The EGFR and TGFβ signaling pathways are important mediators of tumorigenesis, and cross-talk between them contributes to cancer progression and drug resistance. Therapies capable of simultaneously targeting EGFR and TGFβ could help improve patient outcomes across various cancer types. Here, we developed BCA101, an anti-EGFR IgG1 mAb linked to an extracellular domain of human TGFβRII. The TGFβ “trap” fused to the light chain in BCA101 did not sterically interfere with its ability to bind EGFR, inhibit cell proliferation, or mediate antibody-dependent cellular cytotoxicity. Functional neutralization of TGFβ by BCA101 was demonstrated by several in vitro assays. BCA101 increased production of proinflammatory cytokines and key markers associated with T-cell and natural killer–cell activation, while suppressing VEGF secretion. In addition, BCA101 inhibited differentiation of naïve CD4+ T cells to inducible regulatory T cells (iTreg) more strongly than the anti-EGFR antibody cetuximab. BCA101 localized to tumor tissues in xenograft mouse models with comparable kinetics to cetuximab, both having better tumor tissue retention over TGFβ “trap.” TGFβ in tumors was neutralized by approximately 90% in animals dosed with 10 mg/kg of BCA101 compared with 54% in animals dosed with equimolar TGFβRII-Fc. In patient-derived xenograft mouse models of head and neck squamous cell carcinoma, BCA101 showed durable response after dose cessation. The combination of BCA101 and anti-PD1 antibody improved tumor inhibition in both B16-hEGFR–expressing syngeneic mouse models and in humanized HuNOG-EXL mice bearing human PC-3 xenografts. Together, these results support the clinical development of BCA101 as a monotherapy and in combination with immune checkpoint therapy.

Significance:

The bifunctional mAb fusion design of BCA101 targets it to the tumor microenvironment where it inhibits EGFR and neutralizes TGFβ to induce immune activation and to suppress tumor growth.

Since the 1990s, multiple therapeutic molecules involving targeted therapy have been developed and many of them were mAbs as they are naturally stable molecules, highly specific with longer half-lives (1). However, the therapeutic potential of mAbs remains to be fully realized. For example, while Herceptin, which targets HER2-overexpressing breast cancer, has improved patient survival, anti-EGFR (which also belongs to the HER family) mAbs have had limited therapeutic benefit in the clinic (2). The suboptimal clinical efficacy of cetuximab and other EGFR-targeting therapeutics was attributed to both tumor intrinsic and tumor microenvironment (TME) derived acquired (extrinsic) resistance mechanisms (2). Intrinsic factors include activation of other receptor tyrosine kinases such as HER3, MET, and AXL (2), dysregulation of EGFR internalization/degradation, altered steady state EGFR expression, EGFR heterodimerization with HER2 and HER3 leading to transactivation (3), and upregulation of immune checkpoint receptor programmed death ligand 1 (PD-L1; ref. 4). A key extrinsic resistance mechanism to cetuximab is the enhanced secretion of immunosuppressive factors such as IL10, adenosines, and TGFβ1 (here after referred as TGFβ unless otherwise stated; ref. 5).

The biology of TGFβ signaling is complex and activation of the pathway has pleiotropic effects in the context of oncology and fibrosis. In cancer, stimulation of the TGFβ pathway can lead to protumoral or antitumoral effects, depending on the stage of the disease as well as the role of other activated oncogenic pathways, known as “the TGFβ paradox” (6). TGFβ is normally found in an inactive state, as a latent complex in tissues and on platelets, and is activated transiently and locally within the TME. Elevated levels of TGFβ within the TME enable tumor cells to evade antibody-dependent cellular cytotoxicity (ADCC) and resist the antitumor activity of cetuximab in vivo (5). It has also been reported that in patients with head and neck squamous cell carcinoma (HNSCC), the expansion of induced regulatory T cells (iTreg) mediated by extrinsic TGFβ correlated with resistance to cetuximab therapy (4). These iTregs suppressed natural killer (NK)–cell activation and cytotoxicity, leading to immunosuppression within the TME (7). Furthermore, cancer-associated fibroblast (CAF) secrete TGFβ, which in turn regulates the tumor cell invasion and metastasis by inducing mesenchymal markers such as vimentin and N-cadherin by a process known as epithelial-to-mesenchymal transition (EMT). EMT was proposed as one of the mechanisms of resistance to EGFR-targeting therapeutics (8).

Although signaling mechanisms triggered by EGFR and TGFβ are different, the cross-talk between EGFR and TGFβ in cancer progression is well documented. For instance, TGFβ synergizes with EGFR signaling to promote tumor invasiveness and therapy resistance by enhancing the EMT process (6). Importantly, blockade of EGFR with cetuximab leads to increased TGFβ secretion, EMT and CAF formation, which in turn contribute to cetuximab resistance (9). TGFβ and EGF synergistically enhanced EMT phenotype in multiple cancer types, such as oral squamous cancer (10), intestinal epithelial cells (11) and ovarian cancer cells (12). Furthermore, simultaneous inhibition of EGFR and TGFβ1 enhanced the efficacy of cetuximab in head and neck cancer (5), providing a strong rationale for developing a bispecific antibody capable of simultaneously inhibiting EGFR and neutralizing TGFβ.

Here, we describe a first-in-class bifunctional mAb fusion designed to simultaneously target EGFR and sequester TGFβ in the TME. We show that BCA101 has an advantage over cetuximab in activating the immune system known to be suppressed by the presence of TGFβ (13). We also show that BCA101 and thus the TGFβ trap is retained longer in the tumor site in a murine tumor xenograft model compared with TGFβ trap alone, which could improve the efficacy and therapeutic window for BCA101. Proof of principle is further established in combination with an immune checkpoint therapy (anti-PD-1 mAb) in a B16-hEGFR syngeneic mouse tumor model and in HuNOG-EXL humanized PC-3 xenograft model, where the combination with BCA101 showed superiority over BCA101 or anti PD-1 alone. Taken together, our data suggest that BCA101 may exert superior antitumor effects by virtue of improved and thus sustained TGFβ neutralization at the tumor site, while leveraging cotargeting of EGFR and TGFβ signaling pathways to inhibit EMT and improve antitumor immune responses.

BCA101 computational modeling

The BCA101 computational models were developed with software Molecular Operating Environment (MOE) from Chemical Computing Group. The structure for Fab domain was adopted from 1YY8.PDB and a full-length antibody model was developed using the automatic modeling features of the program. The structure of extracellular domain (ECD) of TGFβRII in complex with TGFβ was adopted from 1KTZ.PDB. Structure of full-length antibody and TGFβRII–TGFβ complex was prepared separately as per the automatic feature of the program. Then the two structures were added in the same window and TGFβRII–TGFβ complex was manually positioned such that the C-terminal of each antibody light chains (LC) is in close proximity to each of the N-terminal regions of TGFβRII. The linker regions connecting the C-terminal of LC to N-terminal region of TGFβRII as well as the missing N-terminal residues of TGFβRII were modeled in by the loop modeling feature of MOE.

Cloning and expression and SDS-PAGE

The LC fusion constructs were codon optimized for expression in Chinese hamster ovary (CHO) cells and constructs were obtained by chemical synthesis (GeneArt, Thermo Fisher Scientific). The LC genes were cloned into separate vectors, which contain strong constitutive mammalian promoter for driving the gene expression. After amplifying the plasmids using standard methods, the plasmids were transfected to CHO-S cells, subjected to selection using increasing concentrations of puromycin and methotrexate after 48 hours. ELISA was used to screen for high producer clones in 96-well plate stage and high titer clones were scaled up to 30 mL in 125 mL shake flask. Top clone was selected on the basis of titer, growth, and viability in shake flask stage. The protein was purified from the lead clone by Protein A affinity chromatography. BCA101 and cetuximab was analyzed by SDS-PAGE using the Mini-Protean TGX 4% to 20% gradient precast gel (Bio-Rad). A total of 2 μg of protein was loaded into each lane and electrophoresis was run as per the standard protocol.

ELISA

For EGFR and TGFβ ELISA, maxisorp 96-well plate (NUNC) was coated either with 2 μg/mL of recombinant human (rh) EGFR (R&D Systems) or 1 μg/mL of rh TGFβ (R&D Systems) overnight at 4°C, respectively. Next day, wells were blocked with Superblock (Thermo Fisher Scientific) for 2 hours and probed with various dilutions of drugs for 1 hour. Bound BCA101 or cetuximab (Merck) was probed with either goat anti-human IgG1 F(ab’)2 (Jackson Immuno Research) or goat anti-human IgG1 (H+L)-HRP (horseradish peroxidase) conjugated secondary antibody (Jackson Immuno Research) for 1 hour followed by 3,3′,5,5′-tetramethylbenzidine (Sigma). The reaction was stopped by 1N H2SO4 and the absorbance was measured at 450 nm with reference wavelength at 630 nm in a Biotek plate reader (Biotek). EC50 values were calculated using SoftMax Pro software (6.5 GxP)

For bifunctional ELISA, 96-well plates were coated with recombinant EGFR-Fc chimera. Coated plates were blocked, then incubated with various concentrations of BCA101 for 1 hour that allows BCA101 to bind to the EGFR Fc (R&D Systems) through the Fab region followed by incubation with recombinant TGFβ (R&D Systems), which binds to the TGFβRII-ECD moiety of BCA101. The signal was detected with biotinylated anti-TGFβ antibody (R&D Systems) followed by streptavidin-HRP.

Surface plasmon resonance (Biacore)

The CM5 sensor chip was used to immobilize sEGFR (R&D Systems) using amine coupling kit and EDC and NHS solutions were prepared as per the manufacturer's instructions (GE HealthCare). The kinetic studies of sEGFR-BCA101 or reference molecule cetuximab were carried out using the kinetic analysis wizard. HBS-EP was used as the running buffer. The experiments were carried out at 25°C with a data collection rate of 1 Hz with a constant flow rate. The running methods were created using the Biacore T200 control software. The Sensograms were evaluated using BIA evaluation T200 software.

For Fc gamma R3A receptor (FCR3G, V176F) binding, anti-his Antibody in immobilization buffer (Acetate 4.5) and immobilized on the CM5 chip. Approximately 11,500 RU of immobilization was achieved in both the flow cells. The FCR3G was diluted to 0.5 μg/mL in running Buffer (HBS-P+) and flowed for 60 seconds at 10 μL/minute flow rate over the immobilized anti-his antibody with a stabilization time of 60 seconds. Approximately 90 RU of FCGR3A was captured on Flow cell 2. Flow cell 1 was used for reference subtraction. The antibodies were serially diluted from 2,000 nmol/L to 24.7 nmol/L (3-fold dilution) in running buffer and run along with a buffer blank (0 nmol/L). Diluted antibodies were flowed over the captured ligand at 50 μL/minute flow rate for 120 seconds association followed by 360 seconds dissociation in running buffer. Analysis temperature was 25°C, followed by regeneration of the surface with 90 seconds pulse of 10 mmol/L Glycine HCl pH 1.5 flowed at 30 μL/minute flow rate followed by stabilization of the surface with 60 seconds flow of running buffer. The curves obtained were fitted to 1:1 binding model by setting RI as local parameter.

EGFR binding by flow cytometry

FaDu cells (ATCC) were trypsinized and pellet was resuspended in Eagle Minimum Essential Medium (EMEM; ATCC) with 10% FBS (30-2020, ATCC) and 1% pen strep (ATCC) media. About 0.1 × 106 cells were washed with PBS (SIGMA) at 1,200 rpm, 5 minutes at 4°C and incubated with primary antibodies (BCA101 and cetuximab) for 30 minutes on ice. After incubation, cells were washed with PBS and incubated with florescence-labeled secondary antibodies (SIGMA) for 30 minutes on ice. Tubes were acquired in BD FACS Calibur flow cytometry (BD Biosciences). Data were analyzed by FACS Suite software.

Inhibition of proliferation assay

FaDu cells (ATCC) were trypsinized and resuspended in assay media (EMEM, ATCC) with 1% pen-strep (ATCC) and 1% FBS (ATCC). Drug dilutions were prepared as required and added (80 μL) to a 96-well plate followed by addition of about 5,000 FaDu cells (80 μL) to the respective wells. Plates were incubated at 37°C, 5% CO2 incubator for 72 hours. After incubation, 80 μL of cell media was replaced with 80 μL of CellTiter-Glo 2.0 (Promega) and incubated for 15 minutes at room temperature. Luminescence readout was taken as per manufacturer's instructions using a plate reader (BioTek Synergy H4). Data were analyzed by SoftMax Pro software (6.5 GxP). Anti-VEGF antibody was used as an IgG or negative control.

ADCC reporter assay

The ADCC Reporter Bioassay (Promega) was used for ADCC assay. This assay uses engineered Jurkat cells stably expressing the FcγRIIIa receptor, V158 (high affinity) variant and luciferase reporter that is linked to a promoter responsive to nuclear factor of activated T cells (NFAT). Upon binding of Fc portion of an IgG1, FcγRIIIa activated NFAT, which results in expression of luciferase.

About 1.2 × 104 FaDu cells were seeded in a 96-well white plate (Corning) and incubated overnight. Drug dilutions (3X) and effector ABEC cells were prepared in assay media [RPMI (ATCC) containing 0.5% FBS]. Media was discarded and replenished with 25 μL assay media followed by 25 μL drug diluted media and 25 μL of effector cells (T:E = 1:15) and incubated at 37°C, 5% CO2 incubator for 24 hours. Following incubation, Bio-Glo (Promega) luciferase reagent was added to the wells and incubated for 15 minutes at room temperature. Luminescence readout was taken as per manufacturer's instructions using a plate reader (BioTek Synergy H4). Data were analyzed by SoftMax Pro software. Anti-VEGF antibody was used as an IgG or negative control.

NK cell–based ADCC assay

ADCC of BCA101 was assessed using TGFβ pretreated NK cells and A431 (EGFRhigh) or MCF7 (EGFRlow) cell lines, procured from ATCC. NK cells were isolated from frozen or fresh human peripheral blood mononuclear cells (PBMC) using Human NK-cell negative selection kit (Stemcell Technologies Inc.) and EasySep magnet (Stemcell Technologies Inc.). About 1 million enriched NK cells were left untreated or cultured with 1 ng/mL of TGFβ (BPS Biosciences), TGFβ plus cetuximab or BCA101 (56 nmol/L) for 72 hours in a 12-well plate in NK MACS (Miltenyi Biotec) media containing 1% pen strep (GIBCO, Thermo Fisher Scientific), 1% heat-inactivated human serum (Sigma-Aldrich), and 1% NK supplement (Miltenyi Biotec). These pretreated NK cells were used as effectors in ADCC assay. BCA101 (56 nmol/L) was added to target tumor cells and preincubated for 1–2 hours, after which, pretreated NK cells were added at a ratio of 1:5 (5,000 target cells: 25,000 NK cells) in a 96-well plate. The assay plates were incubated for around 24 hours. Cytotoxicity was measured by evaluating protease release using CytoToxGlo (Promega). Luminescence readout was taken using a plate reader (Cytation5 BioTek plate reader) and data were analyzed using GraphPad Prism software (version9). Data are plotted as percentage change in cytotoxicity over human IgG control group.

TGFβ-SMAD reporter assay

TGFβ SMAD reporter assay was performed as per the manufacturer's instructions (BPS Biosciences, catalog no. 60653). Briefly, HEK293 cells were plated at a density of approximately 35,000 cells per well on to 96-well white clear-bottom plates in growth media containing MEM/EBSS (HyClone Laboratories Inc., GE Healthcare Lifesciences) with 1% non-essential amino acids (HyClone Laboratories Inc., GE Healthcare Lifesciences), 1 mmol/L sodium pyruvate (HyClone Laboratories Inc., GE Healthcare Lifesciences), 1% pen-strep (GIBCO, Thermo Fisher Scientific), and 10% FBS (MP Bio Science Ltd) and allowed to attach overnight. Spent media was discarded and cells were then incubated with recombinant TGFβ (BPS Biosciences) at 20 ng/mL and with different dilutions of BCA101, human IgG isotype antibody (R&D Systems Inc.) or TGFβRII-Fc at equimolar concentration prepared in assay media (growth media containing 0.5% FBS). After 18 hours, cells were incubated with 100 μL of BioGlo luciferase reagent for 15 minutes at room temperature. Luminescence readout was taken as per manufacturer's instructions using a Spectramax M5e plate reader (Molecular Devices) with SoftMax Pro GxP software version 6.5. EC50 was analyzed using SoftMax Pro software and data plotted using GraphPad Prism software version 9.

PBMC activation assay

Frozen human PBMCs, procured from Precision Bioservices, were thawed as per vendor recommended protocol in assay media containing RPMI1640 media (GIBCO, Thermo Fisher Scientific), 10% FBS and 1% pen-strep. Next day, 0.1 × 106 cells/well of PBMCs were plated onto a 96-well white clear-bottom plate and stimulated with 2 ng/mL soluble anti-CD3 (OKT3 clone, eBioscience Inc.) and 10 ng/mL soluble anti-CD28 (eBioscience Inc.) in presence or absence of 10 ng/mL TGFβ (R&D Systems Inc.), followed by addition of 56 nmol/L drugs. The plate was incubated at 37, 5% CO2 incubator for approximately 72 hours. At the end of assay incubation, the cell culture supernatant was collected, and the cells were further processed for surface marker staining. Cells were collected in FACS tubes (BD Biosciences), centrifuged and resuspended in 100 μL of prepared antibody cocktail/FMO control/single color control and incubated at 4°C for 45 minutes in dark. The antibodies BV421 anti-human CD3, APC anti-human NKG2D, FITC anti-human CD8, and aminoactinomycin D (7-AAD) were procured from BD Biosciences, and PE anti-human CD25 from Bio-Rad laboratories. After staining and washing, 7-AAD was added and incubated at room temperature for 15 minutes in dark. After incubation, 500 μL of 1X PBS was added to each tube and proceeded for sample acquisition using a BD FACS Lyric Flow Cytometer with BD FACSuite software version 1.2.1 (BD Biosciences) or CytoFlex Flow Cytometer with CytExpert software version 2.0 (Beckman Coulter). The data were analyzed using FlowJo software (FlowJo, LLC) version 9/10 and GraphPad Prism software version 8/9. During Flow Jo analysis cells were gated on live CD3+ CD8+ T cells.

In anti-PDL1 combination assays, PBMCs were stimulated with 2 ng/mL of soluble anti-CD3 in presence or absence of 10 ng/mL TGFβ. Drugs were added to a final concentration of 56 nmol/L (equivalent to 10 μg/mL of BCA101) and incubated at 37°C, 5% CO2 incubator for 72 hours. IFNγ levels were measured in the supernatant after approximately 72 hours as per manufacturer's instructions using Human IFNγ Quantikine ELISA Kit (R&D Systems Inc.). Absorbance readout was taken using a plate reader (BioTek Synergy HT, BioTek) with Gen 5 software version 2.0. Data were analyzed using GraphPad Prism software version 9.

Tumor cell and PBMC coculture assay

Frozen human PBMCs, procured from Precision Bioservices were thawed as per vendor recommended protocol on the previous day. Next day, drug dilutions (4X), HCT116 cell (ATCC) suspension and PBMC suspension was prepared in assay media containing McCoy's 5A media (GIBCO, Thermo Fisher Scientific), 1% FBS and 1% pen-strep. PBMCs (effector cells) were plated on to a 96-well white clear-bottom plate at a density of 50,000 cells/well and stimulated with soluble anti-CD3 antibody (OKT3 clone) at 2 ng/mL and cocultured with HCT116 cells (target cells) at a density of 5,000 cells/well (E:T = 10:1) in presence of TGFβ spiked at 1 ng/mL. Drugs were added to a final concentration of 56 or 6.67 nmol/L and incubated at 37°C, 5% CO2 incubator for 72 hours. IFNγ levels were measured in the supernatant after approximately 72 hours as per manufacturer's instructions using human IFNγ Quantikine ELISA Kit. Absorbance readout was taken using Cytation5 plate reader. Data were analyzed using GraphPad Prism software.

Luminex-based cytokine/chemokine measurement

A549 spheroids and human PBMC cocultures (T:E = 5,000:25,000) were treated with BCA101, cetuximab or TGFβRII-Fc at 56 nmol/L for 72 hours. Coculture supernatants were collected and analyzed for different cytokines and chemokines expression using a human XL Cytokine Magnetic Luminex Performance Assay 45-plex Fixed Panel kit (R&D Systems) as per the protocol recommended by the manufacturer. Three independent experiments were performed. Algorithm-based data analysis was performed. Baseline difference of each analyte was calculated by comparing individual drug-treated group with cells alone control group. After normalization, the following three groups were categorized: (i) BCA101 versus IgG1; (ii) BCA101 versus TGFβRII-Fc; (iii) BCA101 versus cetuximab. Individual fold difference for each analyte was analyzed for upregulated (>1.4 folds) or downregulated (<0.71 folds). Data shown here show analytes consistently different across three independent experiments while #denotes analytes at least differentially expressed in two out of three independent experiments. Statistical analysis was performed between the groups by a t test and P values are presented next to each analyte in the parenthesis. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Tumor cells—human NK-cell coculture assay

Frozen human peripheral CD56+ NK cells procured from ZenBio Inc. were thawed as per vendor recommended protocol in RPMI1640 media (GIBCO, Thermo Fisher Scientific) containing 10 ng/mL rh IL15 protein (R&D Systems Inc.), 1,000 units/mL rh IL2 protein (R&D Systems Inc.), 10% human serum (Sigma-Aldrich) and 1% pen-strep and incubated at 37°C and 5% CO2 incubator for approximately 48 hours. NK cells were also isolated from frozen as well as freshly isolated human PBMCs using human NK-cell enrichment kit (Stemcell Technologies Inc.) and EasySep magnet (Stemcell Technologies Inc.). Drug dilutions (4X), HCT116 cell suspension, and NK-cell suspension was prepared in assay media containing McCoy's 5A media, 1% FBS and 1% pen-strep. NK cells (effector cells) were plated on to a 96-well white clear-bottom plate (at a density of 50,000 cells/well) and cocultured with HCT116 cells (target cells) at a density of 5,000 cells/well (E:T = 10:1) in presence of TGFβ spiked at 1 ng/mL. Drugs were added to a final concentration of 56 nmol/L and incubated at 37°C, 5% CO2 for 72 hours. IFNγ levels were measured in the supernatant after approximately 72 hours as per manufacturer's instructions using human IFNγ Quantikine ELISA Kit. Absorbance readout was taken using a plate reader (BioTek Synergy HT or Spectramax M5e plate reader. Data were analyzed using GraphPad Prism software.

iTreg differentiation assay

Frozen human PBMCs were revived and rested overnight before enriching naïve CD4+ T cells using a negative selection kit (Stemcell Technologies Inc.). The enrichment was confirmed by flow cytometry using BV421 anti-human CD3, Alexa Fluor 647 anti-human CD4, FITC anti-human CD45RA, PE anti-human CD45RO, and 7-AAD. The enriched cells were seeded as 0.5–1 million cells/well in a 12-well plate and Immunocult Treg differentiation supplement (TDS; Stemcell Technologies Inc.) was added as per the vendor recommendation. Drugs were added at equimolar concentration (56 nmol/L) as indicated and the volume was made up to 1 mL using assay media. A total of 25 μL of Immunocult anti-human CD3/CD28 activator (Stemcell Technologies Inc.) was added to all the wells and the plates were incubated for approximately 168 hours (7 days) in 37°C, 5% CO2 incubator. A total of 500 μL media was added to each well along with TDS on day 5 of the assay set-up. At the end of assay incubation, the cell culture supernatant was collected, and the cells were further processed for intracellular FOXP3 staining using flow cytometry. Cells were collected in FACS tubes, centrifuged, and incubated with Fc block (ebiosciences) for 20 minutes at 4°C.

After washing, fixable viability dye (Invitrogen) was added and incubated at 4°C for 30 minutes followed by staining with surface marker antibodies. FITC anti-human CD45, Alexa Fluor 488 anti-human CD4, BV421 anti-human CD25 were procured from BD Biosciences. After washing, intracellular staining was performed with PE anti-human FOXP3 (BD) using FOXP3 staining buffer kit (BD) as per the vendor recommended protocol. The samples were acquired using a BD FACS Lyric or Beckman CytoFlex flow cytometer. The data were analyzed using FlowJo software (FlowJo, LLC) version 9/10and GraphPad Prism software version 9.

K562 cytolytic assay

Frozen human PBMCs, procured from Precision Bioservices or Zen-Bio Inc. were thawed as per vendor recommended protocol on the previous day in RPMI1640 media with 10% HI-FBS and 100 IU/mL IL2. Next day, PBMCs (1 million per well) were treated with TGFβ (10 ng/mL) along with human IgG/cetuximab or BCA101 at 56 nmol/L in 12-well plates and incubated at 37°C, 5% CO2 incubator for 72 hours.

After incubation, drug pretreated PBMCs (effector cells) were harvested, counted, and reseeded at 0.1 million per well along with K562 (MHC class I–deficient tumor cell line) at 10,000 cells per well in a 96-well plate. Target to effector ratio was maintained as 1:10. Media used was Iscove's modified Dulbecco's medium + 1% HI-FBS + 100 IU/mL of IL2 and the plates were incubated at 37°C, 5% CO2 incubator for 24 hours. Cytotoxicity was measured by evaluating protease release using CytoToxGlo (Promega) reagent. Luminescence readout was taken using a plate reader (Cytation5 BioTek plate reader) and data were analyzed using GraphPad Prism software.

EMT assays

For two-dimensional (2D) culture EMT studies, exponentially growing A549 cells (ATCC) were simultaneously treated with 10 ng/mL TGFβ and test compounds at 56 nmol/L for 72 hours in serum-free F12K (ATCC) media. For three-dimensional (3D) spheroid culture, A549 cells were cultured in ultra-low attachment U-bottom plate (Corning) for 4 days with conditioned media to form spheroids. Spheroids were treated as mentioned above. Cell cultures (2D) and spheroids (3D) were either imaged in Cytation 5 (Biotek) imaging system or further processed for quantitative PCR or Western blot analysis. To show the TGFβ-induced inhibition of EMT and its reversion, A549 2D cells were seeded in a T25 flask (Cell Star) stimulated with or without 2 ng/mL TGFβ in F12K media containing 5% FBS and 1% pen-strep. On day 5, the spent media of the flask was changed to 1% FBS assay media and incubated for another 3 days. On day 8, the cells stimulated with or without 2 ng/mL TGFβ were trypsinized and plated onto a 96-well white clear-bottom plate. Cells were then incubated with recombinant TGFβ at 2 ng/mL and with 56 nmol/L of BCA101, cetuximab (Merck KGaA) and human IgG isotype control (R&D Systems Inc.). After 120 hours, cell culture supernatant was collected from the assay plate and proceeded with IL11 ELISA as per manufacturer's instructions using human IL11 Quantikine ELISA Kit (R&D Systems Inc.). Absorbance readout was taken using BioTek Synergy HT plate reader. Data were analyzed using GraphPad Prism software.

qPCR

A549 cells were treated with 10 ng/mL TGFβ and indicated test compounds at 56 nmol/L for 72 hours. Total RNA was isolated from whole cells using RNeasy Mini Kit (74106, Qiagen) as per the manufacturer's instructions. RNA concentration was measured with NanoDrop and preserved at −80°C for further use. cDNA was synthesized from 2 μg of total RNA by using Omniscirpt Reverse Transcription Kit (205113, Qiagen) and quantitative real-time PCR analysis was performed with Evargreen qPCR Kit (QGHNR-02, QartaBio) in a real-time QuantStudio 5 (Thermo Fisher Scientific). GAPDH (Qiagen) served as an internal housekeeping control. The expression levels of E-cadherin (Qiagen) and vimentin (Qiagen) were determined using the 2−ΔΔCt method. Data were analyzed using GraphPad Prism software.

Immunoblots

All primary antibodies were procured from Cell Signaling Technology, unless otherwise specified. A549 cells were treated with 10 ng/mL TGFβ and indicated test compounds at 56 nmol/L for 72 hours. Cell pellets were lysed with RIPA buffer (Merck), processed, and loaded onto precast SDS-PAGE gels (4%–20%, Bio-Rad) for protein separation. Respective gels were transferred to polyvinylidene difluoride membrane and blocked in 5% BSA in TBST (10 mmol/L Tris, pH 8.0, 150 mmol/L NaCl, 0.5% Tween 20) buffer. Primary antibodies at 1:1,000 dilution against E-cadherin and vimentin were incubated overnight at 4°C. β-actin was used as an internal loading control. Respective HRP-conjugated secondary antibodies (1:2,000) were incubated in 1% nonfat dry milk (Cell Signaling Technology) for 2 hours. Finally, blots were developed with the enhanced chemiluminescence system (Thermo Fisher Scientific) on Chemidoc (Bio-Rad).

EGF and TGF combination studies

A549 spheroids were treated with EGF (5 ng/mL) plus TGFβ (10 ng/mL) followed by BCA101 or cetuximab for 72 hours. Spheroids from three independent experiments were imaged for invasion and supernatant were estimated for VEGF levels by ELISA (R&D Systems) as per the manufacturer's instructions.

Animal husbandry and ethical statement

Mice were kept under 12 hours day and night schedule and maintained as per the regulations of Committee for the Purpose of Control and Supervision of Experiments on Animals, Government of India and Association for Assessment and Accreditation of Laboratory Animal Care guidelines. All animal experiments were approved by Institutional Animal Care and Use Committee and performed under approved protocol.

BCA101 pharmacokinetic studies

Single dose of BCA101 (5 mg/kg) was injected intravenously to 5–6 weeks old female BALB/c mice (n = 3) and mice were bled through retro-orbital plexus at regular intervals. Serum samples were diluted in assay diluent (0.1% PBST plus blocking buffer) and bifunctional ELISA was performed as described above. For TGFβ serum levels, samples were treated with 1N HCl for 10 minutes and neutralized with 1.2N NaOH/0.5 mol/L HEPES buffer. Activated TGFβ levels were measured by quantikine ELISA assay as per the manufacturer's instructions (R&D Systems).

Tumor localization studies (whole animal imaging)

Test compounds BCA101, cetuximab and TGFβRII-Fc were labeled with Alexa Fluor 647 to the similar degree of labeling using a commercially available kit (Thermo Fisher Scientific). Alexa Fluor 647–labeled BCA101 (10 mg/kg), cetuximab (10 mg/kg) or TGFβRII-Fc (10 mg/kg) was administered at a single dose, intraperitoneally to athymic nude mice (n = 1) bearing subcutaneous FaDu xenografts [tumor volume (TV) ∼200 mm3]. To image the florescence from mice, animals were anesthetized using isoflurane vaporizer and placed on the imaging system platform (In Vivo Xtreme, Bruker). In vivo imaging was carried out using fluorescence modality (Ex-650nm; Em:700nm) and coregistered with X-ray at 1, 6, 12, 24, 48, 72, 96, and 168 hours. For quantification of florescence, a constant size region of interest (ROI) was drawn around the tumor of each mouse and the mean florescence intensity within the ROI was measured as photons/second/mm2. Background fluorescence was measured by selecting minimum intensity on the perimeter of ROI as a background parameter. The net intensity of the fluorescence was calculated by subtracting background intensity from the mean fluorescence intensity. For visual presentation of mice, fluorescence images were superimposed with X-ray images and the intensity scale was set 4,000 photons/second/mm2 as minimum and 5,500 photons/second/mm2 as maximum across the groups of mice using Molecular Imaging software from Bruker.

Xenograft studies

Exponentially growing FaDu cells (ATCC HTB-43) were trypsinized and 0.5 × 106 Matrigel (354234, Corning) suspended FaDu cells were injected into the right flank of nude mice. Once TV reached around 100–200 mm3, mice were randomized into test and control groups. Test groups were treated with BCA101 at 50 mg/kg or equimolar doses of cetuximab, TGFβRII-Fc, or combination of BCA101 plus TGFβRII-Fc, intraperitoneally twice a week. TV and mouse body weight were measured twice a week. At the end of the experiments, animals were euthanized as per the standard protocol and photographs were taken for mice and excised tumors.

For in vivo serum and tumor TGFβ neutralization studies, exponentially growing A431 cells (5 × 106) were injected into right flank of mice. Once tumor reached about 100 mm3, mice were randomized into test and control mice. Test group mice were treated with different doses of BCA101 (10, 5 and 1 mg/kg), cetuximab at 10 mg/kg or TGFβRII-Fc at 5 mg/kg for seven doses (twice a week, intraperitoneally). On the day of experiment termination, blood was collected from retro-orbital bleeding and tumors were excised. Serum and tumors were snap frozen and stored at −80°C until TGFβ ELISA analysis as mentioned in the pharmacokinetics section above.

Histologic assessment of apoptotic to mitotic index ratio in xenograft tumor tissues

FaDu xenograft tumor tissues from vehicle, cetuximab, TGFβRII-Fc, BCA101, and cetuximab plus TGFβRII-Fc–treated mice (n = 5, per treatment group; n = 25, total tumor tissues) were fixed in 10% neutral buffered formalin and processed using Leica ASP 300 automated tissue processor. The paraffin-embedded tumor tissue blocks were prepared using Leica EG 1150C, EG 1150H stations. The 4 to 5 μm tissue sections were cut using Leica RM 2255 automated rotary microtome and mounted on poly-L-lysine coated slides. The hematoxylin and eosin (H&E) staining was conducted on two serial sections per tumor tissue using Leica ST 5020 H&E slide stainer followed by mounting and coverslipping using automatic coverslipper (Leica CV5030). The H&E-stained tumor area of two serial sections was evaluated for number of apoptotic and mitotic figures. Briefly, five random images (20× magnification, scale bar 100 μmol/L) were captured from each section using Leica camera (v4.12). In captured images (n = 10 per tumor), the apoptotic and mitotic figures were manually counted using the cell counter tool in Image J software (1.53e). The ratio of mean apoptotic and mitotic figures (per mice of a treatment group) was calculated using Microsoft Excel, and presented as the Box and Whisker plot, and statistically analyzed by one-way ANOVA followed by Dunn multiple comparisons test to compare mean rank of one treatment group with mean rank of every other group using GraphPad Prism 9.

Patient-derived xenograft studies

For patient-derived xenograft (PDX) model, head and neck cancer patient samples were obtained through Mazumdar Shaw Medical Foundation-Narayana Hrudayalaya Biorepository, Bengaluru, India, after appropriate approvals obtained from ethical committee and propagated in NOG mice. Tumor fragments were implanted in NOG mice subcutaneously using trocor, once tumor grew about 100–200 mm3 mice were treated with BCA101 at 10 mg/kg or equimolar doses of cetuximab, TGFβRII-Fc or combination of BCA101 and TGFβRII-Fc. For PDX-1 study, mice were treated for 27 days and left untreated for 52 days; for PDX-2 study, mice were treated for 38 days and left untreated for 70 days; for PDX-3, mice were treated for 32 days and left untreated for 28 days.

Coimplantation studies

For HT29 coimplantation model, about 5 × 106 HT29 cells (ATCC HTB-38) were mixed with human PBMCs at a ratio of 1:2 in Matrigel (BD Biosciences) and implanted into mice. Once TV reached about 100 mm3, mice were randomized and treated as mentioned above.

PC3 humanized mice model

For Hu-NOG EXL model, mice were purchased from Taconic. Prostate cancer PC3 cells were implanted into mice and once tumor grew about 130 mm3, mice were randomized and treated with BCA101 (10 mg/kg), cetuximab (8.1 mg/kg), pembrolizumab (10 mg/kg) or combination of BCA101 and pembrolizumab. TVs and body weight were measured as mentioned above.

B16-EGFR syngeneic model

B16-hEGFR syngeneic mouse model: Human EGFR transfected B16 mouse melanoma cell line (B16-hEGFR) was procured from University of Chicago (Chicago, IL) and maintained in DMEM. About 0.75 million B16-hEGFR cells were injected into right flank of C57 mice. Once tumors were palpable (day 5, nonmeasurable), mice were randomized into control and test groups. Test groups were treated with either BCA101 at 50 mg/kg, cetuximab (40.5 mg/kg), TGFβRII-Fc (23 mg/kg), or combination of BCA101 and TGFβRII-Fc (intraperitoneally, twice a week). For immune checkpoint combination groups, mice were treated with mouse cross-reactive anti-PD1 (10 mg/kg, BioXcell-J43; Cat-BP0033-2) or combination of anti-PD1 with either BCA101 or cetuximab (intraperitoneally, twice a week). TV and mice weight were recorded twice a week.

Cell lines

The source of all cell lines used in this article is acknowledged in the relevant sections. All cell lines have been sourced from a working cell bank and used before passage 20. Mycoplasma testing and cell line authentication has not been routinely performed.

Statistical analysis

Statistical analysis was performed using GraphPad Prism 9 software (GraphPad software). Details of the statistical methods applied for each assay are presented in the respective figure legends. In graphs, only groups showing statistically significant difference are shown, based on levels of significance as asterisks (*). Details are indicated in the figure legends.

Data availability

The data generated in this study are available upon request from the corresponding author.

BCA101 is a bifunctional IgG1 mAb designed to target both EGFR and TGFβ simultaneously

BCA101 is a bifunctional recombinant fusion mAb consisting of a modified cetuximab (devoid of C-terminal lysine in heavy chain), a chimeric anti-human EGFR mAb, fused to human TGFβRII ECD (Fig. 1A). The C-terminus of cetuximab LC is fused to TGFβRII-ECD through a flexible linker consisting of 15 amino acids (G4Sx3), the LC fusion of TGFβRII-ECD, was found to be more stable compared with heavy chain fusion of the receptor trap (Supplementary Fig. S1A). Because binding of TGFβ1 to its natural receptor requires TGFβRII dimerization (14), our structural model (Fig. 1A and B) predicts that BCA101 linker is flexible enough to allow obligatory dimerization and ensures TGFβ neutralization with high avidity. Structural modeling also suggests minimal or no steric hindrance for Fc-mediated functions of the BCA101 because of the extended linker. SDS-PAGE analysis confirmed that under nonreducing conditions cetuximab and BCA101 migrated at an expected molecular weight of 152 and 178.1 kDa, respectively (Fig. 1C, lane #4 and #5). As expected, under reducing conditions, cetuximab migrated as two bands corresponding to 50 kDa heavy chain and 25 kDa LC (Fig. 1C, lane #1), whereas BCA101 migrated as a single band of about 50 kDa (Fig. 1C, lane #2) because of increased size of LC due to TGFβRII ECD fusion.

Figure 1.

BCA101 retains EGFR and TGFβ binding activity in fusion format. A, Computer-generated working model of BCA101 (front view). BCA101 is composed of anti-human EGFR mAb (cetuximab) and TGFβ receptor RII ECD fused to the C terminus of the LC of cetuximab IgG via a flexible (G4S)3 linker. B, A 90° rotation of BCA101 structure showing (Gly4Ser)3 linker connected to TGFβRII-ECD (side view). C, BCA101 reducing and nonreducing SDS-PAGE. Cetuximab and BCA101 were analyzed by SDS-PAGE under both reducing (lane 1 and 2, respectively) and nonreducing (lane 4 and 5, respectively) conditions. Lanes 3 and 6 are protein markers along with the expected molecular weight scale on either side. D, SPR-based BCA101 binding to its cognate targets, EGFR and TGFβ. rh EGFR or TGFβ was immobilized on CM5 chips. BCA101 or controls were run at a specified flow rate for a set of association and dissociation times to determine Kd values (nmol/L). E, FaDu cells were incubated with serial dilutions of BCA101 or cetuximab, followed by a fluorochrome-conjugated anti-human IgG secondary antibody. BCA101 and cetuximab binding to cells was analyzed by flow cytometer and MFI was plotted against concentration. EGFR and TGFβ binding ELISA. F and G, Plates were coated with either EGFR-Fc for EGFR binding (F) or TGFβ for TGFβRII-Fc binding (G) and probed with various concentrations of BCA101, followed by an HRP-conjugated secondary antibody specific to human IgG. Color was developed by adding TMB substrate and plates were read using a plate reader. H, BCA101 bifunctional ELISA. Plates were coated with EGFR-Fc, probed with various concentrations of BCA101, followed by TGFβ spiking, biotinylated anti-TGFβ antibody, and SA-HRP. Plates were developed with substrate (TMB) and read in a plate reader. Cetuximab was used as a negative control. Anti-VEGF mAb was used as an IgG control for EG. Cetuximab was used as a negative control in bifunctional ELISA. For D,F, and G, three independent experiments were performed, and representative data are presented as mean ± SD of triplicate. E, Only one experiment was performed. For H, six independent experiments were performed, and representative data are presented as mean ± SD of triplicate.

Figure 1.

BCA101 retains EGFR and TGFβ binding activity in fusion format. A, Computer-generated working model of BCA101 (front view). BCA101 is composed of anti-human EGFR mAb (cetuximab) and TGFβ receptor RII ECD fused to the C terminus of the LC of cetuximab IgG via a flexible (G4S)3 linker. B, A 90° rotation of BCA101 structure showing (Gly4Ser)3 linker connected to TGFβRII-ECD (side view). C, BCA101 reducing and nonreducing SDS-PAGE. Cetuximab and BCA101 were analyzed by SDS-PAGE under both reducing (lane 1 and 2, respectively) and nonreducing (lane 4 and 5, respectively) conditions. Lanes 3 and 6 are protein markers along with the expected molecular weight scale on either side. D, SPR-based BCA101 binding to its cognate targets, EGFR and TGFβ. rh EGFR or TGFβ was immobilized on CM5 chips. BCA101 or controls were run at a specified flow rate for a set of association and dissociation times to determine Kd values (nmol/L). E, FaDu cells were incubated with serial dilutions of BCA101 or cetuximab, followed by a fluorochrome-conjugated anti-human IgG secondary antibody. BCA101 and cetuximab binding to cells was analyzed by flow cytometer and MFI was plotted against concentration. EGFR and TGFβ binding ELISA. F and G, Plates were coated with either EGFR-Fc for EGFR binding (F) or TGFβ for TGFβRII-Fc binding (G) and probed with various concentrations of BCA101, followed by an HRP-conjugated secondary antibody specific to human IgG. Color was developed by adding TMB substrate and plates were read using a plate reader. H, BCA101 bifunctional ELISA. Plates were coated with EGFR-Fc, probed with various concentrations of BCA101, followed by TGFβ spiking, biotinylated anti-TGFβ antibody, and SA-HRP. Plates were developed with substrate (TMB) and read in a plate reader. Cetuximab was used as a negative control. Anti-VEGF mAb was used as an IgG control for EG. Cetuximab was used as a negative control in bifunctional ELISA. For D,F, and G, three independent experiments were performed, and representative data are presented as mean ± SD of triplicate. E, Only one experiment was performed. For H, six independent experiments were performed, and representative data are presented as mean ± SD of triplicate.

Close modal

To evaluate individual and simultaneous target binding of BCA101, EGFR and TGFβ binding studies were performed using surface plasmon resonance (SPR) and ELISAs. SPR studies show that both EGFR and TGFβ binding affinities were retained in BCA101 (Fig. 1D). To confirm whether BCA101 bound to EGFR with native conformation on a cell surface, FaDu cells were stained with BCA101 or cetuximab and evaluated in a flow cytometer. In this assay, EC50 estimates were similar for cetuximab and BCA101 (Fig. 1E). Similar results were observed by ELISAs, where the EC50 values for EGFR and TGFβ binding for BCA101 were comparable with those of cetuximab and TGFβRII-Fc (Fig. 1F and G), respectively. The higher signal obtained by cetuximab over BCA101 in the EGFR ELISA may be attributed to reduced secondary antibody (anti-human Fc) access to the bispecific. To confirm whether both the arms of BCA101 are concurrently functional, a two-step bifunctional ELISA was performed. In this ELISA, BCA101 was captured by plate-bound EGFR and the ability of BCA101 to bind TGFβ as measured by spiking TGFβ followed by detection by a biotinylated anti-TGFβ antibody (Fig. 1H). BCA101 showed a dose-dependent binding with an EC50 of 0.09 ± 0.01 nmol/L, indicating both the anti-EGFR and TGFβ trap arms are simultaneously functional. In contrast, cetuximab showed a negative profil2e in this assay format as expected.

The Fab and Fc domains of BCA101 are biologically functional

We next evaluated the ability of BCA101 to inhibit EGF-dependent cell proliferation and mediate ADCC against of EGFR-positive tumor cells. In Fig. 2A, FaDu cells preincubated with either BCA101 or cetuximab and spiked with EGF showed comparable inhibition of EGFR phosphorylation (Tyr1068). The sustained inhibition of EGFR phosphorylation at 0.1 μg/mL by BCA101 over cetuximab may be attributed to better steric hindrance to ligand binding (EGF) conferred by BCA101. Quantitative analysis of Western blots presented in Supplementary Fig. S1B. BCA101 showed a dose-dependent inhibition of cell proliferation (IOP) across the indicated tested cell lines (Fig. 2B). EC50 values for different cell lines ranged from 0.08 to 0.33 nmol/L and 0.06 to 0.23 nmol/L for BCA101 and cetuximab, respectively (Fig. 2C), suggesting similar inhibitory potential in EGFR-expressing cell lines for both BCA101 and cetuximab. As expected, BCA101 did not inhibit proliferation of low EGFR-expressing MCF7 cell line (Supplementary Fig. S2A and S2B), indicating the specificity of BCA101.

Figure 2.

BCA101 is functionally active in IOP, ADCC, and TGFβ neutralization assays. A, Western blot analysis of phospho-EGFR in EGF-stimulated FaDu cells. FaDu cells were incubated overnight with indicated concentrations of either BCA101 or cetuximab and then stimulated with EGF (25 ng/mL) for 5 minutes. Cells were lysed and evaluated for pEGFR (Tyr1068) levels by Western blot analysis. Representative blots are presented. B, Inhibition of proliferation assay. Different concentrations of BCA101 or cetuximab were added to FaDu cells in a 96-well plate and incubated for about 72 hours. Cell viability was measured by CellTiter-Glo (Promega). C, Table showing the summary of ADCC and IOP assay EC50 (mean ± SD) for BCA101 and cetuximab across various cell lines. D, ADCC assay. CAL27 cells expressing EGFR were plated in a 96-well plate and allowed to attach overnight. Cells were then incubated with engineered effector Jurkat cells expressing FcγRIIIa receptor and NFAT-RE along with different dilutions of BCA101 or cetuximab. After about 24 hours of incubation, cells were incubated with BioGlo luciferase reagent and plates were read in a plate reader. E, ADCC activity of BCA101 with TGFβ pretreated NK cells. NK cells isolated from healthy donor were incubated for 72 hours with no treatment and with TGFβ at 1 ng/mL along with cetuximab or BCA101 (56 nmol/L).Cetuximab/BCA101 were added at concentration of 56 nmol/L to target cells and were preincubated for 1 to 2 hours. After treatment, pretreated NK cells were added at T:E ratio of 1:5 and cocultured for 24 hours. CytoToxGlo readout was taken to measure the dead cell proteases. Data are plotted as percentage change in cytotoxicity over human IgG control group. Representative data were plotted as mean ± SD of triplicates from single experiment. This experiment is a representative of three independent experiments with three donors. F, TGFβ-SMAD reporter assay. Engineered HEK293 cells were plated on to a 96-well plate and allowed to attach overnight. Cells were then incubated with recombinant TGFβ at 20 ng/mL and with different dilutions of BCA101. After 18 hours of incubation, cells were incubated with BioGlo luciferase reagent and luciferase activity was measured. Anti-VEGF mAb was used as an IgG control for B and C. For AF, at least three independent experiments were performed, and representative data are presented as mean ± SD of triplicate wells wherever applicable. For statistical analysis of E, one-way ANOVA, nonparametric, Kruskal–Wallis test with Dunn multiple comparisons test was performed. Multiple comparison was done comparing hIgG group with all other test groups. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01.

Figure 2.

BCA101 is functionally active in IOP, ADCC, and TGFβ neutralization assays. A, Western blot analysis of phospho-EGFR in EGF-stimulated FaDu cells. FaDu cells were incubated overnight with indicated concentrations of either BCA101 or cetuximab and then stimulated with EGF (25 ng/mL) for 5 minutes. Cells were lysed and evaluated for pEGFR (Tyr1068) levels by Western blot analysis. Representative blots are presented. B, Inhibition of proliferation assay. Different concentrations of BCA101 or cetuximab were added to FaDu cells in a 96-well plate and incubated for about 72 hours. Cell viability was measured by CellTiter-Glo (Promega). C, Table showing the summary of ADCC and IOP assay EC50 (mean ± SD) for BCA101 and cetuximab across various cell lines. D, ADCC assay. CAL27 cells expressing EGFR were plated in a 96-well plate and allowed to attach overnight. Cells were then incubated with engineered effector Jurkat cells expressing FcγRIIIa receptor and NFAT-RE along with different dilutions of BCA101 or cetuximab. After about 24 hours of incubation, cells were incubated with BioGlo luciferase reagent and plates were read in a plate reader. E, ADCC activity of BCA101 with TGFβ pretreated NK cells. NK cells isolated from healthy donor were incubated for 72 hours with no treatment and with TGFβ at 1 ng/mL along with cetuximab or BCA101 (56 nmol/L).Cetuximab/BCA101 were added at concentration of 56 nmol/L to target cells and were preincubated for 1 to 2 hours. After treatment, pretreated NK cells were added at T:E ratio of 1:5 and cocultured for 24 hours. CytoToxGlo readout was taken to measure the dead cell proteases. Data are plotted as percentage change in cytotoxicity over human IgG control group. Representative data were plotted as mean ± SD of triplicates from single experiment. This experiment is a representative of three independent experiments with three donors. F, TGFβ-SMAD reporter assay. Engineered HEK293 cells were plated on to a 96-well plate and allowed to attach overnight. Cells were then incubated with recombinant TGFβ at 20 ng/mL and with different dilutions of BCA101. After 18 hours of incubation, cells were incubated with BioGlo luciferase reagent and luciferase activity was measured. Anti-VEGF mAb was used as an IgG control for B and C. For AF, at least three independent experiments were performed, and representative data are presented as mean ± SD of triplicate wells wherever applicable. For statistical analysis of E, one-way ANOVA, nonparametric, Kruskal–Wallis test with Dunn multiple comparisons test was performed. Multiple comparison was done comparing hIgG group with all other test groups. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01.

Close modal

Another important functional effect of cetuximab is the potential of the antibody to induce ADCC. An ADCC reporter assay was performed to evaluate Fc functions of BCA101. BCA101 showed concentration-dependent ADCC activity in the reporter-based ADCC assay comparable with that of cetuximab (Fig. 2D), the average of EC50 for ADCC across the lines presented in Fig. 2C. In addition, FCGR3A (V176F) binding affinities were also similar for BCA101 (1.25 ± 0.17 μmol/L) and cetuximab (1.11 ± 0.04 μmol/L). In a clinical setting, ADCC is predominantly attributed to the presence of activated NK cells, overexpressing FcγRIIIa, within the TME (15, 16). ADCC activity of BCA101 was evaluated using A431 (EGFRhigh) and MCF7 (EGFRlow) cell lines (Supplementary Fig. S2A). TGFβ pretreated NK cells were cocultured with A431 or MCF7 cells at T:E ratio of 1:5. In the absence of TGFβ, both cetuximab and BCA101 showed comparable NK cell–mediated ADCC activity. Whereas pretreatment of NK cells with TGFβ showed reduction in ADCC, which was rescued when NK cells were pretreated with TGFβ and BCA101 (Fig. 2E), MCF7 (EGFRlow) failed to show ADCC, demonstrating antigen specificity (Supplementary Fig. S2C).

BCA101 neutralizes TGFβ1 and TGFβ3 but not TGFβ2 in TGFβ SMAD reporter assay

TGFβ isoforms neutralizing activity of BCA101 was evaluated in a SMAD responsive reporter assay. BCA101 could neutralize SMAD-mediated luciferase expression induced by TGFβ1 (Fig. 2F) and TGFβ3 (Supplementary Fig. S3A), and not neutralize TGFβ2 in HEK-SMAD reporter assay (Supplementary Fig. S3B). The neutralization was achieved in a concentration-dependent manner with mean EC50 of 1.49 ± 0.09 nmol/L and 3.51 ± 0.19 nmol/L, respectively (Supplementary Fig. S3C). The reason for the inability of BCA101 to neutralize TGFβ2 is unclear but in line with earlier report (17).

BCA101 rescues TGFβ-mediated immune suppression

The immunosuppressive role of TGFβ within the TME is well documented (18, 19). Recent studies have revealed roles for TGFβ in tumor immune evasion and poor responses to cancer immunotherapy (9, 13). Therefore, we evaluated the ability of BCA101 to rescue TGFβ-mediated immune suppression by evaluating activation markers on T cells in PBMCs.

NKG2D is an activation marker on activated cytotoxic NK and CD8+ cells (20, 21). The addition of recombinant TGFβ in PBMCs costimulated with soluble anti-CD3 and anti-CD28 showed a reduction in NKG2D expression [based on median fluorescence intensity (MFI) value] on gated CD8+ T cells. Treatment of only BCA101 showed significant rescue of NKG2D expression compared with hIgG (Fig. 3A). This increase in NKG2D expression was also associated with an enhanced frequency of CD25+ NKG2Dbright CD8+ T cells in the PBMCs treated with BCA101 as compared with hIgG group (Fig. 3B).

Figure 3.

BCA101 shows rescue from TGFβ-mediated immune suppression in PBMC-based immune activation assays and tumor cell coculture assays. A and B, PBMC activation assays. Human PBMCs were stimulated with soluble anti-CD3 (2 ng/mL) and soluble anti-CD28 (10 ng/mL) antibodies in the absence or presence of TGFβ (10 ng/mL) and drugs at 56 nmol/L concentration for about 72 hours, followed by flow cytometry analysis. In A, C, and D, fold increase over hIgG (with TGFβ) is shown. A, Compiled analysis of fold increase in NKG2D marker expression (MFI) gated on live CD3+CD8+ T cells. B, Frequency of CD25+ NKG2Dbright cells for various control and test drugs (gated on live CD3+ CD8+ T cells). C, PBMC and HCT116 coculture assays. Human PBMCs were stimulated with soluble anti-CD3 antibody at 2 ng/mL and cocultured with HCT116 cells (human colorectal carcinoma cell line that expresses EGFR) in the absence or presence of TGFβ spiked at 1 ng/mL. Drugs were added to a final concentration of 56 nmol/L. IFNγ level was measured in the supernatant after 72 hours. D, NK-cell and HCT116 coculture assays. Human NK cells were cultured with HCT116 cells in the absence or presence of TGFβ spiked at 1 ng/mL. IFNγ level was measured in the supernatant after about 72 hours. Drugs were added to a final concentration of 56 nmol/L. E, K562 cytolytic killing assay. Frozen human PBMCs were cultured in presence of TGFβ (10 ng/mL) along with human IgG, TGFβRII-Fc, cetuximab, and BCA101 at 56 nmol/L. After 72 hours, MHC class I–deficient tumor cell line K562 was cocultured with pretreated PBMCs at T:E ratio of 1:10 and incubated for 24 hours. Cytotoxicity was measured by evaluating protease release using CytoToxGlo luminescence readout. F, The frequency of CD4+CD25+FOXP3+ cells treated with human IgG in the absence of TDS and human IgG, cetuximab, TGFβRII-Fc, and BCA101, respectively, in the presence of Immunocult TDS. A,B, and F include data plotted from at least three independent experiments. C,D, and E include data plotted from nine, six, and four independent experiments, respectively, with three replicates in each experiment. Data are plotted as mean ± SD. For statistical analysis, one-way ANOVA, nonparametric, Kruskal–Wallis test with Dunn multiple comparisons test was performed for A,B, and F. In A, while there was a significant difference between BCA101- and hIgG-treated groups, there was no significant difference (P < 0.05) between other groups including BCA101 and TGFβ-RII-Fc. Data from A are represented in B as frequency and none of the groups were significantly different, statistically. Parametric one-way ANOVA was performed for rest of the panels. In all the groups, multiple comparison test where mean rank of each column was compared with every other column. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Figure 3.

BCA101 shows rescue from TGFβ-mediated immune suppression in PBMC-based immune activation assays and tumor cell coculture assays. A and B, PBMC activation assays. Human PBMCs were stimulated with soluble anti-CD3 (2 ng/mL) and soluble anti-CD28 (10 ng/mL) antibodies in the absence or presence of TGFβ (10 ng/mL) and drugs at 56 nmol/L concentration for about 72 hours, followed by flow cytometry analysis. In A, C, and D, fold increase over hIgG (with TGFβ) is shown. A, Compiled analysis of fold increase in NKG2D marker expression (MFI) gated on live CD3+CD8+ T cells. B, Frequency of CD25+ NKG2Dbright cells for various control and test drugs (gated on live CD3+ CD8+ T cells). C, PBMC and HCT116 coculture assays. Human PBMCs were stimulated with soluble anti-CD3 antibody at 2 ng/mL and cocultured with HCT116 cells (human colorectal carcinoma cell line that expresses EGFR) in the absence or presence of TGFβ spiked at 1 ng/mL. Drugs were added to a final concentration of 56 nmol/L. IFNγ level was measured in the supernatant after 72 hours. D, NK-cell and HCT116 coculture assays. Human NK cells were cultured with HCT116 cells in the absence or presence of TGFβ spiked at 1 ng/mL. IFNγ level was measured in the supernatant after about 72 hours. Drugs were added to a final concentration of 56 nmol/L. E, K562 cytolytic killing assay. Frozen human PBMCs were cultured in presence of TGFβ (10 ng/mL) along with human IgG, TGFβRII-Fc, cetuximab, and BCA101 at 56 nmol/L. After 72 hours, MHC class I–deficient tumor cell line K562 was cocultured with pretreated PBMCs at T:E ratio of 1:10 and incubated for 24 hours. Cytotoxicity was measured by evaluating protease release using CytoToxGlo luminescence readout. F, The frequency of CD4+CD25+FOXP3+ cells treated with human IgG in the absence of TDS and human IgG, cetuximab, TGFβRII-Fc, and BCA101, respectively, in the presence of Immunocult TDS. A,B, and F include data plotted from at least three independent experiments. C,D, and E include data plotted from nine, six, and four independent experiments, respectively, with three replicates in each experiment. Data are plotted as mean ± SD. For statistical analysis, one-way ANOVA, nonparametric, Kruskal–Wallis test with Dunn multiple comparisons test was performed for A,B, and F. In A, while there was a significant difference between BCA101- and hIgG-treated groups, there was no significant difference (P < 0.05) between other groups including BCA101 and TGFβ-RII-Fc. Data from A are represented in B as frequency and none of the groups were significantly different, statistically. Parametric one-way ANOVA was performed for rest of the panels. In all the groups, multiple comparison test where mean rank of each column was compared with every other column. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Close modal

BCA101 rescues inhibitory effect of TGFβ and enhances cytolytic activity of PBMCs or NK cells against tumor cells

EGFR-positive HCT-116 cells, a human-derived colon cancer cell line with KRAS mutation, were cultured along with immune cells (PBMCs/NK cells) to evaluate the ability of BCA101 to mediate tumor cell cytolysis in vitro. Although BCA101 and cetuximab bind to these cells, HCT-116 was shown to be insensitive to cetuximab due to the activating KRAS mutation. Therefore, in this experimental setup, the role of TGFβRII-ECD arm of BCA101 in enhancing the direct killing by immune cells, via removing TGFβ from the milieu was evaluated. As expected, the addition of exogenous TGFβ caused an immune suppression as observed by decrease in IFNγ release upon TGFβ spiking (Fig. 3C). BCA101, significantly enhanced IFNγ secretion over cetuximab and hIgG, both in the PBMC (Fig. 3C) or NK-cell (Fig. 3D) coculture with HCT-116. In this immune suppressed milieu, cetuximab showed minimal or equivalent IFNγ secretion when compared with an isotype control mAb (hIgG).

In experiments 3C and 3D, the levels of activation (as judged by IFNγ release) mediated by TGFβRII-Fc was lesser than that of BCA101. This could be attributed to more stable conformational structure of BCA101 in solution as compared with TGFβRII-Fc. The observed difference in these experiments between BCA101 and TGFβRII-Fc as compared with Fig. 1G may be attributed to longer duration of these experiments.

BCA101 also showed an increase in immune cell–mediated cytolytic killing of MHC class I–deficient K562 tumor cell line when compared with cetuximab. K562 cells were cocultured with TGFβ pretreated PBMCs in the presence of human IgG, TGFβRII-Fc, BCA101, and cetuximab. Cytolysis of tumor cell line was evaluated by measuring protease release wherein BCA101 showed enhanced killing of the K562 cells compared with cetuximab and hIgG (Fig. 3E). Though K562 cells do not express EGFR, this assay shows TGFβ-mediated immune suppression and the restoration of cytolytic killing of tumor cells by PBMCs in the presence of BCA101.

Together, these data suggest that BCA101, through its TGFβRII ECD arm neutralizes the inhibitory TGFβ from the milieu and results in immune cell activation and tumor cell killing. Cetuximab, on the other hand, with its inability to sequester TGFβ was unable to eliminate the inhibitory effect of TGFβ. In these assay conditions, BCA101 shows superiority over cetuximab in causing immune cell activation. Therefore BCA101, may have the potential to show better therapeutic efficacy than cetuximab in tumors with high TGFβ levels in their TME.

BCA101 inhibits TGFβ1-mediated differentiation of naïve CD4+ T cells to iTregs

Human Tregs are a subset of CD4+ T cells that play a critical role in regulation of peripheral immune tolerance, preventing autoimmunity and inflammatory diseases (22). However, in the TME, they are involved in the suppression of antitumor immune response, resulting in disease progression (22, 23). Naïve CD4+ T cells differentiate into iTreg (CD4+CD25+FOXP3+) in the presence of TGFβ1 (pluripotent cytokine) and cause immune suppression (24). Depletion of Treg cells in TME leads to enhancement of antitumor immunity (25). In in vitro iTreg differentiation assays, BCA101 showed significant inhibition in the differentiation of naïve CD4+ T cells into iTreg when compared with cetuximab (Fig. 3F). BCA101 treatment resulted in the sequestration of TGFβ, eventually reducing the frequency of CD4+CD25+FoxP3+ cells compared with hIgG group (Fig. 3F). On the other hand, cetuximab, due to the lack of TGFβ trap, was not able to reduce the frequency of CD4+CD25+FoxP3+ T cells. Even though TGFβRII-Fc showed reduction in frequency of CD4+CD25+FoxP3+ cells due to binding/trapping of TGFβ, the inhibition caused was much lower than BCA101, probably due to conformational differences between the molecules resulting in better trapping of TGFβ1 by BCA101, as described earlier. These data indicate that BCA101 is superior to cetuximab in inhibiting the differentiation of naïve CD4+ T cells to iTregs. Hence, BCA101 has a potential therapeutic advantage in reducing the iTreg population, thereby activating the immune system, leading to antitumor immunity.

BCA101 inhibits TGFβ-induced EMT

EMT is a process that allows epithelial cells to undergo morphologic and signaling changes promoting invasiveness and survival (26, 27). TGFβ is known to play a pivotal role in inducing EMT (28, 29). To evaluate whether BCA101 inhibits TGFβ-induced EMT, EGFR-positive A549 cells were treated with TGFβ followed by BCA101 and morphologic and cellular signaling changes were monitored. TGFβ induced several mesenchymal-like changes in A549 cells in a 2D culture format such as spindle-like morphology and cell scattering (Fig. 4A, top) compared with control cells, which were rescued by BCA101 treatment. To evaluate whether BCA101 inhibits EMT in 3D spheroid model, TGFβ along with BCA101 or cetuximab were added to A549 spheroids and morphologic changes were imaged after 72 hours of incubation. TGFβ induced filamentous structures on spheroids indicating induction of EMT morphology and promotion of invasiveness. In comparison, BCA101-treated spheroids in presence of TGFβ showed similar morphology to control cells, indicating that BCA101 rescues TGFβ-induced EMT in 3D spheroid model (Fig. 4A, bottom and Fig. 4B).

Figure 4.

BCA101 inhibits TGFβ-induced EMT. A, BCA101 inhibits TGFβ-induced invasion in 2D and 3D cultures. A549 cells were plated in serum-free media and treated with BCA101, TGFβ-RII-Fc, or cetuximab in the presence of TGFβ (10 ng/mL). After 72 hours of incubation, images were captured using a phase contrast microscope (top). For spheroid imaging, A549 cells were seeded in a 96-well plate and allowed to form spheroids (day 3). Spheroids were treated with BCA101, TGFβRII-Fc, or cetuximab in presence of TGFβ (10 ng/mL), and images were captured using phase contrast microscope after 72 hours of incubation (bottom). B, Spheroid invasion was quantified by ImageJ software and the average of two independent experiments is presented as a box plot. C–G, BCA101 inhibits TGFβ-induced EMT. C, A549 cells were cultured and on day 8, cells were treated with TGFβ at 2 ng/mL along with the test drugs at a concentration of 56 nmol/L and IL11 levels were measured in supernatant on day 13. D, For reversion of EMT, A549 cells were cultured with TGFβ at 2 ng/mL. On day 8, TGFβ at 2 ng/mL along with the test drugs were added at a concentration of 56 nmol/L and IL11 levels were measured in supernatant on day 13. E and F, A549 cells were seeded in serum-free media and treated with TGFβ (10 ng/mL) and BCA101 or cetuximab. After 72 hours of incubation, cells were collected, and mRNA expression of E-cadherin (E) and vimentin (F) was analyzed using a quantitative PCR. G, A549 spheroids were treated with TGFβ (10 ng/mL) along with BCA101 or cetuximab for 72 hours. Spheroids were collected and analyzed for E-cadherin and vimentin by Western blot analysis. In CF, three independent experiments each in triplicate were performed, and data are presented as mean ± SD. H and I, BCA101 inhibits EGF and TGFβ-induced VEGF secretion. A549 spheroids were treated with EGF (5 ng/mL) plus TGFβ (10 ng/mL), followed by BCA101 or cetuximab for 72 hours. Spheroids from three independent experiments were imaged for invasion (H) and supernatant was estimated for VEGF levels (I) by ELISA (R&D Systems) as per the manufacturer's instructions. Statistical analysis for B, one-way ANOVA, nonparametric, Kruskal–Wallis test followed by Dunn multiple comparison test with cells treated with TGFβ as control. For CF, parametric one-way ANOVA was performed. F and I were performed using one-way ANOVA, nonparametric, Kruskal–Wallis test followed by Dunn multiple comparison test. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001. For CF, multiple comparison test was done where mean rank of each column was compared with every other column.

Figure 4.

BCA101 inhibits TGFβ-induced EMT. A, BCA101 inhibits TGFβ-induced invasion in 2D and 3D cultures. A549 cells were plated in serum-free media and treated with BCA101, TGFβ-RII-Fc, or cetuximab in the presence of TGFβ (10 ng/mL). After 72 hours of incubation, images were captured using a phase contrast microscope (top). For spheroid imaging, A549 cells were seeded in a 96-well plate and allowed to form spheroids (day 3). Spheroids were treated with BCA101, TGFβRII-Fc, or cetuximab in presence of TGFβ (10 ng/mL), and images were captured using phase contrast microscope after 72 hours of incubation (bottom). B, Spheroid invasion was quantified by ImageJ software and the average of two independent experiments is presented as a box plot. C–G, BCA101 inhibits TGFβ-induced EMT. C, A549 cells were cultured and on day 8, cells were treated with TGFβ at 2 ng/mL along with the test drugs at a concentration of 56 nmol/L and IL11 levels were measured in supernatant on day 13. D, For reversion of EMT, A549 cells were cultured with TGFβ at 2 ng/mL. On day 8, TGFβ at 2 ng/mL along with the test drugs were added at a concentration of 56 nmol/L and IL11 levels were measured in supernatant on day 13. E and F, A549 cells were seeded in serum-free media and treated with TGFβ (10 ng/mL) and BCA101 or cetuximab. After 72 hours of incubation, cells were collected, and mRNA expression of E-cadherin (E) and vimentin (F) was analyzed using a quantitative PCR. G, A549 spheroids were treated with TGFβ (10 ng/mL) along with BCA101 or cetuximab for 72 hours. Spheroids were collected and analyzed for E-cadherin and vimentin by Western blot analysis. In CF, three independent experiments each in triplicate were performed, and data are presented as mean ± SD. H and I, BCA101 inhibits EGF and TGFβ-induced VEGF secretion. A549 spheroids were treated with EGF (5 ng/mL) plus TGFβ (10 ng/mL), followed by BCA101 or cetuximab for 72 hours. Spheroids from three independent experiments were imaged for invasion (H) and supernatant was estimated for VEGF levels (I) by ELISA (R&D Systems) as per the manufacturer's instructions. Statistical analysis for B, one-way ANOVA, nonparametric, Kruskal–Wallis test followed by Dunn multiple comparison test with cells treated with TGFβ as control. For CF, parametric one-way ANOVA was performed. F and I were performed using one-way ANOVA, nonparametric, Kruskal–Wallis test followed by Dunn multiple comparison test. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001. For CF, multiple comparison test was done where mean rank of each column was compared with every other column.

Close modal

EMT transition in A549 cell is also associated with increased IL11 secretion (30). TGFβ has been shown to be a potent inducer of IL11 transcription (30, 31). A549 cells were cultured in the presence of TGFβ and effect of BCA101 on EMT inhibition and reversion was evaluated. Because A549 is a KRAS-mutant cell line, minimal inhibitory effect of BCA101 was expected via EGFR signaling inhibition. A549 cells showed enhanced IL11 secretion upon TGFβ treatment, which was reduced when cells were treated with BCA101 in the presence of extraneous TGFβ, suggesting BCA101-mediated inhibition of IL11 secretion (Fig. 4C). Cetuximab-treated cells did not show this effect. To assess the effect of prolonged treatment with TGFβ, A549 cells were pretreated with TGFβ for 8 days to induce EMT followed by addition of BCA101 or cetuximab. It was observed that while BCA101 could revert the TGFβ-induced IL11 secretion, cetuximab did not show any reversion of EMT (Fig. 4D).

During EMT, cell-cell contact proteins and cytoskeleton proteins are altered to facilitate local dissemination of cells. E-cadherin is a cell-cell adhesion molecule and inhibits migration and metastasis, while vimentin is an intermediate-size filament that is highly expressed in mesenchymal cells. E-cadherin and vimentin are indicators of epithelial and mesenchymal cells, respectively (32). To evaluate the effect of BCA101 on E-cadherin and vimentin transcription and translational levels, cells were treated with TGFβ and BCA101 or cetuximab or TGFβRII-Fc. mRNA and protein levels were monitored by qPCR and Western blot analysis, respectively. TGFβ spiking significantly reduced mRNA levels of E-cadherin and increased mRNA levels of vimentin by 7-fold each (Fig. 4E and F). In presence of extraneous TGFβ, both BCA101 and TGFβRII-Fc reverted E-cadherin levels to the control levels (Fig. 4E). On the other hand, both BCA101 and TGFβRII-Fc treatment supressed TGFβ-induced vimentin to near control levels (Fig. 4F). In line with the mRNA expression data, similar results were also observed with Western blot analysis where TGFβ treatment decreased E-cadherin protein levels and increased vimentin protein levels in A549 spheroids (Fig. 4G). Upon addition of BCA101, TGFβ was neutralized and E-cadherin protein levels were increased, and vimentin protein levels were restored to control expression (Fig. 4G; Supplementary Fig. S4A and S4B). Neither cetuximab nor hIgG control rescued TGFβ-induced EMT in A549 cells.

BCA101 modulates cytokines/chemokines in 3D spheroid PBMC coculture model

Next, we assessed the effect of BCA101 to modulate the immune responses in a 3D spheroid PBMC/A549 coculture model. Using a Magpix-based cytokine/chemokine release assay, categorized as that mediated by TGFβ trap, cetuximab and unique to BCA101, several proinflammatory and prochemoattractant chemokines, including CCL20/MIP-3α, GROα, GROβ, GCSF, IL5, and IL1α were measured (Table 1). These cytokines can be suppressed by TGFβ and sequestration of the same by BCA101 was associated with activation of these cytokines and chemokines. In addition, BCA101 also activated granzyme B, a potent serine protease typically released from NK and cytotoxic T cells to activate tumor cell killing and is known to be inhibited by TGFβ (5). FGF basic and TGFα were also enhanced by BCA101 treatment. The enhanced secretion of TGFα, a potent ligand for EGFR signaling, may be associated with a feedback loop when EGFR was blocked by BCA101. BCA101 also suppressed VEGF, a potent cytokine associated with TGFβ signaling (13). Because EGF and TGFβ are both known to enhance VEGF expression, we next examined the synergistic effect of BCA101 on VEGF secretion upon combination treatment of EGF and TGFβ in vitro.

Table 1.

Unique immune signature for BCA101 observed in presence of TGFβ in 3D spheroid and PBMC coculture model.

BCA101 relative to hIgG
Upregulated Downregulated 
MIP-3a ** VEGF**** 
GRO-a ****  
GRO-b ****  
GRANZYME Ba  
GCSFa **  
FGF basic  
TGFαa  
IL5a  
IL1aa  
BCA101 relative to TGFβ-RII-Fc 
Upregulated Downregulated 
MIP3a * None 
GCSFa 
IL6a 
TGFα  
FGF basic *  
GRANZYME Ba  
BCA101 relative to cetuximab 
Upregulated Downregulated 
MIP3a ** PDGFAB/BBa
GRO-a ****  
GRO-b ****  
GCSF VEGF **** 
IL5a  
FGF basica  
BCA101 relative to hIgG
Upregulated Downregulated 
MIP-3a ** VEGF**** 
GRO-a ****  
GRO-b ****  
GRANZYME Ba  
GCSFa **  
FGF basic  
TGFαa  
IL5a  
IL1aa  
BCA101 relative to TGFβ-RII-Fc 
Upregulated Downregulated 
MIP3a * None 
GCSFa 
IL6a 
TGFα  
FGF basic *  
GRANZYME Ba  
BCA101 relative to cetuximab 
Upregulated Downregulated 
MIP3a ** PDGFAB/BBa
GRO-a ****  
GRO-b ****  
GCSF VEGF **** 
IL5a  
FGF basica  

Note: Statistical analysis was performed between the groups by a t test and P values are presented next to each analyte. *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

aAnalytes at least differentially expressed in two out of three independent experiments.

BCA101 blocks EGF and TGFβ synergistic effects in cocultures

EGF and TGFβ are known to synergistically enhance the tumor invasion and EMT through activation of PI3K and MAPK pathways (11). Furthermore, EGFR and VEGF share common downstream signaling pathway and upregulated EGFR signaling increased secretion of VEGF in turn contributing to EGFR therapy resistance (33). Recently, addition of bevacizumab to EGFR tyrosine kinase inhibitors (TKI) substantially improved progression-free survival in patients with TKI-naïve EGFR-mutant non–small cell lung cancer (NSCLC; ref. 34). Hence, the effect of BCA101 on EGF and TGFβ synergism on EMT and VEGF secretion was evaluated in A549 cell line.

A549 spheroids were treated with TGFβ and EGF at 10 and 5 ng/mL, respectively, and spheroid were imaged for invasion. Supernatants from the same experiment were collected and estimated for VEGF levels. Results show that TGFβ and EGF combination enhanced the spheroid invasion as well as increased VEGF secretion in A549 spheroids compared with either treatment alone spheroids (Fig. 4H). When EGF and TGFβ combination treated spheroids were incubated with BCA101 for 72 hours, combination-induced invasion and VEGF secretion was restored to control spheroids levels, while cetuximab inhibited VEGF to TGFβ-induced levels only (Fig. 4H and I).

BCA101 localizes to tumor and shows sustained TGFβ neutralization in vivo

To evaluate whether BCA101 could localize into tumor area as compared with TGFβRII-Fc, cetuximab, these molecules were labeled were with Alexa Fluor 647 to the similar degree of labeling and equal dose was administered to FaDu xenograft-bearing mice (10 mg/kg, i.p.). Whole mice dorsal imaging was performed at defined timepoints and net intensity was captured (Fig. 5A). Whole animal imaging showed that florescence detection was mainly confined to abdominal cavity and bladder at initial time points (1 to 12 hours) and subsequently to the tumor at later time points (Fig. 5A). The signal intensity, in BCA101 and cetuximab dosed groups were comparable across the duration of the imaging with signal sustained till 96 hours within the tumor (ROI) and subsequently tapered off similarly at 168 hours (Fig. 5A and B). In the animal dosed with TGFβRII-Fc, the signal intensity was maintained till 48 hours and subsequently tapered off from the ROI and was not observable at 168 hours (Fig. 5A and B). Thus, the ability of both BCA101 and cetuximab to be retained in the tumor beyond 96 hours could be attributed to EGFR target engagement.

Figure 5.

BCA101 preferentially localizes to tumor and neutralizes TGFβ. A and B, Biodistribution of BCA101 compared with cetuximab and TGFβRII-Fc. BCA101, cetuximab, and TGFβRII-Fc were labeled with Alexa Flour 647 to the similar degree of labeling and administered to FaDu xenograft-bearing mice (n = 1, 10 mg/kg, i.p.). Whole animal dorsal imaging (A) was performed at defined time points using In-vivo Xtream imaging station (Bruker) and net intensity from tumor-defined ROI (B) obtained at each time point was plotted against time as ln of photons/second/mm2. C–E, TGFβ neutralization in vivo. C, BALB/c mice (n = 3) were injected with BCA101 at 5 mg/kg intravenously and serum samples were collected at defined time points. BCA101 and TGFβ levels in serum were evaluated by ELISA. Data presented as mean ± SD of 3 mice at each time point. Dotted line, TGFβ levels (∼150 pg/mL) in placebo-treated mice. A431 xenograft-bearing mice were treated (intraperitoneally, twice a week, seven doses) with BCA101, TGFβRII-Fc, or cetuximab (10 mg/kg; n = minimum 9) for 19 days and tumors and serum were collected. Tumor lysates (D) and serum (E) were evaluated for TGFβ levels by ELISA. Data are plotted as mean ± SD. Statistical analysis for D and E was performed using one-way ANOVA, nonparametric, Kruskal–Wallis test followed by Dunn multiple comparison test where mean rank of each column was compared with every other column. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤0.001; ****, P ≤ 0.0001.

Figure 5.

BCA101 preferentially localizes to tumor and neutralizes TGFβ. A and B, Biodistribution of BCA101 compared with cetuximab and TGFβRII-Fc. BCA101, cetuximab, and TGFβRII-Fc were labeled with Alexa Flour 647 to the similar degree of labeling and administered to FaDu xenograft-bearing mice (n = 1, 10 mg/kg, i.p.). Whole animal dorsal imaging (A) was performed at defined time points using In-vivo Xtream imaging station (Bruker) and net intensity from tumor-defined ROI (B) obtained at each time point was plotted against time as ln of photons/second/mm2. C–E, TGFβ neutralization in vivo. C, BALB/c mice (n = 3) were injected with BCA101 at 5 mg/kg intravenously and serum samples were collected at defined time points. BCA101 and TGFβ levels in serum were evaluated by ELISA. Data presented as mean ± SD of 3 mice at each time point. Dotted line, TGFβ levels (∼150 pg/mL) in placebo-treated mice. A431 xenograft-bearing mice were treated (intraperitoneally, twice a week, seven doses) with BCA101, TGFβRII-Fc, or cetuximab (10 mg/kg; n = minimum 9) for 19 days and tumors and serum were collected. Tumor lysates (D) and serum (E) were evaluated for TGFβ levels by ELISA. Data are plotted as mean ± SD. Statistical analysis for D and E was performed using one-way ANOVA, nonparametric, Kruskal–Wallis test followed by Dunn multiple comparison test where mean rank of each column was compared with every other column. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤0.001; ****, P ≤ 0.0001.

Close modal

To evaluate in vivo TGFβ neutralizing efficacy of BCA101, Balb/c mice were injected intravenously with BCA101 and bled at defined time points through retro orbital plexus. Serum levels of BCA101 and TGFβ were measured by ELISA. Single dose of BCA101 reduced TGFβ levels to below detection limits as early as first time point (1 hour) and maintained TGFβ levels below detection levels until day 4, after which, TGFβ expression started increasing and reached pretreatment level by day 42. Whereas control mice showed an average of 150 pg/mL TGFβ expression in serum. On the other hand, BCA101 concentration decreased gradually and reached undetectable values by day 7 (Fig. 5C). The amount of BCA101 to neutralize TGFβ by 50% was 0.2 μg/mL. To monitor TGFβ neutralization activity of BCA101 in tumor xenografts, A431 xenograft-bearing mice were treated with BCA101 for 19 days (twice a week) and TGFβ levels were estimated in xenograft lysates and matching sera by ELISA. These A431 cells secrete TGFβ in cultures (Supplementary Fig. S5). BCA101 neutralized tumor and serum TGFβ1 levels at all doses tested compared with control mice in a dose-dependent manner (Fig. 5D and E). BCA101 at a dose of 10, 5, and 1 mg/kg suppressed tumor TGFβ levels by an average of 90%, 83%, and 56%, respectively. TGFβRII-Fc (5 mg/kg) at BCA101 10 mg/kg molar equivalent dose showed 54% suppression, indicating superior TGFβ neutralization by BCA101. Surprisingly, cetuximab-treated mice also showed reduced tumor TGFβ1 levels (56%) compared with control mice (Fig. 5D). This could be attributed to the direct antitumor effects of cetuximab that lead to partially dead tumor, which contributes to total protein but does not secrete TGFβ; however, this observation demands further investigation. Overall, our data indicate BCA101 efficiently neutralized TGFβ in serum and tumor.

BCA101 shows antitumor efficacy in vivo

To demonstrate in vivo antitumor efficacy of BCA101, a cell line (FaDu) expressing EGFR and TGFβ (Supplementary Fig. S5) but without KRAS, EGFR, and BRAF mutations was selected for xenograft studies in nude mice. In Fig. 6A and B and Supplementary Fig. S6A–S6E, we show that BCA101 is not only superior to cetuximab by preventing FaDu xenograft growth over initial TVs but is also superior to the combination of cetuximab and TGFβ trap at equimolar concentrations. The superior effects of BCA101 over cetuximab and TGFβRII-Fc combination can be seen as early as day 4 and is sustained over the duration of the experiment. These results with FaDu in nude mice were rather unexpected because of the lack of functional T cells in these mice and is probably mediated by activation of innate immune responses, including NK-cell activity in this model. Reasons for the better efficacy observed with BCA101 or the combination therapy could be associated with significantly higher apoptotic bodies to mitotic figure ratio as per evaluated tumor histopathology after H&E staining by a blinded pathologist and scorer (Fig. 6C and D). These results suggest that BCA101 by its ability to target EGFR-expressing tumors shows an early and durable response over cetuximab and/or combination with TGFβRII-Fc.

Figure 6.

BCA101 shows superiority over cetuximab in mice xenograft models. A, BCA101 inhibits tumor xenograft growth in vivo. Nude mice were implanted with FaDu cells on the right flank. Once TV reached about 100 mm3, mice were randomized (n = 10) into test and control groups. Test groups were treated (intraperitoneally, twice a week) with BCA101 (50 mg/kg), or equimolar doses of cetuximab (405 mg/kg), TGFβRII-Fc (23 mg/kg) or the combination of cetuximab and TGFβRII-Fc, twice a week, whereas control mice received PBS. TV and mice weight were recorded twice a week. TV data were plotted as mean ± SEM. B, Excised FaDu xenograft tumors. On the terminal day of xenograft study, mice were humanely sacrificed, and tumors were excised. Photographs of excised tumors were taken for each group. C and D, Apoptotic to mitotic ratios for each group. The representative photomicrographic images at ×20 magnification of H&E-stained FaDu xenograft tumor tissue sections from vehicle, cetuximab, TGFβRII-Fc, BCA101, and cetuximab plus TGFβRII-Fc treatment groups. Representative mitotic cells at different stages and apoptotic cells are indicated in green and red arrowheads, respectively. Scale bar, 100 μm. D, The box and whisker plot show the minimum and maximum ratio of apoptotic and mitotic figures in FaDu xenograft tumor-bearing mice (n = 5). E and F, PDX Study-1. Freshly isolated tumors from patients (head and neck cancers) were minced into small pieces and engrafted into female NOG mice. Once tumor reached about 130 mm3, mice were randomized into control and test groups. Test group mice were treated with either BCA101 (10 mg/kg; E) or equimolar dose of cetuximab (8.1 mg/kg; F), twice a week (intraperitoneally), whereas control animals received placebo alone. Mice were treated for 27 days followed by a treatment-free phase until day 79. TVs and mice weight were recorded twice a week. Vertical dotted line on x-axis indicates treatment cessation day and horizontal dotted line on y-axis shows initial TV. G and H, PDX Study-2. Patient-derived head and neck tumor fragments were implanted into NOG mice and mice were randomized into test and control groups once tumors reached about 120 mm3. Test groups were treated with either BCA101 (G) or cetuximab (H) as mentioned above for 34 days followed by a treatment-free phase until day 108. Individual mouse tumor growth kinetics are presented. First vertical dotted line on x-axis (blue) indicates treatment cessation day and second dotted line on x-axis shows tumor doubling time. Horizontal dotted line on y-axis (green) shows initial TV, and second dotted line (red) shows doubled TV. For A, statistical analysis was performed using repeated measures two-way ANOVA followed by Bonferroni multiple comparison test. *, treated groups were compared with PBS-treated mice; #, BCA101 group was compared with BCA101 and TGFβRII-Fc combination-treated mice. For D, statistical analyses were performed using one-way ANOVA nonparametric test followed by Dunn multiple comparisons tests to compare mean rank of one treatment group with mean rank of every other group: Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.

Figure 6.

BCA101 shows superiority over cetuximab in mice xenograft models. A, BCA101 inhibits tumor xenograft growth in vivo. Nude mice were implanted with FaDu cells on the right flank. Once TV reached about 100 mm3, mice were randomized (n = 10) into test and control groups. Test groups were treated (intraperitoneally, twice a week) with BCA101 (50 mg/kg), or equimolar doses of cetuximab (405 mg/kg), TGFβRII-Fc (23 mg/kg) or the combination of cetuximab and TGFβRII-Fc, twice a week, whereas control mice received PBS. TV and mice weight were recorded twice a week. TV data were plotted as mean ± SEM. B, Excised FaDu xenograft tumors. On the terminal day of xenograft study, mice were humanely sacrificed, and tumors were excised. Photographs of excised tumors were taken for each group. C and D, Apoptotic to mitotic ratios for each group. The representative photomicrographic images at ×20 magnification of H&E-stained FaDu xenograft tumor tissue sections from vehicle, cetuximab, TGFβRII-Fc, BCA101, and cetuximab plus TGFβRII-Fc treatment groups. Representative mitotic cells at different stages and apoptotic cells are indicated in green and red arrowheads, respectively. Scale bar, 100 μm. D, The box and whisker plot show the minimum and maximum ratio of apoptotic and mitotic figures in FaDu xenograft tumor-bearing mice (n = 5). E and F, PDX Study-1. Freshly isolated tumors from patients (head and neck cancers) were minced into small pieces and engrafted into female NOG mice. Once tumor reached about 130 mm3, mice were randomized into control and test groups. Test group mice were treated with either BCA101 (10 mg/kg; E) or equimolar dose of cetuximab (8.1 mg/kg; F), twice a week (intraperitoneally), whereas control animals received placebo alone. Mice were treated for 27 days followed by a treatment-free phase until day 79. TVs and mice weight were recorded twice a week. Vertical dotted line on x-axis indicates treatment cessation day and horizontal dotted line on y-axis shows initial TV. G and H, PDX Study-2. Patient-derived head and neck tumor fragments were implanted into NOG mice and mice were randomized into test and control groups once tumors reached about 120 mm3. Test groups were treated with either BCA101 (G) or cetuximab (H) as mentioned above for 34 days followed by a treatment-free phase until day 108. Individual mouse tumor growth kinetics are presented. First vertical dotted line on x-axis (blue) indicates treatment cessation day and second dotted line on x-axis shows tumor doubling time. Horizontal dotted line on y-axis (green) shows initial TV, and second dotted line (red) shows doubled TV. For A, statistical analysis was performed using repeated measures two-way ANOVA followed by Bonferroni multiple comparison test. *, treated groups were compared with PBS-treated mice; #, BCA101 group was compared with BCA101 and TGFβRII-Fc combination-treated mice. For D, statistical analyses were performed using one-way ANOVA nonparametric test followed by Dunn multiple comparisons tests to compare mean rank of one treatment group with mean rank of every other group: Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.

Close modal

BCA101 has sustainable tumor regression and delayed tumor relapse compared with cetuximab in HNSCC-PDX models

EGFR is an established therapeutic target in HNSCC and cetuximab was approved for HNSCC, with an objective response rate of 13% in the monotherapy setting (35). Unfortunately, patients who respond to cetuximab eventually develop resistance due to various reasons. One of the major factors contributing to cetuximab resistance is TGFβ secreted by CAFs in TME (9, 36, 37). Progression after cetuximab treatment is associated with increased TGFβ signaling secreted by CAFs in stroma and blocking the TGFβ signaling was sufficient to enhance the efficacy of cetuximab in HNSCC PDX-bearing mice (9). Because fresh PDX models retain CAFs from patients, antitumor efficacy of BCA101 was evaluated in HNSCC-PDX models.

To evaluate whether BCA101 has sustainable tumor regression compared with cetuximab, three HNSCC PDX studies were performed: two studies with treatment-naïve samples and one study with metastatic recurrent tumor sample. A summary description of the PDX models is presented in Supplementary Table S1. Basal phosphorylation levels of EGFR (pEGFR) and SMAD (pSMAD2) in the PDX models are presented in Supplementary Fig. S7A and S7B. All PDX models were treated with BCA101, TGFβRII-Fc or cetuximab or combination of cetuximab and TGFβRII-Fc at equimolar doses and tumor relapse or delay in tumor growth was monitored in the treatment-free phase. In PDX-1 study, mice were treated for 27 days with either BCA101 or cetuximab and left untreated until 79 days to monitor the tumor regrowth. Initially, BCA101 and cetuximab treatment resulted in similar tumor growth inhibition (TGI) during the treatment phase (Supplementary Fig. S8). On day 79, BCA101 showed significantly higher antitumor activity compared with cetuximab (TV 67 vs. 146 mm3; P < 0.05). Importantly, during the treatment-free phase, until day 79, only one of nine mice tumors regrew beyond the tumor starting volume (130 mm3) in BCA101-treated mice (Fig. 6E), whereas six of 10 tumors regrew beyond starting volume in cetuximab-treated mice (Fig. 6F) and 2 of 10 mice regrew in cetuximab plus TGFβRII-Fc combination treated mice (Supplementary Fig. S9A), indicating durable tumor regression in BCA101-treated mice compared with cetuximab. A χ2 test with one-tailed distribution (P ≤ 0.05) suggested a difference between the BCA101- and cetuximab-treated groups during the treatment-free phase. On the other hand, in PDX2 study where mice were treated for 34 days and left untreated until day 108 all mice tumors regrew beyond initial volume in BCA101 and cetuximab-treated groups (Fig. 6G and H) and TGI was initially similar for both BCA101 and cetuximab during treatment phase (Supplementary Fig. S10). On the day of dose cessation (day 34), BCA101-treated mice had statistically significant higher antitumor activity compared with cetuximab (TV 134 vs. 210 mm3). Because all tumors regrew in this study, TV doubling time was compared between BCA101 and cetuximab. Results showed that tumor doubling time was 68 days in BCA101 and 47 days in cetuximab-treated group, indicating BCA101 has sustained effect compared with cetuximab (Fig. 6G and H). Doubling time for cetuximab and TGFβRII-Fc group showed 63 days (Supplementary Fig. S9B). Finally, BCA101 and cetuximab antitumor efficacy was compared in a highly treated (chemo and radiation) recurrent metastatic PDX model (PDX-3) of HNSCC. In this model, while TGI profiles were similar for BCA101 and cetuximab (Supplementary Fig. S11), 2 of 10 mice relapsed in the cetuximab-treated group and no tumor relapse was observed in BCA101-treated mice. These data collectively demonstrate that BCA101 treatment can result in sustained antitumor effect and delayed tumor growth compared with cetuximab in HNSCC PDX models.

BCA101 has superior antitumor efficacy than cetuximab in a human immune/tumor cell coimplantation model

BCA101 is not a murine EGFR cross reactive antibody and hence to mimic human immune system in mice, a coimplantation model was developed where human PBMCs were mixed with tumor cells and implanted into nude mice. In HT-29 (a BRAF-mutated colorectal cancer cell line) coimplantation model, BCA101 had significantly higher antitumor efficacy as compared with cetuximab from day 11 to day 17. On day 17, BCA101 showed a 31% TGI whereas cetuximab had a TGI of 8%, indicating superior efficacy of BCA101 over cetuximab (Supplementary Fig. S12A). Because PBMCs have donor to donor variations, we repeated the experiment with a different donor derived PBMCs and obtained similar results (Supplementary Fig. S12B). To confirm that the superior efficacy of BCA101 over cetuximab was attributed to the locally reconstituted immune system, HT-29 cells alone (without PBMCs) were implanted in nude mice and treated with BCA101 and cetuximab. In this study, BCA101 did not show any superiority over cetuximab in antitumor activity, confirming the role of immune system in discerning the mechanism of action for BCA101 (Supplementary Fig. S13).

BCA101 exhibits better antitumor efficacy in combination with PD-1/PD-L1 inhibitor in vitro and in vivo

Literature reports suggest that EGFR signaling on tumor cells results in upregulation of suppressive checkpoint receptor ligand, programmed death ligand (PD-L1), and induces secretion of inhibitory molecules such as TGFβ (5, 38). Targeted removal of TGFβ from TME would enhance the efficacy of immune therapies. Hence, BCA101efficacy was evaluated in combination with immune checkpoint inhibitors.

Combinatorial effect of BCA101 with an anti-PD-L1 mAb was evaluated in vitro human PBMC (Fig. 7A) and PBMC/HCT-116 coculture assays (Fig. 7B). Human PBMCs were stimulated with soluble anti-CD3 in presence or absence of antibody drugs and spiked TGFβ. IFNγ release was measured in the supernatant as a readout of immune activation. TGFβ-induced immune suppression was rescued by BCA101 and was further enhanced when BCA101 was used in combination with anti PD-L1 mAb (Fig. 7A). Combination of BCA101 with anti PD-L1 antibody also showed enhancement of immune response in PBMC/HCT-116 coculture assays (Fig. 7B). In both studies, combinatorial treatment of cells with BCA101 and anti-PD-L1 showed enhanced response than BCA101 or anti-PD-L1 treatment alone (Fig. 7A and B). These results suggested a potential role of BCA101 as a therapeutic option in combination with immune checkpoint therapies.

Figure 7.

BCA101 in combination with immune checkpoint therapeutics. A and B, BCA101 and immune checkpoint inhibitor combination studies in vitro. In A and B, human PBMCs were stimulated with soluble anti-CD3 (2 ng/mL). A, Exogenous TGFβ added at 10 ng/mL and drugs were added to a final concentration of 56 nmol/L. IFNγ release was measured in supernatant after about 72 hours using ELISA. Absorbance values are plotted as mean ± SD from four independent experiments in triplicates. B, Activated human PBMCs were cocultured with HCT116 cells in presence of TGFβ spiked at 1 ng/mL. Drugs were added to a final concentration of 6.667 nmol/L. IFNγ level was measured in the supernatant after about 72 hours and fold increase was calculated over hIgG for the test groups. Data were plotted as mean ± SD of six independent experiments each in triplicate. C and D, Efficacy of BCA101 in B16-EGFR mouse model. B-16 cells expressing human EGFR were implanted into right flank of C57 mice and mice were randomized into test and control groups when tumors were palpable (nonmeasurable). Test group mice were treated with BCA101 (50 mg/kg) and equimolar doses of cetuximab (40.5 mg/kg), TGFβRII-Fc (23 mg/kg), anti-PD1 (10 mg/kg) or combinations as indicated. Mice were treated with test and placebo compounds and TV and body weight were measured twice a week. Kaplan–Meier curves were plotted as a fraction of mice achieving TV ≥ 300 mm3 or death as an event. Blue arrows, dosing day. Mice were treated for 14 days and left untreated until day 28 to check for durability of response. D, Figure showing P values relative to placebo for each group. E, Figure showing number of mice at risk on a given day. Statistical analysis for A and B, one-way ANOVA, parametric with multiple comparisons test was performed, with mean rank of each column compared with every other column. For table in D, χ2 followed by Fisher exact test was performed, with TV ≥ 300 mm3 or death as an event compared with IgG group to each group. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ****, P ≤ 0.0001; ns, nonsignificant.

Figure 7.

BCA101 in combination with immune checkpoint therapeutics. A and B, BCA101 and immune checkpoint inhibitor combination studies in vitro. In A and B, human PBMCs were stimulated with soluble anti-CD3 (2 ng/mL). A, Exogenous TGFβ added at 10 ng/mL and drugs were added to a final concentration of 56 nmol/L. IFNγ release was measured in supernatant after about 72 hours using ELISA. Absorbance values are plotted as mean ± SD from four independent experiments in triplicates. B, Activated human PBMCs were cocultured with HCT116 cells in presence of TGFβ spiked at 1 ng/mL. Drugs were added to a final concentration of 6.667 nmol/L. IFNγ level was measured in the supernatant after about 72 hours and fold increase was calculated over hIgG for the test groups. Data were plotted as mean ± SD of six independent experiments each in triplicate. C and D, Efficacy of BCA101 in B16-EGFR mouse model. B-16 cells expressing human EGFR were implanted into right flank of C57 mice and mice were randomized into test and control groups when tumors were palpable (nonmeasurable). Test group mice were treated with BCA101 (50 mg/kg) and equimolar doses of cetuximab (40.5 mg/kg), TGFβRII-Fc (23 mg/kg), anti-PD1 (10 mg/kg) or combinations as indicated. Mice were treated with test and placebo compounds and TV and body weight were measured twice a week. Kaplan–Meier curves were plotted as a fraction of mice achieving TV ≥ 300 mm3 or death as an event. Blue arrows, dosing day. Mice were treated for 14 days and left untreated until day 28 to check for durability of response. D, Figure showing P values relative to placebo for each group. E, Figure showing number of mice at risk on a given day. Statistical analysis for A and B, one-way ANOVA, parametric with multiple comparisons test was performed, with mean rank of each column compared with every other column. For table in D, χ2 followed by Fisher exact test was performed, with TV ≥ 300 mm3 or death as an event compared with IgG group to each group. Significance is indicated by *, P ≤ 0.05; **, P ≤ 0.01; ****, P ≤ 0.0001; ns, nonsignificant.

Close modal

The syngeneic tumor model utilizing the murine melanoma B16 cell line expressing human EGFR (B16-hEGFR) has been reported earlier (39). We used this model to determine the antitumor efficacy of BCA101 in combination with an anti-PD1 mAb. Additional controls included monotherapy of cetuximab, TGFβRII-Fc, anti-PD1, and combination of cetuximab and TGFβRII-Fc. The experimental design had a treatment phase of 2 weeks and treatment-free phase for an additional 2 weeks to evaluate durability of response. The placebo group showed a robust tumor uptake with 5 of 9 mice crossing the ethical endpoint of TV > 1,500 mm3 by day 20 after initiation of treatment. Kaplan–Meier curves were therefore used to check for efficacy based on achieving an event of TV ≥ 300 mm3 or death. Figure 7C indicates that while anti-PD1 monotherapy is superior to other monotherapies of TGFβRII-Fc, cetuximab, and BCA101, the combination of BCA101 with anti-PD1 showed the best response with only 1 of 9 mice achieving an event. Survival curves indicate a significance difference between groups with P value of 0.0009 (log-rank, Mantel–Cox) test. Figure 7D shows different treatment groups as compared with placebo in a χ2 test for significance based on event occurring as mentioned before, with the best outcome in the BCA101 plus anti-PD1 combination treated group. Spider plots of individual mouse in each group are shown in Supplementary Fig. S14A–S14H. Number of mice at risk at each indicated time point is presented in Fig. 7E. In summary, this syngeneic animal model with B16-hEGFR cells responded to anti-EGFR treatment. This is contrary to reports published earlier (39) and this difference could be attributed to the early treatment setting in the current study.

The improved antitumor efficacy of BCA101 and anti-PD1 mAb combination over monotherapy of BCA101 or anti-PD1 mAb was confirmed in a humanized HuNOG-EXL mouse xenograft model using prostate cancer PC-3 cell line, which is known to be partially responsive to anti-PD1 therapy. Treatment with BCA101, anti-PD1 mAb or combination of BCA101 and anti-PD1 mAb resulted in significant antitumor activity with %TGI of 40, 52 and 78% on day 18, respectively (Supplementary Fig. S15). These treatments were associated with acceptable weight loss within 15% (Supplementary Fig. S16).

The addition of TGFβ trap in BCA101 is not associated with additional toxicologic findings in non-Good Laboratory Practice monkey studies

Following a single intravenous infusion of BCA101 to cynomolgus monkeys, serum mean maximum concentrations and mean exposures of the molecule increased dose proportionally (Supplementary Fig. S17), with no saturation of the routes of elimination. The observed volumes of distribution (33–74 mL/kg) suggest that BCA101 is well distributed to the tissues. Half-life of BCA101 ranged from 12.9 to 84.5 hours for the doses tested.

Once weekly administration of BCA101 to cynomolgus monkeys for 29 days was clinically well tolerated with reversible BCA101-mediated effects. The MTD and no-observed-adverse-effect-level were not established with the doses tested and the highest severely nontoxic dose was 100 mg/kg/occasion. BCA101 toxicity profile was comparable with either cetuximab or anti-TGFβ treatment profiles and no synergistic safety alerts were observed.

BCA101 to the best of our knowledge is a first-in-class bifunctional therapeutic molecule targeting TGFβ and EGFR signaling simultaneously. A cross-talk between EGFR and TGFβ in promoting tumor and immune cell evasion is well documented in the literature. BCA101 is therefore conceptualized to leverage the cooperative/synergistic effects of EGFR and TGFβ signaling, which are known to sustain and promote tumorigenesis.

Over the last two decades, many therapeutic strategies were developed to inhibit EGFR and TGFβ signaling pathways independently. Anti-EGFR therapies such as cetuximab and panitumumab were approved for metastatic colorectal cancers with wild-type KRAS. However, anti-EGFR therapy was effective in only 10% to 15% of unselected patients and is ineffective against mutant KRAS tumors (40). Moreover, resistance to EGFR blockade inevitably occurs with both anti-EGFR targeting mAbs as well as EGFR-targeted small molecules (41). Similarly, different approaches have been developed to block TGFβ signaling. However, none of these strategies have been successful in clinical trials. The minor clinical benefits of TGFβ inhibition were often outweighed by off-target toxicities such as cardiotoxicity and formation of benign tumors (42), due to systemic neutralization of TGFβ isoforms. Considering the potential cross-talk between these two pathways, and involvement of TGFβ in cetuximab resistance, Bedi and colleagues, 2012 (5) evaluated the efficacy of cetuximab and anti-TGFβ antibody combination in a HNSCC tumor xenograft mouse model. Combination of cetuximab and anti-TGFβ not only prevented the emergence of resistant tumor cells but also induced complete regression. However, anti-TGFβ antibody in this combination model still carries the risk of pan-TGFβ inhibition related toxicity. Unlike the above combination, BCA101 delivers TGFβRIIECD preferentially to TME, as demonstrated in whole mice imaging studies and tumor TGFβ neutralization studies. In addition, literature suggests that cetuximab-treated tumors develop and select for high TGFβ-secreting cells, leading to cetuximab resistance and eventually tumor relapse. Furthermore, Yegodayev and colleagues showed that the TGFβ signaling pathway was upregulated in the stromal cells of PDXs that were progressed on cetuximab treatment and TGFβ signaling pathway inhibition by SMAD inhibitor was sufficient to restore the cetuximab sensitivity (9). Consistent with above observations, our PDX and syngeneic tumor (B16-hEGFR) studies indicated that relapse rate of tumors in BCA101-treated mice was minimal compared with cetuximab-treated mice probably due to TGFβ neutralization in tumor stromal compartment, suggesting durability of response with BCA101.

During early tumorigenesis, TGFβ has a complex role in tumor prevention versus tumorigenesis. However, at the later disease stage, TGFβ predominantly contributes to tumorigenesis via multiple mechanisms, including a direct effect of tumor cells by inducing tumor cell proliferation, promoting angiogenesis, EMT, tumor cell metastasis, and drug resistance (43, 44). Consistent with other reports where TGFβ signaling blockade was sufficient to inhibit EMT (45), BCA101 rescued TGFβ-induced EMT in A549 spheroid cultures. Furthermore, TGFβ-mediated angiogenesis was reported to be mediated through VEGF as TGFβ directly regulates VEGF expression. Recently, several reports showed that EGFR signaling ligands TGFα and EGF are also potent activators of proangiogenic factors including VEGF (46, 47). In fact, anti-EGFR agents significantly reduced the secretion of VEGF in tumor cells (48). Hence, presence of both EGF and TGFβ in the TME may further potentiate tumor angiogenesis and metastasis. Our results showed that indeed EGF and TGFβ combination significantly enhanced tumor cell invasion and the secretion of VEGF in A549 spheroids, which was rescued by addition of BCA101, but not cetuximab. Moreover, Derangere and colleagues, showed that VEGF-A can induce resistance toward cetuximab cytotoxicity on KRAS and NRAS wild-type colon cancer cell lines (49). Coupled with anti-EMT potentials and VEGF inhibition properties, BCA101 could potentially reduce the tumor metastasis and resistance compared with cetuximab.

The other increasingly appreciated function of TGFβ is its potent ability to suppress immune cell responses. TGFβ has been shown to switch the polarization of CD4+ T cell from Th1 to Th2 phenotype, driven by IL10 release and direct inhibition of the Th1 response (50, 51). We show BCA101 inhibited TGFβ-mediated immune suppression and increased IFNγ secretion by activated T cells in coculture assay, suggesting an enhanced Th1 response. Three-dimensional spheroid coculture experiments indicated that Th cells, macrophages and neutrophils mediated proinflammatory cytokine/chemokine are significantly elevated by BCA101 potentially enabling the lymphocyte trafficking within the TME. TGFβ is known to suppress cytotoxic CD8+ T lymphocytes (CTL) and NK cells by downregulation of NKG2D expression and inhibition of granzyme B and IFNγ (52–55). BCA101, due to neutralization of TGFβ, showed upregulation of NKG2D expression on activated CD8+ T cells, increased secretion of IFNγ and granzyme B and enhanced cytolytic killing of tumor cells compared with cetuximab. BCA101 also inhibited TGFβ-mediated differentiation of naïve CD4+ T cells to iTreg, reducing Treg suppressive function in vitro (56). Thus, BCA101 can suppress tumor growth and restore antitumor immune cell functions by exploiting several tumor cell intrinsic and extrinsic mechanisms.

EGFR signaling is also shown to contribute to immune escape mechanisms (4) by downregulation of tumor antigen presentation, upregulation of PD-L1 (38, 57) and by regulating the suppressive function of natural Tregs (58). Thus, EGFR-specific mAb therapies not only inhibit intrinsic oncogenic signaling but also activate immune cells that mediate tumor clearance, and thus are being combined with PD-1 blockade therapy to improve clinical efficacy. Consistent with recent clinical trials data (59, 60), our data showed that BCA101 combination enhances the efficacy of immune checkpoint inhibitors, which may be attributed to inhibition of both EGFR and TGFβ-induced immunosuppressive effects.

Currently, a similar molecule M7824, which binds to PD-L1 antigen and sequesters TGFβ though TGF-βRIIECD, is in phase III clinical trial with mixed evidence of clinical trial success. Unlike the more ubiquitous, varied, and inducible expression of PD-L1 on tumor and immune cell types by IFNγ, EGFR is mostly restricted to tumor cells with significant number of patients across tumor types having high expression of EGFR. Therefore, inhibition of EGFR signaling along with potential immune activation by TGFβ neutralization by BCA101 may result in improved clinical efficacy with a better safety profile. We believe that careful patient stratification based on patients expressing EGFR and having an active TGFβ signaling pathway may lead to better clinical success with BCA101.

R. Nair reports a patent for US11060097B2 issued to Biocon Limited. U. Bughani reports a patent for Targeted/immunomodulatory fusion proteins in combination with additional therapeutic agents and methods for making same issued. U. Bughani owns ESOP (employee stocks) as issued by Biocon Limited. S.-L. Tan reports personal fees from Bicara Therapeutics during the conduct of the study and from Bicara Therapeutics outside the submitted work. S.-L. Tan was the Chief Scientific Officer of Bicara Therapeutics during which the research described in this manuscript was performed. P. Nair owns ESOPS (employee stocks) from Biocon Limited. No disclosures were reported by the other authors.

S.R. Boreddy: Conceptualization, formal analysis, supervision, investigation, visualization, methodology, writing–review and editing. R. Nair: Conceptualization, formal analysis, investigation, methodology, writing–review and editing. P.K. Pandey: Formal analysis, investigation, methodology. A. Kuriakose: Formal analysis, investigation, methodology. S.B. Marigowda: Conceptualization, formal analysis, methodology. C. Dey: Formal analysis, investigation, methodology. A. Banerjee: Conceptualization, supervision, investigation, methodology. H. Kulkarni: Investigation, visualization, methodology. M. Sagar: Investigation, visualization, methodology. S.R. Krishn: Investigation, visualization, methodology. S. Rao: Formal analysis, investigation, methodology. M. A R: Formal analysis, investigation, methodology. V. Tiwari: Investigation, methodology. B. Alke: Investigation. P.K. MV: Investigation, visualization, methodology. M. Shri: Investigation, visualization, methodology. C. Dhamne: Investigation, visualization, methodology. S. Patel: Investigation, visualization, methodology. P. Sharma: Investigation, visualization, methodology. S. Periyasamy: Investigation, visualization, methodology. J. Bhatnagar: Conceptualization, validation, investigation, methodology. M.A. Kuriakose: Resources. R.B. Reddy: Resources. A. Suresh: Resources. S. Sreenivas: Resources, formal analysis, investigation, methodology. N. Govindappa: Investigation, visualization, methodology. P.R. Moole: Conceptualization, supervision, investigation, methodology. U. Bughani: Conceptualization, formal analysis, validation, methodology. S.-L. Tan: Conceptualization, validation, writing–review and editing. P. Nair: Conceptualization, data curation, formal analysis, supervision, validation, visualization, methodology, writing–original draft, writing–review and editing.

The authors would like to acknowledge contributions from Dr Sreesha Srinivasa, Dr. Vasan Sambandamurthy, Pallavi Misra, Poulami Manna, and Mary Ros. The authors also would like to thank Syngene for use of animal facility, Prof. Yang-Xin Fu, University of Chicago for B16-EGFR cell line, and Bicara Therapeutics and Biocon Limited for funding.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

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

1.
Joo
WD
,
Visintin
I
,
Mor
G
.
Targeted cancer therapy–are the days of systemic chemotherapy numbered?
Maturitas
2013
;
76
:
308
14
.
2.
Leonard
B
,
Brand
TM
,
O'Keefe
RA
,
Lee
ED
,
Zeng
Y
,
Kemmer
JD
, et al
.
BET inhibition overcomes receptor tyrosine kinase-mediated cetuximab resistance in HNSCC
.
Cancer Res
2018
;
78
:
4331
43
.
3.
Wheeler
DL
,
Huang
S
,
Kruser
TJ
,
Nechrebecki
MM
,
Armstrong
EA
,
Benavente
S
, et al
.
Mechanisms of acquired resistance to cetuximab: role of HER (ErbB) family members
.
Oncogene
2008
;
27
:
3944
56
.
4.
Concha-Benavente
F
,
Ferris
RL
.
Reversing EGFR mediated immunoescape by targeted monoclonal antibody therapy
.
Front Pharmacol
2017
;
8
:
332
.
5.
Bedi
A
,
Chang
X
,
Noonan
K
,
Pham
V
,
Bedi
R
,
Fertig
EJ
, et al
.
Inhibition of TGF-beta enhances the in vivo antitumor efficacy of EGF receptor-targeted therapy
.
Mol Cancer Ther
2012
;
11
:
2429
39
.
6.
Zhao
Y
,
Ma
J
,
Fan
Y
,
Wang
Z
,
Tian
R
,
Ji
W
, et al
.
TGF-beta transactivates EGFR and facilitates breast cancer migration and invasion through canonical Smad3 and ERK/Sp1 signaling pathways
.
Mol Oncol
2018
;
12
:
305
21
.
7.
Jie
HB
,
Schuler
PJ
,
Lee
SC
,
Srivastava
RM
,
Argiris
A
,
Ferrone
S
, et al
.
CTLA-4(+) Regulatory T cells increased in cetuximab-treated head and neck cancer patients suppress NK cell cytotoxicity and correlate with poor prognosis
.
Cancer Res
2015
;
75
:
2200
10
.
8.
Byers
LA
,
Diao
L
,
Wang
J
,
Saintigny
P
,
Girard
L
,
Peyton
M
, et al
.
An epithelial-mesenchymal transition gene signature predicts resistance to EGFR and PI3K inhibitors and identifies Axl as a therapeutic target for overcoming EGFR inhibitor resistance
.
Clin Cancer Res
2013
;
19
:
279
90
.
9.
Yegodayev
KM
,
Novoplansky
O
,
Golden
A
,
Prasad
M
,
Levin
L
,
Jagadeeshan
S
, et al
.
TGF-beta-activated cancer-associated fibroblasts limit cetuximab efficacy in preclinical models of head and neck cancer
.
Cancers
2020
;
12
:
339
.
10.
Richter
P
,
Umbreit
C
,
Franz
M
,
Berndt
A
,
Grimm
S
,
Uecker
A
, et al
.
EGF/TGFbeta1 co-stimulation of oral squamous cell carcinoma cells causes an epithelial-mesenchymal transition cell phenotype expressing laminin 332
.
J Oral Pathol Med
2011
;
40
:
46
54
.
11.
Uttamsingh
S
,
Bao
X
,
Nguyen
KT
,
Bhanot
M
,
Gong
J
,
Chan
JL
, et al
.
Synergistic effect between EGF and TGF-beta1 in inducing oncogenic properties of intestinal epithelial cells
.
Oncogene
2008
;
27
:
2626
34
.
12.
Xu
Z
,
Jiang
Y
,
Steed
H
,
Davidge
S
,
Fu
Y
.
TGFbeta and EGF synergistically induce a more invasive phenotype of epithelial ovarian cancer cells
.
Biochem Biophys Res Commun
2010
;
401
:
376
81
.
13.
Neuzillet
C
,
Tijeras-Raballand
A
,
Cohen
R
,
Cros
J
,
Faivre
S
,
Raymond
E
, et al
.
Targeting the TGFbeta pathway for cancer therapy
.
Pharmacol Ther
2015
;
147
:
22
31
.
14.
Letourneur
O
,
Goetschy
JF
,
Horisberger
M
,
Grutter
MG
.
Ligand-induced dimerization of the extracellular domain of the TGF-beta receptor type II
.
Biochem Biophys Res Commun
1996
;
224
:
709
16
.
15.
Lo Nigro
C
,
Macagno
M
,
Sangiolo
D
,
Bertolaccini
L
,
Aglietta
M
,
Merlano
MC
.
NK-mediated antibody-dependent cell-mediated cytotoxicity in solid tumors: biological evidence and clinical perspectives
.
Ann Transl Med
2019
;
7
:
105
.
16.
Liu
SD
,
Chalouni
C
,
Young
JC
,
Junttila
TT
,
Sliwkowski
MX
,
Lowe
JB
.
Afucosylated antibodies increase activation of FcgammaRIIIa-dependent signaling components to intensify processes promoting ADCC
.
Cancer Immunol Res
2015
;
3
:
173
83
.
17.
Lan
Y
,
Zhang
D
,
Xu
C
,
Hance
KW
,
Marelli
B
,
Qi
J
, et al
.
Enhanced preclinical antitumor activity of M7824, a bifunctional fusion protein simultaneously targeting PD-L1 and TGF-beta
.
Sci Transl Med
2018
;
10
:
eaan5488
.
18.
Huynh
LK
,
Hipolito
CJ
,
Ten Dijke
P
.
A perspective on the development of TGF-beta inhibitors for cancer treatment
.
Biomolecules
2019
;
9
:
743
.
19.
Pang
X
,
Tang
YL
,
Liang
XH
.
Transforming growth factor-beta signaling in head and neck squamous cell carcinoma: Insights into cellular responses
.
Oncol Lett
2018
;
16
:
4799
806
.
20.
Mistry
AR
,
O'Callaghan
CA
.
Regulation of ligands for the activating receptor NKG2D
.
Immunology
2007
;
121
:
439
47
.
21.
Jamieson
AM
,
Diefenbach
A
,
McMahon
CW
,
Xiong
N
,
Carlyle
JR
,
Raulet
DH
.
The role of the NKG2D immunoreceptor in immune cell activation and natural killing
.
Immunity
2002
;
17
:
19
29
.
22.
Vignali
DA
,
Collison
LW
,
Workman
CJ
.
How regulatory T cells work
.
Nat Rev Immunol
2008
;
8
:
523
32
.
23.
Bommireddy
R
,
Doetschman
T
.
TGFbeta1 and treg cells: alliance for tolerance
.
Trends Mol Med
2007
;
13
:
492
501
.
24.
Wan
YY
,
Flavell
RA
. '
Yin-Yang' functions of transforming growth factor-beta and T regulatory cells in immune regulation
.
Immunol Rev
2007
;
220
:
199
213
.
25.
Saleh
R
,
Elkord
E
.
FoxP3(+) T regulatory cells in cancer: Prognostic biomarkers and therapeutic targets
.
Cancer Lett
2020
;
490
:
174
85
.
26.
Kaluri
R
,
Weinberg
RA
.
The basics of epithelial-mesenchymal transition
.
J Clin Invest
2010
;
120
:
1786
.
27.
Kalluri
R
,
Weinberg
RA
.
The basics of epithelial-mesenchymal transition
.
J Clin Invest
2009
;
119
:
1420
8
.
28.
Michael
K
Wendt
TMA
,
Schiemann
WP
.
Mechanisms of epithelial-mesenchymal transition by TGF-β
.
Future Oncol
2009
;
5
:
1145
68
.
29.
Wendt
MK
,
Allington
TM
,
Schiemann
WP
.
Mechanisms of the epithelial-mesenchymal transition by TGF-beta
.
Future Oncol
2009
;
5
:
1145
68
.
30.
Yashiro
R
,
Nagasawa
T
,
Kiji
M
,
Hormdee
D
,
Kobayashi
H
,
Koshy
G
, et al
.
Transforming growth factor-beta stimulates interleukin-11 production by human periodontal ligament and gingival fibroblasts
.
J Clin Periodontol
2006
;
33
:
165
71
.
31.
Tang
W
,
Yang
L
,
Yang
YC
,
Leng
SX
,
Elias
JA
.
Transforming growth factor-beta stimulates interleukin-11 transcription via complex activating protein-1-dependent pathways
.
J Biol Chem
1998
;
273
:
5506
13
.
32.
Nijkamp
MM
,
Span
PN
,
Hoogsteen
IJ
,
van der Kogel
AJ
,
Kaanders
JH
,
Bussink
J
.
Expression of E-cadherin and vimentin correlates with metastasis formation in head and neck squamous cell carcinoma patients
.
Radiother Oncol
2011
;
99
:
344
8
.
33.
Chen
D
,
Gu
K
,
Wang
H
.
Optimizing sequential treatment with anti-EGFR and VEGF mAb in metastatic colorectal cancer: current results and controversies
.
Cancer Manag Res
2019
;
11
:
1705
16
.
34.
Le
X
,
Nilsson
M
,
Goldman
J
,
Reck
M
,
Nakagawa
K
,
Kato
T
, et al
.
Dual EGFR-VEGF pathway inhibition: a promising strategy for patients with EGFR-Mutant NSCLC
.
J Thorac Oncol
2021
;
16
:
205
15
.
35.
Vermorken
JB
,
Trigo
J
,
Hitt
R
,
Koralewski
P
,
Diaz-Rubio
E
,
Rolland
F
, et al
.
Open-label, uncontrolled, multicenter phase II study to evaluate the efficacy and toxicity of cetuximab as a single agent in patients with recurrent and/or metastatic squamous cell carcinoma of the head and neck who failed to respond to platinum-based therapy
.
J Clin Oncol
25
:
16s
,
2007
(
suppl; abstr 2171–7
).
36.
Erdogan
B
,
Ao
M
,
White
LM
,
Means
AL
,
Brewer
BM
,
Yang
L
, et al
.
Cancer-associated fibroblasts promote directional cancer cell migration by aligning fibronectin
.
J Cell Biol
2017
;
216
:
3799
816
.
37.
Guido
C
,
Whitaker-Menezes
D
,
Capparelli
C
,
Balliet
R
,
Lin
Z
,
Pestell
RG
, et al
.
Metabolic reprogramming of cancer-associated fibroblasts by TGF-beta drives tumor growth: connecting TGF-beta signaling with "Warburg-like" cancer metabolism and L-lactate production
.
Cell Cycle
2012
;
11
:
3019
35
.
38.
Chen
N
,
Fang
W
,
Zhan
J
,
Hong
S
,
Tang
Y
,
Kang
S
, et al
.
Upregulation of PD-L1 by EGFR activation mediates the immune escape in EGFR-Driven NSCLC: Implication for optional immune targeted therapy for NSCLC patients with EGFR mutation
.
J Thorac Oncol
2015
;
10
:
910
23
.
39.
Yang
X
,
Zhang
X
,
Fu
ML
,
Weichselbaum
RR
,
Gajewski
TF
,
Guo
Y
, et al
.
Targeting the tumor microenvironment with interferon-beta bridges innate and adaptive immune responses
.
Cancer Cell
2014
;
25
:
37
48
.
40.
Cunningham
D
,
Humblet
Y
,
Siena
S
,
Khayat
D
,
Bleiberg
H
,
Santoro
A
, et al
.
Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer
.
N Engl J Med
2004
;
351
:
337
45
.
41.
Hrustanovic
G
,
Lee
BJ
,
Bivona
TG
.
Mechanisms of resistance to EGFR targeted therapies
.
Cancer Biol Ther
2013
;
14
:
304
14
.
42.
Colak
S
,
Ten Dijke
P
.
Targeting TGF-beta signaling in cancer
.
Trends Cancer
2017
;
3
:
56
71
.
43.
Zhou
C
,
Li
J
,
Lin
L
,
Shu
R
,
Dong
B
,
Cao
D
, et al
.
A targeted transforming growth factor-beta (TGF-beta) blocker, TTB, inhibits tumor growth and metastasis
.
Oncotarget
2018
;
9
:
23102
13
.
44.
Batlle
E
,
Massague
J
.
Transforming growth factor-beta signaling in immunity and cancer
.
Immunity
2019
;
50
:
924
40
.
45.
Miyazono
K
.
Transforming growth factor-beta signaling in epithelial-mesenchymal transition and progression of cancer
.
Proc Jpn Acad Ser B Phys Biol Sci
2009
;
85
:
314
23
.
46.
Schreiber
AB
,
Winkler
ME
,
Derynck
R
.
Transforming growth factor-alpha: a more potent angiogenic mediator than epidermal growth factor
.
Science
1986
;
232
:
1250
3
.
47.
Saryeddine
L
,
Zibara
K
,
Kassem
N
,
Badran
B
,
El-Zein
N
.
EGF-induced VEGF exerts a PI3K-dependent positive feedback on ERK and AKT through VEGFR2 in hematological in vitro models
.
PLoS One
2016
;
11
:
e0165876
.
48.
Pore
N
,
Jiang
Z
,
Gupta
A
,
Cerniglia
G
,
Kao
GD
,
Maity
A
.
EGFR tyrosine kinase inhibitors decrease VEGF expression by both hypoxia-inducible factor (HIF)-1-independent and HIF-1-dependent mechanisms
.
Cancer Res
2006
;
66
:
3197
204
.
49.
Derangere
V
,
Fumet
JD
,
Boidot
R
,
Bengrine
L
,
Limagne
E
,
Chevriaux
A
, et al
.
Does bevacizumab impact anti-EGFR therapy efficacy in metastatic colorectal cancer?
Oncotarget
2016
;
7
:
9309
21
.
50.
Maeda
H
,
Shiraishi
A
.
TGF-beta contributes to the shift toward Th2-type responses through direct and IL-10-mediated pathways in tumor-bearing mice
.
J Immunol
1996
;
156
:
73
8
.
51.
Gorelik
L
,
Constant
S
,
Flavell
RA
.
Mechanism of transforming growth factor beta-induced inhibition of T helper type 1 differentiation
.
J Exp Med
2002
;
195
:
1499
505
.
52.
Thomas
DA
,
Massague
J
.
TGF-beta directly targets cytotoxic T cell functions during tumor evasion of immune surveillance
.
Cancer Cell
2005
;
8
:
369
80
.
53.
Crane
CA
,
Han
SJ
,
Barry
JJ
,
Ahn
BJ
,
Lanier
LL
,
Parsa
AT
.
TGF-beta downregulates the activating receptor NKG2D on NK cells and CD8+ T cells in glioma patients
.
Neuro Oncol
2010
;
12
:
7
13
.
54.
Lee
JC
,
Lee
KM
,
Kim
DW
,
Heo
DS
.
Elevated TGF-beta1 secretion and down-modulation of NKG2D underlies impaired NK cytotoxicity in cancer patients
.
J Immunol
2004
;
172
:
7335
40
.
55.
Trotta
R
,
Dal Col
J
,
Yu
J
,
Ciarlariello
D
,
Thomas
B
,
Zhang
X
, et al
.
TGF-beta utilizes SMAD3 to inhibit CD16-mediated IFN-gamma production and antibody-dependent cellular cytotoxicity in human NK cells
.
J Immunol
2008
;
181
:
3784
92
.
56.
Fu
S
,
Zhang
N
,
Yopp
AC
,
Chen
D
,
Mao
M
,
Chen
D
, et al
.
TGF-beta induces Foxp3 + T-regulatory cells from CD4 + CD25 - precursors
.
Am J Transplant
2004
;
4
:
1614
27
.
57.
Concha-Benavente
F
,
Srivastava
RM
,
Trivedi
S
,
Lei
Y
,
Chandran
U
,
Seethala
RR
, et al
.
Identification of the cell-intrinsic and -extrinsic pathways downstream of EGFR and IFNgamma that induce PD-L1 expression in head and neck cancer
.
Cancer Res
2016
;
76
:
1031
43
.
58.
Wang
S
,
Zhang
Y
,
Wang
Y
,
Ye
P
,
Li
J
,
Li
H
, et al
.
Amphiregulin confers regulatory T cell suppressive function and tumor invasion via the EGFR/GSK-3beta/Foxp3 Axis
.
J Biol Chem
2016
;
291
:
21085
95
.
59.
Boland
PM
,
Hutson
A
,
Maguire
O
,
Minderman
H
,
Fountzilas
C
,
Iyer
RV
.
A phase Ib/II study of cetuximab and pembrolizumab in RAS-wt mCRC
.
J Clin Oncol
2018
;
36
:
834
.
60.
Sacco
AG
,
Chen
R
,
Worden
FP
,
Wong
DJL
,
Adkins
D
,
Swiecicki
P
, et al
.
Pembrolizumab plus cetuximab in patients with recurrent or metastatic head and neck squamous cell carcinoma: an open-label, multi-arm, non-randomised, multicentre, phase 2 trial
.
Lancet Oncol
2021
;
22
:
883
92
.
This open access article is distributed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license.

Supplementary data