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,∗ aCentrodeReferênciaeTreinamentoDST/Aids,SãoPaulo,SP,BrazilbHLeeMoffittCancerCenterandResearchInstitute,DepartmentofCancerEpidemiology,Tampa,FL,USA
cHospitaldasClínicasdaFaculdadedeMedicinadaUniversidadedeSãoPaulo,InstitutodoCâncerdoEstadodeSãoPaulo,Centrode
Investigac¸ãoTranslacionalemOncologia,SãoPaulo,SP,Brazil
dInstitutoLudwigdePesquisaSobreoCâncer,SãoPaulo,SP,Brazil
eUniversidadedeSãoPaulo,InstitutodeQuímica,DepartamentodeBioquímica,SãoPaulo,SP,Brazil fUniversityofVirginiaHealthSystem,Charlottesville,VA,USA
gUniversidadedeSãoPaulo,FaculdadedeMedicina,DepartamentodeRadiologiaeOncologia,SãoPaulo,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
∗ Correspondingauthor.
E-mailaddress:l.villa@hc.fm.usp.br(L.L.Villa). http://dx.doi.org/10.1016/j.bjid.2017.03.004
1413-8670/©2017SociedadeBrasileiradeInfectologia.PublishedbyElsevierEditoraLtda.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).
vaccineweredetectedin82.3%and83.3%ofcondylomaandPeIN,respectively.Thehigh burdenofHPVandhighfrequencyofprogressiontodiseaseunderscorestheneedtooffer HPVprophylacticvaccinationtomentoreducetheoverallburdenofinfectionanddiseases causedbyHPV.
©2017SociedadeBrasileiradeInfectologia.PublishedbyElsevierEditoraLtda.Thisis anopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/ licenses/by-nc-nd/4.0/).
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.2GenitalHPVinfectionisvery com-moninmen3andinfectionsmayprogresstoexternalgenital lesions(EGLs),mostlycondylomataacuminata(genitalwarts orcondyloma)andpenileintraepithelialneoplasia(PeIN)that arethoughttoprecedepenilecancer.4
CondylomaisacommonclinicaloutcomeofHPVinfection inmales,mainly amongmenaged25–29years.5 Nononco-genicHPV-6andHPV-11aretheetiologicagentsofover90%of condylomas.6Althoughaboutonethirdofthesewarts sponta-neouslyregress,recurrenceiscommon,therefore,treatment resultsinhighmedicalcosts.7 Theincidenceofcondyloma amongBrazilianmenispoorlyknown.Whiletherearesome reportsongenitalHPVinfectioninman,8–10theincidenceof HPVinfectionandcondylomainBrazilianmenarelimitedto fewpublications.11–13
Penilecanceroccursmorefrequentlyinmenaged50–70 years,withaglobalestimateof22,000casesperyear.1Though thiscancerisrare, incidenceratesare higherinless devel-opedcountriesrepresentingupto10%oftumorsinmanin somepartsofAfrica,Asia,andSouthAmerica.14InBrazil,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.3Inbrief,
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 Specimensunderwent 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
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.13PeINlesionswerefurther 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
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.
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)
MenwithincidentEGL,no. 25 17 11 3 12
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)
MenwithincidentEGL,no. 29 16 15 3 6
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)
MenwithincidentEGL,no. 4 3 1 0 2
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 Menwith≥1incident,pathologicallyconfirmedHPV-relatedEGLthroughoutthestudyperiod.Formenwith>1EGL,incidenceratesforthe AnyEGLcategoryaredetermined;forthefirstdetectedlesion;thus,menmaycontributefewerperson-monthsinthiscategorythanfor specificpathologicaldiagnoses.
b IncludeslesionssuggestivebutnotdiagnosticofHPVinfectionorcondyloma. c DataconcerningPeINspecimenswerepreviouslydescribed(Sudengaetal.18).
dIncludesvariousHPV-unrelatedbenignskinconditions,suchasseborrheickeratosisandskintags. e Althoughtheinitialcohortincluded1134men,16menwithprevalentEGLswereexcludedinthisanalysis. f Specifiedasthenumberofcasesper100person-years.
g Determinedusingthelog-ranktestandcorrespondtooveralldifferencesinEGLincidenceacrosstheentirefollow-upperiod,byagegroup. 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.
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%)
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 pvalueswerecalculatedusingMonteCarloestimationofexactPearsonchi-squaretestscomparingcharacteristicsofmenwithandwithout EGL.Missingvalueswerenotincludedinpvaluescalculations.
Table3–Incidenceofcondylomaaandpenileintraepithelialneoplasia(PeIN)abyHPVtypedetectedinthelesionbamong
menwiththesameHPVtypedetectedatthegenitalscamongmenfromBrazil.
HPVtyped Incidencerateg(95%CI) Cumulativeincidence(%)
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 Newlyacquired,pathologicallyconfirmedcondylomaandPeIN. b DNAdetectedusingINNOLiPA.
c DNAdetectedusinglineararray.
dPrevalentandincidentgenitalHPVinfections.
e HPVtypes33/35/39/51/56/58/59/68/26/54/69/70/71/73/82didnotprogresstoacondylomalesion;therefore,incidenceratesandcumulative incidencecouldnotbecalculated.
f HPVtypes18/31/33/35/39/45/51/52/56/58/59/68/11/26/40/53/54/66/69/70/71/82didnotprogresstoaPeIN;therefore,incidenceratesand cumu-lativeincidencecouldnotbecalculated.
g Incidencerateiscasesper1000person-months. h VaccineHPVtypes6/11/16/18.
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 0HPV 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%).3AlthoughHPV 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 Inthepresent analysisrestrictedto Brazilian men,we observed thatcondyloma incidence sig-nificantly differed onlyby age(p=0.02):younger men(<30) developed moregenital wartseither whenconsidering the wholepopulation18orsolelyHPV-positiveindividuals. Never-theless,itshouldnotbeignoredthatindividualsaged45–70 yearscontinuedtodevelopcondylomaduringfollow-upwhich highlights thatmenremainsusceptibletodeveloping HPV-relatedEGLsthroughouttheirlives.InthecurrentstudyHPV DNAwasdetectedin87.3%ofcondylomabiopsies,inlinewith previousresultsof91%HPVprevalenceincondylomasamples ofwomenenrolledintheplaceboarmofavaccinetrial.19 HPV-6andHPV-11werethemostprevalentviraltypesdetectedin condylomasinagreementwithpreviousreports,20–23although 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
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-16.24Itisinterestingtonoticethatgenitalswabs 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.13Inaddition, 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.30Furtherstrengthsof 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.16Inourseries,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;Fundac¸ãodeAmparoàPesquisadoEstadode 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 financialsupportforthedataanalysistoFundac¸ãoFaculdade 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
(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.