QingqingHe^a,1,?,BoPeng^b,1,DayongZhuang^a,LeiXiao^a,Luming  Zheng^a,ZiyiFan^a,JianZhu^a,BenmeiXu^b,ChengXu^b,Jiangman  Zhao^b,LimingWu^b,PengZhou^a,LeiHou^a,FangYu^a and  GuoweiZhao^a aDepartmentofThyroidandBreastSurgery,JinanMilitaryGeneralHospitalofPLA,  Jinan,Shandong,China ^bZhangjiangCenterforTranslationalMedicine,Shanghai,China

Abstract.

BACKGROUND:Themolecularclassi?cationofbreastcancermainlyfocusesonestrogenreceptor(ER),Progesteronereceptor(PgR),and  HumanEpidermalGrowthFactorReceptor2(HER2/Neu)statusdetectedbyimmunohistochemistry(IHC)analysis.The  β-tubulinisotypeIII(TUBB3)genewasthoughttobeaMarkeroftaxaneresistanceorcanceraggressiveness.

METHODS:Toevaluatetheclinicopathologicalsigni?canceofTUBB3expressioninbreastcancerpatients,wemeasured  TUBB3mRNAlevelsin92breastcancerpatientsbyQuantitativereversetranscription-polymerasechainreaction  (qRT-PCR),andexaminedtheircorrelationwithER,PgR,andHER2statusdetectedbyIHC.

RESULTS:Weobservedasigni?cantpositivecorrelationbetweentheTUBB3mRNAexpressionandtheimmunohistochemicalpositivityof  bothPgR(p=0.000)andHER2(p=0.001).Inaddition,TUBB3mRNAexpressionwasassociatedwithlymphnodes  status(P=0.008)andtumorstages(0.029),butnocorrelationwasfoundwithotherclinicopathologicalfeatures,suchas  age,pathohistologicalgradesandtumorsize.

CONCLUSIONS:Inconclusion,TUBB3expressioncorrelatedsigni?cantlywithmolecularmarkersofbreastcancer,suchas  PgRandHER2,suggestingthatTUBB3mRNAlevelfacilitatetheidenti?cationofasubsetofpatientswhorespondto  Taxanetreatmentinadditiontohormonaltherapyandtrastuzumab.

Keywords:Breastcancer,ER,PgR,HER2,β-tubulinisotypeIII

1.Introduction

Breastcanceristhemostfrequentlydiagnosedcancerinwomeninwesterncountries,anditisthesecondmostcommon  causeofcancer-relateddeaths[1].Despiterecentimprovementsinthediagnosisandmanagementofearlydisease,  approximately50%ofwomenwithbreastcancerwilldevelopdistantmetastases[2].

Breastcancerisaheterogeneousdiseasethatcanbeclassi?edintosubgroupsonthebasisofhormonereceptorstatus,  HER2expressionlevels,andgeneexpressionpro?ling[3–5].Patientsofsomebreastcancersubgroupsmayhavehigh  responseratestospeci?cchemotherapeuticdrugs,whereasothersmayderivearelativelysmallbene?tfromdrugs.At  thesametimetheyhavetobeexposedtotreatment-relatedtoxicity[6–10].Thisunderscorestheneedforpredictive  biomarkersthatcanfacilitatetheselectionofoptimaltreatmentoralternativeregimenforpatientsofbreast  cancerofvarioussubclasses.Accordingly,predictivebiomarkersofferthepotentialtoimprovethebene?t:risk  ratioofagiventherapeuticagent.

AmongthoseputativemarkersisthestructuralproteinclassIIIβ-tubulin(TUBB3)[11–13].TheTUBB3genewasinitially  thoughttoencodeaneuron-speci?cprotein[14–16]andtobeamarkeroftaxaneresistanceorcanceraggressiveness[17].High  expressionofTUBB3correlateswithlowresponseratesinpatientstreatedwithtaxane-or  vinca-alkaloid-containingregimensandwithreducedsurvivalinpatientswith  nonsmallcelllungcancer (NSCLC) [18,19],breast[20],orovariancancer[21].Althoughoriginallyidenti?edasa  mechanismofdrugresistancetotaxanes[22],recentstudieshaveshownthatTUBB3isinvolvedinanadaptiveresponsetolow  oxygenlevelsandpoornutrientsupplyinagrowingnumberofsolidtumors[17,23].Thisexplainstheinvolvementof  TUBB3indrugresistanceindependentofwhetherthediseaseistreatedwitharegimenthatincludesamicrotubule  targetingagentornot[24].

Clinically,thesigni?canceofalterationofTUBB3expressionhasbeenclearlyestablished.Neverthelessthe  relationshipbetweenTUBB3andhormonereceptors,HER2hasnotbeenveryclearinbreastcancer.Currently,ER,PgR,  andHER2/neustatusaremolecularmarkersusedintheroutinetreatmentofbreastcancer.ThepresenceorabsenceofERand  PgRarevaluableprognosticfactorsprovidingpredictivevalueastothepotentialbene?tsfromhormonaltherapy.  Approximately70%ofmetastaticbreasttumorsareERand/orPgRpositive,andthesepatientstendtohaveagreaterchance  ofeffectivetumorresponseandlongeroverallsurvivalthanpatientswithER/PgR-negativetumorssincetheycanbe  targetedwithtreatmentssuchastamoxifen,aselectiveERmodulator[25–27].Furthermore,patientswith  ER/PgR-positiveearlybreastcancerhaveareducedriskofrecurrenceanddeathfollowingadjuvanthormonaltherapy,whereas  patientswithER/PgR-negativediseasederiveminimalbene?tfromthesetreatments[28].Theepidermalgrowthfactor  receptorHER2/neuisoverexpressedorampli?edinapproximately30%ofbreastcancers,anditspresenceis  associatedwithaworseprognosis[29].Thehumanmonoclonalantibodytrastuzumabwasdevelopedto  bindtheHER2/neureceptorandblockitsactivity[30].AssessmentofHER-2statusalsoisimportantforpredictingresponse  tootherspeci?cchemotherapyregimens.

Inthepresentstudy,westudiedtherelationshipbetweenTUBB3expressionandhormonereceptorsandHER2statusin  breastcancerpatients.

2.Materialsandmethods

2.1.Patients’characteristics

All92enrolledpatientswerecasesofprimaryoperablebreastcancer.Patients’ clinicalhistorydatawasacquiredfrom  the?lesoftheDepartmentofThyroidandBreastSurgery,JinanMilitaryGeneralHospital,Jinan,China.The  patients’agerangedfrom21to74years.Thehistologicaldiagnosisoftheformalin?xed,paraf?nembeddedsections  weredoneandcon?rmedbythreeexperiencedpathologists.Thediagnosisreadas84invasiveductalcarcinomas  (91.3%),1invasivelobularcarcinomas(1.1%),6othertypescarcinoma(6.5%),and1unknownhistologicaltypes  (1.1%).Multipleclinicopathologicalandmolecularcharacteristicswereobtained,includingage,tumorsize,  histologicaltype,histologicalgrade,lymphnodestatus,TNMstage,tumormolecularsubtype(asde?nedbythe  immunohistochemicalexpressionofER,PgR,andHER2).

Theexperimentswereundertakenwiththeunderstandingandwrittenconsentofeachsubject,andthatthestudy  conformswithTheCodeofEthicsoftheWorldMedicalAssociation(DeclarationofHelsinki),printedintheBritish  MedicalJournal(18July1964).Thisstudyhadbeencarriedoutinaccordancewiththeethicalstandardofethicscommittee  ofJinanMilitaryGeneralHospital,Jinan,China.

2.2.Immunohistochemistry(IHC)analysisforER,PgRandHER2proteins

Immunohistochemistry(IHC)wasperformedon3μm-thicktissuesectionspreparedfromformalin-?xed,  paraf?n-embeddedtissue.SectionswerestainedbythestandardmethodusingantibodiesagainstER,PgRandHER-2according  tothemanufacturers’instructions.Brie?y,3-μm-thicksectionswerecutfromparaf?nblockscontainingrepresentative  tumorsamples.Paraf?nsectionsweredewaxedinxylene,rehydratedthroughaseriesofgradedalcohols,placedin10  mMcitratebufferandsubmittedtoheatretrievalusingavaporlockfor40min.Afterheating,theslideswereallowedto  cooltoroomtemperatureandbrie?ywashedwithTris-bufferedsaline.Endogenousperoxidaseactivitywasblockedwith  3%hydrogenperoxideinmethanolfor5min.Normalserumwasusedfor30mininordertoblocknon-speci?cimmunoassaying.  Immunohistochemicalstainingwasperformedusinganavidin-biotinperoxidasesystem.FollowingwashesinPBS,  biotinylateduniversalsecondaryantibodywereappliedfor30min.Thesectionswereincubatedwiththe  avidin-biotincomplexreagentfor30minanddevelopedwith3,3-diaminobenzidinetetrahydrochloride(DAB)in  phosphate-bufferedsaline,pH7.5,containing0.036%hydrogenperoxidefor5min.LightMayer’shematoxylinwas  appliedasacounterstain.TheslideswerethendehydratedinaseriesofethanolandmountedwithPermount.Breast  carcinomatissuesconsistentlyshowingER,PgRorHER2ampli?cationandstrongproteinexpressionby  immunohistochemicalanalysiswasusedasapositivecontrolsample.Normalbreasttissuewasusedinnegativecontrolsamples.

AsemiquantitativescoringofpercentageofpositivecellswithERandPgRnuclearstainingwasused:0=nostaining,  1+=1–10%staining,2+=11–50%staining,and3+=>50%positivenuclearstain.NuclearERandPgRstainintensitywas  gradedasweak,moderate,orstrong.HER-2expressionwasscoredaccordingtothedegreeandtheproportionofmembrane  stainingaccordingtoHercepTestprotocol[31].HER2expressionwasnegativewithascoreof0or1+.Ascoreof0  wasde?nedasnostainingormembranestaininginlessthan10%oftumorcells.Ascore1+comprisedfaintor  partiallystainedmembraneinmoreof10%oftumortissue.OverexpressionofHER-2wasscoredas2+whenweaktomoderate  completemembranestainingwaspresentinmorethan10%oftumorcells.Ascoreof3+wasinterpretedasstrong,  completemembranestaininginmorethan10%ofthetumor.ThepathologicreviewswithIHCstainingfor  allthesurgicalspecimensweredoneprospectivelyandcomprehensivelybytwoexperiencedpathologistsinourhospital.

2.3.qRT-PCRfortheexpressionofTUBB3

Freshfrozenspecimensoftumorandadjacenttissueswereobtainedfrom92patients.Specimensweremicroscopically  examinedtoassessqualityandtoverifythehistopathology.Specimenswerepulverizedbypulpre?nerunderTrizol  reagent(LifeTechnologies,California,USA).TotalRNAwasextractedwithTaKaRaRNAisoReagent(TakaraBio,Inc.,  Otsu,Shiga,Japan).Quantitativereversetranscriptionpolymerasechainreaction(qRT-PCR)wasappliedforthe  assessmentofTUBB3gene.TotalRNAwerereversetranscribedwithRevertAidTMFirstStrandcDNASynthesisKit  (Fermentas,ThermoScienti?c,Massachusetts,USA)forgenerationofcDNAwithABI9700PCR(AppliedBiosystems,  Carlsbad,California,USA).Fortycyclesofnucleicacidampli?cationwereappliedusingABI7500Real-TimePCR(Applied  Biosystems,Carlsbad,California,USA),andthecyclethreshold(CT)valueofthetargetgenewasidenti?ed.CT  valueswerenormalisedbysubtractingtheCTvalueofthehousekeepinggeneGAPDHfromtheCTvalueofthetarget  gene(CT).RNAresultswerethenreportedas40-CTvalues,whichwouldcorrelateproportionallytothemRNA  expressionlevelofthetargetgene[32].ExpressionoftheTUBB3gene,aswellasthereferencegeneGAPDH,was  assessedintriplicatebyPCRusingtheSYBRgreenSysteminanABIPRISM7900HT(AppliedBiosystems,Darmstadt,  Germany).ThePrimersetsusedforampli?cationofTUBB3andGAPDHwereasfollows(5’-3’):

TUBB3ForwardPrimer:
AGTCGCCCACGTAGTTGC 
ReversePrimer:
CGCCCAGTATGAGGGAGAT
GAPDHForwardPrimer:
GCCACATCGCTCAGACACC  ReversePrimer: 
GATGGCAACAAT
ATCCACTTTACC

Intheabsenceofareliablegoldstandardandfollowingdistributionalstudies,weusedthe25thpercentileof  observedTUBB3mRNAexpressionlevelsasthresholdsforcategorizationoftumorstopositiveor  negativecases[32].Thechosencut-offswerefoundtobeclosetothenaturalcut-offsindistributionalstudies.Moreover,  cut-offsonthequartilesoffereasily interpretable, 
 reproducibleand objectiveresults.

2.4.Statisticalanalysis

Intheabsenceofareliablegoldstandardandfollowingdistributionalstudies,weusedthe25thpercentileof  observedTUBB3mRNAexpressionlevelsasthresholdsforcategorizationoftumorstopositiveor  negativecases[32].Thechosencut-offswerefoundtobeclosetothenaturalcut-offsindistributionalstudies.Moreover,  cut-offsonthequartilesoffereasilyinterpretable,reproducibleandobjectiveresults.

2.4.Statisticalanalysis

DescriptivestatisticscomparingTUBB3expressionwiththeclinicopathologicalcharacteristicswereanalyzedbythe  chi-squaretestor,whennecessary,byFisher’sexacttest.AcomputerprogrampackageStataTM(Version10.0,  StataCorp,CollegeStation,TX,USA)wasusedforallstatisticaltestingandmanagementofthedatabase,anda  signi?cantlevelof5%wasconsideredstatisticallysigni?cant.

3.Results

3.1.Patientsandtumourcharacteristics

Themainclinicopathologiccharacteristicsofthe92patientswereshowninTable1.92patients,allwomen  (agerange21–74years,meanage50.6years)wereincludedinthestudy.TheimmunohistochemicalstainingforER,PgRand  HER2proteinswereshowninFig.1.Tumorsof60(65.2%)womenwereERpositiveand30(32.6%)ERnegative,withtheremaining  twopatients’ERstatusoftumorwerenotidenti?ed.Tumorsof53(57.6%)patientswerePgRpositiveand38  (41.3%)werePgRnegative.TumorPgRstatuswasnotdeterminedinonepatient.HER-2positivestainingwas  observedintumorsof48(52.2%)patientsandnegativestainingintumorsof41(44.6%)patients,HER-2  statuscouldnotbedeterminedintheremaining3patients.Tumorsof16patientsshowedbothERandHER-2negativestaining,  tumorsofelse22patientswereERnegativebutHER-2positive.Tumorsin44patientsexpressedbothERandHER-2positive,  tumorsofadditional38patientsrevealedERpositivebutHER-2negative.

From92patients,90(97.8%)weretreatedbyendocrineand/orchemotherapy:1(1.1%)patientsweretreatedonlyby  endocrinetherapy;48(53.3%)patientsreceivedonlychemotherapy;and41(44.6%)patients  weretreatedbybothendocrinetherapyandchemotherapy.1(1.1%)patientswerenottreatedbyendocrinetherapyneitherby  chemotherapy.Treatmentdataregardingendocrineandchemotherapywerenotavailablefor1(1.1%)patients.

3.2.CorrelationbetweenTUBB3expressionandclinicalfeatures

ThedistributionofTUBB3mRNAvaluesisshowninFig.2.ThecorrelationsofTUBB3mRNAstatus(positiveor  negative)withstandardclinicopathologicalfactorsareshowninTable2.Ingeneral,negativeTUBB3mRNAexpression  wassigni?cantlyassociatedwithtumorhighermalignancystate,suchasadvancedstage(p=0.029),positivenodes  (p=0.008).Speci?cally,TNMstagesofII–IIIwereseenin84%TUBB3mRNA-negativeversus59.3%TUBB3  mRNA-positivetumors,whilepositivenodeswereobservedin45.6%vs.76.9%ofTUBB3mRNApositiveversusnegativetumors,  respectively.ButtheexpressionofTUBB3wasnotsigni?cantlyrelatedtoage(P=0.397),tumorsize(P=0.446)or  menopausecondition(p=0.474).

3.3.AssociationbetweenTUBB3expressionandER,PgR,HER2status

ThecorrelationsofTUBB3mRNAstatus(positiveornegative)withER,PgRandHER2areshowninTable3.Ingeneral,  positive-mRNAexpressionofTUBB3genewassigni?cantlyassociatedwithPgRpositivity(P=0.000)andHER2  positivity(P=0.001).TherewasnocorrelationbetweentheexpressionofTUBB3andERstatus(P=0.246).  Speci?cally,HER2positivetumorswereobservedin66.7%TUBB3mRNA-positivetumorsversus26.9%inTUBB3  mRNA-negativetumorswhilepositiveexpressionofPgRwasobservedin72.1%TUBB3mRNA-positivetumorsversus26.9%  inTUBB3mRNA-negativetumors.

4.Discussion

OurstudyanalyzedtheassociationbetweenclinicopathologicalfeaturesandtheexpressionofTUBB3in  aseriesofbreastcancerpatients.TUBB3mRNApositiveexpressionwasassociatedwithlymphnodesstatus(P  =0.008)andtumorstages(p=0.029).ThemRNAexpressionofTUBB3inbreasttumorswasalso  observedtobeassociatedwiththepresenceofPgR(P=0.000)andHER2(P=0.001)proteins,buttherewasnocorrelation  betweentheexpressionofTUBB3andERstatus(P=0.246).

Invitrostudiesorinclinicalinvestigations,TUBB3seemstoplayacrucialroleinthedevelopmentof  chemoresistancetoantimicrotubuleagentsandbeaprognosticmarkerofdiseaseevolution.TUBB3tumorexpressioncould  thusbeconsideredasapredictivemarkerofpaclitaxelresistanceinbreastcancer[33].Furthermore,high  expressionlevelsofTUBB3havebeenshowntobeassociatedwithpoorprognosisinovariancancerpatientstreated  bypaclitaxel-basedchemotherapy[21]andincarcinomasofunknownprimarysite[19].

Here,themostinteresting?ndinginthisstudyisthatthehighexpressionofTUBB3geneissigni?cantlyrelated withPgRpositive  breasttumors.Progesteroneis  wellknownforitsabilitiestomodulatedirectlytheexpressionofgrowthfactorreceptorpathwaysanddownstreamcellcycle  regulatorygenesknownasnuclearproto-oncogenes[34].Progesteronecandirectmammaryepithelialgrowth,  differentiation,andsurvival.PgRisageneregulatoryproteinthatisdimeric.PgRstatushasanestablishedrole  inthetumorresponsetohormonaltherapy.Basedonthe?ndingthathighexpressionofβIII-tubulinwasassociated  withpaclitaxelresistanceinadvancedbreastcancer,ourresults  indicatedthatPgRpositivebreastcancerpatientsmayhavemorelikelihoodinpaclitaxelresistancebecauseofthe  co-relationshipofPgRandTUBB3expression.

ItisnoteworthythatinformationintheliteratureregardingathoroughevaluationofbothHER2andTUBB3isvery  limited.JunM[35]etal.,foundthatoverexpressionofTUBB3andampli?edHER2genespredictedgoodresponseand  favorableprogressionfreesurvivalinHER2-positivebreastcancerpatientstreatedwithpaclitaxeland  trastuzuamab.OurresultsalsoshowedthatpositiveHER2wassignificantlyrelatedwiththepresenceofTUBB3mRNA  whichwasconsistentwithGeorge’s?nding[32].However,somestudiesreportedthattherewasnosigni?cantassociation  foundbetweentubulinexpressionandHER2/neuoverexpression[36].Possiblereasonofthediscrepancycouldbedueto  diverseretrospectivedesigns,samplesize,orheterogeneityoftheeligible  patients(withdifferentnumbersofpreviouslinesoftreatment).Inthisstudy,wedidnotanalyzethecorrelationsof  TUBB3geneexpressionandoverallsurvival(OS),disease-freesurvival(DFS),wecouldnotevaluatetheprognostic  andpredictiveutilityofTUBB3geneinbreastcancerpatients.However,determinationofTUBB3statusinbreastcancer  withHER2expressionwouldhelpidentifyasubsetofpatientswhomighthavearesponsetoTaxaneinadditiontotrastuzumab.

OurresultsshowedthatTUBB3correlateswithPgRstatus,butdoesnotcorrelatewithERstatus.Thisdoesnotcontradictwiththe  factthatPRisahighlyERdependentgene,aspreviousresearcheshavenotonlyshowntheevidenceofdiscordant  expressionofPRandER,butalsoprovidedatleastfourreasonsforthisdiscordance.First,ERdoesnotalways  activatePR.ERfunctionsviatwoseparatedways,nuclearinitiatedsteroidsignaling(NISS)andmembraneinitiated  steroidsignaling(MISS).TheformerisknownasclassicalandgenomicERactivity,causingtranscriptionof  PR,whilethelatterisknownasnon-classicalandnongenomicERactivity,causingPRdownregulation.InaclassicalER-PR  pathway,ERlocatesinsidethenucleus,bindstoestrogen,recruitsacoregulatorcomplex,andregulates  transcriptionofPR.However,inanon-classicalER-PRpathway,ERislocatedoutsidethenucleus,bindstoestrogen  ontheplasmamembrane,andactivatesgrowthfactorreceptors(EGFRorHER2)andtheirdownstreamsignaling  pathways,suchasphosphoinositide3-kinase(PI3K)/Akt/mammaliantargetofrapamycinm(mTOR).Thisleadsto  downregulationofPR[37–40].FactorswhichpotentiateMISSincludehyperactivegrowthfactorsignaling,  phosporylationofER,lowlevelsofnuclearER,andsoon[41,42].Second,activationofPRbyERdependson  estrogenlevel.Previousstudieshavereportedthatlowcirculatinglevelofestrogeninpostmenopausalwomenisnot  suf?cienttoinducePRexpression[43,44].Inthepresentstudy,>40%subjectsarepostmenopausalwomen,thusthe  ER-PRpathwaymaybeinactive.Third,otherproteinsandgrowingfactorscanalsoregulatePRexpressiondirectly.For  example,Insulin-likegrowthfactor-1(IGF-1)andEGFRhavebeenprovedtoregulateexpressionofPRindependentofER  levelsoractivityintwodifferentstudies[45,46].Horwitzetal.,alsoobservedthatsome  breastcancercelllinescontinuouslyexpresshighlevelofPRindependentofestrogensorER[47].Forth,the  limitationoftestingtechniquesmayleadtofalsedeterminationofERandPRstatus.Immunohistochemistry(IHC)is  routinelyusedforhormonereceptoranalysiscurrently,thelimitationsofwhichhavebeenmentionedbyseriesof  publications[48–53].Factors,suchaspre-analyticissuesstorage,?xationmethod,  intensityofantigenretrieval,typeofantibody,lackofapositiveinternalcontrolsignal,variabilityinsystemcontrol  samples,mayhaveeffectsonIHCresults[52,53].Therefore,itisreasonablethatexpressionofTUBB3correlates  withPRstatusbutdoesnotcorrelateswithERstatus.

Insummary,ourworkhasmanifestedthesigni?canceofthecorrelationofTUBB3mRNAlevelwithPgR,HER2status.  Still,theirfunctionandrelationtootherbiomoleculesshouldbefurtherinvestigated,whiletheirprognosticand  potentiallypredictivevalueremainstobeestimated.

Acknowledgements

TheauthorsthankallstaffsofShanghaiBiotecanPharmaceuticalsCo.,Ltd.fortheirtechnicalassistanceofTUBB3  geneexpressionanalysis.ThisworkwassupportedbyShandongProvincialNatural  ScienceFoundation(China)(No.ZR2012HM072)andthePresidentFundingofJinanMilitaryGeneralHospital(No.2011M03).

References

[1]C.Bosetti,P.Bertuccio,F.Levi,F.Lucchini,E.NegriandC.LaVecchia,CancermortalityintheEuropean  Union,19702003,withajoinpointanalysis,AnnOncol19(2008),631-40.

[2]V.H.DeVita,S.;Rosenberg,SA.,Cancer:principles&practiceofoncology,Philadelphia:LippincottWilliams&  Wilkins,2008.

[3]M.CianfroccaandW.Gradishar,Newmolecularclassi?cationsofbreastcancer,CACancerJClin59(2009),  303-13.

[4]S.J.Schnitt,Classi?cationandprognosisofinvasivebreastcancer:frommorphologytomoleculartaxonomy,ModPathol  23Suppl2,S60-4.

[5]T.Sorlie,C.M.Perou,R.Tibshirani,T.Aas,S.Geisler,H.  Johnsen,T.Hastie,M.B.Eisen,M. vandeRijn,S.S.Jeffrey,T.Thorsen,H.Quist,J.C.Matese,P.O.Brown,D.Botstein,P.E.  LonningandA.L.Borresen-Dale,Geneexpressionpatternsof breastcarcinomasdistinguishtumorsubclasseswithclinical  implications,ProcNatlAcadSciUSA98(2001),10869-74.

[6]R.Rouzier,C.M.Perou,W.F.Symmans,N.Ibrahim,M.Cristofanilli,K.Anderson,K.R.Hess,J.Stec,M.Ayers,P.  Wagner,P.Morandi,C.Fan,I.Rabiul,J.S.Ross,G.N.HortobagyiandL.Pusztai,Breastcancermolecularsubtypes  responddifferentlytopreoperativechemotherapy,ClinCancerRes11(2005),5678-85.

[7]S.J.Isakoff,Triple-negativebreastcancer:roleofspeci?cchemotherapyagents,CancerJ16,53-61.

[8]N.Koshy,D.Quispe,R.Shi,R.MansourandG.V.Burton,Cisplatin-gemcitabinetherapyinmetastatic  breastcancer:Improvedoutcomeintriplenegativebreastcancerpatientscomparedtonon-triplenegativepatients,  Breast19,246-8.

[9]L.A.Carey,E.C.Dees,L.Sawyer,L.Gatti,D.T.Moore,F.Collichio,D.W.Ollila,C.I.Sartor,M.L.Grahamand  C.M.Perou,Thetriplenegative paradox:Primarytumorchemosensitivity ofbreastcancersubtypes,ClinCancerRes13  (2007),2329-34.

[10]C.Liedtke,C.Mazouni,K.R.Hess,F.Andre,A.Tordai,J.A.Mejia,W.F.Symmans,A.M.Gonzalez-Angulo,B.  Hennessy,M.Green,M.Cristofanilli,G.N.HortobagyiandL.Pusztai,Responsetoneoadjuvanttherapyandlong-term  survivalinpatientswithtriple-negativebreastcancer,JClinOncol26(2008),1275-81.

[11]S.Hasegawa,Y.Miyoshi,C.Egawa,M.Ishitobi,T.Taguchi,Y.Tamaki,M.MondenandS.Noguchi,Predictionof  responsetodocetaxelbyquantitativeanalysisofclassIandIIIbeta-tubulinisotypemRNAexpressioninhuman  breastcancers,ClinCancerRes9(2003),2992-7.

[12]K.M.Lee,D.Cao,A.Itami,P.M.Pour,R.H.Hruban,A.MaitraandM.M.Ouellette,ClassIIIbeta-tubulin,amarkerof  resistancetopaclitaxel,isoverexpressedinpancreaticductaladenocarcinomaandintraepithelialneoplasia,  Histopathology51(2007),539-46.

[13]P.SeveandC.Dumontet,IsclassIIIbeta-tubulinapredictivefactorinpatientsreceivingtubulin-binding  agents?LancetOncol9(2008),168-75.

[14]C.D.Katsetos,A.Legido,E.Perentes andS.J.Mork,ClassIIIbeta-tubulinisotype:akeycytoskeletalproteinatthe  crossroadsofdevelopmentalneurobiologyandtumorneuropathology,JChildNeurol18(2003),851-66;discussion  867.

[15]S.S.Easter,Jr.,L.S.RossandA.Frankfurter,Initialtractformationinthemousebrain,JNeurosci13(1993),  285-99.

[16]V.Kukharskyy,V.Sulimenko,L.Macurek,T.Sulimenko,E.DraberovaandP.Draber,Complexesofgamma-tubulin  withnonreceptorproteintyrosinekinasesSrcandFynindifferentiatingP19embryonalcarcinomacells,ExpCellRes  298(2004),218-28.

[17]G.Raspaglio,F.Filippetti,S.Prislei,R.Penci,I.DeMaria,L.Cicchillitti,S.Mozzetti,G.ScambiaandC.  Ferlini,HypoxiainducesclassIIIbeta-tubulingeneexpressionbyHIF-1alphabindingtoits3’?ankingregion,Gene   409(2008),100-8.

[18]P.Seve,T.Reiman,R.Lai,J.Hanson,C.Santos,L.Johnson,L.Dabbagh,M.Sawyer,C.DumontetandJ.R.Mackey,  ClassIIIbeta-tubulinisamarkerofpaclitaxelresistanceincarcinomasofunknownprimarysite,CancerChemother  Pharmacol60(2007),27-34.

[19]P.Seve,S.Isaac,O.Tredan,P.J.Souquet,Y.Pacheco,M.Perol,L.Lafanechere,A.Penet,E.L.PeillerandC.  Dumontet,ExpressionofclassIII{beta}-tubulinispredictiveofpatientoutcomeinpatientswithnon-smallcell  lungcancerreceivingvinorelbine-basedchemotherapy,ClinCancerRes11(2005),5481-6.

[20]C.Bernard-Marty,I.Treilleux,C.Dumontet,F.Cardoso,A.Fellous,D.Gancberg,M.C.Bissery,M.Paesmans,D.  Larsimont,M.J.PiccartandA.DiLeo,Microtubule-associatedparametersaspredictivemarkersofdocetaxelactivity  inadvancedbreastcancerpatients:resultsofapilotstudy,ClinBreastCancer3(2002),341-5.

[21]G.Ferrandina,G.F.Zannoni,E.Martinelli,A.Paglia,V.Gallotta,S.Mozzetti,G.ScambiaandC.Ferlini,  ClassIIIbetatubulinoverexpressionisamarkerofpoorclinicaloutcomeinadvancedovariancancerpatients,ClinCancer  Res12(2006),2774-9.

[22]W.B.Derry,L.Wilson,I.A.Khan,R.F.LuduenaandM.A.Jordan,Taxoldifferentiallymodulatesthedynamicsof  microtubulesassembledfromunfractionatedandpuri?edbetatubulinisotypes,Biochemistry36(1997),3554-62.

[23]G.Raspaglio,I.DeMaria,F.Filippetti,E.Martinelli,G.F.Zannoni,S.Prislei,G.Ferrandina,S.Shahabi,G.  ScambiaandC.Ferlini,HuRregulatesbeta-tubulinisotypeexpressioninovariancancer,CancerRes70,  5891-900.

[24]J.A.McCarroll,P.P.Gan,M.LiuandM.Kavallaris,betaIIItubulinisamultifunctionalproteininvolvedindrug  sensitivityandtumorigenesisinnon-smallcelllungcancer,CancerRes70,4995-5003.

[25]C.K.Osborne,Tamoxifeninthetreatmentofbreastcancer,NEnglJMed339(1998),1609-18.

[26]C.K.Osborne,H.ZhaoandS.A.Fuqua,Selectiveestrogenreceptormodulators:structure,function,andclinical  use,JClinOncol18(2000),3172-86.

[27]B.Fisher,C.Redmond,E.R.FisherandR.Caplan,Relativeworthofestrogenorprogesteronereceptorand  pathologiccharacteristicsofdifferentiationasindicatorsofprognosisinnodenegativebreastcancerpatients:  ?ndingsfromNationalSurgicalAdjuvantBreastandBowelProjectProtocolB-06,JClinOncol6(1988),1076-87.

[28]Tamoxifenforearlybreastcancer:anoverviewoftherandomisedtrials.EarlyBreastCancerTrialists’  CollaborativeGroup,Lancet351(1998),1451-67.

[29]D.J.Slamon,G.M.Clark,S.G.Wong,W.J.Levin,A.UllrichandW.L.McGuire,Humanbreastcancer:correlationof  relapseandsurvivalwithampli?cationoftheHER-2/neuoncogene,Science235(1987),177-82.

[30]P.Carter,L.Presta,C.M.Gorman,J.B.Ridgway,D.Henner,W.L.Wong,A.M.Rowland,C.Kotts,M.E.Carverand  H.M.Shepard,Humanizationofananti-p185HER2antibodyforhumancancertherapy,ProcNatlAcadSciUSA89  (1992),4285-9.

[31]T.W.Jacobs,A.M.Gown,H.Yaziji,M.J.BarnesandS.J.Schnitt,Speci?cityofHercepTestindetermining  HER-2/neustatusofbreastcancersusingtheUnitedStatesFoodandDrugAdministration-approvedscoringsystem,J  ClinOncol17(1999),1983-7.

[32]G.Pentheroudakis,A.Batistatou,K.T.Kalogeras,R.Kronenwett,R.M.Wirtz,E.Bournakis,A.G.Eleftheraki,D.  Pectasides,M.Bobos,I.Papaspirou,S.Kamina,H.Gogas,A.K.Koutras,N.PavlidisandG.Fountzilas,Prognostic  utilityofbeta-tubulinisotypeIIIandcorrelationswithothermolecularandclinicopathologicalvariablesin  patientswithearlybreastcancer:atranslationalHellenicCooperativeOncologyGroup(HeCOG)study,BreastCancer  ResTreat127,179-93.

[33]S.Tommasi,A.Mangia,R.Lacalamita, A.Bellizzi,V.Fedele,A.Chiriatti,C.Thomssen,N.Kendzierski,A.Latorre,V.  Lorusso,F.Schittulli,F.Zito,M.KavallarisandA.Paradiso,Cytoskeletonandpaclitaxelsensitivityinbreast  cancer:theroleofbeta-tubulins,IntJCancer120(2007),2078-85.

[34]M.Tewari,A.KrishnamurthyandH.S.Shukla,Predictivemarkersofresponsetoneoadjuvantchemotherapyin  breastcancer,SurgOncol17(2008),301-11.

[35]M.Jung,J.S.Koo,Y.W.Moon,B.W.Park,S.I.Kim,S.Park,S.H.Lee,S.Hong,S.Y.Rha,H.C.Chung,J.H.Kim  andJ.Sohn,OverexpressionofclassIIIbetatubulinandampli?edHER2genepredictgoodresponsetopaclitaxeland  trastuzumabtherapy,PLoSOne7,e45127.

[36]Y.Wang,J.A.Sparano,S.Fineberg,L.Stead,J.Sunkara,S.B.HorwitzandH.M.McDaid,Highexpressionofclass  IIIbeta-tubulinpredictsgoodresponsetoneoadjuvant taxaneanddoxorubicin/cyclophosphamide-basedchemotherapy  inestrogenreceptor-negativebreastcancer,ClinBreastCancer13,103-8.

[37]X.Cui,R.Schiff,G.Arpino,C.K.OsborneandA.V.Lee,Biologyofprogesteronereceptorlossinbreastcancer  anditsimplicationsforendocrinetherapy,JClinOncol23(2005),7721-35.

[38]I.Nemere,R.J.PietrasandP.F.Blackmore,Membranereceptorsforsteroidhormones:signaltransductionand  physiologicalsigni?cance,JCellBiochem88(2003),438-45.

[39]C.K.Osborne,J.Shou,S.MassarwehandR.Schiff, Crosstalkbetweenestrogenreceptorandgrowthfactorreceptor  pathwaysasacauseforendocrinetherapyresistanceinbreastcancer,ClinCancerRes11(2005),865s-70s.

[40]J.P.ThakkarandD.G.Mehta,Areviewofanunfavorablesubsetofbreastcancer:estrogenreceptorpositiveprogesterone  receptornegative,Oncologist16(2011),276-85.

[41]R.X.Song,R.A.McPherson,L.Adam,Y.Bao,M.Shupnik,R.KumarandR.J.Santen,LinkageofrapidestrogenactiontoMAPK  activationbyERalpha-ShcassociationandShcpathwayactivation,MolEndocrinol16(2002),116-27.

[42]Z.Zhang,B.Maier,R.J.SantenandR.X.Song,Membraneassociationofestrogenreceptoralphamediatesestrogen  effectonMAPKactivation,BiochemBiophysResCommun294(2002),926-33.

[43]N.D.Bloom,E.H.Tobin,B.SchreibmanandG.A.Degenshein,Theroleofprogesteronereceptorsinthemanagement  ofadvancedbreastcancer,Cancer45(1980),2992-7.

[44]G.A.Colditz,B.A.Rosner,W.Y.Chen,M.D.HolmesandS.E.Hankinson,Riskfactorsforbreastcanceraccordingto  estrogenandprogesteronereceptorstatus,JNatlCancerInst96(2004),218-28.

[45]X.Cui,P.Zhang,W.Deng,S.Oesterreich,Y.Lu,G.B.MillsandA.V.Lee,Insulin-likegrowthfactor-Iinhibits  progesteronereceptorexpressioninbreastcancercellsviathephosphatidylinositol3-kinase/Akt/mammalian  targetofrapamycinpathway:progesteronereceptorasapotentialindicatorofgrowthfactoractivityinbreast  cancer,MolEndocrinol17(2003),575-88.

[46]Y.Zhang,H.Su,M.Rahimi,R.TochiharaandC.Tang,EGFRvIII-inducedestrogen-independence,  tamoxifenresistancephenotypecorrelateswithPgRexpressionandmodulationofapoptoticmoleculesinbreastcancer,  IntJCancer125(2009),2021-8.

[47]K.B.Horwitz,M.B.MockusandB.A.Lessey,VariantT47Dhumanbreastcancercellswithhigh  progesterone-receptorlevelsdespiteestrogenandantiestrogenresistance,Cell28(1982),633-42.

[48]S.S.Badve,F.L.Baehner,R.P.Gray,B.H.Childs,T.Maddala, M.L.Liu,S.C.Rowley,S.Shak,E.A.Perez,L.J.Shulman,  S.Martino,N.E.Davidson,G.W.Sledge,L.J.GoldsteinandJ.A.Sparano,Estrogen-andprogesterone-receptorstatus  inECOG2197:Comparisonofimmunohistochemistrybylocalandcentrallaboratoriesandquantitativereverse  transcriptionpolymerasechainreactionbycentrallaboratory,JClinOncol26(2008),2473-81.

[49]P.L.Fitzgibbons,D.A.Murphy,M.E.Hammond,D.C.AllredandP.N.Valenstein,Recommendationsforvalidating  estrogenandprogesteronereceptorimmunohistochemistryassays,ArchPatholLabMed134(2010),930-5.

[50]G.D.Francis,M.Dimech,L.GilesandA.Hopkins,Frequencyandreliabilityofoestrogenreceptor,progesterone  receptorandHER2inbreastcarcinomadeterminedbyimmunohistochemistryinAustralasia:resultsoftheRCPAQuality  AssuranceProgram,JClinPathol60(2007),1277-83.

[51]R.D.Gelber,S.Gelber,G.InternationalBreastCancerStudyandG.BreastInternational,Facilitatingconsensus  byexaminingpatternsoftreatmenteffects,Breast18Suppl3(2009),S2-8.

[52]S.K.Mohsin,H.Weiss,T.Havighurst,G.M.Clark,M.Berardo,D.Roanhle,T.V.To,Z.Qian,R.R.LoveandD.C.  Allred,Progesteronereceptorbyimmunohistochemistryandclinicaloutcomeinbreastcancer:Avalidationstudy,Mod  Pathol17(2004),1545-54.

[53]R.L.Theriault,R.W.Carlson,C.Allred,B.O.Anderson,H.J.Burstein,S.B.Edge,W.B.Farrar,A.Forero,S.H.  Giordano,L.J.Goldstein,W.J.Gradishar,D.F.Hayes,C.A.Hudis,S.J.Isakoff,B.M.Ljung,D.A.Mankoff,P.K.  Marcom,I.A.Mayer,B.McCormick,L.J.Pierce,E.C.Reed,L.S.Schwartzberg,M.L.Smith,H.Soliman,G.Somlo,J.H.Ward,  A.C.Wolff,R.Zellars,D.A.Shead,R.KumarandN.NationalComprehensiveCancer,Breastcancer,version3.2013:  featuredupdatestotheNCCNguidelines,JNatlComprCancNetw11(2013),753-60;quiz761