Research Synergistic Antitumor Activity of Sorafenib in ... · extension step at 72°C. PCR...

Can

SynwithCol

ErikaConceS. Ga

Abst

Tumdepeninducgrowt

AuthorClinicaFarmaUniverOncoloPatologNapoli,Health

Note: SResear

Erika M

CorresChirurgLanzar80131fortuna

doi: 10

©2010

Clin C4990

Dow

Published OnlineFirst September 1, 2010; DOI: 10.1158/1078-0432.CCR-10-0923

Clinical
Canceresearch
cer Therapy: Preclinical

ergistic Antitumor Activity of Sorafenib in CombinationEpidermal Growth Factor Receptor Inhibitors in

R

orectal and Lung Cancer Cells

Martinelli1, Teresa Troiani1, Floriana Morgillo1, Gabriella Rodolico2, Donata Vitagliano3, Maria Pia Morelli1,4,
tta Tuccillo1, Loredana Vecchione1, Anna Capasso1, Michele Orditura1, Ferdinando De Vita1, il Eckhardt4, Massimo Santoro3, Liberato Berrino2, and Fortunato Ciardiello1
ractPur

inducExp

in vitrH129Res

colorelorectA, B, aCom

betwein alland Aand recombmentdelay

ed angioh factors

s' Affiliatioe Sperime

cologia, Dsità degli Sgia Sperimia CellularItaly; and 4

Sciences C

upplementach Online (h

artinelli and

ponding Aico di Intea”, SecondNapoli, Italyto.ciardiello

.1158/1078-

American A

ancer Res

nloaded

pose: Cancer cell survival, invasion, andmetastasis dependon cancer cell proliferation andon tumor-ed angiogenesis.We evaluated the efficacy of the combination of sorafenib and erlotinib or cetuximab.erimental Design: Sorafenib, erlotinib, and cetuximab, alone or in combination, were testedo in a panel of non–small cell lung cancer (NSCLC) and colorectal cancer cell lines and in vivo in9 tumor xenografts.ults: Epidermal growth factor receptor (EGFR) ligand mRNAs were expressed in all NSCLC andctal cancer cell lines with variable levels ranging from 0.4- to 8.1-fold as compared with GEO co-al cancer cells. Lung cancer cells had the highest levels of vascular endothelial growth factors (VEGF)nd C, and of VEGF receptors as compared with colorectal cancer cells.bined treatments of sorafenib with erlotinib or cetuximab produced combination index values

en 0.02 and 0.5, suggesting a significant synergistic activity to inhibit soft agar colony formationcancer cell lines, which was accompanied by a marked blockade in mitogen-activated protein kinaseKT signals. The in vitro migration of H1299 cells, which expressed high levels of both VEGF ligandsceptors, was inhibited by treatment with sorafenib, and this effect was significantly increased by theination with anti-EGFR drugs. In nude mice bearing established human H1299 xenografts, treat-with the combination of sorafenib and erlotinib or cetuximab caused a significant tumor growthresulting in 70 to 90 days increase in mice median overall survival as compared with single-agentnib treatment.
sorafe
Conclusions: Combination treatment with sorafenib and erlotinib or cetuximab has synergistic anti-tumor effects in human colorectal and lung cancer cells. Clin Cancer Res; 16(20); 4990–5001. ©2010 AACR.

velopgrowt

or cell growth, survival, invasion, and metastasisd on efficient tumor cell proliferation and tumor-

genesis. Several autocrine and paracrineregulate these processes during cancer de-

non–sare thtors a(EGFRing togrowtand HVas

role indotheferenVEGF(PlGFdifferVEGFgiogened prVEGFof di

ns: 1Dipartimento Medico-Chirurgico di Internisticantale “F. Magrassi e A. Lanzara”, and 2Sezione diipartimento di Medicina Sperimentale, Secondatudi di Napoli, and 3Istituto di Endocrinologia edentale del CNR, c/o Dipartimento di Biologia ee e Molecolare, Università di Napoli “Federico II”,Division of Medical Oncology, University of Coloradoenter, Aurora, Colorado

ry data for this article are available at Clinical Cancerttp://clincancerres.aacrjournals.org/).

Teresa Troiani contributed equally to this study.

uthor: Fortunato Ciardiello, Dipartimento Medico-rnistica Clinica e Sperimentale “F. Magrassi e A.a Università degli Studi di Napoli, Via Pansini 5,. Phone: 39-81-5666745; Fax: 39-815666732; E-mail:@unina2.it.

0432.CCR-10-0923

ssociation for Cancer Research.

; 16(20) October 15, 2010

Research. on September 11,clincancerres.aacrjournals.org from

ment and progression. In this respect, key autocrineh regulators for human epithelial cancers, includingmall cell lung cancer (NSCLC) and colorectal cancer,e epidermal growth factor (EGF)-related growth fac-nd their cognate receptors (1). The EGF receptor) is a transmembrane growth factor receptor belong-a family of four related proteins [human epidermalh factor receptor 2 (HER2)/ERBB2, HER3/ERBB3,ER4/ERBB4; refs. 2, 3].cular endothelial growth factors (VEGF) play a keycancer-induced neoangiogenesis by promoting en-

lial cell proliferation, migration, and functional dif-tiation (4). Several ligands, including VEGF-A,-B, VEGF-C, VEGF-D, and placental growth factor) mediate their effects through the activation of threeent VEGF tyrosine kinase receptors, VEGFR-1,R-2 and VEGFR-3, on endothelial cells. The proan-ic actions of VEGFs in endothelial cells are mediat-imarily through the binding and activation of
R-2 (5–8). The recent discovery of the productionfferent VEGF ligands and of the expression of
2020. © 2010 American Association for Cancer

http://clincancerres.aacrjournals.org/

VEGFdirectcontroexampmanand thsociatand gcell pVEGFof VEand wVEGFmigrawas re3 withblocksatic aThe

tors b(TGFαman cpropepart, bsis byblockaothergrowtseveration itiallyTheref

bettercominblockErlo

sine kspeciftreatmtuximclonahighe(21).ceptorout rcurrenthe trcancetherapneckSor

hibitothe wFLT-3growtclinicabroadassayssphoPDGFtreatmvancetion iNSCLIn t

tion o10 huaim of developing models for a rational multitargeted can-cer tre

Mate

DrugSor

It waworkiby MelutionItalysoluti

Cell lThe

NSCLHCT1obtainwere

Translational Relevance

In this study we provide experimental evidence thatepidermal growth factor receptor (EGFR)-dependentand the vascular endothelial growth factor receptor(VEGFR)-dependent pathways are activated and func-tionally linked in human non–small cell lung cancer(NSCLC) and colorectal cancer cell lines. The activa-tion of VEGF-dependent signaling could be responsi-ble of intrinsic resistance to anti-EGFR therapies.Therefore, in addition to anti-EGFR drugs the simulta-neous blockade of RAF and VEGF signaling with amultitargeted inhibitor such as sorafenib could obtaina significant antitumor inhibition of cancer cellgrowth. We think this study could have a relevant clin-ical interest because it is an extensive attempt in a largepanel of selected human NSCLC and colorectal cancercell lines to define a rational approach to combinedmolecular targeted therapies. The in vitro and in vivoresults could be used for the design of translationalresearch-based clinical studies with sorafenib in com-bination with erlotinib and cetuximab in patients withNSC

Combining Molecular Targeted Drugs against Tumors

www.a

Dow


R-1 and VEGFR-3 in epithelial cancer cells suggests arole for these ligands and receptors in the autocrinel of some biological processes in cancer cells. Forle, VEGFR-1 expression has been detected in hu-colorectal cancer and pancreatic cancer cell lines,e activation of this receptor by VEGF has been as-ed with enhanced cancer cell invasion, migration,rowth in soft agar, without direct effects on cancerroliferation (9, 10). Further evidence for a role ofRs in cancer cell migration has been the detectionGF-C and of VEGFR-3 in NSCLC patient sampleshat has been shown to be an active VEGF-C/

R-3 autocrine pathway related to lung cancer celltion in vitro and in vivo (11). In this respect, itcently shown that targeting VEGFR-1 and VEGFR-cediranib, a pan-VEGFR tyrosine kinase inhibitor,in vitro migration and invasion of human pancre-

nd hepatic carcinoma cell lines (12).activation of EGFR and related growth factor recep-y EGF, heregulin, or transforming growth factor α) can upregulate the production of VEGF-A in hu-ancer cells (13, 14), suggesting that the oncogenicrties of the EGFR-driven pathway may, at least ine mediated by the stimulation of tumor angiogene-upregulating angiogenic growth factors. Thus, EGFRde causes inhibition of the secretion of VEGF and ofangiogenic growth factors, including basic fibroblasth factor, interleukin 8, and TGFα (15, 16). Moreover,l studies have shown the role of VEGF-A upregula-n the acquired resistance to EGFR treatment in ini-

LC and colorectal cancer.

EGFR inhibitor–sensitive cancer cells (17, 18).ore, targeting both these pathways could provide a

menteicillin

acrjournals.org

Research. on September 11,clincancerres.aacrjournals.org nloaded from

anticancer therapeutic strategy, especially for over-g the acquired resistance of cancer cells to EGFRade (19, 20).tinib is a small molecule that inhibits EGFR tyro-inase activity, thus blocking EGFR activation byic ligands. Erlotinib is currently available for theent of metastatic chemorefractory NSCLC (3). Ce-ab is a chimeric human-mouse anti-EGFR mono-l antibody that binds to EGFR with a two-logr affinity than the natural ligands TGFα and EGFThe binding of cetuximab to EGFR promotes re-internalization and subsequent degradation with-

eceptor phosphorylation and activation (3). It istly used in combination with chemotherapy ineatment of wild-type K-RAS metastatic colorectalr and in combination with radiotherapy or chemo-y in locally advanced and in metastatic head andcancer (22, 23).afenib is a small molecule multitargeted kinase in-r that blocks the activation of C-RAF, B-RAF (bothild-type and the activated V600E mutant), c-KIT,, RET, VEGFR-2, VEGFR-3, and platelet-derivedh factor receptor β (PDGFR-β; refs. 24, 25). Pre-l studies have shown that sorafenib is active in aspectrum of tumor types (26). In in vitro cellular, sorafenib inhibits the ligand-induced autopho-rylation of VEGFR-1, VEGFR-2, VEGFR-3, andR-β (27). Sorafenib is currently approved for theent of metastatic renal cell carcinoma and ad-

d hepatocellular carcinoma, and is under investiga-n phase II/III trials in other malignancies, includingC (25).his study, we evaluated the efficacy of the combina-f sorafenib and erlotinib or cetuximab in a panel ofman NSCLC and colorectal cancer cell lines with the

atment approach.

rials and Methods

safenib was kindly provided by Bayer Schering Italy.s dissolved in sterile DMSO, and a 10 mmol/Lng solution was prepared. Cetuximab was suppliedrck-Serono Pharma Italy, and a 2 mg/mL stock so-was used. Erlotinib was kindly given by Roche

and was diluted in DMSO to 10 μmol/L as stockon.

ineshuman GLC82, H1299, A549, CALU3, and H460C cell l ines and the human HT29, HCT15,16, SW480, and GEO colorectal cancer cell lines wereed from the American Type Culture Collection. Cellscultured in RPMI 1640 medium (Cellgro) supple-
d with 10% fetal bovine serum (Hyclone), 1% pen-/streptomycin (Invitrogen), and 1% nonessential
Clin Cancer Res; 16(20) October 15, 2010 4991



aminolial cewith E(Lonzincubwere(Myco

Real tEGF

ERBBand VmRNAin colPCR w(Life T4 μgprocephosphouse3 minThe aextenon 1%ethidiisolateto thetase w

GrowCel

Difcocultu(18).of sorwere soniesexpericentrasorafe0.01-5eachsoraferesulterlotimethoware p

EvalufactorThe

in thecommassaymanuResultand reSubse

CM wserumconce

EvaluCel

fore tdescriscribeSignalEGFRcers aSTAT3humapp44bodiebodyImmuchemexperi

WounCel

tic ditip ansorafe(2.5 μtinibphotohealintweenSysteMeOHpresenoriginwas d

ApopH1

72 hocell dthat sfor prKit (Isortincytomwashepropi

TumoFou

mice wreseartainedthe SeComm

Martinelli et al.

Clin C4992

Dow


acid (Invitrogen). Human umbilical vein endothe-lls were obtained from Lonza Inc. and were culturedGM SingleQuots growth factor supplemented mediaa, Inc.). All cell lines were grown in a humidifiedator with 5% carbon dioxide (CO2) at 37°C. androutinely screened for the presence of mycoplasmaAlert, Cambrex Bio Science).

ime-PCR analysis, TGFα, amphiregulin, ERBB1, ERBB2, ERBB3,

4, VEGF-A, VEGF-B, VEGF-C, VEGFR-1, VEGFR-2,EGFR-3 genes were amplified by real time-PCR fromusing appropriate primers to detect their expression

orectal cancer and in NSCLC cell lines. Real time-as done on total RNA extracted with Trizol reagentechnologies). First-strand cDNA was prepared fromof total RNA according to the Pharmacia Biotechdure and amplified by PCR. Glyceraldehyde-3-hate dehydrogenase (GAPDH) was used as a controlkeeping gene. Each sample was denatured (95°C;utes), and PCR was carried out using 40 cycles.mplification was terminated by a final 10-minutesion step at 72°C. PCR products were separatedagarose gels and visualized under UV light after

um bromide staining. The fragments detected wered and sequenced to confirm that they correspondedgenes listed above. In controls, the reverse transcrip-as omitted.

th in soft agarls (104 cells/well) were suspended in 0.5 mL 0.3%Noble agar (Difco) supplemented with completere medium and seeded as previously describedCells were treated with different concentrationsafenib, erlotinib, or cetuximab, and after 14 daystained with nitroblue tetrazolium (Sigma,) and col->0.05 mm were counted. For the combinationments, cancer cells were treated with different con-tions of erlotinib (range, 0.02-10 μmol/L) plusnib (range, 0.01-5 μmol/L) and sorafenib (range,μmol/L) plus cetuximab (range, 0.02-10 μg/mL)

day, for a total of 3 days, at the fixed drug rationib:erlotinib or sorafenib:cetuximab of 1:2. Thes of the combined treatment with sorafenib andnib or cetuximab were analyzed according to thed of Chou and Talalay by using the CalcuSyn soft-rogram (Biosoft), as previously reported (28).

ation of EGF- and VEGF-related growthsecretionconcentrations of EGF, TGFα, VEGF-A, and VEGF-Cconditioned medium (CM) were measured usingercially available enzyme-linked immunosorbent(ELISA) kits (R&D Systems Inc.) according to thefacturer's instructions. Assays were done in triplicate.s were normalized for the number of producing cells
ported as pg of ligands per 106 cells per 72 hours.quently, the secretions of VEGF-A and VEGF-C in the
Unive1 wee

ancer Res; 16(20) October 15, 2010


ere evaluated after stimulation for 24 hours in-free medium of cancer cells with two differentntrations of EGF (10 and 20 ng/mL).

ation of protein expression by Western blottingls were seeded into 6-multiwell plates 24 hours be-reatment with each drug or drug combination asbed above. The analysis was done as previous de-d (18). We used the following antibodies from Celling Technology: rabbit polyclonal antibody against, phosphorylated EGFR, P-AKT, AKT, signal transdu-nd activators of transcription 3 (STAT3), and p-(Tyr 705); mouse monoclonal antibodies against

n p44/42 mitogen activated protein kinase (MAPK)/42 MAPK (Thr202/Tyr20); goat polyclonal anti-s against p44/42 MAPK; and goat polyclonal anti-against β-actin (Santa Cruz Biotechnology).noreactive proteins were visualized by enhancediluminescence. (Amersham International). Eachment was done in triplicate.

d-healing scratch assayl monolayers grown to confluence on gridded plas-shes were wounded by scratching with a pipetted then cultured in the presence or absence ofnib (1 μmol/L), erlotinib (1 μmol/L), cetuximabg/mL), or the combination of sorafenib and erlo-or cetuximab for 24 hours. The wounds weregraphed (10× objective) at 24 hours (10), andg was quantified by measuring the distance be-the edges using Adobe Photoshop (v.8.0.1; Adobe

ms, Inc.) after being washed, fixed with 90%, and stained with crystal violet. The results areted as the percentage of the total distance of theal wound filled by cancer cells. Each experimentone in triplicate.

tosis assay and cell cycle299 cells were collected in 6-well plates, treated forurs, and stained with Annexin V–FITC. Apoptoticeath was assessed by counting the numbers of cellstained positive for Annexin V–FITC and negativeopidium iodide using an Apoptosis Annexin V–FITCnvitrogen), coupled with fluorescence-activated cellg analysis. Cell cycle analysis was done by using flowetry on cell pellets that were fixed in 70% ethanol,d in PBS, and mixed with RNase (Invitrogen) anddium iodide (Invitrogen).

r xenografts in nude micer- to six-week-old female BALB/c athymic (nu+/nu+)ere purchased from Charles River Laboratories. The

ch protocol was approved and the mice were main-in accordance with the institutional guidelines ofcond University of Naples Animal Care and Useittee. The mice were acclimatized at the Second

rsity of Naples Medical School Animal Facility fork prior to being injected with cancer cells. The mice

Clinical Cancer Research



were isuspe7 day0.3 cmwith odays 1of erland/oweek)ment,umediame

StatisStu

icanceof staBMDPcrosof

Resu

ExprecognaFirs

coloreEGF rthreevolvedtativepositipressevariabparedamphHCT1GLC8in allwhichpressecells.rectalfour NpresenFinallwas foof GLWe

gandsfour hto serditionTGFαof TGand SFig. 1EGFRcolore

Littledetectwhichtheseof thethreeNSCLsion cselecte

ExpreNSCLWe

VEGFVEGFthe pshowntectediablecolorecell linpressimRNAcer ceHCT1in H1contracanceVEGFNSCLabsenand Cinto thsion fsion lbothencescanceand recell linNSCLand reBec

tion ccell litreatmdose-VEGFwith Etary Fcross-of theand c

Effectcolor


www.a

Dow


njected s.c. with 107 H1299 cells that had been re-nded in 200 μL Matrigel (Becton Dickinson). Afters, when established tumors of approximately 0.2 to3 diameter were detected, the mice were treatedral administration of sorafenib (50 mg/kg/day on-5 of each week) and/or with oral administrationotinib (75 mg/kg/day on days 1-5 of each week)r i.p. with cetuximab (1 mg on days 2-5 of eachfor the indicated time periods. For each experi-treatment groups comprised 10 mice. Tumor vol-was measured using the formula π/6 × largerter × (smaller diameter)2.

tical analysisdent's t test was used to evaluate the statistical signif-of the results. All P values represent two-sided tests

tistical significance. All analyses were done with theNew System statistical package version 1.0 for Mi-

t Windows (BMDP Statistical Software).

lts

ssion of EGF-related growth factors and theirte receptorst we characterized a panel of 10 human NSCLC andctal cancer cell lines for the expression of the foureceptors (EGFR, ERBB2, ERBB3, and ERBB4) andligands (amphiregulin, EGF, and TGFα) that are in-in EGFR-driven autocrine growth by using quanti-

real time-PCR. GEO colon cancer cells were used as ave control. The mRNAs of the three ligands were ex-d in all NSCLC and colorectal cancer cell lines withle levels ranging between 0.4- and 8.1-fold as com-with GEO colorectal cancer cells (Fig. 1A). Of note,iregulin expression was higher in CALU3 and16 cells, whereas EGF was more expressed in2, H460, and GEO cells. TGFα was widely detectedcancer cell lines, with the exception of A549 cells init was barely detectable. All cancer cell lines ex-d EGFR mRNA with the highest levels in GLC-82Low levels of ERBB2 mRNA were measured in colo-cancer cell lines and in CALU3, whereas the otherSCLC cell lines were negative. ERBB3 mRNA wast in all cell lines, with highest levels in GEO cells.y, low or no detectable expression of ERBB4 mRNAund in all cancer cell lines tested, with the exceptionC82 cells (Fig. 1A).next analyzed the secretion of two EGFR-specific li-, EGF and TGFα, into the cell culture media. Twenty-ours after cell seeding, culture medium was switchedum-free medium and CM was collected after an ad-al 72 hours. As shown in Fig. 1B, both EGF andwere secreted by all cell lines, with the highest levelsFα production in human CALU3, HT29, HCT116,W480 colorectal cancer cells. As illustrated inC, Western blot analysis showed the presence of
in all cancer cell lines. ERBB2 was expressed in allctal cancer cell lines and in CALU3 NSCLC cells.
Togrowt

acrjournals.org


or no ERBB3 and ErbB4 protein expression wased in all cancer cell lines, except for GLC-82 cells,were positive for ERBB4 (Fig. 1C). Taken together,

finding show that a high variability in the expressionfour EGFR-related growth factor receptors and ofEGFR ligands is observed in colorectal cancer andC cell lines, suggesting that this differential expres-ould influence cancer cell biology and response tod molecular targeted inhibitors.

ssion of VEGFRs and their specific ligands inC and colorectal cancer cell linesnext evaluated the expression of VEGFR (VEGFR-1,R-2, VEGFR-3) and their ligand (VEGF-A, VEGF-B,-C) mRNAs by using quantitative real time-PCR inanel of NSCLC and colorectal cancer cell lines. Asin Fig. 2A, VEGF-A and VEGF-B mRNAs were de-in all NSCLC and colorectal cancer cell lines at var-levels. VEGF-C mRNA expression was absent inctal cancer cell lines and was detected in all NSCLCes with the exception of CALU3 cancer cells. The ex-on of VEGFRs was more limited. In fact, VEGFR-2was not detectable in GLC82, A549, and GEO can-lls; it was barely measurable in HT29, HCT15,16, and SW480 cancer cells; and it was expressed299, CALU3, and H460 NSCLC cells (Fig. 2A). Inst, VEGFR-1 mRNA expression was observed in allr cell lines with the exception of A549 cells. Finally,R-3 mRNA expression was limited to four of fiveC cell lines (GLC82, H1299, A549, H460) and wast in colorectal cancer lines. As illustrated in Fig. 2B, we analyzed the secretion of VEGF-A and VEGF-Ce cell culture media and the levels of protein expres-or VEGFR-1, VEGFR-2, and VEGFR-3. Protein expres-evels confirmed the results observed for mRNAs forligands and receptors. In this respect, several differ-could be observed between NSCLC and colorectalr cells. With the exception of VEGFR-1, all ligandsceptors were expressed at a higher degree in NSCLCes both at mRNA and at protein level. In particular,C H1299 cells express high levels of all VEGF ligandsceptors (Fig. 2A-C).ause it was previously shown that VEGF-A produc-ould be induced by EGFR activation in some cancernes (7, 8), we evaluated if EGFR stimulation by EGFent could induce VEGF-A and VEGF-C secretion. Adependent upregulation in the secretion of both-A and VEGF-C into the CM following treatmentGF was observed in all cancer cell lines (Supplemen-ig. S1). These results suggest that there is a functionaltalk between the EGFR pathway and the productionse two ligands of the VEGF family in human NSCLColorectal cancer cells.

of erlotinib and cetuximab on NSCLC andectal cancer cell growth in soft agar
evaluate the effects of EGFR inhibition on cancer cellh, soft agar anchorage-independent growth assays



were dconcecountof anservedsensit

NSCLbothcells wlorect

Fig. 1. EquantitSW480ErbB4 mELISA areceptors. Thirty micrograms of cell protein extracts were fractionated through 4% to 20% SDS-PAGE, transferred to nitrocellulose filters, and incubatedwith the active

Martinelli et al.

Clin C4994

Dow


one. Cells were treated for 72 hours with differentntrations of erlotinib or cetuximab. Colonies wereed after 10 to 14 days. A dose-dependent inhibitionchorage-independent growth in soft agar was ob-

appropriate antibodies as described in Materials and Methods. Immunore

in all cancer cells lines with a different degree ofivity, as illustrated in Fig. 3A and B. Among the

and c1 μmo



C cell lines, CALU3 cells were the most sensitive toerlotinib and cetuximab. On the contrary, H1299ere intrinsically resistant to EGFR inhibition. All co-al cancer cell lines were sensitive to both erlotinib

proteins were visualized by enhanced chemiluminescence.

xpression of EGF-related growth factors and growth factor receptors in human NSCLC and colorectal cancer cells. A, specific mRNA expression byative real time-PCR: total RNA was extracted from NSCLC (GLC82, H1299, A549, CALU3, H460) and colorectal cancer (HT29, HCT15, HCT116,, GEO) cell lines and quantitative real time-PCR was done to assess the expression of amphiregulin, EGF, TGFα, EGFR, ErbB2, ErbB3, andRNA. B, secretion of EGF and TGFα into the conditioned medium. EGF and TGFαprotein levels weremeasured in cell culturemedia by using specifics described in Materials and Methods. HUVEC, human umbilical vein endothelial cells. C, Western blotting analysis of EGFR-related growth factor

etuximab with an IC50 ranging between 0.3 andl/L for erlotinib and 1 to 3.6 μg/mL for cetuximab.




Effectcell gWe

A dosgrowtcells lfromMu

ly 40NSCLto EGtion (clusivof 10ence obored

was mNSCLity toB-RAF

EffectinhibgrowtTo

tinib ocer ceerloti0.01-5cetuxi

Fig. 2.quantitSW480VEGFRmedia bmicrogantibod


www.a

Dow


s of sorafenib on NSCLC and colorectal cancerrowth in soft agarnext tested the antiproliferative effects of sorafenib.e-dependent inhibition of anchorage-independenth in semisolid medium was observed in all cancerines treated with sorafenib with the IC50 ranging0.2 to 2.8 μmol/L (Fig. 3C).tations within the K-RAS that occur in approximate-% of colorectal cancer and in 15% to 25% ofC (22) have been correlated to intrinsic resistanceFR inhibitors (29). An activating B-RAF gene muta-V600E) is less frequent and generally mutually ex-e with K-RAS mutations (30). We tested the panelNSCLC and colorectal cancer cell lines for the pres-
f such mutations. Eight of 10 cancer cell lines har- 3 days
rams of cell protein extracts were fractionated through 4% to 20% SDS-PAGE, tranies as described in Materials and Methods. Immunoreactive proteins were visua

acrjournals.org


utated only in HT29 colorectal cancer and H1299C cells. No significant correlation between sensitiv-sorafenib or to anti-EGFR drugs and K-RAS orgene status was observed (Supplementary Fig. S2).

s of sorafenib in combination with EGFRitors on NSCLC and colorectal cancer cellh in soft agarevaluate the interaction between sorafenib and erlo-r cetuximab, combination analyses were done. Can-lls were treated with different concentrations ofnib (range, 0.02-10 μmol/L) and sorafenib (range,μmol/L) or sorafenib (range, 0.01-5 μmol/L) and

mab (range, 0.02-10 μg/mL) each day, for a total of
at a fixed drug ratio sorafenib:erlotinib or sorafenib:
an activating K-RAS gene mutation, whereas B-RAF cetuximab of 1:2. Colonies were counted after 14 days.

Expression of VEGF-related growth factors and growth factor receptors in human NSCLC and colorectal cells. A, specific mRNA expression byative real time-PCR. Total RNA was extracted from NSCLC (GLC82, H1299, A549, CALU3, H460) and colorectal cancer (HT29, HCT15, HCT116,, GEO) cell lines, and quantitative real time-PCR was done to assess the expression of VEGF-A, VEGF-B, VEGF-C, VEGFR-1, VEGFR-2, and-3 mRNA. B, secretion of VEGF-A and VEGF-C into the conditioned medium. VEGF-A and VEGF-C protein levels were measured in cell culturey using specific ELISA as described in Materials and Methods. C, Western blotting analysis of VEGFR-related growth factor receptors. Thirty

sferred to nitrocellulose filters, and incubated with the appropriatelized by enhanced chemiluminescence.




Exper(CI) vTalalathe Catreatmcausedwith evaluesand 0to inhThesementswere o

EffectinhibcanceSev

tion wincreaagentFig. Stive atributtreatm

. 3.ctsatioraumcerterias wer

uximafene corese inh done in triplicate.

Martinelli et al.

Clin C4996

Dow


iments were done in triplicate. Combination indexalues were calculated according to the Chou andy mathematical model for drug interactions usinglcusyn software (28). In all 10 cancer cell lines tested,ent with the multitargeted inhibitor sorafenibsynergistic growth inhibitory effects in combinationach EGFR inhibitor (Fig. 4A and B). In fact, the CIfor the combined treatments ranged between 0.02.5, suggesting a significant synergism in the abilityibit soft agar colony formation in all cancer cell lines.results were comparable for both combination treat-
. Similar results in anchorage-dependent growthbtained (data not shown).
G1 phof sor



s of sorafenib in combination with EGFRitors on apoptosis and cell cycle in H1299r cellsenty-two-hour treatment with sorafenib in combina-ith either erlotinib or cetuximab induced a markedse in apoptosis as compared with control or single-treatments for both combinations (Supplementary3A). To further explore this synergistic antiprolifera-ctivity, we investigated the effects on cell cycle dis-ion. As compared with erlotinib or cetuximab,ent with sorafenib induced an increase in the G0-

Figeffeformor sin hcanMacellwithcetsorwerThethreeac

ase of the cell cycle. In particafenib plus erlotinib caused

Clini

2020. © 2010 American Ass

Evaluation of the inhibitoryon soft agar colonyon of erlotinib, cetuximab,fenib given as single agentan NSCLC and colorectalcells. As described inls and Methods, cancerere treated for 72 hourslotinib (0.01-5 μmol/L),ab (0.01-5 μg/mL), orib (0.01-5 μmol/L). Coloniesunted after 10 to 14 days.ults are average ± SD ofdependent experiments

ular, the combinationan almost complete

cal Cancer Research

ociation for Cancer


arrestthe G

EffectinhibH129To

obtainbitorstivatioand/ofromperiodor witWe setant tharboof bothat ca decrhibitiTreatmor cetthe levP-MA

maintbinedstreaminhibipressotein 1(pp70werewithinlatedtreatedrugs

EffectinhibcanceWe

cells cagenteithercells band Vand V

Fig. 4.effectsformaticombincetuximcoloreccells wconcenerlotinibcetuximMateriawere coExperimtriplicataccordmodelthe Calfractionin Mate


www.a

Dow


of cells in the G0-G1 phase with further reduction in2-M and S phases (Supplementary Fig. S3B).

s of sorafenib in combination with EGFRitors on intracellular mitogenic signaling in9 cancer cellsdetermine whether the synergistic growth inhibitioned by the combination of sorafenib and EGFR inhi-was due to a more effective inhibition of EGFR ac-n and the intracellular signaling through MAPKr AKT, Western blots were done on protein extractsH1299 NSCLC cells that were treated for differents of time with sorafenib, cetuximab, or erlotinib,h combinations of sorafenib and an anti-EGFR drug.lected these cells because they are intrinsically resis-o treatment with anti-EGFR drugs as single agent,r a V600E B-RAF mutation, and express high levelsth VEGF ligands and receptors. Figure 5 illustratesetuximab or erlotinib treatment, although causingease on EGFR phosphorylation, has little effect in in-ng downstream mitogenic and prosurvival signals.ent with sorafenib in combination with erlotinib

uximab resulted in a more pronounced decrease in
els of examined proteins phosphorylation (P-EGFR,PK, P-AKT) after 60 minutes of tr
assays

rials and Methods.

acrjournals.org


ained up to 120 minutes of treatment. The com-treatment also affected P-AKT–dependent down-signaling (mTOR) as suggested by the sustained

tion of the phosphorylation of the translational re-r protein eukaryotic initiation factor 4E binding pro-(P4E-BP1) and of the ribosomal protein S6 kinaseS6K). Furthermore, total levels of STAT-3 proteinnot substantially changed by treatments. However,30 minutes a marked reduction in the phosphory-

form of STAT-3 was observed in H1299 cancer cellsd with sorafenib in combination with anti-EGFR(Supplementary Fig. S3C).

s of sorafenib in combination with EGFRitors on in vitro migratory capabilities of H1299r cellsassessed if the migratory potential of H1299 NSCLCould be affected by treatment with the multitargetedsorafenib or by EGFR blockade by treatment witherolotinib or cetuximab. We selected H1299 cancerecause they express high levels of VEGF-A, VEGF-B,EGF-C, as well as high levels of VEGFR-1, VEGFR-2EGFR-3. As determined by in vitro scratch motility
, the 24-hour migration ability of H1299 cancer cells
eatment, which was was not affected by single-agent treatment with either

Evaluation of the inhibitoryon soft agar colonyon of sorafenib ination with erlotinib or withab in human NSCLC andtal cancer cells. Cancerere treated with differenttrations of sorafenib plus(A) or of sorafenib plusab (B), as described inls and Methods. Coloniesunted after 14 days.ents were done ine. CI values were calculateding to the Chou and Talalayfor drug interactions usingcusyn software for differents affected (fa), as described




erlotinminedover,cetuxi90%)was pVEGF(Supp

AntituEGFRWe

afenibH129whenwereweeksmab omor gtumormenteffectH129servedcontra

fromsorafewerewith nFig. Stumowith eof 145mice (sorafesurvivtreate(Tablmentsdose acol wother

Discu

In tligandpanel

Fig. 5.Analysi1 μmol/SDS-PAprotein

Martinelli et al.

Clin C4998

Dow


ib or cetuximab, whereas sorafenib treatment deter-a partial (30-50%) inhibition of migration. More-the combined treatment with sorafenib andmab or erlotinib induced an almost complete (80-suppression on H1299 migratory behavior, whichrobably due to the simultaneous inhibition of both- and EGF-dependent pathways in these cellslementary Fig. S4).

mor activity of sorafenib in combination withinhibitors on H1299 tumor xenograftsnext investigated the in vivo antitumor activity of sor-and/or cetuximab or erlotinib in nude mice bearing

9 lung adenocarcinoma s.c. xenografts. After 7 days,established H1299 tumors of approximately 0.3 cm3

detectable, groups of 10 mice were treated for 4with each agent alone or in combination. Cetuxi-r erlotinib as single agents were unable to affect tu-rowth, whereas a 70% to 80% inhibition of H1299growth was caused by sorafenib single-agent treat-at the end of the 4 weeks of therapy. However, thiswas reversible upon cessation of treatment because9 tumors resumed a growth rate similar to that ob-
in control untreated mice within 3 to 4 weeks. Inst, a prolonged tumor growth inhibition resulted
the ptors. W

GE, transferred to nitrocellulose filters, and incubated with the appropriate antibs were visualized by enhanced chemiluminescence.



the combination of sorafenib and cetuximab or ofnib and erlotinib. As a surrogate of survival, micesacrificed when tumor burden was not compatibleormal life (approximately 2.5 cm3; Supplementary

4). In Table 1 and in Supplementary Fig. S4, H1299rs in mice treated with sorafenib in combinationrlotinib for 4 weeks survived for an average period(SD, ±5) days as compared with control, untreated35 ± 2 days), erlotinib-treated mice (37 ± 2 days), ornib-treated mice (75 ± 5 days). A similar increase inal was observed in H1299 xenograft-bearing miced with the combination of sorafenib and cetuximabe 1 and Supplementary Fig. S5). Combined treat-with sorafenib and cetuximab or erlotinib at thend schedule tested for the 4-week treatment proto-ere well tolerated by mice, with no weight loss orsigns of acute or delayed toxicity (data not shown).

ssion

he present study we investigated the expression ofs and receptors of the EGF and VEGF families in aof colorectal cancer and NSCLC cell lines, to assess
otential functional role of these ligands and recep-hereas EGF-related growth factors and their cognate
Western blot analysis of protein expression following treatment with sorafenib, erlotinib, cetuximab, alone or in combination in H1299 NSCLC cells.s of intracellular signaling pathways was done by Western blotting in H1299 cancer cells treated with 2.5 μg/mL cetuximab, 2 μmol/L erlotinib,L sorafenib, alone or with the indicated combinations for the indicated time periods. Total cell protein extracts were fractionated through 4% to 20%

odies as described in Materials and Methods. Immunoreactive




receptin themetasin epirespecof cancer papressiThe

EGFRgulin,cancelinesis actiand EmostlNSCLlevelsticulaVEGFVEGFpresenman cway pendotquiredfirst recer ceVEGFthe mof VEan ausuggecells,cance

VEGFstimuaffectinvasiFurthto thecanceinhibVEGFwith vand imacquirThe

cross-way adepencondifollowThe

EGFRdepenin vitrtreatmnib, cof C-Bothtive tThis ein thecantanti-Eobservsorafe

Table of sorafen r cetuximacanc

Treatm Averaafter t

Avera

ControErlotinSorafeCetErloCet

NOTfolloErloSoraCetTreaMice were sacrificed when tumor volume reached 2.5 cm3 (approximately 10% of a nude mouse body weight).


www.a

Dow


ors are widely recognized as key autocrine regulatorscontrol of cancer cell proliferation, invasion, and

tasis (1), the role of VEGFR expression and functionthelial cancer cells is more controversial (9). In thist, although VEGFRs are expressed in a wide varietycer cell lines as well as in tumor specimens from can-tients, little is known on the significance of their ex-on on cancer cell functions (8).results of the present study show the expression ofand three EGFR-specific ligands, i.e., EGF, amphire-and TGFα, in a panel of 10 NSCLC and colorectalr cell lines, suggesting that in all these cancer cella potential EGFR-driven autocrine growth pathwayve. On the contrary, expression of ERBB2, ERBB3,RBB4 is more restricted, with ERBB2 being expressedy in colorectal cancer cells. We also showed thatC and colorectal cancer cell lines express differentof mRNA and proteins of the VEGF families. In par-r, NSCLC cells had the highest levels of VEGF-A,-B, and VEGF-C, and of VEGFR-1, VEGFR-2, andR-3 as compared with colorectal cancer cells. Thece of active VEGF/VEGFR signaling pathways in hu-ancer cells is an important issue because this path-lays a critical role in angiogenesis by promoting thehelial cell proliferation, invasion, and migration re-for neovascularization (5). Ellis and colleagues (8)ported the expression of VEGFR-1 in colorectal can-ll lines and suggested that VEGFR-1 activation by-A and VEGF-B could increase the migration andetastatic capabilities of these cells. In fact, expressionGFR-1 and VEGFR-2 and their ligands may sustaintocrine loop in some human cancers, as has beensted in melanoma, mesothelioma, and leukemia
in which exogenous VEGF stimulation inducesr cell proliferation and migration by activating
mentsmetho

acrjournals.org


R-2 (31–33). It was also shown recently that ligand-lated activation of VEGFR-1 and VEGFR-3 does notcell proliferation but induces in vitro migration andon of human gastrointestinal cancer cell lines (9).ermore, overexpression of VEGFR-1 may contributeacquisition of anti-EGFR drug resistance in humanr cells. In this respect, VEGFR-1 gene silencing withition of VEGFR-1 protein expression or blockade ofR-1 autophosphorylation and signaling by treatmentandetanib restored the sensitivity to anti-EGFR drugspairedmigration abilities in human cancer cells withed resistance to gefitinib or to cetuximab (34).results of the present study also suggest a functionaltalk between the EGFR-dependent autocrine path-nd the control of angiogenesis because a dose-dent increase in VEGF-A and VEGF-C release in thetioned media of all cancer cell lines was observeding treatment with EGF.refore, to obtain a simultaneous blockade of the-dependent autocrine pathway and of VEGFR-dent signaling in cancer cells, we evaluated theo and in vivo antitumor activity of the combinedent with selective anti-EGFR drugs, such as erloti-etuximab, and sorafenib, a multitargeted inhibitorRAF and B-RAF and of all three VEGFRs (3, 24).NSCLC and colorectal cancer cell lines were sensi-o the antiproliferative effects of sorafenib in vitro.ffect does not correlate with gene mutations with-RAS-RAF signaling pathway, because no signifi-

correlation between sensitivity to sorafenib or toGFR drugs and K-RAS or B-RAF gene status wased. To evaluate the degree of interaction betweennib and erlotinib or cetuximab, combination treat-

1. Antitumor activityer xenografts

ib in combination with erlotinib o

were analyzed according td. A synergistic inhibition

Clin Cancer Re

2020. © 2010 American A

b in H1299 human

ent
ge tumor volume (±SD) on day 35umor cell injection, cm3
ge survival (±SD), days

l
2.5 (±0.2) 35 (±2) ib 2.5 (±0.3) 37 (±4) nib 0.56 (±0.01) 75 (±5) mab 2.55 (±0.2) 34 (±5) uxi
tinib plus sorafenib 0.15 (±0.04) 145 (±5)uximab plus sorafenib 0.10 (±0.03) 165 (±10)

E: Each treatment (erlotinib, sorafenib, cetuximab, erlotinib plus sorafenib, or cetuximab plus sorafenib) was started on day 7wing H1299 tumor cell s.c. injection when the average tumor volume was 0.30 (±0.05) cm3.tinib treatment: 75 mg/kg/dose orally 5 days/week starting on the day 7 following tumor cell injection.fenib treatment: 50 mg/kg/dose orally 5 days/week starting on the day 7 following tumor cell injection.

uximab treatment: 1 mg/dose i.p. twice weekly starting on the day 7 following tumor cell injection.tment was done for 4 weeks. Each group consisted of 10 mice.

o the Chou and Talalayof colony formation in

s; 16(20) October 15, 2010 4999

ssociation for Cancer


soft athe comoreowereagentWe

becauanti-EV600EpressiTGFαlevelsVEGFthat call thbeen ttaneofor abinedafenibcancerbitionin thefects oin H1inhibiinhibtheraptivationalinresista(18, 3In v

EGFR

xenogactiviusedwith ticantmatelsurvivIn c

bothpathwity oftion oresistNSCLthe siwith atain acancecept tlinkedstrateg

Discl

No p

Grant

GranThe

paymenadvertiindicate

Refe1. Sa

facCri

2. Mecan

3. CiaEn

4. Ferica

5. Hicpa10

6. Ze49:

7. WerecOn

8. Ellgrocat

9. Faencel

10. Legro

Martinelli et al.

Clin C5000

Dow


gar was observed in all cancer cell lines followingmbination with sorafenib plus anti-EGFR drugs;ver, antiproliferative effects with the combinationobserved, although maximally inhibitory single-dose was used.selected H1299 NSCLC cells for further experimentsse they are intrinsically resistant to treatment withGFR drugs as single agent (35, 36), harbor aactivating B-RAF mutation, and have the lowest ex-on of amphiregulin and EGF and high levels of, as ligands for EGFR (37); indeed they present highof both VEGF ligands and VEGFR-1, VEGFR-2, andR-3. This expression may sustain an autocrine loopould be related with cell migration. In fact, amonge NSCLC and colorectal cancer cell lines that haveested, H1299 cells exhibit the highest level of spon-us migration. The migratory activity is a crucial stepcancer cell in the development of metastasis. Com-treatment with sorafenib plus erlotinib or with sor-plus cetuximab significantly inhibited H1299cell migration, probably for the simultaneous inhi-of both the VEGF and the EGF signaling pathwaysse cells. The antiproliferative and antimigratory ef-f the combination of anti-EGFR drugs and sorafenib299 cancer cells were accompanied by a significanttion of the MAPK and AKT pathways. The sustainedition of both the AKT and the MAPK pathways iseutically relevant because the EGFR-independent ac-n of these growth factor receptor downstream sig-g molecules may represent one of the causes ofnce to EGFR-targeted therapies in human cancer4, 38, 39).
ivo experiments in nude mice bearing established Rece
ls. Oncogene 2005;24:2647–53.sslie DP, Summy JM, Parikh NU, et al. Vascular endothelialwth factor receptor-1 mediates migration of human colorectal

ca20

11. Suan

12. MogroonCa

13. Gilind(VPac

14. Godeproma

15. Brumaitopa

16. Hirgetyr

17. Vilothe



rafts showed a significant cooperative antitumorty when sorafenib and cetuximab or erlotinib werein combination. In this respect, a 4-week treatmenthe two agents was well tolerated and caused a signif-tumor growth delay which determined an approxi-y 70- to 90-day increase in mice median overallal as compared with single-agent sorafenib treatment.onclusion, the results of the current study show thatthe EGFR-dependent and the VEGFR-dependentays are activated and functionally linked in a major-human NSCLC and colorectal cell lines. The activa-f VEGF-dependent signaling could drive an intrinsicance to anti-EGFR therapies, as shown in H1299C cells. Therefore, in addition to anti-EGFR drugsmultaneous blockade of RAF and VEGF signalingmultitargeted inhibitor such as sorafenib could ob-sustained and significant antitumor inhibition of

r cell growth in vitro and in vivo, supporting the con-hat molecular targeting of distinct but functionallycancer cell pathways could represent a suitabley in the treatment of cancer.

osure of Potential Conflicts of Interest

otential conflicts of interest were disclosed.

Support

t form the Associazione Italiana per la Ricerca sul Cancro (AIRC).costs of publication of this article were defrayed in part by thet of page charges. This article must therefore be hereby markedsement in accordance with 18 U.S.C. Section 1734 solely tothis fact.

ived 04/16/2010; revised 07/30/2010; accepted 08/18/2010;
inhibitor–resistant H129 9 lung adenocarcinoma published OnlineFirst 09/01/2010.
renceslomon DS, Brandt R, Ciardiello F, Normanno N. Epidermal growthtor-related peptides and their receptors in human malignancies.t Rev Oncol Hematol 1995;19:183–232.ndelsohn J, Baselga J. The EGF receptor family as targets forcer therapy. Oncogene 2000;19:6550–65.rdiello F, Tortora G. EGFR antagonists in cancer treatment. Ngl J Med 2008;358:1160–74.rara N. The role of vascular endothelial growth factor in patholog-l angiogenesis. Breast Cancer Res Treat 1995;36:127–37.klin DJ, Ellis LM. Role of the vascular endothelial growth factorthway in tumor growth and angiogenesis. J Clin Oncol 2005;23:11–27.tter BR. Angiogenesis and tumor metastasis. Annu Rev Med 1998;407–24.y JS, Stoeltzing O, Ellis LM. Vascular endothelial growth factoreptors: expression and function in solid tumors. Clin Adv Hematolcol 2004;2:37–45.is LM, Takahashi Y, Liu W, Shaheen RM. Vascular endothelialwth factor in human colon cancer: biology and therapeutic impli-ions. Oncologist 2000;5 Suppl 1:11–5.n F, Wey JS, McCarty MF. Expression and function of vasculardothelial growth factor receptor-1 on human colorectal cancer

rcinoma cells by activation of Src family kinases. Br J Cancer06;94:1710–7.JL, Yang PC, Shih JY. The VEGF-C/Flt-4 axis promotes invasiond metastasis of cancer cells. Cancer Cell 2006;9:209–23.relli MP, Brown AM, Pitts TM. Targeting vascular endothelialwth factor receptor-1 and -3 with cediranib (AZD2171): effectsmigration and invasion of gastrointestinal cancer cell lines. Molncer Ther 2009;8:2546–58.l J, Swerlick RA, Caughman SW. Transforming growth factor -uced transcriptional activation of the vascular permeability factorF/VEGF) gene requires AP2-dependent DNA binding and trans-tivation. EMBO J 1997;16:750–9.ldman CK, Kim J, Wong WL, King V, Brock T, Gillespie GY. Epi-rmal growth factor stimulates vascular endothelial growth factorduction by human malignant glioma cells: a model of glioblasto-multiforme pathophysiology. Mol Biol Cell 1993;4:121–33.ns CJ, Solorzano CC, Harbison MT, et al. Blockade of the epider-l growth factor receptor signaling by a novel tyrosine kinase inhib-r leads to apoptosis of endothelial cells and therapy of humanncreatic carcinoma. Cancer Res 2000;60:2926–35.ata A, Ogawa S, Kometani T. ZD1839 (Iressa) induces antiangio-nic effects through inhibition of epidermal growth factor receptor
osine kinase. Cancer Res 2002;62:2554–60.ria-Petit A, Crombet T, Jothy S, et al. Acquired resistance toantitumor effect of epidermal growth factor receptor-blocking



antRe

18. Ciaa vtorgro

19. Todestu

20. NadotorEG

21. Kimog

22. NoCiapie

23. Maepthe

24. WiPreboMo

25. Tak(Neclin

26. Lyokin

27. Wispewaang

28. TroSeanZDMo

29. Ba

ma12

30. RaVerep

31. BylatCe

32. Diaofan

33. StranJ P

34. Biafacfac14

35. Mooftortio

36. ThtiolineCa

37. TaphcaOn

38. Janepcaptoreg9:2

39. Ma


www.a

Dow


ibodies in vivo: a role for altered tumor angiogenesis. Cancers 2001;61:5090–101.rdiello F, Bianco R, Caputo R, et al. Antitumor activity of ZD6474,ascular endothelial growth factor receptor tyrosine kinase inhibi-, in human cancer cells with acquired resistance to antiepidermalwth factor receptor therapy. Clin Cancer Res 2004;10:784–93.rtora G, Ciardiello F, Gasparini G. Combined targeting of EGFR-pendent and VEGF-dependent pathways: rationale, preclinicaldies and clinical applications. Nat Clin Pract Oncol 2008;5:521–30.umov GN, Nilsson MB, Cascone T, et al. Combined vascular en-thelial growth factor receptor and epidermal growth factor recep-(EGFR) blockade inhibits tumor growth in xenograft models ofFR inhibitor resistance. Clin Cancer Res 2009;15:3484–94.ES, Khuri FR, Herbst RS. Epidermal growth factor receptor biol-

y (IMC-C225). Curr Opin Oncol 2001;13:506–13.rmanno N, Tejpar S, Morgillo F, De Luca A, Van Cutsem E,rdiello F. Implications for KRAS status and EGFR-targeted thera-s in metastatic CRC. Nat Rev Clin Oncol 2009;6:519–27.rtinelli E, De Palma R, Orditura M, De Vita F, Ciardiello F. Anti-idermal growth factor receptor monoclonal antibodyes in cancerrapy. Clin Exp Immunol 2009;158:1–9.lhelm SM, Adnane L, Newell P, Villanueva A, Llovet JM, Lynch M.clinical overview of sorafenib, a multikinase inhibitor that targetsth Raf and VEGF and PDGF receptor tyrosine kinase signaling.l Cancer Ther 2008;7:3129–40.imoto CH, Awada A. Safety and anti-tumor activity of sorafenibxavar®) in combination with other anti-cancer agents: a review ofical trials. Cancer Chemother Pharmacol 2008;61:535–48.ns JF, Wilhelm S, Hibner B, Bollag G. Discovery of a novel Rafase inhibitor. Endocr Relat Cancer 2001;8:219–25.lhelm SM, Carter C, Tang L, et al. BAY 43–9006 exhibits broadctrum oral antitumor activity and targets the RAF/MEK/ERK path-y and receptor tyrosine kinases involved in tumor progression andiogenesis. Cancer Res 2004;64:7099–109.iani T, Lockerbie O, Morrow M, Ciardiello F, Eckhardt SG.quence-dependent inhibition of human colon cancer cell growthd of prosurvival pathways by oxaliplatin in combination with
6474 (Zactima), an inhibitor of VEGFR and EGFR tyrosine kinases.l Cancer Ther 2006;5:1883–94.rdelli A, Siena S. Molecular mechanisms of resistance to cetuxi-
facofJ C

acrjournals.org


b and panitumumab in colorectal cancer. J Clin Oncol 2010;28:54–61.jagopalan H, Bardelli A, Lengauer C, Kinzler KW, Vogelstein B,lculescu VE. Tumorigenesis: RAF/RAS oncogenes and mismatch-air status. Nature 2002;418:934.zova TV, Goldman CK, Pampori N, et al. A mechanism for modu-ion of cellular responses to VEGF: activation of the integrins. Molll 2000;6:851–60.s S, Shmelkov SV, Lam G, Rafii S. VEGF (165) promotes survivalleukemic cells by Hsp90-mediated induction of Bcl-2 expressiond apoptosis inhibition. Blood 2002;99:2532–40.izzi L, Catalano A, Vianale G. Vascular endothelial growth factor isautocrine growth factor in human malignant mesothelioma.athol 2001;193:468–75.nco R, Rosa R, Damiano V, et al. Vascular endothelial growthtor receptor-1 contributes to resistance to anti-epidermal growthtor receptor drugs in human cancer cells. Clin Cancer Res 2008;:5069–80.rgillo F, Woo JK, Kim ES, Hong WK, Lee HY. Heterodimerizationinsulin-like growth factor receptor/epidermal growth factor recep-and induction of survivin expression counteract the antitumor ac-n of erlotinib. Cancer Res 2006;66:10100–11.omson S, Buck E, Petti F, et al. Epithelial to mesenchymal transi-n is a determinant of sensitivity of non-small lung carcinoma cells and xenograft to epidermal growth factor receptor inhibition.ncer Res 2005;65:9455–62.bernero J, Cervantes A, Rivera F, et al. Pharmacogenomic andarmacoproteomic studies of cetuximab in metastatic colorectalncer: biomarker analysis of a phase I dose-escalation study. J Clincol 2010;28:1181–9.maat ML, Kruyt FA, Rodriguez JA, Giaccone G. Response toidermal growth factor receptor inhibitors in nonsmall cell lungncer cells: limited antiproliferative effects and absence of apo-sis associated with persistent activity of extracellular signal-ulated kinase or Akt kinase pathways. Clin Cancer Res 2003;316–26.gne N, Fischel JL, Dubreuil A, et al. Influence of epidermal growthtor receptor (EGFR), p53 and intrinsic MAP kinase pathway status
tumour cells on the antiproliferative effect of ZD1839 (“Iressa”). Brancer 2002;86:1518–23.



2010;16:4990-5001. Published OnlineFirst September 1, 2010.Clin Cancer Res Erika Martinelli, Teresa Troiani, Floriana Morgillo, et al. and Lung Cancer Cells

Colorectalwith Epidermal Growth Factor Receptor Inhibitors in Synergistic Antitumor Activity of Sorafenib in Combination

Updated version

10.1158/1078-0432.CCR-10-0923doi:

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/16/20/4990.full#ref-list-1

This article cites 39 articles, 21 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/16/20/4990.full#related-urls

This article has been cited by 6 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] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/16/20/4990To request permission to re-use all or part of this article, use this link

Research. on September 11, 2020. © 2010 American Association for Cancerclincancerres.aacrjournals.org Downloaded from


http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-10-0923

http://clincancerres.aacrjournals.org/content/16/20/4990.full#ref-list-1

http://clincancerres.aacrjournals.org/content/16/20/4990.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/16/20/4990


Research Synergistic Antitumor Activity of Sorafenib in ... · extension step at 72°C. PCR...

Documents

Transcript of Research Synergistic Antitumor Activity of Sorafenib in ... · extension step at 72°C. PCR...