sequence-dependent inhibition of human colon cancer cell...

Sequence-dependent inhibition of human colon cancercell growth and of prosurvival pathways by oxaliplatinin combination with ZD6474 (Zactima), an inhibitorof VEGFR and EGFR tyrosine kinases

Teresa Troiani,1 Owen Lockerbie,1 Mark Morrow,1

Fortunato Ciardiello,2 and S. Gail Eckhardt1

1University of Colorado Cancer Center, Aurora, Colorado and2Cattedra di Oncologia Medica, Dipartimento Medico-Chirurgicodi Internistica Clinica e Sperimentale ‘‘F Magrassi e A Lanzara,’’Seconda Universita degli Studi di Napoli, Naples, Italy

AbstractTo date, clinical studies combining the new generation oftargeted therapies and chemotherapy have had mixedresults. Preclinical studies can be used to identify potentialantagonism/synergy between certain agents, with thepotential to predict the most efficacious combinations forfurther investigation in the clinical setting. In this study,we investigated the sequence-dependent interactions ofZD6474 with oxaliplatin in two human colon cell linesin vitro. We evaluated the in vitro antitumor activity ofZD6474, an inhibitor of vascular endothelial growth factorreceptor (VEGFR), epidermal growth factor receptor (EGFR)and RET tyrosine kinase activity, and oxaliplatin usingthree combination schedules: ZD6474 before oxaliplatin,oxaliplatin before ZD6474, and concurrent exposure. Cellproliferation studies showed that treatment with oxalipla-tin followed by ZD6474 was highly synergistic, whereasthe reverse sequence was clearly antagonistic as wasconcurrent exposure. Oxaliplatin induced a G2-M arrest,which was antagonized if the cells were previously orconcurrently treated with ZD6474. ZD6474 enhancedoxaliplatin-induced apoptosis but only when added afteroxaliplatin. The sequence-dependent antitumor effectsappeared, in part, to be based on modulation of compen-satory prosurvival pathways. Thus, expression of total andactive phosphorylated EGFR, as well as AKT and extra-cellular signal-regulated kinase, was markedly increased

by oxaliplatin. This increase was blocked by subsequenttreatment with ZD6474. Furthermore, the synergisticsequence resulted in reduced expression of insulin-likegrowth factor-I receptor and a marked reduction in secre-tion of vascular endothelial growth factor protein. ZD6474in combination with oxaliplatin has synergistic antiproli-ferative properties in human colorectal cancer cell linesin vitro when oxaliplatin is administered before ZD6474.[Mol Cancer Ther 2006;5(7):1883–94]

IntroductionColon cancer is the third most common type of canceramong men and women in the United States (1). Thetreatment of advanced colorectal cancer consisted offluoropyrimidine-based chemotherapy for several decades.Fortunately, in the last few years, the standard care forpatients with metastatic colon cancer has shifted to moreactive regimens consisting of combinations of irinotecan,oxaliplatin, and 5-fluorouracil. The addition of irinotecanor oxaliplatin to 5-fluorouracil/leucovorin resulted inimproved response rates and progression-free survival inlarge, randomized trials; moreover, irinotecan-containingregimens resulted in improved overall survival (2–4).Although traditional chemotherapy has improved out-comes for patients with advanced colorectal cancer, theseagents are still associated with limited efficacy, indicatingthat new therapies to combat this disease are urgentlyneeded. Several novel targeted agents are being investigat-ed both as single agents and in combination withchemotherapy to assess the potential for increased efficacy.Two of these, cetuximab, a monoclonal antibody (mAb)against the epidermal growth factor receptor (EGFR), andbevacizumab, a mAb against the vascular endothelialgrowth factor (VEGF) protein, have already shown clinicalbenefit and are commercially approved for use in advancedcolorectal cancer. Thus, agents that target the EGFR andVEGF pathways have been clinically benchmarked in thisdisease.

The EGFR is overexpressed in many epithelial tumors,including 65% to 70% of colorectal cancer. Increased levelsof the EGFR have been associated with more aggressivedisease and a poor prognosis as manifested by reduceddisease-free and overall survival. For these reasons, theEGFR is being exploited as an antitumor target (5–9).Recently, the results of a phase II trial revealed thesignificant clinical activity of cetuximab, a chimeric IgG1mAb that binds competitively to the external domain ofEGFR, in patients with advanced, irinotecan-refractorycolorectal cancer. Patients who received combination

Received 2/13/06; revised 4/11/06; accepted 5/16/06.

Grant support: AstraZeneca Pharmaceuticals (Macclesfield, UnitedKingdom).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

Requests for reprints: S. Gail Eckhardt, University of Colorado CancerCenter, 12801East17thAvenue,Aurora,CO80010.Phone:303-724-3850;Fax: 303-724-3892. E-mail: [email protected]

Copyright C 2006 American Association for Cancer Research.

doi:10.1158/1535-7163.MCT-06-0055

1883

Mol Cancer Ther 2006;5(7). July 2006

Research. on February 15, 2019. © 2006 American Association for Cancermct.aacrjournals.org Downloaded from

http://mct.aacrjournals.org/

therapy with cetuximab and irinotecan experienced asuperior response rate (22.9% versus 10.8%) and prolongedtime to tumor progression (4 versus 1.5 months) whencompared with those who received monotherapy withcetuximab (10).

Another important pathway in the pathogenesis andprogression of colorectal cancer is the VEGF pathway,which is associated with the induction of angiogenesis.Tumor-related neovascularization plays a critical role incolorectal cancer progression, and this has been welldocumented (11, 12). Several studies have shown thatVEGF expression is increased in the progression fromnonmalignant to malignant colon tumors (13–15). In fact,the expression of this growth factor was highest incarcinomas followed by adenomas and was lowest innormal mucosa (15). Both microvessel density and expres-sion of VEGF have been associated with progression ofcolorectal cancer, development of metastases, and a poorprognosis (16–23). The approaches that have been pro-posed for blocking the VEGF receptor (VEGFR)/ligandsystem include a neutralizing mAb to VEGF (bevacizumab)and inhibitors of the VEGFR2 tyrosine kinase (SU11248,ZD6474, AZD2171, and PTK787/ZK22854). Many of theseare currently in preclinical and clinical development,whereas bevacizumab, recently, received regulatory ap-proval for the treatment of metastatic colorectal cancer(24–33). In fact, results from a randomized phase III trialshowed that the addition of bevacizumab to irinotecan,5-fluorouracil, and leucovorin yielded significant increasesin the objective response rate (45% versus 35%) and inoverall survival (20.3 versus 15.6 months) as well asimprovement in progression-free survival (10.6 versus 6.2months; ref. 34).

ZD6474 is an orally bioavailable, small-molecule inhibitorof VEGFR signaling (IC50 = 60 nmol/L for VEGF-dependent in vitro human umbilical vascular endothelialcell proliferation) and EGFR signaling (IC50 = 170 nmol/Lfor EGF-dependent in vitro human umbilical vascularendothelial cell proliferation; refs. 35, 36). ZD6474 alsohas activity against both VEGFR3 tyrosine kinase and RETsignaling (37). Chronic oral administration of ZD6474 hasbeen shown to produce a dose-dependent inhibition oftumor growth against a histologically diverse panel ofhuman tumor xenograft models (breast, colon, lung, andprostate vulva; ref. 35). In preclinical studies, the in vivoantitumor and antiangiogenic activity of ZD6474 is poten-tiated by its use in combination with paclitaxel (36). Inphase I studies, daily oral ZD6474 was shown to begenerally well tolerated at doses up to 300 mg/d. Commondose-related adverse events included diarrhea, rash, andasymptomatic QTc prolongation. Pharmacokinetic varia-bles revealed that absorption and elimination of ZD6474after a single oral dose was slow with a half-life of f120hours (38). ZD6474 is currently undergoing clinical evalu-ation and showed prolongation of progression-free survivalin two recent phase II trials in non–small cell lung cancer (39,40). Phase III trials of ZD6474 in non–small cell lung cancerhave been initiated.

In certain preclinical and clinical scenarios, the additionof agents that target VEGFR or EGFR signaling pathways tostandard chemotherapy seems to provide greater activitythan standard chemotherapy alone. However, despite theextensive number of clinical trials conducted with combi-nations of chemotherapy and targeted agents, relativelylittle is known about the sequence-dependent antitumoreffects of combining targeted therapies and chemotherapy.We hypothesized that sequence-dependent antitumoreffects of combining ZD6474 with oxaliplatin may existand that these effects may be, in part, based on modulationof the EFGR and its downstream effector pathways. Agreater understanding of the combinatorial effects of che-motherapy and agents targeting specific survival pathwaysin cancer may lead to strategies of chemopotentiation thatare distinct from those currently being tested in the clinic.

Materials andMethodsDrugsZD6474 [N-(4-bromo-2-fluorophenyl)-6-methoxy-7-[(1-

methylpiperidin-4-yl)methoxy]quinazolin-4-amine] waskindly provided by AstraZeneca Pharmaceuticals (Maccles-field, United Kingdom). A 10-mmol/L working solution inDMSO was prepared and stored in aliquots at �20jC.Oxaliplatin was obtained from the pharmacy of theUniversity of Colorado Health Science Center (Aurora,CO) and dissolved in water. C225 (cetuximab) anti-EGFRhuman-mouse chimeric anti-EGFR IgG1 mAb was suppliedby ImClone Systems (New York, NY). UO126 (1,4-diamino-2,3-dicyano-1,4-bis[2-aminophenylthio]butadiene), a selec-tive inhibitor of mitogen-activated protein/extracellularsignal-regulated kinase (ERK) kinase (MEK) 1/2, wasprovided by Cell Signaling (Beverly, MA). Stock solutionswere prepared at 10 mmol/L in DMSO and stored inaliquots at �80jC. These drugs were diluted in culturemedium to obtain the desired concentrations immediatelybefore use.

Cell LinesThe human colon cancer cell lines HCT-116 (p53 mutated;

Ras wild-type) and HT29 (p53 wild-type; Ras mutated)were obtained from the American Type Culture Collection(Manassas, VA). Cells were routinely cultured in RPMI1640 supplemented with 10% fetal bovine serum, 1%nonessential amino acids, and 1% penicillin/streptomycinand were maintained at 37jC in an incubator under anatmosphere containing 5% CO2. HT29-MEK1 (R4F) cellswere kindly provided by Dr. Pamela L. Rice (University ofColorado Health Science Center, Denver, CO) and culturedin RPMI 1640 supplemented with 600 Ag/mL hygromycin.The cells were routinely screened for the presence ofMycoplasma (MycoAlert, Cambrex Bioscience, Baltimore)and exposed to the drugs when they reached f70%confluence.

Evaluation of Cytotoxicity and Combination EffectsCellular cytotoxicity was determined by the 3-(4,5-

dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromideassay. Briefly, cells in the exponential growth phase were

ZD6474 and Oxaliplatin in Colon Cancer Cell Lines1884




transferred to 96-well flat-bottomed plates with lids. Cellsuspensions (100 AL) containing 3,000 viable cells for HT29and 2,000 for HCT-116 were plated into each well andincubated overnight before exposure with different con-centrations of drugs. Initially, HT29 and HCT-116 cellswere exposed to increasing concentrations of ZD6474 (1–21Amol/L for 48 hours) and oxaliplatin alone (3–30 Amol/Lfor 24 and 48 hours) in addition to these three sequences:(a) oxaliplatin followed by ZD6474 (OXA-ZD6474), cellswere exposed to oxaliplatin for 24 hours and then themedium containing drug was removed and cells wererinsed once with HBSS followed by exposure to ZD6474 for48 hours; (b) ZD6474 followed by oxaliplatin (ZD6474-OXA), cells were exposed to ZD6474 for 48 hours and thenthe medium containing drug was removed and cells wererinsed once with HBSS followed by exposure to oxaliplatinfor 24 hours; and (c) concurrent drug exposure (ZD6474 +OXA), cells were exposed to both ZD6474 and oxaliplatinfor 48 hours.

Each drug sequence was seeded into three wells and wasdone in triplicate. At the end of each time point, 40 AL/wellof a tetrazolium compound, 3-(4,5-dimethylthiazol-2.yl)-5-(3-carboxymethxyphenyl)-2-(4sulfopheyl)-2H-tetrazolium,were added (Promega, Madison, WI). Cells were thenincubated at 37jC with 5% CO2 for 4 hours. Plates wereread on a plate reader set at 490 nm absorbance. The resultsof the combined treatment were analyzed according to themethod of Chou and Talalay by using the Calcusynsoftware program (Biosoft, Ferguson, MO). The resultingcombination index (CI) is a quantitative measure of thedegree of interaction between different drugs. When CI = 1,it denotes additivity; if CI >1, it denotes antagonism; if CI <1,it indicates synergism.

Flow Cytometric Analysis of Cell Cycle Distributionand of Induction of Apoptosis

Apoptosis was measured using YO-PRO-1/propidiumiodide (Molecular Probes, Eugene, OR). YO-PRO-1 is agreen fluorescent DNA dye that is incorporated intoapoptotic cells, whereas propidium iodide is taken up onlyby necrotic cells. Cells stained with YO-PRO-1 andpropidium iodide were analyzed by flow cytometry. Cellswere cultured in T-75 flasks and, after 3 days, harvested bytrypsinization and plated in six-well plates. After overnightincubation, the cells were exposed to each drug or drugcombination sequence. At the end of each time point, bothcells and medium were collected, washed once in Ca/Mg-containing PBS, and aliquoted for apoptosis and cell cycleanalysis. For apoptosis measurements, cells were resus-pended in 1 mL of Ca/Mg-containing PBS followed by theaddition of 1 AL of each YO-PRO-1 and propidium iodide.Cells were placed on ice and analyzed by flow cytometrywithin 30 minutes. For cell cycle analysis, cells wereresuspended in Krishan’s stain and allowed to incubate at4jC for a minimum of 6 hours before analysis. Eachexperiment was done in triplicate.

Flow Cytometry of VEGFRsUntreated cells were cultured in T-75 flasks and, after

3 days, harvested by trypsinization. HCT-116 and HT29

cells were washed once with wash buffer (1% fetal bovineserum, 0.1% azide in PBS). Cells were stained for30 minutes on ice with one of the following primary anti-bodies: goat anti-human VEGFR1 (flt-1) at 5 Ag/mL, mouseanti-humanVEGFR2 (flk-1/KDR) mAb at 2.5 Ag/mL, andmouse anti-humanVEGFR3 biotinylated mouse mAb at2.5 Ag/mL (all from R&D Systems, Inc., Minneapolis, MN).Cells were then washed twice with wash buffer andincubated with the appropriate biotinylated secondaryantibody on ice. After 30 minutes, the cells were againwashed twice and all were incubated with streptavidin-phycoerythrin (BD PharMingen, San Diego, CA) at0.25 Ag/mL for 30 minutes on ice. Cells were washedtwice and then analyzed by flow cytometry. Each experi-ment was done in triplicate.

Flow Cytometry of EGFRPathway ProteinsCells were seeded into 100-mm tissue culture dishes 2

days before treatment and then exposed to drugs asdescribed above. After treatment, cells were harvestedand fixed with 0.5% formaldehyde for 10 minutes at37jC, washed once in PBS, permeabilized with 90%methanol, and placed on ice for 30 minutes. The sampleswere then stored at �20jC and processed the followingday. After one wash with 0.5% bovine serum albumin inPBS (wash buffer), the cells were incubated for 30minutes at room temperature with one of the followingantibodies: anti-EGFR rabbit polyclonal antibody at 1:200,anti-phosphorylated EGFR (Tyr1068) rabbit polyclonalantibody at 1:50, anti-ERK rabbit polyclonal antibody at1:25, anti-phosphorylated ERK (Thr202/Tyr204) rabbitpolyclonal antibody at 1:100, anti-phosphorylated AKT(Ser473) mouse monoclonal at 1:40, or anti-AKT polyclonalat 1:50 (all from Cell Signaling). Samples were thenwashed once with wash buffer and incubated with theappropriate FITC-conjugated secondary antibody dilutedin wash buffer for 30 minutes at room temperature. Cellswere then washed twice with wash buffer and analyzedby flow cytometry. Each experiment was done intriplicate.

ImmunoblottingCells were seeded into six-well plates 24 hours before

treatment with each drug or drug combination as describedabove. After treatment, cells were scraped into radio-immunoprecipitation assay buffer containing protease inhib-itors, EDTA, NaF, and sodium orthovanadate. The totalprotein in samples was determined using the bicinchoninicacid total protein assay reagent from Pierce (Rockford, IL).Total protein (30 Ag) per well was electrophoresed on a 7.5%,10%, or 12% SDS-PAGE and electrophoretically transferredto Immobilon membrane (Hybond-P, Amersham Biosciences,Buckinghamshire, United Kingdom). Membranes wereblocked for 1 hour at room temperature with 5% nonfat drymilk in TBS containing 0.1% Tween 20 before overnightincubation at 4jC with one of the following antibodies: anti-EGFR rabbit polyclonal antibody at 1:2,000, anti-phosphory-lated EGFR (Tyr1068) rabbit polyclonal antibody at 1:2,000,anti-p21 mouse mAb at 1:2,000, anti-cyclin D1 mouse mAbat 1:2,000, anti-phosphorylated ChK2 rabbit mAb at 1:2,000,

Molecular Cancer Therapeutics 1885




anti-Bax and anti-Bcl-xL rabbit polyclonal at 1:10,000, anti-cleaved poly(ADP-ribose) polymerase (Asp214) rabbit poly-clonal at 1:1,000, anti-survivin mouse mAb at 1:1,000, anti-insulin-like growth factor-I receptor (IGF-IR) polyclonalantibody at 1:1,000, or anti-phosphorylated IGF-IR (Tyr1131)polyclonal antibody at 1:1,000 (all from R&D Systems). Afterthe primary antibody, blots were washed thrice for 15 minutesin TBS containing 0.1% Tween 20 and were incubated with theappropriate secondary anti-rabbit IgG1 horseradish peroxi-dase–linked antibody at 1:2,000 (Amersham Biosciences) oranti-mouse IgG1 horseradish peroxidase–linked antibody at1:2,000 (Amersham Biosciences) for 1 hour at room temper-ature. After three additional washes, the blots were developedby an enhanced chemiluminescence detection system (ECL-plus, Amersham Biosciences). Each experiment was done intriplicate.

Evaluation of VEGF-A SecretionHCT-116 and HT29 cells were plated in 24-well plates

before the treatment with each drug or drug combination.The concentration of VEGF-A in the conditioned mediumwas measured using a commercially ELISA kit (R&DSystems) according to the manufacturer’s instructions.Assays were done in triplicate. Results were normalizedfor the number of producing cells and reported as pg ofVEGF per 3 � 105 cells per 72 hours.

Statistical AnalysisWe used a commercially available statistical program

(InStat 3, GraphPad, San Diego, CA) to do ANOVAparametric analysis for comparing the means of differenttreatments.

ResultsEGFRandVEGFRExpression by HCT-116 and HT29Before carrying out the cytotoxicity assay, we charac-

terized the EGFR and VEGFR expression of HCT-116 andHT29 cells. As depicted in Fig. 1A, the two colon cancercell lines express higher amounts of EGFR/phosphory-lated EGFR than the CEM cells (internal negativecontrol). Although the HCT-116 cells express a slightlyhigher level of EGFR/phosphorylated EGFR than HT29cells (1.75- and 2-fold higher, respectively), the propor-tions were similar. We also evaluated the expression ofthree VEGFR receptors (VEGFR1, VEGFR2, and VEGFR3)in the colon cell lines. Interestingly, both HT29 and HCT-116 showed expression of VEGFR1 and VEGFR3, whereasVEGFR2 expression was only detected in the humanumbilical vascular endothelial cells (internal positivecontrol; Fig. 1B). These results were confirmed byWestern blot (data not shown).

Sequence Dependence of Oxaliplatin and ZD6474Interaction on Proliferation of Colorectal Cancer Cells

To evaluate the antiproliferative effects of oxaliplatin orZD6474 alone and in combination using three differenttreatment sequences, we first did a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. Treatmentwith ZD6474 alone for 48 hours resulted in a dose-dependent inhibition of growth with an IC50 value of 16

Amol/L for both cell lines. Similarly, a dose-dependentinhibition of growth by oxaliplatin was observed with IC50

of 20 to 25 Amol/L and 7 to 14 Amol/L against HT29 andHCT-116 cells at 24 and 48 hours, respectively. We thenassessed combinations of ZD6474 and oxaliplatin in thethree different exposure sequences. As shown in Fig. 2, a24-hour exposure to oxaliplatin followed by a 48-hourexposure to ZD6474 resulted in clear synergy with a CIbetween 0.2 and 0.5 in both HT29 and HCT-116 cells. Thiswas significantly different (P < 0.001) to the reversesequence that was clearly antagonistic in both cell lines,with CI between 1.5 and 1.9. Likewise, concurrent exposurewas also antagonistic with CI between 1.3 and 1.6 (P <0.001, compared with synergistic sequence).

Cell Cycle Arrest by Oxaliplatin Is Modulated bySequence-Dependent Combinationwith ZD6474

To explore the mechanism of this sequence-dependentsynergy, we investigated the effects of oxaliplatin and

Figure 1. Flow cytometry for EGFR, phosphorylated EGFR, and VEGFRreceptor expressions in colon cell lines. HCT-116, HT29, CEM, and humanumbilical vascular endothelial cells were plated in T-75 flasks and, whenthey reached f70% confluence, harvested, fixed with formaldehyde, andpermeabilized with methanol as described in Materials and Methods. Thefollowing day, the cells were stained with EGFR, phosphorylated EGFR(pEGFR ; A), VEGFR1, VEGFR2, or VEGFR3 (B) polyclonal antibody andanalyzed by flow cytometry. CEM and acute lymphoblastic leukemia cellsare the negative control, whereas human umbilical vascular endothelialcells (HUVEC ) were used as a positive control. Results are meanfluorescent intensity (MFI ) change from no primary antibody control(i.e., secondary antibody only). Columns, mean of three independentexperiments; bars, SD.





ZD6474 in the three different exposure sequences on cellcycle distribution. As depicted in Fig. 3A, in HT29 cells thetreatment with oxaliplatin for 24 hours resulted in a G2-Mphase arrest (39%) compared with controls (10%; P < 0.001).This effect was increased with a longer (48 hours) exposureto oxaliplatin (58%). As reported with other EGFRantagonists (41), the treatment with ZD6474 induced asmall increase in G1 phase (70%) compared with controls(60%). Interestingly, the oxaliplatin-induced block in G2-Mwas maintained with the OXA-ZD6474 sequence (40%) butwas antagonized in the reverse sequence (10%) or withconcurrent exposure (22%). A similar, although lesspronounced, effect was obtained with HCT-116 cells(Fig. 3B). Several proteins, such as ChK2, p21, and cyclinD1, are involved in the regulation of the cell cycle (42, 43).We next evaluated the effects of the three sequences on the

expression of these proteins. The results depicted inFig. 4 reflect that the expression of phosphorylated ChK2in HT29 cells was markedly increased by treatment withoxaliplatin alone, although relatively unchanged byZD6474 alone. The oxaliplatin-induced phosphorylation ofChK2 was maintained with the synergistic sequence(OXA-ZD6474) but reduced by the concurrent exposure(oxaliplatin + ZD6474) and completely blocked by the reversesequence (ZD6474-OXA). Activation of ChK2 is known tolead to increased expression of cyclin-dependent kinaseinhibitor p21 expression (42). Figure 4 shows thatoxaliplatin induction of p21 was strongly potentiatedby the OXA-ZD6474 sequence compared with thereverse sequence. As also depicted in Fig. 4, cyclin D1expression was reduced by oxaliplatin and slightly byZD6474, consistent with the G1 arrest we observed. The

Figure 2. Effect of the combinationof ZD6474 and oxaliplatin on colon cellline. Approximately 3 � 103 HT29 cellsand 2 � 103 HCT-116 cells werecultured in 96-well plates. Cells weretreated with increasing concentrationof ZD6474 and/or oxaliplatin usingthree different sequences: oxaliplatinfor 24 h followed by ZD6474 for48 h (A and B), ZD6474 for 48 h fol-lowed by oxaliplatin for 24 h (C and D),and simultaneous exposure of drugsfor 48 h (E and F). CI values werecalculated according to the Chou andTalalay mathematical model for druginteractions using Calcusyn software.Columns, average of three independentexperiments; bars, SD.





synergistic OXA-ZD6474 sequence further reducedcyclin D1 expression, whereas the other two sequencesseemed to antagonize the oxaliplatin effect on cyclin D1expression.

Oxaliplatin-Induced Apoptosis of Colorectal CancerCells Is Potentiated by ZD6474

Next, we assessed the effect of these three sequences onapoptosis and apoptotic regulatory proteins. As expected,treatment with oxaliplatin for 24 or 48 hours resulted in anincrease in apoptosis compared with ZD6474 for 48 hours(Fig. 5). In the OXA-ZD6474 sequence, a significantinduction of apoptosis was observed, which was greaterthan either single-agent oxaliplatin or ZD6474 exposure aswell as the other two exposure sequences. Notably, theantagonistic ZD6474-OXA sequence inhibited the inductionof apoptosis observed with oxaliplatin at 24 hours. Toelucidate further the mechanism of these proapoptoticeffects by the synergistic sequence, we evaluated the

expression of Bcl-xL, Bax, poly(ADP-ribose) polymerase,and survivin in HT29 cells (Fig. 6). We hypothesized thatoxaliplatin would induce the expression of proapoptoticproteins because of DNA damage but at the same timeproduce a tumor survival response through increasing theexpression of antiapoptotic proteins. Further, we hypothe-sized that ZD6474 would modulate the oxaliplatin-inducedchanges in expression of prosurvival and proapoptoticproteins in a sequence-dependent manner. Consistent withour hypothesis, poly(ADP-ribose) polymerase cleavage wasstrikingly induced by the OXA-ZD6474 sequence comparedwith the other exposure sequences and single-agent expo-sure. The proapoptotic protein Bax was induced by single-agent oxaliplatin, and this induction was maintained inOXA-ZD6474 and oxaliplatin + ZD6474 sequences. Notably,there was a reduction in Bax expression by the ZD6474-OXAsequence. Interestingly, the antiapoptotic protein Bcl-xLwas induced by oxaliplatin alone, suggesting a prosurvivalresponse, whereas this induction was markedly abrogatedby the synergistic OXA-ZD6474 sequence and partiallymaintained by the other sequences. Interestingly, theantiapoptotic protein survivin was strongly inducedby single-agent treatment with oxaliplatin or ZD6474, andalthough all three sequences led to a reduction insurvivin expression compared with oxaliplatin alone, thereduction was most pronounced by the antagonistic ZD6474-OXA sequence. These results suggest that the sequence-dependent effects of the combination of oxaliplatin andZD6474 may at least be partially explained by the ability ofZD6474 to block prosurvival responses in HT29 cells thatoccur as a result of oxaliplatin exposure.

Oxaliplatin and ZD6474 Interaction Is EGFRDependent

Because oxaliplatin seems to induce a prosurvivalresponse in vitro , a setting in which the EGFR-based effectsof ZD6474 would be expected to predominate, we nextassessed the effect of the exposure sequences on EGFR

Figure 3. Effects of exposure sequences on cell cycle distribution.HT29 and HCT-116 cells were treated with ZD6474 alone for 48 h (13Amol/L for both cell lines), oxaliplatin alone for 24 h (20 and 15 Amol/L,respectively) and 48 h (10 and 5 Amol/L, respectively), and variouscombinations of ZD6474 and oxaliplatin [ZD6474 (48 h; 13 Amol/L forboth cell lines) plus oxaliplatin (48 h; 10 Amol/L for HT29 and 5 Amol/L forHCT-116), oxaliplatin (24 h; 20 Amol/L for HT29 and 15 Amol/L for HCT-116) followed by ZD6474 (48 h; 13 Amol/L for both cell lines), andZD6474 (48 h; 13 Amol/L for both cell lines) followed by oxaliplatin (24 h;20 Amol/L for HT29 and 15 Amol/L for HCT-116)]. After the treatment, thecells were harvested and stained for DNA content with propidium iodidefollowed by flow cytometric analysis. Columns, mean of three indepen-dent experiments; bars, SD. ND, untreated; OXA, oxaliplatin;ZD6474M+OXA, ZD6474 (48 h) plus oxaliplatin (48 h); OXA�ZD6474,oxaliplatin (24 h) followed by ZD6474 (48 h); ZD6474�OXA, ZD6474(48 h) followed by oxaliplatin (24 h). OXA�ZD6474 sequence versusZD6474 + OXA sequence (P < 0.01 for HT29 and P < 0.05 forHCT-116). OXA�ZD6474 sequence versus ZD6474�OXA sequence(P < 0.001 for HT29 and P < 0.001 for HCT-116).

Figure 4. Effects of exposure sequences on the expression of p21 andcyclin D1. HT29 cells were treated as described previously, and after thetreatment, total cell proteins were fractionated through SDS-PAGE,transferred to nitrocellulose filters, and incubated with the appropriateantibodies as described in Materials and Methods. The experiment wasdone in triplicate. Actin protein was used as a protein loading control.





expression and activation as well as on the downstreameffectors AKT, phosphorylated AKT, ERK, and phosphor-ylated ERK. As a consequence of prosurvival pathwayactivation, oxaliplatin treatment for 24 and 48 hoursmarkedly increased the expression of total and phosphor-ylated EGFR in both cell lines as depicted in Fig. 7. In HT29cells, this induction was blocked when oxaliplatin wascombined with ZD6474 in all three sequences but with thegreater reduction of EGFR activity being observed with thesynergistic OXA-ZD6474 sequence. By contrast, in HCT-116cells, the synergistic and concurrent treatments completelyblocked the oxaliplatin-induced expression of total andactivated EGFR, whereas the antagonistic sequence waswithout effect (Fig. 7). As expected, oxaliplatin exposuredramatically increased the expression of both total andphosphorylated ERK and AKT in HT29 and HCT-116 celllines (Fig. 7). Subsequent but not prior treatment withZD6474 significantly reduced the oxaliplatin-induced in-crease in the expression levels of total and phosphorylatedERK and AKT (Fig. 7). These data support the hypothesisthat oxaliplatin promotes an EGFR-dependent cell survivalresponse that is blocked by subsequent exposure toZD6474, which may be responsible for the observedsynergistic activity of this sequence. To support thishypothesis and to isolate further the EGFR dependence ofthese effects, we used two approaches. First, we showed inboth HT29 and HCT-116 cells that similar sequence-dependent effects on cell cycle and apoptosis were obtainedif oxaliplatin treatment is followed by C225 (an antibody-based inhibitor of EGFR) or U0126 (an inhibitor of MEK1;Fig. 8). Second, we made use of HT29 cells (R4F) that stablyexpressing a constitutively active form of MEK1 (44). TheR4F cells when treated with the OXA-ZD6474 sequenceboth failed to undergo apoptosis (Fig. 9A) or a G2-M arrest(Fig. 9B). Together, these data suggest that the block ofoxaliplatin-induced EGFR pathway by ZD6474 contributesfunctionally to the marked synergism exhibited by theOXA-ZD6474 treatment.

Oxaliplatin and ZD6474 Interaction Involves OtherProsurvival Pathways

IGF-IR-dependent signaling is another important prosur-vival pathway in colorectal cancer, and activation of thisreceptor leads to the production of angiogenic factors,including members of the VEGF family (45). We thereforeelucidated the involvement of the IGF-IR pathway in thesequence-dependent interaction of oxaliplatin and ZD6474.As depicted in Fig. 10, oxaliplatin clearly reduced theexpression but not the activation of IGF-IR. Interestingly,ZD6474 alone reduced the expression of both total andphosphorylated IGF-IR (Fig. 10). We observed thatalthough the oxaliplatin-induced reduction of total andphosphorylated IGF-IR expression was maintained in bothantagonistic sequences with ZD6474, it was markedlypotentiated by the synergistic sequence (Fig. 10). Becauseactivation of IGF-IR leads to the production of angiogenicfactors, including VEGF-A, we measured VEGF-A secretionby ELISA from both HT29 and HCT-116 cells followingtreatment with single-agent oxaliplatin or ZD6474 and in thethree sequences (Fig. 11). Oxaliplatin treatment alone inhi-bited VEGF-A secretion at 24 and 48 hours particularly inHCT-116 cells (P < 0.05 at 24 hours and P < 0.001 at 48 hoursin HT29; P < 0.001 at 24 and 48 hours in HCT-116) as didZD6474, which was more prominent in HT29 (P < 0.001).The block in VEGF-A secretion by single agents was clearlypotentiated when the drugs were administrated in the syner-gistic sequence and produced an almost complete suppres-sion (>98%) of VEGF-A secretion in both cell lines. By contrast,the other two sequences reduced VEGF secretion in a mannersimilar to single-agent oxaliplatin and ZD6474 (Fig. 11).

DiscussionThe possibility of combining cytotoxic drugs with molec-ularly targeting agents that specifically interfere with key

Figure 5. Effect of exposure sequences on apoptosis. HT29 and HCT-116 cells were treated as described previously. After the treatment, thecells were harvested and stained with YO-PRO-1 and propidium iodide andanalyzed by flow cytometry. Columns, average of triplicate determina-tions; bars, SD. Data are ratio of percentage of apoptotic cells topercentage of live cells. OXA�ZD6474 sequence versus ZD6474 + OXAsequence (P < 0.001 for both cell lines). OXA�5ZD6474 sequenceversus ZD6474�OXA sequence (P < 0.001 for both cell lines).

Figure 6. Effect of exposure sequence on expression of apoptoticregulatory protein. Total whole-cell proteins were extracted from HT29cells treated as described previously. Equal amounts of cellular protein (30Ag/lane) were loaded onto a SDS-polyacrylamide gel, transferred tonitrocellulose filters, and incubated with anti-Bcl-xL, anti-Bax, anti-cleavedpoly(ADP-ribose) polymerase (PARP ), and anti-survivin antibodies. Actinprotein was blotted as a control.





pathways controlling proliferation, invasion, and/or meta-static spreading has generated wide interest. Activation ofthe EGFR signaling pathway is of known importance in thedevelopment and progression of human epithelial cancers.For this reason, several approaches to the inhibition ofEGFR have been developed and many preclinical andclinical trials have been conducted with the aim of testingthe antitumor activity of selective EGFR antagonists in

combination with cytotoxic therapies. This class of targetedinhibitors has all shown successful preclinical activity,which, unfortunately, has been tempered by limited clinicalsuccess. In fact, several large phase III clinical trials of theEGFR tyrosine kinase inhibitors gefitinib (Iressa) orerlotinib (Tarceva) have failed to show any advantage ofcombining these agents with chemotherapy over single-agent therapy (46–48). A consideration that has arisen from

Figure 7. Effect of exposure sequences on EGFR pathway. HT29 and HCT-116 cells were treated as described previously, harvested and fixed withformaldehyde, and permeabilized with methanol. The following day, the cells were stained with EGFR, phosphorylated EGFR, ERK, phosphorylated ERK(pERK ), AKT, and phosphorylated AKT (pAKT ) polyclonal antibodies and analyzed by flow cytometry. Results are mean fluorescence intensity changefrom no primary antibody control (i.e., secondary antibody only). Columns, average of three independent experiments; bars, SD.





the ‘‘negative’’ results of these clinical trials is that theempirical combination of biological agents and chemother-apy may not yield positive results and that a betterunderstanding of the mechanistic effects of these agentsin combination is required. Because the focus of our groupis on the development of targeted agents against gastroin-testinal malignancies, we decided to evaluate in vitro theoptimum way to combine oxaliplatin, a DNA-damagingagent, and ZD6474, an inhibitor of VEGFR and EGFRsignaling, against two different human colon cell lines,HT29 and HCT-116. We first characterized the expressionof EGFR and VEGFR on these cell lines. As previouslyreported, we found that both cell lines significantly expressEGFR and phosphorylated EGFR (41). Although recentstudies have suggested that VEGFRs are expressed not onlyon endothelial cells but also on tumor cells, includingbreast, colon, and lung adenocarcinoma, we found that theHT29 and HCT-116 cell lines were negative for VEGFR2expression (49–51). Interestingly, they both showed ex-pression of VEGFR1 and VEGFR3. The functional signifi-cance of the expression of these receptors on colorectalcancer is currently under investigation in our group, but ithas been shown by others that VEGFR1 on colorectal canceris coupled to cell invasion and migration but not toproliferation (49). We next evaluated the antiproliferativeeffect of oxaliplatin and ZD6474 using three differentschedules. Our results provide evidence of a strongsynergistic antiproliferative effect in both cell lines betweenthe two drugs only when oxaliplatin treatment precededZD6474, whereas the reverse sequence and concurrenttreatment were clearly antagonistic. Similar to our findings,Xu et al. have shown a sequence-dependent antiprolifer-

ative synergy between gefitinib and both oxaliplatin andirinotecan, a topoisomerase I inhibitor (41, 52). In addition,Morelli et al. (53) have shown in an in vitro model of anEGFR-positive human esophageal cancer epithelial cell linewith a functional EGFR-dependent autocrine growthpathway that the potentiation of the antiproliferativeactivity of selected cytotoxic drugs, such as three platinumderivatives (cisplatin, carboplatin, and oxaliplatin) andtaxanes (paclitaxel and docetaxel), by combined treatmentwith three EGFR inhibitors (gefitinib, cetuximab, orZD6474) is schedule and sequence dependent.

To further investigate the mechanisms underlying thesynergistic interaction of ZD6474 with oxaliplatin, weexamined the effect of the three sequences on cell cyclephase progression and regulatory proteins. As previouslyreported, treatment of cells with oxaliplatin, a DNA-damaging agent, induced a G2-M arrest allowing forDNA repair before continued cell cycle progression (41).We observed that the oxaliplatin-induced G2-M phaseblock was maintained in the OXA-ZD6474 sequence,whereas antagonized in both reverse and concurrenttreatments. The G2-M phase is regulated by variouscheckpoint proteins, including ChK2 and p21, a cyclin-dependent kinase inhibitor. Agents that induce DNAdamage activate ChK2 by phosphorylation at Thr68, whichleads to increased expression of p21 and ultimately a blockin cell cycle progression (42). Consistent with this, weshowed that oxaliplatin induced ChK2 phosphorylationand p21 expression and this was either maintained orpotentiated in the OXA-ZD6474 sequence but antagonizedin the other two sequences when ZD6474 precededoxaliplatin or when both agents were given concurrently.

Figure 8. Effect of C225 andUO126 on cell cycle distribution.HT29 and HCT-116 cells were platedin six-well plates and treated withthree different sequences: sequentialoxaliplatin (20 and 15 Amol/L, re-spectively) 1 d followed by ZD6474(13 Amol/L for both cell lines) 2 d;oxaliplatin (20 and 15 Amol/L, re-spectively) 1 d followed by C225(20 Ag/mL) or UO126 (10 Amol/L) for2 d. After the treatment, cells wereharvested and samples were stainedfor cell cycle and apoptosis anal-ysis as indicated in Materials andMethods.





As reported with other EGFR signaling inhibitors, wefound that ZD6474 alone increased the proportion of cellsin G1, associated with a small decrease in cyclin D1expression (43). These observations may explain, in part,the synergy and the antagonism observed. Induction of G1

arrest by pretreatment with ZD6474 before oxaliplatin maylimit the DNA damage induced by oxaliplatin. Conversely,treatment with ZD6474 following oxaliplatin would beexpected to potentiate the G2-M phase arrest caused byoxaliplatin and further limit the ability of the cells to repairdamaged DNA and progression through the cell cycle. Insupport of this, we found that the synergistic sequence ledto a clear potentiation of oxaliplatin-induced apoptosis,suggesting irreparable DNA damage, whereas the othertwo sequences, particularly the reverse sequence, antago-nized this activity. The apoptosis data were then confirmedby monitoring the expression of proteins involved inmediating cell death. Thus, poly(ADP-ribose) polymerasecleavage was strongly induced only in the OXA-ZD6474sequence, whereas oxaliplatin-induced expression of theproapoptotic protein Bax was maintained by the synergisticsequence but completely blocked by the reverse sequences.A key observation in the present study was that oxaliplatin

also induced a prosurvival response as reflected by anincrease in the expression of antiapoptotic proteins Bcl-xLand survivin. The oxaliplatin-induced expression of Bcl-xLwas completely blocked by OXA-ZD6474 sequence. Inter-estingly, although oxaliplatin-induced survivin expressionwas reduced by all three sequences, the largest reductionwas observed with ZD6474-OXA sequence. Collectively,these data suggest that treatment with ZD6474 followingoxaliplatin alters the balance of proapoptotic and prosur-vival proteins, ultimately leading to potentiation ofoxaliplatin-induced apoptosis in human colorectal cancercells in vitro . In these in vitro studies, we hypothesized thatthe prosurvival response induced by oxaliplatin wasmediated by EGFR pathway. Indeed, other studies haveshown activation of this pathway in response to treatmentwith DNA-damaging agents, such as doxorubicin andcisplatin, but not with microtubule-binding agents, suchas paclitaxel and vinblastine (54). Consistent with this,oxaliplatin produced a marked increase in both theexpression and the activation of EGFR and the downstreameffectors AKT and ERK that was antagonized by subse-quent treatment with ZD6474. This was then substantiatedby the finding that similar synergistic effects were obtainedwhen the cells were exposed to oxaliplatin followed byC225 (an antibody inhibitor of EGFR) or UO126 (aninhibitor of MEK1). In addition, maintenance of the G2-Mblock and potentiation of apoptosis by the OXA-ZD6474sequence were completely blocked in HT29-MEK1 (R4F)cells expressing a constitutively active form of MEK (44).Together, these data strongly suggest that the antitumorsynergistic interaction between oxaliplatin and ZD6474depends, in large part, on reversal of ZD6474-mediatedantagonism of the activation of the EGFR prosurvivalpathway by oxaliplatin. EGFR activation is likely servingas a cell survival response in colorectal cancer cells exposedto DNA-damaging chemotherapeutic agents, such asoxaliplatin.

By examining IGF-IR expression and activation, wefurther assessed stimulation of prosurvival pathways in

Figure 9. Effect of enforced activation of ERK on OXA-ZD6474sequence HT29-MEK1 (R4F), which express constitutively active MEK1,and HT29 cells were exposed to 25 Amol/L oxaliplatin (24 h) followed by8 or 13 Amol/L ZD6474 (48 h), respectively. After the treatment, the cellswere harvested and stained for apoptosis and cell cycle analysis.Columns, average of triplicate determinations; bars, SD. Data are ratioof percentage of apoptotic cells to percentage of live cells (A) or as apercentage of cells in G2-M phase (B). Asterisks, significantly less thanvalues obtained for HT29 cells.

Figure 10. Effect of exposure sequences on expression and activationof EGFR and IGF-IR. HT29 cells were treated with oxaliplatin or ZD6474and different combination of these drugs as described previously. After thetreatment, proteins were extracted and run on SDS-PAGE and incubatedwith the appropriate antibodies as described in Materials and Methods.Immunoreactive proteins were visualized by enhanced chemilumines-cence. The experiment was done in triplicate. Actin protein was blotted asa control.





colorectal cancer cells exposed to oxaliplatin. A variety oftumors, including colon, show altered expression of IGF-Iand its receptor, IGF-IR. Recently, it has been shown thatIGF-I, acting through its receptor, plays an important rolein multiple mechanisms that mediate human colon cancergrowth, including regulation of VEGF-A secretion andangiogenesis (45). We showed that IGF-IR expression, butnot activation, was down-regulated by oxaliplatin treat-ment. This effect was strongly potentiated only whenoxaliplatin preceded ZD6474. Moreover, the modulation ofIGF-IR expression and activation by the synergisticsequence paralleled regulation of VEGF-A secretion. Themarked block of VEGF-A production by the synergisticOXA-ZD6474 sequence is particularly interesting as thiswould be expected to result in antiangiogenesis effectsin vivo through modulation of both tumor and endothelialcell compartments. Experiments to this effect are ongoing.There is an urgent need in cancer therapy to define theoptimal schedule of administration of chemotherapy withbiologically targeted therapeutic agents. There may bedistinct roles for the use of biological agents in colorectalcancer, one which exploits strategies that potentiate thecytotoxic effects of chemotherapy and another that max-imizes independent biological effects. The doses andschedules with these two approaches may be quitedifferent. In conclusion, our study shows that ZD6474possesses antiproliferative antitumor activity in vitro thatcan act in a sequence-dependent manner with oxaliplatin inhuman colon cancer cell lines. These results, if confirmedin vivo , would suggest that rational combination strategiescould be investigated in the clinic using EGFR signalinginhibitors to potentiate the effects of oxaliplatin-basedchemotherapy by modulating prosurvival responses andenhancing proapoptotic pathways.

Acknowledgments

We thank Dr. Anderson Ryan (AstraZeneca Pharmaceuticals) for thegenerous gift of ZD6474 and for the helpful discussion and Drs. Pamela L.Rice and Dennis J. Ahnen for the generous gift of HT29-MEK1 (R4F) cellline.

References

1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. Cancer JClin 2005;1:10–30.

2. Douillard JY, Cunningham D, Roth AD, et al. Irinotecan combined withfluorouracil compared with fluorouracil alone as first-line treatment formetastatic colorectal cancer: a multicentre randomised trial. Lancet 2000;355:1041–7.

3. De Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracilwith or without oxaliplatin as first-line treatment in advanced colorectalcancer. J Clin Oncol 2000;16:2938–47.

4. Grothey A, Deschler B, Kroening H, et al. Phase III study of bolus5-fluorouracil (5-FU)/folinic acid (FA) (Mayo) versus weekly high dose24 h 5-FU infusion/FA + oxaliplatin (OXA) in advanced colorectal cancer(CRC). Proc Am Soc Clin Oncol 2002;21:129.

5. Messa C, Russo F, Caruso MG, Di Leo A. EGF, TGF-a, and EGF-R inhuman colorectal adenocarcinoma. Acta Oncol 1998;37:285–9.

6. Porebska I, Harlozinska A, Bojarowski T. Expression of the tyrosinekinase activity growth factor receptors (EGFR, ERB B2, ERB B3) incolorectal adenocarcinomas and adenomas. Tumour Biol 2000;21:105–15.

7. Salomon DS, Brandt R, Ciardiello F, Normanno N. Epidermal growthfactor-related peptides and their receptors in human malignancies. Crit RevOncol Hematol 1995;19:183–232.

8. Goldstein NS, Armin M. Epidermal growth factor receptor immunohis-tochemical reactivity in patients with American Joint Committee onCancer Stage IV colon adenocarcinoma: implications for a standardizedscoring system. Cancer 2001;92:1331–46.

9. Mayer A, Takimoto M, Fritz E, Schellander G, Kofler K, Ludwig H. Theprognostic significance of proliferating cell nuclear antigen, epidermalgrowth factor receptor, and mdr gene expression in colorectal cancer.Cancer 1993;71:2454–60.

10. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapyand cetuximab plus irinotecan in irinotecan-refractory metastatic colorec-tal cancer. N Engl J Med 2004;4:337–45.

11. Andre T, Kotelevets L, Vaillant JC, et al. Vegf, Vegf-B, Vegf-C andtheir receptors KDR, FLT-1, and FLT-4 during the neoplastic progression ofhuman colonic mucosa. Int J Cancer 2000;2:174–81.

12. Ellis LM, Liu W. Vascular endothelial growth factor (VEGF) expressionand alternate splicing in non-metastatic and metastatic human coloncancer cell lines. Proc Am Assoc Cancer Res 1995;36:88.

13. Wong MP, Cheung N, Yuen ST, Leung SY, Chung LP. Vascularendothelial growth factor is up-regulated in the early pre-malignant stageof colorectal tumour progression. Int J Cancer 1999;6:845–50.

14. Lee JC, Chow NH, Wang ST, Huang SM. Prognostic value of vascularendothelial growth factor expression in colorectal cancer patients. Eur JCancer 2000;6:748–53.

15. Shiraishi A, Ishihwata T, Shoji T, Asano G. Expression of PCNA, basicfibroblast growth factor, FGF-receptor, and vascular endothelial growthfactor in adenomas and carcinomas of human colon. Acta HistochemCytochem 1995;28:21–9.

16. Takahashi Y, Kitadai Y, Bucana CD, Cleary KR, Ellis LM. Expression ofvascular endothelial growth factor and its receptor, KDR, correlates withvascularity, metastasis, and proliferation of human colon cancer. CancerRes 1995;18:3964–8.

17. Hicklin DJ, Ellis LM. Role of the vascular endothelial growthfactor pathway in tumor growth and angiogenesis. J Clin Oncol 2005;5:1011–27.

18. Amaya H, Tanigawa N, Lu C, et al. Association of vascular endothelialgrowth factor expression with tumor angiogenesis, survival, and thymi-dine phosphorylase/platelet-derived endothelial cell growth factor expres-sion in human colorectal cancer. Cancer Lett 1997;119:227–35.

19. Tokunaga T, Oshika Y, Abe Y, et al. Vascular endothelial growthfactor (VEGF) mRNA isoform expression pattern is correlated with livermetastasis and poor prognosis in colon cancer. Br J Cancer 1998;77:998–1002.

Figure 11. Effect of exposure sequences on VEGF secretion. HT29 andHCT-116 cells were treated with ZD6474 alone for 48 h (13 Amol/L forboth cell lines), oxaliplatin alone for 24 h (20 and 15 Amol/L, respectively)and 48 h (10 and 5 Amol/L, respectively), and various combinations ofZD6474 and oxaliplatin. After the treatment, the medium was collectedand assay was formed according to the manufacturer’s instructions.Columns, average of two independent experiments done in duplicate;bars, SD. OXA�ZD6474 sequence versus ZD6474 + OXA sequence(P < 0.001 for HT29 and P < 0.01 for HCT-116). OXA�ZD6474 sequenceversus ZD6474�OXA sequence (P < 0.001 for both cell lines).





20. Ishigami SI, Arii S, Furutani M, et al. Predictive value of vascularendothelial growth factor (VEGF) in metastasis and prognosis of humancolorectal cancer. Br J Cancer 1998;78:1379–84.

21. Cascinu S, Staccioli MP, Gasparini G, et al. Expression of vascularendothelial growth factor can predict event-free survival in stage II coloncancer. Clin Cancer Res 2000;6:2803–7.

22. Lee JC, Chow NH, Wang ST, et al. Prognostic value of vascularendothelial growth factor expression in colorectal cancer patients. Eur JCancer 2000;36:748–53.

23. Takahashi Y, Tucker SL, Kitadai Y. Vessel counts and expression ofvascular endothelial growth factor as prognostic factors in node-negativecolon cancer. Arch Surg 1997;5:541–6.

24. Rosen LS. Inhibitors of the vascular endothelial growth factorreceptor. Hematol Oncol Clin North Am 2002;5:1173–87.

25. Presta LG, Chen H, O’Connor SJ. Humanization of an antivascularendothelial growth factor monoclonal antibody for the therapy of solidtumors and other disorders. Cancer Res 1997;57:4593–9.

26. Fong TA, Shawver LK, Sun L, et al. SU5416 is a potent and selectiveinhibitor of the vascular endothelial growth factor receptor (Flk-1/KDR)that inhibits tyrosine kinase catalysis, tumor vascularization, and growthof multiple tumor types. Cancer Res 1999;59:99–106.

27. Wedge SR, Ogilvie DJ, Dukes M. ZD4190: an orally active inhibitor ofvascular endothelial growth factor signaling with broad-spectrum anti-tumor efficacy. Cancer Res 2000;60:970–5.

28. Prewett M, Huber J, Li Y, et al. Antivascular endothelial growth factorreceptor (fetal liver kinase 1) monoclonal antibody inhibits tumorangiogenesis and growth of several mouse and human tumors. CancerRes 1999;59:5209–18.

29. Brekken RA, Overholser JP, Stastny VA, Waltenberg J, Minna JD,Thorpe PE. Selective inhibition of vascular endothelial growth factor(VEGF) receptor 2 (KDR/Flk-1) activity by a monoclonal anti-VEGFantibody blocks tumor growth in mice. Cancer Res 2000;60:5117–24.

30. Kunkel P, Ulbricht U, Bohlen P, et al. Inhibition of glioma angiogenesisand growth in vivo by systemic treatment with a monoclonal antibodyagainst vascular endothelial growth factor receptor-2. Cancer Res 2001;61:6624–8.

31. Inoue K, Slaton JW, Davis D, et al. Treatment of human metastatictransitional cell carcinoma of the bladder in a murine model with theantivascular endothelial growth factor receptor monoclonal antibodyDC101 and paclitaxel. Clin Cancer Res 2000;6:2635–43.

32. Schlaeppi JM, Wood JM. Targeting vascular endothelial growthfactor (VEGF) for anticancer therapy, by anti-VEGF neutralizing monoclo-nal antibodies and VEGF receptor tyrosine-kinase inhibitors. CancerMetastasis Rev 1999;18:473–81.

33. Cherrington JM, Strawn LM, Shawver LK. New paradigms for thetreatment of cancer: the role of anti-angiogenesis agents. Adv Cancer Res2000;79:1–38.

34. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plusirinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. NEngl J Med 2004;23:2335–42.

35. Wedge SR, Ogilvie DJ, Dukes M, et al. ZD6474 inhibits vascularendothelial growth factor signaling, angiogenesis, and tumor growthfollowing oral administration. Cancer Res 2002;62:4645–55.

36. Ciardiello F, Caputo R, Damiano V, et al. Antitumor effects ofZD6474, a small molecule vascular endothelial growth factor receptortyrosine kinase inhibitor, with additional activity against epidermal growthfactor receptor tyrosine kinase. Clin Cancer Res 2003;9:1546–56.

37. Carlomagno F, Vitagliano D, Guida T, et al. ZD6474, an orally

available inhibitor of KDR tyrosine kinase activity, efficiently blocksoncogenic RET kinases. Cancer Res 2002;62:7284–90.

38. Holden SN, Eckhardt SG, Basser R, et al. Clinical evaluation ofZD6474, an orally active inhibitor of VEGF and EGF receptor signaling, inpatients with solid, malignant tumors. Ann Oncol 2005;16:1931–9.

39. Heymach JV, Johnson BE, Rowbottom JA, et al. A randomized,placebo-controlled phase II trial of ZD6474 plus docetaxel, in patients withNSCLC. Proc Am Soc Clin Oncol 2005;24:3023.

40. JohnsonB,MaP,WestH, et al. Preliminary phase II safety evaluation ofZD6474 in combination with carboplatin and paclitaxel, as first linetreatment in patients with NSCLC. Proc Am Soc Clin Oncol 2005;74:7102.

41. Xu JM, Azzariti A, Colucci G. The effect of gefitinib (Iressa, ZD1839)in combination with oxaliplatin is schedule-dependent in colon cancer celllines. Cancer Chemother Pharmacol 2003;6:442–8.

42. Chen CR, Wang W, Rogoff HA, Xiaotong Li, Mang W, Chiang Li. Dualinduction of apoptosis and senescence in cancer cells by ChK2 activation:checkpoint activation as a strategy against cancer. Cancer Res 2005;14:6017–21.

43. Kalish LH, Kwong RA, Cole IE, Gallagher RM, Sutherland RL,Musgrove EA. Deregulated cyclin D1 expression is associated withdecreased efficacy of the selective epidermal growth factor receptortyrosine kinase inhibitor gefitinib in head and neck squamous cellcarcinoma cell lines. Clin Cancer Res 2004;22:7764–74.

44. Rice PL, Beard KS, Driggers LJ, Ahnen DJ. Inhibition of extracel-lular-signal regulated kinases 1/2 is required for apoptosis of humancolon cancer cells in vitro by sulindac metabolites. Cancer Res 2004;22:8148–51.

45. Reinmuth N, Liu W, Fan F, et al. Blockade of insulin-like growth factorI receptor function inhibits growth and angiogenesis of colon cancer. ClinCancer Res 2002;10:3259–69.

46. Herbst RS, Giaccone G, Schiller JH, et al. Gefitinib in combinationwith paclitaxel and carboplatin in advanced non-small-cell lung cancer: aphase III trial-INTACT 2. J Clin Oncol 2004;22:785–94.

47. Giaccone G, Herbst RS, Manegold C, et al. Gefitinib in combinationwith gemcitabine and cisplatin in advanced non-small-cell lung cancer: aphase III trial—INTACT 1. J Clin Oncol 2004;22:777–84.

48. Herbst RS, Prager D, Hermann R, et al. TRIBUTE—a phase III trial oferlotinib HCl (OSI-774) combined with carboplatin and paclitaxel (CP)chemotherapy in advanced non-small cell lung cancer (NSCLC). Proc AmSoc Clin Oncol 2004;23:7011.

49. Fan F, Wey JS, McCarty MF, et al. Expression and function ofvascular endothelial growth factor receptor-1 on human colorectal cancercells. Oncogene 2005;16:2647–53.

50. Ryden L, Linderholm B, Nielsen NH, Emdin S, Jonsson PE, LandbergG. Tumor specific VEGF-A and VEGFR2/KDR protein are co-expressed inbreast cancer. Breast Cancer Res Treat 2003;3:147–54.

51. Tanno S, Ohsaki Y, Nakanishi K, Toyoshima E, Kikuchi K. Humansmall cell lung cancer cells express functional VEGF receptors, VEGFR-2and VEGFR-3. Lung Cancer 2004;1:11–9.

52. Azzariti A, Xu JM, Porcelli L, Paradiso A. The schedule-dependentenhanced cytotoxic activity of 7-ethyl-10-hydroxy-camptothecin (SN-38)in combination with Gefitinib (Iressa, ZD1839). Biochem Pharmacol 2004;1:135–44.

53. Morelli MP, Cascone T, Troiani T, et al. Sequence-dependentantiproliferative effects of cytotoxic drugs and epidermal growth factorreceptor inhibitors. Ann Oncol 2005;16:61–8.

54. Benhar M, Engelberg D, Levitzki A. Cisplatin-induced activation of theEGF receptor. Oncogene 2002;57:8723–31.





2006;5:1883-1894. Mol Cancer Ther Teresa Troiani, Owen Lockerbie, Mark Morrow, et al. and EGFR tyrosine kinasescombination with ZD6474 (Zactima), an inhibitor of VEGFRgrowth and of prosurvival pathways by oxaliplatin in Sequence-dependent inhibition of human colon cancer cell

Updated version

http://mct.aacrjournals.org/content/5/7/1883

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/5/7/1883.full#ref-list-1

This article cites 40 articles, 13 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/5/7/1883.full#related-urls

This article has been cited by 9 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. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://mct.aacrjournals.org/content/5/7/1883To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/content/5/7/1883.full#ref-list-1

http://mct.aacrjournals.org/content/5/7/1883.full#related-urls

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

mailto:[email protected]



sequence-dependent inhibition of human colon cancer cell...

Documents