58.4mail.126.com

Zhangetal.
JTranslMed(2018)16:171https://doi.
org/10.
1186/s12967-018-1544-1RESEARCHAprospectivestudyonthechangesandclinicalsignificanceofpreoperativeandpostoperativecirculatingtumorcellsinresectablegastriccancerQiyueZhang1,FeiShan2,ZiyuLi2,JingGao1,YilinLi1,LinShen1,JiafuJi2*andMingLu1*AbstractBackground:Circulatingtumorcells(CTCs)havebeensuggestedaspotentialprognosticindicatorsformultipletumors,includinggastriccancer;however,pre-andpost-operativeCTCchangesinresectablegastriccancerandpos-siblecorrelationstopost-operativerecurrencehavenotbeenevaluated.
Methods:Subjects(n=93)withresectablegastriccancerwereprospectivelyreviewedfromJuly2013toDecember2014atPekingUniversityCancerHospital.
TheproportionofCTCswereevaluatedbefore(n=93)andafter(n=63)radicaloperationusingastandardizedCellSearchsystem.
Results:CTCs≥1weremeasuredinthepre-operativebloodof31(33.
3%)patientsandinthepost-operativebloodof21patients(33.
3%).
Patientswithrelativelypoorclinicopathologicalfeatureshadmorepre-andpost-operativeCTCs.
The3-yeardisease-freesurvival(DFS)rateforpatientswithCTCs≥5/7.
5mlwassignificantlylowerthanforpatientswithCTCs4577(82.
8%)50(79.
4%)PrimarysiteaEGJ29(31.
2%)19(30.
2%)Non-EGJ64(68.
8%)44(69.
8%)DifferentiationbGood22(23.
7%)15(23.
8%)Poor71(76.
3%)48(76.
2%)LaurenIntestinal36(38.
7%)25(39.
7%)Diffuse23(24.
7%)16(25.
4%)Mixed34(36.
6%)22(34.
9%)TNMclassificationI24(25.
8%)15(23.
8%)II24(25.
8%)16(25.
4%)III45(48.
4%)32(50.
8%)DepthofpenetrationT1/T230(32.
3%)19(30.
2%)T334(36.
6%)25(39.
7%)T429(31.
2%)19(30.
2%)LymphnodeN031(33.
3%)16(25.
4%)N1/N231(33.
3%)26(41.
3%)N331(33.
3%)21(33.
3%)Lymph-vascularinvasionYes45(48.
4%)34(54.
0%)No48(51.
6%)29(46.
0%)CEAclevelbeforeoperation(ng/ml)0.
05)correla-tionbetweenCTCsandage,differentiation,lymphnodemetastasis,andCEAlevelspriortosurgery(Fig.
1e–h).
Moreover,patientsnoolderthan45years-of-age,withlowdifferentiationandlymphnodemetastasisweremorelikelytohavemoreCTCs.
Additionalfile1:TableS1depictspatientsCTCsdatapre-andpost-surgery.
Post-operativeCTCswerecorrelatedwithpatientcharacteris-ticsinamannersimilartopre-operativeCTCs(Table3).
Patientswithrelativelypoorclinicopathologicalfeatureshadmorepost-operativeCTCs,andthiswassignificantlycorrelatedwithdepthofpenetration(p0.
05;Fig.
2).
Tenpatientshad≥3CTCs/7.
5mlinpost-oper-ativesamplesandmosthadpoordifferentiation(n=8),stageIIB/III(n=10),orT3/T4(n=10)penetration.
CorrelationsofCTCsenumerationwithDFSandOSinresectablegastriccancerUntilNovember2016,medianfollow-upwas36.
4[inter-quartile33.
9–38.
7]months.
Intotal,31(33.
3%)patientsrelapsed,24(25.
8%)patientsdied.
PatientsCTCandDFSdataarepresentedinAdditionalfile2:TableS2.
Inourstudy,patientswithgastroesophagealjunc-tion,T3/T4penetration,lymphnodemetastases,CEAlevel≥5ng/mlandhigherpre-orpostoperativeCTCshadlower3-yearDFS(Table2andFig.
3a).
The3-yearDFSwas100.
0%inpatientswithstageItumors,68.
3%inpatientswithstageIItumors,44.
0%inpatientswithstageIIItumors(Table2).
AsshowninFig.
3a,patientswhohadmoreCTCshadlower3-yearDFS.
Obviously,asthethresholdincreased,the3-yearDFSofpatientswithpost-operativeCTCsaboveeachcut-offvaluedecreasedmarkedly.
Moreover,3-yearDFSinpatientswithCTCs≥5/7.
5mlwassignificantlylowerthaninpatientswithCTCs45(n=77)66.
225(32.
5%)9(11.
7%)5(6.
5%)4(5.
2%)4(5.
2%)pvalue0.
2440.
6980.
3170.
0700.
2741.
000PrimarysiteaEGJ(n=29)50.
49(31.
0%)3(10.
3%)2(6.
9%)2(6.
9%)2(6.
9%)Non-EGJ(n=64)71.
222(34.
3%)10(15.
6%)7(10.
9%)4(6.
3%)3(4.
7%)pvalue0.
0020.
8730.
7870.
8731.
0000.
635DifferentiationbGood(n=22)71.
84(18.
2%)1(4.
5%)1(4.
5%)0(0.
0%)0(0.
0%)Poor(n=71)62.
627(38.
0%)12(16.
9%)8(11.
3%)6(8.
5%)5(7.
0%)pvalue0.
1740.
0840.
2680.
6040.
3300.
335LaurenIntestinal(n=36)68.
08(22.
2%)2(5.
6%)1(2.
8%)0(0.
0%)0(0.
0%)Diffuse(n=23)67.
46(26.
1%)5(21.
7%)5(21.
7%)3(13.
0%)2(8.
7%)Mixed(n=34)60.
017(50.
0%)6(17.
6%)3(8.
8%)3(8.
8%)3(8.
8%)pvalue0.
4340.
0330.
1180.
0600.
0680.
148TNMclassificationI(n=24)100.
06(25.
0%)0(0.
0%)0(0.
0%)0(0.
0%)0(0.
0%)II(n=24)68.
38(33.
3%)4(16.
7%)2(8.
3%)1(4.
2%)1(4.
2%)III(n=45)44.
017(37.
8%)9(20.
0%)7(15.
6%)5(11.
1%)4(8.
9%)pvalue0.
0000.
5630.
0470.
1030.
3060.
443DepthofpenetrationT1/T2(n=30)90.
08(26.
7%)1(3.
3%)1(3.
3%)1(3.
3%)1(3.
3%)T3(n=34)56.
912(35.
3%)5(14.
7%)1(2.
9%)0(0.
0%)0(0.
0%)T4(n=29)47.
711(37.
9%)7(24.
1%)7(24.
1%)5(17.
2%)4(13.
8%)pvalue0.
0000.
6270.
0640.
0070.
0090.
026LymphnodeN0(n=31)96.
87(22.
6%)2(6.
5%)1(3.
2%)0(0.
0%)0(0.
0%)N1/N2(n=31)58.
49(29.
0%)3(9.
7%)2(6.
5%)2(6.
5%)2(6.
5%)N3(n=31)38.
915(48.
4%)8(25.
8%)6(19.
4%)4(12.
9%)3(9.
7%)pvalue0.
0000.
0810.
1090.
1330.
1600.
362Lymph-vascularinvasionYes(n=45)53.
020(44.
4%)9(20.
0%)6(13.
3%)3(6.
7%)3(6.
7%)No(n=48)75.
811(22.
9%)4(8.
3%)3(6.
3%)3(6.
3%)2(4.
2%)pvalue0.
0010.
0280.
1050.
3071.
0000.
671CEAclevelbeforeoperation(ng/ml)45(n=50)60.
616(32.
0%)9(18.
0%)8(16.
0%)6(12.
0%)3(6.
0%)pvalue0.
5650.
9121.
0001.
0001.
0001.
000PrimarysiteaEGJ(n=19)43.
57(36.
8%)3(15.
8%)3(15.
8%)3(15.
8%)1(5.
3%)Non-EGJ(n=44)67.
214(31.
8%)8(18.
2%)7(15.
9%)5(11.
4%)3(6.
8%)pvalue0.
0010.
6981.
0001.
0000.
9431.
000DifferentiationbGood(n=15)66.
05(33.
3%)3(20.
0%)2(13.
3%)1(6.
7%)0(0.
0%)Poor(n=48)57.
816(33.
3%)8(16.
7%)8(16.
7%)7(14.
6%)4(8.
3%)pvalue0.
2441.
0001.
0001.
0000.
7190.
564LaurenIntestinal(n=25)62.
08(32.
0%)5(20.
0%)4(16.
0%)2(8.
0%)0(0.
0%)Diffuse(n=16)65.
74(25.
0%)3(18.
8%)3(18.
8%)3(18.
8%)2(12.
5%)Mixed(n=22)52.
49(40.
9%)3(13.
6%)3(13.
6%)3(13.
6%)2(9.
1%)pvalue0.
1150.
5800.
8440.
9130.
5050.
167TNMclassificationI(n=15)100.
02(13.
3%)0(0.
0%)0(0.
0%)0(0.
0%)0(0.
0%)II(n=16)54.
75(31.
3%)4(25.
0%)3(18.
8%)2(12.
5%)1(6.
3%)III(n=32)43.
114(43.
8%)7(21.
9%)7(21.
9%)6(18.
8%)3(9.
4%)pvalue0.
0620.
1170.
1060.
1610.
2490.
798DepthofpenetrationT1/T2(n=19)89.
52(10.
5%)0(0.
0%)0(0.
0%)0(0.
0%)0(0.
0%)T3(n=25)51.
210(40.
0%)4(16.
0%)3(12.
0%)3(12.
0%)1(4.
0%)T4(n=19)41.
49(47.
4%)7(36.
8%)7(36.
8%)5(26.
3%)3(15.
8%)pvalue0.
0000.
0360.
0080.
0050.
0430.
178LymphnodeN0(n=16)100.
04(25.
0%)2(12.
5%)1(6.
3%)0(0.
0%)0(0.
0%)N1/N2(n=26)54.
58(30.
8%)6(23.
1%)6(23.
1%)5(19.
2%)2(7.
7%)N3(n=21)33.
39(42.
9%)3(14.
3%)3(14.
3%)3(14.
3%)2(9.
5%)pvalue0.
0010.
4880.
7610.
3750.
1850.
663Lymph-vascularinvasionYes(n=34)49.
810(29.
4%)5(14.
7%)5(14.
7%)3(8.
8%)2(5.
9%)No(n=29)71.
411(37.
9%)6(20.
7%)5(17.
2%)5(17.
2%)2(6.
9%)pvalue0.
0020.
4750.
5331.
0000.
4531.
000CEAclevelbeforeoperation(ng/ml)<5(n=51)62.
618(35.
3%)9(17.
6%)9(17.
6%)8(15.
7%)4(7.
8%)≥5(n=12)48.
63(25.
0%)2(16.
7%)1(8.
3%)0(0.
0%)0(0.
0%)pvalue0.
0460.
7341.
0000.
7220.
3241.
000Page9of12Zhangetal.
JTranslMed(2018)16:171Fig.
33-yearDFSwaslowerinpatientswhohadmoreCTCs(a),thefilledblackcolumnsrepresentthe3-yearDFSofpatientswhohadCTCsnomorethanthecut-offvalue,andtheemptywhitecolumnsrepresent3-yearDFSofpatientswhohadCTCmorethanthecutoffvalue.
Patientswhohave≥5CTCs/7.
5mlbloodpre-operativelyhadshorterDFS(b,d)andOS(c,e)Page10of12Zhangetal.
JTranslMed(2018)16:171combineothermethodsofmolecularanalysistobet-teridentifytheprognosticvalueofCTCsinresect-ablegastriccancer.
Third,numerousstudiesagreethatintra-operativeCTCssheddingoccurswithtumormanipulation[24].
However,ourdatasuggestedthatpost-operativeCTCenumerationwithin1weekaftersurgerywasnotincreasedbysurgery.
Althoughintra-operativeCTCswerenotdetectedinourstudy,stud-iesofothertumorsreportedthatCTCdetectionrateandnotCTCenumerationincreasedinintra-operativesamples[24].
Moreover,studiesalsodemonstratedthatincreasedintra-operativeCTCsnormalizepost-operativelywithindaystoweeks[24].
Fourth,wecol-lectedpost-operativesampleswithin1weekaftersurgery(2–7daysaftersurgery).
WearealsointerestedinhowCTCscanchangewhenthelengthofobser-vationisincreasedtotwoor3weeks,whichwewilladdressinfollow-upstudies.
Interestingly,KragandcolleaguesreportedthatCTCsrapidlydeclinedduring48hpost-operativelyinmostpatientswithoperableFig.
4Amonghematogenousmetastasessubjects,patientswhohadpost-operativeCTCs≥2/7.
5ml(a)andpatientswhohadmarkedlyincreasedCTCsaftersurgery(b)hadmarkedlyshorterDFSPage11of12Zhangetal.
JTranslMed(2018)16:171breastcancer[25].
Animalstudiesalsofoundthesimi-larphenomenon[26].
Kragetal.
alsodemonstratedthat30%patientshaddetectableCTCspersistentlyupto14daysaftersurgery.
Theseresultsareconsist-entwiththenotionthatonlyaminorityofCTCscansurviveandhavecapacitytobeclonogenic.
Inaddi-tion,thoseauthorsfurtherputforwardaview,whichissimilartous,namely,thepersistentpresenceofcancercellsafterradicalsurgeryisastrongindicatorofrecurrence.
Therefore,wehypothesizethatCTCsarestablefrom2daysto2weeks,andCTCsshouldbemonitoredbeforeandaftersurgerytopredictcan-cerrecurrenceandfordiseasestagingandtreatmentmanagement.
ConclusionInsummary,weanalyzedtheutilityofpost-operativeCTCsinresectablegastriccancer.
Inaddition,tothebestofourknowledge,thisisthefirstprospectivestudythathasanalyzedthecorrelationsbetweenpost-operativeCTCsandchangesinCTCsduringsurgerywithclinicopathologicalcharacteristics,prognosis,andrecurrencepatterns.
Wefoundthatpost-opera-tiveCTCsmightbeamoredirectandefficientrecur-rencemarkerthanpre-operativeCTCsandincreasedpost-operativeCTCsmightbecorrelatedwithhema-togenousrecurrence.
Monitoringthepost-operativeCTCsandchangesinCTCsduringsurgeryisofgreatimportanceforbetterpost-operativeadjuvanttreat-mentdecisions.
AbbreviationsCTC:circulatingtumorcell;CEA:carcinoembryonicantigen;OS:overallsurvival;EpCAM:epithelialcelladhesionmolecule;CK:cytokeratin;DAPI:4′,6-diamidino-2-phenylindole;DFS:disease-freesurvival;3-yearDFS:3-yeardisease-freesurvivalrate.
Authors'contributionsMLdesignedthestudy.
QZandFSsummarizedthedataanddraftedthemanuscript.
Alloftheauthorscontributedtotherevisionofmanuscript.
Allauthorsreadandapprovedthefinalmanuscript.
AdditionalfilesAdditionalfile1:TableS1.
PatientcharacteristicsandCTCnumbersin63patients.
Additionalfile2:TableS2.
PatientcharacteristicsandCTCnumbersin31relapsedpatients.
Additionalfile3:FigureS1.
Correlationsofpre-operativeCTCswithpatientrecurrencepatterns.
Additionalfile4:FigureS2.
Patientswhohave≥4CTCs/7.
5mlbloodpre-operativelyhadshorterDFS(a,c)andOS(b,d).
AcknowledgementsWethankLetPubforitslinguisticassistanceduringthepreparationofthismanuscript.
CompetinginterestsTheauthorsdeclarethattheyhavenocompetinginterests.
AvailabilityofdataThedatasetsanalyzedwithinthecurrentstudyareavailablefromthecor-respondingauthoruponreasonablerequest.
ConsentforpublicationNotapplicable.
EthicsapprovalandconsenttoparticipateAlloftheproceduresfollowedwereinaccordancewiththeethicalstandardsoftheresponsiblecommitteeonhumanexperimentation(institutionalandnational)andwiththeHelsinkiDeclarationof1964andlaterversions.
Informedconsentoranappropriatesubstitutewasobtainedfromallofthepatientspriortoinclusioninthestudy.
FundingThisworkwasfundedbyCapitalHealthResearchandDevelopment(No.
2014-4-1023),theBeijingNaturalScienceFoundation(7161002),andtheBeijingMunicipalAdministrationofHospitalClinicalMedicineDevelopmentofSpecialFundingSupport(ZYLX201701).
Publisher'sNoteSpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpub-lishedmapsandinstitutionalaffiliations.
Received:26January2018Accepted:7June2018References1.
SiegelR,MaJ,ZouZ,JemalA.
Cancerstatistics,2014.
CACancerJClin.
2014;64:9–29.
2.
SasakoM,SakuramotoS,KataiH,etal.
Five-yearoutcomesofarandomizedphaseIIItrialcomparingadjuvantchemotherapywithS-1versussurgeryaloneinstageIIorIIIgastriccancer.
JClinOncol.
2011;29:4387–93.
3.
ParkSH,SohnTS,LeeJ,etal.
PhaseIIItrialtocompareadjuvantchemo-therapywithcapecitabineandcisplatinversusconcurrentchemoradio-therapyingastriccancer:finalreportoftheadjuvantchemoradiotherapyinstomachtumorstrial,includingsurvivalandsubsetanalyses.
JClinOncol.
2015;33:3130–6.
4.
SunZ,WangZN,ZhuZ,etal.
EvaluationoftheseventheditionofAmericanJointCommitteeonCancerTNMstagingsystemforgastriccancer:resultsfromaChinesemonoinstitutionalstudy.
AnnSurgOncol.
2012;19:1918–27.
5.
LiZ,ZouX,XieL,etal.
PersonalizingriskstratificationbyadditionofPAK1expressiontoTNMstaging:improvingtheaccuracyofclinicaldecisionforgastroesophagealjunctionadenocarcinoma.
IntJCancer.
2015;136:1636–45.
6.
DuffyMJ,LamerzR,HaglundC,etal.
Tumormarkersincolorectalcancer,gastriccancerandgastrointestinalstromalcancers:europeangroupontumormarkers2014guidelinesupdate.
IntJCancer.
2014;134:2513–22.
7.
CristescuR,LeeJ,NebozhynM,etal.
Molecularanalysisofgastriccanceridentifiessubtypesassociatedwithdistinctclinicaloutcomes.
NatMed.
2015;21:449–56.
8.
YooCH,NohSH,ShinDW,ChoiSH,MinJS.
Recurrencefollowingcurativeresectionforgastriccarcinoma.
BrJSurg.
2000;87:236–42.
9.
ReymondN,d'guaBB,RidleyAJ.
Crossingtheendothelialbarrierduringmetastasis.
NatRevCancer.
2013;13:858–70.
10.
AcetoN,BardiaA,MiyamotoDT,etal.
Circulatingtumorcellclus-tersareoligoclonalprecursorsofbreastcancermetastasis.
Cell.
2014;158:1110–22.
Page12of12Zhangetal.
JTranslMed(2018)16:17111.
CristofanilliM,BuddGT,EllisMJ,etal.
Circulatingtumorcells,diseaseprogression,andsurvivalinmetastaticbreastcancer.
NEnglJMed.
2004;351:781–91.
12.
CohenSJ,PuntCJ,IannottiN,etal.
Relationshipofcirculatingtumorcellstotumorresponse,progression-freesurvival,andoverallsurvivalinpatientswithmetastaticcolorectalcancer.
JClinOncol.
2008;26:3213–21.
13.
deBonoJS,ScherHI,MontgomeryRB,etal.
Circulatingtumorcellspredictsurvivalbenefitfromtreatmentinmetastaticcastration-resistantprostatecancer.
ClinCancerRes.
2008;14:6302–9.
14.
KrebsMG,SloaneR,PriestL,etal.
Evaluationandprognosticsignificanceofcirculatingtumorcellsinpatientswithnon-small-celllungcancer.
JClinOncol.
2011;29:1556–63.
15.
YuM,BardiaA,AcetoN,etal.
Cancertherapy.
Exvivocultureofcirculat-ingbreasttumorcellsforindividualizedtestingofdrugsusceptibility.
Science.
2014;345:216–20.
16.
LiuMC,ShieldsPG,WarrenRD,etal.
Circulatingtumorcells:ausefulpredictoroftreatmentefficacyinmetastaticbreastcancer.
JClinOncol.
2009;27:5153–9.
17.
LiY,ZhangX,GeS,etal.
Clinicalsignificanceofphenotypingandkaryo-typingofcirculatingtumorcellsinpatientswithadvancedgastriccancer.
Oncotarget.
2014;5:6594–602.
18.
SoteloMJ,SastreJ,MaestroML,etal.
RoleofcirculatingtumorcellsasprognosticmarkerinresectedstageIIIcolorectalcancer.
AnnOncol.
2015;26:535–41.
19.
IinumaH,WatanabeT,MimoriK,etal.
Clinicalsignificanceofcirculat-ingtumorcells,includingcancerstem-likecells,inperipheralbloodforrecurrenceandprognosisinpatientswithDukes'stageBandCcolorectalcancer.
JClinOncol.
2011;29:1547–55.
20.
LucciA,HallCS,LodhiAK,etal.
Circulatingtumourcellsinnon-metastaticbreastcancer:aprospectivestudy.
LancetOncol.
2012;13:688–95.
21.
GazzanigaP,deBerardinisE,RaimondiC,etal.
Circulatingtumorcellsdetectionhasindependentprognosticimpactinhigh-risknon-muscleinvasivebladdercancer.
IntJCancer.
2014;135:1978–82.
22.
PalSK,HeM,WilsonT,etal.
Detectionandphenotypingofcirculatingtumorcellsinhigh-risklocalizedprostatecancer.
ClinGenitourinCancer.
2015;13:130–6.
23.
UenosonoY,ArigamiT,KozonoT,etal.
Clinicalsignificanceofcirculatingtumorcellsinperipheralbloodfrompatientswithgastriccancer.
Cancer.
2013;119:3984–91.
24.
KauffmanEC,LeeMJ,AlarconSV,etal.
Lackofimpactofrobotic-assistedlaparoscopicradicalprostatectomyonintraoperativelevelsofprostatecancercirculatingtumorcells.
JUrol.
2015;195:1136–42.
25.
KragDN,AshikagaT,MossTJ,etal.
Breastcancercellsintheblood:apilotstudy.
BreastJ.
1999;5:354–8.
26.
ButlerTP,GullinoPM.
Quantitationofcellsheddingintoefferentbloodofmammaryadenocarcinoma.
CancerRes.
1975;35:512–6.