFα and TNFβ. Notably, IL-12 [38], IL-15 [39], IL-18 [40] and IL-21[41, 42] are reported to improve CAR T cell proliferation, persistence and anti-tumor activity, but were not changed in any of the MET-CAR T cells after co-culture with G159, G215 or U87 cells, suggesting that administration of these cytokines in combination with MET-CAR T cells may improve the efficacy as a future strategy.

Fig. 5
figure 5

Multi-panel cytokine release profiling of MET-CAR T cells upon GBM stimulation. A Heatmap of cytokine release by MET-CAR T cells in response to GBM cell stimulation. For each cell line model, MET-CAR T cells generated from at least 3 independent healthy donors were tested for cytokine release in response to GBM cell stimulation. A total of 21 cytokines were analyzed. Python data visualization libraries Matplotlib 3.5.3 and Seaborn 0.12.1 were used to generate the heat-map of cytokine expression levels using the log2-fold change value, which is measured by comparing each cytokine expression level of MET-CAR T cells to that of the paired NT cells. B Box plot analysis of the most significantly upregulated cytokines after MET-positive GBM stimulation. Out of 21 cytokines 12 showed consistent upregulation in MET-CAR.CD28ζ and MET-CAR.41BBζ T cells across the 3 MET-positive GBM cell lines. Fold change compared with MET-CAR∆ T cells *p < 0.001; **p < 1e−5

Full size image

MET-CAR T cells inhibit MET-positive GBM orthotopic tumor growth

HGF-autocrine activation, METamp, and ZM fusion all count for MET overexpression in GBM. To test whether MET-CAR T cells inhibit GBM tumor growth in an antigen-dependent manner we evaluated the therapeutic efficacy of MET-CAR.CD28.ζ T cells against U87, G159, and G215 orthotopic tumor growth in mice. With U87, a single dose of i.c.-delivered MET-CAR.CD28ζ T cells at 10 days after tumor inoculation significantly inhibited U87 orthotopic tumor growth and significantly prolonged survival time as compared with MET-CAR.∆ T cells (Fig. 6A, C, p < 0.05). Similar tumor inhibition also was found with the G159 orthotopic model: after a single dose treatment at day 8, 5 out of 7 mice in the MET-CAR.∆ group were euthanized due to significant tumor growth while all 7 mice in MET-CAR.CD28ζ group survived until the termination of the study at day 85 (Fig. 6B, C, p < 0.05). Since MET-CAR.CD28ζ T cells potently inhibited U87 and G159 tumor growth (Fig. 4), we expected MET-CAR T cells to inhibit G215 tumor growth at a later time point of injection. However, the efficacy turned out to be much lower. One-time i.c. delivery of MET-CAR.CD28ζ T cells at day 13 significantly inhibited G215 tumor growth at days 20 and 24 but failed to show a significant tumor inhibition thereafter (Fig. 7A, B). Most animals died of tumor growth by day 27 after tumor initiation. Overall, there was no survival benefit in MET-CAR.CD28ζ group (Fig. 7A). Noting that G215 tumors grew faster than U87 tumors in the control settings (Figs. 6 vs 7), we repeated the G215 experiment with earlier MET-CAR T cell delivery at 9 days after tumor inoculation (Fig. 7C). Results from the 2nd cohort again showed significant tumor growth inhibition at day 17 (Fig. 7D, MET-CAR.CD28ζ vs. MET-CAR.∆, p < 0.05) although several tumors in the MET-CAR.CD28ζ group grew back after that. After combining the data from the two cohorts, results support that MET-CAR.CD28ζ T cell treatment benefit the survival time of G215 tumor bearing mice as compared to MET-CAR.∆ T cells (Fig. 7E).

Fig. 6
figure 6

Inhibition of U87 and G159 orthotopic tumor growth by MET-CAR T cells. A U87 and B G159 orthotopic tumor growth measured by BLI. The arrow indicates the days post tumor initiation when CAR T cells were injected. Note that one U87 MET-CAR.CD28ζ mouse died after day 23 due to anesthesia at the imaging and was excluded from the statistics. C Survival time of U87 and G159 orthotopic mice as shown in A and B

Full size image
Fig. 7
figure 7

Inhibition of G215 orthotopic tumor growth by MET-CAR T cells. A 1 st cohort of G215 orthotopic tumor growth measured by BLI. Arrow indicates the days post tumor initiation when CAR T cells were injected. B The BLI signal intensity of individual mice at each time point until day 27 from A was quantified using Bruker MI SE software to plot a tumor growth curve for each mouse. Darker line indicates the average signal intensity of each group. Vertical bars with brackets refer to standard deviation. Note that average BLI signal intensity at days 20 and 24 in MET-CAR.CD28ζ group was significantly lower than that in MET-CAR∆ group (student t test, one tail, * p < 0.05). C 2nd cohort of G215 orthotopic tumor growth measured by BLI. Arrow indicates the days post tumor initiation when CAR T cells were injected. D The BLI signal intensity of individual mice at each time point until day 24 from C was quantified using Bruker MI SE software to plot a tumor growth curve for each mouse. Darker line indicates the average signal intensity of each group. Vertical bars with brackets refer to standard deviation. Note that average BLI signal intensity at day 17 in MET-CAR.CD28ζ group was significantly lower than that in MET-CAR∆ group (student t test, one tail, * p < 0.05). E Survival time of G215 orthotopic mice after combining the results from the two cohorts

Full size image

Discussion

Based on TCGA datasets, approximately 4% of primary GBM patients harbor METamp [6]. Amplification of multiple RTKs within the same tumor also is common, which complicates treatment options [43]. Despite that GBM are highly heterogeneous, all neuro-sphere cells isolated from G159 PDX showed METamp along with GSC marker expression, indicating METamp is a driving force of GSC stemness and a potential therapeutic target. We previously reported that gastric and liver cancer models harboring METamp are highly responsive to MET TKIs [37]. Here, we show that cabozantinib and V4084 significantly inhibited G159 proliferation and MET downstream signaling activation independent of HGF stimulation. These results demonstrate that METamp can be a biomarker indicating GBM sensitivity to MET TKIs.

Low-grade gliomas often progress into secondary GBM (sGBM) with limited therapeutic options. Experimentally, ZM fusion overexpression in U87 cells promoted MET pathway activation and tumor growth which can be significantly inhibited by MET inhibitor PLB-1001 [8, 24]. However, we observed that MET signaling was not active in the G215 ZM fusion model (Fig. 1B). Consequently, MET inhibitors did not decrease G215 cell survival (Fig. 3A). Notably, U87 cells express both HGF and MET forming an HGF autocrine activation loop in the tumor which is sensitive to MET TKIs or neutralizing antibodies [9, 37, 44]. Thus, the endogenous ZM fusion by itself may not indicate a sensitivity to TKIs unless the MET signaling is activated by the ligand or environmental stimulation.

From past studies we have learned that effective TKIs require MET to be active in tumors, making patient stratification difficult. Given that overexpression of MET commonly occurs in primary and bevacizumab-resistant GBM patients, MET-CAR T cell therapy may become an effective strategy for targeting MET overexpression. In this study, we show that both MET-CAR T cells specifically inhibited MET-positive U87, G159, and G215 cell growth but had no effect on MET-negative G91 or HEK293 cells (Fig. 4), which is consistent to our previous findings with HCC preclinical models [28]. Consistent to the efficacy data, we observed significant up-regulation of 12 out of 21 cytokines released by MET-CAR T cells after co-culture with MET-positive GBM cells, including GM-CSF, IL-2, interferon gamma (IFNγ) and granzyme A, which are common measures of T cell activation (Fig. 5A, B). These results demonstrate strong activation of both MET-CAR.CD28ζ and MET-CAR.4-1BBζ T cells in response to antigen stimulation by the PDX tumor cells.

Given that MET-CAR.CD28ζ demonstrated better efficacy than MET-CAR.41BBζ in inhibiting MET-positive HCC tumor growth [28], we selected MET-CAR.CD28ζ for evaluation of therapeutic efficacy against GBM orthotopic xenografts in vivo. The fact that U87, G159 and G215 tumors all responded to MET-CAR.CD28ζ T cell therapy suggest that HGF-autocrine activation, METamp and ZM fusion may serve as biomarkers to predefine GBM patients for MET-CAR T cell therapy. Comparing with U87 and G159 by which a single dose of intracranial delivery of MET-CAR.CD28ζ T cells significantly inhibited orthotopic tumor growth (Fig. 6, p < 0.05), the inhibition of G215 tumors was less potent, most likely due to a faster tumor growth phenotype (G215 vs. U87 and G159). We anticipate that increasing MET-CAR T cell numbers and/or dosing at an earlier time point after G215 orthotopic tumor injection will improve the therapeutic efficacy. A second dose at 3 weeks after tumor initiation may further improve the inhibitory effects. Taken together, our results suggest that MET-CAR.CD28ζ T cells have the therapeutic potential to effectively target MET-positive GBMs.

Although several CAR T cell targets under development for treating GBM are showing promising preclinical results, limited antitumor responses were observed in clinical trials. This is largely due to the immunosuppressive nature of tumor microenvironment (TME), including the presence of inhibitory cytokines, regulatory T cells and myeloid-derived suppressor cells, leading to CAR T cell dysfunction and exhaustion [20, 45]. Therefore, approaches to improve MET-CAR T cell expansion, persistence and survival are the key to the success of MET-CAR T cell therapy. Recent studies have shown that genetic editing of CAR T cells to overexpress cytokines such as IL-12 [38], IL-15 [39], IL-18 [40] and IL-21[41, 42] enhanced therapeutic efficacy in various solid tumor preclinical models. Thus, optimization of MET-CAR T cells to armor cytokine production may serve as a combination strategy to improve MET-CAR T cell efficacy. Of note, we also observed IL-4 and IL10 upregulation in activated MET-CAR T cells (Fig. 5). Both IL-4 and IL-10 are type 2 cytokines contributing to an immunosuppressive TME by activating tumor-associated M2 macrophage [46]; however, recent studies have suggested controversial roles on T cell functions. While Steward et al. reported that IL-4 drives CD8+ CAR T cell exhaustion [47], Feng et al. showed that a mutant IL-4 fusion protein with an extended circulating half-life rejuvenated terminally exhausted CD8+ T cells, leading to enhanced antitumor efficacy of Her2-targeting CAR T cell therapy in a colon cancer xenograft model [48]. Similarly, Ravi et al. reported that myeloid cells released IL-10 which drives T cell exhaustion in GBM [49]; however, Zhao et al.; showed that engineering CAR T cells to produce IL-10 improved CAR T cell expansion and memory cell phenotype, leading to enhanced anti-tumor efficacy [50]. Thus, further investigations are needed to understand their functions during MET-CAR T cell therapy.

In summary, we characterized GBM PDX models harboring METamp and ZM fusion and tested their sensitivity to MET TKIs and specific MET-CAR T cells. Whereas MET inhibitors are effective in MET-active tumors, MET-CAR T cells eradicate MET-positive GBM growth in an antigen-dependent manner, demonstrating a novel therapeutic approach for GBM treatment. While MET-CAR.CD28ζ T cells have shown promising therapeutic potential against GBM tumor growth in preclinical animal models, further optimization of MET-CARs and development of combination strategies will improve the MET-CAR T cell therapy.

Availability of data and materials

Data are available upon reasonable request.

Abbreviations

bFGF:

Basic fibroblast growth factor

BLI:

Bioluminescence imaging

CAH:

Carbonic anhydrase-like

CAR:

Chimeric antigen receptor

DMEM:

Dulbecco’s Modified Eagle’s Medium

E:T:

Effector T cell:tumor cell

ECD:

Extracellular domain

EGF:

Epidermal growth factor

EGFRvIII:

Epidermal growth factor receptor variant III

FBS:

Fetal bovine serum

GBM:

Glioblastoma

GFAP:

Glial fibrillary acidic protein

HCC:

Hepatocellular carcinoma

HER2:

Human epidermal growth factor receptor 2

HGF:

Hepatocyte growth factor

i.c :

Intracranial

IF:

Immunofluorescence

IL:

Interleukin

IFNγ:

Interferon gamma

IPT:

Ig-like, plexin, transcription factor

METamp :

MET amplification

MetMab:

Anti-MET monoclonal antibody

METΔ7-8:

MET exon 7-8 deletion

METΔ14:

Exon 14 deletion

NT:

Non-transduced

PBMCs:

Peripheral blood mononuclear cells

NSCs:

Neuro stem cells

PDX:

Patient-derived xenografts

PSI:

Plexins, Semaphorins, Integrins

PTPRZ1:

Protein tyrosine phosphatase receptor type Z1

RTK:

Receptor tyrosine kinases

sGBM:

Secondary GBM

TME:

Tumor microenvironment

TKI:

Tyrosine kinase inhibitor

ZM fusion:

PTPRZ1-MET fusion

VEGFR2:

Vascular epithelial growth factor receptor 2

References

  1. Xie Q, Mittal S, Berens ME. Targeting adaptive glioblastoma: an overview of proliferation and invasion. Neuro Oncol. 2014;16(12):1575–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J, Dirks PB. Identification of a cancer stem cell in human brain tumors. Cancer Res. 2003;63(18):5821–8.

    CAS  PubMed  Google Scholar 

  3. Liu G, Yuan X, Zeng Z, Tunici P, Ng H, Abdulkadir IR, Lu L, Irvin D, Black KL, Yu JS. Analysis of gene expression and chemoresistance of CD133+ cancer stem cells in glioblastoma. Mol Cancer. 2006;5:67.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q, Hjelmeland AB, Dewhirst MW, Bigner DD, Rich JN. Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature. 2006;444(7120):756–60.

    Article  CAS  PubMed  Google Scholar 

  5. Lee J, Kotliarova S, Kotliarov Y, Li A, Su Q, Donin NM, Pastorino S, Purow BW, Christopher N, Zhang W, et al. Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines. Cancer Cell. 2006;9(5):391–403.

    Article  CAS  PubMed  Google Scholar 

  6. Cancer Genome Atlas Research N. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature. 2008;455(7216):1061–8.

    Article  Google Scholar 

  7. Navis AC, van Lith SA, van Duijnhoven SM, de Pooter M, Yetkin-Arik B, Wesseling P, Hendriks WJ, Venselaar H, Timmer M, van Cleef P, et al. Identification of a novel MET mutation in high-grade glioma resulting in an auto-active intracellular protein. Acta Neuropathol. 2015;130(1):131–44.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Bao ZS, Chen HM, Yang MY, Zhang CB, Yu K, Ye WL, Hu BQ, Yan W, Zhang W, Akers J, et al. RNA-seq of 272 gliomas revealed a novel, recurrent PTPRZ1-MET fusion transcript in secondary glioblastomas. Genome Res. 2014;24(11):1765–73.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Xie Q, Bradley R, Kang L, Koeman J, Ascierto ML, Worschech A, De Giorgi V, Wang E, Kefene L, Su Y, et al. Hepatocyte growth factor (HGF) autocrine activation predicts sensitivity to MET inhibition in glioblastoma. Proc Natl Acad Sci U S A. 2012;109(2):570–5.

    Article  PubMed  Google Scholar 

  10. Olmez OF, Cubukcu E, Evrensel T, Kurt M, Avci N, Tolunay S, Bekar A, Deligonul A, Hartavi M, Alkis N, et al. The immunohistochemical expression of c-Met is an independent predictor of survival in patients with glioblastoma multiforme. Clin Transl Oncol. 2014;16(2):173–7.

    Article  CAS  PubMed  Google Scholar 

  11. Petterson SA, Dahlrot RH, Hermansen SK, Munthe SKA, Gundesen MT, Wohlleben H, Rasmussen T, Beier CP, Hansen S, Kristensen BW. High levels of c-Met is associated with poor prognosis in glioblastoma. J Neurooncol. 2015;122(3):517–27.

    Article  CAS  PubMed  Google Scholar 

  12. Liu W, Fu Y, Xu S, Ding F, Zhao G, Zhang K, Du C, Pang B, Pang Q. c-Met expression is associated with time to recurrence in patients with glioblastoma multiforme. J Clin Neurosci. 2011;18(1):119–21.

    Article  CAS  PubMed  Google Scholar 

  13. Carvalho B, Lopes JM, Silva R, Peixoto J, Leitao D, Soares P, Fernandes AC, Linhares P, Vaz R, Lima J. The role of c-Met and VEGFR2 in glioblastoma resistance to bevacizumab. Sci Rep. 2021;11(1):6067.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Qin Y, Musket A, Kou J, Preiszner J, Tschida BR, Qin A, Land CA, Staal B, Kang L, Tanner K, et al. Overexpression of HGF/MET axis along with p53 inhibition induces de novo glioma formation in mice. Neurooncol Adv. 2020;2(1):vdaa067.

    PubMed  PubMed Central  Google Scholar 

  15. Camacho CV, Todorova PK, Hardebeck MC, Tomimatsu N, Gil del Alcazar CR, Ilcheva M, Mukherjee B, McEllin B, Vemireddy V, Hatanpaa K, et al. DNA double-strand breaks cooperate with loss of Ink4 and Arf tumor suppressors to generate glioblastomas with frequent Met amplification. Oncogene. 2015;34(8):1064–72.

    Article  CAS  PubMed  Google Scholar 

  16. Li Y, Li A, Glas M, Lal B, Ying M, Sang Y, Xia S, Trageser D, Guerrero-Cazares H, Eberhart CG, et al. c-Met signaling induces a reprogramming network and supports the glioblastoma stem-like phenotype. Proc Natl Acad Sci U S A. 2011;108(24):9951–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Joo KM, Jin J, Kim E, Ho Kim K, Kim Y, Gu Kang B, Kang YJ, Lathia JD, Cheong KH, Song PH, et al. MET signaling regulates glioblastoma stem cells. Cancer Res. 2012;72(15):3828–38.

    Article  CAS  PubMed  Google Scholar 

  18. Garber K. MET inhibitors start on road to recovery. Nat Rev Drug Discov. 2014;13(8):563–5.

    Article  CAS  PubMed  Google Scholar 

  19. Qin A, Musket A, Musich PR, Schweitzer JB, Xie Q. Receptor tyrosine kinases as druggable targets in glioblastoma: do signaling pathways matter? Neurooncol Adv. 2021;3(1):vdab133.

    PubMed  PubMed Central  Google Scholar 

  20. Land CA, Musich PR, Haydar D, Krenciute G, Xie Q. Chimeric antigen receptor T-cell therapy in glioblastoma: charging the T cells to fight. J Transl Med. 2020;18(1):428.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Brown CE, Starr R, Aguilar B, Shami AF, Martinez C, D’Apuzzo M, Barish ME, Forman SJ, Jensen MC. Stem-like tumor-initiating cells isolated from IL13Ralpha2 expressing gliomas are targeted and killed by IL13-zetakine-redirected T Cells. Clin Cancer Res. 2012;18(8):2199–209.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Morgan RA, Johnson LA, Davis JL, Zheng Z, Woolard KD, Reap EA, Feldman SA, Chinnasamy N, Kuan CT, Song H, et al. Recognition of glioma stem cells by genetically modified T cells targeting EGFRvIII and development of adoptive cell therapy for glioma. Hum Gene Ther. 2012;23(10):1043–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Ahmed N, Salsman VS, Kew Y, Shaffer D, Powell S, Zhang YJ, Grossman RG, Heslop HE, Gottschalk S. HER2-specific T cells target primary glioblastoma stem cells and induce regression of autologous experimental tumors. Clin Cancer Res. 2010;16(2):474–85.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Hu H, Mu Q, Bao Z, Chen Y, Liu Y, Chen J, Wang K, Wang Z, Nam Y, Jiang B, et al. Mutational landscape of secondary glioblastoma guides MET-targeted trial in brain tumor. Cell. 2018;175(6):1665-1678 e1618.

    Article  CAS  PubMed  Google Scholar 

  25. Kwak Y, Kim SI, Park CK, Paek SH, Lee ST, Park SH. C-MET overexpression and amplification in gliomas. Int J Clin Exp Pathol. 2015;8(11):14932–8.

    PubMed  PubMed Central  Google Scholar 

  26. Merchant M, Ma X, Maun HR, Zheng Z, Peng J, Romero M, Huang A, Yang NY, Nishimura M, Greve J, et al. Monovalent antibody design and mechanism of action of onartuzumab, a MET antagonist with anti-tumor activity as a therapeutic agent. Proc Natl Acad Sci U S A. 2013;110(32):E2987-2996.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Cloughesy T, Finocchiaro G, Belda-Iniesta C, Recht L, Brandes AA, Pineda E, Mikkelsen T, Chinot OL, Balana C, Macdonald DR. Randomized, double-blind, placebo-controlled, multicenter phase II study of onartuzumab plus bevacizumab versus placebo plus bevacizumab in patients with recurrent glioblastoma: efficacy, safety, and hepatocyte growth factor and O (6)-methylguanine-DNA methyltransferase biomarker analyses. J Clin Oncol. 2017;35(3):343–51.

    Article  CAS  PubMed  Google Scholar 

  28. Qin A, Qin Y, Lee J, Musket A, Ying M, Krenciute G, Marincola FM, Yao ZQ, Musich PR, Xie Q. Tyrosine kinase signaling-independent MET-targeting with CAR-T cells. J Transl Med. 2023;21(1):682.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Shah PD, Huang AC, Xu X, Orlowski R, Amaravadi RK, Schuchter LM, Zhang P, Tchou J, Matlawski T, Cervini A, et al. Phase I trial of autologous RNA-electroporated cMET-directed CAR T cells administered intravenously in patients with melanoma and breast carcinoma. Cancer Res Commun. 2023;3(5):821–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Johnson J, Ascierto ML, Mittal S, Newsome D, Kang L, Briggs M, Tanner K, Marincola FM, Berens ME, Vande Woude GF, et al. Genomic profiling of a Hepatocyte growth factor-dependent signature for MET-targeted therapy in glioblastoma. J Transl Med. 2015;13:306.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Stein RA, McHaourab HS. SPEACH_AF: sampling protein ensembles and conformational heterogeneity with Alphafold2. PLoS Comput Biol. 2022;18(8): e1010483.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Mirdita M, Schutze K, Moriwaki Y, Heo L, Ovchinnikov S, Steinegger M. ColabFold: making protein folding accessible to all. Nat Methods. 2022;19(6):679–82.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Bouyain S, Watkins DJ. Identification of tyrosine phosphatase ligands for contactin cell adhesion molecules. Commun Integr Biol. 2010;3(3):284–6.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Bouyain S, Watkins DJ. The protein tyrosine phosphatases PTPRZ and PTPRG bind to distinct members of the contactin family of neural recognition molecules. Proc Natl Acad Sci U S A. 2010;107(6):2443–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Uchikawa E, Chen Z, Xiao GY, Zhang X, Bai XC. Structural basis of the activation of c-MET receptor. Nat Commun. 2021;12(1):4074.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Vaubel RA, Tian S, Remonde D, Schroeder MA, Mladek AC, Kitange GJ, Caron A, Kollmeyer TM, Grove R, Peng S, et al. Genomic and phenotypic characterization of a broad panel of patient-derived xenografts reflects the diversity of glioblastoma. Clin Cancer Res. 2020;26(5):1094–104.

    Article  CAS  PubMed  Google Scholar 

  37. Kou J, Musich PR, Staal B, Kang L, Qin Y, Yao ZQ, Zhang B, Wu W, Tam A, Huang A, et al. Differential responses of MET activations to MET kinase inhibitor and neutralizing antibody. J Transl Med. 2018;16(1):253.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Chmielewski M, Kopecky C, Hombach AA, Abken H. IL-12 release by engineered T cells expressing chimeric antigen receptors can effectively Muster an antigen-independent macrophage response on tumor cells that have shut down tumor antigen expression. Cancer Res. 2011;71(17):5697–706.

    Article  CAS  PubMed  Google Scholar 

  39. Krenciute G, Prinzing BL, Yi Z, Wu MF, Liu H, Dotti G, Balyasnikova IV, Gottschalk S. Transgenic expression of IL15 improves antiglioma activity of IL13Ralpha2-CAR T cells but results in antigen loss variants. Cancer Immunol Res. 2017;5(7):571–81.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Jaspers JE, Khan JF, Godfrey WD, Lopez AV, Ciampricotti M, Rudin CM, Brentjens RJ. IL-18-secreting CAR T cells targeting DLL3 are highly effective in small cell lung cancer models. J Clin Invest. 2023;133(9): e166028.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Zhu W, Zhang Z, Chen J, Chen X, Huang L, Zhang X, Huang X, Ma N, Xu W, Yi X, et al. A novel engineered IL-21 receptor arms T-cell receptor-engineered T cells (TCR-T cells) against hepatocellular carcinoma. Signal Transduct Target Ther. 2024;9(1):101.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Liu Y, Dang Y, Zhang C, Liu L, Cai W, Li L, Fang L, Wang M, Xu S, Wang G, et al. IL-21-armored B7H3 CAR-iNKT cells exert potent antitumor effects. iScience. 2024;27(1): 108597.

    Article  CAS  PubMed  Google Scholar 

  43. Snuderl M, Fazlollahi L, Le LP, Nitta M, Zhelyazkova BH, Davidson CJ, Akhavanfard S, Cahill DP, Aldape KD, Betensky RA, et al. Mosaic amplification of multiple receptor tyrosine kinase genes in glioblastoma. Cancer Cell. 2011;20(6):810–7.

    Article  CAS  PubMed  Google Scholar 

  44. Martens T, Schmidt NO, Eckerich C, Fillbrandt R, Merchant M, Schwall R, Westphal M, Lamszus K. A novel one-armed anti-c-Met antibody inhibits glioblastoma growth in vivo. Clin Cancer Res. 2006;12(20 Pt 1):6144–52.

    Article  CAS  PubMed  Google Scholar 

  45. Akhavan D, Alizadeh D, Wang D, Weist MR, Shepphird JK, Brown CE. CAR T cells for brain tumors: lessons learned and road ahead. Immunol Rev. 2019;290(1):60–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Mirlekar B. Tumor promoting roles of IL-10, TGF-beta, IL-4, and IL-35: Its implications in cancer immunotherapy. SAGE Open Med. 2022;10:20503121211069012.

    Article  PubMed  PubMed Central  Google Scholar 

  47. Stewart CM, Siegler EL, Sakemura RL, Cox MJ, Huynh T, Kimball B, Mai L, Can I, Manriquez Roman C, Yun K, et al. IL-4 drives exhaustion of CD8(+) CART cells. Nat Commun. 2024;15(1):7921.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Feng B, Bai Z, Zhou X, Zhao Y, Xie YQ, Huang X, Liu Y, Enbar T, Li R, Wang Y, et al. The type 2 cytokine Fc-IL-4 revitalizes exhausted CD8(+) T cells against cancer. Nature. 2024;634(8034):712–20.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Ravi VM, Neidert N, Will P, Joseph K, Maier JP, Kuckelhaus J, Vollmer L, Goeldner JM, Behringer SP, Scherer F, et al. T-cell dysfunction in the glioblastoma microenvironment is mediated by myeloid cells releasing interleukin-10. Nat Commun. 2022;13(1):925.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. Zhao Y, Chen J, Andreatta M, Feng B, Xie YQ, Wenes M, Wang Y, Gao M, Hu X, Romero P, et al. IL-10-expressing CAR T cells resist dysfunction and mediate durable clearance of solid tumors and metastases. Nat Biotechnol. 2024;42(11):1693–704.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We thank Drs. Stephen Gottschalk from St. Jude Children’s Research Hospital for providing CAR vectors. We thank The Mayo Clinic GBM Xenograft National Resource (Dr. Jann Sarkaria) for providing the GBM-PDX models. We thank Dr. Regenia Campbell at ETSU molecular biology core for performing the multiplex cytokine analysis.

Funding

This study is funded by an ETSU College of Medicine Research Enhancement Award, an ETSU RDC award (825-30), an ETSU MBCF award (2024) and a NIH/NS120062-01 A1 (to Q.X.), and Graduate Student Research Grants (E85078 and E85096, to A.M.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author information

Author notes

  1. Anna Qin and Anna Musket contributed equally to this work.

Authors and Affiliations

  1. Department of Biomedical Sciences, Quillen College of Medicine, East Tennessee State University, 1276 Gilbreath Dr, Johnson City, TN, 37614, USA

    Anna Qin, Anna Musket, Phillip R. Musich & Qian Xie

  2. Cytogenetics Laboratory, Greenwood Genetic Center, Greenwood, SC, 29646, USA

    Benjamin Hilton

  3. Department of Pathology, Quillen College of Medicine, East Tennessee State University, Johnson City, TN, 37604, USA

    Johanna Preiszner

  4. Department of Bone Marrow Transplantation and Cellular Therapy, St. Jude Children’s Research Hospital, Memphis, TN, 38105, USA

    Giedre Krenciute

  5. Clinical Genomics and Therapeutics Division, Translational Genomics Research Institute, Phoenix, AZ, 85004, USA

    Michael E. Berens

  6. Department of Neurology, Hugo W. Moser Research Institute at Kennedy Krieger, Johns Hopkins University School of Medicine, Baltimore, MD, USA

    Mingyao Ying

  7. Center of Excellence for Inflammation, Infectious Disease and Immunity, East Tennessee State University, Johnson City, TN, 37614, USA

    Qian Xie

Authors
  1. Anna Qin
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Anna Musket
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Benjamin Hilton
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Johanna Preiszner
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Giedre Krenciute
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Michael E. Berens
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Mingyao Ying
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Phillip R. Musich
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Qian Xie
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

QX conceived the project. AQ, AM and QX characterized GBM-PDX and performed MET-CAR T cell analysis. BH performed FISH analysis. JP did the pathological staining and PRM performed protein structure analysis. AQ, AM, PRM and QX analyzed the data. QX and PRM wrote and edited the original manuscript. GK, MEB, and YM reviewed and commented on the manuscript. All authors read and approved of the final manuscript.

Corresponding author

Correspondence to Qian Xie.

Ethics declarations

Ethics approval and consent to participate

This study used human whole blood from de-identified healthy donors purchased from a commercial vendor and therefore is not a human subject research.

Consent for publication

All authors give their consent to publish this manuscript.

Competing interests

This manuscript is associated with a patent (US Patent No. 11,866,493).

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

12967_2025_6475_MOESM1_ESM.tif

Supplementary Material 1. Figure 1A. RT-PCR results of wild-type MET and ZM fusion in GBM cells. B. Sanger sequencing analysis of RT-PCR product from G215 cells reveals ZM fusion at PTPRZ1 exon 8 to MET exon 2. C. Amino acid sequence of the G215 ZM protein ECD including the PTPRZ1 CAH domain and its SEMA domain. The first 309 residues represent the amino terminal portion of the PTPRZ1 protein, residues 310-314 result from the fusion event and M315 is the initiation codon for the MET protein. The signal peptide regions of PTPRZ1 and MET are indicated in Italic font with reduced size. The 793 residues that were included in the AlphaFold-2 modeling analysis are marked in blue; residues of the loop region between PTPRZ1 and the MET SEMA domain are underlined. Residues G831-T1264 represent the single PSI domain and the four IPT domains of MET’s ECD. D. Structural comparison of MetMab in binding to MET-SEMA or ZM-SEMA. The MetMab antibody H- and L-chains bind to MET-SEMA domain β-folds 5 & 6 as indicated in yellow. This overlaps the HGF-α binding sites in both MET-SEMA and ZM-SEMA domains as indicated in brown. In both cases, red indicates the HGF-β binding domain spanning β-folds 2, 3 and part of 4, which is not affected by the MetMab binding.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Qin, A., Musket, A., Hilton, B. et al. Efficacy of MET-targeting CAR T cells against glioblastoma patient-derived xenograft models. J Transl Med 23, 460 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-025-06475-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-025-06475-6

Keywords

Journal of Translational Medicine

ISSN: 1479-5876

Contact us