Diagnostic excision of the cervix in women over 40 years with human papilloma virus persistency and normal cytology

(1)

Diagnostic

excision

of

the

cervix

in

women

over

40 years

with

human

papilloma

virus

persistency

and

normal

cytology

Riina

Aarnio

a,

*

,

Ingrid

Wikström

a

,

Inger

Gustavsson

b

,

Ulf

Gyllensten

b

,

Matts

Olovsson

a

DepartmentofWomen'sandChildren'sHealth,UppsalaUniversity,SE-75185Uppsala,Sweden

b

DepartmentofImmunology,GeneticsandPathology,BiomedicalCenter,SciLifeLabUppsala,Box815,UppsalaUniversity,SE-75108Uppsala,Sweden

ARTICLE INFO Articlehistory:

Received28November2018 Receivedinrevisedform16April2019 Accepted6May2019

Availableonline8May2019 Keywords:

Humanpapillomavirus Cervicalintraepithelialneoplasia Colposcopy

Loopelectricalexcisionprocedure Transformationzone

ABSTRACT

Objective:Persistentinfectionwithhumanpapillomavirus(HPV)isrecognizedasthemainriskfactorof cervicalcancer.InvestigationviacytologyandcolposcopyhavelowersensitivitythanHPVtestinginthe diagnosis of high-grade cervical intraepithelial neoplasia (CIN2+). Despite normal cytology and colposcopyﬁndingswomenwithpersistentHPVinfectionhaveanincreasedriskofCIN2+.Theaimofthe studywastoevaluatetheproportionofhistologicallyconﬁrmedCIN2+inwomenwithpersistentHPV infectionandnormalPapsmears.

Studydesign:FromApril2013untilMarch2016weprospectivelyrecruited91womenover40yearswith persistent HPV infection without any abnormalities in cytology. Of these, 40 women attended a gynecologicalexaminationincludinganHPVtest,Papsmear,endocervicalcytology,colposcopywith biopsiesanddiagnosticloopelectrosurgicalexcisionprocedure(LEEP).BiopsyandLEEPsampleswere subjectedtohistologicalexamination

Results:CIN2+wasveriﬁedbyhistologicalexaminationoftheLEEPsamplein6/40(15%)ofthewomen. AllthecytologicalsampleswerenormalandnoneofthebiopsiesconﬁrmedCIN2+.Only19/40women stillhadapersistentHPVinfectionatthestudyvisit.Noneofthe21/40womenwhohadclearedtheirHPV infectionatthestudyvisithadCIN2+inhistologyoftheLEEPsample.

Conclusions:ApersistentHPVinfectionneedstobemonitoreddespitenormalPapsmears,since6/40 (15%) women older than40 years, was revealed to have an undiagnosed CIN2+ when LEEP was performed.Counselingwomenregardingtheriskofcervicalcancerandtheexpectedeffectofaneventual LEEPcanhelpthemtomakeanoptimalinformedchoice.

Introduction

Persistent infection with high-risk human papillomavirus (HPV) is a prerequisite for the developmentof cervical cancer [1].PersistenceofHPVisconsistentlyandstronglyassociatedwith theriskofdevelopinghigh-gradecervicalintraepithelialneoplasia (CIN2+)[2]whichinturnincludeanelevatedrisktoprogressto cervicalcancer[3].HPVtestinghasgreatersensitivityinrevealing CIN2+than cytology[4], and ispresently recommendedas the primaryscreeningmethodforcervicalcancerinEurope[5]. HPV-positive women are recommended cytological triage, the less sensitivemethodfordetectionofCIN2+.Thiswillresultinagroup of HPV-positivewomen with normal cytologyand the optimal clinicalhandlingofthesewomenisnotknown[6].

Although some persistent infections clear spontaneously, womenwithnormalcytologywhoare positiveforHPVhavea muchhigherriskofdevelopingCIN3thanHPV-negativewomen [7].Katkietal.havefounda7.4%5-yearriskofCIN3+inwomen withpersistentHPVinfectionandnormalcytology[8].Kjaeretal didapopulationbasedprospectivecohortstudyonHPVpositive women<30yearswithnormalPapsmearat baselineshowing risksofdevelopingCIN3in12yearsindifferentHPVgenotypesas following:HPV1626,7%,HPV1819,1%andHPV31/33over14%[9]. The Swedescreen study, involving women aged 32–38 years, describedthatamongwomenwithanormalPapsmearattending organized screening, the positive predictive value of HPV persistence as regards detection of biopsy-conﬁrmed CIN2+ was 29% [10]. Long-term follow-up of this study pointed out thatalltheHPV-positivewomenwithinitiallynormalcytology eitherbecome HPV-negative or developedCIN2+ within seven years[11].Mittaletaladditionallydescribehighestincidencerate ofCIN2+developingfrompersistentHPVinfectioninwomenover 50years[12].

* Correspondingauthorat:UppsalaUniversity,SE-75185Uppsala,Sweden. E-mailaddress:riina.aarnio@kbh.uu.se(R.Aarnio).

http://dx.doi.org/10.1016/j.eurox.2019.100042

ContentslistsavailableatScienceDirect

European

Journal

of

Obstetrics

&

Gynecology

and

Reproductive

Biology:

X

(2)

Cytologicalscreeningislesssensitiveinwomenof50yearsofage ormore comparedwith youngerwomen [13]. Colposcopy have moderate sensitivity for detection of CIN2+ in premenopausal women,andthesensitivitydecreasesfurtherinpostmenopausal women[14].Incolposcopicexaminationthetypeofthe transfor-mationzone(TZ)iscrucialforinterpretationoftheexamination.In types1and2theTZisfullyvisible.InTZtype3theupperlimitisnot visibleandthisisacommonfindinginthepostmenopausalperiod whentheTZoftenretractsintotheendocervix[15].Performanceof biopsieslackingtheTZisinsufficient[16].Itisalsomoredifficultto obtainadequateamountsoftissuefromtheendocervixand the sensitivity of cytobrush sampling in detecting dysplasia varies between44-93%,andisevenlowerwithendocervicalcurettagefor histopathologicalsamples[17,18].Inaddition,thepositivepredictive rate of colposcopic examinationis betteras regards high-grade cervicallesionsandlessaccurateforlow-gradecervicallesions[19]. Petry etalhavedescribedahighercolposcopyfailurerateinthe HPV-positive/Pap-normalgroupthanintheHPV-positive/Pap-abnormal group[16]. The most reliable method to obtain representative samplesistoexcisethewholeTZsurgically,forexamplebyusinga loop electrosurgical excision procedure (LEEP), for histological analysis.Thisprocedureisalreadyrecommendedasclinicalpractice forwomen withlow-gradecolposcopic changesorTZ3incolposcopy associatedwithhigh-gradecytologicalchanges[20].

ThemanagementofwomenwithpersistentHPVinfectionand no cytological or colposcopic evidence of CIN represents an unsolved clinical problem. We therefore performed LEEP on a groupofwomenover40yearswithHPVpersistencybutnormal cytology,todeterminetheprevalenceofhistologicallyconﬁrmed CIN2+.

Materialsandmethods

This prospectivestudywas performedfrom April2013until March2016andcarriedoutatthegynecologicalout-patientclinic, UppsalaUniversityHospital,Sweden.Werecruited91womenwith persistentHPVinfectionbutwithoutabnormalitiesincytologyby sendingthemaninvitationletter(Fig. 1).Weexcludedwomenwho hadplansforfuturepregnancies,whocouldnotunderstandthe informationinSwedish,andwhereLEEPwas regardedasbeing

technically difﬁculttoperform,since wewished tocarry itout usinglocalanesthesia.Forty-fourwomenwereinterestedintaking partinthestudyandthesewomenwerecontactedbytelephone formoreinformationandtoscheduleastudyvisit.Infourwomen LEEP could not be performed for anatomical reasons. All postmenopausalwomenweretreatedwithlocalestradiolforat leasttwoweeksbeforethevisittooptimizethevaginalmucosa andminimizethepossibleriskofpostoperativecervicalstenosis.A signedinformedconsentdocumentwasobtainedatthevisit.The examinationwasperformedbyanexperiencedcolposcopist(RA) and included a) a cytobrush sample for HPV analysis, b) a conventional Pap smear, c) a separate endocervical cytobrush sample on a glass slide, and d) colposcopic evaluation with applicationof5%aceticacidandiodinesolution,identiﬁcationof squamocolumnar junction and transformation zone and punch biopsysampling.Thetissueswiththemostpathological appear-ancewerebiopsiedandintheabsenceofabnormalityarandom biopsysamplewastaken.Finally,diagnosticLEEPwasperformed inlocalanesthesiaforhistologicalanalysis.Allwomenthatwere HPV-positiveatthestudyvisitunderwentfollow-up6–12months afterLEEP,withaPapsmearandanHPVtest.Womenthatdidnot participateinthestudywerefollowedupwithannualPapsmears andHPVtests.

All cytology and histology was performed at the Clinic of PathologyandCytology,UppsalaUniversityHospital,Uppsala.The highest histological grade found in each patient was used for interpretation of the results. In HPV-testing cervical cytobrush sampleswereappliedtoanindicatingFTAelutemicro-cardTM(GE Healthcare,UnitedKingdom,art.noWB129308).TheFTAcards wereprocessedusinga dedicatedautomatedlaboratorysystem (easyPunchSTARlet,HamiltonRobotics,USA)whichcollectseach card,takesaphotographofthesamplecollectionarea,identiﬁes thepartsofthesamplingdepositionareawiththehighestamount ofcellularmaterialusingamachinelearningsoftware,andthen takes 4 punches with a 3-mm diameter knife from the area containingmostmaterialanddepositsthepunchesinasinglewell ina96-wellmicrotiterplate.DNAextractionfrompuncheswas performed as described earlier [21]. Testing for HPV was performedusingamultiplexreal-timePCRassay(hpVIR),which detectsthefollowinghigh-riskHPVtypes:16,18,31,33,35,39,45,

(3)

51,52,56,58and59(18and45aredetectedtogether,and33,52 and58asonegroup)andalsomeasuresahumansinglecopygene (HMBS),whichservesas acontrolforthat thesamplescontain sufﬁcient amounts of cellular material for the test to be informative.Thelimit of detection(LOD) forthenuclear single copygeneHMBSandHPVwasbothsetto10copiesperPCR.The FTAcardswereanalyzedattheHPVlaboratory,UppsalaUniversity. TheendpointofthestudywashistologicallyidentiﬁedCIN2+in LEEP samples from women with persistent HPV infection but withoutevidentCINineithercytologyorcolposcopy.

ThestudywasapprovedbytheRegionalEthicsCommitteein Uppsala(Dnr2012/460).

Results

In the 40 women who underwent complete examination includingLEEP,themeandurationofknownHPVpersistencewas 20months(median12,range4–93).Themeanageofthewomen was58years(median59,range41–77)and33/40womenwere postmenopausal.Beforethestudy,25/40womenhadtaken1Pap smear,13/40 had taken 2 Pap smears, 1/40 had taken 3 Pap smearsand1/40hadtaken7Papsmears.Theywereallnormal. All the cytological analysis of Pap smears and endocervical samplesobtainedatthestudyvisitwerenormal.Only3/40Pap smears were reported lacking columnar cells which can be interpretedthatin37/40casesthejunctionwasreached.Noneof thebiopsysamplesconfirmedCIN2+,butsixshowedCIN1.Five caseslackabiopsysamplebecauseitwasdifficulttoobtaindue totechnicalreasons.Notably,for28/40(70%)ofthewomenhada TZ type 3. The mean depth of excisionwas 12.3mm (median 12mm,range8–18mm)and34/40histologicalsamplesincluded the whole TZ. No complications were reported. Histological evaluationoftheLEEPsamplesshowedthat20womenhadno CIN,14womenhadCIN1 andonly 3 ofthese samplesdidnot include thewhole TZ. Further two womenhadCIN2 andfour womenhadCIN3.NoneoftheLEEPsamplesdisplayedinvasive diseaseandfiveoutofsixCIN2+excisionshadfreeendocervical margins.FouroutofsixwomenwithCIN2+hadTZtype3buttwo womenwithCIN2+hadafully visiblenormalTZ.HPVanalysis revealed that 21/40 women had cleared their HPV infection, while19/40womenstillhadapersistent HPVinfectionatthe study visit. The most common HPVtypes thatpersisted were HPV16(n=5)andtheHPV33/52/58group(n=5).Allthewomen withCIN2+wereHPV-positive atthestudyvisitwithdifferent HPVtypes.Theknowndurationof HPVpersistence amongthe womenwith CIN2+ was between 7 and 20 months (mean 13 months)andtheknowndurationofHPVpersistenceamongthe womenshowingnoCINwasbetween7and93months(mean26 months).Noneofthe21womenwhowereHPV-negativeatthe studyvisitshowedCIN2+in histology.Allresultsfromwomen thatunderwentLEEParepresentedinTable1.

Atfollow-up6–12monthsafterLEEP,allsixwomenwithCIN2+ hadbecomeHPV-negativeandshowed normalcytology.Among eightHPV-positivewomenwithnodysplasiaintheLEEPsample sixwomenstillhadapersistentHPV-infectionatfollow-up(Fig. 1). Discussion

Inourstudy6/40(15%)womenwithpersistentHPVinfection andnodetectableabnormalitiesincytologyhadCIN2+veriﬁedby histologicalexaminationoftheLEEPsample.Twothirdsofthese women had a TZ3. HPVgenotyping showed differenthigh-risk typesinallsixofthesewomen,aﬁndingalsonoticedinanearlier studyfromourgroup[22],whichindicatesthatinordertocover the HPV types associated with persistence, it is necessary to genotype for more than HPV16/18 in this olderage group. An

importantnoticeisthatonly19womeninthiscohortstilltested HPV-positiveatthestudyvisitandalltheCIN2+womenwerein thisgroup.Bycontrast,noneofthe21womenwithclearedHPV infectionhadCIN2+,afindingthatwasalsonotedinfollow-upin the Swedescreenstudy[11]. However, a recentDutch post-hoc analysisdescribeshigherriskforCIN3+infiveyearsforwomen thatwereHPVpositiveinthefirstscreeningroundandthenturned HPVnegativeatfollow-up[23].Acontinuedsurveillanceofthese womenmight thusbeneeded.Our resultsfurtherhighlightthe importanceofcarryingoutfollow-upof womenwithpersistent HPVinfectionandanormalcytology,sinceamoderateproportion haveCIN2+andthusareatriskofdevelopingcervicalcancer.

The optimal managementof womenwith histologicalCIN1 is surveillance, since at least 70% of these lesions will resolve spontaneously and only few will progress [24]. While CIN2+ is consideredasatrueprecursorofcervicalcancerandisassociatedwith oncogenictypesofHPV,CIN1ismerelyaninsensitive histopatholog-icalsignofHPVinfection[25]thatcanbecausedevenbylow-riskHPV types[26].EvenifLEEPperformedbecauseofCINtendstoeliminate HPVinfection[27],treatmentofCIN1isassociatedwithahigher HPV-positiverateatfollow-upwhencomparedwithtreatmentofCIN2+ [28].Inourstudyalso,LEEPperformedinconnectionwithCIN2+ tendedtocleartheHPVinfectionearlierthaninwomenwithCIN1.In addition,itdoesnotseempossibletotreatapersistentHPVinfection withoutCINbymeansofLEEP,sincemostofthesewomenstilltested HPV-positiveatfollow-up.In women withCIN1ornormalhistologyin theexcisionthefutureriskofdevelopmentofCINorcancerremains unclear,butexcisionsamplesincludingthewholeTZ(29/34women) excludeanexistingCIN2+.

Thenumberofstudysubjectsisrathersmallsinceonly44/91 (48%)oftheinvitedwomenchosetoparticipate.Wedonotﬁnd this unexpected since the study visit included an invasive treatment demanding local anesthesia. Also the usually self-healingnatureoftheHPV-infectionandatrusttoanormalPap smearcouldexplainthelowparticipationrate. Wedohowever believethattheobtaineddatagivesa quitefairestimateofthe CIN2+prevalenceinthisgroupofwomen.

The present cohort represents a highly selected groupwith previously identiﬁed persistent HPV infection and no former evidence ofCIN. Currently it isnot clearhow tomanagethese womenandeventhoughapersistentHPVinfectionisariskfactor ofcervicalcancer,veryfewHPV-positivewomenwilldevelopthe disease.AbouthalformoreofHPVinfectionswillclearwithina year[29]astheyalsodidinthisstudy.Itisimportanttoconsider the balance between overtreatment of women with transient infectionsagainsttheriskofnotdetectingthosewomenwhomay develop cervical cancer. Unnecessary treatments are not only uncomfortable but have short-term risks such as infection or bleedingandcaninthelong-termalsocausepretermlabor[30]or cervicalstenosis[31].Knowledgeofhavingapotentiallydangerous HPVinfectioncanalsocauseanxiety[32].However,theremaining HPVinfectionsmaypersistlongerandtheriskofprogressionfrom CINtocancerisconsiderable[33]andespeciallyincaseofTZ3the LEEPmustkeptinmind.

There is still no consensus concerning the deﬁnition of persistencyof anHPVinfection.AColumbianprospectivestudy on HPV persistence proposed the deﬁnition that persistent infectionsarethoselastingmorethanthemedianduration,which wasforexample9,5monthsforHPV16inwomen>30years[34].In alargemeta-analysis,Koshioletal.[2]remarkedthateventesting intervalsofsixmonthsproducestrongsummativerelativerisks asregardstheassociationbetweenHPVpersistenceandCIN2+.Itis therefore suggestedthat repeatHPV testingat six months isa valuable way to identify women at increased risk of cervical precancerandcancer.InourstudythewomenwithCIN2+hada detectedHPVpersistenceforsevento20months.Follow-upinthe

(4)

Table1

IndividualdataonallthewomenwhounderwentLEEPpresentedinorderofCINﬁndingswithhighestgradeﬁrst.

Baselinedataatinvitation Testresultatstudyvisit Followup6

monthsafter studyvisit Age Former excisionfor CIN Smoking Persistent HPVtype Persistency (months) HPV PAP-smear Endocervical cytology TZ Histology inbiopsy Histology inexcision Marginstatus cervical/vaginal Excision height (mm) HPV PAP-smear

59 0 0 16 20 16 normal normal 3 normal CIN3 neg/neg 13 neg normal

56 0 0 31 7 31 normal normal 3 normal CIN3 neg/neg 15 neg normal

41 0 1 39 10 39 normal normal 3 0 CIN3 neg/neg 15 neg normal

59 0 0 33/52/58 10 33/ 52/ 58 normal normal 1 or 2

CIN1 CIN3 pos/pos 13 neg normal

60 0 ? 56 19 56 normal normal 3 0 CIN2 neg/neg 14 neg normal

54 0 1 59 11 59 normal normal 1

or 2

0 CIN2 neg/neg 12 neg normal

59 0 1 16 7 16 normal normal 1

or 2

CIN1 CIN1 neg/neg 12 neg normal

45 0 0 18/45 57 18/

45

normal normal 3 normal CIN1 neg/neg 11 35 0

45 0 0 18/45 11 18/

45

normal normal 3 CIN1 CIN1 neg/neg 13 neg normal

61 0 0 33/52/58 29 33/

52/ 58

normal normal 3 CIN1 CIN1 pos/pos 10 neg normal

71 1 0 33/52/58 18 33/ 52/ 58 normal normal 1 or 2

normal CIN1 neg/neg 9 33/

52/ 58

ASCUS

62 0 0 59 33 neg normal normal 1

or 2

CIN1 CIN1 neg/neg 12 0 0

67 1 0 16 8 neg normal normal 3 CIN1 CIN1 neg/pos 12 neg normal

or 2

normal CIN1 neg/neg 18 0 0

62 0 0 16 4 neg normal 0 3 normal CIN1 neg/neg 12 0 0

60 0 1 39 12 neg normal normal 3 0 CIN1 neg/neg 11 neg normal

58 1 0 56 10 neg normal normal 3 normal CIN1 neg/neg 14 0 0

59 0 ? 56 12 neg normal normal 3 normal CIN1 neg/neg 15 neg normal

47 1 0 33/52/58 12 neg normal normal 1

or 2

normal CIN1 neg/neg 12 0 normal

59 0 0 18/45 12 neg normal normal 1

or 2

normal CIN1 neg/neg 13 neg normal

59 0 0 16 45 16 normal normal 3 normal normal 12 16 normal

68 0 0 16 33 16 normal normal 1

or 2

normal normal 15 16 normal

46 1 0 16 7 16 normal normal 3 normal normal 15 16 normal

56 0 0 39 12 39 normal normal 3 normal normal 8 neg normal

69 0 0 56 29 56 normal normal 3 normal normal 14 56 ASCUS

53 0 1 56 10 56 normal notpossible 3 normal normal 11 neg normal

54 0 0 33/52/58 25 33/

52/ 58

normal normal 3 0 normal 9 33/

52/ 58 normal 62 0 0 33/52/58 93 33/ 52/ 58

normal notpossible 3 normal normal 9 33/

52/ 58

ASCUS

61 0 0 16 44 neg normal normal 3 normal normal 14 0 0

or 2

normal normal 10 0 0

64 0 0 18/45 28 neg normal normal 3 normal normal 11 0 0

or 2 normal normal 12 0 0 41 1 0 51 12 neg normal 0 1 or 2 normal normal 9 0 0

62 0 0 33/52/58 7 neg normal normal 3 normal normal 10 0 0

77 1 0 33/52/58 16 neg normal notpossible 3 normal normal 10 0 0

56 0 0 31 10 neg normal normal 3 normal normal 15 neg normal

(5)

Swedescreenstudypointedoutthatallinitiallycytology-negative womenwithpersistentHPVinfectiondevelopedcolposcopically veriﬁedCIN2+withinsevenyears[11].Thewomeninthatstudy werebetween32and38yearsofageatentry,incontrasttoour studywherethemajoritywerepostmenopausal.Thereisaneed forfurtherstudiesondifferentagegroupstoformaﬁrmbasisfor futuredecisionsonhowtohandlethesepatients.

Conclusion

A persistent HPV infection needs to be monitored despite normalPapsmears,since6/40(15%)womenolderthan40years, was revealed to have an undiagnosed CIN2+ when LEEP was performed.Counselingwomenregardingtheriskofcervicalcancer andtheexpectedeffectofaneventualLEEPcanhelpthemtomake anoptimalinformedchoice.

Conﬂictofintereststatement

Theauthorsdeclarethattherearenoconﬂictsofinterest. Acknowledgement

LionsCancerFoundation,Uppsala,Sweden.GrantID1050023. References

[1]WalboomersJM,JacobsMV,ManosMM,etal.Humanpapillomavirusisa necessarycauseofinvasivecervicalcancerworldwide.JPathol1999;189 (1):12–9.

[2]KoshiolJ,LindsayL,PimentaJM,PooleC,JenkinsD,SmithJS.Persistenthuman papillomavirusinfectionandcervicalneoplasia:asystematicreviewand meta-analysis.AmJEpidemiol2008;168(2):123–37.

[3]OstörAG.Naturalhistoryofcervicalintraepithelialneoplasia:acriticalreview. IntJGynecolPathol1993;12(2):186–92.

[4]MayrandMH,Duarte-FrancoE,RodriguesI,etal.HumanpapillomavirusDNA versusPapanicolaouscreeningtestsforcervicalcancer.NEnglJMed2007;357 (16):1579–88.

[5]Europeanguidelinesforqualityassuranceincervicalcancerscreening.second edition2015.http://www.gisci.it/documenti/news/EW0115451ENN_002.pdf. [6]KatkiHA,KinneyWK,FettermanB,etal.Cervicalcancerriskforwomen

undergoingconcurrenttestingforhumanpapillomavirusandcervical cytology:apopulation-basedstudyinroutineclinicalpractice.LancetOncol 2011;12(7):663–72.

[7]RozendaalL,WalboomersJM,vanderLindenJC,etal.PCR-basedhigh-riskHPV testincervicalcancerscreeninggivesobjectiveriskassessmentofwomenwith cytomorphologicallynormalcervicalsmears.IntJCancer1996;68(6):766–9. [8]KatkiHA,SchiffmanM,CastlePE,etal.Five-yearrisksofCIN3+andcervical

canceramongwomenwhotestPap-negativebutareHPV-positive.JLowGenit TractDis2013;17(5Suppl1):S56–63.

[9]KjaerSK,FrederiksenK,MunkC,IftnerT.Long-termabsoluteriskofcervical intraepithelialneoplasiagrade3orworsefollowinghumanpapillomavirus infection:roleofpersistence.JNatlCancerInst2010;102(19):1478–88. [10]ElfgrenK,RylanderE, RadbergT, etal. Colposcopic andhistopathologic

evaluationofwomenparticipatinginpopulation-basedscreeningforhuman papillomavirusdeoxyribonucleicacidpersistence.AmJObstetGynecol 2005;193(3Pt1):650–7.

[11]ElfgrenK,ElfstromKM,NauclerP,Arnheim-DahlstromL,DillnerJ.Management ofwomenwithhumanpapillomaviruspersistence:long-termfollow-upofa randomizedclinicaltrial.AmJObstetGynecol2017;216(3)e1–7264.

[12]MittalS.Riskofhigh-gradeprecancerouslesionsandinvasivecancersin high-riskHPV-positivewomenwithnormalcervixorCIN1atbaseline-A population-basedcohortstudy.IntJCancer2017;140(8):1850–9.

[13]Gustafsson L, SparenP, Gustafsson M,et al. Lowefﬁciency of cytologic screeningforcancerinsituofthecervixinolderwomen.IntJCancer1995;63 (6):804–9.

[14]CostaS,NuzzoMD,RubinoA,etal.Independentdeterminantsofinaccuracyof colposcopicallydirectedpunchbiopsyofthecervix.GynecolOncol2003;90 (1):57–63.

[15]Gilani SM,Mazzara PF. Cytohistologic correlation inpremenopausal and postmenopausalwomen.ActaCytol2013;57(6):575–80.

[16]PetryKU,LuytenA,ScherbringS.Accuracyofcolposcopymanagementto detectCIN3andinvasivecancerinwomenwithabnormalscreeningtests: resultsfromaprimaryHPVscreeningprojectfrom2006to2011inWolfsburg, Germany.GynecolOncol2013;128(2):282–7.

[17]HoffmanMS,SterghosJrS,GordyLW,GunasekaranS,CavanaghD. Evaluationofthe cervicalcanalwiththeendocervicalbrush.ObstetGynecol1993;82(4Pt1):573–7. [18]BoardmanLA,MeinzH,SteinhoffMM,HeberWW,BlumeJ.Arandomizedtrial

ofthesleevedcytobrushandtheendocervicalcurette.ObstetGynecol 2003;101(3):426–30.

[19]Hopman EH, Kenemans P, Helmerhorst TJ. Positive predictive rate of colposcopicexaminationofthecervixuteri:anoverviewofliterature. ObstetGynecolSurv1998;53(2):97–106.

[20]McCordML,StovallTG, SummittJrRL, Ling FW.Discrepancyof cervical cytologyandcolposcopicbiopsy:iscervicalconizationnecessary?Obstet Gynecol1991;77(5):715–9.

[21]GustavssonI,LindellM,WilanderE,StrandA,GyllenstenU.UseofFTAcardfor drycollection,transportationandstorageofcervicalcellspecimentodetect high-riskHPV.JClinVirol2009;46(2):112–6.

[22]GyllenstenU,GustavssonI,LindellM,WilanderE.Primaryhigh-riskHPV screeningforcervicalcancerinpost-menopausalwomen.GynecolOncol 2012;125(2):343–5.

[23]PolmanNJ,VeldhuijzenNJ,HeidemanDAM,SnijdersPJF,MeijerC,BerkhofJ. HPV-positivewomenwithnormalcytologyremainatincreasedriskofCIN3 afteranegativerepeatHPVtest.BrJCancer2017;117(10):1557–61. [24]PetryKU.Managementoptionsforcervicalintraepithelialneoplasia.BestPract

ResClinObstetGynaecol2011;25(5):641–51.

[25]Schiffman M,Castle PE,JeronimoJ, Rodriguez AC,WacholderS.Human papillomavirusandcervicalcancer.Lancet(London,England)2007;370 (9590):890–907.

[26]StanleyM.PathologyandepidemiologyofHPVinfectioninfemales.Gynecol Oncol2010;117(2Suppl):S5–10.

[27]KuceraE,SliutzG,CzerwenkaK,BreiteneckerG,LeodolterS,ReinthallerA.Is high-riskhumanpapillomavirusinfectionassociatedwithcervical intraepithelialneoplasiaeliminatedafterconizationbylarge-loopexcisionof thetransformationzone?EurJObstetGynecolReprodBiol2001;100(1):72–6. [28]LegoodR,SmithM,LewJB,etal.Costeffectivenessofhumanpapillomavirus testofcureaftertreatmentforcervicalintraepithelialneoplasiainEngland: economicanalysisfromNHSSentinelSitesStudy.BMJ2012;345:e7086. [29]Rodriguez AC,SchiffmanM,HerreroR, et al. Rapidclearance ofhuman

papillomavirusandimplicationsforclinicalfocusonpersistentinfections.J NatlCancerInst2008;100(7):513–7.

[30]Tempfer C, Polterauer S,Grimm C, Bentz EK, Reinthaller A, Heﬂer LA. Endocervicalcytobrushforthedetectionofcervicaldysplasiabeforelargeloop excisionofthetransformationzone(LLETZ).AnticancerRes2008;28 (5B):3131–4.

[31]PennaC,FambriniM,FallaniMG,PieralliA,ScarselliG,MarchionniM.Laser CO2conizationinpostmenopausalage:riskofcervicalstenosisand unsatisfactoryfollow-up.GynecolOncol2005;96(3):771–5.

[32]HendryM,PasterﬁeldD,LewisR,etal.Arewomenreadyforthenewcervical screeningprotocolinEngland?Asystematicreviewandqualitativesynthesis ofviewsabouthumanpapillomavirustesting.BrJCancer2012;107(2):243–54. [33]PlummerM,SchiffmanM,CastlePE,Maucort-BoulchD,WheelerCM.A2-year prospectivestudyofhumanpapillomaviruspersistenceamongwomenwitha cytologicaldiagnosisofatypicalsquamouscellsofundeterminedsigniﬁcanceor low-gradesquamousintraepitheliallesion.JInfectDis2007;195(11):1582–9. [34]MunozN,Hernandez-SuarezG,MendezF,etal.PersistenceofHPVinfection

andriskofhigh-gradecervicalintraepithelialneoplasiainacohortof Colombianwomen.BrJCancer2009;100(7):1184–90.

Table1(Continued)

Baselinedataatinvitation Testresultatstudyvisit Followup6

monthsafter studyvisit Age Former excisionfor CIN Smoking Persistent HPVtype Persistency (months) HPV PAP-smear Endocervical cytology TZ Histology inbiopsy Histology inexcision Marginstatus cervical/vaginal Excision height (mm) HPV PAP-smear