Increased TGF-a as a Mechanism of Acquired Resistance to ...Authors' Afﬁliations: 1Oncologia...

Cancer Therapy: Preclinical

Increased TGF-a as a Mechanism of Acquired Resistance tothe Anti-EGFR Inhibitor Cetuximab through EGFR–METInteraction and Activation of MET Signaling in Colon CancerCells

Teresa Troiani1, Erika Martinelli1, Stefania Napolitano1, Donata Vitagliano1, Loreta Pia Ciuffreda3,Sara Costantino3, Floriana Morgillo1, Anna Capasso1, Vincenzo Sforza1, Anna Nappi1, Raffaele De Palma2,Elena D'Aiuto2, Liberato Berrino3, Roberto Bianco4, and Fortunato Ciardiello1

AbstractPurpose: Although cetuximab, an anti-EGF receptor (EGFR) monoclonal antibody, is an effective

treatment for patients with KRAS wild-type metastatic colorectal cancer (mCRC), its clinical use is limited

by onset of resistance.

Experimental Design: We characterized two colorectal cancer models to study the mechanisms of

acquired resistance to cetuximab.

Results: Following chronic treatment of nude mice bearing cetuximab-sensitive human GEO colon

xenografts, cetuximab-resistantGEO(GEO-CR)cellswereobtained. InGEO-CRcells,proliferationandsurvival

signals were constitutively active despite EGFR inhibition by cetuximab treatment. Whole gene expression

profiling identifiedaseriesofgenes involved in thehepatocytegrowthfactor (HGF)-MET–dependentpathways,

whichwereupregulatedinGEO-CRcells.Furthermore,activated,phosphorylatedMETwasdetectedinGEO-CR

cells. A second colorectal cancer cell line with acquired resistance to cetuximab was obtained (SW48-CR).

InhibitionofMET expression by siRNA restored cetuximab sensitivity inGEO-CR and SW48-CR cells, whereas

exogenousactivationofMETbyHGFstimulationincetuximab-sensitiveGEOandSW48cells inducedresistance

to cetuximab. Treatment ofGEO-CRand SW48-CR cellswith PHA665752, a selectiveMET inhibitor, inhibited

cell growth, proliferation, and survival signals and impaired cancer cell migration. Overexpression of TGF-a, aspecificEGFR ligand,was involved in theacquisitionof cetuximab resistance inGEO-CRandSW48-CRcells. In

fact, TGF-a overexpression induced the EGFR–MET interaction, with subsequent MET phosphorylation andactivation of MET downstream effectors in GEO-CR and SW48-CR cells.

Conclusions: These results suggest that overexpression of TGF-a through induction of EGFR–METinteraction contributes to cetuximab resistance in colorectal cancer cells. The combined inhibition of EGFR

and MET receptor could represent a strategy for preventing and/or overcoming cetuximab resistance in

patients with colorectal cancer. Clin Cancer Res; 19(24); 6751–65. �2013 AACR.

IntroductionColorectal cancer is one of the most frequently diagnosed

malignant diseases in Europe and one of the leading causes of

cancer-related death worldwide (1). Recent therapeutic strate-gies formetastatic colorectal cancer (mCRC)havebeen focusedon developing molecularly targeted therapies. The EGFR isexpressed in 60% to 80% colorectal cancer and plays a keyrole in the development and progression of human cancersand for this reason it has been proposed as a target for antican-cer therapies (2, 3). Cetuximab, an anti-EGF receptor (EGFR)blockingmonoclonal antibody (mAb), is aneffective treatmentas single agent and in combination with standard chemother-apy regimens for patients with KRAS wild-type mCRC (4).

Cetuximab treatment is effective in a subgroup of patientswithmCRCbecause resistance to anti-EGFR therapies couldbe due to the constitutive activation of signaling pathwaysacting downstream of the EGFR. Several retrospective clin-ical studies have demonstrated that patients with mCRCwith tumors harboring a KRAS mutation have no clinicalbenefit following cetuximab treatment, whereas, in patients

Authors' Affiliations: 1Oncologia Medica and 2Immunologia Clinica,Dipartimento Medico-Chirurgico di Internistica Clinica e Sperimentale F.Magrassi eA. Lanzara; 3Sezionedi Farmacologia, Dipartimento diMedicinaSperimentale, Seconda Universit�a degli Studi di Napoli; and 4OncologiaMedica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica,Universit�a di Napoli Federico II, Naples, Italy

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Teresa Troiani, Dipartimento Medico-Chirurgico diInternistica Clinica e Sperimentale "F. Magrassi e A. Lanzara," Seconda Uni-versit�a degli Studi di Napoli, Italy; Via S. Pansini 5, 80131 Naples, Italy. Phone:39-08-1566-6725; Fax: 39-08-1566-6732; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-13-0423

�2013 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 6751

on July 7, 2021. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 11, 2013; DOI: 10.1158/1078-0432.CCR-13-0423

http://clincancerres.aacrjournals.org/

with wild-type KRAS tumors, a significant clinical benefit interms of progression-free survival and overall survival hasbeen observed (5, 6). The evidence that KRAS mutationswere associated with the lack of response to cetuximab orpanitumumab, another anti-EGFR blocking mAb, inpatients with mCRC has led the U.S. Food and DrugAdministration and the European Medicines Agency torestrict the use of anti-EGFR mAbs to patients with KRASwild-type tumors. In addition to KRAS gene mutations,several retrospective studies in patients with chemorefrac-tory mCRC have provided evidence that primary resistanceto cetuximab could be correlated with mutations in otherintracellular downstream effectors of EGFR activation,such as BRAF, NRAS, and PIK3CA exon 20 genes (7, 8).However, even among the molecularly enriched subset ofpatients with colorectal cancer with KRAS, BRAF, NRAS,and PIK3CA exon 20 wild-type genes, cetuximab is notuniformly clinically effective, suggesting that there areother undefined mechanisms of cetuximab resistance(7–10). Therefore, both intrinsic and acquired resistancemechanisms significantly limit the efficacy of anti-EGFRmAbs in the medical management of patients with mCRC.In this scenario, HER2 gene amplification has recentlybeen identified as a potential mechanism of resistance tocetuximab in a subset of patients with mCRC with wild-type KRAS/NRAS/BRAF/PIK3CA genes (11–13). A proof-of-concept, a multiarm preclinical study in mice bearinghuman coloreactal cancer tumors with HER2 amplifica-tion ("xenopatients"), revealed that the combined inhibi-tion of HER2 and EGFR induced long-lasting tumor regres-sion (12). Other mechanisms of resistance have recentlybeen reported, such as MET and KRAS amplification (14,15). Therefore, an understanding of these mechanismsis necessary to design effective therapies for patients to

prevent or to overcome clinical resistance to cetuximabtreatment.

MET is a cell membrane tyrosine kinase receptor for thehepatocyte growth factor (HGF). Deregulation and conse-quent aberrant MET signaling may occur by differentmechanisms and it has been reported that MET is over-expressed in a variety of cancers including colorectal cancer(16, 17). High expression of MET in colorectal is associatedwith development of distantmetastases (18). Increased andderegulated MET signaling has been identified as one of thekey pathways to bypass growth inhibition caused by drugstargeting the EGFR in lung adenocarcinoma (19, 20). Sev-eral clinical studies have evaluated the combination ofMETand EGFR inhibitors in patients with non–small cell lungcancer (NSCLC; refs. 21–23).

Furthermore, a functional link between EGFR and METhas been suggested (24). In this respect, expression of EGFRand MET correlate in multiple malignancies such as chor-doma, prostate and ovarian carcinomas, and gastrinoma(25). EGFR has been implicated in HGF-induced hepato-cyte proliferation and is required for MET-mediated coloncancer cell invasiveness (26). Despite these studies, theunderlying mechanisms of EGFR-induced MET phosphor-ylation are not yet well understood.

In the present study, we have generated and characterizedtwo human colon cancer cell models of acquired resistanceto cetuximab to elucidate the molecular mechanisms ofresistance. We have found that in cetuximab-resistant colo-rectal cancer cells, cell proliferation, and survival pathwaysare activated by MET. Interestingly, enhanced expression ofthe selective EGFR ligand TGF-a in cetuximab-resistantcolorectal cancer cells is responsible for EGFR–MET inter-action and subsequent EGFR-induced MET phosphoryla-tion and activation. Finally, treatment of these cells with aselective MET inhibitor restores cetuximab sensitivity, sug-gesting that the combined inhibition of both EGFR andMET receptor tyrosine kinases (RTK) could represent arational therapeutic strategy for preventing and/or over-coming cetuximab resistance in patients with mCRC.

Materials and MethodsDrugs

Cetuximab, an anti-EGFR human–mouse chimeric mAbwas kindly provided by Merck Serono. PHA665752, aselective MET tyrosine kinase inhibitor (TKI), was pur-chased from Santa Cruz Biotechnology. PHA665752 wasdissolved in sterile dimethyl sulfoxide and a 10 mmol/Lworking solution was prepared and stored in aliquots at�20�C. Working concentrations were diluted in culturemedium just before each experiment.

Generation of cetuximab-resistant GEO and SW48 cellsFour- to six-week-old female balb/c athymic (nuþ/nuþ)

mice were purchased from Charles River Laboratories. Theresearch protocol was approved and mice were maintainedin accordance with the institutional guidelines of the Sec-ond University of Naples Animal Care and Use Committee

Translational RelevanceCetuximab, an anti-EGFR monoclonal antibody

(mAb), has proven to be effective in combination withchemotherapy or as a single agent for treatment ofpatients with KRAS wild-type metastatic colorectal can-cer (mCRC). Clinical data indicate that even the bestresponses are transient and eventually all patients devel-op acquired resistance. This evidence triggered a series ofstudies on the molecular mechanisms of primary andacquired resistance to cetuximab. In the present study,we have demonstrated that resistance to cetuximab incolorectal cancer cells is mediated by TGF-a overexpres-sion, which induced the EGFR–MET interaction withsubsequent MET pathway activation. Blockade of bothEGFR andMET receptor tyrosine kinases could representa strategy for preventing and/or overcoming cetuximabresistance in patients with colorectal cancer. Theseresults could be of relevant clinical interest for the designof translational research-based clinical studies withcetuximab in combination with MET inhibitors.

Troiani et al.

Clin Cancer Res; 19(24) December 15, 2013 Clinical Cancer Research6752




(Naples, Italy). Mice were acclimatized at the Second Uni-versity of Naples Medical School Animal Facility for 1 weekbefore being injected with cancer cells. Mice were injectedsubcutaneously with 2.5 � 106 GEO cells that had beenresuspended in 200 mL of Matrigel (BD Biosciences). Whenestablished tumors of approximately 200 to 300 mm3 indiameter were detected, mice were treated continuously byintraperitoneal (i.p.) injection with cetuximab (1 mg twiceweekly) for the indicated time periods. Each treatmentgroup consisted of 8 mice. Tumor size was evaluated twiceper week by calliper measurements using the followingformula: p/6 � larger diameter � (smaller diameter)2. Theanimals were sacrificed when the tumor diameter exceeded1,500 mm3. Tumor from the treated group, were removed,digested, and suspended as a single cell, which were prop-agated in in vitro culture. For a period of 6 months SW48cells were continuously exposed to increasing concentra-tions of cetuximab. The starting dose was the dose causingthe inhibition of 50%of cancer cell growth (IC50). The drugdose was progressively increased to 1 mg/mL in approxi-mately 2 months, to 5 mg/mL after other 2 months, andfinally, to 10 mg/mL after additional 2 months. The estab-lished cetuximab-resistant SW48 cancer cell line (SW48-CR) was then maintained in continuous culture with thismaximally achieved dose of cetuximab that allowed cellularproliferation.

Cell linesThe human GEO colon cancer cell was a gift of Dr. N.

Normanno (National Cancer Institute, Naples, Italy). Thehuman SW48 colon cancer cell line was obtained from theAmerican Type Culture Collection. Four GEO cetuximab-resistant clones were established as in vitro cell lines aftercancer cell recovery and enzymatic treatment (27) from invivoGEO tumor xenografts in mice, as previously described(27). GEO andGEO cetuximab-resistant clones were growninDulbecco’s Modified EagleMedium (Lonza) supplemen-ted with 20% FBS (Lonza), 1% penicillin/streptomycin(Lonza) and maintained in a humidified atmosphere of95% air and 5% carbon dioxide (CO2) at 37

�C. SW48 andSW48-CR cells were grown in RPMI-1640 (Lonza) supple-mented with 10% FBS, 1% penicillin/streptomycin. All celllines were routinely screened for the presence of mycoplas-ma (Mycoplasma Detection Kit, Roche Diagnostics).

Proliferation assayCancer cells were seeded in 24-well plates and were

treatedwith different concentrations of PHA665752 (range,0.01–10nmol/L), cetuximab (range, 0.01–20mg/mL) aloneor in combination for 96 hours. Cell proliferation wasmeasured with MTT. The IC50 was determined by interpo-lation from the dose-response curves. Results represent themedian of three separate experiments, each performed inquadruplicate.

Apoptosis assayGEO, GEO-CR, and SW48-CR cells were seeded in 6-well

plates, treated for 72 hours, and stained with Annexin

V–fluorescein isothiocynate (FITC). Apoptotic cell deathwas assessed by counting the numbers of cells that stainedpositive for Annexin V–FITC and negative for propidiumiodide using theApoptosis AnnexinV-FITCKit (Invitrogen),coupled with fluorescence-activated cell sorting analysis.

Western blotting and immunoprecipitationGEO, SW48, GEO-CR, and SW48-CR cells were seeded

into 100 mm3 dishes and treated for 24 hours. Proteinlysates containing comparable amounts of proteins, esti-mated by a modified Bradford assay (Bio-Rad), were sub-jected to immunoprecipitation or direct Western blot anal-ysis. Immunocomplexes were detected with the enhancedChemiluminescence Kit (Pierce Biotechnology Inc.). Anti-EGFR, phospho-EGFR (Tyr1068), HER2, HER3, HER4,p44/42MAPK, phospho-p44/42MAPK, AKT, phospho-AKT(Ser473), MET, phospho-MET (Tyr 1234/1235), STAT3,phospho-STAT3, FAK, and phospho-FAK were purchasedfrom Cell Signaling Technology. Monoclonal anti-a-tubu-lin antibody was purchased from Sigma Chemical Co.Secondary antibodies coupled to horseradish peroxidasewere from GE Healthcare. Each experiment was done intriplicate. Two milligrams of protein lysates were immuno-precipitated with the required antibodies; immunocom-plexes were recovered with protein G Sepharose (RocheDiagnostics) and detected by Western blotting.

Microarray gene expression analysisAgilent (Agilent Technologies) microarray analyses were

performed to assess baseline gene expression profile forGEO and GEO-CR cells using a one color labeling micro-array system as previously described (28). The absoluteamount and purity (A260/280 nm ratio) of total RNAsamples were determined by spectrophotometry (Nano-drop, Thermofisher) and the size distribution was assessedby Agilent Bioanalyzer. Eight hundred nanograms of totalRNA were converted into labeled cRNA with nucleotidescoupled to a fluorescent dye (either Cy3 or Cy5) followingthemanufacturer’s protocol (Quick AmpKit, Agilent). Yieldand purity (A260/280 nm ratio) of cRNAs were determinedby spectrophotometry (Nanodrop, Thermofisher). Of note,825 ng of cRNA labeled from GEO colon cancer cell lineswere hybridized to AgilentHumanWholeGenome4� 44 kMicroarrays. Data were extracted from slide image usingAgilent Feature Extraction software (v.10.5). The raw dataand associated sample information were loaded and pro-cessed by Gene Spring 11.5X (Agilent Technologies). Foridentification of genes significantly altered in resistant cells,total detected entities were filtered by signal intensity value(upper cutoff 100th and lower cutoff 20th percentile) andflag to remove very low signal entities. Experiments werecarried out in triplicate and data were analyzed usingStudent t test (P < 0.05) with a Benjamani–Hochbergmultiple test correction to minimize selection of false posi-tives. Of the significantly differentially expressed RNA, onlythose with greater than 2-fold increase or 2-fold decrease ascompared with the controls were used for further analysis.Functional and network analyses of statistically significant

Resistance Mechanisms to Anti-EGFR Inhibitors

www.aacrjournals.org Clin Cancer Res; 19(24) December 15, 2013 6753




gene expression changes were performed using IngenuityPathways Analysis (IPA) 8.0 (Ingenuity Systems; http://www.ingenuity.com). Analysis considered all genes fromthe data set that met the 2-fold (P < 0.05) change cutoff andthat were associated with biologic functions in the Ingenu-ity Pathways Knowledge Base. The significance of the asso-ciation between the data set and the canonical pathway wasmeasured in twoways: (i) ratio of the number of genes fromthe data set that map to the pathway divided by the totalnumber of genes that map to the canonical pathway isdisplayed; (ii) Fisher exact test was used to calculate a Pvalue determining the probability that the associationbetween the genes in the data set and the canonical pathwayis explained by chance alone.

RNA interference and PCR analysisThe small inhibitor duplex RNAs (siRNA) (ON-target

plus SMARTpool) siMET (human: #L-003156-00), siTGF-a (human: #L019737), and siHBEGF (human: #L019624)were from Dharmacon. The siCONTROL Non-targetingPool (#D-001206-13-05) was used as a negative (scram-bled) control. Cells were transfected with 100 nmol/LsiRNAs using Dharmafect reagent following the manufac-turer’s instructions. The day before transfection, the cellswere plated in 35 mm dishes at 40% of confluence inmedium supplemented with 5% FBS without antibiotics.Cells were harvested 48 hours after transfection. Westernblot analysis for MET, and PCR for TGF-a and HBEGFexpression was done. RNA extraction was performed by theRNeasy Kit (Qiagen) following the manufacturer’s instruc-tions. The RNA was quantified by Nanodrop (ThermoScientific) and RNA integrity was analyzed by the 2100Bioanalyzer (Agilent Technologies). For semiquantitativePCR reactions, random-primed first-strand cDNA was syn-thesized in a 50 mL reaction volume starting from 2 mg RNAby using the GeneAmp RNA PCR Core Kit (Applied Bio-systems). PCR amplification was performed using the Gen-eAmp RNA PCR Core Kit system following the manufac-turer’s instructions. To exclude DNA contamination, eachPCR reaction was also performed on untranscribed RNA.We used the amplimers for TGF-a, MET, and HBEGF andlevels of b-actin transcripts were used for normalization.Sequence of amplimers used is available on request. ThesiRNA effects on cell signaling were evaluated by Westernblot analysis. siRNA effects on cell proliferation and apo-ptosis were evaluated by MTT and apoptotic assay as pre-viously described. Briefly, cells were seeded into 24-multi-well cluster dishes and transfected with MET, TGF-a, andHBEGF siRNA. Of note, 96 hours after transfection, cellstreated withMET, TGF-a, andHBEGF siRNA received 5 mg/mL of cetuximab and cell proliferation and apoptosis weredetermined 24 hours later.

Evaluation of growth factor secretionThe concentrations of HGF, TGF-a, HBEGF in the con-

ditioned medium were analyzed by using Luminex tech-nology (29) analysis with multiplex beads suspensionarray plates according to the manufacturer’s instructions

(Invitrogen). Assays were done in triplicate. Results werenormalized for the number of producing cells and reportedas pg of ligands per 106 cells per 72 hours.

Migration assayCell migration was assessed using a commercially avail-

able chemotaxis assay. Briefly, cells were incubated inserum-free medium for 24 hours and were left untreatedor treated with the indicated doses of cetuximab,PHA665752, or their combination following which theywere detached from flasks, suspended in quenching medi-um (serum-free medium containing 5% bovine serumalbumin) and EDTA, and seeded into Boyden migrationchamber inserts placed in a 24-well plate. The insertscontain a microporous membrane with an 8 mm pore size.Inserts were placed over wells containing serum-free mediaplus chemo-attractant (10%FBS). After a 48-hour treatmentperiod, cells/media were discarded from the top side of themigration chamber insert and the chamber was placed inthe wells of a new 24-well plate containing cell detachmentsolution. Following incubation for 30 minutes at 37�C, theinsert was discarded, and a solution of lysis buffer andCyQuant GR dye was added to each well. CyQuant is agreen fluorescent dye that exhibits strong enhancement offluorescence when bound to cellular nucleic acids releasedby the lysis buffer, enabling assessment of the relativenumber of migrated cells. Fluorescence was determinedwith a fluorimeter at 480/520 nm. Assays were performedin triplicate.

Gene mutation analysisGEO, SW48, GEO-CR, and SW48-CR cells were seeded

into 100 mm3 dishes for 72 hours. DNA extraction wasperformed by the QIAamp DNA Mini Kit (Qiagen) fol-lowing the manufacturer’s instructions. The DNA wasquantified by Nanodrop (Thermo Scientific). Ten nano-grams of GEO, SW48, GEO-CR, and SW48-CR genomicDNA was analyzed by next generation sequencing usingthe Ion AmpliSeq colon and lung panel as previouslydescribed (30).

Statistical analysisThe statistical analyses of in vitro and in vivo data were

carried out using Prism version 4.02 (GraphPad Software,Inc). The Student t test was used to evaluate the statisticalsignificance of the results. All P values represent two-sidedtests of statistical significance with P value

therapeutic dose of the drug (Fig. 1A). Continuous treat-ment caused tumor growth suppression within the firstweeks of treatment.However, after approximately 12weeks,in most mice GEO tumors resumed growth despite cetux-imab treatment, reaching a growth rate comparable tountreated control GEO tumors within 18 to 20 weeks (Fig.1A). Four cetuximab-resistant tumors were surgicallyremoved and homogenized into single-cell suspensionsused to generate in vitro GEO-CR cell lines, which dis-played resistance to the growth inhibitory effects of cetux-imab treatment (Fig. 1B). Only one of these cell lines

(named GEO-CR cells) was further characterized and usedfor the subsequent experiments. As illustrated in Fig. 1Band C, GEO-CR cells were not sensitive cetuximab treat-ment at doses up to 20 mg/mL, as compared with parentalGEO cells.

The investigation of EGFR-dependent intracellular path-ways revealed that treatment of GEO cells with cetuximabcompletely blocked EGFR phosphorylation and conse-quently the activation of downstream mitogen-activatedprotein kinase (MAPK) and AKT signals (Fig. 1D).Although cetuximab treatment markedly reduced EGFR

Figure 1. Establishment andcharacterization of cetuximab-resistant colorectal cancer cells.A, mice bearing GEO cells weretreated continuously byintraperitoneal injection withvehicle or cetuximab (1 mg dailytwice weekly; n ¼ 8/group).Treatments started when tumorreached volumes of 200 to 300mm3. Animals were sacrificedwhen tumor reached 1,500 mm3 insize. Tumors from the cetuximab-treated group were removed,digested, and suspended as asingle cell, which was propagatedin in vitro culture. B, parental andcetuximab-resistant GEO cloneswere treated with increasingconcentrations of cetuximab(0.01–20 mg/mL) for 96 hours andevaluated for proliferation by MTTstaining, as described in Materialsand Methods. The results are theaverage�SDof three independentexperiments each done inquadruplicate. C, apoptosis wasevaluated with Annexin V staining,as described in Materials andMethods. GEO and GEO-CR weretreated with cetuximab for 72hours. The rate of apoptosis wasexpressed as a percentage of thetotal cells counted. Columns:means of three independentexperiments. D, GEO and GEO-CRwere treated with cetuximab (5mg/mL) for 24 hours. Cell extractswere assayed by Western blottingto detect the indicated proteins.Experiments were repeated threetimes with similar results.






phosphorylation in GEO-CR cells, the downstream MAPKand AKT pathways were not completely blocked (Fig. 1D).

Identificationof gene expressionprofiles that correlatewith acquired resistance to cetuximab in GEO-CR cells

To investigate the potential molecular pathways involvedin cetuximab resistance, mRNAs from GEO and GEO-CRcells were extracted and assessed for global gene expressionchanges bymicroarray analysis. In GEO-CR cells, 39 and 54genes were identified whose expression was upregulated or

downregulated, respectively, as compared with GEO cells(P < 0.05). Among the 39 upregulated genes in GEO-CRcells, we identified 17 genes involved in the HGF–MET-dependent pathways (Supplementary Table S1 and Fig. 2A)including MET (31–34). Moreover, the mRNAs of twoEGFR selective ligands were upregulated in GEO-CR cellsas compared were GEO cells: TGF-a and heparin-bindingEGF–like growth factor (HB-EGF). Among the 54 down-regulated genes in GEO-CR cells, 7 genes were involvedin DNA repair and 4 genes were involved in cell-cycle

Figure 2. Schematic representation of genes whose expression is upregulated in GEO-CR cells as compared with GEO cells. A, a total of 1 � 106 GEO andGEO-CR cancer cells were seeded in 100-mm3 dishes for 96 hours before RNA extraction. The procedure was performed as described in MaterialsandMethods. Agilent microarray analyses were performed to assess baseline gene expression profile for GEO andGEO-CR cancer cell lines, as described inMaterials and Methods. In yellow are highlighted genes with upregulated expression in GEO-CR as compared with GEO, which are involved in MET-dependent signaling. B, thirtymicrograms of cell protein extractswere fractionated through 7%SDS–PAGE, transferred to nitrocellulose filters, and incubatedwith the appropriate antibodies, as described in Materials and Methods. Immunoreactive proteins were visualized by enhanced chemiluminescence.MET mRNA levels were measured by PCR, as described in Materials and Methods.

Troiani et al.





regulation (Supplementary Table S2A and S2B). ByWesternblot analysis, we found a moderate increase in total METprotein expression, which was not accompanied by a sig-nificant increase in MET mRNA in GEO-CR cells as com-pared with parental GEO cells. However, higher levels ofactivated phospho-MET protein were depicted (Fig. 2B).

Inhibition of MET expression by siRNA restoressensitivity to cetuximabTo determine whether MET activation could be involved

in the acquisition of cetuximab resistance in GEO-CR cells,we investigated whether reduction ofMET expression couldrestore cetuximab sensitivity. Transfection with a specificMET siRNA for 48 hours significantly reduced MET proteinexpression in GEO-CR cells, as shown in Fig. 3A. As illus-trated in Fig. 3B and C, MET siRNA treatment slightlyreduced cell growth with a little increase in apoptosis inGEO-CR cells. Although single-agent cetuximab treatmentdid not affect GEO-CR proliferation, cetuximab treatmentin combination withMET silencing determined a statisticalsignificant cell growth inhibition and proapoptotic effectsin GEO-CR cells. MET silencing also restored the ability ofcetuximab to inhibit MAPK and AKT activation in GEO-CRcells, as shown by downregulation of both phospho-MAPKand phospho-AKT levels (about 50% of inhibition, res-pectively; Fig. 3D).To further evaluate whether MET could confer resistance

in adifferentmodel of cetuximab-sensitive colorectal cancercells,weused the SW48colorectal cancer cell line, harboringKRAS,NRAS, BRAF, and PIK3CAwild-type genes, known tobe cetuximab sensitive. Following continuous exposure toincreasing doses of cetuximab for up to 6 months of SW48cells, SW48-CR cells were obtained. As depicted in Supple-mentary Fig. S1A, SW48-CR cells were not sensitive to thegrowth inhibitory effect of cetuximab treatment at doses upto 20 mg/mL. The EGFR-dependent intracellular signalingpathways were not inhibited by cetuximab treatment inSW48-CR cells as compared with parental SW48 cells (Sup-plementary Fig. S1B). It has recently suggested that a mech-anismof acquired resistance to cetuximab continuous treat-ment in human colorectal cancer cell lines that are initiallysensitive, is the appearance of cancer clones with a mutatedRAS gene (9, 10). To test this hypothesis, we have evaluatedthe presence of mutations in both the KRAS and the NRASgenes in GEO-CR and SW48-CR cells as compared withparental GEO and SW48 cells by next generation sequenc-ingusing the IonAmpliSeq colonand lung cancer panel thatevaluates the presence of mutations in 22 genes includingexon 2, 3, and 4 KRAS and exon 2 and 3 NRAS (30). Noadditional mutations were observed in KRAS and NRASgenes in GEO-CR and SW48-CR cells (data not shown).Moreover, increased MET protein phosphorylation wasobserved in SW48-CR but not in parental SW48 cells,(Supplementary Fig. S1B). Similarly to GEO-CR cells, theinhibition of MET expression by siRNA treatment in com-bination with cetuximab determined a statistical significantcell growth inhibition and proapoptotic effects in SW48-CR(Supplementary Fig. S2A–S2C). Moreover, MET protein

downregulationby siRNA treatment also restored the abilityof cetuximab to inhibit MAPK and AKT activation in SW48-CR cells (Supplementary Fig. S2D).

Activation of MET by HGF treatment can rescuecetuximab-sensitive GEO and SW48 cells fromcetuximab-induced cell growth inhibition

We next evaluated whether MET activation could conferresistance in two cetuximab-sensitive colorectal cancer celllines (GEO and SW48). These colorectal cancer cell lineswere extremely sensitive to cetuximab cell growth inhibi-tion with IC50 of approximately 0.1 and 0.2 mg/mL, respec-tively (Fig. 3E). Treatment with the MET ligand HGF,significantly reduced cetuximab-induced cell growth inhi-bition in both cancer cell lines (Fig. 3F). In addition, despitetreatment with cetuximab, HGF treatment caused METphosphorylation and partial reactivation of MAPK and AKTin both GEO (Fig. 3G) and SW48 cells (Fig. 3H) comparedwith cetuximab treatment alone.

Treatment with a selective MET TKI sensitizes GEO-CRand SW48-CR cells to cetuximab

To further evaluate the role of MET activation in cetux-imab resistance, treatment of GEO-CR and SW48-CR cellswith PHA665752, a selective MET TKI, was performed.PHA665752 treatment of GEO-CR and SW48-CR cellsinduced a dose-dependent inhibition of cell growthwith anIC50 of approximately 0.5 and 5 nmol/L, respectively (Fig.4A and C). Moreover, altough cetuximab treatment had noeffect on cell growth in GEO-CR and SW48-CR cells, thecombined treatment with PHA665752 restored the sensi-tivity of GEO-CR and SW48-CR cells to cetuximab in a dose-dependent manner (Fig. 4B and D). In particular, in GEO-CR cells the IC50s for treatment with cetuximab in combi-nation with PHA665752 (0.01 or 0.05 nmol/L) were 0.1and 0.01 mg/mL, respectively (Fig. 4B). We next assessed byWestern blot analysis the effects of treatment of GEO-CRandSW48-CR cellswith cetuximaband/or the selectiveMETinhibitor on the expression and activation of key proteins,which act downstream to EGFR and MET. As shownin Fig. 4E and F, single-agent cetuximab or single-agentPHA665752 treatment were able to completely inhibitEGFR andMET phosphorylation, respectively, while induc-ing a modest decrease of MAPK and AKT phosphorylation.However, the combined treatmentwith both drugswas ableto fully abrogate phosphorylationof bothMAPKandAKT inGEO-CR and SW48-CR cells. (Fig. 4E and F).

GEO and SW48 cells are differentiated colon carcinomacells with little or no ability to migrate or to invade in vitro(35). GEO-CR and SW48-CR cells acquired the ability tomigrate in an in vitro migration assay (Fig. 4G and H). Toinvestigate whether the migration properties of GEO-CRand SW48-CR cells could be due to MET activation, weperformed a migration assay on GEO-CR and SW48-CRcells in the presence of cetuximab, PHA665752 or theircombination. Although single-agent cetuximab treatmenthad no effect on GEO-CR and SW48-CR cancer cell migra-tion, PHA665752 significantly reduced the number of






Figure 3. Inhibition ofMET expression restores cetuximab sensitivity in GEO-CR cells, andHGF-dependentMET activation rescues GEOand SW48 cells fromcetuximab inhibition. A, GEO-CR cells were transfectedwith either a specific siRNA targetingMET orwith a control (scrambled) RNA sequence and harvested96hours after transfection.Western blot analysis forMETexpressionwasdoneasdescribed inMaterials andMethods.B,GEO-CRcellswere transfectedwitha specific siRNA targetingMET. Twenty-four hours after transfection, cells were treated with cetuximab, 5 mg/mL. Viable cells were counted after 24 hours oftreatment and plotted relative to untreated control. The results are average�SDof three independent experiments each done in duplicate. Cetuximab plus si-MET versus si-MET (��, P < 0.005), cetuximab plus si-MET versus cetuximab (��, P < 0.0005). (Continued on the following page.)

Troiani et al.





cancer cells that were able to invade through the migrationchambers. Moreover, this effect was markedly potentiatedby the combined inhibition of EGFR and MET (Fig. 4Gand H).

TGF-a overexpression mediates acquired resistance tocetuximab by causing active EGFR–MET interaction inGEO-CR and SW48-CR cellsOnce demonstrated the MET role in the acquired cetux-

imab resistance of GEO-CR and SW48-CR cells, we furtherevaluated the potential mechanism(s) responsible for METactivation.Onemajormechanismof cellmembrane growthfactor receptor activation in cancer cells is the specific ligandstimulation. In this respect, HGF-driven autocrine or para-crine activation ofMET has been observed in several cancers(16). However, little or no HGF secretion was detected inGEO, SW48, GEO-CR, and SW48-CR cells, suggesting thatautocrine MET activation was not occurring in these cancercell models (Fig. 5A and E). Moreover, it has been shownthat several members of the EGFR family can transactivateMET by inducing the formation of heterodimers even inabsence ofHGF (36). Therefore, aWestern blot analysis wasperformed to evaluate the expression of the four membersof the EGFR family. Although EGFR was expressed in GEO,SW48, GEO-CR, and SW48-CR cells, no expression ofERBB2, ERBB3, and ERBB4 was found in these cancer celllines (data not shown). According to these results, wehypothesized that MET activation could be caused by inter-action with the EGFR in GEO-CR and SW48-CR cells. Bymicroarray gene expression analysis, the two specific EGFRligands TGF-a and HB-EGF, were upregulated in GEO-CRcells as compared with GEO cells (Fig. 2A and Supplemen-tary Table S1). ELISA assays confirmed that both growthfactors were significantly secreted in the conditioned medi-um derived from GEO-CR cells as compared with parentalGEO cells (Fig. 5A). Furthermore TGF-a levels were alsosignificantly higher in the conditioned medium derivedfrom SW48-CR cells (Fig. 5E). To examine whether TGF-aand/or HB-EGF enhanced expression could be causallyinvolved in cetuximab resistance, GEO-CR cells were trans-fectedwith specific siRNAs forTGF-aorHB-EGF (Fig. 5B andC). TGF-a silencing, but not HB-EGF, was functionallyrelevant because it sensitized GEO-CR cells to the antipro-liferative effects of cetuximab (Fig. 5B). Similarly, TGF-adownregulationby siRNAsensitized SW48-CR cells to cetux-imab (Fig. 5F andG). Furthermore, inhibiting TGF-a expres-

sion by siRNA together with cetuximab treatment of GEO-CR and SW48-CR cells caused a significant suppression ofMET phosphorylation as well as of downstream effectorsMAPK, AKT, STAT3, and FAK (Fig. 5D and H).

Taken together, these results suggest a role for TGF-aoverexpression in acquired cetuximab resistance inGEO-CRand SW48-CR cells. To test this hypothesis, further experi-ments were carried out in GEO and SW48 parental cells toevaluatewhether exogenous TGF-a treatment by itself couldlead to the acquisition of resistance to cetuximab. TGF-astimulation of both cell lines resulted in decreased cetux-imab sensitivity, because TGF-a could partially overcomethe cetuximab induced growth inhibition (Fig. 6A). More-over, TGF-a treatment of GEO and SW48 cells led to bothEGFR and MET phosphorylation with an increase in MAPKactivation and did not show evident effects on AKT (Fig.6B). However, combined treatment with cetuximab partial-ly rescues the cetuximab-mediated effects on MAPK andAKT phosphorylation. (Fig. 6B).

Wenext analyzedwhether a functional cross talk betweenEGFR and MET in GEO-CR cells could be due to theinteraction of the two growth factor receptors. For thispurpose, GEO and GEO-CR protein extracts were immu-noprecipitated with a specific anti-MET antibody andassayed by Western blotting with a specific anti-EGFR anti-body. As shown in Fig. 6C, EGFR immunoprecipitatedtogether with MET in GEO-CR cells, but not in GEO cells.Moreover, to elucidate the potential role of TGF-a in induc-ing EGFR–MET interaction, GEO, SW48, GEO-CR, andSW48-CR cells were treated with TGF-a in the presence orin the absence of cetuximab and lysates were immune-precipitated with the anti-MET antibody and then assayedby Western blotting with the anti-EGFR antibody. Asreported in Fig. 6D, TGF-a treatment induced EGFR–METinteraction in GEO and SW48 cells and increased theinteraction GEO-CR and SW48-CR cells. The EGFR–METinteraction was also observed following combined treat-ment of cells with both TGF-a and cetuximab, suggestingthat, even in the presence of cetuximab, TGF-a could inducea cross-interaction between EGFR and MET.

DiscussionElucidating the mechanisms of cancer cell resistance to

anticancer drugs is critical for the development of effectivetherapies. In this respect, mechanistic insights gained from

(Continued.) C, apoptosiswasevaluatedwithAnnexin Vstaining, asdescribed inMaterials andMethods.GEO-CRcellswere transfectedwith a specific siRNAtargeting MET. Ninety-six hours after transfection, cells received 5 mg/mL of cetuximab and apoptosis was measured after 24 hours of treatment. Therateof apoptosiswasexpressedasapercentageof the total cells counted.Columns:meansof three independent experiments.Cetuximabplus si-METversussi-METor cetuximab (��,P

Figure 4. PHA655752 restoressensitivity of GEO-CR and SW48-CR cells to cetuximab. A and C,GEO, GEO-CR (A), SW48, andSW48-CR (C) cells were treatedwithincreased concentrations ofPHA665752 (0.01–10 nmol/L) for 96hours and evaluated for cellproliferation by MTT staining, asdescribed in Materials andMethods. The results are theaverage � SD of three independentexperiments, each done inquadruplicate. B and D, GEO-CR(B) and SW48-CR (D) cells weretreated with 2 doses of PHA665752(0.01 or 0.05 nmol/L for GEO-CRand 5 or 10 nmol/L for SW48-CR),with increasing concentrations ofcetuximab (0.01–10mg/mL) orwith acombination of both drugs for 96hours and evaluated for cellproliferation by MTT staining, asdescribed in Materials andMethods. The results are theaverage � SD of three independentexperiments, each done inquadruplicate. E and F, GEO-CR (E)and SW48-CR (F) cells were treatedwith PHA665752, cetuximab, ortheir combination at the indicatedconcentrations for 24hours. Thecelllysates were assayed by Westernblotting with the indicatedantibodies, as described inMaterials and Methods. G and H,GEO-CR (G) and SW48-CR (H) cellswere treated with PHA665752,cetuximab, or their combination atthe indicated concentrations andanalysis of in vitro cell migrationwasperformed as described inMaterialsand Methods.

Troiani et al.





Figure 5. Inhibition of TGF-aexpression reverts cetuximabresistance in GEO-CR and SW48-CR cells. A and E, TGF-a, HB-EGF,and HGF protein levels weremeasured in cell culture media byusing Luminex technology, asdescribed in Materials andMethods. TGF-a in GEO versusGEO-CR (�, P < 0.05), HBEGF inGEO versus GEO-CR (�, P < 0.05),TGF-a in SW48 versus SW48-CR(��,P < 0.005). B and F, GEO-CR (B)and SW48-CR (F) cells weretransfected with a specific TGF-a–targeting siRNA, or with aspecific HB-EGF–targeting siRNA(only GEO-CR), or with ascrambled, control siRNA. TGF-aand HBEGF mRNA levels weremeasured by PCR, as described inMaterials and Methods. C and G,GEO-CR (C) andSW48-CR (G) cellswere transfected with a specificTGF-a–targeting siRNA, or with aspecific HB-EGF–targeting siRNA(only GEO-CR), or with ascrambled, control siRNA. Twenty-four hours after transfection, cellswere treated with cetuximab, 5 mg/mL. Viable cells were counted after24 hours of treatment and plottedas a percentage untreated controlcells. The results are the average�SD of three independentexperiments, each done inquadruplicate. D and H, Westernblot analysis of protein expressionin GEO-CR (D) and in SW48-CR (H)cells transfected with a specificTGF-a–targeting siRNA or with ascrambled, control siRNA andsubsequently treated with theindicated dose of cetuximab for 24hours. Total cell protein extractswere subjected to immunoblottingwith the indicated antibodies, asdescribed in Materials andMethods.






preclinical models can be used to design novel treatmentstrategies. In the past few years, an extensive effort has beenmade to understand the resistance mechanism(s) that can-cer cells develop to overcome the anticancer efficacy ofEGFR inhibitors (3, 4, 7).

The aim of this study was to examine the signalingmechanisms operating in human colorectal cancer cells thatare sensitive to the antitumor activity of the anti-EGFR mAbcetuximab and that become resistant following continuousexposure to the drug. In fact, although cetuximab is aneffective anticancer agent in the treatment of patients withmCRC, the occurrence of acquired resistance is a majorclinical limitation. It hasbeen shown that intrinsic resistanceto cetuximab as well as of the other anti-EGFR mAb pani-

tumumab canbe the result of constitutive activationofKRASsignaling, due to the presence of KRAS gene mutations (4).For this reason, cetuximab therapy is limited to patients withmCRC with a tumor harboring wild-type KRAS gene. More-over, the constitutive activation of other proteins by specificgenemutations are responsible of intrinsic resistance to anti-EGFR therapies in patients with colorectal cancer, includingBRAF,NRAS, PI3KCA (exon 20) genes, or inactivation of thePTENphosphatase (7, 37). It has recentlybeen shown that ininitially responding patients with colorectal cancer with aKRAS wild-type tumor, the resistance to anti-EGFR antibo-dies may occur by selection of cancer cell clones harboring aKRAS gene mutation (9, 10). However, approximately 25%of patients with colorectal cancer not responding to EGFR

Figure 6. TGF-a treatment of GEOcells renders these cells resistantto cetuximab, and TGF-a inducesEGFR–MET interaction. A, GEOand SW48 cells were treated withcetuximab (1 mg/mL), TGF-a (100ng/mL), or their combination. Cellproliferation was determined after96 hours. The results are theaverage�SDof three independentexperiments, each done inquadruplicate. B, Western blotanalysis of protein expression inGEO and SW48 cells treated withcetuximab (1 mg/mL), TGF-a (100ng/mL), or their combination. Totalcell protein extracts weresubjected to immunoblotting withthe indicated antibodies, asdescribed in Materials andMethods. C, two milligrams ofGEO-CR cells or of GEOcell protein extracts wasimmunoprecipitated with a specificanti-MET antibody and thenimmunoblottedwith a specific anti-EGFR antibody, as described inMaterials and Methods. D, GEO,SW48, GEO-CR, and SW48-CRcells were treated with cetuximab,1 mg/mL, for 24 hours. Twentyminutes before harvesting thecells, TGF-a (100 ng/mL) wasadded to the cell cultures. Twomilligrams of GEO, SW48, GEO-CR, and SW48-CR cell proteinextracts was immunoprecipitatedwith a specific anti-MET antibodyand then immunoblotted with aspecific anti-EGFR antibody, asdescribed in Materials andMethods.

Troiani et al.





inhibitors arewild-type forKRAS,BRAF,NRAS,PIK3CA, andPTEN genes and the mechanism of resistance in thesepatients is still unknown (38). Therefore, both primary andacquired resistance mechanisms significantly limit the effi-cacy of anti-EGFR mAbs in the medical management ofpatients with colorectal cancer.In the present study, we have generated and characterized

twomodels of cetuximab resistance in human colon cancercells. In the parental, cetuximab-sensitive, GEO, and SW48colorectal cancer cells, cetuximab treatment was able toinduce cell growth inhibition and apoptosis with MAPKand AKT phosphorylation reduction. On the contrary, thiswas not observed in the derived GEO-CR and SW48-CRcells, in which activation of MAPK and AKT, was notblocked, despite EGFR inhibition by cetuximab treatment.MAPK activation that could bypass EGFR inhibition is oneof the potential mechanisms of acquired resistance to anti-EGFR therapies (39). In this respect, it has recently beenshown that one mechanism of acquired resistance to cetux-imab treatment couldbe theHER2 (ERBB2) activation (11).Similarly, Bertotti and colleagues identified HER2 geneamplification as a mechanism of resistance to cetuximabin mCRC that harbor wild-type KRAS, NRAS, BRAF, andPIK3CA genes (12). In this respect, although HER2 geneamplification was found to occur in only 2% to 3% ofunselectedmCRCspecimens, a significantly higher frequen-cy was found in metastatic KRAS wild-type patients withcolorectal cancer that did not benefit from treatment withanti-EGFRmAbs (12). For this reason, we decided to exam-ine the expression and the activation of HER2 and of otherEGFR family members in GEO-CR and SW48-CR cells.However, no gene amplification and/or enhanced expres-sion of HERB2, HERB3, or ERBB4 was found.Our results suggest that MET activation may play a rel-

evant role in determining acquired resistance to cetuximab.Inparticular, gene expressionprofilingofGEOandGEO-CRcells identified a series of genes involved in HGF–MET-dependent pathway, that were upregulated in GEO-CR cellsas compared with GEO cells. Intriguingly, inhibition ofMET correlated with a partially restored sensitivity to cetux-imab in GEO-CR and SW48-CR cancer cells. In fact, METsilencing restores cetuximab ability to inhibit MAPK andAKT and cell proliferation. In contrast, HGF-mediated acti-vation of MET in parental GEO and SW48 cells reduced thesensitivity cetuximab. The role of MET in EGFR resistancewas also confirmed by using the selective MET inhibitorPHA665752 in combinationwith cetuximab that overcomethe resistance in GEO-CR and SW48-CR cells.Several studies have demonstrated that in lung adeno-

carcinoma–derived cells the EGFR inhibition can be over-come by MET (19, 22). Moreover, HGF-dependent METactivation also proved to be a mechanism of intrinsicresistance to gefitinib in NSCLC cells with EGFR-activatingmutations and no MET gene amplification (22). METamplification is associated with acquired resistance toanti-EGFR treatment in patients with mCRC who have notselected a KRAS mutation during the therapy (14). Inagreement with these results, it has been recently shown

thatHGF stimulation rescues cetuximab-sensitive colorectalcancer cells from EGFR inhibition by preventing cell-cyclearrest and inducing cell proliferations (40). Similarly, inHER2-overexpressing breast cancer cells, MET contributesto trastuzumab resistance (12). Conversely,MET-amplifiedgastric cancer cells were shown to be resistant to a MET-specific TKI when stimulated with EGF or heregulin-b (11).In these studies treatment of cancer cells with bothMET andEGFR inhibitors could overcome resistance to a singleinhibitor.

We have investigated the potentialmechanisms by whichMET can be activated. AlthoughHGF-driven autocrine loopas amechanism ofMET activation was not demonstrated inGEO-CR and SW48-CR cells, here we provide evidence of across talk between EGFR and MET. In fact, we found thatEGFR immunoprecipitated together with MET in GEO-CRand SW48-CR cells, but not in GEO and SW48 cells. Thisinteraction was observed in parental cells following TGF-atreatment. In agreement with these results, several studieshave shown an interaction between EGFR and MET thatallows activation of MET following stimulation of differentcell lines (A431, AKN-1, HepG2, AKN-1, HuH 6, andMRC5) with the EGFR selective ligands EGF or TGF-a(24). Stimulation of cells expressing both MET and EGFRwith EGF resulted in phosphorylation of MET, and stimu-lation with ligands for both receptors resulted in synergisticactivation of downstream modulators, indicating mutualactivation of these two pathways (41).

EGFR ligands have been studied in several cancers aspotential biomarkers for EGFR-targeted therapy. However,the results have been controversial depending on differenttumor types anddifferent clinical specimens used for testing(42, 43). In a study by Cohen and colleagues, changes inserum TGF-a levels in patients with head and neck cancertreated with gefitinib, were not associated with clinicalresponse to gefitinib (43). In a study by Mutsaers andcolleagues, TGF-a levels were increased in the plasma ofpatients with colorectal cancer during cetuximab treatment(44). However, increased TGF-a levels did not associatewith cetuximab response (45). In contrast, more consistentresults support the view that increased mRNA levels ofamphiregulin (AREG) and epiregulin (EREG) in the tumorsof patients with colorectal cancer are associated with cetux-imab sensitivity. Khambata-Ford and colleagues were thefirst to demonstrate that AREG and EREG were significantlyupregulated in chemotherapy refractory patients with colo-rectal cancer that obtained a clinical benefit from cetuximabtreatment (42). The predictive value of EREG and AREGexpression for cetuximab sensitivity was confirmed byJacobs and colleagues, which analyzed primary tumorsfrom refractory patients with mCRC treated with cetuxi-mab-based therapies (46). Furthermore, in the 045 phase Istudy of cetuximab monotherapy as first-line treatment ofpatients with mCRC, Tabernero and colleagues found thatAREG and EREG mRNA levels were significantly higher intumor samples from patients with clinical response tocetuximab as compared with patients with disease progres-sion (47). Although high level of expression of AREG and






EREG ligands may predict a better response to treatmentwith cetuximab in patients withmCRC, HB-EGF and TGF-aoverexpression could confer lack of sensitivity to EGFRinhibitors. In fact, in 045 clinical trial, TGF-a mRNA levelswere significantly elevated in the tumors from patients withmCRC with disease progression after cetuximab therapy ascompared with patient with clinical response (47). Further-more, increased HB-EGF may be correlated with cetuximabresistance in head and neck squamous cell carcinoma (48).

Taken together, these results suggest that enhancedexpression of different EGFR-specific ligands could have adifferent effect on the clinical activity of cetuximab. In thisrespect, the observation of TGF-a overexpression with acti-vation of MET signaling in human colorectal cancer cellswith acquired resistance to cetuximab treatment, asdescribed in the present study, is in agreement with theabove clinical findings.

The results of the present study suggest that overexpres-sion of TGF-a, which induces EGFR–MET interaction, playsa relevant role in determining the development of acquiredresistance in cetuximab-treated colorectal cancer cells.Therefore, the combined inhibition of both EGFR andMETRTKs could represent a rational therapeutic strategy forpreventing and/or overcoming cetuximab resistance inpatients with colorectal cancer.

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

Authors' ContributionsConception and design: T. Troiani, E. Martinelli, A. Capasso, A. Nappi,L. Berrino, R. Bianco, F. CiardielloDevelopment of methodology: T. Troiani, E. Martinelli, S. Napolitano,D. Vitagliano, A. Nappi, R. Bianco, F. CiardielloAcquisition of data (provided animals, acquired and managed pati-ents, provided facilities, etc.): T. Troiani, F. Morgillo, V. Sforza, A. Nappi,L. Berrino, F. CiardielloAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): T. Troiani, E. Martinelli, L.P. Ciuffreda,S. Costantino, F. Morgillo, V. Sforza, A. Nappi, R. De Palma, E. D’Aiuto,R. Bianco, F. CiardielloWriting, review, and/or revision of the manuscript: T. Troiani, E. Marti-nelli, F. Morgillo, A. Nappi, R. De Palma, L. Berrino, F. CiardielloAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): A. Nappi, F. CiardielloStudy supervision: R. De Palma, R. Bianco, F. Ciardiello

Grant SupportThis researchhas been supported by a grant from theAssociazione Italiana

per la Ricerca sul Cancro (AIRC).The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received February 14, 2013; revised September 20, 2013; acceptedSeptember 29, 2013; published OnlineFirst October 11, 2013.

References1. Malvezzi M, Arfe A, Bertuccio P, Levi F, La Vecchia C, Negri E.

European cancer mortality predictions for the year 2011. Ann Oncol2011;22:947–56.

2. Galizia G, Lieto E, De Vita F, Orditura M, Castellano P, Troiani T, et al.Cetuximab, a chimeric human mouse anti-epidermal growth factorreceptor monoclonal antibody, in the treatment of human colorectalcancer. Oncogene 2007;26:3654–60.

3. Ciardiello F, Tortora G. EGFR antagonists in cancer treatment. N EnglJ Med 2008;358:1160–74.

4. NormannoN, Tejpar S,Morgillo F, De LucaA, VanCutsemE,CiardielloF. Implications for KRAS status and EGFR-targeted therapies inmetastatic CRC. Nat Rev Clin Oncol 2009;6:519–27

5. Karapetis CS, Khambata-Ford S, Jonker DJ, O'Callaghan CJ, Tu D,Tebbutt NC, et al. K-ras mutations and benefit from cetuximab inadvanced colorectal cancer. N Engl J Med 2008;359:1757–65.

6. Amado RG, Wolf M, Peeters M, Van Cutsem E, Siena S, Freeman DJ,et al. Wildtype KRAS is required for panitumumab efficacy in patientswith metastatic colorectal cancer. J Clin Oncol 2008;26:1626–34.

7. De Roock W, Claes B, Bernasconi D, De Schutter J, Biesmans B,Fountzilas G, et al. Effects of KRAS, BRAF, NRAS, and PIK3CAmutations on the efficacy of cetuximab plus chemotherapy in chemo-therapy-refractory metastatic colorectal cancer: a retrospective con-sortium analysis. Lancet Oncol 2010;11:753–62.

8. Van Cutsem E, K€ohne CH, L�ang I, Folprecht G, Nowacki MP, CascinuS, et al. Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-line treatment for metastatic colorectal cancer: updated analysis ofoverall survival according to tumor KRAS and BRAF mutation status.J Clin Oncol 2011;29:2011–9.

9. Misale S, Yaeger R, Hobor S, Scala E, Janakiraman M, Liska D, et al.Emergence of KRAS mutations and acquired resistance to anti-EGFRtherapy in colorectal cancer. Nature 2012;486:532–6.

10. Diaz LA Jr, Williams RT, Wu J, Kinde I, Hecht JR, Berlin J, et al. Themolecular evolution of acquired resistance to targeted EGFR blockadein colorectal cancers. Nature 2012;486:537–40.

11. Yonesaka K, Zejnullahu K, Okamoto I, Satoh T, Cappuzzo F, Sougla-kos J, et al. Activation of ERBB2 signaling causes resistance to the

EGFR-directed therapeutic antibody cetuximab. Sci Transl Med 2011;3:99ra86.

12. Bertotti A, Migliardi G, Galimi F, Sassi F, Torti D, Isella C, et al. Amolecularly annotated platform of patient-derived xenografts ("xeno-patients") identifies HER2 as an effective therapeutic target in cetux-imab-resistant colorectal cancer. Cancer Discov 2011;1:508–23.

13. Martin V, Landi L, Molinari F, Fountzilas G, Geva R, Riva A, et al. HER2gene copy number status may influence clinical efficacy to anti-EGFRmonoclonal antibodies in metastatic colorectal cancer patients. Br JCancer 2013;108:668–75.

14. Bardelli A, Corso S, Bertotti A, Hobor S, Valtorta E, Siravegna G, et al.Amplification of the MET receptor drives resistance to anti-EGFRtherapies in colorectal cancer. Cancer Discov 2013;3:658–73.

15. Valtorta E, Misale S, Sartore-Bianchi A, Nagtegaal ID, Paraf F, Laur-icellaC, et al., KRASgeneamplification in colorectal cancer and impacton response to EGFR-targeted therapy. Int J Cancer 2013;133:1259–66.

16. Cecchi F, Rabe DC, Bottaro DP. Targeting the HGF/Met signallingpathway in cancer. Expert Opin Ther Targets 2012;16:553–72.

17. Kammula US, Kuntz EJ, Francone TD, Zeng Z, Shia J, Landmann RG,et al. Molecular co-expression of the c-Met oncogene and hepatocytegrowth factor in primary colon cancer predicts tumor stage and clinicaloutcome. Cancer Lett 2008;248:219–28.

18. Di Renzo MF, Olivero M, Giacomini A, Porte H, Chastre E, Mirossay L,et al. Overexpression and amplification of the met/HGF receptor geneduring the progression of colorectal cancer. Clin Cancer Res 1995;1:147–54.

19. Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland C, Park JO,et al. MET amplification leads to gefitinib resistance in lung cancer byactivating ERBB3 signaling. Science 2007;316:1039–43.

20. Turke AB, Zejnullahu K, Wu YL, Song Y, Dias-Santagata D, Lifshits E,et al. Preexistence and clonal selection of MET amplification in EGFRmutant NSCLC. Cancer Cell 2010;17:77–88.

21. Scagliotti GV,Novello S, Schiller JH,HirshV, Sequist LV, Soria JC, et al.Rationale and design of MARQUEE: a phase III, randomized, double-blind study of tivantinib plus erlotinib versus placebo plus erlotinib in

Troiani et al.





previously treated patients with locally advanced or metastatic, non-squamous, non–small cell lung cancer. Clin Lung Cancer 2012;5:391–5.

22. Goldman JW, Laux I, Chai F, Savage RE, Ferrari D, Garmey EG, et al.Phase 1 dose-escalation trial evaluating the combination of the selec-tive MET (mesenchymal-epithelial transition factor) inhibitor tivantinib(ARQ 197) plus erlotinib. Cancer 2012;118:5903–11.

23. Spigel DR, Edelman MJ, Mok T, O'Byrne K, Paz-Ares L, Yu W, et al.Treatment rationale study design for the MetLung trial: a randomized,double-blind phase iii study of onartuzumab (MetMAb) in combinationwith erlotinib versus erlotinib alone in patients who have receivedstandard chemotherapy for stage IIIB or IV met-positive non–small-cell lung cancer. Clin Lung Cancer 2012;6:500–4.

24. Jo M, Stolz DB, Esplen JE, Dorko K, Michalopoulos GK, Strom SC.Cross-talk between epidermal growth factor receptor and c-Met signalpathways in transformed cells. J Biol Chem 2000;275:8806–11.

25. Weinberger PM, Yu Z, Kowalski D, Joe J, Manger P, Psyrri A, et al.Differential expression of epidermal growth factor receptor, c-Met, andHER2/neu in chordoma compared with 17 other malignancies. ArchOtolaryngol Head Neck Surg 2005;131:707–11.

26. Pai R, Nakamura T, MoonWS, Tarnawski AS. Prostaglandins promotecolon cancer cell invasion; signaling bycross-talk between twodistinctgrowth factor receptors. FASEB J 2003;17:1640–7.

27. Ciardiello F, Bianco R, CaputoR, Caputo R, Damiano V, Troiani T, et al.Antitumor activity of ZD6474, a vascular endothelial growth factorreceptor tyrosine kinase inhibitor, in human cancer cells with acquiredresistance to antiepidermal growth factor receptor therapy. Clin Can-cer Res 2004;10:784–93.

28. Troiani T, VecchioneL,Martinelli E,CapassoA,CostantinoS,CiuffredaLP, et al. Intrinsic resistance to selumetinib, a selective inhibitor ofMEK1/2, by cAMP-dependent protein kinase A activation in humanlung and colorectal cancer cells. Br J Cancer 2012;106:1648–59.

29. Szodoray P, Alexy P, Brunz JG, Centolay M, Jonsson R. Circulatingcytokines in primary Sj€ogren's syndrome determined by a multiplexcytokine array system. Scand J Immunol 2004;59:592–9.

30. Normanno N, Petraroli R, Rico A, Kurth H, Tops B, Amato E, et al. TheOncoNetwork Consortium: A European Collaborative Research studyon the development of an Ion AmpliSeq gene panel targeting hot spotsin colon and lung cancers [abstract]. In: Proceedings of the 104thAmericanAssociation forCancerReasearchAnnualMeeting; 2013Apr6–10; Philadelphia, PA. Washington, DC: AACR; 2013. Abstract nr 36.

31. Ivanovska I, Zhang C, Liu AM, Wong KF, Lee NP, Lewis P, et al. Genesignatures derived from a c-MET-driven liver cancer mouse modelpredict survival of patients with hepatocellular carcinoma. PLoS ONE2011;6:e24582.

32. BoonEMJ, van derNeut R, vandeWeteringM,CleversH, Pals ST.Wntsignaling regulates expression of the receptor tyrosine kinase met incolorectal cancer. Cancer Res 2002;62:5126–28.

33. KnowlesLM,Stabile LP,Egloff AM,RothsteinME, ThomasSM,GubishCT , et al. HGF and c-met participate in paracrine tumorigenic path-ways in head and neck squamous cell cancer. Clin Cancer Res2009;15:3740–50.

34. Tanaka H, Yamamoto M, Hashimoto N, Miyakoshi M, Tamakawa S,Yoshie M , et al. hypoxia-independent overexpression of hypoxia-

inducible factor 1A as an early change in mouse hepatocarcinogen-esis. Cancer Res 2006;66:11263–70.

35. Martinelli E, Troiani T, Morgillo F, Rodolico G, Vitagliano D, Morelli MP,et al. Synergistic antitumor activity of sorafenib in combination withepidermal growth factor receptor inhibitors in colorectal and lungcancer cells. Clin Cancer Res 2010;16:4990–5001.

36. OrganSL, TsaoMS. Anoverviewof the c-MET signaling pathway. TherAdv Med Oncol 2011;3:S7–19.

37. Laurent-Puig P, Cayre A, Manceau G, Buc E, Bachet JB, Lecomte T.Analysis of PTEN, BRAF, and EGFR status in determining benefit fromcetuximab therapy in wild-type KRAS metastatic colon cancer. J ClinOncol 2009;27:5924–30.

38. Bardelli A, Siena S. Molecular mechanisms of resistance to cetuximaband panitumumab in colorectal cancer. J Clin Oncol 2010;28:1254–61.

39. Ciardiello F, Normanno N. HER2 signaling and resistance to the anti-EGFR monoclonal antibodt cetuximab: a further step toward person-alized medicine for patients with colorectal cancer. Cancer Discov2011;1:472–4.

40. Liska D, Chen CT, Bachleitner-Hofmann T, Christensen JG, WeiserMR. HGF rescues colorectal cancer cells fromEGFR inhibition viaMETactivation. Clin Cancer Res 2011;17:472–82.

41. Puri N, Salgia R. Synergism of EGFR and c-Met pathways, cross-talkand inhibition, in non–small cell lung cancer. J Carcinog 2008;7:9.

42. Khambata-Ford, Garrett CR,Meropol NJ, BasikM,HarbisonCT,WuS,et al. Expression of epiregulin and amphiregulin and K-ras mutationstatus predict disease control in metastatic colorectal cancer patientstreated with cetuximab. J Clin Oncol 2007;25:3230–7.

43. Cohen EE, Lingen MW, Martin LE, Harris PL, Brannigan BW, HaserlatSM, et al. Response of some head and neck cancers to epidermalgrowth factor receptor tyrosine kinase inhibitors may be linked tomutation of ERBB2 rather than EGFR. Clin Cancer Res 2005;11:8105–8.

44. Mutsaers AJ, Francia G, Man S, Lee CR, Ebos JM, et al. Dose-dependent increases in circulating TGF-alpha and other EGFR ligandsact as pharmacodynamic markers for optimal biological dosing ofcetuximab and are tumor independent. Clin Cancer Res 2009;15:2397–405.

45. Chan E, LaFleur B, Rothenberg ML, Merchant N, Lockhart AC,Trivedi D, et al. Dual blockade of the EGFR and COX-2 pathways:a phase II trial of cetuximab and celecoxib in patients with chemo-therapy refractory metastatic colorectal cancer. Am J Clin Oncol2011;34:581–6.

46. Jacobs B, De RoockW, PiessevauxH, VanOirbeek R, Biesmans B, DeSchutter J, et al. Amphiregulin and epiregulin mRNA expression inprimary tumors predicts outcome in metastatic colorectal cancertreated with cetuximab. J Clin Oncol 2009;27:5068–74.

47. Tabernero J, Cervantes A, Rivera F, Rodriguez-Braun E, Ramos FJ,Martinelli E, et al. Pharmacogenomic and pharmacoproteomic studiesof cetuximab in metastatic colorectal cancer: biomarker analysis of aphase I dose-escalation study. J Clin Oncol 2010;28:1181–9.

48. HatakeyamaH,ChengH,WirthP,Counsell A,MarcromSR,WoodCB ,et al. Regulation of heparin-binding EGF-like growth factor bymiR-212and acquired cetuximab-resistance in head and neck squamous cellcarcinoma. PLoS ONE 2010;5:e12702.






2013;19:6751-6765. Published OnlineFirst October 11, 2013.Clin Cancer Res Teresa Troiani, Erika Martinelli, Stefania Napolitano, et al. Activation of MET Signaling in Colon Cancer Cells

MET Interaction and−Anti-EGFR Inhibitor Cetuximab through EGFR as a Mechanism of Acquired Resistance to the αIncreased TGF-

Updated version

10.1158/1078-0432.CCR-13-0423doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2013/10/09/1078-0432.CCR-13-0423.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/19/24/6751.full#ref-list-1

This article cites 47 articles, 22 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/19/24/6751.full#related-urls

This article has been cited by 21 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/19/24/6751To request permission to re-use all or part of this article, use this link


http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-13-0423http://clincancerres.aacrjournals.org/content/suppl/2013/10/09/1078-0432.CCR-13-0423.DC1http://clincancerres.aacrjournals.org/content/19/24/6751.full#ref-list-1http://clincancerres.aacrjournals.org/content/19/24/6751.full#related-urlshttp://clincancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://clincancerres.aacrjournals.org/content/19/24/6751http://clincancerres.aacrjournals.org/

Increased TGF-a as a Mechanism of Acquired Resistance to ...Authors' Afﬁliations: 1Oncologia...

Documents

Transcript of Increased TGF-a as a Mechanism of Acquired Resistance to ...Authors' Afﬁliations: 1Oncologia...