PrognosticSignificanceofTissueFactorinPancreatic ... · vival (hazard ratio, 2.014; P = 0.0076). Moreover, patients withTF-negative tumors had a sig-nificantly better prognosis even

Prognostic Significance ofTissue Factor in PancreaticDuctal AdenocarcinomaNobuhiro Nitori,1,5 Yoshinori Ino,1 Yukihiro Nakanishi,1 Tesshi Yamada,2 Kazufumi Honda,2

Kazuyoshi Yanagihara,3 Tomoo Kosuge,4 Yae Kanai,1 Masaki Kitajima,5 and Setsuo Hirohashi1

Abstract Tissue factor (TF) is a transmembrane glycoprotein that plays roles in the blood coagulationand intracellular signaling pathways, and has also been suggested to modulate the biologicalbehavior of cancer cells. In order to examine the clinicopathologic significance of TF expres-sion in pancreatic ductal adenocarcinoma, TF expression was determined by immunohisto-chemistry using a newly raised anti-TF monoclonal antibody in 113 patients who hadundergone surgical resection of pancreatic ductal adenocarcinoma. According to the incidenceof tumor cell immunopositivity, patients were divided into ‘‘negative TF’’ (0%), ‘‘weak TF’’(<25%), or ‘‘high TF’’ (25% or more) groups, which accounted for 11.6% (n = 13), 44.2%(n = 50), and 44.2% (n = 50) of the total, respectively. Increased TF expression was correlat-ed with the extent of the primary tumor (P = 0.0043), lymph node metastasis (P = 0.0043),lymphatic distant metastasis (P = 0.0039), advanced tumor-node-metastasis stage (P =0.0002), and high tumor grade (P = 0.0164). Multivariate analysis using the Cox proportionalhazards model showed that high TF expression was an independent negative predictor for sur-vival (hazard ratio, 2.014; P = 0.0076). Moreover, patients with TF-negative tumors had a sig-nificantly better prognosis even if lymph node metastasis was present (P < 0.0001). We alsoshowed that TF knockdown by RNA interference suppressed the invasiveness of a pancreaticadenocarcinoma cell line in vitro. These results indicate that TF expression may contribute tothe aggressiveness of pancreatic ductal adenocarcinoma by stimulating tumor invasiveness,and that evaluation of the primary tumor for TF expression may identify patients with a poorprognosis.

Tissue factor (TF) is a transmembrane glycoprotein thatfunctions as a cellular receptor for coagulation factor VII(FVII) and modulates it to produce the activated form, FVIIa.The TF/FVIIa complex is regarded as the initiator of theextrinsic blood coagulation cascade, which ultimately leads tothe generation of thrombin (1). In normal human tissues, TFis expressed only in extravascular cells, including the vascularadventitia and organ capsules (2). Based on this cellulardistribution, under physiologic conditions, TF is thought toact mainly as a hemostatic barrier to prevent blood loss. Inaddition to its role as a hemostatic initiator, the binding of

FVIIa with TF has been suggested to be involved inintracellular signaling mechanisms (3), such as the mitogen-activated protein kinase pathway (4) and the Src familymember/PI3K/Rac-dependent signaling pathway (5), at leastin some cell types.

TF is also involved in many pathophysiologic conditions,such as inflammation, atherosclerosis, and malignancies.With regard to malignancies, it has been well recognizedthat patients with malignant diseases are predisposed tohypercoagulation since Trousseau (6) first reported theincreased frequency of thrombosis in patients with gastroin-testinal cancers, and this hypercoagulable state is associatedwith TF (7). Immunohistochemical analysis has revealed thatTF is expressed in a wide variety of malignancies (8).Metastatic melanoma cells express higher levels of TF thannonmetastatic cells (9), and a metastatic rectal carcinomasubline showed enhanced TF expression in comparison to itsparental line (10). Transfection of TF promoted themetastasis of melanoma in a mouse model (11), andenhanced primary tumor growth in a pancreatic adenocarci-noma cell line (12). Therefore, TF not only contributes to thedevelopment of a hypercoagulable state in cancer patientsbut also modulates the biological behavior of cancer cells.

Pancreatic adenocarcinoma is one of the most clinicallyaggressive malignancies; indeed, the 3-year survival rate aftersurgical resection of the primary tumor has been reported asonly 17% (13). Therefore, identification of molecules that

www.aacrjournals.org Clin Cancer Res 2005;11(7) April 1, 20052531

Authors’Affiliations: 1Pathology, 2Chemotherapy Divisions, and 3Central AnimalLaboratory, National Cancer Center Research Institute, 4Department of Surgery,National Cancer Center Hospital, and 5Department of Surgery, Keio UniversitySchool of Medicine, Tokyo, JapanReceived 5/4/04; revised 12/21/04; accepted 12/28/04.Grant support:Grant-in-Aid for the SecondTerm Comprehensive10-Year Strategyfor Cancer Control from the Ministry of Health, Labor and Welfare ofJapan, and byResearch Resident Fellowships from the Foundation for Promotion of CancerResearch in Japan.The costs of publication of this article were defrayed in part by the payment of pagecharges.This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.Requests for reprints:TsuoTirchashi, Pathology Division, National Cancer CenterResearch Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan. Phone: 81-3-3542-2511; Fax: 81-3-3248-2463; E-mail: [email protected].

F2005 American Association for Cancer Research.

Imaging, Diagnosis, Prognosis

Research. on April 14, 2017. © 2005 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

might predict a poor prognosis is important in selectingpatients who would benefit from radical treatment ormolecular targeting therapy. Although a few immunohisto-chemical studies on TF expression in pancreatic ductalcarcinoma have been done (8, 14, 15), no detailedclinicopathologic study using multivariate-type analysis hasbeen carried out to date. In the present immunohistochem-ical study, we used a newly raised anti-TF antibody namedNCC-7C11 to examine TF expression in a large series ofsurgically resected pancreatic ductal adenocarcinomas, andinvestigated the correlations between TF expression andvarious clinicopathologic parameters, including the clinicaloutcome. Furthermore, we investigated the effect of TFknockdown on the invasiveness of a pancreatic cancer cellline using RNA interference, a new gene-silencing technique.

Materials andMethods

Production of the monoclonal antibody. Female BALB/c (nu/nu)mice were immunized with the scirrhous gastric carcinoma cell lineHSC-44PE by means of a rejection method, and hybridomas wereproduced as described previously (16). The hybridomas were thenselected on the basis of their immunohistochemical reactivity withvarious cancerous tissues, and a hybridoma that produced themonoclonal antibody (mAb) NCC-7C11 (IgG1, k), which reactedwith the invasive front of pancreatic ductal adenocarcinoma, wasobtained.

Cell lines and reagents. All pancreatic cancer cell lines (BxPC-3,SU 86.86., AsPC-1, Capan-1, Capan-2, PK-59, HPAC, MPanc-96,CFPAC-1, PANC-1, and MIAPaCa-2) were obtained from theAmerican Type Culture Collection (Rockville, MD). The scirrhousgastric carcinoma cell line HSC-44PE was established by Yanagihara(17). The cells were maintained in RPMI 1640 (BxPC-3, SU86.86.,AsPC-1, Capan-1, PK-59, HPAC, CFPAC-1 and HSC-44PE) or DMEM(Capan-2, MPanc-96, PANC-1, and MIAPaCa-2), supplemented witheither 20% (Capan-1) or 10% (others) heat-inactivated fetal bovineserum (Sigma Chemical Co., St. Louis, MO), 100 units/mL penicillinand 100 Ag/mL streptomycin (Invitrogen Corp., Carlsbad, CA) at37jC in a humidified atmosphere containing 5% carbon dioxide.Another murine anti-human TF mAb (TFE), recombinant human TFapoprotein, and normal murine IgG1k were purchased from EnzymeResearch Laboratories, Inc. (South Bend, IN), Angiopharm (O’Fallon,MO), and Becton Dickinson and Company (Franklin Lakes, NJ),respectively.

Immunoprecipitation. The BxPC-3 pancreatic carcinoma cell linewas used for immunoprecipitation. The cells were washed with ice-cold Ca2+/Mg2+-free PBS and treated with radioimmunoprecipitationassay buffer containing a proteinase inhibitor cocktail (RocheMolecular Biochemicals, Mannheim, Germany) on ice for 30minutes. After centrifugation (15,000 rpm for 30 minutes), thesupernatant was collected and precleared with protein G sepharose(50% slurry) at 4jC overnight. To conjugate the primary antibodies,1 Ag primary antibody and 25 AL protein G sepharose beadssuspended in RIPA buffer were incubated with mixing at 4jCovernight. After centrifugation, f500 Ag of total cellular proteinfrom the precleared supernatant and the antibody-sepharose conju-gate were incubated with mixing at 4jC for 3 hours. Theimmunoprecipitates were collected by centrifugation at 2,500 rpmfor 5 minutes at 4jC. After washing four times with RIPA buffer, thesupernatant was carefully removed and the pellets were resuspendedin 40 AL of 2� electrophoresis sample buffer.

Protein identification by mass spectrometry. The protein immuno-precipitated by mAb NCC-7C11 from the BxPC-3 lysate wassubjected to SDS-PAGE. The protein was visualized using a negativegel stain kit (Wako Pure Chemical Industries, Ltd., Japan) and its

band was excised from the gel. In-gel digestion was carried out withtrypsin (Promega, Madison, WI), as described in the literature (18).Mass spectrometric analyses of the trypsin digests were done usingVoyager (Applied Biosystems, Framingham, MA), and peptide massmapping was carried out with reference to the MASCOT database.

Western blot analysis. Samples were subjected to SDS-PAGE andtransferred to polyvinylidene difluoride membranes (Millipore, Bed-ford, MA). After blocking, the filters were incubated with the primaryantibodies, then with peroxidase-conjugated secondary antibodies(Amersham Biosciences Corp., Piscataway, NJ). The peroxidase-labeled bands were visualized using an electrochemiluminescencekit (Amersham Biosciences). As a loading control, the samemembrane was reprobed with an anti-h-actin mAb (Sigma-Aldrich),as described in the literature (19).

www.aacrjournals.orgClin Cancer Res 2005;11(7) April 1, 2005 2532

Fig. 1. Identification of the antigen recognized by mAb NCC-7C11. A,Western blotanalysis of the NCC-7C11antigen in various pancreatic cancer cell lines. Lane1,BxPC-3; lane2 , SU 86.86.; lane 3, AsPC-1; lane4, Capan-1; lane5, Capan-2; lane6,PK-59; lane 7, HPAC; lane 8, MPanc-96; lane 9, CFPAC-1; lane10, PANC-1; lane11,MIAPaCa-2; lane12 , HSC-44 (scirrhous gastric carcinoma cell line; Immunogen).Forty micrograms of whole cell lysate were applied to each lane and separated bySDS-PAGE under reducing condition. Position of the 45 kDa molecular size marker(right); B, identification of the protein immunoprecipitated by NCC-7C11using massspectrometry. After trypsin digestion, the ion peak spectra matched the sevenpeptide sequences ofTF (P13726); C, reciprocal coimmunoprecipitations fromBxPC-3 cells. Immunoprecipitates with NCC-7C11 (lane1), a commercially availableanti-TF mAb (TFE; lane 2) and mouse immunoglobulin (lane 4; negative control)were subjected to SDS-PAGE under nonreducing conditions.Whole cell lysatesserved as a positive control (lane 3); D, reactivity of antibodies with recombinanthumanTF apoprotein (0.5 Ag/lane) byWestern blot analysis under reducingcondition. Lane1, NCC-7C11; lane 2 , the anti-TF mAbTFE; lane 3, mouseimmunoglobulin.




Patients and tissue specimens. Formalin-fixed, paraffin-embedded

tumor specimens were obtained from a series of 113 consecutivepatients with pancreatic ductal adenocarcinoma who had undergone

surgical resection at the National Cancer Center Hospital in Tokyo,

Japan between 1990 and 1999. Patients with pancreatic tumors of aspecial type, such as mucinous cystadenocarcinoma, intraductal

papillary-mucinous adenocarcinoma, or adenosquamous carcinoma,were excluded. Three patients who died in the immediate postop-

erative period were also excluded. The patients consisted of 72 men

(63.7%) and 41 women (36.3%), who ranged in age from 45 to 82years, with a mean age of 63.1 years. The median duration of follow-

up was 16 months (range 2.9-72 months). The surgical procedureswere total pancreatectomy in 6 patients, distal pancreatectomy in 35

patients, pylorus-preserving pancreaticoduodenectomy in 20 patients,

and pancreaticoduodenectomy in 52 patients. Intraoperative radia-tion was done in 77 patients and postoperative chemotherapy was

given to 44 patients. The resected specimens were staged according tothe International Union against Cancer tumor-node-metastasis

(TNM) classification (20). Histologic grading of the tumors was

done according to the WHO classification system (21). Otherpathologic variables (lymphatic invasion, vascular invasion, perineu-

ral invasion, and growth pattern) were based on the Japan Pancreas

Society’s classification system for pancreatic carcinoma (22).Immunohistochemistry. The avidin-biotin-peroxidase complex

method was used for immunostaining, as described in the literature(23). Briefly, formalin-fixed, paraffin-embedded sections (4 Am thick)containing the maximum diameter of the tumor were deparaffinizedusing a graded ethanol and xylene series, treated with 0.3% hydrogenperoxide in methanol and immersed in 10 mmol/L citrate buffer (pH6.0). After autoclaving, the sections were incubated with normalswine serum for 10 minutes to block nonspecific antibody reactions,exposed to the primary antibody (final concentration, 1 Ag/mL)overnight at 4jC, then incubated sequentially with biotinylated goatanti-mouse IgG and avidin-biotinylated-peroxidase complex assupplied in the Vectastain ABC kit (Vector Laboratories, Burlingame,CA). The color reaction was developed over 5 minutes usingdiaminobenzidine tetrahydrochloride and 0.02% hydrogen peroxide,

and nuclear counterstaining with hematoxylin was done. The positivecontrol included in every assay was a section composed of formalin-fixed, paraffin-embedded cell pellets of the human pancreaticcarcinoma cell line BxPC-3, which was confirmed to express theNCC-7C11 antigen by Western blot analysis. Negative controlstaining, which was done using the same class of mouse immuno-globulin as the primary antibody, yielded negative results in everyspecimen.

RNA interference, immunocytochemistry, and invasion assays. Thesequences used to design the small interfering RNAs (siRNA) wereselected according to a previously described strategy (24–26). ThesiRNA sequences chosen to target TF (Genbank accession numberNM 001993) were positions 489 to 509 (siRNATF489) and 653 to673 (siRNATF653), numbered from the start codon, and the siRNAswere purchased from Dharmacon, Inc. (Lafayette, CO). Controlexperiments were done using two unrelated siRNAs. siRNALuc wasCy3 labeled siRNA directed against Luciferase mRNA (Dharmacon)and siRNANC (mock) was Nonspecific Control Duplex X (Dharma-con). The sequence of the latter (5V-NNATTCTATCACTAGCGTGAC-3V)was confirmed to have no homology with any known mRNA by aBLAST search; however, it had the same GC content as siRNATF.

At first, we examined transfection efficiencies among the TF-

positive cell lines BxPC-3, SU 86.86., and AsPC-1 by using Cy3-

labeled siRNA against luciferase. In >60% of BxPC-3 cells, Cy3 was

observed by fluorescence microscopy, and therefore the BxPC-3 cell

line was selected. This Cy3-labeled siRNA against luciferase was used

as a negative control in each experiment, so we confirmed the

transfection efficiency every time we did the siRNA knockdown and

invasion assay. Reduction of TF expression on the surface of cells was

confirmed by immunocytochemistry using anti-TF antibody NCC-

7C11, biotinylated goat anti-mouse IgG, and avidin-FITC (Vector

Laboratories) under fluorescence microscopy.RNA interference and invasion assays were done as described in

the literature (27). BxPC-3 cells were exposed to 40 nmol/L siRNA, inthe presence of Lipofectamine 2000 (Invitrogen), for 6 hours. Thetransfected cells were subjected to either immunoblot assays orinvasion assays 24 hours after the removal of the transfection reagent.


Fig. 2. Immunohistochemical stainingpattern of the anti-TF mAb NCC-7C11.A, NCC-7C11reacted preferentially with theinvasive tumor front, as shown in thismoderately differentiated pancreatic ductaladenocarcinoma (arrowheads);B, pancreatic ductal adenocarcinomacells were stained by NCC-7C11,whereas the adjacent normal pancreaticducts showed no immunoreactivity(arrows). Representative staining patterns(C and D); C, this moderately differentiatedadenocarcinoma was classified asshowing lowTF expression; D, this poorlydifferentiated adenocarcinoma wasmarkedly stained by NCC-7C11and wasclassified as showing highTF expression.Bar, 100 Am.

Tissue Factor in Pancreatic Ductal Adenocarcinoma




Table1. Association betweenTF expression and clinicopathologic variables

TF expression

LowTF(<25%, n = 63)

VariablesNegativeTF(0%, n = 13)

WeakTF(0%<, n = 50)

HighTF(zz 25%, n = 50)

P value(Low vs. High)

Age (y)<65 8 26 24>_65 5 24 26 0.5284

GenderMale 7 31 34Female 6 19 16 0.3989

Extent of the primary tumor spreadpT1 3 3 1pT2 3 9 5pT3 4 20 15pT4 3 18 29 0.0043*

Lymph node metastasispN0 2 7 5pN1a 10 20 11pN1b 1 23 34 0.0043*

Distant metastasispM0 9 42 28pM1 4 8 22 0.0041*

M1 (LYM)c 2 7 19 0.0039*

M1 (HEP)c 1 1 1 0.9999M1 (PER)c 1 0 2 0.5508

StageI 1 4 1II 0 2 2III 5 23 7IVA 3 13 18IVB 4 8 22 0.0002*

Histopathologic tumor gradeG1 7 24 16G2 5 23 23G3 1 3 11 0.0164*

Lymphatic invasionb

Negative 4 5 8Positive 9 45 42 0.8001

Vascular invasionb


Perineural invasionb


Growth patternb

Expansive + intermediate 9 30 21Infiltrative 4 20 29 0.0392*

Surgical marginNegative 10 34 30Positive 3 16 20 0.2744

*Significant.cLYM, lymphatic metastasis; HEP, hepatic metastasis; PER, peritoneal metastasis.bClassified according to the classification of Pancreatic Carcinoma ofJapan Pancreas Society.




The relative density of the chemiluminescence signal was determinedusing Image Gauge Software (Fuji Photo Film Co., Ltd., Japan) andstandardized by using the relative density of the h-actin signal. Forthe invasion assays, Biocoat Matrigel Invasion Chambers (BectonDickinson Labware) were utilized according to the manufacturer’sinstructions. We used Accutase (Innovative Cell Technologies, Inc.,San Diego, CA) to harvest cells for use in the invasion assay, and theharvested cells were washed with ice-cold PBS containing 0.1%bovine serum albumin before seeding. Transfected cells (4� 105) in500 AL RPMI 1640 containing 0.1% bovine serum albumin wereseeded into each insert chamber. Then, 750 AL RPMI 1640supplemented with 10% heat-inactivated fetal bovine serum wasadded to each lower chamber, and the plates were incubated at 37jCin a 5% CO2/95% air incubator for 18 hours. After incubation, thenoninvading cells were carefully removed from the top of each insertchamber with a cotton swab. The invading cells were then fixed andstained using a Diff-Quik kit (Sysmex Corp., Japan), and the totalnumber of invading cells was counted under a microscope. Each runwas done in triplicate, and the experiment was repeated indepen-dently thrice.

Statistical analysis. Correlations between TF immunoreactivityand patients’ clinicopathologic variables were analyzed using theMann-Whitney U test for the extent of the primary tumor spread(pT), lymph node metastasis, histologic tumor grade, and pTNMstage, and either the m2 test or Fisher’s exact test for the remainingvariables. The Kaplan-Meier method was used to generate survivalcurves, and differences in survival were analyzed using the log-ranktest, based on the TF expression status. Univariate and multivariateanalyses were done using the Cox proportional hazards model.Matrigel invasion assays and densitometric analyses were comparedusing the Mann-Whitney U test. Probability values <0.05 wereconsidered statistically significant. All analyses were done usingstatistical analysis software (Statview, version 5.0; SAS Institute, Inc.,Cary, NC).

Results

Monoclonal antibody characterization. Western blottingunder reducing condition showed that about half of thepancreatic cancer cell lines expressed moderate to high levelsof the NCC-7C11 antigen (Fig. 1A). A peptide massfingerprint of tryptic digests of the antigen immunoprecipi-tated from the BxPC-3 cell lysates was obtained by massspectrometry and a search of the MASCOT databaseidentified this antigen as TF (Fig. 1B). To confirm theidentity of TF, we did reciprocal coimmunoprecipitationassays using a commercially available anti-TF mAb TFE undernonreducing conditions (Fig. 1C). We also showed thereactivity of NCC-7C11 and TFE mAbs to recombinant TFapoprotein by immunoblotting (Fig. 1D). Together, thesedata confirmed that NCC-7C11 was an anti-TF mAb. Weexamined the TF expression pattern of the cell lines byWestern blotting with a commercially available polyclonalantibody against TF (clonal, American Diagnostic, Inc,Greenwich, CT), and thus confirmed the results of ourWestern blot analysis (data not shown).

Immunohistochemical analysis of tissue factor expression inpancreatic ductal adenocarcinoma. The immunostaining pat-tern of NCC-7C11 is shown in Fig. 2. TF expression occurredpreferentially at the invasive front of the tumor (Fig. 2A),whereas no TF was expressed in adjacent normal ductal cells(Fig. 2B), as previously described in the literature (14).According to the proportion of TF-positive cancer cells, TFexpression was classified as ‘‘low TF’’ (0-25% of cells showing

immunopositivity, Fig. 2C) or ‘‘high TF’’ (25% or more ofcells showing immunopositivity, Fig. 2D). Low TF includedpatients with completely TF-negative tumors (‘‘negative TF’’,0% of cells showing immunopositivity), and those withweakly TF-positive tumors (‘‘weak TF’’, >0% and <25% ofcells showing immunopositivity). The cutoff point for weak/high TF was set at the median value for the entire samplewithout the TF-negative sample. When comparing the highTF group with the low TF group, increased TF expression waspositively correlated with the extent of primary tumor spread,lymph node metastasis, the presence of lymphatic distantmetastasis, high tumor grade, advanced TNM stage, and aninfiltrative growth pattern (Table 1).

Prognostic significance of tissue factor expression. The surviv-al curves of the patients, grouped according to the level of TFstaining in their tumors, are shown in Fig. 3A . The high TFexpression group had a significantly poorer prognosis than the


Fig. 3. Kaplan-Meier survival curves for patients who had undergone surgicalresection of pancreatic ductal adenocarcinoma, stratified according to the level ofexpression of TF in their tumors. A, overall survival of patients with pancreaticductal adenocarcinoma (lowTF, 0-25%; highTF, 25% or more of the cells showingimmunopositivity; log-rank test, P < 0.0001); B, overall survival of patients whohad tumors with lymph node metastasis (negativeTF, 0%; weakTF, >0% and<25%; highTF, 25% or more of the cells showing immunopositivity; log-ranktest, P < 0.0001).




low TF expression group (log-rank test, P < 0.0001). Uponunivariate analysis with the Cox proportional hazards model,the extent of the primary tumor (P = 0.0497), lymph nodemetastasis (P = 0.0102), distant metastasis (P = 0.0027),histologic tumor grade (P = 0.0070), growth pattern(P = 0.0173), and TF immunopositivity (P < 0.0001) wereall positively correlated with a poor prognosis. Multivariateanalyses indicated that TF expression was an independentpredictor of an unfavorable prognosis (P = 0.0076; risk ratio,2.014; 95% confidence interval, 1.205-3.366), as were thepresence of lymph node metastasis (P = 0.0103) andhistologic tumor grade (P = 0.0154; Table 2). The survivalof the patients with lymph node metastasis was furtheranalyzed, grouped according to three TF staining levels, i.e.,negative TF, weak TF, and high TF (Fig. 3B). The survival ofthe TF-negative group was markedly better and increased TFexpression was significantly correlated with a poor prognosis(log-rank test, P < 0.0001).

The effects of small interfering RNAs targeted against tissuefactor on tumor invasion. TF overexpression proved to belinked with the aggressiveness of pancreatic cancer in ourimmunohistochemical analysis. In order to determine whether

down-regulation of endogenous TF would suppress theinvasive behavior of pancreatic cancer, we synthesized siRNAsthat, when transfected into cells, target TF mRNA fordegradation, thus reducing the expression of TF protein. Hightransfection efficiency of siRNAs into BxPC-3 cells has beenachieved with Lipofectamine 2000 (Fig. 4A, top) and reductionof TF expression by siRNATF653 against TF, compared withcontrol siRNANC, has been ascertained under fluorescencemicroscopy by immunocytochemistry (Fig. 4A, middle andbottom).Densitometricanalyses(Fig.4B)andinvasionassays(Fig.4C)(Fig. 4C) showed that transfection with either siRNATF489 orsiRNATF653 significantly reduced TF expression by, and theinvasiveness of, BxPC-3 cells compared with mock-transfectedcells (siRNANC), whereas transfection with a siRNA targeted toan unrelated mRNA (siRNALuc) had no effect on TF expressionor invasiveness.

Discussion

In the present study, we showed the clinicopathologicsignificance of TF expression in pancreatic ductal adenocarci-noma in an immunohistochemical analysis using a newly raised


Table 2. Prognostic factors in Cox’s proportional hazards model

Univariate Multivariate

Variables Hazard ratio95% Confidence

interval P Hazard ratio95% Confidence

interval P

Age (y)>_ 65/<65 1.202 0.782-1.846 0.4014

GenderFemale/male 0.909 0.582-1.420 0.6762

Extent of the primarytumor spreadpT4/pT1-pT3 1.545 1.001-2.385 0.0497* 1.280 0.793-2.066 0.3125

Lymph node metastasispN1a, pN1b/pN0 2.770 1.274-6.023 0.0102* 2.953 1.292-6.752 0.0103*

Distant metastasispM1/pM0 2.301 1.279-3.223 0.0027* 1.501 0.912-2.471 0.1101

Histologic tumor gradeG2, G3/G1 1.845 1.182-2.879 0.0070* 1.882 1.128-3.318 0.0154*

Lymphatic invasionPositive/negative 1.429 0.757-2.700 0.2708

Vascular invasionPositive/negative 1.412 0.877-2.273 0.1554

Tumor diameter (cm)>_ 3.5/<3.5 1.366 0.884-2.111 0.1604

Growth patternInfiltrative/expansive,intermediate

1.638 1.096-2.584 0.0173* 1.211 0.742-1.976 0.4446

Surgical marginPositive/negative 1.168 0.747-1.824 0.4959

ChemoradiotherapyNot received/received 0.957 0.562-1.630 0.8708

TF expressionHighTF/lowTF 2.723 1.748-4.243 <0.0001* 2.014 1.205-3.366 0.0076*

*Significant.




anti-TF antibody. Our findings indicate that TF has prognosticsignificance in patients with resectable tumors. Moreover, weconfirmed that TF contributed to the invasiveness of apancreatic cancer cell line by inhibiting TF expression usingthe RNA interference technique in vitro.

It is well recognized that cancer cells at the invasive frontexpress invasion-related molecules such as matrix metal-loproteinases (28) and the laminin g2 chain (29, 30). Weconfirmed that TF is another of these invasion-relatedmolecules, since TF immunopositivity was clearly observed atthe invasive fronts of the pancreatic ductal adenocarcinomas.Our immunohistochemical study also showed that TFexpression in the primary tumors was correlated significantlywith many aggressiveness-related factors, including the extentof primary tumor spread, lymph node metastasis, lymphaticdistant metastasis, TNM stage, tumor grade, and growthpattern. Among previous immunohistochemical studies of TFexpression in pancreatic ductal adenocarcinoma, only thatreported by Kakkar et al. (14) showed correlations betweenTF expression and clinicopathologic characteristics, showingthat TF expression is correlated with histologic tumor gradeand possibly with lymph node metastasis. In agreement withtheir results, the present study clarified that TF expressionwas indeed correlated with tumor grade and the extent oflymph node metastasis. Although there was a tendency for TFto be frequently expressed in G3 cells, it was also expressedin some well or moderately differentiated tumors. Moreover,it is very disconcerting that the least differentiated cell linesexamined, such as MIAPaCa-2 and Panc-1, proved TF-negative. However, in agreement with the present study,MIAPaCa-2 and Panc-1 have actually been reported toexpress hardly any TF mRNA (31). Therefore, we speculatethat TF is not merely an indicator of grade. It is unclear whatvalue this spectrum of cell lines adds to the current proposaland whether they are incapable of expressing TF. Furtheranalysis will be needed to reconcile this discrepancy betweenin vitro and in situ conditions. On the other hand, TFexpression in lymph node metastases is of great interest sinceour immunohistochemical analysis seemed to indicate thatTF was involved in lymph node metastasis. Therefore, we

have additionally examined 10 lymph node metastases todetermine whether TF expression is enriched in comparisonwith the expression in the primary tumor. We found that TFexpression in lymph node metastases reflected that in theprimary tumor, although it was not necessarily enriched(data not shown). Immunohistochemical studies on othercancers have also revealed correlations between TF expressionand clinicopathologic characteristics. In colorectal carcinoma,TF expression was positively correlated with lymph nodemetastasis, liver metastasis, and Dukes’ stage (32). In non–small cell lung cancers, TF expression was also associatedwith hematogenous or lymphogenous metastasis (33). Theseobservations are consistent with our findings, in that TF


Fig. 4. Effect ofTF knockdown by RNA interference on the invasiveness of humanpancreatic cancer cells. BxPC-3 cells were transiently transfected with shortinterfering RNAs and subjected to eitherWestern blot analysis or Matrigelinvasion assays. siRNATF489 and siRNATF653 are directed againstTF. Controlexperiments were done with a Cy3-labeled siRNA directed against an unrelatedmRNA (Luciferase; siRNALuc) and an irrelevant siRNA (siRNANC; used as amock-transfectant).Transfection efficiency was confirmed by using Cy3-labeledsiRNALuc in each assay, and representative pictures obtained by phase-contrastmicroscopy and fluorescence microscopy revealed a high efficiency of transfectionof siRNA into BxPC-3 cells (A, top). Immunocytochemistry under fluorescencemicroscopy shows that many cells lackTF expression on their surface as a result ofknockdown by siRNATF653 againstTF (A, middle), whereas control siRNANC

has no effect onTF surface expression (A, bottom). Reduction ofTF proteinexpression by siRNA againstTF was determined byWestern blot analysis anddensitometric analysis.The relative density of the chemiluminescence signal wasmeasured and standardized using the relative density of the h-actin signal.Transfection with either siRNATF489 or siRNATF653 significantly reducedTFcompared with mock-transfected cells (siRNANC), whereas transfection with asiRNA targeted to an unrelated mRNA (siRNALuc) had no effect onTF expression(B). For the invasion assays, the transfectants were seeded onto Matrigel-coatedinvasion chambers and incubated for18 hours, then the total number of cells on theunderside of each filter was determined. Invading cells were significantly suppressedby siRNA againstTF, as reflected in the observed reduction of protein expression(C). Columns, means; bars, SE (n = 9); *, P < 0.01compared with bothcontrol groups.




expression was significantly correlated with lymphatic distantmetastasis and TNM stage. In our series, TF expression didnot correlate with either hepatic or peritoneal metastasis,but only with lymphatic distant metastasis, suggesting apotential specificity of this protein’s role in invasion.However, it is rare for pancreatic tumors with distantmetastasis, except lymphatic distant metastasis, to becomeoperable. Therefore, it is difficult to conclude that there is nocorrelation between TF expression and distant metastasisbesides lymphatic distant metastasis. The present study alsorevealed that high TF expression was associated with theextent of the primary tumor and an infiltrative growthpattern, suggesting that TF overexpression has a proinvasiveeffect.

The clinical significance of high-level TF expression wasfurther substantiated by its correlation with a shorter overallsurvival time. Univariate analysis showed that TNM status,tumor grade, tumor size, growth pattern, and TF expressionwere all significantly correlated with patient survival. More-over, multivariate analysis also showed that TF expressionwas an independent prognostic factor. Therefore, TF hadsignificant predictive value for overall survival, suggesting thatits expression could be a useful predictor of poor prognosis.Although the hazard ratio of lymph node status was higherthan that of TF expression in multivariate analysis, lymphnode status and TF expression were proven to be statisticallysignificant and independent prognostic factors. Therefore, webelieve that both factors are almost equally important inpredicting prognosis in patients with pancreatic cancer.Indeed, among patients with lymph node metastasis, thosewith TF-negative tumors had a markedly better prognosis,and increased TF was also significantly correlated with apoorer prognosis. Thus, our findings suggest that TFcontributes to the aggressiveness of pancreatic ductal adeno-carcinoma. To our knowledge, this is the first study to haveshown the clinicopathologic significance of TF expression inpancreatic ductal adenocarcinoma using multivariate-typeanalysis.

The present study revealed that knockdown of endogenousTF could suppress the invasiveness of a pancreatic adenocar-cinoma cell line in vitro , suggesting that TF plays animportant role in tumor invasion. The potential role ofcoregulation of TF and effector proteases such as matrixmetalloproteinases has been reported previously for other celltypes (34, 35). In a small cell lung cancer cell line, thetransition of a small cell lung cancer from a suspension toadherent and aggressive growth was accompanied by expres-sion of TF as well as matrix metalloproteinases-2 and -9 (35).Other mechanisms by which TF promotes tumor invasionhave been suggested previously. Taniguchi et al. (31) showedthat binding of FVIIa to TF induced overexpression of theurokinase plasminogen activator receptor gene, which isinvolved in proteolytic extracellular matrix degradation,resulting in increased migration of pancreatic cancer cells,whereas blockade of TF activity with neutralizing monoclonalantibodies inhibited FVIIa-dependent tumor invasion. Ottet al. (36) showed that the role of TF in cell migration andadhesion is mediated by an interaction with actin-bindingprotein. TF has also been shown to mediate intracellularsignaling leading to the development of lamellipodia andfilopodia (5). In our invasion assay, however, the number of

invading control cells observed was higher than the levelsreported previously (37). One reason for the high invasionmay have been that the seeding density we used was morethan 10 times higher than that reported previously. Anotherreason might be that we used Accutase to harvest the cellsfrom culture, although Accutase has also been reportedlyutilized for the invasion assay in a study of another cell type(38). Since Accutase is reported to maintain most cell surfaceantigens and some antibodies including anti-TF antibody andanti–urokinase plasminogen activator receptor antibody workwell with Accutase according to the manufacturer (data notshown), cells treated with Accutase might retain their invasiveability. On the other hand, Accutase is a mixture of invasion-relevant proteases that are directly capable of degrading thereconstituted basement membrane used as a barrier in theinvasion assay. So, although the cells were washed beforebeing seeded, we cannot rule out the possibility that thisassay might not represent an examination of the capabilityof BxPC-3 cells to invade de novo , but rather their ability touse extrinsic enzymes to effect invasion. Although thepresent study could not prove the mechanism by which TFpromotes tumor invasion, our finding of a distinct associ-ation between TF and tumor invasiveness may havetherapeutic as well as prognostic implications. Since retinoicacid (39), resveratol (40), vitamin D3 (41), and pentoxifyl-line (42) have all been reported to down-regulate TF, theeffects of these agents on TF expression in pancreatic cancercells are worth evaluating. Recently, the relationship betweenTF expression and angiogenesis in various types of malig-nancies has also been emphasized (43–45); this may occurthrough regulation of the vascular endothelial growth factor(46). Therefore, down-regulation of TF expression might leadto the suppression of not only tumor invasiveness but alsoangiogenesis. However, although TF seems to be an attractivetarget for potential treatments of pancreatic ductal adeno-carcinoma, we must always be concerned about the possibleside effects of TF targeting therapy, including an increasedbleeding tendency.

Finally, Kakkar et al. showed that the level of TF washigher in the plasma of cancer patients, including thosewith pancreatic cancer, than in healthy controls (47).Furthermore, the plasma concentration of TF was shownto reflect tumor TF, which was correlated with the prognosisof patients with breast cancer (48). Hence, measurement ofthe plasma TF concentration might be of predictive valuefor prognosis or selecting candidates for TF-targetingtherapy, even in patients with inoperable pancreatic ductalcarcinoma.

In conclusion, our present findings indicate that there is asignificant association between TF expression and tumoraggressiveness in pancreatic ductal adenocarcinoma andsuggest that TF expression is a useful prognostic marker inpostoperative patients. In addition, TF expression maycontribute to the aggressiveness of pancreatic ductal adeno-carcinoma by stimulating tumor invasiveness.

Acknowledgments

The authors are grateful to A. Miura and F. Kaiya for their expert technical assis-tance. N. Nitori is a recipient of a Research Resident Fellowship from the Foundationfor Promotion of Cancer Research in Japan.






References1. Nemerson Y. Tissue factor and hemostasis. Blood

1988;71:1 ^ 8.2. DrakeTA, Morrissey JH, EdgingtonTS. Selective cel-

lular expression of tissue factor in human tissues.Implications for disorders of hemostasis and thrombo-sis. AmJPathol 1989;134:1087 ^ 97.

3. Ruf W, Mueller BM. Tissue factor signaling. ThrombHaemost1999;82:175 ^ 82.

4. Poulsen LK, Jacobsen N, Sorensen BB, et al. Signaltransduction via the mitogen-activated protein kinasepathway induced by binding of coagulation factorVIIato tissue factor. J Biol Chem 1998;273:6228 ^ 32.

5.Versteeg HH, Hoedemaeker I, Diks SH, et al. FactorVIIa/tissue factor-induced signaling via activation ofSrc-like kinases, phosphatidylinositol 3-kinase, andRac. J Biol Chem 2000;275:28750 ^ 6.

6.Trousseau A. Phlegmasia alba dolens. Clinique Medi-cale de l’Hotel-Dieu de Paris 3. Paris, France: JB Bal-liere et Fils; 1865. p. 654 ^ 712.

7. Ueda C, HirohataY, Kihara Y, et al. Pancreatic cancercomplicated by disseminated intravascular coagula-tion associated with production of tissue factor.J Gastroenterol 2001;36:848 ^50.

8. Callander NS,Varki N, Rao LV. Immunohistochemicalidentification of tissue factor in solid tumors. Cancer1992;70:1194 ^ 201.

9. Mueller BM, Reisfeld RA, Edgington TS, Ruf W. Ex-pression of tissue factor by melanoma cells promotesefficient hematogenous metastasis. Proc Natl AcadSci U S A1992;89:11832^ 6.

10. Kataoka H, Uchino H, AsadaY, et al. Analysis of tis-sue factor and tissue factor pathway inhibitor expres-sion in human colorectal carcinoma cell lines andmetastatic sublines to the liver. Int J Cancer 1997;72:878 ^ 84.

11. Bromberg ME, Konigsberg WH, Madison JF,Pawashe A, Garen A. Tissue factor promotes mela-noma metastasis by a pathway independent ofblood coagulation. Proc Natl Acad Sci U S A 1995;92:8205 ^ 9.

12. Kakkar AK, Chinswangwatanakul V, Lemoine NR,Tebbutt S, Williamson RC. Role of tissue factor expres-sion on tumour cell invasion and growth of experimen-tal pancreatic adenocarcinoma. Br J Surg 1999;86:890 ^ 4.

13. Niederhuber JE, Brennan MF, Menck HR. The Na-tional Cancer Data Base report on pancreatic cancer.Cancer 1995;76:1671 ^ 7.

14. Kakkar AK, Lemoine NR, Scully MF, Tebbutt S,Williamson RC. Tissue factor expression correlateswith histological grade in human pancreatic cancer.Br J Surg 1995;82:1101^ 4.

15.Wojtukiewicz MZ, Rucinska M, Zacharski LR, et al.Localization of blood coagulation factors in situ inpancreatic carcinoma. Thromb Haemost 2001;86:1416 ^ 20.

16.Watanabe M, Hirohashi S, ShimosatoY, et al. Carbo-hydrate antigen defined by a monoclonal antibodyraised against a gastric cancer xenograft. Jpn J Can-cer Res 1985;76:43 ^ 52.

17.Yanagihara K,Tanaka H,Takigahira M, et al. Establish-ment of two cell lines from human gastric scirrhouscarcinoma that possess the potential to metastasizespontaneously in nude mice. Cancer Sci 2004;95:575 ^ 82.

18. Rosenfeld J, Capdevielle J, Guillemot JC, Ferrara P.In-gel digestion of proteins for internal sequence anal-ysis after one- or two-dimensional gel electrophore-sis. Anal Biochem 1992;203:173 ^ 9.

19. Krajewski S, ZapataJM, Reed JC. Detection of mul-tiple antigens on Western blots. Anal Biochem 1996;236:221 ^ 8.

20. Sobin LH, Wittekind Ch. TNM classification of ma-lignant tumours. 5th ed. NewYork: Wiley-Liss; 1997.

21. Kloppel H, Solcia E, Longnecker DS, Cappella C,Sobin LH. Histological typing of tumors of the exo-crine pancreas. International histological classificationof tumors. 2nd ed. Berlin: Springer-Verlag; 1996.

22. Japan Pancreas Society. Classification of pancreaticcarcinoma. 1st English ed. Tokyo: Kanehara & Compa-ny, Ltd.; 1996.

23. Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidasetechniques: a comparison between ABC and unla-beled antibody (PAP) procedures. J Histochem Cyto-chem 1981;29:577 ^80.

24. Harborth J, Elbashir SM, Bechert K,Tuschl T,WeberK. Identification of essential genes in cultured mam-malian cells using small interfering RNAs. J Cell Sci2001;114:4557 ^ 65.

25. Paddison PJ, Caudy AA, Hannon GJ. Stable sup-pression of gene expression by RNAi in mammaliancells. Proc Natl Acad Sci U S A 2002;99:1443 ^ 8.

26. Hannon GJ. RNA interference. Nature 2002;418:244 ^ 51.

27. Krishnamachary B, Berg-Dixon S, Kelly B, et al.Regulation of colon carcinoma cell invasion byhypoxia-inducible factor 1. Cancer Res 2003;63:1138 ^ 43.

28.Yamamoto H, Itoh F, Iku S, et al. Expression of matrixmetalloproteinases and tissue inhibitors of metallopro-teinases in human pancreatic adenocarcinomas: clini-copathologic and prognostic significance of matrilysinexpression. J Clin Oncol 2001;19:1118 ^ 27.

29. Ono Y, Nakanishi Y, Ino Y, et al. Clinicopathologicsignificance of laminin-5 g2 chain expression insquamous cell carcinoma of the tongue: immunohis-tochemical analysis of 67 lesions. Cancer 1999;85:2315 ^ 21.

30. Koshikawa N, Moriyama K, Takamura H, et al.Overexpression of laminin g2 chain monomer in in-vading gastric carcinoma cells. Cancer Res 1999;59:5596 ^ 601.

31. Taniguchi T, Kakkar AK, Tuddenham EG, WilliamsonRC, Lemoine NR. Enhanced expression of urokinasereceptor induced through the tissue factor-factor VIIapathway in human pancreatic cancer. Cancer Res1998;58:4461 ^ 7.

32. Shigemori C, Wada H, Matsumoto K, Shiku H,Nakamura S, Suzuki H. Tissue factor expression andmetastatic potential of colorectal cancer. ThrombHaemost 1998;80:894 ^ 8.

33. Sawada M, Miyake S, Ohdama S, et al. Expressionoftissue factor in non-small-cell lung cancers and its rela-tionship to metastasis. BrJCancer1999;79:472 ^ 7.

34. Aljada A, Ghanim H, Mohanty P, Syed T,BandyopadhyayA, Dandona P. Glucose intake inducesan increase in activator protein 1and early growth res-ponse 1binding activities, in the expression of tissue

factor and matrix metalloproteinase in mononuclearcells, and in plasma tissue factor and matrix metallo-proteinase concentrations. Am J Clin Nutr 2004;80:51 ^ 7.

35. Salge U, Seitz R, Wimmel A, Schuermann M,Daubner E, Heiden M. Transition from suspension toadherent growth is accompanied by tissue factor ex-pression and matrix metalloproteinase secretion in asmall cell lung cancer cell line. J Cancer Res ClinOncol 2001;127:139 ^ 41.

36. Ott I, Fischer EG, Miyagi Y, Mueller BM, Ruf W. Arole for tissue factor in cell adhesion and migrationmediated by interaction with actin-binding protein280. J Cell Biol 1998;140:1241 ^ 53.

37. Maehara N, Matsumoto K, Kuba K, Mizumoto K,Tanaka M, NakamuraT. NK4, a four-kringle antagonistof HGF, inhibits spreading and invasion of human pan-creatic cancer cells. BrJ Cancer 2001;84:864 ^ 73.

38. Staff AC, Ranheim T, Henriksen T, Halvorsen B.8-Iso-prostaglandin f(2a) reduces trophoblast inva-sion and matrix metalloproteinase activity. Hyper-tension 2000;35:1307 ^ 13.

39. Tenno T, Botling J, Oberg F, Jossan S, Nilsson K,Siegbahn A. The role of RAR and RXR activation inretinoid-induced tissue factor suppression. Leukemia2000;14:1105 ^ 11.

40. Pendurthi UR, Meng F, Mackman N, Rao LV. Mech-anism of resveratrol-mediated suppression of tissuefactor gene expression. Thromb Haemost 2002;87:155 ^ 62.

41. Ohsawa M, Koyama T, Yamamoto K, Hirosawa S,Kamei S, Kamiyama R. 1a,25-dihydroxyvitamin D(3)and its potent synthetic analogs downregulate tissuefactor and upregulate thrombomodulin expressionin monocytic cells, counteracting the effects of tumornecrosis factor and oxidized LDL. Circulation 2000;102:2867 ^ 72.

42. Amirkhosravi A, Meyer T, Warnes G, et al. Pentoxi-fylline inhibits hypoxia-induced upregulation of tumorcell tissue factor and vascular endothelial growth fac-tor.Thromb Haemost1998;80:598 ^602.

43. Ohta S,Wada H, NakazakiT, et al. Expression of tis-sue factor is associated with clinical features and an-giogenesis in prostate cancer. Anticancer Res 2002;22:2991 ^ 6.

44. Nakasaki T,Wada H, Shigemori C, et al. Expressionof tissue factor and vascular endothelial growth factoris associated with angiogenesis in colorectal cancer.AmJHematol 2002;69:247 ^54.

45. Poon RT, Lau CP, Ho JW, Yu WC, Fan ST, Wong J.Tissue factor expression correlates with tumor angio-genesis and invasiveness in human hepatocellular car-cinoma. Clin Cancer Res 2003;9:5339 ^ 45.

46. Abe K, Shoji M, ChenJ, et al. Regulation of vascularendothelial growth factor production and angiogene-sis by the cytoplasmic tail of tissue factor. Proc NatlAcad Sci U S A1999;96:8663 ^ 8.

47. Kakkar AK, DeRuvo N, Chinswangwatanakul V,Tebbutt S,Williamson RCN. Extrinsic-pathway activa-tion in cancer with high factor VIIa and tissue factor.Lancet 1995;346:1004 ^ 5.

48. Ueno T, Toi M, Koike M, Nakamura S,Tominaga T.Tissue factor expression in breast cancer tissues: itscorrelation with prognosis and plasma concentration.BrJ Cancer 2000;83:164 ^ 70.




2005;11:2531-2539. Clin Cancer Res Nobuhiro Nitori, Yoshinori Ino, Yukihiro Nakanishi, et al. Ductal AdenocarcinomaPrognostic Significance of Tissue Factor in Pancreatic

Updated version

http://clincancerres.aacrjournals.org/content/11/7/2531

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/11/7/2531.full.html#ref-list-1

This article cites 43 articles, 16 of which you can access for free at:

Citing articles

/content/11/7/2531.full.html#related-urls

This article has been cited by 17 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

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications


http://clincancerres.aacrjournals.org/content/11/7/2531

http://clincancerres.aacrjournals.org/content/11/7/2531.full.html#ref-list-1

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

mailto:[email protected]

mailto:[email protected]


Documents

PrognosticSignificanceofTissueFactorinPancreatic ... · vival (hazard ratio, 2.014; P = 0.0076). Moreover, patients withTF-negative tumors had a sig-nificantly better prognosis even