Migration in last decade to high-risk

(1)

Disponibleenlignesur

ScienceDirect

www.sciencedirect.com

ORIGINAL ARTICLE

Migration in last decade to high-risk

prostate cancer after radical prostatectomy

Migration des stades après prostatectomie totale au cours des 10 dernières années vers un cancer de la prostate à haut risque

P.-M. Patard

^a

, M. Roumiguié

^a

, T. Prudhomme

^a

, N. Doumerc

^a

, M. Thoulouzan

^a

, X. Gamé

^a

,

A. de la Taille

^b

, P. Rischmann

^a

, M. Soulié

^a

, L. Salomon

^b

, J.-B. Beauval

^a^,∗

aDepartmentofurology,andrologyandrenaltransplantation,CHURangueil,Paul-Sabatier university,1,avenueJPoulhès,31059Toulousecedex,France

bDepartmentofurology,Henri-MondorHospital,Assistance—PubliqueHôpitauxdeParis,51, avenueduMaréchaldeLattre-de-Tassigny,94010Créteil,France

Received27March2018;accepted17September2018 Availableonline15October2018

KEYWORDS Prostatecancer;

Radical prostatectomy;

Stagemigration

Summary

Objective.—Thereiscontroversyaroundprostatecancer(PCa)screeningthroughtheuseof PSA, duetotheriskofovertreatment.The currenttrendobservedinvariousEuropeanand Americanstudiesisadecreaseinthenumberofradicalprostatectomy(RP)inlow-riskPCaand anincreaseforintermediateorlocallyadvanceddiseases.Theobjectiveofthisstudywasto observethemigrationofthepathologicalstagesfromradicalprostatectomy(RP)over10years inFrancethrough2Frenchcenters.

Methods.—Itwas amulticentricretrospectivestudy,where alltheRP realized in2French tertiarycenters,inalaparoscopicorretropubicapproachforeachoftheyears2005,2010and 2015wereincluded.Preoperativedata(age,PSA,clinicalstage,numberofpositivebiopsies, Gleasonbiopsyscore)andpostoperativedata(pTNM,pathologicalGleasonscore(pGS))were analyzedandcompared.

∗Correspondingauthor.

E-mailaddresses:pm.patard@gmail.com(P.-M.Patard),roumiguiemathieu@yahoo.fr(M.Roumiguié),prudhomme-thomas@hotmail.fr (T.Prudhomme),doumerc.n@chu-toulouse.fr(N.Doumerc),thoulouzan.m@chu-toulouse.fr(M.Thoulouzan),gamexavier@gmail.com(X.

Gamé),adelataille@hotmail.com (A.de la Taille), rischmann.p@chu-toulouse.fr (P. Rischmann),soulie.m@chu-toulouse.fr (M. Soulié), laurent.salomon@aphp.fr(L.Salomon),jbbeauval@gmail.com(J.-B.Beauval).

https://doi.org/10.1016/j.purol.2018.09.008

(2)

Results.—Inall,1282RPwererealized(503in2005,403in2010,376in2015).Respectively between2005,2010,2015theaveragenumberofpositivebiopsyincreasedsigniﬁcantlyfrom 2.30vs.2.88vs.5.3(P=0.0001).ThedistributionofD’Amico’sriskevolveswithtime:low-risk:

49.9vs.44.4vs.15.7%(P=0.0001);intermediate risk:40.95vs.43.92 vs.64.1% (P=0.0001) andhigh-risk:9.15vs.11.66vs.20.2%(P=0.0001)between2005,2010and2015respectively.

pGSevolvedtohigherscorewithSG<7:22.8vs.29.9vs.7.1%etSG≥7:77.2vs.70.1vs.92.9%

(P=0.001).Also,pTNMincreasedtonon-organ-conﬁneddisease:pT2:66.9vs.51.9vs.48.7%;

pT3:33.1vs.48.1vs.51.3%(P=0.0001).

Conclusion.—This studyshowed achangeinthemanagementofPCa sincenewrecommen- dationsfrommedicalauthoritiesaboutPSAscreeningandevolvingofconservativetreatment.

NumberofRPincreaseforhigherriskPCa.Thischangecorrespondstobetterpatientselection forRP:decreaseforlow-riskandincreaseforhigh-riskorgan-conﬁneddisease.

Levelofevidence.—3.

MOTSCLÉS Cancerdela prostate; Prostatectomie; Migrationdesstades

Résumé

Objectif.—Il existe une controverse autour du dépistage du cancer de prostate(CaP) par l’utilisationduPSA,enraisondurisquedesurtraitement.Latendanceactuellementobservée dansdifférentesétudesEuropéennesetAméricainesestunediminutiondunombredeprosta- tectomietotale (PT)dansles CaPde basrisqueetune augmentationpour lesmaladies de pronostiqueintermédiaireoulocalementavancée.L’objectifdecetteétudeétaitd’observer lamigrationdes stadespathologiques sur10 ansdes patients subissantuneprostatectomie totale,aﬁnd’évaluerlatendanceFranc¸aiseàpartirde2centresuniversitairesdehautvolume chirurgical.

Méthodes.—Il s’agissait d’une étude rétrospective, multicentrique où toutes les PT réal- isées,dans2centresuniversitairesFranc¸aisetquelquesoitlavoied’abord,laparoscopique ourétropubienne,pourchacunedesannées2005,2010et2015étaientincluses.Lesdonnées préopératoires(âge,PSA, stadeclinique,nombre debiopsies positives,score debiopsie de Gleason)etpostopératoires(pTNM,scoredeGleasonpathologique(pGS))ontétéanalyséeset comparées.

Résultats.—Autotal,1282PTontétéréalisées(503en2005,403en2010,376en2015).Res- pectivemententre2005, 2010et 2015,lenombremédiandebiopsies positivesaaugmenté signiﬁcativementde2,30vs2,88vs 5,3(p=0,0001).Larépartitiondesrisquesselonlaclas- siﬁcationdeD’Amicoévoluaitavecletempsavec:unrisquefaible:49,9vs44,4vs 15,7% (p=0,0001);risqueintermédiaire:40,95vs43,92vs64,1%(p=0,0001);etrisqueélevé:9,15vs 11,66vs20,2%(p=0,0001)entre2005,2010et2015respectivement.LescorepGSévoluait versunscoreplusélevéavecSG<7:22,8vs29,9vs7,1%etSG≥7:77,2vs70,1vs92,9% (p=0,001).DepluslesstadespTNMévoluentavecnotammentuneextensionextraprostatique: pT2:66,9vs51,9vs48,7%;pT3:33,1vs48,1vs51,3%(p=0,0001).

Conclusion.—CetteétudeamontréunchangementdanslapriseenchargeduCaPdepuisles nouvellesrecommandationssurledépistageduPSAetl’évolution destraitementsconserva- teurs.LenombredePTpourlesCaPàrisqueplusélevéaugmente.Cechangementcorrespond àunemeilleuresélectiondespatientséligiblesàlaPT:diminutionpourlesmaladiesàfaible risqueetaugmentationpourlesmaladieslocaliséesàhautrisque.

Niveaudepreuve.— 3.

Introduction

Prostatecancer(PCa)incidenceincreasedsigniﬁcantlysince the establishment of PSA screening in the 1980s [1]. In this context,age, PSA and clinical stageat the diagnosis decreaseforpatientstreatedbyradicalprostatectomy(RP)

[2].Moreover,thepathologicalresultsafterRPreportedan increasingrateoforgan-conﬁneddiseasefrom50%to80%in USAbetween1988and2001andfrom50%to75%inEurope between 1988and 2005 [3]. Afteran increase inthe rate ofRPanddebateabouttheriskofovertreatment,thereis currently a decreasein the number of RP in Francefrom

(3)

25,000peryearbetween2009 and2011 to19,600in2014 [4] whereas the absolute number of RP from 5 European institutionsincreasedfrom2344to2504patientsbetween 2010 and 2015 [5]. Gallina etal. [3] in USA and recently Budausetal.[6]inGermanyreportedastagemigrationto higherriskandnon-organ-conﬁneddisease,from30%to45%

inUSAbetween2001and2005andfrom19to33%inGer- manybetween2003and2009.Beauvaletal.observedthe sametrendwithmorepT3onpathologicalresultsbetween 2005and2010fromaFrenchcohort[7].Evenmorerecently, aEuropeanmulticenteranalysisshowedadecreasein the percentageofpatientshavingonlylocalizedGleason6dis- ease aftersurgery from34% in 2005 to8% in 2015 for all patientswithacorrespondingincreaseintheproportionof patientswithlocallyadvanceddisease(≥pT3and/orGlea- son≥7)from 66% to 92% over the same years.The same studyalsoreportedadecreaseinthepercentageofpatients whowereeligiblefor activesurveillancefrom54% to38%

between2005and2015[5].

The objective of this study wasto observe the migra- tionofpathologicalstagesfrompatientsundergoingradical prostatectomyover10years,toseethecontemporarytrend in2Frenchreferraltertiaryinstitutions.

Methods

Forthisstudy,theRP databaseswere retrospectivelycol- lectedfrom2highvolumetertiaryreferralinstitutions,were combinedandanalyzedforeachofthe3years,2005,2010, 2015.Between2005and2015,1282 patientswithlocalize prostatecancerweretreatedbyRPafterapprovalfromour InstitutionalReviewBoard(503RPin2005,402RPin2010, 376RPin2015).

Treatment

All patients were treated with RP associated or not with extendedpelviclymph-nodedissection.RPwereperformed bylaparoscopic, robot-assistedor openretropubic prostatectomy.

Endpoint

All complete clinical and pathologic data were recorded, including age, year of surgery, preoperative PSA, clinical stage,biopsyGleasonscore,numberofbiopsycores,num- berofpositivecores,percentageofpositive biopsycores, pathologicstage,pathologicGleasonscore,seminal-vesicle invasion,surgical-marginstatus,andlymph-nodeinvasion.

The clinical stage was assigned according to the 2002 TNM stagingsystem, prostate biopsy cores were obtained under transrectal ultrasound guidance, using a >12-core biopsy protocol, and some were realized using an image fusionsystem(Koelis,guided-biopsy)andpre-treatmentPSA wasmeasuredbeforedigitalrectalexamination.Dedicated genitourinary pathologists assessed biopsy and pathologic grading in each center without central pathology review accordingtotheGleasongradingsystembefore2005andthe modiﬁedISUPGleasonscoreafter2005.Thereanalyzedthe prostatetotheStanfordprotocolandthepTstageaccording tothe2002AJCCstagingsystemforPCa.

Patients were stratiﬁed using the D’Amico classiﬁca- tion, in low, intermediate and high-risk of progression.

Low-risk PCa was defined as: ≤cT1a, cN0/X, cM0/X and PSA<10ng/mL, and clinical Gleason 6 disease.Intermedi- aterisk wasdefined as: cT2b or PSA≥10 and≤20ng/mL orclinical Gleasonscore=7.High-riskdiseasewasdefined as:≥cT2c,cN1,cM1,PSA>20ng/mL,and/or clinicalGlea- son≥8. Anyofthese criteriawouldmake apatienttobe consideredhighrisk.Organconfineddiseasewasdefinedas:

≤pT2andISUP1(Gleasonscore6)whereasnoorgancon- ﬁnedwas>pT2or≥ISUP2(≥Gleason 7).Gandagliaandal in2015 deﬁnedpatientswitha betterprognosis (pT2and GS≤6)whowouldbeeligibleforactivesurveillanceinwhom surgicaltreatmentmaybequestionable,orinaconservative approach[8].

Statistical analysis

Dataweresummarizedusingdescriptivestatistics.Categor- icalvariableswerepresentedascontingencytablesintothe statisticalsoftware, i.e,number andpercentage for each category of variable, and number of missing data. Con- tinuous variables were presented as median, range, and numberofmissingdata.Comparisonsbetweengroupswere performedusingtheChi-squaredorFisher’sexacttest for categoricalvariablesandthestudentttestorAnovaforcon- tinuousvariables.Allreportedp-valuesweretwo-sidedwith asigniﬁcancelevelatP<0.05.Statisticalanalysiswasper- formedusingPRISM(GraphPadPrism^®version5.0,Software, Inc,California,USA).

Results

The absolute number of RP performed including these 3 years(2005,2010,2015)was1282.Fortheyear2005,503 RPwererealized,incomparisonto403RPin2010and376in 2015,whichcorrespondedtoa19%decreaseinRPbetween 2005and2010and25%between2005and2015.

Theclinicalandpathologicalcharacteristicsaresumma- rizedinTable1.

ThepreoperativemedianPSAincreasedfrom8.55ng/mL to8.99ng/mL and10.14ng/mL (P=0.047) respectively in 2005, 2010, 2015. The median number of positive cores increasedsigniﬁcantlyfrom2.29in2005to2.86in2010and 5.3in2015aswellasthebiopsyGS≥7whichincreasedfrom 46.8in2005to76.3%in2015.

TheclinicalstagesweresigniﬁcantlydifferentandcT1c decreasedfrom77.8% in2005 to73.0% in2010 and57.5%

in2015(P=0.0001)whilecT2increasedfrom20.7%in2005 to23.0% in 2010and 37.9% in 2015 (P=0.0001).Also cT3 increasedfrom 1.0% in 2005 to4.0% in 2010 and 4.6% in 2015(P=0.0001)(Fig.1).

Specimen GS increased signiﬁcantly to GS≥7, from 77.2% in 2005 to 70.1% in 2010 and 92.9% in 2015 (P=0.0001).Mainly,thepathologicalstagesevolvedtowards non-localizedtumors,withadecreaseofpT2from66.9%to 51.9%and 48.7% (P=0.001) andan increase for pT3 from 33.1%to48.1%and51.3%(P=0.0001),respectivelyin2005, 2010and2015(Fig.2).Thus,in2015,morepT3(51.3%)than pT2wereobserved(48.7%).

(4)

Table1 Clinicalandpathologicalcharacteristicsofpatientsaccordingtothedateofsurgery.

2005 2010 2015 P

Age(yr) 62.85 62.52 64.48 0.0001

PSA(ng/mL) 8.55 8.99 10.14 0.047

Numberofpositivescores 2.2 2.8 5.3 <0.0001

BiopsyGleasonScore(n)

<7 302(63.0) 242(60.3) 88(23.7)

7 154(32.0) 136(33.9) 246(66.3) <0.0001

>7 24(5.0) 23(5.7) 37(10.0)

cTNM(n,%)

T1c 383(77.4) 298(74.0) 211(57.5)

T2a 82(16.7) 58(14.4) 97(26.4)

T2b 20(4.0) 27(6.7) 32(8.7)

T2c 0(0) 4(0.9) 10(2.6) <0.0001

T3 5(1.0) 16(4.0) 17(4.6)

T4 2(0.4) 0(0) 0(0)

PathologicGleasonscore

<7 113(22.8) 118(29.4) 26(7.0) <0.0001

7 336(67.9) 247(61.4) 303(81.9) <0.0001

>7 46(9.3) 37(9.2) 41(11.1) 0.6

pTNM(n,%)

pT2a 42(8.3) 32(7.9) 23(6.2) 0.37

pT2b 43(8.5) 7(1.7) 6(1.6) <0.0001

pT2c 250(49.7) 170(42.2) 152(40.9) 0.013

pT3 166(33.0) 194(48.1) 191(51.3) <0.0001

pT4 2(0.4) 0(0) 0(0)

Prostatevolume(cc) 55.56 48.86 53.27 0.0001

Figure1. EvolutionofclinicalstagecTNMinpatientstreatedby radicalprostatectomy.

Afterstratiﬁcation, accordingtoD’Amico’srisk groups, therewasasigniﬁcant reductionin theproportionoflow- risk PCa from 49.9% in 2005 to 44.4% in 2010 and 15.7%

in 2015 (P<0.0001) and therefore a signiﬁcant increase towardshigh-risk PCafrom9.1% in 2005to 11.6%in 2010 and20.2%in2015(P<0.0001).(Fig.3).Thereforein2015, wehadmoreRPforhigh-riskthanlow-risk.

Wealsoobservedindetail,theevolutionofeachparam- eteroftheD’Amicohigh-riskclassiﬁcation(PSA,cTNM,GS biopsy)foreachofthese3years.Thus,wefoundthatforthe high-risk group, the proportionof Gleason 8-10 increased from 14 patients (30.4%) in 2005 to 27 patients (35.5%) in2015 (P=0.003). Therewas nosigniﬁcant differencein

Figure2. EvolutionofpathologicalstagepTNMin10yearsafter radicalprostatectomy.

theproportionof patients≥cT2c(P=0.051)or thosewith PSA>20ng/mL(P=0.06)(Fig.4).

AccordingtoGandagliastratiﬁcation,weobservedasig- niﬁcant decrease of ‘‘favorable pathology’’ from 21.1 to 19.4and5.0%in2005,2010and2015respectively(Fig.5).

Discussion

Thereis aconﬁrmeddecreaseofthenumbers inRPwhile our study concern 2 high volumetertiary referral institu- tions.Van denBerghetal.observedunlikeanincrease of absolutenumberofRPsbasedon5tertiaryEuropeancenters

(5)

Figure3. EvolutionofD’Amico’sriskgroupin10years.

Figure4. Evolutionofhigh-riskD’Amico’sgroupitems.

Figure5. Evolutionoffavorable/unfavorablegroupsaccordingto Gandaglianomogramm.

between2000and2015.Indetails,therateofhigh-riskPCa (HRPCa)increasedaround30%(from1in10patientsto1 in3patients)andevenmorefromGermanyandNetherland wheretheratedeHRPCarepresentedaround40%ofRPsin 2015.InFrancetheresultsweresimilartoourstudywith20%

ofHRPCa(from1in10patientsin2005to1in5patientsin 2015).Also,ourstudyconﬁrmedalaterdecreasinginFrance

ofRPsfororganconﬁneddiseasecomparedwithEuropean studies[5,9].Thistrendseemedtostartafter2004—2005.To putthistiminginperspective,theﬁrstreportsonAS from theEuropean RandomizedStudy of Screeningfor Prostate CancerandinitiationofPRIASdatebackto2006—2007.

ThevariationinthenumberofRPscanalsobeexplained because of the large referral centers in others European countries (as Germany with more than 20,000 RPs in 15 years)[5].

We can then assume that the decrease in RP is even greaterinotherinstitutionswithavariablerateamongthe Frenchregionstestiﬁedourresultwithadecreaseof6.7%

between2010—2015whereasthedecreaseofRPinFrance was21%between 2008 and2016 [10].The global number of RPs increasedcontinuously through 2007 and the drop of20% between 2007 and2010 wasexplainedby thesta- bilization of PCa incidence and the introduction of new recommendationsby the French Association of Urologyin 2007,highlightingalternativeoptionssuchasactivesurveil- lanceandbrachytherapy[11,12].

Indeed, the increase in PCa screening led to earlier diagnosisand thustolower riskdiseases, localizedtothe prostate [2]. These patients are often eligible for active surveillancewithanincreasingincidencethatcouldpartly explain the decrease of RP and especially the risk of overtreatment. These patients will then have a delayed curativetreatmentwhenprogression[12].

More recently, PCafocaltreatment techniques such as HIFUand cryotherapy have been developed which,in the contextofprotocols,essentiallytreatedpatientswithlow riskofdevelopmentwithanincreasingproportion[13,14].

These2centershadthepossibilitytoperformfocaltreat- mentssuchasHIFUbetween2010and2015orcryotherapy foroneofthem.

Migrationfromstagingtoorgan-conﬁned-diseaseaswell asadecreaseinageandPSAatdiagnosishasbeenreported sincethegeneralizationofPSAscreening[2,15].Forthepast decade,therehasbeenaninversemigrationofthepatholog- icalstagesofpatientstreatedbyRPresultinginanincrease ofhigh-gradestageonpathologicalspecimenandGS≥ISUP2 [3,5—7,16,17]. Thisstage migrationiswell representedin ourstudywheretheproportionoflowriskdecreasedcon- siderablysincein2015agreaternumberofRPareexecuted forhigh-riskPCa(20.2%)thanlowrisk(15.7%)(P<0.0001).

ThesametrendwasalsoobservedinaUKregionaltertiary referralcenterover10yearswithanincreasedproportion ofhigh-riskdiseasefrom11.6%between2005-2008to33.6%

between2013—2015[17].

This could represent toa better selection of patients, where RP is reserved for higher stages and also a better lowriskselectionwiththedevelopmentofunfavorableSA criteriasuchasinvasionpercentageonbiopsyandMRIthat arerecentlyapplied.

Inaddition,MRIhasanincreasinglyimportantroleinthe diagnosis of PCa withthe use of a fusion imaging system inthesetwocenterstoperformtargetedbiopsies.This is reflectedin thesignificant increaseinthe numberofpos- itivebiopsyinourstudy 2.29to2.86and5.3respectively in2005,2010,2015(P<0.0001).Severalstudieshaveshown theinterestofMRIfor thedetectionofsignificantcancers inordertocarryoutbiopsyofsuspiciousareasanddonot misunderstandamoreaggressivedisease(ISUP>1)[18,19].

(6)

Heretheuseoftargetedbiopsiesstartedbetween2010and 2015inthese2centersandcouldpartlyexplaintheincrease inthenumberofpositivebiopsiestestiﬁedbytherandom- izedstudyPRECISIONwheretherateofpositivebiopsieswas 44%fortargetedbiopsiescomparedtotherateof 18%for standardbiopsies.Moreover,thissamestudyobserved38%

ofsigniﬁcantcancers(≥ISUP2)inthetargetedbiopsygroup against26%forstandardbiopsies(P=0.005)[20].

The French and European associationsof urology have expandedtherecommendationsforRPforlocallyadvanced stages through a good survival [21,22]. Indeed the spe- ciﬁcsurvivalat5,10and15yearsandevaluatedbetween 90—99%,85—92%,62—84%andtheoverallsurvivalat5and 10yearswas90—96and76—77%,respectively[23—27].Free- landandalreportedaspeciﬁcsurvivalof 84%at15 years afterRPaloneforT3[25].

Our results could be in line withthe current trend of applyingthisradicaltreatmentonlytothemostbeneﬁcial populationandsoavoidingsideeffectstoothers.

ThischangeinindicationsofRPisstillevolvingandcur- rently,localtreatment isalsodeveloping indiseases even more advanced or metastatic [24,25]. Indeed, RP can be practicedforPCawithclinicallypelviclymphnode-positive (cN1) disease toprovide local treatment in a multimodal approach.Localtreatment±androgendeprivationtherapy (ADT) wasassociated with a significant overall mortality- freesurvivalbenefit(hazardratio=0.31)andnodifference was highlighted between RT and RP [28]. In addition, localcontrol withcytoreductive RP in a contextof oligo- metastaticdiseasemaybeapartofamultimodalapproach inawell-selectedpopulation.Somestudiesobservedasig- nificantlybetterclinicalsymptom-freesurvival(38.6versus 26.5months)andcancer-specificsurvivalrates(95.6%ver- sus84.2%)withasimilaroverallsurvivalcomparedtoADT alone[29].

Inthefuture,developmentofnewtoolswillbehelpful toevaluatepreoperative risk.Forexampleapreoperative nomogramforidentiﬁcationofpathologicallyfavorabledis- easeinintermediateriskhasbeendevelopedwithastrong accuracy(AUC=82%)[30].Thisvalidatedriskcalculatorcan helpphysiciantodistinguishfavorableintermediateriskPCa thatcanbetreatedbyconservativeapproach.

In last decade, several studies investigating diagnostic biomarkers (PCA3, Prostate Heath Index, 4K score) with promisingresultsthatcanhelpphysicianindecision-making inchallengingclinicalsettings[31].

Weak points of our study included the fact that this wasa multifactorialstage migration aschange in screen- inganddiagnosispractice.Indeedtheriskdistributioncan alsobeexplainedbythemodification ofthe classification oftheGleasonscoresince2005whereevenatinyfraction of more aggressive tumor is reported in the final analy- siswhich makesitpossibletoreclassifyadiseasetowards higherscores [32].Also,MRImayhavereclassifiedpatient asnon-confinedorgandiseasewithbetterdetectionofT3.

Theincreasinguseoftargetedbiopsymayhaveresultedin ﬁndingmore high Gleason scores. In addition, we had no objectiveevidence onthe use of other treatments espe- ciallyforthelowrisk.Wecansuggestahigherrateoffocal treatmentandASwithoutdatatoconﬁrmitinourdatabase andsome patients mayhave been initially managed with AS before undergoing surgery. However, the incidence of

PCaremained stablein France,at around0.4% each year between2009and2014,whilealackofspeciﬁctreatment (AS or watchfulwaiting) increased from20.8% in 2009 to 26.9% in 2012 [4]. This could be an additional argument in favor of a change in the selection of these patients.

Theseresultsshouldbetakenwithcautionbecausethiswas an observational study evaluatingtheabsolute number of RP and the evolution of the pathological proﬁle of these patients without specifying theproportion of other treat- mentsappliedorthenumberofPCadetectedeachyear.

Finally,ourresultsreflectedthepracticefrom2tertiary referralinstitutionsandcannotbeextrapolatedtoalloth- erscentersthat donotnecessarily haveother treatments availabletomanagelowriskdisease.Long-termstudiesare needed toassess theoverallandspecific survivalofthese groupsinordertoconfirmthisbetterpatientselection.The potential riskof this high-risk migrationwouldbe tomiss thewindowofcurability.

Conclusion

Over the past10 years,therehas been a stagemigration tohigherriskdiseases.In2015moreRPareperformedfor intermediateandhighrisksthanlowrisks.Thisincreasein RPforhigher-riskorlocallyadvanceddiseasescorresponds to improve patient selection and because of recognized efﬁcacy of RP inlocal controland theemergence of new therapiesforlow-riskPCa.

Disclosure of interest

Theauthorsdeclarethattheyhavenocompetinginterest.

References

[1]JemalA,SiegelR,WardE,HaoY,XuJ,MurrayT,etal.Cancer statistics,2008.CACancerJClin2008;58(2):71—96.

[2]CooperbergMR,LubeckDP, MengMV, MehtaSS,Carroll PR.

The changing face of low-risk prostate cancer: trends in clinicalpresentationand primarymanagement.JClinOncol 2004;22(11):2141—9.

[3]Gallina A, Chun FK-H, Suardi N, Eastham JA, Perrotte P, Graefen M, et al. Comparison of stage migration patterns between Europe and the USA: an analysis of 11 350 men treatedwithradicalprostatectomyfor prostatecancer.BJU Int2008;101(12):1513—8.

[4]Tuppin P, Leboucher C, Samson S, Peyre-Lanquar G, Pierre GabachP,RebillardX.InVS.Décembre2015.Towardanevo- lutionofprostatecancerdetectionandmanagementpractices amongmenaged40yearsoldandmoreinFrance(2009—2014)?

[5]vandenBerghR,GandagliaG,TilkiD,BorgmannH,OstP,Surcel C,etal.Trendsinradicalprostatectomyriskgroupdistribution inaeuropeanmulticenteranalysisof28,572patients:towards tailoredtreatment.EurUrolFocus2017[S2405-4569(17)30188- 8].

[6]BudäusL,SpethmannJ,IsbarnH,SchmitgesJ,BeeschL,Haese A,etal.Inversestagemigrationinpatientsundergoingradi- calprostatectomy:resultsof8916Europeanpatientstreated withinthelastdecade.BJUInt2011;108(8):1256—61.

[7]BeauvalJ-B,RoumiguiéM,DoumercN,ThoulouzanM,Huyghe E,AlloryY,etal.[Migrationofpathologicalstageafterradical

(7)

prostatectomy to higher risk tumors of relapse: compara- tive two-center study between 2005 and 2010]. Prog Urol 2012;22(16):1015—20.

[8]GandagliaG, SchiffmannJ, Schlomm T, FossatiN, Moschini M,SuardiN,et al.Identiﬁcation ofpathologicallyfavorable diseaseinintermediate-riskprostatecancerpatients:Impli- cationsforactivesurveillancecandidatesselection.Prostate 2015;75(13):1484—91.

[9]SchröderFH,HugossonJ,RoobolMJ,TammelaTLJ,CiattoS, NelenV,etal.Screeningandprostate-cancermortalityinaran- domizedEuropeanstudy.NEnglJMed2009;360(13):1320—8.

[10] CornuJ-N, TerrasaJ-B, CussenotO,Haab F,LukacsB. Evo- lutionofthe landscapeofsurgical management ofprostate cancerinFrance:thecaseofradicalprostatectomy.EurUrol 2012;62(4):735—6.

[11] Rébillard X, Grosclaude P, Lebret T, Patard J-J, Pﬁster C, Richaud P, et al. [Projected incidence and mortality from urologiccancerinFrancein2010.].ProgUrol2010;20(Suppl 4):S211—4.

[12] CooperbergMR,BroeringJM,CarrollPR.Timetrendsandlocal variationinprimarytreatmentoflocalizedprostatecancer.J ClinOncol2010;28(7):1117—23.

[13] TurpenR,RosserCJ.Focaltherapyforprostatecancer:revo- lutionorevolution?BMCUrol2009;9:2.

[14] RischmannP,GeletA, RicheB,Villers A,PasticierG,Bondil P,etal.Focalhighintensityfocusedultrasoundofunilateral localizedprostatecancer:aprospectivemulticentrichemiab- lationstudyof111patients.EurUrol2017;71(2):267—73.

[15] ShaoY-H,DemissieK, Shih W,MehtaAR, SteinMN, Roberts CB,etal.Contemporaryriskproﬁleofprostatecancerinthe UnitedStates.JNatlCancerInst2009;101(18):1280—3.

[16] Huland H, Graefen M. Changing trends insurgical manage- mentofprostatecancer:theendofovertreatment?EurUrol 2015;68(2):175—8.

[17] GnanapragasamVJ,ThurtleD,SrinivasanA,VolanisD,George A,LophatananonA,etal.Evolutionandoncologicaloutcomes of a contemporary radical prostatectomy practice in a UK regionaltertiaryreferralcentre.BJUInt2016;118(5):779—84.

[18] FüttererJJ, Briganti A, De Visschere P,Emberton M, Gian- nariniG,KirkhamA,etal.Canclinicallysigniﬁcantprostate cancerbedetectedwithmultiparametricmagneticresonance imaging? A systematic review of the literature. Eur Urol 2015;68(6):1045—53.

[19] SiddiquiMM,Rais-BahramiS,TurkbeyB,GeorgeAK,Rothwax J,ShakirN,etal.ComparisonofMR/ultrasoundfusion-guided biopsy with ultrasound-guided biopsy for the diagnosis of prostatecancer.JAMA2015;313(4):390—7.

[20]Kasivisvanathan V, Rannikko AS, Borghi M, Panebianco V, Mynderse LA, Vaarala MH, et al. MRI-targeted or stan- dard biopsy for prostate-cancer diagnosis. N Engl J Med 2018;378(19):1767—77.

[21]MottetN,BellmuntJ,BollaM,BriersE,CumberbatchMG,De SantisM,etal.EAU-ESTRO-SIOGGuidelinesonprostatecancer.

Part1:screening,diagnosis,andlocaltreatmentwithcurative intent.EurUrol2017;71(4):618—29.

[22]Rozet F,HennequinC,Beauval J-B, BeuzebocP,Cormier L, FromontG,etal.[CCAFUfrenchnationalguidelines2016-2018 onprostatecancer].ProgUrol2016;27(Suppl1):S95—143.

[23]Carver BS, Bianco FJ, Scardino PT, Eastham JA. Long-term outcomefollowingradicalprostatectomyinmenwithclinical stageT3prostatecancer.JUrol2006;176(2):564—8.

[24]GerberGS,ThistedRA,ChodakGW,SchroderFH,Frohmuller HG, Scardino PT, et al. Results of radical prostatectomyin menwithlocallyadvancedprostatecancer:multi-institutional pooledanalysis.EurUrol1997;32(4):385—90.

[25]FreedlandSJ,PartinAW,HumphreysEB,MangoldLA,WalshPC.

RadicalprostatectomyforclinicalstageT3a disease.Cancer 2007;109(7):1273—8.

[26]HsuC-Y,JoniauS,OyenR,RoskamsT,VanPoppelH.Outcome ofsurgeryforclinicalunilateralT3aprostatecancer:asingle- institutionexperience.EurUrol2007;51(1):121—8[discussion 128-129].

[27]WardJF,SlezakJM,BluteML,BergstralhEJ,ZinckeH.Radical prostatectomyfor clinically advanced(cT3)prostatecancer sincetheadventofprostate-speciﬁcantigentesting:15-year outcome.BJUInt2005;95(6):751—6.

[28]SeisenT,VetterleinMW,KarabonP,JindalT,SoodA,NoceraL, etal.Efﬁcacyoflocaltreatmentinprostatecancerpatients with clinically pelvic lymph node-positive disease at initial diagnosis.EurUrol2017.

[29]Heidenreich A, Pﬁster D, Porres D. Cytoreductive radical prostatectomyinpatientswithprostatecancerandlowvolume skeletalmetastases:resultsofafeasibilityand case-control study.JUrol2015;193(3):832—8.

[30]BeauvalJ-B,CabarrouB,GandagliaG,PatardP-M,OuzzaneA, delaTailleA,etal.Externalvalidationofanomogramforiden- tiﬁcationofpathologicallyfavorablediseaseinintermediate riskprostatecancerpatients.Prostate2017;77(8):928—33.

[31]KretschmerA,TilkiD.Biomarkersinprostatecancer—Current clinicalutilityandfutureperspectives.CritRevOncolHematol 2017;120:180—93.

[32]AlbertsenPC,HanleyJA,BarrowsGH,PensonDF,Kowalczyk PDH,SandersMM,etal.ProstatecancerandtheWillRogers phenomenon.JNatlCancerInst2005;97(17):1248—53.