aManuscript:/esps/HelpDesk:/esps/helpdesk.aspxDOI:10.3748/wjg.v20.i19.5826 WorldJGastroenterol2014May21;20(19):5826-5838ISSN1007-9327(print)ISSN2219-2840(online) ©2014BaishidengPublishingGroupInc.Allrightsreserved. ORIGINALARTICLE Identificationofbiomarkersforhepatocellularcarcinomabysemiquantitativeimmunocytochemistry HongMu,Kai-XuanLin,HongZhao,ShuXing,CongLi,FangLiu,Hai-ZhenLu,ZeZhang,Yu-LinSun,Xi-YunYan,Jian-QiangCai,Xiao-HangZhao HongMu,FangLiu,Yu-LinSun,Xiao-HangZhao,StateKeyLaboratoryofMolecularOncology,CancerInstituteHospital,ChineseAcademyofMedicalSciencesandPekingUnionMedicalCollege,Beijing100021,ChinaKai-XuanLin,ZeZhang,Xiao-HangZhao,CenterofBasicMedicalSciences,NavyGeneralHospitalofChinesePLA,Beijing100048,ChinaHongZhao,CongLi,Jian-QiangCai,DepartmentofAbdominalSurgery,CancerInstituteHospital,ChineseAcademyofMedicalScienceandPekingUnionMedicalCollege,Beijing100021,ChinaHai-ZhenLu,DepartmentofPathology,CancerInstituteHospital,ChineseAcademyofMedicalSciencesandPekingUnionMedicalCollege,Beijing100021,ChinaShuXing,Xi-YunYan,KeyLaboratoryofProteinandPeptidePharmaceutical,NationalLaboratoryofBiomacromolecules,InstituteofBiophysics,ChineseAcademyofSciences,Beijing100101,ChinaAuthorcontributions:MuHandLinKXperformedthemajorityoftheworkandwrotethemanuscript;ZhaoH,LiC,LiuF,LuHZ,ZhangZ,andSunYLassistedintheexperimentsanddataanalysis;XingSanalyzedthethreedimensionalreconstructions;ZhaoXH,YanXY,andCaiJQsupervisedthiswork;ZhaoXHdesignedthestudy.SupportedbyGrantsfromtheNationalHigh-techRandDProgramNo.2012AA020206,theKeyProjectfortheInfectiousDiseasesNo.2012ZX10002-017andNo.2013ZX10002009-001-004,theStateKeyProjectsforBasicResearchNo.2011CB910703,theNationalNaturalScienceFoundationNo.81372591,andNo.81321091ofChinaandtheCenterforMarineMedicineandRescueofTsinghuaUniversityCorrespondenceto:Xiao-HangZhao,MD,PhD,StateKeyLaboratoryofMolecularOncology,CancerInstituteHospital,ChineseAcademyofMedicalSciencesandPekingUnionMedicalCollege,PanjiayuanNanli17,ChaoyangDistrict,Beijing100021,China.zhaoxh@Telephone:+86-10-67709015Fax:+86-10-87778360Received:December31,2013Revised:February14,2014epted:April5,2014Publishedonline:May21,2014 Abstract AIM:Toinvestigatetheexpressionofkeybiomarkersinhepatomacelllines,tumorcellsfrompatients’bloodsamples,andtumortissues. METHODS:Weperformedthebiomarkertestsintwosteps.First,cellsplatedoncoverslipswereusedtoassessbiomarkers,andfluorescenceintensitieswerecalculatedusingtheNIHImageJsoftware.ThemeasuredvalueswereanalyzedusingtheSPSS19.0softwaretoparisonsamongeightcelllines.Second,eighty-fourindividualsampleswereusedtoassessthebiomarkers’expression.Negativeenrichmentofthebloodsampleswasperformed,andkaryocyteswereisolatedanddroppedontopretreatedglassslidesforfurtheranalysisbyimmunofluorescencestaining.Fluorescenceintensitiesparedamonghepatocellularcarcinoma(HCC)patients,chronicHBV-infectedpatients,andhealthycontrolsfollowingmethodssimilartothoseusedforcelllines.TherelationshipsbetweentheexpressionofbiomarkersandclinicalpathologicalparameterswereanalyzedbySpearmanrankcorrelationtests.Inaddition,westudiedthedistinctbiomarkers’expressionwiththree-dimensionallaserconfocalmicroscopyreconstructions,andKaplan-Meiersurvivalanalysiswasperformedtounderstandtheclinicalsignificanceofthesebiomarkers. RESULTS:Microscopicexaminationandfluorescenceintensitycalculationsindicatedthatcytokeratin8/18/19(CK)expressionwassignificantlyhigherinsixofthesevenHCCcelllinesexaminedthaninthecontrolcells,andtheexpressionlevelsofasialoglycoproteinreceptor(ASGPR)andglypican-3(GPC3)werehigherinallsevenHCCcelllinesthaninthecontrol.CellsobtainedfromHCCpatients’bloodsamplesalsodisplayedsig- WJG| 5826 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers nificantlyhigherexpressionlevelsofASGPR,GPC3,andCKthancellsfromchronicHBV-infectedpatientsorhealthycontrols;theseproteinsmaybevaluablesurfacebiomarkersforidentifyingHCCcirculatingtumorcellsisolatedandenrichedfromthebloodsamples.Thestemcell-likeandepithelial-mesenchymaltransition-relatedbiomarkerscouldbedetectedonthekaryocyteslides.ASGPRandGPC3wereexpressedathighlevels,andthusthree-dimensionalreconstructionswereusedtoobservetheirexpressionindetail.ThisanalysisindicatedthatGPC3waslocalizedinthelasmandmembrane,butthatASGPRhadapolarlocalization.SurvivalanalysesshowedthatexpressionofGPC3andASGPRisassociatedwithapatient’soverallsurvival(OS). CONCLUSION:ASGPR,GPC3,andCKmaybevaluableHCCbiomarkersforCTCdetection;theexpressionofASGPRandGPC3mightbehelpfulforunderstandingpatients’OS. ©2014BaishidengPublishingGroupInc.Allrightsreserved. Keywords:Hepatocellularcarcinoma;Biomarker;Immunocytochemistry;Semiquantitativeanalysis;Threedimensionalreconstruction Coretip:Wereportanovelworkflowtodetectpotentiallyvaluablebiomarkersforhepatocellularcarcinoma(HCC).Wemeasuredimmunofluorescenceintensityandperformedstatisticalanalysestoassesstheexpressionofbiomarkersincelllinesandpatientbloodsamples.Furthermore,wedeterminedtheexpressionofbiomarkersviathree-dimensionalreconstructions.Theseanalysesindicatedthatasialoglycoproteinreceptor(ASGPR),glypican-3(GPC3),andcytokeratin(CK)maybevaluableHCCbiomarkersfordetectingcirculatingtumorcells(CTCs).Inaddition,theexpressionofASGPRandGPC3mightcorrelatewithpatients’prognoses,andourCTCdetectionmethodcanincludeepithelialcelladhesionmolecule-andvimentin-positivetumorcells,andwillthussupplementpreviousstudiesandpotentiallyhelppredictfuturetumorrecurrenceandmetastasis. MuH,LinKX,ZhaoH,XingS,LiC,LiuF,LuHZ,ZhangZ,SunYL,YanXY,CaiJQ,ZhaoXH.Identificationofbiomarkersforhepatocellularcarcinomabysemiquantitativeimmunocytochemistry.WorldJGastroenterol2014;20(19):5826-5838Availablefrom:URL:/1007-9327/full/v20/i19/5826.htmDOI:/10.3748/wjg.v20.i19.5826 INTRODUCTION Hepatocellularcarcinoma(HCC)isoneofthemonmalignanttumorsintheworld,rankingseventhandthirdinmorbidityandmortality,respectively,forall cancers[1,2].Althoughmostoftheburdenliesindevelopingcountries,wherealmost85%ofHCCcasesur,largeincreasesinincidence,particularlyinyoungeragegroups,havebeenreportedintheUnitedStatesandEuropeoverthepasttwodecades[2].Currently,approximately50%oftheannualincidenceursinChina,makingHCCthesecondleadingcauseofcancer-relateddeaths.Thistrendislikelytopersistinthenearfuture,giventhat93millionhepatitisBvirus(HBV)carriersand980000HBVpatientsliveinChina[3,4].Atpresent,hepatectomyisthemaincurativetreatment,buterativeriskscanonlybepredictedmonclinicalandpathologicalparameters,suchasalpharotein(AFP)andtumordifferentiation;prognosisremainspoorduetothehighincidenceofrecurrenceandmetastasis[5,6].ConsideringthattherearenoreliablediagnosticandprognosticbiomarkersforHCCpatients,avarietyofstudieshavebeenperformedtoidentifynewbiomarkers,mainlyusingtumorsectionsorpatients’serumfortheanalyses[7-10]. Inrecentyears,thenumberofcirculatingtumorcells(CTCs)hasbeenreportedtobeassociatedwithapoorprognosis[11-13].ThedetectionofCTCsmayprovideamethodforassessingdiseaseprognosisandamodelforthebiologyofcancermetastasis.TheCellSearch™system(VeridexLLC,Warren,NJ,UnitedStates),basedonthepositivecaptureofepithelialcelladhesionmolecule(EpCAM),hasbeenapprovedbytheUnitedStatesFoodandDrugAdministration(FDA)forthedetectionofCTCsinpatientswithmetastaticbreast,prostate,andcolorectalcancers[14].In2012,thissystemwasapprovedbytheChinaFoodandDrugAdministrationforassessingpatientswithbreastcancer[15];however,fewstudieshaveexaminedHCCCTCs.WearguedthattheexpressionofEpCAMinHCCtissueswasaslowas30%,andcouldbeconsideredacancerstemcell-likebiomarkerforHCC.Theepithelial-mesenchymaltransition(EMT),consideredaninitiationprocessforcancermetastasis,involvesthelossofepithelialbiomarkerssuchasEpCAM,whichmeansthattheCellSearch™systemmayoverlookHCCCTCs[16,17].AlthoughseveralstudieshaveinvestigatedasuitablemethodforidentifyingHCCCTCs[18,19],furthereffortsarerequired.Here,weanalyzedtheexpressionofpotentialcellularbiomarkersfortheidentificationofHCCCTCsinsevencelllinesandinkaryocytesisolatedfromHCCpatients’peripheralblood.WeincludedHCC-relatedbiomarkers,suchascytokeratin8/18/19(CK),glypican-3(GPC3),andasialoglycoproteinreceptor(ASGPR);stemcell-relatedbiomarkers,suchasEpCAMandCD133;andEMTrelatedbiomarkers,suchasvimentin.Cytokeratinsareproteinsinthelasmiccytoskeleton,andCKexpressionhasbeenusedasabiomarkerinhepatomaathology[20,21].GPC3attachedtothecellsurfaceisoverexpressedinmostHCCfociandundetectableinnormalliversandbenignliverdiseases[7].ASGPR,asamembranereceptor,canspecificallyinteractwiththepre-S1domainofHBV[22].EpCAMwasreportedasa WJG| 5827 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers Table1Clinicalcharacteristicsofindividualsenrolledinthestudyn(%) Clinicalcharacteristics Ageatbaseline,years≥60<60MedianRange SexMaleFemale KPSscore>8060-80 HepatitisinfectionHBVpositiveHBVnegative Tumorsize(cm)≤3.0>3.0 PrimaryfociSingleMultiple CarcinomacellembolusYesNo AJCCstageatenrollmentEarly(stage1+2)Late(stage3+4) HCCnumber 20(32.3)42(67.7) 5529-76 53(85.5)9(14.5) 31(50.0)31(50.0) 59(95.2)3(4.8) 21(33.9)41(66.1) 54(87.1)8(12.9) 1(1.6)61(98.4) 55(88.7)7(11.3) Controlnumber 6(27.3)16(72.7) 5120-72 11(50.0)11(50.0) 22(100.0)0(0.0) 7(31.2)15(68.2) NA NA NA NA HCC:Hepatocellularcarcinoma;KPS:Karnofskyperformancestatus;HBV:HepatitisBvirus;AJCC:AmericanJointCommitteeonCancer;NA:Notapplicable. HCCstemcell-likebiomarker,andCD133ismonstemcellbiomarker.EpCAM-andCD133-positivecellscanpotentiallyundergoself-renewalanddifferentiation[23,24].VimentinisapredominantmesenchymalmarkerintheEMT[25].Giventhatnormalepithelialcellsandhepatocytesinthecirculationofhealthyadultsarerare,anyofthesecellsidentifiedbyCK,GPC3,orASGPRarelikelytobetumorcells[7,19,26].Stemcell-relatedandEMTmarker-positivetumorcellsmaybeassociatedwithrecurrenceandmetastasis,whichcorrespondswiththeclinicalsignificanceofCTCsinseveralcancers[27-30]. Theprimaryaimofthisstudywastoassessbiomarkerexpressioninhepatomacelllinesandenrichedpatientbloodcellsusingasemiquantitativeathologicalworkflow.Secondly,thepresentstudyanalyzesifexceptionalbiomarkerexpressioninHCCpatientscouldbevaluablefordeterminingcancerprognosis. MATERIALSANDMETHODS CellcultureandpreparationofcellsplatedoncoverslipsHumanhepatomacelllinesHepG2(ATCCHB-8065),Hep3B(ATCCHB-8064),andSK-HEP-I(ATCCHTB-52)cellswereobtainedfromtheAmericanTypeCultureCollection(ATCC,Manassas,VA,UnitedStates),andBel7402,Bel7404,SMMC7721,andthehumanhepaticcelllineL02cellswerepurchasedfromtheInstituteofBiochemistryandCellBiology,ChineseAcademyofSci- ences,China.Huh7cellswereobtainedfromtheHumanScienceResearchResourcesBank(Osaka,Japan).HepG2,Hep3B,andSK-Hep-IcellswereculturedinMEM(Hyclone,Logan,UT,UnitedStates)supplementedwith10%FBS(Hyclone,Logan,UT,UnitedStates),1%200mmolglutamine,and1%100mmolpyruvicacidsodium.Huh7cellswereculturedino’smodifiedEaglemedium(Hyclone,Logan,UT,UnitedStates)supplementedwith10%FBS.Bel7402,Bel7404,andSMMC7721cellswereculturedinRPMI-1640(Hyclone,Logan,UT,UnitedStates)supplementedwith10%FBS.Cellsweremaintainedat37 ℃inahumidifiedatmospherecontaining5%CO2,andwereharvestedwithtrypsinbeforeplatingforexperimentsintissuecultureplatesandoncoverslips.Thecellswereseededontocoverslipspretreatedwith0.01%poly-L-lysineandfixedwith4%paraformaldehydewhentheconfluencereachedapproximately60%-70%. BloodsamplepreparationEighty-fourindividuals,including62HCCpatients,7chronicHBV-infectedpatients,and15healthyindividuals,wererecruitedinthepresentstudy.Atotalof7.5mLofperipheralbloodwascollectedinBDvacutainertubescontainingacidcitratedextrose(BectonDickinson,FranklinLakes,NJ,UnitedStates).Writteninformedconsentwasobtainedfromeachparticipant,andthestudyprotocolconformedtotheethicalguidelinesofthe1975DeclarationofHelsinki,asreflectedinaprioriapprovalbytheReviewBoardattheCancerHospitalaffiliatedwiththeChineseAcademyofMedicalSciences,PekingUnionMedialCollege,andNavyGeneralHospital(Beijing,China).Toavoidepithelialcellcontaminationduringvenouspuncture,allsampleswerecollectedafterdiscardingthefirst2mLofblood.Sampleswereprocessedwithin24hofcollection.Diagnoseswerepathologicallyconfirmedusingsurgicalspecimens.TheclinicalcharacteristicsofHCCpatientsaresummarizedinTable1.HCCpatientswereclassifiedordingtotheseventheditionofthecancerstagingsystempublishedbytheAmericanJointCommitteeonCancer(AJCC)andtheUnionforInternationalCancerControl(UICC). Thekaryocyteenrichmentmethodwassimilartopreviousdescriptions[31].Bloodwastransferredtoa50mLcentrifugetube.Thecollectingtubeswererinsedtwicewithwashbuffer(137mmol/LNaCl,2.7mmolKCl,10mmol/LNa2HPO4,2mmol/LKH2PO4,2mmol/LEDTA,0.5%BSA,pH=7.4)tobinedvolumeof45mL.Bloodsampleswerecentrifugedat1400rpmfor5min,andthesupernatantwasaspirated.Redbloodcells(RBCs)weremixedwith37.5mLoflysisbuffer(155mmol/LNH4Cl,10mmol/LKHCO3,0.1mmol/LEDTA),rotatedfor8min,andcentrifugedat1400rpmfor5min.Theprocedurewasrepeatedtwice.Theresultingcellpelletwasresuspended,washed,andincubatedwithCD45microbeads(MiltenyiBiotec,BergischGladbach,Germany)ataproportionof20μLper107totalwhitebloodcells(WBCs)for15min.WBCsboundtomicrobeadswereremovedwithanLScolumninaMidi- WJG| 5828 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers MACS™separator(MiltenyiBiotec,BergischGladbach,Germany).Supernatantsweretransferredintoanewtubeandcentrifugedat1400rpmfor5min.Cellpelletswerefixedwith4%paraformaldehydeonSuperFrostPlusslides(ThermoFisherScientific,Pittsburgh,PA,UnitedStates),withimmunofluorescence(IF)stainingthenbeingperformed. ImmunofluorescencestainingandmicroscopicexaminationCellsoncoverslipsandkaryocyteslidesenrichedfrombloodsampleswereblockedusing2%BSA(SigmaAldrich,St.Louis,MO,UnitedStates)for45min.DirectandindirectIFstainingwasperformedatroomtemperaturewiththefollowingHCC-relatedbiomarkers(1:100dilutedin2%BSA):anti-CK8/18/19-FITC(MiltenyiBiotec,BergischGladbach,Germany),anti-GPC3(SantaCruzBiotechnologyInc,Dallas,TX,UnitedStates),antiASGPR(Sigma-Aldrich,St.Louis,MO,UnitedStates),anti-AFP(ZymedLaboratoryInc.,SouthSanFrancisco,CA,UnitedStates),andanti-CD45-PE(MiltenyiBiotec,BergischGladbach,Germany).EMT-relatedandstemcell-relatedbiomarkers(1:100dilutedin2%BSA)werealsotested,includinganti-vimentin(ThermoFisherScientific,Pittsburgh,PA,UnitedStates),anti-EpCAMFITC(MiltenyiBiotec,BergischGladbach,Germany),andanti-CD133-PE(MiltenyiBiotec,BergischGladbach,Germany).Coverslipsandslideswereincubatedfor60minwithprimaryantibodiesforIFstainingandwithAlexaFluor-labeledsecondaryantibody(LifeTechnologies,GrandIsland,NY,UnitedStates)foranother60minforindirectIFstaining.Theywerethenwashedthreetimeswith0.2%BSAfor3min.Cellsweremountedwithmediumcontainingthenucleardye4’,6-diamidino-2-phenylindole(DAPI)(Sigma-Aldrich,St.Louis,MO,UnitedStates).BlindedreviewofthefluorescentimageswasperformedbythreetechniciansusingaNikon80i3-colorfluorescentmicroscope(NikonCo,Tokyo,Japan)andaLeicaTCSSP2confocalmicroscope(LeicaCo,Berlin,Germany).FluorescenceintensitiesweremeasuredusingtheImageJsoftware(NIH,Bethesda,MD,UnitedStates).Relativebiomarkerintensitiesoncellswereassessedusingthefollowingformulathatcalculateshowmanytimeshigherthebiomarkerstainingintensityofthecellisthanthebackground,whichwasbasedontheCellSearch™modelforHER2stainingofCTCs[32,33]: BiomarkerRelativeIntensity=[ForegroundIntensity(Cellstaining)/SurfaceArea]/[BackgroundIntensity/SurfaceArea] Theexperimentswereperformedintriplicate.Representativeimagesof5cellsfromeachcelllineand3cellsfromeachpatientwereanalyzed,andthemeanintensitywascalculated.Onlycellsfrompatientswhometthefollowingstringentcriteriawereanalyzed:intactcellswithdiametersabove10μm,biomarker-positive,CD45negative,andDAPI-positive.Thepositiveratiosofbiomarkerexpressionwerecalculatedfromtheobservationof500cells. ImmunohistochemicalstainingoflivercancersamplesTissuearraysposedoftumorsamplesandpara-carcinomalivertissuescollectedfrom32patientswhowerediagnosedwithHCCandunderwentsurgicalresection(ShanghaiNationalEngineeringResearchCenterforBiochip,Shanghai,China).Fromeachparaffinblock,4-μm-thicksectionswerecut,clearedinxylene,andrehydrated.Slideswerethenmovedfromwatertoplasticslideholders,fullyimmersedin10mmolsodiumcitratebuffer(pH=6.0),andheatedfor20minat120 ℃inmerciallyavailablepressurecooker.Aftercoolingtoroomtemperatureinthesodiumcitratebuffer,slidesweretreatedwithasolutionof3%hydrogenperoxide(H2O2)for30minatroomtemperaturetoabolishendogenousperoxidaseactivity.Sectionswerethenincubatedfor10mininamoistchamberwithnon-immunegoatserumdilutedto5%inPBS(pH=7.4)toreducenon-specificbackgroundstaining.Sectionswerethenincubatedovernightat4 ℃withtheprimaryantibody(anti-GPC3oranti-ASGPR)dilutedto1:100.Sectionsweresubsequentlyincubatedfor30minwithbiotin-labeledsecondaryantibody(SantaCruzBiotechnologyInc,Dallas,TX,UnitedStates)dilutedto1:600inPBS,andthenincubatedinastreptavidin-biotin-peroxidaseplex(ZhongshanJinqiaoBiotechnologyCo,Beijing,China)for30min.Theimmunologicreactionwasvisualizedusing3,3′-diaminobenzidinesubstrate,andsampleswerecounterstainedwithhematoxylin,dehydrated,andmountedwithCanadabalsam(Sigma-Aldrich,St.Louis,MO,UnitedStates).Anegativecontrolwasperformedbyomittingtheprimaryantibody. ThetumorexpressionofASGPRandGPC3wasevaluatedbytwopathologists(HMandHZL)blindedtoclinicaldata,anddiscrepancieswereresolvedbyconsensus.ImageswerevisualizedusinganOlympusBX40microscope(OlympusCo,Tokyo,Japan).Tenrandomfieldsperbiomarkerwereselected,andthepercentageofimmunoreactivecellsinatotalof1000tumorcellswasdetermined[34].Asdescribedpreviously,theintensityofabiomarkerandthepercentageofpositivecellswerebothgradedwithdifferentscores.Theproductofthetwoscoreswasusedtoevaluatethebiomarker’sstaining[35,36].Anintensityscorerepresentingtheaverageintensityofpositivecells(0,none;1,weak;2,intermediate;3,strong)wasassigned.Aproportionscorerepresentingtheestimatedproportionofpositive-stainingcells(0,0-5%positivecells;1,6%-25%;2,26%-50%;3,51%-75%;4,76%-100%)wasassigned.Theproportionandintensityscoresweremultipliedtocreateanimmunoreactivityscore(IS).TheISwasfurtherdividedasfollows:0-1(−);2-4(+);5-7(++);>8(+++).“-”and“+”wereconsideredlowlevelsofimmunoreactivity;“++”and“+++”wereconsideredhighlevelsofimmunoreactivity. StatisticalanalysisAnalyseswereperformedusingSPSSversion19.0(IBM,NewYork,NY,UnitedStates).TheMann-Whitneytestwasusedparecontinuousvariablesbetweenthe WJG| 5829 May21,2014|Volume20|Issue19| MuHetal.SemiquantitativeanalysisofhepatocellularcarcinomabiomarkersA Bel7402 Bel7404 SMMC7721 Hep3B HepG2
B Bel7402 Huh7Bel7404 SK-HEP-
I LO2 SMMC7721 Hep3B HepG2
C Bel7402 Huh7Bel7404 SK-HEP-
I LO2 SMMC7721 Hep3B HepG2 Huh7 SK-HEP-
I LO2 WJG| 5830 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers D60b 50 b 40 Relativeintensity/times 30 b b 20 b bbbbb bb 10 b b b bb
0 Bel74 Bel74 SMMCHep3 HepG Huh7 02 04 772
B 2
1 bbbSK-HeP-
I ASGPRGPC3CK LO2 Figure1Asialoglycoproteinreceptor,glypican-3andcytokeratinexpression,andfluorescenceintensitiesineightcelllines.A-C:Representativeimages ofcellsonthecoverslipswithDAPI-stainednuclei(blue)andasialoglycoproteinreceptor(ASGPR)staining(green)/glypican-3(GPC3)staining(red)/cytokeratin(CK) staining(green).Thescalebaris5μm.D:Thebargraphoffluorescenceintensitiesineightcelllines.FluorescenceintensitiesweremeasuredusingNIHImageJ software.Biomarkerrelativeintensitieswerecalculatedasthedifferencebetweenthebiomarkerstainingintensityofthecellandthebackgroundintensity.parisonsbetweenHCCcelllinesandthecontrolcelllinewereanalyzedusingtheMann-Whitneytest(bP<0.01vscontrol). Bel7402 Bel7404 SMMC7721 Hep3B HepG2 Huh7 SK-HEP-
I LO2 Figure2Epithelialcelladhesionmoleculeexpressionineightcelllines.RepresentativeimagesofcellsonthecoverslipswithDAPI-stainednuclei(blue)andEpCAMstaining(green).Thescalebaris5μm. twogroups.Fluorescenceintensitiesandclinicalcharacteristics,suchasage,AFP,ALT,AST,tumorsize,differentiation,andAJCCstage,weresubjectedtoSpearmanrankcorrelationanalysis.SurvivalanalysiswasperformedusingtheKaplan-Meiermethod.Pvalueslessthan0.05wereconsideredstatisticallysignificant,andPvalueslessthan0.1wereconsideredpotentiallysignificant. RESULTS ExpressionofHCC-relatedbiomarkersincelllinesIFstainingindicatedthatASGPR(Figure1A)andGPC3(Figure1B)wereexpressedinalllivercancer-relatedcelllines,andfluorescenceintensitywassignificantlyhigherinthesevenHCCcelllinesthaninL02,thecontrolcells(Figure1D).ConfocalmicroscopicexaminationsuggestedthatASGPRislocatedinthenucleiornucleoli oftheBel7404andSMMC7721cells.Inaddition,GPC3expressionwassignificantlystrongerincelllinesderivedfromAsiansthanfromWesterners(10.41±6.67vs5.61±2.89,P<0.05).CKwasexpressedinallcelllinesandlocatedinthelasm(Figure1C).Whenthefluorescenceintensitiespared,CKexpressioninsixofthesevenHCCcelllineswassignificantlyhigherthaninthecontrolcellline,L02(Figure1D).AFPwasonlyexpressedinHep3B,HepG2,andSK-HEP-Icelllines. ExpressionofEMT-andstemcell-relatedbiomarkersincelllinesIFstainingindicatedthattherewereconsiderablyfewercellsexpressingEpCAMthantherewerecellsexpressingtheabovethreebiomarkers(Figure2).ThecellsexpressingCD133wereevenrarerthanthoseexpressingEpCAM.Table2presentsthepositiveratiosinthecelllines WJG| 5831 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers Table2Positiveratiosofepithelialcelladhesionmolecule/CD133expressioninhepatocellularcarcinomacells % EpCAMCD133 Bel7402 0.40.4 Bel7404 1.20.4 EpCAM:Epithelialcelladhesionmolecule. SMMC7721 0.20.2 Huh7 10.2 HepG2 0.60.2 Hep3B 0.80.4 SK-Hep-
I L02
2 0 0.6
0 tested.Vimentinwasexpressedatverylowlevels,andratiosinthecelllinesdeterminedbyIFstaining. ExpressionofbiomarkersinbloodsamplesBasedontheaboveresults,threeHCC-relatedbiomarkerswerefurtherdetectedinpatientsamples.IFstainingindicatedthatGPC3,ASGPR,andCKwereexpressedinmostHCCpatients(Figure3),independentofHBVinfection.PositivestainingwasrarelyobservedonthekaryocyteslidesfromchronicHBV-infectedpatientsorhealthycontrols. paredthefluorescenceintensitiesofthreebiomarkersbetweenHCCpatients,chronicHBV-infectedpatients,andhealthycontrols.TheexpressionofthethreebiomarkerswassignificantlyincreasedinHCCparedtochronicHBV-infectedpatientsandhealthycontrols.ThefluorescenceintensitiesofGPC3andASGPRwereconsiderablyhigherthanthatofCK.Theconcordanceratesbetweenabiomarker’sexpressionandpathologicaldiagnosiswere90%forGPC3,93.3%forASGPR,and63.3%forCK,whereastheconcordanceratebetweenAFPserumdetectionandpathologicaldiagnosiswas46.7%.Thepositiveandnegativepredictivevaluesofthesethreebiomarkerswere90%and71.4%forGPC3,93.1%and75%forASGPR,and82.6%and28.6%forCK,respectively. Tounderstandtheclinicalsignificanceofthebiomarkers’expressionlevels,weusedSpearmanrankcorrelationanalysistoassesstherelationshipbetweenfluorescenceintensitiesandclinicalpathologicalparameters.Nosignificantcorrelationswerefound. VimentinwaspositivelyexpressedonsomeofthekaryocyteslidesandwaspaniedbytheexpressionofEpCAM(Figure4).BecauseEpCAMisassociatedwithstemcell-likepropertiesinHCCpatients,itwasunnecessarypareitsfluorescenceintensitiesinthemannerperformedfortheHCC-relatedbiomarkersdiscussedpreviously.EpCAMwasexpressedin34.4%andvimentinwasexpressedin24.1%ofthecellsonthekaryocyteslides. Three-dimensionalreconstructionforGPC3andASGPRThree-dimensionalreconstructionsweregeneratedtoevaluateASGPRandGPC3localization.ThecellswererevolvedaroundtheX-,Y-,andZ-axes,whichindicatedthatGPC3waslocalizedinthelasmandonthemembrane(mov1),whereasASGPRhadapolarlocalization(mov2). SurvivalanalysisforGPC3andASGPRTheexpressionofGPC3andASGPRisshowninFigure5.Kaplan-MeiersurvivalanalysesindicatedthatASGPRexpressionwassignificantlyassociatedwiththeoverallsurvival(OS)(P=0.017),andGPC3expressionexhibitedatrendtowardsignificantassociationwithOS(P=0.068).ThemeanOSofpatientswithhighandlowlevelsofASGPRwas24.1and48.5mo,respectively.ThemeanOSofpatientswithhighandlowlevelsofGPC3was31.0and48.9mo,respectively. DISCUSSION HCCranksasthethirdmostfrequentcauseofcancerrelateddeathworldwide[2].InChina,itisthethirdmostprevalentcancer,thesecondleadingcauseofcancerrelateddeath,andnewcasesinthecountryountfor55%ofthosereportedglobally[4].CurrentlyserumAFP,asecretoryprotein,iswidelyusedfordetectingHCCpatientsandmonitoringdiseaseprogression,butithasasensitivityrangingfrom39%to97%andaspecificityrangingfrom76%to95%,evenwhenusedtoscreenhigh-riskpopulations[37-39].Inapreviousstudy,AFPwasusedtoidentifyCTCsinpatientswithlivercancer[18].However,giventheingsofAFP,theremaybebiomarkersthatcandetectCTCswithahighersensitivity. Inthisstudy,andbasedonourprevioustests,weassessedtheexpressionofpotentialkeybiomarkersinsevenHCCcelllinesandkaryocytesisolatedfrompatientbloodsamples.Cytokeratinsareproteinsthatconsistofkeratin-containingintermediatefilamentsinthelasmiccytoskeleton,andtheirexpressionprimarilydependsonthetypeofepithelia,themomentofterminaldifferentiation,andthestageofdevelopment[40].Inmanycases,cytokeratinexpressionintumorsandperipheralbloodhasprognosticsignificanceforcancerpatients,andthelevelsofCK8/18/19expressioninHCChavebeenusedasbiomarkersinathology[20,21].Inadditiontoconventionalathology,CK8/18/19havebeenusedasdiagnosticbiomarkersinCTCsforbreast,prostate,andcolorectalcancers[11-13].ExpressioninHCCcellsobtainedfromperipheralbloodwasexploredinthisstudyinordertobroadenknowledgeregardingathology.Positiveresultsfromcelllinesandbloodsamplesprovideevidenceforpotentialapplicationsofbiomarkers,althoughadditionalinvestigationmaybeneededinthefuture. WJG| 5832 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers
A CD45 DAPI Merge GPC3 ASGPR CK Relativeintensity/timesRelativeintensity/timesRelativeintensity/times B 20 bb 15 10
5 0HCC HBVControl b 20 b 15 10
5 0HCC HBVControl 20 15 10 b b
5 0HCC HBVControl HCCHBVControl Figure3Representativeimagesandfluorescenceintensitiesofcellsobtainedfrompatients.A:Representativeimageofcirculatingtumorcells(CTCs)withDAPI-stainednuclei(blue),positivestainingofASGPR/GPC3/CK(green),andnegativestainingforCD45.Thescalebaris5μm.Atotalof7.5mLofperipheralbloodwascollectedfromeachindividual,mixedwithredbloodcelllysisbuffer,andincubatedwithCD45microbeadstodepletethewhitebloodcells.TheremainingcellpelletswerefixedonSuperFrostPlusslidesforimmunofluorescencestaining;B:ThebargraphofASGPR/GPC3/CK(fromlefttoright)fluorescenceintensitiesin30HCCpatients,7chronicHBV-infectedpatients,and15healthycontrols.FluorescenceintensitiesweremeasuredusingNIHImageJsoftware.Biomarkerrelativeintensitieswerecalculatedasthedifferencebetweenthebiomarkerstainingintensityofthecellandthebackgroundintensity.parisonsbetweenHCCpatientsandhealthycontrolswereanalyzedbytheMann-Whitneytest(bP<0.01vscontrol). ASGPRislocatedonlivercellsandwaspreviouslyknownasahepaticgalactose/N-acetylglucosamine(GlcNAc)receptor.Itsprimaryphysiologicalfunctionistobind,internalize,andsubsequentlyclearglycoproteinscontainingterminalgalactoseorGlcNAcresiduesfromcirculation[41-43].Severalinvestigationshavesupported ASGPRasamultifunctionalmembranereceptorinvolvedintheremovalofapoptoticcells[44],acarrieroflow-densitylipoprotein(LDL)[45],andanentrypointforhepatotropicviruses[46,47].HBVDNAandvirionscanreportedlybetransferredintohepatocytesviaASGPR[48].Zhangetal[22]foundthatASGPRonhumanhepatocytes WJG| 5833 May21,2014|Volume20|Issue19| MuHetal.SemiquantitativeanalysisofhepatocellularcarcinomabiomarkersA EpCAM
B Vimentin DAPI Merge EpCAM Vimentin DAPI Merge Figure4Representativeimagesofepithelial-mesenchymaltransitionmarker-relatedcellsobtainedfromhepatocellularcarcinomapatients.A:CTCswithdye4’,6-diamidino-2-phenylindole,(DAPI)-stainednuclei(blue),positivestainingofepithelialcelladhesionmolecule(EpCAM)(green),andpositivestainingofvimentin(red).B:CTCswithDAPI-stainednuclei(blue),positivestainingofEpCAM(green),andnegativestainingofvimentin.Thescalebaris5μm. caninteractspecificallyanddirectlywiththepre-S1domainofHBVinvitroandinvivo[22].TheIFimagesoflivercancercelllinesandpatientCTCsinthisstudyindicatedstrongASGPRexpression,whichsuggesteditspotentialuseforidentifyingHCCcellsincirculation,especiallyforHBV-relatedHCC. GPC3isaproteoglycanattachedtothecellsurfacebyaglycosylphosphatidylinositol(GPI)anchor[7,49].GPC3mRNAlevelshavebeenreportedtobesignificantlyelevatedinmostHCCparedwithnormalliverandnon-malignantliverlesions,andtheyareelevatedmorefrequentlythanAFPinHCC[50].Similarresultshavebeenobtainedattheproteinlevel.Usingimmunohistochemicaltechniques,Capurroetal[7]foundthatGPC3wasspecificallyoverexpressedinmostHCCfoci,butwasundetectableinhepatocytesfromnormalliverandbenignliverdiseases.Stainingwaslocalizedtothecellmembraneand/orthelasm.Zhangetal[8]foundthat87.1%(128/147)ofsurgicallyexcisedHCCsampleswereGPC3positive,anddidnotobservedGPC3expressioninpara-carcinomaorcirrhotictissues.ThedetectionofHCCcelllinesandkaryocytesinpatients’bloodinourstudieswasconsistentwiththeaboveresultsforbothmRNAandproteinexpression,displayingpositiveparedtothecontrols.Additionally,tumorcellscanbeidentifiedbyGPC3inpatients’blood,inwhichAFPtestscouldbepositiveornegative.ThisresultwasalignedwithapreviousreportthattestedforGPC3inHCCpatients’serum[51].TheGPC3biomarkertestresultscorrespondedbettertothepathologicaldiagnosisinourstudythantheserumAFPtest. TheresultsfromSpearmanrankcorrelationanalysis(Table3)indicatedthattherewasnosignificantcorrela- tionbetweenfluorescenceintensitiesandclinicalpathologicalparameters,whichsuggestedthatfluorescenceintensitymaybeunaffectedbyclinicalpathologicalparameters.Fromouranalysesofthebloodsamples,thefluorescenceintensitiesoftheabovethreebiomarkersweresignificantlyincreasedinHCCparedtochronicHBV-infectedpatientsandhealthycontrols,althoughtheelevationofCKexpressionwasnotaspronouncedastheothertwobiomarkers.ThisfindingindicatesthatCKexpressionmaybeperturbedinchronicliverdisease,whereasGPC3andASGPRcouldprovideabetterdistinctionbetweenHCCandchronicliverdisease.AFPwasonlyexpressedinthecelllinesderivedfromWesternersinourstudy.AFPwasreportedtobeexpressedin52.3%ofHCCpatients(23/44)fortheidentificationofCTCs,whereasthethreebiomarkersthatweconsideredweremonlydetectedthanpositivechangesinAFPintheserum[18].Inthisstudy,survivalanalysisindicatedthatpatientswithhighASGPRexpressionlevelshadpoorerOS,andasimilartrendwasobservedforGPC3expression.TheseresultssuggestthatGPC3andASGPRmaycorrelatewithpatientprognosis. EMTisaprocessbywhichepithelialcellslosetheircellpolarityandcell-celladhesion,andgainmigratoryandinvasivepropertiestoemesenchymalcells[17].Thistransitionisconsideredoneofthemechanismsunderlyingcancermetastasisandrecurrence.EMTandMETformtheinitiationpletionoftheinvasion-metastasiscascade[27],andCTCsinthevesselsactas“intermediates”betweenprimarytumorsandmetastaticfoci.WemeasuredtheexpressionofanEMT-relatedbiomarkeronthepatients’karyocyteslidestounderstandthebiologyofCTCs.Vimentin,awidelyusedbiomarker WJG| 5834 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers
A B
C D E 1.0 0.8 0.6 0.4 Survivalfunctions ASGPRexpressioninHCCHighlevelsLowlevelsHighlevels-censoredLowlevels-censored F 1.0 0.8 0.6 0.4 Survivalfunctions GPC3expressioninHCCHighlevelsLowlevelsHighlevels-censoredLowlevels-censored CumsurvivalCumsurvival 0.2 0.2 0.00 20 40 60 t/mo 0.00 20 40 60 t/mo Figure5Representativeimagesofasialoglycoproteinreceptor/glypican-3expressionontissuearrays,andKaplan-Meiersurvivalcurvesfromfollow-upstudies.A-B:TheexpressionofASGPRinHCC(A)andparacarcinomatous(B)tissues.C-D:TheexpressionofGPC3inHCC(C)andmatchedparacarcinomatous(D)tissues(A-D:×100;Inner×200).Thesurvivalcurvesfromsixandahalfyearsoffollow-upindicateasignificantdifferencebetween32HCCpatientswithhighandlowlevelsofASGPR(E)andexhibitatrendtowardsignificancebetween32HCCpatientswithhighandlowlevelsofGPC3(F). Table3Spearmanrankcorrelationanalysisforfluorescenceintensities Pvalue ASGPRGPC3CK Age 0.9450.8630.889 AFP 0.9090.9110.173 ALT 0.1310.5490.573 AST 0.2810.8460.771 TBIL 0.8100.2880.970 Size 0.5370.3570.173 Differentiation 0.4180.6230.974 AJCCstage 0.7960.8490.425 ASGPR:Asialoglycoproteinreceptor;GPC3:Glypican-3;CK:Cytokeratin8/18/19;AFP:rotein;ALT:Alanineaminotransferase;AST:Aspartateaminotransferase;TBIL:Totalbilirubin;AJCC:AmericanJointCommitteeonCancer. forEMT,wasexpressedonsomeCTCs,whereasEpCAM,anepithelium-specificbiomarker,wasexpressedatvariouslevels.TheseresultssuggestthatCTCsmaybeassociatedwithcancermetastasisandrecurrenceviaEMTmechanisms.Somerecentevidenceindicates thattumorcellsundergoingEMTmightgainstemcell- likeproperties,thereforegivingrisetocancerstemcells(CSCs)[27,30],andEpCAM-positiveHCCcellshavebeen reportedtohavestemcell-likefeatures,includingselfrenewalanddifferentiationcapacities[23,29,52].Previously, WJG| 5835 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers EpCAMwasonlyknownasanepithelialbiomarkerinmanyhumancancerswithanepitheliumorigin[53],andithasthusbeenusedasatargettocaptureCTCsbytheCellSearch™system.OurresultsdemonstratethatEpCAMwasnotwidelyexpressedintheHCCcelllines,andtheexpressionratioonthekaryocyteslideswasconsistentwithhistologicalreports[35],whichimpliedthattheCellSearch™systemmaynotbeidealforthedetectionofHCCCTCs[52].TheexpressionlevelsofvimentinandEpCAMwereapproximately20%and30%,respectively,ofthetotalCTCsidentifiedbyGPC3,ASGPR,andCK,whichindicatesthatourselectedbiomarkerscoulddetectmoretumorcells,includingtheEpCAM-positiveandEMT-relatedcells. Ofthebiomarkersdetected,GPC3,ASGPR,andCKwereallsignificantlyexpressedinHCCcelllinesandkaryocytesfrompatientbloodsamples,suggestingthattheseproteinscouldbeusedasvaluablebiomarkersofcirculatingHCCcells.ConsideringthatGPC3andASGPRmaybespecifictoHCCfromtestsonclinicalsamples,weappliedthree-dimensionalreconstructionforfurtherobservation,revealingtheirlocalization.Additionally,thesurvivalanalysisindicatedthatGPC3andASGPRexpressioncorrelatewithpatientprognoses.Toourknowledge,fewstudieshaveevaluatedbiomarkerexpressionusingasimilarworkflow.TheresultspresentedinthisstudyimprovetheunderstandingofHCCbiomarkersfromaathologicalperspectivegiventhatserumandtissuesarewidelyusedinotherclinicalexaminationsforbiomarkers.TheHCC-relatedbiomarkersmayhelptoidentifytumorcellsintheperipheralblood,whichmayberelevanttopatientprognosis[11-13,54].Inaddition,theexpressionofEMT-relatedbiomarkerssuggeststhatCTCsmayhelpelucidatemetastaticmechanisms,andtheexpressionofEpCAMsupportsitsnovelfunctionassociatedwithstemcell-likeproperties.Althoughourstudywaslimitedbythesmallsamplesizeof84individualsinthecohort,thebiomarkerstestedheremaybeusedascapturingtargetsforHCCCTCsandmayassistinthedevelopmentofeffectiveHCCtherapiesandindividuallytargetedtreatmentsbasedonbiomarkerexpressionandlocalization. CCOOMMMMEENNTTSS Background Hepatocellularcarcinoma(HCC)iscurrentlyoneofthemonmalignanttumorsintheworld.Althoughthedetectionofcirculatingtumorcells(CTCs)willprovideuswithprognosticinformationforseveraltumors,itisdifficulttoidentifyHCCCTCsbycellularsurfacebiomarkers,thusmakingitcrucialtodetecttheexpressionofbiomarkersinanovelworkflow. Researchfrontiers ThestudyofCTCsisanincreasinglyimportantfieldinseveralcancers,andmayberelevanttometastasisandrecurrence.However,thesearchforbiomarkersforHCCCTCidentificationhasnotbeenunequivocallyaddressed.Inthisstudy,theauthorsdemonstratethatasialoglycoproteinreceptor(ASGPR),glypican-3(GPC3),andcytokeratin(CK)canidentifyCTCsonkaryocyteslides.TheexpressionofGPC3andASGPRisclinicallysignificant. Innovationsandbreakthroughs RecentreportshavehighlightedtheimportanceofCTCdetection.ForHCCin particular,itisimportanttoidentifyCTCsintheblood.ThisstudyisthefirsttodetectseveralbiomarkersofHCCinasemiquantitativeimmunocytochemicalworkflow.Furthermore,theresultssuggestthatGPC3,ASGPR,andCKcanidentifyCTCsonkaryocyteslidesandthatGPC3andASGPRexpressioncorrelatestotumorprognosis. Applications ByidentifyingHCCCTCsviathebiomarkersthattheauthorsdetected,thisstudymayrepresentafuturestrategyformonitoringCTCchangesandassessingthetherapeuticprognosisofHCC. Terminology CTCsarecellsthathavedislocatedfromaprimarytumortocirculateintheblood,andtheymaybeabletoseedinanstoeametastaticfocus.Thus,theyarethoughttoberelevanttometastasisandrecurrence.However,potentialbiomarkerstoidentifyHCCCTCsmustbeidentified. Peerreview TheauthorsdetectedtheexpressionlevelsofpotentialcellularbiomarkersforHCCanddemonstratedthatGPC3,ASGPR,andCKmaybevaluablesurfacebiomarkersfortheidentificationofHCCCTCs.Furthermore,GPC3andASGPRcanbetterdistinguishHCCpatientsfromchronicHBVpatientsandhealthycontrols,andtheirexpressionisclinicallysignificant.Inaddition,theexpressionofEMT-andstemcell-relatedbiomarkersmightprovideuswithhelpfulinformationregardingmetastasisandrecurrence.TheresultsareinterestingandmayrepresentastrategyforidentifyingHCCCTCsanddevelopingeffectivetargetedtreatments. REFERENCES 1JemalA,BrayF,CenterMM,FerlayJ,WardE,FormanD.Globalcancerstatistics.CACancerJClin2011;61:69-90[PMID:21296855DOI:10.3322/caac.20107] 2InternationalAgencyforResearchonCancer.GLOBOCAN2012:Estimatedcancerincidence,mortalityandprevalenceworldwidein2012.Availablefrom:URL:http://globocan.iarc.fr/factsheet.asp 3El-SeragHB.Epidemiologyofviralhepatitisandhepatocellularcarcinoma.Gastroenterology2012;142:1264-1273.e1[PMID:22537432DOI:10.1053/j.gastro.2011.12.061] 4KongLZ,ZhaoP.ReportoncancersinChina:RetrospectivesurveyonmalignanttumorinChina.1sted.Beijing:People’sMedicalPublishingHouse,2010:37-38 5MaWJ,WangHY,TengLS.Correlationanalysisofpreoperativeserumrotein(AFP)levelandprognosisofhepatocellularcarcinoma(HCC)afterhepatectomy.WorldJSurgOncol2013;11:212[PMID:23981851DOI:10.1186/1477-7819-11-212] 6ShahSA,ClearySP,WeiAC,YangI,TaylorBR,HemmingAW,LangerB,GrantDR,GreigPD,GallingerS.Recurrenceafterliverresectionforhepatocellularcarcinoma:riskfactors,treatment,andes.Surgery2007;141:330-339[PMID:17349844DOI:10.1016/j.surg.2006.06.028] 7CapurroM,WanlessIR,ShermanM,DeboerG,ShiW,MiyoshiE,FilmusJ.Glypican-3:anovelserumandhistochemicalmarkerforhepatocellularcarcinoma.Gastroenterology2003;125:89-97[PMID:12851874DOI:10.1016/S0016-5085(03)00689-9] 8ZhangL,LiuH,SunL,LiN,DingH,ZhengJ.Glypican-3asapotentialdifferentialdiagnosismarkerforhepatocellularcarcinoma:Atissuemicroarray-basedstudy.ActaHistochem2012;114:547-552[PMID:22119409DOI:10.1016/j.acthis.2011.10.003] 9TomodaT,NousoK,MiyaharaK,KobayashiS,KinugasaH,ToyosawaJ,HagiharaH,KuwakiK,OnishiH,NakamuraS,IkedaF,MiyakeY,ShirahaH,TakakiA,YamamotoK.Prognoticimpactofserumfollistatininpatientswithhepatocellularcarcinoma.JGastroenterolHepatol2013;28:1391-1396[PMID:23432377DOI:10.1111/jgh.12167] 10YangH,ChenGD,FangF,LiuZ,LauSH,ZhangJF,LauWY,YangLY.Dickkopf-1:asadiagnosticandprognosticserummarkerforearlyhepatocellularcarcinoma.IntJBiolMarkers2013;28:286-297[PMID:23568769DOI:10.5301/jbm.5000015] WJG| 5836 May21,2014|Volume20|Issue19| MuHetal.Semiquantitativeanalysisofhepatocellularcarcinomabiomarkers 11CristofanilliM,BuddGT,EllisMJ,eckA,MateraJ,MillerMC,ReubenJM,DoyleGV,AllardWJ,TerstappenLW,HayesDF.Circulatingtumorcells,diseaseprogression,andsurvivalinmetastaticbreastcancer.NEnglJMed2004;351:781-791[PMID:15317891DOI:10.1056/NEJMoa040766] 12deBonoJS,ScherHI,MontgomeryRB,ParkerC,MillerMC,TissingH,DoyleGV,TerstappenLW,PientaKJ,RaghavanD.Circulatingtumorcellspredictsurvivalbenefitfromtreatmentinmetastaticcastration-resistantprostatecancer.ClinCancerRes2008;14:6302-6309[PMID:18829513DOI:10.1158/1078-0432.CCR-08-0872] 13CohenSJ,PuntCJ,IannottiN,SaidmanBH,SabbathKD,GabrailNY,PicusJ,MorseM,MitchellE,MillerMC,DoyleGV,TissingH,TerstappenLW,MeropolNJ.Relationshipofcirculatingtumorcellstotumorresponse,progression-freesurvival,andoverallsurvivalinpatientswithmetastaticcolorectalcancer.JClinOncol2008;26:3213-3221[PMID:18591556DOI:10.1200/JCO.2007.15.8923] 14FoodandDrugAdministration.510(k)substantialequivalencedetermination.Availablefrom:URL:essdata.fda.gov/cdrh_docs/reviews/K040898.pdf 15ChinaFoodandDrugAdministration.Datasearchforinstruments.Availablefrom:URL:/datasearch/face3/base.jsp 16YamashitaT,uesM,WangW,KimJW,YeQ,JiaH,BudhuA,tiKA,ChenY,QinLX,TangZY,WangXW.EpCAMandroteinexpressiondefinesnovelprognosticsubtypesofhepatocellularcarcinoma.CancerRes2008;68:1451-1461[PMID:18316609DOI:10.1158/0008-5472.CAN-07-6013] 17KalluriR,WeinbergRA.Thebasicsofepithelial-mesenchymaltransition.JClinInvest2009;119:1420-1428[PMID:19487818DOI:10.1172/JCI39104] 18VonaG,EstepaL,BéroudC,DamotteD,CapronF,NalpasB,MineurA,FrancoD,LacourB,PolS,BréchotC,PaterliniBréchotP.Impactofcytomorphologicaldetectionofcirculatingtumorcellsinpatientswithlivercancer.Hepatology2004;39:792-797[PMID:14999698] 19XuW,CaoL,ChenL,LiJ,ZhangXF,QianHH,KangXY,ZhangY,LiaoJ,ShiLH,YangYF,WuMC,YinZF.Isolationofcirculatingtumorcellsinpatientswithhepatocellularcarcinomausinganovelcellseparationstrategy.ClinCancerRes2011;17:3783-3793[PMID:21527564DOI:10.1158/1078-0432.CCR-10-0498] 20KakehashiA,KatoA,InoueM,IshiiN,OkazakiE,WeiM,TachibanaT,WanibuchiH.Cytokeratin8/18asanewmarkerofmouseliverpreneoplasticlesions.ToxicolApplPharmacol2010;242:47-55[PMID:19796649DOI:10.1016/j.taap.2009.09.013] 21TsuchiyaK,KomutaM,YasuiY,TamakiN,HosokawaT,UedaK,KuzuyaT,ItakuraJ,NakanishiH,TakahashiY,KurosakiM,AsahinaY,EnomotoN,SakamotoM,IzumiN.Expressionofkeratin19isrelatedtohighrecurrenceofhepatocellularcarcinomaafterradiofrequencyablation.Oncology2011;80:278-288[PMID:21734420DOI:10.1159/000328448] 22ZhangX,LinSM,ChenTY,LiuM,YeF,ChenYR,ShiL,HeYL,WuLX,ZhengSQ,ZhaoYR,ZhangSL.AsialoglycoproteinreceptorinteractswiththepreS1domainofhepatitisBvirusinvivoandinvitro.ArchVirol2011;156:637-645[PMID:21207081DOI:10.1007/s00705-010-0903-x] 23TerrisB,CavardC,PerretC.EpCAM,anewmarkerforcancerstemcellsinhepatocellularcarcinoma.JHepatol2010;52:280-281[PMID:20006402DOI:10.1016/j.jhep.2009.10.026] 24YuX,LinY,YanX,TianQ,LiL,LinEH.CD133,StemCells,andCancerStemCells:MythorReality?
CurrColorectalCancerRep2011;7:253-259[PMID:22131911] 25LiM,ZhangB,SunB,WangX,BanX,SunT,LiuZ,ZhaoX.Anovelfunctionforvimentin:thepotentialbiomarkerforpredictingmelanomahematogenousmetastasis.JExpClinCancerRes2010;29:109[PMID:20701774DOI:10.1186/1756 -9966-29-109]26ErdbrueggerU,HaubitzM,WoywodtA.Circulatingen- dothelialcells:anovelmarkerofendothelialdamage.ClinChimActa2006;373:17-26[PMID:16836991DOI:10.1016/a.2006.05.016]27ChafferCL,WeinbergRA.Aperspectiveoncancercellmetastasis.Science2011;331:1559-1564[PMID:21436443DOI:10.1126/science.1203543]28YamashitaT,JiJ,BudhuA,uesM,YangW,WangHY,JiaH,YeQ,QinLX,WauthierE,ReidLM,MinatoH,HondaM,KanekoS,TangZY,WangXW.EpCA