在12枚阳性前哨淋巴结中国乳腺癌人群中

WorldJSurg.Sep1.[Epubaheadofprint]

PredictingNon-sentinelLymphNodeMetastasisinaChineseBreastCancerPopulationwith1-2PositiveSentinelNodes:DevelopmentandAssessmentofaNewPredictiveNomogram.

ChenJY,ChenJJ,XueJY,ChenY,LiuGY,HanQX,YangWT,ShenZZ,ShaoZM,WuJ.

DepartmentofBreastSurgery,FudanUniversityShanghaiCancerCenter,DongAnRoad,Shanghai,,China.

BACKGROUND:Wehavedevelopedanewnomogramtopredicttheprobabilityofapatientwith1-2metastaticsentinellymphnodes(SLNs)topresentfurtheraxillarydisease.

METHODS:Datawerecollectedfrompatientswhowerediagnosedwith1-2positivelymphnodesandthusunderwentaxillarylymphnodedissectionbetweenMarchandJune.Clinicalandpathologicalfeaturesofthepatientswereassessedwithmultivariablelogisticregression.TheShanghaiCancerCenterNon-SLNnomogram(SCC-NSLN)wascreatedfromthelogisticregressionmodel.ThisnewmodelwassubsequentlyappliedtopatientsfromJulytoDecember.ThepredictiveaccuracyoftheSCC-NSLNnomogramwasmeasuredbycalculatingtheareaunderthereceiveroperatingcharacteristiccurve(AUC).

RESULTS:Basedontheresultsoftheunivariateanalysis,thevariablesthatweresignificantlyassociatedwiththeincidenceofnon-SLNmetastasisinanSLN-positivepatientincludedlymphovascularinvasion,neuralinvasion,thenumberofpositiveSLNs,thenumberofnegativeSLNs,andthesizeofSLNmetastasis(P0.05).Usingmultivariateanalysis,lymphovascularinvasion,thenumberofpositiveSLNs,thenumberofnegativeSLNs,andthesizeofSLNmetastasiswereidentifiedasindependentpredictorsofnon-SLNmetastasis.TheSCC-NSLNnomogramwasthendevelopedusingthesefourvariables.Thenewmodelwasaccurateanddiscriminatingonboththemodelingandvalidationgroups(AUC:0.vs0.).Thefalse-negativeratesoftheSCC-NSLNnomogramwere3.54and9.29%forthepredictedprobabilitycut-offpointsof10and15%whenappliedtopatientswhohave1-2positiveSLNs.

CONCLUSION:TheSCC-NSLNnomogramcouldserveasanacceptableclinicaltoolinclinicaldiscussionswithpatients.TheomissionofALNDmightbepossibleiftheprobabilityofnon-SLNinvolvementis10and15%inaccordancewiththeacceptableriskdeterminedbymedicalstaffandpatients.

PMID:

DOI:10./s---z