Selective Targeting of Glioblastoma with EGFRvIII/EGFR … · Research Article Selective Targeting of Glioblastoma with EGFRvIII/EGFR Bitargeted Chimeric Antigen Receptor T Cell Hua

Research Article

Selective Targeting of Glioblastoma withEGFRvIII/EGFR Bitargeted Chimeric AntigenReceptor T CellHua Jiang1, Huiping Gao1, Juan Kong1, Bo Song2, Peng Wang2, Bizhi Shi1,Huamao Wang2, and Zonghai Li1,2

Abstract

The heterogeneous expression of EGFRvIII [variant IIImutant of epidermal growth factor receptor (EGFR)] inglioblastoma has significant impact on the clinicalresponse to the treatment of EGFRvIII-specific chimericantigen receptor–engineered T (CAR T) cells. We hypoth-esized that CAR T cells that could target both EGFRvIII andthe form of EGFR expressed on tumor cells, but not EGFRon normal cells, would greatly improve efficacy withoutinducing on-target, off-tumor toxicity. Therefore, we devel-oped a humanized single-chain antibody, M27, with asingle specificity that binds to an epitope found both onwild-type EGFR- and EGFRvIII-overexpressing tumor cells,but not EGFR-expressing normal cells, including primarykeratinocytes, on which wild-type EGFR is highlyexpressed. M27-derived CAR T cells effectively lysed EGFR-vIII- or EGFR-overexpressing tumor cells, but showed no

observable toxicity on normal cells. Inclusion of theCD137 (4-1BB) costimulatory intracellular domain in theM27-28BBZ CAR provided CAR T cells with higher tumorlysis activity than when not included (as in the M27-28ZCAR). The growth of established EGFR- or EGFRvIII-over-expressing glioma xenografts was suppressed by M27-28BBZ CAR T cells as well. The growth of heterogeneicxenograft tumors, created by mixing EGFR- and EGFR-overexpressing glioblastoma cells, was also effectivelyinhibited by M27-28BBZ CAR T cells. The survival of micein the orthotopic models was significantly prolonged afterM27-28BBZ CAR T-cell infusion. These results suggestedthat tumor-selective, bitargeted anti-EGFR/EGFRvIII CAR Tcells may be a promising modality for the treatment ofpatients with EGFR/EGFRvIII-overexpressing glioblastoma.Cancer Immunol Res; 6(11); 1314–26. �2018 AACR.

IntroductionGlioblastoma is the most common and aggressive brain

malignancy without effective treatment options (1, 2). In theUnited States, approximately 10,000 new cases are diagnosedannually (3). Standard-of-care procedures include surgical resec-tion followed by radiotherapy and/or chemotherapy withthe alkylating agent temozolomide (4). Despite therapeuticadvances, the prognosis of glioblastoma remains very poor, witha median survival of less than 2 years (5). The resistance ofglioblastoma to standard therapies results in a high rate of tumorrecurrence, and the lack of tumor specificity of the treatment mayresult in a significant damage to healthy brain tissues (6). There isconsequently a high unmet medical need for a safe and effica-cious tumor-selective therapy against glioblastoma.

Chimeric antigen receptor (CAR)–engineered T (CAR T) cellsare a promising strategy for cancer treatment (7). A classicCAR T structure includes an extracellular tumor-targeting sin-gle-chain variable fragment from an antibody (scFv domain), ashort transmembrane domain, and a tandemly assembledintracellular costimulatory domain with intracellular T-cellsignaling moieties. Preclinical and clinical studies have shownCAR T cells redirected to IL13Ra2 and HER2 could inducetumor regression in glioblastoma (8–11), suggesting that CAR Tcells could be developed as next-generation therapeutics forglioblastoma treatment.

Epidermal growth factor receptor (EGFR) is amplified inapproximately 40% to 60% of glioblastomas (12, 13). EGFRvIII,a cancer-specific variant of EGFR, is found in approximately 31%of glioblastoma patients (12). Antibodies to EGFR and small-molecule inhibitors targeting EGFR have been used to treatpatients with glioblastoma, but were not clinically efficacious(14, 15). CAR T cells targeting EGFRvIII can selectively eliminateEGFRvIII-expressing glioblastoma cells (14, 16, 17). AlthoughCAR T cells efficiently infiltrated and eliminated EGFRvIII tumorcells in some patients, neither partial nor complete responseswere achieved in the clinical trials (18).

An increasing body of data suggests that intratumoral het-erogeneity is one of the principal causes of tumor relapse(19, 20). The poor clinical outcome of glioblastoma patientstreated with EGFRvIII-targeted CAR T cells might be attributed,in part, to the fact that EGFRvIII was only expressed in asubpopulation of tumor cells (21, 22). We hypothesized that

1State Key Laboratory of Oncogenes and Related Genes, Shanghai CancerInstitute, Renji Hospital, Shanghai Jiao Tong University School of Medicine,Shanghai, China. 2CARsgen Therapeutics, Shanghai, China.

Note: Supplementary data for this article are available at Cancer ImmunologyResearch Online (http://cancerimmunolres.aacrjournals.org/).

Corresponding Author: Zonghai Li, Shanghai Cancer Institute, Renji Hospital,Shanghai Jiaotong University School of Medicine, No. 25/Ln2200, XieTu Road,Shanghai 200032, China. Phone: 86-21-64436601; Fax: 86-21-64432027; E-mail:[email protected]

doi: 10.1158/2326-6066.CIR-18-0044

�2018 American Association for Cancer Research.

CancerImmunologyResearch

Cancer Immunol Res; 6(11) November 20181314

on March 30, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst September 10, 2018; DOI: 10.1158/2326-6066.CIR-18-0044

http://crossmark.crossref.org/dialog/?doi=10.1158/2326-6066.CIR-18-0044&domain=pdf&date_stamp=2018-10-16

http://cancerimmunolres.aacrjournals.org/

bitargeted CAR T cells that recognize both wild-type EGFR andEGFRvIII overexpressed by glioma cells would show betterefficacy than CAR T cells targeting EGFRvIII alone. Monoclonalantibody 806 (mAb 806) targets a conformationally exposedepitope, EGFR287-302, that is found in both wild-type EGFRthat is expressed on tumor cells and the mutated oncogenicform of EGFR, EGFRvIII (23, 24). ABT-806, a humanized vari-ant of mAb 806, shows minimal skin toxicity in treated patients(25). ABT-414, an antibody–drug conjugate composed ofABT-806 and a potent antimicrotubule agent, monomethylauristatin F (MMAF) had no skin toxicity in human studieswith an objective response rate of approximately 6.8% (26).This evidence suggests that the immune-oncology therapeuticstargeting EGFR287–302 have potential for improved safety andefficacy. Therefore, here we generated CAR T cells recognizingEGFR287–302 and characterized their safety and efficacy for thetreatment of glioblastoma.

Materials and MethodsCell lines

The human glioblastoma cell lines U87MG, U251, humanepidermoid carcinoma cell line A431, human oral adenosqua-mous carcinoma cell line CAL27, and the human embryonickidney cell line 293T were obtained from ATCC. Human hepa-tocyte L02 cell and human prostate epithelial RWPE-1 cell werepurchased from the Cell Bank of the Chinese Academy of Sciences(Shanghai, China). U87MG-EGFR and U251-EGFR cells stablyexpressing human EGFR protein were generated in our laboratoryby transducing the U87MG and U251 cells with a VSV-G pseu-dotyped lentiviral vector encoding wild-type EGFR. U87MG-EGFRvIII and U251-EGFRvIII cells stably expressing EGFRvIIIprotein were generated in our laboratory by transducing theU87MG and U251 cells with a VSV-G pseudotyped lentiviralvector encoding exon 2–7-deleted EGFR. U251-luci-EGFR andU251-luci-EGFRvIII cells expressing firefly luciferase were alsoestablished by lentiviral transduction. All cells were culturedin Dulbecco's Modified Eagle Medium with 10% FCS (LifeTechnologies) with 100 mg/mL penicillin and 100 U/mL strep-tomycin (Life Technologies). Human primary keratinocytes wereisolated from healthy human skin and cultured in EpiLifeMediumwith 60 mmol/L calcium (Life Technologies) withHumanKeratinocyte Growth Supplement (HKGS; Life Technologies). Thecell lines were authenticated by using short tandem repeat analysis.Mycoplasma contamination testing was routinely performed byusing PCR every 3 months in our laboratory. These cell lines hadbeen cultured for 3 to 4 years in our lab and were cryopreservedto create a working cell bank in which each vial of cells wassubject to subculture for up to 4 weeks after recovery. All cellswere maintained at 37�C in humidified air with 5% CO2.

EGFR CAR constructionM27-scFv, an anti–EGFR-specific single-chain variable frag-

ment (scFv), was derived from the humanized EGFR antibodyM27 that recognizes the residues 287–302 of EGFR. The sequenceencoding the M27-scFv antibody in the VL-VH orientationwas cloned by PCR amplification from a plasmid encodingscFv-M27-Fc. Another EGFR-specific scFv, 806-scFv, was derivedfrom the chimeric antibody ch806 (24) that binds amino acids287 to 302 of EGFR as well. The sequence encoding the 806scFv antibody in the VL-VH orientation was obtained by PCR

amplification from a plasmid encoding scFv-806-Fc. M27-28BBZand 806-28BBZ CARs contained extracellular domains of thehuman CD8a signal peptide (nucleotides 1–63, GenBankNM 001768.6), M27-scFv or 806-scFv, CD8a hinge (nucleotides412–546, GenBank NM 001768.6), and CD28 transmembranedomain (nucleotides 457–537, GenBank NM 006139.3). Theintracellular domains consisted of CD28 ("28"; nucleotides538–660,GenBankNM006139.3), CD137 (4-1BB, "BB"; nucleo-tides 640–765, GenBank NM 001561.5) costimulatory domainand CD3z ("Z"; nucleotides 154–492, GenBank NM 198253.2)costimulatory polypeptide. M27-28Z CAR had an extracellulardomain that consisted of human CD8a signal peptide (nucleo-tides 1–63, GenBank NM 001768.6), M27-scFv, and a CD8ahinge (nucleotides 412–546, GenBank NM 001768.6). The intra-cellular signaling domain of M27-28Z CAR consisted of CD28(nucleotides 538–660, GenBank NM 006139.3) and CD3z(nucleotides 154–492, GenBank NM 198253.2) molecules. The28BBZ and 28Z fragments were produced by PCR amplificationwith a plasmid template encoding the corresponding frag-ments (27). The DNA sequences of scFv-28BBZ or scFv-28Z werefurther cloned by PCR amplification. Both of these fragmentswere designed to have a MluI site at the 50 end and a SalI site atthe 30 end. The synthesized fragments were digested with MluIand SalI restriction enzymes (New England Biolabs) and theninserted into a similarly digested pRRLSIN.cPPT-GFP.WPREvector plasmid. The sequence integrity of all the vectors describedin this paper was confirmed by DNA sequencing. The mockconstruct was transduced by using the pRRLSIN.cPPT-GFP.WPRElentiviral vector.

Lentivirus productionHuman embryonic kidney 293T cells were seeded at 1.5� 107

per 15-cm dish for 24 hours prior to transfection. Expressionvector pRRLSIN.cPPT-GFP.WPRE (mock), or pRRLSIN-M27-28Z,or pRRLSIN-M27-28BBZ, or pRRLSIN-806-28BBZ was mixedwith two lentiviral packaging plasmids pMDLg/pRRE andpRSV-Rev plus an envelope expressing plasmid pCMV-VSV-G(from Addgene) to reconstitute a transfection DNA mixture inthe polyethylenimine-based DNA transfection reagent. 293T cellswere transfected with the reconstituted DNA mixture as men-tioned above. The viral supernatants were harvested and filtered(0.45-mm filter) at 72 hours after transfection. The lentiviralparticles were subsequently concentrated 30-fold by ultracentri-fugation (BeckmanOptima XL-100 K) for 2 hours at 28,000 rpm.

Transduction and culture of primary T cellsPeripheral blood mononuclear cells (PBMC) were cultured

in AIM-V medium (Life Technologies) in the presence of 2%humanAB serum (Huayueyang Biotechnology) and recombinanthuman IL2 (HuaxinHigh Biotechnology). For the transduction ofprimary T cells, PBMCs were activated for 48 hours by DynabeadsHuman T-Activator CD3/CD28 (Life Technologies) at a bead:cell ratio of 2:1 before infection. The activated T cells weretransduced with lentiviral vectors at a multiplicity of infectionof 15 in a 24-well plate coated with RetroNectin (Takara). Thetransduced T cells were cultured at a density of 5� 105 cells/mL inthe presence of rhIL2 (500 IU/mL).

Flow-cytometric analysisTo determine the binding of scFv to target cells, 1 � 106 cells

were incubated with 5 mg/mL scFv-M27-Fc, scFv-806-Fc, or

EGFR/EGFRvIII Bitargeted CAR T Cells for Cancer Treatment

www.aacrjournals.org Cancer Immunol Res; 6(11) November 2018 1315




scFv-C225-Fc antibody for 45 minutes at 4�C. PBS was usedas a negative control. After washing with FACS buffer (coldphosphate-buffered saline containing 1% newborn calf serum),the cells were incubated with FITC-conjugated goat antihumanIgG (H þ L; Catalog Number: SA00003-12; Proteintech) for45 minutes at 4�C. To quantify CAR expression, CAR T cells wereincubated with 20 mg/mL biotinylated antihuman-EGFR-F(ab0)2fragments at 4�C for 45 minutes. PBS was used as a negativecontrol. After FACS buffer wash, the cells were incubated withPE-conjugated streptavidin (eBioscience) for 45 minutes at4�C. Fluorophore-labeled CAR T cells were determined by usinga BD FACSCelesta flow cytometer. Data were analyzed usingFlowJo 7.6 software. Data are representative of three independentexperiments.

Analysis of CAR T-cell persistence in mouse peripheralblood lymphocytes

To examine the persistence of CAR T cells, whole mouseblood collected by retro-orbital bleeding was analyzed asfollowing: 50 mL of blood was incubated with the antibodyagainst CD3-PerCP/CD4-FITC/CD8-PE in the dark for 15 min-utes at room temperature. After lysis of red blood cells, cellswere fixed with 0.45 mL 1 � BD FACS Lysing Solution (BDBiosciences) for 15 minutes at room temperature. The fixedcells were then subjected to flow-cytometric analysis by usingFACSCelesta flow cytometer (BD Biosciences). Data from flowcytometry were further analyzed by using FlowJo 7.6 software.Absolute cell numbers per microliter of blood were determinedby using TruCount tubes (BD Biosciences) as described in themanufacturer's instruction.

Cytotoxicity assays in vitroEach line of glioma cells was cocultured with target-selective

CAR T cells or mock T cells at effector:target (E:T) ratios of 3:1,1:1, and 1:3. After 18-hour culture, supernatant lactate dehy-drogenase (LDH) activity was determined by using the CytoTox96 Nonradioactive Cytotoxicity Kit (Promega) as previouslydescribed (27).

Cytokine release assayCAR T cells or mock T cells were cocultured with glioma cells at

a 3:1 ratio in a 96-well culture plate. After 24-hour coculture, therelease of IFNg , TNFa, and IL2 cytokines from activated CAR Tcells or from mock T cells was determined by using an ELISAkit (MultiSciences Biotech) as described in the manufacturer'sinstructions.

In vivo efficacy studiesTo generate subcutaneous xenograft models, 6- to 8-week-old

female NOD/SCID mice were subcutaneously inoculated with3 � 106 U251-EGFR or 2 � 106 U251-EGFRvIII cells at the rightflank. After the tumor volume increased to 75 to 120 mm3, themice were randomly divided into two groups (n ¼ 6–10 pergroup) for receiving treatment regimen. The control group ofmice received intravenous (i.v.) injection of mock T cells(activated T cells transduced with eGFP-CAR). The treatmentgroup received i.v. injection of M27-28BBZ CAR T cells (acti-vated T cells transduced with EGFR CAR). Twenty-four hoursprior to CAR T-cell infusion, the mice were intraperitoneally(i.p.) injected with 100 mg/kg cyclophosphamide to depletehost lymphocytes and to enhance the tumor treatment efficacy

of the administered T cells (28). Two doses of 1 � 107

M27-28BBZ CAR T cells were i.v. injected via the tail vein in200 mL of PBS on days 14 and 17. Tumors were measured byusing calipers, and tumor volumes were calculated by using theformula V ¼ (length � width2)/2. Animal body weights andtumor volume measurements were carried out twice weekly.The mice were euthanized when their body weight loss wasgreater than 20% of the initial weight, the mean tumor volumeexceeded 2,000 mm3, or the tumors became ulcerated in themock control groups.

To generate the U251-luci-EGFR orthotopic model, 1 � 106

U251-luci-EGFR cells were intracranially implanted into 6- to8-week-old female NOD/SCID mice (n ¼ 8 mice per group).To generate the U251-luci-EGFR/U251-luci-EGFRvIII orthoto-pic model, the mixture of 7.5 � 105 U251-luci-EGFR and2.5 � 105 U251-luci-EGFRvIII cells were intracranially implant-ed into 6- to 8-week-old female NOD/SCID mice (n ¼ 7 miceper group). To generate the U251-luci-EGFRvIII orthotopicmodel, 5 � 105 U251-luci-EGFRvIII cells were intracraniallyimplanted into 6- to 8-week-old female NOD/SCID mice (n ¼6 mice per group). Implantation was performed by using astereotactic surgical device with injection of tumor cells at1 mm right and at 1 mm anterior to the bregma and at 3 mminto the brain on day 0. Seven days after surgery, the mice wererandomly divided into two groups. On day 8, the mice wereintravenously injected with a single dose of 1 � 107 CAR T cellsin 200 mL of PBS via the tail vein. Bioluminescent measure-ments were used as surrogates for tumor volume. The trans-duction efficiency of CAR T cells used in the assays was �50%.All animal experiments were performed by following the pro-tocol approved by the Shanghai Cancer Institute ExperimentalAnimal Care Commission.

Bioluminescence imagingBioluminescence imaging was performed by using the

IVIS system (IVIS, Xenogen). Briefly, tumor-bearing mice wereintraperitoneally injected with D-luciferin (150 mg/kg) andimaged under isoflurane anesthesia after 10 minutes. Theimages were quantified by using Living Image software (CaliperLife Sciences).

IHC analysisTo assess the infiltration of human T cells into tumors,

formalin-fixed and paraffin-embedded tumor tissues wereimmunostained by using anti-CD3 (Thermo Fisher Scientific).A normal rabbit IgG was used as an isotype-matched control.The procedures were performed as previously described (27).Briefly, after deparaffinization and rehydration, the sectionswere exposed to 3% H2O2 in methanol to eliminate endoge-nous peroxidase activity and then blocked with bovine serumalbumin (1%) for 30 minutes at room temperature (RT). Afterblocking, the sections were incubated with primary rabbitantihuman CD3 mAb overnight at 4�C. After PBS wash, thesections were incubated with peroxidase-conjugated second-ary antibodies (ChemMate DAKO EnVision Detection Kit,Peroxidase/DAB, Rabbit/Mouse) for 45 minutes at RT. Thesections were visualized by using a diaminobenzidine stainingkit (Dako Corporation) and then counterstained with hema-toxylin, dehydrated, cleared, mounted, and photographed. TheDAB-immunostained sections were analyzed by bright-fieldmicroscopy using an Olympus microscope (OLYMPUS IX71)

Jiang et al.

Cancer Immunol Res; 6(11) November 2018 Cancer Immunology Research1316




equipped with an image analysis software. CD3þ cells werequantified by measuring the number of stained T cells in eachsection by using the Image-Pro Plus (Media Cybernetics) soft-ware. Sections from 3 mice in each group were subjected todetermination of T-cell infiltration for statistical analysis. Themean count of the three areas was obtained and expressed asthe absolute number of CD3þ cells per 0.95 mm2 (200�magnification).

Statistical analysisAll data were presented as the mean � SE. Statistical signif-

icance was determined by using a two-way or one-way ANOVAwith Bonferroni posttest for multisample comparisons or theunpaired two-tailed Student t test for comparisons betweentwo samples. The overall survival statistics were calculated byusing the log-rank test. GraphPad Prism 5.0 was used forstatistical calculations. P < 0.05 was considered statisticallysignificant.

ResultsHumanized scFv recognizes overexpressed EGFR andEGFRvIII in tumor cells

To avoid anaphylaxis caused by mouse scFv-derived CAR Tcells (29), we generated a panel of humanized antibody frag-ments derived from a mouse EGFR monoclonal antibody 7B3

that specifically bound EGFR287-302 (Supplementary Table S1;Supplementary Fig. S1). EGFR- and EGFRvIII-transfectedU87MG and U251, two cancer cell lines with endogenousEGFR overexpression (A431 and CAL27), normal cells L-02(Human hepatocyte cell) and RWPE-1 (human prostate epi-thelial cell) as well as primary keratinocytes were used in thefollowing biological assays. The expression of EGFR was con-firmed by Western blot (Supplementary Fig. S2). After tworounds of affinity maturation and binding specificity enrich-ment, M27 scFv clone was selected for further characterizationof its binding specificity. The scFv derived from anti-EGFRantibody C225 (cetuximab) and mAb 806 were used as con-trols (Supplementary Fig. S3). Flow-cytometric analysis showedthat C225 scFv bound to most of the cells except for U87MGcells (Fig. 1A). The binding curves of C225 scFv to U251 cellsand to the three healthy keratinocytes were similar. 806scFv showed observable binding on EGFR- and EGFRvIII-overexpressing U87MG and U251 cells and three primarykeratinocytes. The mean fluorescence intensity (MFI) of 806scFv on U87MG-EGFRvIII, U251-EGFRvIII, U87MG-EGFR, andU251-EGFR cells was 22.6 � 0.42, 58.1 � 0.85, 1.57 � 0.05,and 9.81 � 0.08, respectively. The MFI of 806 scFv on threeprimary keratinocytes was 0.71 � 0.04, 0.57 � 0.02, and0.60 � 0.03. Humanized M27 scFv showed distinguish-able binding on EGFRvIII-overexpressing cells and EGFR-overexpressing U87MG and U251 cells. The MFI of M27 scFv

Figure 1.

FACS analysis demonstrated the binding of scFv-M27-Fc, scFv-806-Fc, and scFv-C225-Fc to the indicated target cells. To determine the binding of scFv-Fcproteins to untransfected or transfected U87MG, U251 glioblastoma cells, and three primary keratinocytes, 1 � 106 cells were isolated and incubatedwith 5 mg/mL of scFv-M27-Fc, scFv-806-Fc, and scFv-C225-Fc antibodies for 45 minutes at 4�C. After wash with FACS buffer, the cells were stainedwith an FITC-conjugated goat ant human IgG (HþL) for 45 minutes at 4�C. A, Fluorescence was assessed by using a BD FACSCelesta flow cytometer.B, The mean fluorescence intensity (MFI) of scFv proteins binding to cells was determined by flow-cytometric analysis.






on U87MG-EGFRvIII, U251-EGFRvIII, U87MG-EGFR, andU251-EGFR cells was 12.2 � 0.31, 19.4 � 0.50, 0.39 � 0.01,and 0.78� 0.02, respectively. The EC50 of M27 and 806 scFv onU87 MG-EGFR cell was 3.9 and 2.4 mmol/L respectively,whereas the EC50 of M27 and 806 scFv on U87 MG-EGFRvIIIcell was 99.5 nmol/L and 66.4 nmol/L, respectively (Supple-mentary Fig. S4). These results indicated that M27 scFv had alower affinity to EGFR- or EGFRvIII-overexpressing cells than806 scFv. However, M27 scFv did not bind to any of the threeprimary keratinocytes tested. In addition, all anti-EGFR scFvdemonstrated binding to endogenous EGFR-overexpressingcell lines A431 and CAL27. M27 scFv, however, showed rela-tively less affinity than other two scFv in binding to A431 andCAL27 (Fig. 1A). The MFI of different scFv proteins on thecells tested is shown (Fig. 1B). Additionally, we also evaluatedthe binding ability of M27 scFv to normal cells L-02 (humanhepatocyte cell) and RWPE-1 (human prostate epithelial cell),which endogenously express EGFR. The results showed thatM27 scFv also could not bind to these two cells (SupplementaryFig. S5). These data indicate that the M27 scFv specificallybound EGFR and EGFRvIII overexpressed on tumor cells butnot EGFR expressed on normal cells.

Molecular construction of CAR cassette and generation ofCAR T cells

We generated a series of recombinant lentiviral vectorsencoding various CAR cassettes that had an EGFR287–302 scFvantibody, an intracellular human T-cell costimulatory domainderived from human CD28, CD137, and CD3z chains, and alinker domain of human CD8 hinge and CD28 transmembraneregions (M27-28BBZ and 806-28BBZ). We also constructed aCAR cassette lacking the intracellular CD137 signaling domain(M27-28Z). The scheme of recombinant lentiviral constructs isshown in Fig. 2A.

Todetermine the expressionof the EGFRCARon the geneticallymodified T-cell surface, different CAR-transduced T cells weredetermined by flow cytometry using biotinylated human-EGFR-F(ab0)2 fragment antibody and PE-conjugated streptavidin. Thetransduction efficiency ranged from51.9% to 74.8%. As a control,mock T cells were transduced with the lentiviral vector encodingGFP gene. The transduction efficiency of mock T cells was deter-mined by eGFP expression (Fig. 2B).

M27-derived CAR T cells lyse EGFR- and/or EGFRvIII-overexpressing tumor cells

To determine whether CAR T cells targeting EGFR/EGFRvIIIcould selectively recognize and eliminate EGFR/EGFRvIII-posi-tive human glioblastoma cells, cytotoxicity assayswere performedby incubating the genetically modified CAR T cells with eithercontrol cells or EGFR- or/and EGFRvIII-overexpressing glioblas-toma cells. The results showed that both M27-28Z and M27-28BBZ CAR T cells efficiently lysed the EGFR/EGFRvIII-overex-pressing glioblastoma cells but not untransfected U87MG andU251 cells (Fig. 2C and D). Mock T cells showed very weakcytotoxicity compared with M27-derived CAR T cells (Fig. 2C).The M27-28BBZ CAR T cells showed higher cytotoxic activitiesthan M27-28Z CAR T cells in lysis of both EGFR- and EGFRvIII-overexpressing U87MG or U251 cells at 1:1 and 3:1 E:T ratio(P < 0.05; Fig. 2C and D). These data suggested that M27-28BBZCAR T cells would be a better candidate for further evaluation anddevelopment.

To determine the selective cytotoxicity of M27-28BBZ CAR Tcells in lysis of tumor cells, CAR T cells were incubated withprimary keratinocytes and EGFR/EGFRvIII-overexpressing tumorcells, respectively. 806 scFv-derived 806-28BBZ CAR T cells werealso included as a control. The data showed that both 806- andM27-derived CAR T cells could efficiently lyse EGFR- and EGFR-vIII-overexpressing U87MG and U251 cells. The cytotoxicitiesof both CAR T cells were similar in lysis of U251-EGFRvIII cells.806-28BBZ CAR T cells showed slightly higher cytotoxic effecton U87MG-EGFRvIII cells than on M27-28BBZ CAR T cells(806-28BBZ vs. M27-28BBZ at 1:3 E:T: 24.3% � 2.0% vs.16.2% � 0.9%; 806-28BBZ vs. M27-28BBZ at 1:1 E:T: 50.2% �3.7% vs. 43.0% � 2.5%; 806-28BBZ vs. M27-28BBZ at 3:1E:T: 68.9% � 4.4% vs. 57.8% � 4.7%; Fig. 3A and B). UnlikeM27-28BBZ, 806-28BBZ CAR T cells demonstrated a cytotoxiceffect on untransfected U251 cells at a 3:1 E:T ratio. At a 3:1 E:Tratio, both M27-28BBZ and 806-28BBZ CAR T cells couldlyse almost all the U251-EGFR cells. At an E:T ratio of 1:1,M27 CAR T cells had a lower lysis capacity than the 806 CAR Tcells in EGFR-transfected U87MG and U251 cells (Fig. 3A and B).In addition, both CAR T cells could effectively induce lysisof endogenous EGFR-overexpressing cancer cell lines A431 andCAL27. At E:T ratio of 3:1, no significant difference on the lysiscapacity was observed between these two CAR T cells (Fig. 3C).The M27-28BBZ CAR T cells had no measurable cytotoxic effecton all three primary keratinocytes; however, the 806-28BBZCAR T cells showed significantly higher cytotoxic effect onall three keratinocyte lines at 3:1 E:T ratio (806-28BBZ vs.M27-28BBZ on keratinocyte-1: 31.9% � 1.6% vs. 16.9% �0.8%, P < 0.001; 806-28BBZ vs. M27-28BBZ on keratinocyte-2:38.8% � 2.0% vs. 11.7% � 1.3%, P < 0.001; 806-28BBZ vs.M27-28BBZ on keratinocyte-3: 54.1%� 3.3% vs. 20.6%� 1.1%,P < 0.001; Fig. 3D). Additionally, M27-28BBZ CAR T cells had nocytotoxic effect on other normal cells L-02 and RWPE-1 (Supple-mentary Fig. S6). This indicated that M27-28BBZ CAR T cellscould selectively lyse EGFR- and/or EGFRvIII-overexpressingtumor cells but not EGFR-expressing normal cells.

M27-28BBZ CAR T cells produce cytokines in the presenceof target cells

Cytokine secretion byCART cells in response to a target antigenindicates activation and maintenance of an antigen-specificimmune response. The secretion of TNFa, IL2, and IFNg fromCAR T cells was determined to evaluate the activation of CAR Tcells by antigen-expressing tumor cells. Activated M27-28BBZCAR T cells secreted significantly more TNFa, IL2, and IFNg thandid mock T cells after incubation with EGFR/EGFRvIII-overex-pressing tumor cells (P < 0.05, Fig. 3E). In addition, M27-28BBZCAR T cells produced greater concentrations of cytokines inthe presence of EGFRvIII-overexpressing tumor cells than in thepresence of EGFR-overexpressing tumor cells. Secretion of thesecytokines from CAR T cells was negligible after incubation withuntransfected U87MG and U251 cells (Fig. 3E). Cytokine secre-tion was not induced in M27-28BBZ CAR T cells cocultured withthree primary keratinocytes, further supporting that M27-28BBZCAR T cells would not cross react with keratinocytes.

M27-28BBZ CAR T cells suppress growth of EGFR/EGFRvIII-overexpressing tumors

To determine the therapeutic efficacy of genetically modifiedT cells, M27-28BBZ CAR T cells and mock T cells were used

Jiang et al.





to treat mice bearing U251-EGFR or U251-EGFRvIII subcuta-neous tumors. The results demonstrated that M27-28BBZ CART cells could significantly inhibit tumor growth in both xeno-grafts. On day 42 following tumor cell inoculation, significantreduction of tumor volume (P < 0.001, Fig. 4A) and tumorweight (P < 0.0001, Fig. 4B) of the U251-EGFRvIII tumorswas observed in the group treated with M27-28BBZ CAR Tcells when compared with the group of mock T-cell treatment.On day 49 following tumor cell inoculation, the tumorvolume and tumor weight of U251-EGFR xenografts in the

M27-28BBZ CAR T-cell group was also significantly lower thanthose in the mock T-cell group (P < 0.05; Fig. 4C and D). Theseresults indicated that M27-28BBZ CAR T cells could efficientlysuppress the growth of both U251- EGFR and U251-EGFRvIIIcells in vivo.

Efficacy of M27-28BBZ CAR T cells in orthotopic glioblastomaxenograft models

Given that the subcutaneous microenvironment might dis-turb the activity of CAR T cells, we evaluated the antitumor

Figure 2.

Construction and characterization of EGFR/EGFRvIII bitargeted CAR T cells. A, Schematic representation of CARs. M27-28Z CAR consists of an scFv, ahuman CD8a hinge and a transmembrane (TM) region derived from human CD28, and an intracellular signaling domain from human CD28 and humanCD3z. M27-28BBZ or 806-28BBZ CAR comprises an scFv, a human CD8a hinge, a CD28 transmembrane domain, and an intracellular signalingdomain derived from human CD28, CD137, and CD3z. B, Expression of EGFR-specific CARs on the lentivirus-transduced human T cells was analyzedusing flow cytometry. Transduction efficiency is shown. C and D, In vitro cytotoxic activities of EGFR/EGFRvIII-targeted CAR T cells (M27-28Z or M27-28BBZ).Primary human T cells transduced with the indicated lentiviral vectors were incubated with the untransfected and transfected glioblastoma cells atthe various E:T ratios for 18 hours. Cell lysis was determined using a standard nonradioactive cytotoxic assay. Data are representative of threeindependent experiments. Each data point reflects the mean � SEM of triplicates (� , P < 0.05; �� , P < 0.01; �� , P < 0.001; two-tailed Student t test).






Figure 3.

In vitro cytotoxic activity of and cytokine production by M27-28BBZ T cells in the presence of tumor cells. Primary human T cells transduced with theindicated lentiviral vectors were incubated with cell lines at different E:T ratios for 18 hours. Cell lysis was determined using a standard nonradioactivecytotoxic assay. Data are representative of three independent experiments and shown as the mean � SEM of triplicates. A, The cytotoxic activitiesof M27-28BBZ and 806-28BBZ to U87MG, U87MG-EGFR, and U87MG-EGFRvIII cells. B, The cytotoxic assay of M27-28BBZ and 806-28BBZ using U251,U251-EGFR, and U251-EGFRvIII cells. C, The specific lysis of M27-28BBZ and 806-28BBZ of primary keratinocytes from three different human skinspecimens. D, The specific lysis of M27-28BBZ and 806-28BBZ to A431 and CAL27 cells. Data are representative of three independent experiments andshown as the mean � SEM of triplicates (� , P < 0.05; ��, P< 0.01; �� , P< 0.001; two-tail Student t test). E, Human T cells transduced with lentiviralvectors were cocultured with indicated cell lines at a 3:1 E:T ratio. Culture supernatants harvested at 24 hours after coculture were assayed for TNFa, IL2,and IFNg by ELISA. Data are representative of three independent experiments and presented as the mean � SEM of triplicates.

Jiang et al.





activities of M27-28BBZ CAR T cells in an intracranial glioblas-toma xenograft that contained luciferase-transfected U251-EGFR and U251-EGFRvIII tumor cells. Previous studies havereported that the expression of EGFRvIII in some cells is fre-quently associated with amplified expression of EGFR and thatthe coexpression of both receptors within the tumor mass con-fers a worse prognosis in patients with glioblastoma (12, 30).To model heterogeneous expression of EGFR and EGFRvIII inglioblastoma, U251-luci-EGFR cells and U251-luci-EGFRvIIIcells were mixed together at 3:1 ratio (EGFR-expressing cellsgrew much more slowly than EGFRvIII-expressing cells). Themixed cells were intracranially implanted into mice (day 0).After 6 days since glioblastoma xenografts were established, theengraftment ofU251-EGFR andU251-EGFRvIII cellswas confirmedby bioluminescence imaging. On day 7 following initial tumorengraftment, the mice were intravenously injected with one dose of1 � 107 M27-28BBZ CAR T cells or mock T cells. At day 28 aftertumor inoculation (21 days after T-cell injection), the tumors in theU251-EGFR xenograft model showed a significant reduction in

volume after treatment of M27-28BBZ CAR T cells compared withmock T cells (Fig. 5A and B). The survival of mice in M27-28BBZCAR T cell–treatment group was significantly prolonged comparedwith that in the mock T cell–treatment group (P < 0.05, Fig. 5C). Inthe U251-EGFRvIII and U251-EGFR/EGFRvIII xenograft models,tumor growth in the M27-28BBZ treatment group was suppressedcompared with that in the mock group at day 21 after tumorinoculation (14 days after T-cell injection; Fig. 5A and B). In bothmodels, some mice treated with M27-28BBZ CAR T cells survivedfor greater than 80 days (U251-EGFR/EGFRvIII model: 2/6 mice,U251-EGFRvIII model: 1/6 mice). In M27-28BBZ CAR T cell–treatment groups, median survival time was increased from 26 to60 days in the mice with U251-EGFRvIII xenografts and changedfrom 24 to 56 days in the mice with U251-EGFR/EGFRvIII xeno-grafts comparedwithmice receivingmock T cells (P < 0.05; Fig. 5C).

In vivo persistence of M27-28BBZ CAR T cellsTo evaluate CAR T-cell persistence in vivo, tumor-bearing mice

were euthanized at day 15 after intracranial tumor implantation

Figure 4.

In vivo antitumor activities of M27-28BBZ CAR T cells on established subcutaneous EGFR- or EGFRvIII-overexpressing glioblastoma xenografts. A, NOD/SCIDmice were subcutaneously inoculated with 2 � 106 U251-EGFRvIII cells on day 0. Two doses of 1 � 107 M27-28BBZ CAR T cells or mock T cells wereintravenously administered on days 12 and 15. Data are presented as the mean tumor volume � SEM. Statistically significant differences betweenM27-28BBZ CAR T cells and mock T cells are marked with asterisks (�� , P < 0.001; two-way ANOVA with Bonferroni posttest). B, On day 42 after tumorcell inoculation, the mice were euthanized. Tumor weight was measured. Endpoint was defined by the tumors reaching a mean volume of 2,000 mm3,more than 20% body weight loss in mice, tumor ulceration, or inability of the mice to ambulate. Efficacy was evaluated by measuring the reduction ofthe mean tumor weight relative to mock T cells (�� , P < 0.001; two-tailed Student t test). C, NOD/SCID mice were subcutaneously inoculated with3 � 106 U251-EGFR cells. When the tumor volume was approximately 100–120 mm3, mice bearing established subcutaneous tumors were treated withtwo doses of 1 � 107 M27-28BBZ CAR T cells or mock T cells on days 14 and 17. Data are presented as the mean tumor volume � SEM (� , P < 0.05; two-wayANOVA with Bonferroni posttest). D, On day 49 after tumor cell inoculation, the mice were euthanized. Tumor weight was measured. Efficacy wasevaluated by measuring the reduction of the mean tumor weight relative to mock control (� , P < 0.05; two-tailed Student t test).






(8 days after injection of CAR T cells). The persistence of CART cells in the murine peripheral blood and the infiltration ofCAR T cells into brain tissues were determined in the orthotopicglioblastoma xenograft models. Flow-cytometric analysis show-ed a significant increase (P<0.05) of bothCD4þ andCD8þT cells,with a predominance of CD8þ T cells, in the M27-28BBZgroup compared with the mock group in all three models(Fig. 6A). More T cells in the EGFR/EGFRvIII heterogeneitymodel persisted than in EGFRvIII and EGFR groups. The infiltra-

tion of human T cells was determined in the tumor-bearingmouse brain tissues by immunostaining with antibody to humanCD3 (Fig. 6B–D). Human CD3þ cells robustly infiltrated residualtumors after M27-28BBZ CAR T-cell infusion. In contrast, veryfew T cells could be detected in tumor lesions after mockT-cell treatment (Fig. 6B–D). More infiltrating T cells werefound in EGFRvIII-overexpressing tumor lesions than in EGFR-overexpressing tumors (U251-EGFRvIII vs. U251-EGFR: 101.8 �19.9 vs. 62.1 � 20.7 CD3þ cells/mm2, P < 0.01; Fig. 6E).

Figure 5.

Antitumor activities of M27-28BBZ T cells on established intracranial glioblastoma xenografts in vivo. For an orthotopic model of glioblastoma, 1 � 106

U251-luci-EGFR cells were implanted intracranially into NOD/SCID mice (n ¼ 8 mice per group). For U251-luci-EGFR/U251-luci-EGFRvIII models,7.5 � 105 U251-luci-EGFR and 2.5 � 105 U251-luci-EGFRvIII cells were implanted intracranially into mice (n ¼ 7 mice per group). For U251-luci-EGFRvIIImodels, 5 � 105 U251-luci-EGFRvIII cells were implanted intracranially into mice on day 0 (n ¼ 6 mice per group). On day 7, mice were injectedintravenously with 1 � 107 M27 CAR T cells or mock T cells. A, Representative IVIS images depicting 6–7 of the mice from each of the groups treatedwith the indicated CAR T cells. Mice were imaged weekly. Endpoint was defined by mice inability to ambulate. B, Tumor burden was quantified astotal flux in units of photons/second. Bars indicate means � SE (� , P < 0.05; �� , P < 0.01; �� , P < 0.001; two-way ANOVA with Bonferroni posttest). C, Survivalwas plotted using a Kaplan–Meier curve; statistically significant differences between the experimental groups were determined using log-rankMantel–Cox test (� , P < 0.05).

Jiang et al.





These data suggested that the injected M27-28BBZ CAR T cellsinfiltrated antigen-expressing brain tumors in vivo and wereamplified and maintained.

DiscussionMounting evidence indicates that intratumoral hetero-

geneity may be a cause of drug resistance and tumor recurrence

Figure 6.

The persistence of human T cells from mice treated with the indicated genetically modified T cells in orthotopic glioblastoma xenografts models. A, Thequantification of CD4þ and CD8þ T cells in murine peripheral blood of intracranial tumor-bearing mice on day 15. Fifty microliters of murine bloodwas drawn at 8 days after indicated T-cell infusion to determine T-cell counts. T cells were analyzed by flow cytometry using anti-CD3-PerCP/anti-CD4-FITC/anti-CD8-PE and were enumerated using the count beads. Data indicate means � SEM number of T cells per microliter of blood asmeasured by flow cytometry. Statistically significant differences were calculated by the two-tailed Student t test (� , P < 0.05). B–D, The representativeimmunostaining images of CD3þ T-cell infiltration in tumor-bearing mice brains. The brain tissues were harvested from intracranial tumor-bearingmice on day 15. Formalin-fixed, paraffin-embedded mice brain tissue sections were consecutively cut and stained for human CD3 (brown). [Theimages were taken with the microscope (BX41, Olympus) and camera (DP70) under � 200 magnifications. Scale bar, 100 mm]. E, The quantificationof T-cell infiltration in brain tissues of intracranial tumor-bearing mice treated with M27-28BBZ and mock T cells. Data are expressed as mean � SEM(� , P < 0.05; �� , P < 0.01; (��); �� , P < 0.001; two-tailed Student t test).






in cancer treatment (31). Glioblastoma is a highly heteroge-neous tumor (32, 33) and one of the first cancers profiled inThe Cancer Genome Atlas project (34). Genetic alterationsincluding mutations, rearrangements, alternative splicing, andfocal amplifications of EGFR have been found in 57% ofglioblastomas (35). In addition, EGFR amplification and over-expression has been reported in approximately 50% to 60%of glioblastomas (36, 37). EGFR variants including EGFRvIIIand de4 EGFR (38, 39) could increase proliferation andinvasiveness of glioblastoma cells. One study demonstratedthat EGFRvIII is invariably accompanied by EGFR amplifica-tion, although EGFR overexpression does not necessarily giverise to the EGFRvIII variant (23, 24). This partially explainswhy EGFRvIII-specific therapy was unable to completely elim-inate glioblastoma. Another study showed that EGFRvIIIexpression was lost in 80% of relapsed tumors followingvaccination with PEPvIII-KLH in Freund's complete adjuvant,suggesting that the EGFRvIII-negative cell population predo-minated at the time of tumor recurrence (40). Additionally,neither partial nor complete responses were observed ina total of 10 glioblastoma patients treated with EGFRvIII-specific CAR T cells, although EGFRvIII antigen completelydisappeared in 2 patients after CAR T-cell infusion (18).This paradoxical clinical readout could be ascribed to hetero-geneity of EGFRvIII expression. To overcome the hetero-geneity of EGFRvIII, CAR-NK cells that could recognize bothEGFR and EGFRvIII were developed (41, 42). However, theantibodies used in the CARs, such as C225, could bind tonormal EGFR, which is expressed in normal brain tissues(39, 43). Thus, these CAR-NK cells may induce on-target,off-tumor toxicities in patients with glioblastoma even treatedby locally administration.

Severe toxicities including death have been reported in CART-cell immunotherapy due to on-target, off-tumor toxicities(44, 45). Thus, using cancer-selective antibodies as the targetingmoiety of the CARs may enhance the safety of this therapy.MAb 806, which binds to the EGFR287–302 epitope, can bind toEGFR and to EGFRvIII overexpressed in cancer cells, but not toEGFR in normal cells. Unexpectedly, our study revealed that806 scFv could bind to primary keratinocytes, although with amuch lower binding capacity than cetuximab. At doses up to24 mg/kg, ABT-806 (25) only induced mild dermatitis acnei-form in 12% (6/49) of the patients and one grade 3 morbilli-form rash. Even more promising is that ABT-414, at a dose of1.25 mg/kg, showed no skin toxicity and induced an objectiveresponse rate of 6.8% in the 66 patients treated (26). Incombination with radiation and temozolomide, ABT-414 waswell tolerated and showed promising clinical benefit in pati-ents with newly diagnosed glioblastoma (46). These studiessupport using antibodies targeting the EGFR287-302 epitope asan immunotherapeutic strategy for patients with glioblastoma.However, the study of ABT-414 had ocular grade 3/4 adverseevents in 33% of patients overall, which may be ascribed tothe accumulation of the toxic MMAF. Another considerationwith antibody–drug conjugates is that they need internalizationthrough receptor-mediated endocytosis. Its antitumor activ-ities depend on the binding avidity of the antibody and theendocytosis capacity of the receptor. Unlike antibody–drugconjugates, CAR T cells do not possess the same compound-associated toxicities and are capable of killing cancer cellsindependent of receptor-mediated endocytosis. A single CAR

T-cell may express thousands of CAR copies, which mayincrease its binding avidity on target cells and maybe the reasonwhy CAR T cells comprising 806 scFv, a low-affinity EGFRbinder, are capable of lysing normal keratinocytes in vitro. Inthis study, we identified a humanized scFv M27, which did notbind to normal cells, but could bind to both EGFR andEGFRvIII in tumor cells at a relatively lower affinity than806. Like previous reports that EGFR antibodies could be tunedto low affinity to reduce CAR T cell–associated toxicityagainst normal cells (47, 48), M27 CAR T cells showed notoxicity against normal cells while retaining their tumor lysiscapabilities. Additionally, M27-28BBZ CAR T cells have rela-tively higher cytotoxicity against cancer cells with EGFR- orEGFRvIII-overexpression, compared with that of M27-28Z CART cells. These data indicated that M27-28BBZ CAR T cellsdemonstrated effective antitumor activity with minimal clini-cally relevant on-target, off-tumor toxicity.

In intracranial tumor xenograft models, M27-28BBZ CART cells efficiently suppressed the growth of EGFR- andEGFRvIII-overexpressing tumors and increased survival ofthe mice. Its antitumor activity against EGFRvIII tumorsappeared more potent than those against EGFR tumors. Itmight be explained by higher binding affinity of M27 scFv toEGFRvIII than to EGFR. The xenografts in the EGFR/EGFRvIIImixture group were eradicated by M27-28BBZ CAR T cells,and this may be ascribed to the enhanced amplificationor persistence of M27-28BBZ CAR T cells in the presence ofEGFRvIII-positive tumor cells. T-cell numbers in both bloodand tumor tissues in the EGFR/EGFRvIII group were higherthan those in the EGFR group, and increased T-cell infiltrationand persistence might facilitate the elimination of tumorcells with EGFR overexpression.

Given their distinct tumor specificity and antitumor activ-ities, M27-28BBZ CAR T cells represent a new bitargetedantitumor agent for the treatment of patients with glioblas-toma. Additionally, as many other tumor types, includingnon–small cell lung cancer, breast cancer, and head and necksquamous carcinoma, are characterized by EGFR amplifica-tion and/or EGFRvIII mutations (23, 49, 50), M27-28BBZCAR T cells also have the potential for the treatment of thesecancer types.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: Z. LiDevelopment of methodology: H. Jiang, H. Gao, J. Kong, B. SongAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): P. Wang, B. Shi, H. WangAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): H. Jiang, H. GaoWriting, review, and/or revision of the manuscript: H. Jiang, Z. LiAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): H. WangStudy supervision: Z. Li

AcknowledgmentsThis work was supported by funding from the Supporting Programs

of Shanghai Science and Technology Innovation Action Plan (No.18431902900), Shanghai Municipal Commission of Health and FamilyPlanning (No. 201740124), the National Natural Science Foundation ofChina (No. 81672724 and No. 81472573), a grant from the State Key

Jiang et al.





Laboratory of Oncogenes and Related Genes (No. 91-17-17), and the seedfund of Renji Hospital (No. RJZZ17-021).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received January 30, 2018; revised June 21, 2018; accepted August 28, 2018;published first September 10, 2018.

References1. Bhat KP, Salazar KL, Balasubramaniyan V, Wani K, Heathcock L,

Hollingsworth F, et al. The transcriptional coactivator TAZ regulatesmesenchymal differentiation in malignant glioma. Genes Dev 2011;25:2594–609.

2. Paugh BS, Qu C, Jones C, Liu Z, Adamowicz-Brice M, Zhang J, et al.Integrated molecular genetic profiling of pediatric high-grade gliomasreveals key differences with the adult disease. J Clin Oncol 2010;28:3061–8.

3. Wen PY, Kesari S. Malignant gliomas in adults. N Engl J Med 2008;359:492–507.

4. Lefranc F, Sadeghi N, Camby I, Metens T, Dewitte O, Kiss R. Present andpotential future issues in glioblastoma treatment. Expert Rev Anticancer2006;6:719–32.

5. Louis DN, Ohgaki H, Wiestler OD, CaveneeWK, Burger PC, Jouvet A, et al.The 2007 WHO classification of tumours of the central nervous system.Acta Neuropathol 2007;114:97–109.

6. Bonnet D, Dick JE. Human acute myeloid leukemia is organized as ahierarchy that originates from a primitive hematopoietic cell. Nat Med1997;3:730–7.

7. Couzin-Frankel J. Breakthrough of the year 2013. Cancer immunotherapy.Science 2013;342:1432–3.

8. Ahmed N, Brawley V, Hegde M, Bielamowicz K, Kalra M, Landi D, et al.HER2-specific chimeric antigen receptor-modified virus-specific T cellsfor progressive glioblastoma: A Phase 1 Dose-Escalation Trial. JAMAOncol 2017;3:1094–101.

9. Ahmed N, Salsman VS, Kew Y, Shaffer D, Powell S, Zhang YJ, et al.HER2-specific T cells target primary glioblastoma stem cells andinduce regression of autologous experimental tumors. Clin Cancer Res2010;16:474–85.

10. Brown CE, Alizadeh D, Starr R, Weng L, Wagner JR, Naranjo A, et al.Regression of glioblastoma after chimeric antigen receptor T-celltherapy. N Engl J Med 2016;375:2561–9.

11. Kahlon KS, Brown C, Cooper LJ, Raubitschek A, Forman SJ, Jensen MC.Specific recognition and killing of glioblastomamultiforme by interleukin13-zetakine redirected cytolytic T cells. Cancer Res 2004;64:9160–6.

12. Heimberger AB, Hlatky R, Suki D, Yang D, Weinberg J, Gilbert M, et al.Prognostic effect of epidermal growth factor receptor and EGFRvIII inglioblastoma multiforme patients. Clin Cancer Res 2005;11:1462–6.

13. Saikali S, Avril T, Collet B,Hamlat A, Bansard JY,DrenouB, et al. Expressionof nine tumour antigens in a series of human glioblastoma multiforme:interest of EGFRvIII, IL-13Ralpha2, gp100 and TRP-2 for immunotherapy.J Neurooncol 2007;81:139–48.

14. Johnson LA, Sampson JH. Immunotherapy approaches for malig-nant glioma from 2007 to 2009. Curr Neurol Neurosci Rep 2010;10:259–66.

15. Learn CA, Hartzell TL, Wikstrand CJ, Archer GE, Rich JN, FriedmanAH, et al. Resistance to tyrosine kinase inhibition by mutant epi-dermal growth factor receptor variant III contributes to the neoplas-tic phenotype of glioblastoma multiforme. Clin Cancer Res 2004;10:3216–24.

16. Morgan RA, Johnson LA, Davis JL, Zheng Z, Woolard KD, Reap EA, et al.Recognition of glioma stem cells by genetically modified T cells targetingEGFRvIII and development of adoptive cell therapy for glioma. HumGene Ther 2012;23:1043–53.

17. Sampson JH,Choi BD, Sanchez-Perez L, SuryadevaraCM, SnyderDJ, FloresCT, et al. EGFRvIII mCAR-modified T-cell therapy cures mice with estab-lished intracerebral glioma and generates host immunity against tumor-antigen loss. Clin Cancer Res 2014;20:972–84.

18. O'Rourke DM, Nasrallah MP, Desai A, Melenhorst JJ, Mansfield K,Morrissette JJD, et al. A single dose of peripherally infused EGFRvIII-directed CAR T cells mediates antigen loss and induces adaptive resistancein patients with recurrent glioblastoma. Sci Transl Med 2017;9:eaaa0984.

19. Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E,et al. Intratumor heterogeneity and branched evolution revealed by multi-region sequencing. N Engl J Med 2012;366:883–92.

20. Yachida S, Jones S, Bozic I, Antal T, Leary R, Fu B, et al. Distant metastasisoccurs late during the genetic evolution of pancreatic cancer. Nature2010;467:1114–7.

21. Ding H, Shannon P, Lau N, Wu X, Roncari L, Baldwin RL, et al. Oligoden-drogliomas result from the expression of an activated mutant epidermalgrowth factor receptor in a RAS transgenic mouse astrocytoma model.Cancer Res 2003;63:1106–13.

22. Wikstrand CJ, Hale LP, Batra SK, Hill ML, Humphrey PA, Kurpad SN, et al.Monoclonal antibodies against EGFRvIII are tumor specific and reactwith breast and lung carcinomas and malignant gliomas. Cancer Res1995;55:3140–8.

23. Gan HK, Burgess AW, Clayton AH, Scott AM. Targeting of a conforma-tionally exposed, tumor-specific epitope of EGFR as a strategy for cancertherapy. Cancer Res 2012;72:2924–30.

24. Scott AM, Lee FT, Tebbutt N, Herbertson R, Gill SS, Liu Z, et al. A phase Iclinical trial with monoclonal antibody ch806 targeting transitional stateand mutant epidermal growth factor receptors. Proc Natl Acad Sci U S A2007;104:4071–6.

25. Cleary JM, Reardon DA, Azad N, Gandhi L, Shapiro GI, Chaves J, et al. Aphase 1 study of ABT-806 in subjects with advanced solid tumors. InvestNew Drugs 2015;33:671–8.

26. Van den Bent M, Gan HK, Lassman AB, Kumthekar P, Merrell R, ButowskiN, et al. Efficacy of depatuxizumabmafodotin (ABT-414) monotherapy inpatients with EGFR-amplified, recurrent glioblastoma: results from amulti-center, international study. Cancer Chemother Pharmacol 2017;80:1209–17.

27. Gao H, Li K, Tu H, Pan X, Jiang H, Shi B, et al. Development of T cellsredirected to glypican-3 for the treatment of hepatocellular carcinoma.Clin Cancer Res 2014;20:6418–28.

28. Pinthus JH, Waks T, Malina V, Kaufman-Francis K, Harmelin A,Aizenberg I, et al. Adoptive immunotherapy of prostate cancer bonelesions using redirected effector lymphocytes. J Clin Invest 2004;114:1774–81.

29. Maus MV, Haas AR, Beatty GL, Albelda SM, Levine BL, Liu X, et al. T cellsexpressing chimeric antigen receptors can cause anaphylaxis in humans.Cancer Immunol Res 2013;1:26–31.

30. Shinojima N, Tada K, Shiraishi S, Kamiryo T, Kochi M, NakamuraH, et al. Prognostic value of epidermal growth factor receptorin patients with glioblastoma multiforme. Cancer Res 2003;63:6962–70.

31. Maley CC, Galipeau PC, Finley JC,Wongsurawat VJ, Li X, Sanchez CA, et al.Genetic clonal diversity predicts progression to esophageal adenocarcino-ma. Nat Genet 2006;38:468–73.

32. Furnari FB, Cloughesy TF, Cavenee WK, Mischel PS. Heterogeneity ofepidermal growth factor receptor signalling networks in glioblastoma.Nat Rev Cancer 2015;15:302–10.

33. Nathanson DA, Gini B, Mottahedeh J, Visnyei K, Koga T, Gomez G, et al.Targeted therapy resistance mediated by dynamic regulation of extrachro-mosomal mutant EGFR DNA. Science 2014;343:72–6.

34. Cancer Genome Atlas Research Network. Comprehensive genomic char-acterization defines human glioblastoma genes and core pathways. Nature2008;455:1061–8.

35. Brennan CW, Verhaak RG, McKenna A, Campos B, Noushmehr H, SalamaSR, et al. The somatic genomic landscape of glioblastoma. Cell 2013;155:462–77.

36. Libermann TA, Nusbaum HR, Razon N, Kris R, Lax I, Soreq H, et al.Amplification, enhanced expression and possible rearrangement of EGFreceptor gene in primary human brain tumours of glial origin. Nature1985;313:144–7.






37. Pelloski CE, Ballman KV, Furth AF, Zhang L, Lin E, Sulman EP, et al.Epidermal growth factor receptor variant III status defines clinicallydistinct subtypes of glioblastoma. J Clin Oncol 2007;25:2288–94.

38. Wang H, ZhouM, Shi B, Zhang Q, Jiang H, Sun Y, et al. Identification of anexon 4-deletion variant of epidermal growth factor receptor with increasedmetastasis-promoting capacity. Neoplasia 2011;13:461–71.

39. Zhou M, Wang H, Zhou K, Luo X, Pan X, Shi B, et al. A novel EGFRisoform confers increased invasiveness to cancer cells. Cancer Res2013;73:7056–67.

40. Heimberger AB, Archer GE, Crotty LE, McLendon RE, Friedman AH,Friedman HS, et al. Dendritic cells pulsed with a tumor-specificpeptide induce long-lasting immunity and are effective againstmurine intracerebral melanoma. Neurosurgery 2002;50:158–64;discussion 64–6.

41. Genssler S, Burger MC, Zhang C, Oelsner S, Mildenberger I, Wagner M,et al. Dual targeting of glioblastoma with chimeric antigen receptor-engineered natural killer cells overcomes heterogeneity of targetantigen expression and enhances antitumor activity and survival.Oncoimmunology 2016;5:e1119354.

42. Han J, Chu J, Keung Chan W, Zhang J, Wang Y, Cohen JB, et al. CAR-engineeredNK cells targetingwild-type EGFRandEGFRvIII enhance killingof glioblastoma and patient-derived glioblastoma stem cells. Sci Rep2015;5:11483.

43. Siddiqui S, Fang M, Ni B, Lu D, Martin B, Maudsley S. Central role of theEGF receptor in neurometabolic aging. Int J Endocrinol 2012;2012:739428.

44. Lamers CH, Sleijfer S, Vulto AG, Kruit WH, Kliffen M, Debets R, et al.Treatment of metastatic renal cell carcinoma with autologous T-lympho-cytes genetically retargeted against carbonic anhydrase IX: first clinicalexperience. J Clin Oncol 2006;24:e20–2.

45. Morgan RA, Yang JC, Kitano M, Dudley ME, Laurencot CM, Rosenberg SA.Case report of a serious adverse event following the administration ofT cells transduced with a chimeric antigen receptor recognizing ERBB2.Mol Ther 2010;18:843–51.

46. Reardon DA, Lassman AB, van den Bent M, Kumthekar P, Merrell R, ScottAM, et al. Efficacy and safety results of ABT-414 in combinationwith radiation and temozolomide in newly diagnosed glioblastoma.Neuro Oncol 2017;19:965–75.

47. Caruso HG, Hurton LV, Najjar A, Rushworth D, Ang S, Olivares S, et al.Tuning sensitivity of CAR to EGFR density limits recognition of normaltissue while maintaining potent antitumor activity. Cancer Res 2015;75:3505–18.

48. Liu X, Jiang S, Fang C, Yang S, Olalere D, Pequignot EC, et al. Affinity-tunedErbB2 or EGFR chimeric antigen receptor T cells exhibit an increasedtherapeutic index against tumors in mice. Cancer Res 2015;75:3596–607.

49. JungbluthAA, Stockert E,HuangHJ, Collins VP,CoplanK, IversenK, et al. Amonoclonal antibody recognizing human cancers with amplification/overexpression of the human epidermal growth factor receptor. Proc NatlAcad Sci U S A 2003;100:639–44.

50. Reis-Filho JS, Pinheiro C, Lambros MB, Milanezi F, Carvalho S, Savage K,et al. EGFR amplification and lack of activating mutations in metaplasticbreast carcinomas. J Pathol 2006;209:445–53.


Jiang et al.




2018;6:1314-1326. Published OnlineFirst September 10, 2018.Cancer Immunol Res Hua Jiang, Huiping Gao, Juan Kong, et al. Chimeric Antigen Receptor T CellSelective Targeting of Glioblastoma with EGFRvIII/EGFR Bitargeted

Updated version

10.1158/2326-6066.CIR-18-0044doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerimmunolres.aacrjournals.org/content/suppl/2018/09/08/2326-6066.CIR-18-0044.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerimmunolres.aacrjournals.org/content/6/11/1314.full#ref-list-1

This article cites 50 articles, 22 of which you can access for free at:

Citing articles

http://cancerimmunolres.aacrjournals.org/content/6/11/1314.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerimmunolres.aacrjournals.org/content/6/11/1314To request permission to re-use all or part of this article, use this link



http://cancerimmunolres.aacrjournals.org/lookup/doi/10.1158/2326-6066.CIR-18-0044

http://cancerimmunolres.aacrjournals.org/content/suppl/2018/09/08/2326-6066.CIR-18-0044.DC1

http://cancerimmunolres.aacrjournals.org/content/6/11/1314.full#ref-list-1

http://cancerimmunolres.aacrjournals.org/content/6/11/1314.full#related-urls

http://cancerimmunolres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerimmunolres.aacrjournals.org/content/6/11/1314


Documents

Selective Targeting of Glioblastoma with EGFRvIII/EGFR … · Research Article Selective Targeting of Glioblastoma with EGFRvIII/EGFR Bitargeted Chimeric Antigen Receptor T Cell Hua