HPV-related external genital lesions among men residing in Brazil

(1)

w w w . e l s e v ie r . c o m / l o c a t e / b j i d

The

Brazilian

Journal

of

INFECTIOUS

DISEASES

Original

article

HPV-related

external

genital

lesions

among

men

residing

in

Brazil

Roberto

Jose

Carvalho

da

Silva

a

_,

_Staci

_Lynn

_Sudenga

b

_,

_Laura

_Sichero

c

_,

Maria

Luiza

Baggio

c

_,

_Lenice

_Galan

d

_,

_Ricardo

_Cintra

e

_,

_Benji

_Nelson

_Torres

b

_,

_Mark

_Stoler

f

_,

Anna

Regina

Giuliano

b

_,

_Luisa

_Lina

_Villa

c,g,∗ a_Centro_de_Referência_e_Treinamento_DST/Aids,_São_Paulo,_SP,_Brazil

b_H_Lee_Mofﬁtt_Cancer_Center_and_Research_Institute,_Department_of_Cancer_{Epidemiology,}_Tampa,_FL,_USA

c_Hospital_das_Clínicas_da_Faculdade_de_Medicina_da_Universidade_de_São_Paulo,_Instituto_do_Câncer_do_Estado_de_São_Paulo,_Centro_de

Investigac¸ãoTranslacionalemOncologia,SãoPaulo,SP,Brazil

d_Instituto_Ludwig_de_Pesquisa_Sobre_o_Câncer,_São_Paulo,_SP,_Brazil

e_Universidade_de_São_Paulo,_Instituto_de_Química,_Departamento_de_Bioquímica,_São_Paulo,_SP,_Brazil f_University_of_Virginia_Health_System,_{Charlottesville,}_VA,_USA

g_Universidade_de_São_Paulo,_Faculdade_de_Medicina,_Departamento_de_Radiologia_e_Oncologia,_São_Paulo,_SP,_Brazil

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received19September2016 Accepted21March2017 Availableonline8April2017

Keywords: Men Genitallesions PeIN Condyloma HPV

a

b

s

t

r

a

c

t

Theaimsofthisstudyweretodeterminetheincidenceofexternalgenitallesions(EGLs), specificallyhistologicallyconfirmedcondyloma(genitalwarts)andPenileIntraepithelial Neoplasia(PeIN),andgenitalHPVinfectionprogressiontoEGLsamonghealthymenaged 18–73residinginBrazil.Subjectsincluded1118menenrolledintheHPVInfectioninMen(HIM) studybetweenJuly2005andJune2009.Ateachvisit,EGLswerebiopsiedandsubjected topathologicalevaluation.HPVstatusingenitalswabsandbiopsieswasdeterminedby LinearArrayandINNO-LiPA,respectively.Age-specificEGLsincidenceandtheproportion andmediantimetoEGLdevelopmentwereestimated.Kaplan–Meiercumulativeincidence ratesat6,12,and24monthsweredetermined.Duringfollow-up,73mendevelopedan incidentEGL.MencoulddevelopmultipleEGLsandtherewere36menwithcondyloma, 27menwithlesionssuggestiveofcondyloma,sixmenwithPeIN,and20menwith non-HPVlesions.HPV-positivemenwhodevelopedEGLswereyounger(p=0.002)thanmenthat didnotdeveloplesions.Amongthe815menwithHPVinfection,4%progressedtoEGL withthesame HPVdetectedintheswab.Duringfollowup,15.7%ofgenitalHPV-6and HPV-11infectionsprogressedtocondyloma(medianprogressiontimeofninemonthsfor HPV-6versus6.8monthsforHPV-11).Approximately1%ofHPV-16infectionsprogressedto PeINwithamedianprogressiontimeof25months.HPVtypescoveredbythe4-valentHPV

∗ _{Corresponding}_author.

E-mailaddress:l.villa@hc.fm.usp.br(L.L.Villa). http://dx.doi.org/10.1016/j.bjid.2017.03.004

(2)

vaccineweredetectedin82.3%and83.3%ofcondylomaandPeIN,respectively.Thehigh burdenofHPVandhighfrequencyofprogressiontodiseaseunderscorestheneedtooffer HPVprophylacticvaccinationtomentoreducetheoverallburdenofinfectionanddiseases causedbyHPV.

Introduction

Humanpapillomavirus(HPV)isthemostcommonsexually transmittedinfectionworldwidewithsexualactivity consti-tuting the major risk factor for infection.1 Most infections areasymptomaticandclearspontaneouslywithintwoyears. However,persistenceofhigh-risk HPVisastrong predictor ofcervical and other HPV-relatedcancers development. To date,morethan200HPVgenotypeshavebeenidentiﬁed,40 ofwhichare routinelydetectedinanogenitalspecimensof bothfemalesandmales.2_Genital_HPV_infection_is_very com-moninmen3_and_infections_may_progress_to_external_genital lesions(EGLs),mostlycondylomataacuminata(genitalwarts orcondyloma)andpenileintraepithelialneoplasia(PeIN)that arethoughttoprecedepenilecancer.4

CondylomaisacommonclinicaloutcomeofHPVinfection inmales,mainly amongmenaged25–29years.5 Nononco-genicHPV-6andHPV-11aretheetiologicagentsofover90%of condylomas.6_Although_about_one_third_of_these_warts sponta-neouslyregress,recurrenceiscommon,therefore,treatment resultsinhighmedicalcosts.7 _The_incidence_of_condyloma amongBrazilianmenispoorlyknown.Whiletherearesome reportsongenitalHPVinfectioninman,8–10_the_incidence_of HPVinfectionandcondylomainBrazilianmenarelimitedto fewpublications.11–13

Penilecanceroccursmorefrequentlyinmenaged50–70 years,withaglobalestimateof22,000casesperyear.1_Though thiscancerisrare, incidenceratesare higherinless devel-opedcountriesrepresentingupto10%oftumorsinmanin somepartsofAfrica,Asia,andSouthAmerica.14_In_Brazil,_this neoplasiaaccountsforabout2%ofallcancersinman,where boththeNorthandNortheastregionsaremostlyaffected.15 Bothbenignandmalignant HPV-relatedtumorsinmenare potentiallypreventablewiththe4-valentHPVvaccine.16,17

ThepurposeofthisstudywastoestimateEGLincidence and assess the rate of progression from HPV infection to histopathologicallyconﬁrmedEGLsinotherwisehealthymen fromBrazilparticipatinginaninternationalHPVnatural his-torystudy,theHPVInfectioninMen(HIM)Study.

Materials

and

methods

Studydesign

TheHIM Study includesover 4,000 men aged18–70 years enrolledbetween2005and 2009inSãoPaulo(Brazil), Cuer-navaca(Mexico),and Tampa,Florida(USA). Studymethods anddesignhavebeendescribedindetailelsewhere.3_In_brief,

everysixmonthsparticipantsunderwentinterview,a physi-calexam,andlaboratoryanalysisofanogenitalspecimensfor HPVdetection.In2009abiopsyandpathologyprotocolwas implementedallowingforpathologicalconﬁrmationandHPV genotypingoflesions.Menwhohadtwoormorestudyvisits aftertheimplementationofthisprotocolwereincludedinthe presentstudy.

In São Paulo, the largest city of Brazil, a total of 1443 menwererecruitedattechnicalandhighereducation institu-tions,incommunities,companiesandmilitarycorporations, throughleaﬂetsposters,lettersande-mailsaddressedto par-ticipantsofotherstudies.Menwerealsoinvitedthroughsome radioandinternetadsandduringcounselingsessionsona publicSTD(sexuallytransmitteddisease)clinic.Menin treat-ment forSTDs or condyloma, andHIV-infected individuals wereexcludedfromparticipatinginthestudy.Retentionrates inBrazilwerethehighestamongallthreesitesparticipatingat theHIMstudy:80%ofallenrolledmencompleted10visitsof follow-upinapproximatelyﬁveyears,accordingtothestudy protocol(six-monthintervalvisits).Thestudywasapproved bytheethicalreviewboardsoftheInstitutionalReviewBoards attheUniversityofSouthFlorida(Tampa,FL,USA),theLudwig InstituteforCancerResearch(SãoPauloBranch,Brazil),Centro deReferênciaeTreinamentoemDST/Aids(SãoPaulo,Brazil), andtheInstitutoNacionaldeSaludPublica(Cuernavaca, Mex-ico).Informedconsentwasobtainedfromallparticipants.

Clinical

specimens,

HPV

detection

and

genotyping

Ateachvisit,threedifferentpre-wettedDacronswabswere usedtosampletheexternalgenitalia(coronalsulcus,glans penis,penileshaft)andscrotumofparticipants,andwerelater combined toforma singlesample.3 _Specimens_underwent DNAextractionusingtheQiagenMediakit(Qiagen,Venlo,The Netherlands),andHPVdetectionandtypingwasconducted usingtheLinearArray® HPVGenotypingTest(Roche Molecu-larDiagnosis,Pleasanton,USA),capableofidentifying37HPV types,classiﬁedashigh-risk(HR-HPV:16,18,31,33,35,39,45, 51,52,56,58,59,68)orlow-risk(LR-HPV:6,11,26,40,42,53, 54,55,61,62,64,66,67,69,70,71,72,73,81,82,IS39,83,84, 89).

Furthermore,ateach clinicvisitparticipantsunderwent clinicalexaminationunder3×lightmagniﬁcationbyatrained clinician in order to detect EGLs. Whenever a lesion was observed, atissue samplewas obtainedbyshaveexcision, ﬁxedandsentforhistopathologicalevaluation.AllEGLsthat appeared to be HPV-related or had an unknown etiology based onvisualinspectionwere sentforHPVtesting.EGLs

(3)

werecategorizedascondyloma,suggestiveofcondyloma(very early condylomas that have some but have not yet devel-oped all of the pathological features of condyloma), PeIN, or not HPV-related (such as skin tags, seborrheic kerato-sis,chronicbalanitis,Molluscumcontagiosum),accordingto criteriapreviouslypublished.13_PeIN_lesions_were_further cat-egorizedinPeINI(low-gradesquamousintraepitheliallesion [SIL]),PeINII(high-gradeSIL),PeINII/III(high-gradeSIL),and PeINIII(high-gradeSIL).DNAsfromformalin-ﬁxed, parafﬁn-embedded(FFPE)tissueswere extracted usingthe QIAamp DNAFFPETissue Kit(Qiagen,Venlo,TheNetherlands),and HPVdetectionandgenotypingwasconductedusingthe INNO-LiPAHPVGenotypingExtraassay(Fujirebio,Göteborg,Sweden) whichdetects28viraltypes(HR-HPV:16,18,31,33,35,39,45, 51,52,56,58,59,68;orLR-HPV:6,11,26,40,43,44,53,54,66, 69,70,71,73,74,82).DNAsfromswabsandtissueswere con-sideredadequateandincludedinthisstudyiftestedpositive for␤-globinoranyHPVgenotype.

Statistical

analysis

EGLincidence

Allmenwithprevalentlesionswereexcludedfromthebiopsy cohort. Age-specific analyses were conducted among men whodevelopedanincidentEGL,andagegroupswerestratified as18–30,31–44,and45–73years.ForEGLincidenceanalyses, onlythefirstdetectedEGLwasconsidered.Timetoincident EGLwascalculatedfrombiopsycohortbaselinedatetothe dateoffirst EGLdetection. Person-time incidenceand 95% confidenceintervals(CIs)werecalculatedbasedonthe num-ber of events modeled as a Poisson variable for the total numberofperson-months.Kaplan–MeiercurvesforEGL inci-dencewere generated, andthe incidenceofEGL overtime acrossthethreeagegroupswascomparedusingthelog-rank test.

ToevaluateHPVdistributioninEGLs,allprevalentand inci-dentlesionswereincluded.InadditiontospeciﬁcHPVtypes, infectionspositivefor≥1typewereincludedintheanyHPV group,thosepositivefor≥1high-riskHPVtypewereincluded inthehigh-riskHPVgroup,andthosepositivefor≥1low-risk typeswereincludedinthelow-riskHPVgroup.Independent analyseswereconductedforhigh-riskandlow-riskinfections. Additionally,EGLsthatwerepositivefor≥1high-risktypeand ≥1low-risktypewereincludedinbothHR/LRHPVgroup.

HPVprogressiontoEGL

Sociodemographicandsexualbehavior characteristicswere describedforHPVpositivemenwhodidordidnotdevelopan EGLduringfollow-up,andcomparisonswereperformedusing theMonteCarloestimationofexactPearsonchi-squaretests. Menwithanincidentor prevalentgenitalHPVinfection and without a prevalent condyloma or PeIN lesion at the biopsyprotocolbaselinevisitwereincludedintheanalyses designedtoassess the rateofprogressionand the propor-tionofinfections progressingtodisease.HPVinfectionwas reportedbygenotypeorgrouped(any,HR-HPV,LR-HPV,and vaccine(HPVs6/11/16/18)).TheclassiﬁcationofanyHPVtype

wasdefinedasapositivetestresultforatleastoneof25(HPV types43/44/74arenotdetectedbytheLinearArrayassay)HPV genotypesdetectedbyINNO-LiPA.HPVinfectionswithsingle ormultipleHR-HPVtypeswereclassifiedasHRandthosewith atleastoneLR-HPVtypewereclassifiedasLR.

Time-to-event approach was appliedto assess the time fromtype-speciﬁcgenitalHPVpositivitytoEGLincidence har-boringthesameHPVtypewithinthelesion.Theanalytical unitforthisstudyisinfection.HPVgenitalinfectionsthatdid notprogresstoEGLwere censoredatthe lastvisit.The6-, 12-,and24-monthcumulativeincidenceofEGLsandmedian timetoEGLdevelopmentforindividualgenitalHPVtypeswere estimatedusingtheKaplan–Meiermethod.Forgrouped gen-ital HPVs, weadjusted forwithin-subject correlation using theclusteredKaplan–Meiermethodasmencouldhavebeen infectedwithmultipleHPVtypeswithinadeﬁnedgroup.The overall EGLincidencerateduringthestudyperiodwasalso calculated.MultipleHPVtypescouldbedetectedinasingle EGL,andamancoulddevelopmultipleEGLs.

Results

EGLincidence

Overall1134BrazilHIMStudyparticipantsaged18–73years were included in the biopsy cohort. Sixteen men with a prevalentEGLwereexcludedfromtheEGLincidence analy-sis(Fig.1).Amongtheremaining1118men,73developedat leastoneincidentEGLduringfollowup.Menprimarily devel-opedcondyloma(n=36),suggestiveofcondyloma(n=27),PeIN (n=6),ornon-HPVrelatedEGL(n=20).

Table 1presents EGL incidence stratiﬁed by age. There wasadeclineinthe12-monthcumulativeincidenceforany EGLbyage(3.4%,1.9%,and0.9%forages18–30,31–44,and 45–73years,respectively;p=0.02;Fig.2).Whenstratiﬁedby lesion type,the 12-monthcumulativecondylomaincidence washigheramongmen<30 yearsofage(1.8%amongmen aged 18–30, p=0.06) compared to older men. In contrast, amongmenwhodevelopedPeIN,the12-monthcumulative incidenceremainedconstantacrosscategoriesofage(0.6%, 0.0%,and0.0%forages18–30,31–44,and45–73years, respec-tively;p=0.38acrosstheentirefollow-upperiod).

Any HPVDNAwas detectedin87.3%ofcondyloma, and 83.5%werepositiveforlow-riskviraltypes.HPV-6andHPV-11 weredetectedin48.1%and36.7%ofcondylomas,respectively (Fig.3).Lessfrequentlydetectedlow-risktypesincludeHPVs 26,40,44,54,66,74.Among48EGLscategorizedas sugges-tiveofcondyloma,72.9%werepositiveforanyHPV,andthe mostcommontypedetectedwasHPV-6(58.3%).Amongthe sixprevalentandincidentPeINlesions,100%wereHPV posi-tiveand66.7%werepositivefor≥1HR-HPVgenotype.HPV-16 andHPV-18weredetectedintwo(33.3%)andone(16.7%)of thePeINlesions,respectively.AdditionallytwootherHR-HPVs werealsodetectedinPeINlesions:HPV-39wasfoundintwo sampleswhereasHPV-68wasdetectedinonesample(data notshown).LR-HPVtypesweremostlyfoundinPeINlesions asco-infectionswithaHR-HPV;howeverHPV-6wasdetected asasingleinfectioninonePeINIandonePeINIIIlesions.PeIN lesionsoccurredatthecoronalsulcus(twocases),theglans

(4)

Men age 18-70 years N=1134

N=1118∗

∗_{Excluded 16 men}

with prevalent EGL

EGL incidence cohort

Men developed EGL N=73

Men did not develop EGL N=1045

Men HPV+ progress to EGL with same HPV types

N=35

Men HPV+ did not progress to EGL

with same HPV types N=780 Men HPV positive

with 2 or more visits N=815

HPV progression to EGL cohort

Fig.1–MenincludedintheHIMStudyBrazilBiopsyCohort.

penis(twocases),themeatusandtheshaftleftdorsal(1case each)(Table1).AtleastoneoftheHPVtypesinthe4-valent HPVvaccinewasdetectedin82.3%,70.8%,and83.3%of condy-loma,suggestiveofcondyloma,andPeINlesions,respectively.

HPVprogressiontoEGL

Among HPV-positive men, EGL incidence was associated only with younger age (p=0.002) (Table 2). No other

KM estimate of EGL development

with number of subjects at risk

12% 11% 10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0% 0 12 24

Months of EGL development

Age group Ages 18-30 Ages 31-44 Ages 45-70 345 550 223 305 494 194 264 422 173 200 330 144 106 176 69 2 6 7

Ages 18-30 Ages 31-44 Ages 45-70

36

Cumulative probability of EGL

development

48 60

Log-rank p=0.0214

Fig.2–Kaplan–Meiercurveshowingdifferencesincumulativeincidenceofexternalgenitallesions(EGLs)overtimebyage group.p-Valuesweredeterminedusingthelog-ranktestanddenotedifferencesacrosstheentirefollow-upperiod.Values <0.05areconsideredstatisticallysigniﬁcant.

(5)

Table1–Age-speciﬁcincidenceofpathologicallyconﬁrmedcondylomaamongmenfromBrazil. Pathologicaldiagnosis

Anya _Condyloma _Suggestiveb _PeINc _Otherd

Allages(n=1118)e

MenwithincidentEGL,no. 58 36 27 6 20

Person-months 41,109 41,760 41,743 42,407 42,079

Incidenceratef_(95%_CI) _1.69_{(1.31–2.19)} _1.03_{(0.75–1.43)} _0.78_{(0.53–1.13)} _0.17_{(0.08–0.38)} _0.57_{(0.37–0.88)}

12-monthincidence 2.2(1.4–3.3) 1.1(0.6–2) 1(0.6–1.9) 0.2(0–0.7) 0.7(0.3–1.4)

18–30y(n=345)

Person-months 12,504 12,794 12,822 13,092 12,895

Incidenceratef_(95%_CI) _2.4_{(1.62–3.55)} _1.59_{(0.99–2.56)} _1.03_{(0.57–1.86)} _0.28_{(0.09–0.85)} _1.12_{(0.63–1.97)}

12-monthincidence 3.4(1.9–6.1) 1.8(0.8–4.1) 1.2(0.5–3.2) 0.6(0.2–2.4) 1.2(0.5–3.2)

31–44y(n=550)

Person-months 20,322 20,646 20,611 20,968 20,851

Incidencerate(95%CI) 1.71(1.19–2.46) 0.93(0.57–1.52) 0.87(0.53–1.45) 0.17(0.06–0.53) 0.35(0.16–0.77)

12-monthincidence 1.9(1–3.5) 0.9(0.4–2.3) 1.1(0.5–2.5) 0.0(0.0–0.0) 0.6(0.2–1.8)

45–73y(n=223)

Person-months 8283 8320 8310 8347 8333

Incidencerate(95%CI) 0.58(0.22–1.54) 0.43(0.14–1.34) 0.14(0.02–1.03) 0.0(0.0–0.0) 0.29(0.07–1.15)

12-monthincidence 0.9(0.2–3.8) 0.5(0.1–3.3) 0.5(0.1–3.3) 0.0(0.0–0.0) 0.0(0.0–0.0)

p-Valueg _0.02 _0.06 _0.10 _0.38 _0.02

a _Men_with_≥1_incident,_{pathologically}_confirmed_HPV-related_EGL_throughout_the_study_period._For_men_with_>1_EGL,_incidence_rates_for_the AnyEGLcategoryaredetermined;forthefirstdetectedlesion;thus,menmaycontributefewerperson-monthsinthiscategorythanfor specificpathologicaldiagnoses.

b _Includes_lesions_suggestive_but_not_diagnostic_of_HPV_infection_or_condyloma. c _Data_concerning_PeIN_specimens_were_previously_described_(Sudenga_et_al.18_).

d_Includes_various_{HPV-unrelated}_benign_skin_conditions,_such_as_seborrheic_keratosis_and_skin_tags. e _Although_the_initial_cohort_included₁₁₃₄_men,₁₆_men_with_prevalent_EGLs_were_excluded_in_this_analysis. f _Speciﬁed_as_the_number_of_cases_per₁₀₀_{person-years.}

g _Determined_using_the_log-rank_test_and_correspond_to_overall_differences_in_EGL_incidence_across_the_entire_follow-up_period,_by_age_group. Values<0.05areconsideredstatisticallysigniﬁcant.CI,conﬁdenceinterval;PeIN,penileintraepithelialneoplasia(I–III).

sociodemographicorsexualbehaviorvariablewasassociated withEGLincidence.

TheHPV-6positivecondylomaincidencerateamongmen withapreviouslydetectedHPV-6infectionwas5.9per1000 person-months (Table 3). Among HPV-6 infections, 7.34% progressed to HPV-6-positive condyloma during the ﬁrst sixmonthsoffollow-up. ForHPV-11, theincidenceratefor

HPV-11-positivecondylomawashigherthanHPV-6at6.7per 1000person-months,andthe6-monthcumulativeincidence wasalsohigherat10.07%(Fig.4).TheHPV-16positivePeIN incidenceratewas0.4per1000person-monthsamongmen with a prior HPV-16 infection. By 24 months of follow-up 0.79%ofmenwithagenital HPV-16infection progressedto anHPV-16-positivePeIN. 100 90 80 70 60 50 40 30 20 10 0

Condyloma (n=79) Suggestive of condyloma (n=48) PeIN (n=6) Other (n=30)

Any HPV HR-HPV HPV -16 HPV -18 LR-HPV HPV -6 HPV -11 HR/LR-HPV _{4-valent HPV} Prevalence, %

Fig.3–PrevalenceofmucosalHumanPapillomavirus(HPV)DNAwithinthetissueofprevalentandincidentexternal genitallesions(EGLs)among89menfromBrazil.

(6)

Table2–Comparisonofsociodemographicandsexualbehaviorcharacteristicsbetweenhumanpapillomavirus-positive menwhodidanddidnotdevelopanexternalgenitallesion(EGL)duringfollow-upintheHIMstudyinBrazil.

NoEGLincidence AnyEGLincidence pa

n(%) n(%) Age 0.002 18–30 240(30.8%) 19(54.3%) 31–44 392(50.3%) 15(42.9%) 45–74 148(19%) 1(2.9%) Race 0.90 White 469(60.1%) 23(65.7%) Black 230(29.5%) 10(28.6%) Asian/PI 14(1.8%) 0(0%) Other 58(7.4%) 2(5.7%) Refused 9(1.2%) 0(0%) Ethnicity 0.68 Hispanic 193(24.7%) 8(22.9%) Non-Hispanic 570(73.1%) 27(77.1%) Refused 17(2.2%) 0(0%) Yearsofeducation 0.82

Completed12yearsorless 379(48.6%) 15(42.9%)

13–15years 150(19.2%) 8(22.9%)

Completedatleast16years 247(31.7%) 12(34.3%)

Refused 4(0.5%) 0(0%) Missing 0(0%) 0(0%) Maritalstatus 0.23 Single 302(38.7%) 19(54.3%) Married/cohabiting 381(48.8%) 12(34.3%) Divorced/separated/widowed 95(12.2%) 4(11.4%) Refused 2(0.3%) 0(0%) Missing Circumcised 0.82 Notcircumcised 645(82.7%) 28(80%) Circumcised 135(17.3%) 7(20%)

Vaginalcondomuse 0.21

Nosex 105(13.5%) 1(2.9%)

Always 140(17.9%) 6(17.1%)

Sometimes 297(38.1%) 18(51.4%)

Never 207(26.5%) 9(25.7%)

Missing 31(4%) 1(2.9%)

Analcondomuse 0.12

NoSex 397(50.9%) 13(37.1%) Always 130(16.7%) 10(28.6%) Sometimes 121(15.5%) 8(22.9%) Never 126(16.2%) 4(11.4%) Missing 6(0.8%) 0(0%) Smokingstatus 0.48 Current 161(20.6%) 10(28.6%) Former 239(30.6%) 9(25.7%) Never 380(48.7%) 16(45.7%) Missing 0(0%) 0(0%)

Alcoholpermonth 1.00

0 155(19.9%) 7(20%) 1–30 314(40.3%) 14(40%) >30 285(36.5%) 13(37.1%) Missing 26(3.3%) 1(2.9%) Sexualorientationa _0.61 MSM 41(5.3%) 1(2.9%) MSMW 210(26.9%) 10(28.6%) MSW 501(64.2%) 24(68.6%) Missing 28(3.6%) 0(0%)

(7)

Table2 –(Continued)

NoEGLincidence AnyEGLincidence pa

n(%) n(%)

Totalnumberoffemalepartners 0.38

0–1 74(9.5%) 4(11.4%)

2–9 167(21.4%) 4(11.4%)

10–49 380(48.7%) 21(60%)

50+ 131(16.8%) 6(17.1%)

Refused 28(3.6%) 0(0%)

Totalnumberofmalepartners 1.00

0 514(65.9%) 24(68.6%)

1–9 164(21%) 7(20%)

10+ 87(11.2%) 4(11.4%)

Missing 15(1.9%) 0(0%)

a _p_values_were_calculated_using_Monte_Carlo_estimation_of_exact_Pearson_chi-square_tests_comparing_{characteristics}_of_men_with_and_without EGL.Missingvalueswerenotincludedinpvaluescalculations.

Table3–Incidenceofcondylomaa_and_penile_{intraepithelial}_neoplasia_(PeIN)a_by_HPV_type_detected_in_the_lesionb_among

menwiththesameHPVtypedetectedatthegenitalsc_among_men_from_Brazil.

HPVtyped _Incidence_rateg_(95%_CI) _Cumulative_incidence_(%)

6m(95%CI) 12m(95%CI) 24m(95%CI)

Condylomaa,e Any 0.5(0.4–0.7) 0.70(0.42–1.15) 1.05(0.69–1.59) 1.74(1.23–2.44) Highrisk 0.1(0.1–0.3) 0.08(0.01–0.57) 0.17(0.04–0.67) 0.39(0.14–1.05) 16 0.4(0.1–1.5) 0.0(0.0–0.0) 0.64(0.09–4.43) 1.36(0.34–5.35) 18 0.4(0.1–2.7) 1.14(0.16–7.79) 1.14(0.16–7.79) 1.14(0.16–7.79) 45 0.4(0.1–3.1) 0.0(0.0–0.0) 0.0(0.0–0.0) 2.13(0.30–14.16) 52 0.2(0.0–1.3) 0.0(0.0–0.0) 0.0(0.0–0.0) 0.0(0.0–0.0) Lowrisk 1.0(0.7–1.5) 1.48(0.88–2.49) 2.17(1.41–3.35) 3.44(2.39–4.93) 6 5.9(3.7–9.4) 7.34(3.73–14.15) 10.27(5.81–17.79) 17.83(11.35–27.39) 11 6.7(3.3–13.3) 10.07(4.31–22.53) 12.31(5.72–25.43) 17.59(9.12–32.4) 40 0.6(0.1–4.0) 1.69(0.24–11.43) 1.69(0.24–11.43) 1.69(0.24–11.43) 53 0.2(0–1.3.0) 0.0(0.0–0.0) 0.0(0.0–0.0) 0.77(0.11–5.33) 66 0.4(0.1–1.5) 0.0(0.0–0.0) 1.3(0.33–5.09) 1.3(0.33–5.09) Vaccineh _2.4_(1.6–3.4) _3.37_{(2.01–5.63)} _4.67_(3–7.24) _7.61_{(5.29–10.90)}

Penileintraepithelialneoplasiaa,f

Any 0.1(0.0–0.2) 0.05(0.01–0.34) 0.05(0.01–0.34) 0.11(0.03–0.44) Highrisk 0.1(0.0–0.2) 0(0–0) 0(0–0) 0.11(0.02–0.78) 16 0.4(0.1–1.5) 0(0–0) 0(0–0) 0.79(0.11–5.5) Lowrisk 0.1(0.0–0.3) 0.11(0.02–0.77) 0.11(0.02–0.77) 0.11(0.02–0.77) 6 0.3(0.0–2.0) 0.92(0.13–6.33) 0.92(0.13–6.33) 0.92(0.13–6.33) 73 0.4(0.1–3.0) 0(0–0) 0(0–0) 0(0–0) Vaccineh _0.2_(0.1–0.7) _0.25_{(0.03–1.74)} _0.25_{(0.03–1.74)} _0.57_{(0.14–2.27)}

CI,conﬁdenceinterval.

a _Newly_acquired,_{pathologically}_conﬁrmed_condyloma_and_PeIN. b _DNA_detected_using_INNO_LiPA.

c _DNA_detected_using_linear_array.

d_Prevalent_and_incident_genital_HPV_infections.

e _HPV_types_{33/35/39/51/56/58/59/68/26/54/69/70/71/73/82}_did_not_progress_to_a_condyloma_lesion;_therefore,_incidence_rates_and_cumulative incidencecouldnotbecalculated.

f _HPV_types_{18/31/33/35/39/45/51/52/56/58/59/68/11/26/40/53/54/66/69/70/71/82}_did_not_progress_to_a_PeIN;_therefore,_incidence_rates_and cumu-lativeincidencecouldnotbecalculated.

g _Incidence_rate_is_cases_per₁₀₀₀_{person-months.} h _Vaccine_HPV_types_6/11/16/18.

(8)

KM estimate of progression from HPV to Condy

with number of subjects at risk

70%

60%

50%

40%

30%

20%

10%

0%

0

12

24 Months of EGL development

HPV type

HPV6 HPV11 HPV16 HPV18 115 51 177 88 90 39 149 79 61 23 118 53 34 10 75 28 13 3 28 13 1 0 0 0 0

HPV 6

HPV 11

HPV 16

HPV 18

36 EGL

development

48

60

72 Log-rank p<0.0001

Fig.4–Kaplan–MeiercurvesshowingprogressionfromHPVtocondyloma.p-Valuesweredeterminedusingthelog-rank testanddenotedifferencesacrosstheentirefollow-upperiod.Values<0.05areconsideredstatisticallysigniﬁcant.

Discussion

TheHIMstudyistheﬁrstinternationalcomparisonofthe dis-tributionof37HPVgenotypesinthegenitalsofmenenrolled inthreedifferent countries. Amongtheseindividuals, HPV prevalenceatthegenitalswashigherinBrazil(72.3%),than intheUSA(61.3%)andMexico(61.9%).3_Although_HPV infec-tionisacommonﬁndinginmen,whenconsideringallHIM participants,only5%oftheseincident viralinfections pro-gressed to and EGL during follow-up.13 _We _herein _present datacomparingEGLincidenceandprogressionofHPV infec-tiontoEGLamongmenenrolledinBrazil.Duringfollow-up 73menfromBrazildevelopedanincidentEGLincluding36 condyloma,sixPeIN,and20non-HPVrelatedEGLs(e.g.skin tags,benignkeratosis).HPV-positivemenwhodevelopedEGLs wereyounger(p=0.002)thanmenthatdidnotdeveloplesions. HPVinfectionwasdetectedin815men;4%ofinfections pro-gressedtoanEGL.Duringthefollowup,15.7%genitalHPV-6 andHPV-11infectionsprogressedtocondyloma(median pro-gressiontimeofninemonthsforHPV-6versus6.8monthsfor HPV-11).Approximately1%ofHPV-16infectionsprogressedto PeINwithamedianprogressiontimeof25months.HPVtypes coveredbythe4-valentHPVvaccineweredetectedin82.3%, 70.8%,and83.3%ofcondyloma,suggestiveofcondyloma,and PeIN,respectively.

Wepreviouslyobservedthat forall menenrolled inthe HIMstudy,factorsstronglyassociatedwithcondyloma devel-opment included age, high lifetime number of female or male sexualpartners.5 _In_the_present _analysis_restricted_to Brazilian men,we observed thatcondyloma incidence sig-niﬁcantly differed onlyby age(p=0.02):younger men(<30) developed moregenital wartseither whenconsidering the wholepopulation18_or_solely_HPV-positive_individuals. Never-theless,itshouldnotbeignoredthatindividualsaged45–70 yearscontinuedtodevelopcondylomaduringfollow-upwhich highlights thatmenremainsusceptibletodeveloping HPV-relatedEGLsthroughouttheirlives.InthecurrentstudyHPV DNAwasdetectedin87.3%ofcondylomabiopsies,inlinewith previousresultsof91%HPVprevalenceincondylomasamples ofwomenenrolledintheplaceboarmofavaccinetrial.19 HPV-6andHPV-11werethemostprevalentviraltypesdetectedin condylomasinagreementwithpreviousreports,20–23_although HPV-6detectionsurpassedthatofHPV-11(48.1%versus36.7%) inthesesamples.

About16%ofHPV-6andHPV-11infectionsprogressedtoa condylomawiththesameviraltypeduringfollow-up;however mediantimetoprogressionwashigheramongHPV-6positive men(ninemonthsforHPV-6versus6.8monthsforHPV-11). Ourresultsareinlinewithpreviousestimatesoftwoweeksto eightmonthsasthetimerequiredfrominitialHPV-6orHPV-11 infectionstoprogressiontocondyloma.7

(9)

Although PeIN was a rare event in this population, we wereabletodetectandgenotypesixprevalentandincident PeINlesions.Histopathologically,threecaseswereclassiﬁed asPeINIII,onecaseasPeINII,andtwocasesasPeINI.All PeINlesionswereHPV-positive;HPV-16wasdetectedintwo cases,inagreementwithpreviousstudiesreporting88–100% ofsamplespositiveforHPVDNA,andabout40%ofPeIN posi-tiveforHPV-16.24_It_is_interesting_to_notice_that_genital_swabs collectedpriortoHPV-16positivePeINharboredmultipleHPV infections.AmongBrazilianmenenrolledintheHIMstudy, 1.1%ofHPV-16infectionsprogressedtoHPV-16-positivePeIN inamediantimeof25.6months.AmongLR-HPVgenital infec-tions,oneHPV-6andoneHPV-73infectionsprogressedtoPeIN. However,whereas theHPV-6infectionprogressedtoaPeIN within6.7months,theHPV-73infectionprogressionoccurred after30.5months.Similarresultsareobservedwhenpooling dataacrossthethreeparticipatingcountriesoftheHIMstudy, namelyBrazil,Mexico,andUSA:1.6%ofHPV-16infections pro-gressedtoPeINinamediantimeof19months.13_In_addition, inoneofthePeINcases,wedetectedexclusivelyHPV-6and timefrominfectiontoprogressionwas6.7months.The pres-enceoflow-riskHPVsinhigh-gradeprecancerousanogenital lesionshasbeenalsoobservedbyothers.25–27 _{Nevertheless,} thelimitednumberofPeINdetectedlimitsfurther consider-ations.

Considering the HIM study cohort as a whole, during all six years of follow-up, 10 incident PeIN lesions were detected,fromtheUSA(two)andfromMexicoandBrazil(four each).13 _Regional _differences _concerning _the _rates _of HPV-relatedtumorsinthesecountriesarereported.InBrazilthe incidencerateofpenilecanceris0.7–2.3/100,000inhabitants, oneofworld’s highestincidencerates forthis neoplasia.28 Thecity ofRecife inPernambuco, Northeast region ofthe countryhasanincidencerateofpenilecancerof4.6/100,000 inhabitants.Unfortunately,thereisstillveryincompletedata availableinregardstoHPVprevalenceinpenilecancerandits precursorlesionsinBrazil.Inonestudy,HR-HPVwasdetected intwocasesofcondylomaandtwocasesofPeIN(onePeIN IandonePeINIII),whereasjustonecaseofPeINIwasHPV negativebyHybridCapture.29

Aswithanyepidemiological study,there arelimitations andstrengthstothepresentstudy.Althoughmostprevious reportsofHPVincidenceinEGLswerebasedondetectionfrom thesurfaceofthelesion,thisanalysistookadvantageofHPV detectionwithinbiopsytissues,whichismorelikelyto repre-senttheviraltypepresentinthelesion.30_Further_strengths_of thisstudyincludetheclinicaldatafromalargesamplesizeof menfollowedforgenitaldiseaseinBrazilwithveryhigh reten-tionrates(>80%atﬁveyearsoffollow-up).Moreover,menwere followed-upeverysixmonthstoallowforearlydetectionof lesions,whichwereavailableforhistopathologicalevaluation andHPVgenotyping.Nevertheless,ourstudymaybe under-estimatingHPV infection progressiontocondyloma, which could haveoccurred duringthe sixmonths betweenstudy visits.

Takentogether,approximately 16%ofgenital HPV-6and HPV-11infectionsprogressedtocondylomaswiththesame HPVtypewhilelessthan1%ofgenitalHPV-16infections pro-gressedtoanHPV-16-positivePeINamongBrazilianmen.In addition, mencontinuetobeatriskofEGLsacross allage

groups.Thisstudyemphasizestheneedforvaccine-related immunityinmalesandforthatimmunitytobelonglasting acrossthelifespantoensureareductionintheoverallburden ofinfectionanddiseasescausedbyHPV.

The4-valentvaccineoffersprotectionagainstHPVs6,11, 16, and 18has been demonstrated tobeefficacious in the reductionofHPV-relateddiseasesinmen.16_In_our_series,_HPV vaccinetypesweredetectedin82.3%and83.3%incondyloma andPeINcases,respectively.Todate,fewcountriesglobally havenationalvaccine programsthat includemale vaccina-tion, specifically the USA, Austria, Israel, Switzerland, and Australia.AsmaleHPVinfectionsignificantlycontributesto infectionandsubsequentcervicaldiseaseinwomenand dis-easeinmen,ourdatasupporttheimplementationinBrazil ofgender-neutralnationalimmunizationpolicies,crucialto reducetheoverallHPVinfectionburdenandultimately, dis-easeinbothwomenandmen.

Ethical

approval

Ethical approval was given by the University of South Florida(IRB#102660),CentrodeReferênciaeTreinamentoem DST/AIDS,SãoPaulo,Brazil,andInstitutoNacionaldeSalud PúblicadeMéxico.

Funding

statement

This work was supportedbythe Ludwig Institute for Can-cerResearch;FundaçãodeAmparoàPesquisadoEstadode São Paulo(FAPESP)[Grantnumber 08/57889-1 toLLV]; Con-selhoNacionaldeDesenvolvimentoCientíficoeTecnológico (CNPq)[Grantnumber573799/2008-3toLLV].The infrastruc-tureoftheHIMStudycohortwassupportedthroughagrant fromtheNationalCancerInstitute(NCI),NationalInstitutes ofHealth(NIH)(R01CA098803toARG).Merck&Co.provided financialsupportforthedataanalysistoFundaçãoFaculdade deMedicinadaUniversidadedeSãoPauloonbehalfofLLV.

Conﬂicts

of

interest

ARG (IISP39582)andSLS(IISP53280)arecurrentrecipientof grantfundingfromMerck,andARGandLLVaremembersof theMerckAdvisoryBoardforHPVprophylacticvaccines.RJCS receivedtravelgrantsfromMerck,Sharp&Dohme.No con-ﬂictsofinterestaredeclaredforanyoftheremainingauthors.

Acknowledgements

Thisresearchwassupportedinpartbyresearchfundingfrom MerckSharp&DohmeCorp.Theopinionsexpressedinthis paperarethoseoftheauthorsanddonotnecessarily repre-sentthoseofMerckSharp&DohmeCorp.

The authors would like to thank the HIM StudyTeams intheUSA(MofﬁttCancerCenter,Tampa,FL),Brazil(Centro deReferênciaeTreinamentoemDST/AIDS,Fundac¸ão Facul-dadedeMedicina,InstitutodoCâncerdoEstadodeSãoPaulo, LudwigInstituteforCancerResearch,SãoPaulo)andMexico

(10)

(InstitutoMexicanodelSeguroSocial,InstitutoNacionalde SaludPública,Cuernavaca).

r

e

f

e

r

e

n

c

e

s

1. DeMartelC,FerlayJ,FranceschiS,etal.Globalburdenof cancersattributabletoinfectionsin2008areviewand syntheticanalysis.LancetOncol.2012;13:607–15. 2. BouvardV,BaanR,StraifK,etal.Areviewofhuman

carcinogens-PartB:biologicalagents.LancetOncol. 2009;10:321–2.

3. GiulianoAR,Lazcano-PonceE,VillaLL,etal.Thehuman papillomavirusinfectioninmenstudyhumanpapillomavirus prevalenceandtypedistributionamongmenresidingin Brazil,Mexico,andtheUnitedStates.CancerEpidemiol BiomarkersPrev.2008;17:2036–43.

4. CubillaAL,VelazquezEF,YoungRH.Epitheliallesions associatedwithinvasivepenilesquamouscellcarcinomaa pathologicstudyof288cases.IntJSurgPathol.

2004;12:351–64.

5. AnicGM,LeeJH,VillaLL,etal.Riskfactorsforincident condylomainamultinationalcohortofmentheHIMstudy.J InfectDis.2012;205:789–93.

6. IARCMonographsAreviewofhumancarcinogens.PartB: Biologicalagents/IARCWorkingGroupontheEvaluationof CarcinogenicRiskstoHumans,(2012:Lyon,France). 7. StebenM,GarlandSM.Genitalwarts.BestPractResClin

ObstetGynaecol.2014;28:1063–73.

8. NicolauSM,CamargoCG,StávaleJN,etal.Human papillomavirusDNAdetectioninmalesexualpartnersof womenwithgenitalhumanpapillomavirusinfection. Urology.2005;65:251–5.

9. RombaldiRL,SeraﬁniEP,VillaLL,etal.Infectionwithhuman papillomavirusesofsexualpartnersofwomenhavingcervical intraepithelialneoplasia.BrazJMedBiolRes.2006;39:177–87. 10.SilvaRJ,CassebJ,AndreoliMA,VillaLL.Persistenceand

clearanceofHPVfromthepenisofmeninfectedand non-infectedwithHIV.JMedVirol.2011;83:127–31. 11.MoreiraEDJr,GiulianoAR,PalefskyJ,etal.Incidence,

clearance,anddiseaseprogressionofgenitalhuman papillomavirusinfectioninheterosexualmen.JInfectDis. 2014;210:192–9.

12.GiulianoAR,LeeJH,FulpW,etal.Incidenceandclearanceof genitalhumanpapillomavirusinfectioninmen(HIM):a cohortstudy.Lancet.2011;377:932–40.

13.SudengaSL,InglesDJ,PierceCampbellCM,etal.Genital humanpapillomavirusinfectionprogressiontoexternal genitallesions:theHIMstudy.EurUrol.2016;69:166–73. 14.ParkinDM,AlmonteM,BruniL,CliffordG,CuradoMP,Pineros

M.Burdenandtrendsoftype-speciﬁchumanpapillomavirus infectionsandrelateddiseasesintheLatinAmericaand Caribbeanregion.Vaccine.2008;26Suppl.11:L1–15.

15.FavoritoLA,NardiAC,RonalsaM,ZequiSC,SampaioFJ,Glina S.EpidemiologicstudyonpenilecancerinBrazil.IntBrazJ Urol.2008;34:587–91.

16.GiulianoAR,PalefskyJM,GoldstoneS,etal.Efﬁcacyof quadrivalentHPVvaccineagainstHPVInfectionanddisease inmales.NEnglJMed.2011;364:401–11.

17.VillaLL,CostaRL,PettaCA,etal.Prophylacticquadrivalent humanpapillomavirus(types6,11,16,and18)L1virus-like particlevaccineinyoungwomenarandomiseddouble-blind placebo-controlledmulticentrephaseIIefﬁcacytrial.Lancet Oncol.2005;6:271–8.

18.SudengaSN,TorresB,FulpW,etal.Countryspeciﬁc HPV-relatedgenitallesions,amongmenresidinginBrazil, MexicoandtheUnitesStates:HIMstudy.(abstractSS12-8). In:Programandabstracts:Eurogin2016(Salzburg),2016. 19.GarlandSM,StebenM,SingsHL,etal.Naturalhistoryof

genitalwarts:analysisoftheplaceboarmof2randomized phaseIIItrialsofaquadrivalenthumanpapillomavirus (types6,11,16,and18)vaccine.JInfectDis.2009;15:805–14. 20.BrownDR,SchroederJM,BryanJT,StolerMH,FifeKH.

Detectionofmultiplehumanpapillomavirustypesin Condylomataacuminatalesionsfromotherwisehealthyand immunosuppressedpatients.JClinMicrobiol.

1999;37:3316–22.

21.ChanPK,LukAC,LukTN,etal.Distributionofhuman papillomavirustypesinanogenitalwartsofmen.JClinVirol. 2009;44:111–4.

22.FormanD,DeMartelC,LaceyCJ,etal.Globalburdenof humanpapillomavirusandrelateddiseases.Vaccine.2012;30 Suppl.5:F12–23.

23.InglesDJ,PierceCampbellCM,MessinaJA,etal.HPV genotypeandage-speciﬁcanalysesofexternalgenitallesions amongmenintheHPVinfectioninmen(HIM)study.JInfect Dis.2015;211:1060–7.

24.AlemanyL,CubillaA,HalecG,etal.Penilecancer:roleof humanpapillomavirusinpenilecarcinomasworldwide.Eur Urol.2016,pii:S0302-2838(15)01215-4.

25.AynaudO,IonescoM,BarrassoR.Penileintraepithelial neoplasia.Speciﬁcclinicalfeaturescorrelatewithhistologic andvirologicﬁndings.Cancer.1994;74:1762–7.

26.KrustrupD,JensenHL,VanDenBruleAJ,etal.Histological characteristicsofhumanpapilloma-virus-positiveand negativeinvasiveandinsitusquamouscelltumoursofthe penis.IntJExpPathol.2009;90:182–9.

27.RubinMA,KleterB,ZhouM,etal.Detectionandtypingof humanpapillomavirusDNAinpenilecarcinoma:evidencefor multipleindependentpathwaysofpenilecarcinogenesis.Am JPathol.2001;159:1211–8.

28.BruniL,Barrionuevo-RosasL,AlberoG,etal.ICOinformation centreonHPVandcancer(HPVInformationCentre)human papillomavirusandrelateddiseasesinBrazil.Summary Report.2015:12–23.

29.GiraldoPC,EleutérioJJr,CavalcanteDI,Gonc¸alvesAK,Romão JA,EleutérioRM.Theroleofhigh-riskHPV-DNAtestinginthe malesexualpartnersofwomenwithHPV-inducedlesions. EurJObstetGynecolReprodBiol.2008;137:88–91.

30.PierceCampbellCM,MessinaJL,StolerMH,etal.Cutaneous humanpapillomavirustypesdetectedonthesurfaceofmale externalgenitallesionsacaseserieswithintheHPVinfection inmenstudy.JClinVirol.2013;58:652–9.