Cost-effectiveness of human papillomavirus vaccination in Germany.

Damm e al. Cos E Resou Alloc (2017) 15:18 DOI 10.1186/s12962-017-0080-9 RESEARCH Cos -e ec i eness o human papilloma i us accina ion inGe many Oli e Damm1* , Johannes Ho n2, Ra ael T. Mikolajczyk2,3,4, Mi jam E. E. K e zschma 5,6, And eas M. Kau mann7, Y onne Dele é8, Be nha d Ul sch9, Ole Wichmann9, Alexande K äme 10 and Wol gang G eine 1 Abs ac Backg ound: The aim o his s udy was o assess he cos -e ec i eness o human papilloma i us (HPV) accina ion in addi ion o he cu en ce ical cance sc eening p og amme in Ge many using a dynamic ansmission model. Me hods: Based on a ma hema ical model simula ing he ansmission dynamics and he na u al his o y o HPV in ec ion and associa ed diseases (ce ical in aepi helial neoplasia, ce ical cance , and geni al wa s), we es ima ed he epidemiological and economic consequences o HPV accina ion wi h bo h he quad i alen and bi alen ac- cines. In ou base case analysis, we assessed he cos -e ec i eness o accina ing 12-yea -old gi ls wi h a 3-dose schedule. In sensi i i y analysis, we also e alua ed he use o a 2-dose schedule and assessed he impac o accina - ing boys. Resul s: F om a heal h ca e paye pe spec i e, inc emen al cos -e ec i eness a ios (ICERs) o a 3-dose schedule we e €34,249 pe quali y-adjus ed li e yea (QALY) o he bi alen and €14,711 pe QALY o he quad i alen accine. Inclusion o indi ec cos s dec eased ICERs by up o 40%. When adop ing a heal h ca e paye pe spec i e, ICERs o a 2-dose app oach dec eased o €19,450 pe QALY o he bi alen and o €3645 pe QALY o he quad i alen accine. F om a socie al pe spec i e, a 2-dose app oach using he quad i alen accine was a cos -sa ing s a egy while using he bi alen accine esul ed in an ICER o €13,248 pe QALY. I espec i e o he pe spec i e adop ed, addi ional ac- cina ion o boys esul ed in ICERs exceeding €50,000 pe QALY, excep o scena ios wi h low co e age (20%) in gi ls. Conclusions: Ou model esul s sugges ha ou ine HPV accina ion o 12-yea -old gi ls wi h h ee doses is likely o be cos -e ec i e in Ge many. Due o he addi ional impac on geni al wa s, he quad i alen accine appea ed o be mo e cos -e ec i e han he bi alen accine. A 2-dose schedule o he quad i alen accine migh e en lead o cos sa ings when adop ing a socie al pe spec i e. The cos -e ec i eness o addi ional accina ion o boys was highly dependen on he co e age in gi ls. Keywo ds: HPV, Vaccina ion, Economic e alua ion, Cos -e ec i eness, Dynamic ansmission model, Ge many © The Au ho (s) 2017. This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/ publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Backg ound Pe sis en in ec ion wi h high- isk (oncogenic) ypes o human papilloma i us (HPV) is he main cause o ce i- cal cance and i s p ecu so s [1]. In ec ion wi h high- isk HPV ypes can also cause o he ypes o anogeni al can- ce (i.e. aginal, ul a , anal, and penile cance ) and o o- pha yngeal cance [2, 3], whe eas in ec ion wi h low- isk (non-oncogenic) HPV ypes is p ima ily associa ed wi h geni al wa s [4]. Clinical s udies ha e demons a ed high e icacy o p o- phylac ic HPV accines in he p e en ion o HPV in ec- ions, p emalignan anogeni al lesions (ce ical, aginal, ul a , and anal lesions), and geni al wa s [5–7]. In Ge - many, he e a e cu en ly h ee HPV accines a ailable o p e en HPV in ec ion and ela ed diseases: a bi alen accine (Ce a ix®), a quad i alen accine (Ga dasil®), and a 9- alen accine (Ga dasil® 9). All h ee accines p o ec agains he high- isk geno ypes 16 and 18, which cause app oxima ely 70% o all ce ical cance cases Open Access Cos E ec i eness and Resou ce Alloca ion *Co espondence: oli e [email p o ec ed] 1 Depa men o Heal h Economics and Heal h Ca e Managemen , School o Public Heal h, Biele eld Uni e si y, Uni e si ä ss aße 25, 33615 Biele eld, Ge many Full lis o au ho in o ma ion is a ailable a he end o he a icle Page 2 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 [8]. The quad i alen accine is addi ionally di ec ed agains he low- isk geno ypes 6 and 11, which accoun o app oxima ely 90% o geni al wa s [9]. The ecen ly app o ed 9- alen e sion o Ga dasil® p o ec s agains he ou s ains o he quad i alen accine and i e addi- ional high- isk s ains (31, 33, 45, 52, and 58). All a ail- able accines a e licensed o use in emales and males. In Ge many, ou ine HPV accina ion wi h wo doses is cu en ly ecommended o emales aged 9–14yea s [10, 11]. Howe e , h ee doses a e necessa y when ac- cina ing adolescen gi ls aged 15–17 yea s (due o delayed ini ia ion o se ies comple ion) o when he in e al be ween he i s and he second dose alls below 6mon hs. Be o e Augus 2014, h ee HPV accine doses we e ecommended o emales aged 12–17yea s. Bo h ecommenda ions ha e p ima ily aimed a p e en ing ce ical cance . Since 1971, a ce ical cance sc eening p og amme is implemen ed in Ge many, which is cu - en ly based on an annual Pap smea beginning a he age o 20yea s. No independen ly unded dynamic ansmission model analysing he cos -e ec i eness o HPV accina ion in he Ge man heal h ca e se ing, which can suppo he na ional decision-making p ocess o he S anding Vacci- na ion Commi ee (STIKO), has been published ye . The pu pose o his s udy was h ee old: (a) o assess he cos -e ec i eness o he bi alen and quad i a- len accines in addi ion o he exis ing ce ical can- ce sc eening p og amme in Ge many using a dynamic ansmission model, (b) o quan i y he economic impac o swi ching om a 3-dose o a 2-dose schedule, and (c) o compa e ou esul s wi h indings o p e iously pub- lished s udies on he cos -e ec i eness o HPV accina- ion in Ge many. Me hods O e iew We de eloped a ma hema ical model simula ing he ansmission dynamics and he na u al his o y o HPV in ec ion and associa ed diseases o e alua e he epi- demiological and economic consequences o HPV ac- cina ion in Ge many. The age-s uc u ed model akes accoun o he occu ence o ce ical in aepi helial neoplasia (CIN), ce ical cance , and geni al wa s. I was calib a ed using Ge man cance s a is ics and o he da a. In he base case analysis, which was conduc ed om bo h a heal h ca e paye and a socie al pe spec i e, HPV accina ion o 12yea old gi ls in conjunc ion wi h he exis ing cy ological sc eening p og amme was com- pa ed o sc eening alone. The ime ho izon was se o 100yea s a e he in oduc ion o HPV accina ion. Epi- demiological and economic pa ame e es ima es we e ob ained om published li e a u e and supplemen ed by expe in e iews. The model was p og ammed using he so wa e R. All cos calcula ions we e pe o med using Mic oso Excel. De ails on he me hods and da a sou ces used o cons uc he model a e desc ibed in he ollowing sec ions. A mo e de ailed desc ip ion o he unde lying epidemiological model and he co espond- ing model pa ame e s ega ding he ansmission o HPV, he na u al his o y o ce ical HPV in ec ion, and he simula ed sc eening p og amme has been published p e iously [12]. Model s uc u e The basic model s uc u e combines an age-s uc u ed de e minis ic compa men al model, which simula es he sexual ansmission o HPV, wi h a model ha ep esen s he na u al his o y o ce ical cance . The ansmission dynamics a e desc ibed by a SIRS- model whe e he popula ion is di ided in o suscep ible indi iduals (S), in ec ious indi iduals (I), and eco e ed (and he e o e immune) indi iduals (R). Ou model con- side s six g oups o i al s ains: HPV 16, HPV 18, phy- logene ically ela ed high- isk ypes wi h po en ial o c oss-p o ec ion (HPV 31/33/35/39/45/51/52/56/58/59), o he high- isk ypes wi hou po en ial o c oss-p o ec- ion, HPV 6/11, and o he low- isk ypes. We neglec ed any syne gis ic and an agonis ic in e ac ions be ween di - e en HPV ypes as mul i-s ain in e ac ions a e subjec o con o e sy. To es ima e he long- e m consequences o HPV in ec- ion, he ansmission model was complemen ed by a module co e ing he ce ical ca cinogenesis. This pa o he model includes ype-speci ic heal h s a es o ha ing CIN o h ee g ades (CIN 1, CIN 2, o CIN 3), ha ing ca - cinoma insi u (CIS), and ha ing in asi e ce ical cance o di e en s ages acco ding o he In e na ional Fede a- ion o Gynecology and Obs e ics (FIGO) classi ica ion sys em. CIS and in asi e cance s a es we e u he sub- di ided in o unde ec ed and de ec ed disease s a es. Fo women who had unde gone hys e ec omy, only ansi- ions be ween he suscep ible, in ec ed, and immune ( eco e ed) s a es we e pe mi ed. The e o e, wi h he excep ion o sexual mixing and backg ound mo ali y, hose women we e modelled iden ically o males who we e only allowed o mo e be ween he suscep ible, in ec ed, and immune s a es. The occu ence o geni- al wa s in bo h gende s was modelled as an e en ha was linked wi h inciden in ec ions wi h low- isk ypes, bu only a p opo ion o hose newly in ec ed indi idu- als we e assumed o de elop clinical symp oms o geni al wa s and o seek medical ea men . A simpli ied low diag am o he model s uc u e ep esen ing he na u al his o y o HPV in ec ion and ce ical cance in women is ou lined in Fig.1. Page 3 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 Da a sou ces andmodel inpu s Demog aphic, beha iou al, and ansmission pa ame e s The model popula ion is based on a cons an numbe o one million bi hs pe yea and he 2008 mo ali y a es in Ge many de i ed om he Fede al S a is ical O ice. Ou model ocusses on he e osexual ansmission o HPV in ec ion only. We assumed he sexual debu o be a he age o 12yea s and di ided he sexually ac i e pop- ula ion in o en age g oups (12–13, 14–15, 16–17, 18–24, 25–34, 35–44, 45–54, 55–64, 65–74, and 75–100yea s) and h ee sexual ac i i y-based isk g oups (low, mod- e a e, and high ac i i y). The p opo ions o indi iduals belonging o he di e en ac i i y classes we e 80, 15, and 5% o he low, mode a e, and high ac i i y g oups, espec i ely [13]. The mixing pa e ns be ween indi- iduals o he di e en age and isk g oups we e de e - mined by wo mixing pa ame e s. These pa ame e s o he asso a i eness o mixing by age and by isk g oup we e se o 0.4 and 0.3, espec i ely [14]. An asso a i e- ness by age o 0.4 means ha 40% o new sexual pa ne s a e p e e en ially chosen om he same age g oup while he emaining sexual pa ne s a e chosen p opo ionally om all age g oups depending on hei size. An asso a- i eness by isk g oup o 0.3 means ha 30% o pa ne s a e p e e en ially chosen wi hin he same sexual ac i i y g oup while he emaining sexual pa ne s a e chosen p opo ionally om all sexual ac i i y g oups depending on hei size. As obus Ge man da a on pa e ns o sex- ual beha iou we e lacking, we used da a om he ans- mission model by Zechmeis e e al. [15], which had been o iginally de i ed om B i ish and No wegian su eys o sexual beha iou . Na u al his o y The ini ial alues o he epidemiological model pa am- e e s ega ding he na u al his o y o ce ical cance we e ob ained om he published li e a u e, pa icula ly om p e ious modelling s udies [14, 16–20]. Whe eas some pa ame e alues we e di ec ly inco po a ed in ou model, o he alues we e de i ed h ough calib a ion. De ails on he pa ame e alues used in ou model a e p o ided in an a icle by Ho n e al. [12]. No e e y HPV in ec ion is ollowed by he de elop- men o a ype-speci ic esis ance. Hence, we only allowed a ac ion o he in ec ed popula ion o be immune a e in ec ion o a ce ain pe iod o ime [21]. The p opo ion o indi iduals expe iencing a se ocon e sion was es i- ma ed o be 60% [22]. The ac ion o se ocon e ed indi- iduals de eloping a na u ally acqui ed immuni y was es ima ed o be 30% [23]. Waning o na u ally acqui ed immuni y was assumed o be 10% pe yea esul ing in an a e age ype-speci ic immuni y o 10yea s. Well (suscepble) Well ( eco e ed) HPV-in ec ed CIN 1 CIN 2 CIN 3 FIGO IV FIGO III FIGO II FIGO I CIS FIGO III FIGO II FIGO I CIS Ce icalcance (unde ec ed) Ce icalcance (de ec ed) FIGO IV Benign hys e ec omy Dea h F om all non- cance s a es F om all s a es Fig. 1 Simpli ied model s uc u e (adap ed om Ho n e al. [12]) Page 4 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 In ou model, he CIS s a e was subdi ided in o wo compa men s as p oposed by Insinga e al. [24] o a oid sho ansi ion imes om CIS o in asi e cance esul - ing om he exponen ial dis ibu ion. Sc eening p og amme The sc eening module o ou model was cons uc ed o e lec he cu en sc eening p ac ice in Ge many. We applied age-speci ic sc eening co e age a es ha we e ob ained om Ge man da a [25] and assumed he p o- po ion o women who ne e pa icipa e in sc eening o be 7% as done in o he modelling s udies [18, 19]. To simula e he u u e e ec s o he s epwise inc ease in sc eening co e age in Ge many du ing he 1990s and o p e en an o e es ima ion o he accine-induced educ- ion in ce ical cance - ela ed bu den o disease, we inco po a ed his o ical changes in sc eening up ake in ou model. Fo his pu pose, we combined he cu en age-speci ic sc eening co e age a es wi h he obse ed changes in pa icipa ion in ea ly cance de ec ion p o- g ammes o e ime [26]. Diagnos ic ollow-up (including managemen o abno mal ce ical cy ology and his ol- ogy) was modelled by he use o in eg a ed decision ees. The cons uc ion o hese decision ees was based on clinical guidelines, empi ical s udies, heal h economic models, and an in e iew wi h membe s o he Ge man wo king g oup on ce ical pa hology and colposcopy ha was conduc ed explici ly o his pu pose. Vaccina ion In he base case scena io, we analysed he cos -e ec- i eness o accina ing 12-yea -old gi ls wi h h ee doses. Vaccina ion co e age was assumed o be 50% based on se e al s udies assessing HPV accine up ake in Ge many [27–32]. In hese s udies, epo ed up ake, de ined as eceip o a leas one HPV accine dose, anged om 30 o 60%. When conside ing only hose emales who ecei ed he ull cou se o h ee doses, up ake anged om 27 o 48%. We assumed comple ion o he 3-dose se ies o all gi ls who ini ia ed accina- ion. All analyses we e ca ied ou o bo h he bi a- len and quad i alen accines. Vaccine e icacy agains HPV 16/18 in ec ion was es ima ed o be 98% o bo h accines, and accine e icacy in p e en ing HPV 6/11 in ec ion was es ima ed o be 100% o he quad i alen accine. C oss-p o ec ion agains non- accine onco- genic HPV- ypes was no conside ed in he base case scena io bu was explo ed in sensi i i y analysis aking he di e en c oss-p o ec ion p o iles o he accines in o accoun . Cu en ly, he maximum du a ion o ac- cine p o ec ion is unknown. Clinical e icacy agains ype-speci ic in ec ion and associa ed diseases has been demons a ed up o 9.4yea s pos - accina ion [33], bu long- e m pe sis ence o an ibody esponses has been p edic ed by s a is ical modelling o indi idual an ibody da a [34, 35]. Taking bo h he e idence based on lim- i ed ollow-up o adolescen gi ls and women in clini- cal ials and he p edic ions o s a is ical models in o accoun , we decided o assume a 10yea s las ing ini ial pe iod o sus ained accine p o ec ion ollowed by a pe iod o waning immuni y using a waning a e o 10% pe yea . This app oach esul ed in an a e age du a ion o accine-induced immuni y o 20yea s. We sepa a ely examined he impac o li elong p o ec ion as well as he in luence o adminis e ing a boos e dose in sensi- i i y analysis. While in he base case scena io accina- ion was es ic ed o gi ls, immunisa ion o boys was assessed in sensi i i y analysis. A 2-dose schedule was also examined in sensi i i y analysis. All accina ion- ela ed inpu da a a e p esen ed in Table1. Resou ce u ilisa ion anddi ec heal h ca e cos s Ou model akes in o accoun esou ce use associa ed wi h accina ion, sc eening, managemen o abno mal cy ological sc eening esul s, managemen and ea - men o biopsy-con i med CIN, and ea men o ce ical cance and geni al wa s. Di ec heal h ca e cos s we e calcula ed conside ing all ele an cos componen s ha a e eimbu sed by he s a u o y heal h insu ance. These componen s include medica ion, physician consul a- ions, ou pa ien diagnos ic p ocedu es, labo a o y es - ing, he apeu ic appliances and ou pa ien heal h ca e se ices p o ided by non-physicians (i.e. medical com- p ession igh s and manual lympha ic d ainage as pa o lymphoedema ea men ), and hospi alisa ions. T ea men pa e ns and ela ed esou ce consump ion we e mainly de i ed om clinical guidelines and pub- lished s udies on he cu en managemen o ce ical cance and i s p ecu so s [36–39]. Li e a u e-based e i- dence was supplemen ed by expe opinion o accoun o missing da a o Ge many and a ia ion in clinical p ac ice. Fo ins ance, he expe s we e asked o es ima e he s age-speci ic equency o u ilisa ion o di e en ou pa ien and inpa ien ea men p ocedu es o ce ical cance . A o al o six expe in e iews wi h gynaecolo- gis s we e conduc ed by elephone o in w i en o m. Table2 gi es an o e iew o he assumed ea men pa - e ns and esou ce u ilisa ion in he ea men and pos - ea men ollow-up o ce ical cance . Mos uni cos s we e based on o icial Ge man p ice lis s, ee scales, o ca alogues. Vaccine p ices and d ug cos s we e ob ained om he pha maceu ical da abase LAUER-TAXE® [40]. The cos pe dose o bo h accines was es ima ed a €150.41. The mean accine adminis a- ion ee was calcula ed o be €7.50 pe dose, based on a e iew o he immunisa ion ee scales o all egional Page 5 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 Associa ions o S a u o y Insu ance Physicians in Ge - many. Cha ges o ou pa ien isi s as well as ou pa ien diagnos ic and ea men p ocedu es we e based on he physician ee scale (Einhei liche Bewe ungsmaßs ab, EBM) o he s a u o y heal h insu ance. Hospi alisa ion cos s we e e ie ed om he Ge man diagnosis- ela ed g oup (DRG) ca alogue using a base a e o €2935.78 and he cos weigh s o a ious DRGs (N01A, N01E, N03B, N09Z, N15Z, and N60A). Cos s o inpa ien pal- lia i e ca e and ea men we e calcula ed assuming a leng h o s ay o 30days and combining he DRG N60B wi h a supplemen a y ee o pallia i e ca e (ZE60.01). All di ec cos s we e adjus ed o pa ien co-paymen s. Cos es ima es o ea ing CIN 3 and CIS we e aken om a Ge man esou ce use s udy p o iding in e en ion cos s associa ed wi h a PAP IV diagnosis o he Munich Cy o- logical Classi ica ion, which co esponds o se e e dys- plasia and CIS [41]. Cos s o ea ing geni al wa s we e based on own calcula ions using da a om a Ge man cos -o -illness s udy [42]. All cos s a e epo ed in 2010 eu os. Whe e 2010 p ices we e no a ailable, p ices we e in la ed o 2010 alues using he Ge man consume p ice index (CPI). The base case alues o he agg ega ed di ec heal h ca e cos s a e summa ised in Table3. Indi ec cos s Indi ec cos s in e ms o p oduc ion losses we e con- side ed when adop ing a socie al pe spec i e. Da a on absence om wo k due o HPV-associa ed illness we e ob ained om he s a is ics o a Ge man sickness und [43] using yea 2008 in o ma ion. Indi ec cos s we e calcula ed by he ic ion cos  app oach assum- ing he ic ion pe iod equal o he a e age du a ion o a acan job posi ion. Acco ding o a epo o he Fede al Employmen Agency [44], his pe iod was assumed o be 63days. Cos pe wo k day los was es ima ed a €85.13 using 2010 da a on mone a y compensa ion and num- be o employees in Ge many om he Fede al S a is i- cal O ice [45]. Indi ec cos s due o CIN and ce ical cance we e weigh ed by age-speci ic employmen a es o women o a oid an o e es ima ion o he p oduc ion losses. All indi ec cos inpu s a e summa ised in Table4. Heal h s a e u ili ies In he absence o u ili y alues ha a e speci ic o Ge - many, he da a o calcula ing quali y-adjus ed li e yea s (QALYs) we e aken om he in e na ional li e a u e and p e ious heal h economic models. These s udies applied di e en me hods o elici ing QALY weigh s including he use o he EQ-5D ques ionnai e [46], he ime ade- o echnique [47], and an expe -based applica ion o he Heal h U ili y Index (HUI) Ma k II [48]. The selec ion o u ili y alues was guided by he model s uc u e. The u il- i y alues used in ou model a e p esen ed in Table5. We assumed he baseline u ili y alue o no mal heal h o be 1.0. Es ima es o he du a ion o educ ions in quali y o li e we e mainly based on expe opinion. QALY losses associa ed wi h Pap smea -based sc eening and diagnos- ic ollow-up o cy ological and his ological esul s we e no conside ed in ou modelling app oach. Discoun ing In he base case analysis, u u e cos s and heal h e ec s we e discoun ed a an annual a e o 3% as ecommended by guidelines o he Ins i u e o Quali y and E iciency in Heal h Ca e [49] and he STIKO [10]. O he Ge man ecommenda ions on heal h economic e alua ion, also e e ed o as Hano e Consensus, a ou a discoun a e Table 1 Vaccina ion- ela ed inpu a iables HPV human papilloma i us Pa ame e Value Sou ce Vaccine e icacy HPV 16/18 in emales 98% [81, 82] HPV 6/11 in emales (quad i alen accine only) 100% [83] HPV 16/18 in males 90.4% [84] HPV 6/11 in males (quad i alen accine only) 90.4% [84] C oss-p o ec ion p o ided by he quad i alen accine (conside ed in sensi i i y analysis only) HPV 31/33/35/39/45/51/52/56/58/59 32.5% [85, 86] C oss-p o ec ion p o ided by he bi alen accine (conside ed in sensi i i y analysis only) HPV 31/33/35/39/45/51/52/56/58/59 68.4% [86] Du a ion o ull p o ec ion 10 yea s Assump ion Waning (a e he du a ion o ull p o ec ion) 0.1 pe yea Assump ion Vaccina ion co e age 50% Assump ion Age a accina ion 12 yea s Assump ion Boos e accina ion No boos e accina ion in he base case analysis Assump ion Page 6 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 o 5% [50], which was conside ed in sensi i i y analysis. In addi ion, we assessed he impac o di e en ial dis- coun ing (3% o cos s and 1.5% o heal h e ec s) [51]. Analy ic s a egy andsensi i i y analysis To de e mine he cos -e ec i eness o he in oduc ion o HPV accina ion in Ge many, we calcula ed inc emen al Table 2 T ea men pa e ns and esou ce u ilisa ion in he ea men andpos - ea men ollow-up o ce ical cance (pe cen ages a e a e age alues based onexpe s’ esponses) FIGO In e na ional Fede a ion o Gynecology and Obs e ics Ce ical cance s age (FIGO classi ica ion) o ea men phase T ea men pa e ns and esou ce u ilisa ion FIGO IA1 Conisa ion 60% Conisa ion wi h pel ic lymph node dissec ion 10% Simple hys e ec omy 20% Simple hys e ec omy wi h pel ic lymph node dissec ion 10% FIGO IA2 Conisa ion wi h pel ic lymph node dissec ion 20% Radical achelec omy wi h pel ic lymph node dissec ion 10% Simple hys e ec omy 10% Simple hys e ec omy wi h pel ic lymph node dissec ion 60% FIGO IB1 Radical hys e ec omy wi h pel ic lymph node dissec ion 64% Radical hys e ec omy wi h pel ic lymph node dissec ion and adju an chemo adio he apy 16% Radical achelec omy wi h pel ic lymph node dissec ion 5% Chemo adio he apy 15% FIGO IB2 Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion 49% Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion and adju an chemo adio- he apy 21% Chemo adio he apy 30% FIGO IIA Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion 35% Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion and adju an chemo adio- he apy 15% Chemo adio he apy 50% FIGO IIB Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion 12% Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion and adju an chemo adio- he apy 18% Chemo adio he apy 70% FIGO IIIA Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion 2% Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion and adju an chemo adio- he apy 8% Chemo adio he apy 90% FIGO IIIB Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion 2% Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion and adju an chemo adio- he apy 8% Chemo adio he apy 90% FIGO IVA Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion 2% Radical hys e ec omy wi h pel ic and pa aao ic lymph node dissec ion and adju an chemo adio- he apy 8% Chemo adio he apy 80% Exen e a ion 2% Exen e a ion and adju an chemo adio he apy 8% FIGO IVB Chemo adio he apy 40% Pallia i e chemo he apy 60% Pos - ea men ollow-up Ou pa ien isi s, Pap-smea s, and pel ic and abdominal ul asonog aphy 100% Ho mone eplacemen he apy (women <50 yea s) 50% Manual lympha ic d ainage and medical comp ession igh s 10–30% Page 7 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 Table 3 Di ec heal h ca e cos s Pa ame e Di ec cos s (€, 2010 p ice le el) Vaccina ion cos s Vaccine (ini ial se ies o 3 doses) 451.23 Adminis a ion (ini ial se ies o 3 doses) 22.50 Boos e sho (pe dose) 150.41 Adminis a ion o boos e sho (pe dose) 7.50 Cos s o sc eening, managemen o abno mal cy ological sc eening esul s, and obse a ional ollow-up o CIN 1 and CIN 2 Cy ological sc eening (Pap smea ) 25.23 Follow-up smea (including qua e ly Gynaecologis ’s ee and op ional colposcopy) ≤59 yea s 20.15 60+ yea s 20.33 HPV es (including qua e ly Gynaecologis ’s ee) ≤59 yea s 44.77 60+ yea s 44.94 HPV es and ollow-up smea (including qua e ly Gynaecologis ’s ee) ≤59 yea s 50.55 60+ yea s 50.73 Colposcopy (including qua e ly Gynaecologis ’s ee) ≤59 yea s 14.37 60+ yea s 14.54 Biopsy and his ology 129.47 Cos s o CIN/CIS ea men and pos - ea men ollow-up Conisa ion o he ce ix (CIN 1 and CIN 2) ≤39 yea s 525.05 40–59 yea s 531.22 60+ yea s 534.17 T ea men o CIN 3 and CIS 1621.53 Pos - ea men ollow-up o CIN/CIS (yea 1 and 2 a e ea men ) ≤59 yea s 70.71 60+ yea s 71.06 Cos s o ce ical cance ea men and pos - ea men ollow-up Diagnos ics o symp om-de ec ed ce ical cance ≤59 yea s 323.03 60+ yea s 324.08 Diagnos ics o sc een-de ec ed ce ical cance ≤59 yea s 179.18 60+ yea s 180.06 T ea men o ce ical cance (FIGO I) ≤59 yea s 7586.98 60+ yea s 7591.22 T ea men o ce ical cance (FIGO II) ≤59 yea s 11,455.22 60+ yea s 11,456.94 T ea men o ce ical cance (FIGO III) ≤59 yea s 12,380.21 60+ yea s 12,380.70 T ea men o ce ical cance (FIGO IV) ≤59 yea s 10,615.72 60+ yea s 10,616.21 Pos - ea men ollow-up o ce ical cance (yea 1 a e ea men ) ≤49 yea s 841.53 50–59 yea s 835.65 60+ yea s 836.35 Pos - ea men ollow-up o ce ical cance (yea 2 a e ea men ) ≤49 yea s 428.13 50–59 yea s 422.25 60+ yea s 422.95 Pos - ea men ollow-up o ce ical cance (yea 3 a e ea men ) ≤49 yea s 352.42 50–59 yea s 346.54 60+ yea s 347.24 Pos - ea men ollow-up o ce ical cance (yea 4 and 5 a e ea men ) ≤49 yea s 282.50 50–59 yea s 276.62 60+ yea s 276.97 Page 8 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 cos -e ec i eness a ios (ICERs) using li e yea s (LYs) gained and QALYs gained as ou come measu es. The base case analysis was ca ied ou om he Ge man heal h ca e paye pe spec i e, which is he pe spec i e o he s a u- o y heal h insu ance unds ( aking accoun o eimbu sed di ec cos s only), and om he socie al pe spec i e (con- side ing eimbu sed di ec cos s as well as indi ec cos s). Pa ien co-paymen s we e no included in bo h pe spec- i es since no all ele an sou ces p o ided su icien de ails on ha aspec . We e alua ed he long- e m heal h and economic e ec s o accina ing 12-yea -old gi ls agains HPV alongside he cu en cy ology-based ce ical cance sc eening p o- g amme compa ed o an exclusi e con inua ion o he cy ological sc eening p og amme. De e minis ic sensi i - i y analyses we e pe o med o es he obus ness o he esul s o changes in model inpu da a and assump ions. The impac o a ying single model pa ame e s was exam- ined by one-way sensi i i y analyses. Pa ame e s a ied we e cha ac e is ics o he accina ion p og amme, cos s, u ili ies, and he discoun a e. Fu he mo e, we assessed he implica ion o inco po a ing accine-speci ic c oss-p o- ec ion and e alua ed he addi ional impac o accina ing boys. Since a 2-dose schedule showed equi alen an ibody esponse and simila e icacy o he s anda d 3-dose egi- men [52–54], we also analysed he cos -e ec i eness o adminis e ing only wo doses a he age o 12yea s assum- ing he same le el o p o ec ion. Mul i a ia e sensi i i y analyses we e ca ied ou in e ms o bes -case (6/11/16/18 e icacy: 100%; accine-speci ic c oss-p o ec ion: 32.5 o 68.4%; li elong p o ec ion; 20% inc ease in sc eening and ea men cos ; 20% inc ease in quali yo li e de imen s; socie al pe spec i e) and wo s -case (6/11/16/18 e icacy: 80%; no c oss-p o ec ion; a e age du a ion o p o ec ion: 15yea s; 20% dec ease in sc eening and ea men cos ; 20% dec ease in quali yo li e de imen s; heal h ca e paye pe spec i e) analyses. Model calib a ion and alida ion Model calib a ion is he p ocess o adjus ing inpu pa ame e alues un il he simula ion ou pu ma ches empi ical da a. Ou model was calib a ed o e lec obse a ions on age-speci ic p e alence o HPVin ec- ion [55–59], age-speci ic p e alence o CIN [60], age- speci ic incidence and mo ali y o ce ical cance [61], as well as HPV ype-dis ibu ion in di e en ce ical disease s a es [57, 62–65]. When da a o Ge many we e no a ailable, we used da a om o he coun ies. The modi ica ion o pa ame e alues and he subsequen compa ison o he simula ion esul s wi h he obse ed da a we e pe o med manually. Fu he mo e, we used age-speci ic adjus men ac o s o ou g oups o pa ame e s (p og ession p obabili ies o cance , eg es- sion p obabili ies, de ec ion o cance by symp oms, and mo ali y o ce ical cance ) o achie e a be e i o he obse ed da a. Mo e de ails ega ding bo h he model calib a ion p ocess and he esul s o he model alida- ion ha e been p e iously published [12]. Sys ema ic li e a u e e iew We pe o med a sys ema ic li e a u e e iew o compa e ou esul s wi h indings o p e iously published s udies on he cos -e ec i eness o HPV accina ion in Ge many. A PubMed-based li e a u e sea ch was conduc ed using he ollowing sea ch e ms: (“HPV” OR “human papillo- ma i us”) AND (“ accine” OR “ accina ion” OR “immu- nisa ion” OR “immuniza ion”) AND (“cos -e ec i eness” OR “economic”) AND “Ge many”. This sea ch was com- plemen ed by scanning e e ence lis s o p e iously iden- i ied ull- ex a icles. A s udy was included i i me he ollowing c i e ia: (i) i was an economic e alua ion o HPV accina ion in Ge many, (ii) i was conduc ed om a heal h ca e paye o a socie al pe spec i e, (iii) i was w i en in English o Ge man, and (i ) i was published as a ull- ex a icle. Table 3 con inued Pa ame e Di ec cos s (€, 2010 p ice le el) Pos - ea men ollow-up o ce ical cance ( om yea 6 a e ea men onwa ds) ≤49 yea s 262.35 50–59 yea s 256.47 60+ yea s 256.65 Inpa ien pallia i e ca e and ea men 7518.09 Cos s o geni al wa s ea men T ea men o geni al wa s in emales 572.14 T ea men o geni al wa s in males 396.69 CIN ce ical in aepi helial neoplasia, CIS ca cinoma insi u, FIGO In e na ional Fede a ion o Gynecology and Obs e ics Page 9 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 Table 4 Indi ec cos s CIN ce ical in aepi helial neoplasia, CIS ca cinoma insi u, FIGO In e na ional Fede a ion o Gynecology and Obs e ics a A e age du a ion o absence om wo k in pa ien s who missed wo k because o illness b Weigh ed by age-speci ic employmen a es o women c No weigh ed by age-speci ic employmen a es as he ac ion o geni al wa s pa ien s who missed wo k was es ima ed di ec ly on he basis o a Ge man s udy [42] Pa ame e A e age absence omwo k (days)aIndi ec cos s (€, 2010 p ice le el)b T ea men o CIN 1 and CIN 2 15.9 15–19 yea s 336.50 20–24 yea s 835.26 25–29 yea s 965.33 30–34 yea s 977.31 35–39 yea s 1009.89 40–44 yea s 1063.18 45–49 yea s 1060.82 50–54 yea s 1010.30 55–59 yea s 853.97 60–64 yea s 410.89 T ea men o CIN 3 and CIS 21.3 15–19 yea s 450.79 20–24 yea s 1118.93 25–29 yea s 1293.18 30–34 yea s 1309.23 35–39 yea s 1352.87 40–44 yea s 1424.25 45–49 yea s 1421.10 50–54 yea s 1353.42 55–59 yea s 1144.00 60–64 yea s 550.44 T ea men o ce ical cance (all FIGO s ages) 44.4 15–19 yea s 939.67 20–24 yea s 2332.42 25–29 yea s 2695.65 30–34 yea s 2729.11 35–39 yea s 2820.06 40–44 yea s 2968.87 45–49 yea s 2962.30 50–54 yea s 2821.22 55–59 yea s 2384.68 60–64 yea s 1147.39 Dea h due o ce ical cance (all FIGO s ages) 63 ( ic ion pe iod) 15–19 yea s 1333.31 20–24 yea s 3309.52 25–29 yea s 3824.91 30–34 yea s 3872.38 35–39 yea s 4001.44 40–44 yea s 4212.58 45–49 yea s 4203.26 50–54 yea s 4003.08 55–59 yea s 3383.67 60–64 yea s 1628.06 T ea men o geni al wa s in emales 7.7 15–64 yea s 30.81c T ea men o geni al wa s in males 8.7 15–64 yea s 28.14c Page 16 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 Compa ison wi ho he models In o he models e lec ing he Ge man heal h ca e se - ing, he cos pe QALY o quad i alen HPV accina ion wi h h ee doses anged be ween €10,000 and €20,000 om a heal h ca e paye pe spec i e when assuming a 20yea s las ing du a ion o p o ec ion and conside ing e ec s on CIN, ce ical cance , and geni al wa s [67, 69]. Ou es ima e o €14,711 pe QALY alls wi hin his ange o es ima es. When adop ing a socie al pe spec i e, quad i alen HPV accina ion o 12-yea -old gi ls wi h wo doses esul ed in cos sa ings in ou model. This also compa es well wi h p e ious indings [66]. In con as , he s udy by Soe gel e al. [68] allows no di ec compa i- son wi h ou es ima es because i ocussed on conisa ion- associa ed neona al mo bidi y and mo ali y. Howe e , Soe gel e al. [68] showed ha accina ing gi ls agains HPV 16/18 migh e en be cos -e ec i e wi hou consid- e ing he impac on ce ical cance . S eng hs andlimi a ions An impo an s eng h o ou s udy is ha we p esen ed esul s o a wide ange o scena ios a ying he a ge g oup ( emale accina ion s. no accina ion; emale and male accina ion s. emale accina ion), co e age a e, and numbe o accine doses. Ano he s eng h is ha we used a dynamic ansmission model o es ima e he cos - e ec i eness o HPV accina ion in Ge many. Compa ed o s a ic models, which accoun only o di ec e ec s o accina ion, dynamic models also cap u e indi ec e ec s in e ms o he d p o ec ion [75]. Fu he mo e, we included he impac o his o ical changes in pa icipa ion a es o he cy ological ce ical cance sc eening in Ge many. This is one eason why ou model migh p o ide mo e accu a e esul s han he dynamic model om Schobe e al. [67]. In addi ion, we applied a mo e comp ehensi e calib a ion app oach han he a o emen ioned model and used age- speci ic epidemiological da a o se e al calib a ion a ge s. We acknowledge ha ou model also has se e al limi a- ions. Fi s , we did no include o he cance s han ce ical cance . O he modelling s udies showed ha conside - ing he impac on aginal, ul a , penile, anal, and head and neck cance could imp o e he cos -e ec i eness o HPV accina ion [76, 77]. Howe e , ou model e lec s he goal o he cu en Ge man HPV immunisa ion ec- ommenda ion, which is de ined (by STIKO) as he educ- ion in disease bu den caused by ce ical cance . Second, ou model e alua ed he cos -e ec i eness o he bi alen and quad i alen accines, bu did no include he new 9- alen HPV accine, which was in oduced in Ge many a e comple ion o his s udy. Howe e , we in es iga ed he impac o including accine-speci ic c oss-p o ec ion agains non- accine high- isk HPV ypes in sensi i - i y analyses, and co esponding esul s migh gi e an imp ession o he cos -e ec i eness o highe - alen ac- cines. We ecommend including he 9- alen HPV ac- cine in u u e modelling s udies. Thi d, ou model did no accoun o po en ial c oss-p o ec ion o he bi alen ac- cine agains low- isk HPV ypes. Resul s o an ecological s udy sugges ha he e migh be a mode a e c oss-p o- ec i e e ec o he bi alen accine agains geni al wa s since he a es o geni al wa s ha e declined a e he in oduc ion o a na ional HPV accina ion p og amme using he bi alen accine in England [78]. In addi ion, a pos hoc analysis o a clinical ial showed ha he bi alen accine p o ides a mode a e e icacy agains pe sis en in ec ion wi h low- isk HPV ypes [79]. Ne e heless, he unde lying biological mechanisms o hese indings ha e no ye been cla i ied conclusi ely. Fou h, in ou base case analysis, we assumed a accina ion age o 12yea s in o de o ensu e compa abili y wi h o he s udies al hough mos gi ls in Ge many ha e ecei ed he i s dose in he age o 13 o 14yea s [28]. Howe e , he upda ed immuni- sa ion ecommenda ion a ou s a younge accina ion age ( om 9 yea s), which migh impac u u e heal h and eco- nomic e ec s o HPV accina ion. Fi h, modelling o ce - ical cance sc eening was based on cy ological sc eening, which p edominan ly e lec s he cu en sc eening p ac- ice in Ge many. Howe e , u u e changes o ce ical cance sc eening may in ol e p ima y HPV es ing [80]. Six h, inpu da a on sexual beha iou we e based on su - eys om o he Eu opean coun ies, and ac ual beha iou in Ge many migh di e om ha in o he coun ies. Se en h, due o he lack o Ge man-speci ic u ili y al- ues, he u ili y es ima es we used in ou model we e aken om in e na ional s udies wi h some o hem es - ing on expe opinion. Fu he mo e, we assumed a base- line u ili y alue o 1.0 in he absence o HPV-associa ed diseases, which migh lead o a po en ial o e es ima ion o HPV- ela ed QALY losses. Howe e , sensi i i y analy- ses showed ha a ia ions in u ili ies had limi ed impac on he esul s. Eigh h, since we used a s able popula ion app oach, ou model did no ake accoun o demog aphic ends and hei implica ions. Conclusions Conside ing he o en-ci ed h eshold o €50,000 pe QALY, ou model esul s sugges ha ou ine HPV acci- na ion o 12-yea -old gi ls wi h h ee doses is likely o be cos -e ec i e in Ge many. Due o he addi ional impac on HPV 6/11- ela ed diseases (mos ly geni al wa s), he quad i alen accine appea ed o be mo e cos -e ec i e han he bi alen accine, e en when conside ing he highe c oss-p o ec ion o he bi alen accine. Mos o hese indings a e consis en wi h esul s p edic ed by p e iously published indus y- unded models o HPV accina ion in Ge many. Ou model also showed ha a Page 17 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 2-dose schedule o he quad i alen accine could esul in cos sa ings when assuming an equi alen le el o p o- ec ion and adop ing a socie al pe spec i e. Addi ional accina ion o boys was ound o be a cos -e ec i e s a - egy in scena ios wi h low co e age in gi ls. Howe e , he e is a need o an ex ended e sion o his model ha also accoun s o he po en ial impac on non-ce ical cance ypes in bo h gende s since he conside a ion o his aspec migh lead o mo e a ou able esul s ega d- ing he addi ional accina ion o boys in scena ios wi h mode a e o high co e age in gi ls. Abb e ia ions BCR: bene i –cos a io; CBA: cos –bene i analysis; CEA: cos -e ec i eness analysis; CIN: ce ical in aepi helial neoplasia; CIS: ca cinoma in si u; CPI: consume p ice index; CUA: cos -u ili y analysis; DRG: diagnosis- ela ed g oup; EBM: Einhei liche Bewe ungsmaßs ab; FIGO: In e na ional Fede a ion o Gynecology and Obs e ics; HPV: human papilloma i us; HUI: Heal h U ili y Index; ICER: inc emen al cos -e ec i eness a io; QALY: quali y-adjus ed li e yea ; MSM: men who ha e sex wi h men; RKI: Robe Koch Ins i u e; SIRS: suscep ible-in ec ious- eco e ed-suscep ible; STIKO: S ändige Imp kommis- sion (S anding Vaccina ion Commi ee). Au ho s’ con ibu ions OD and JH designed he s udy, de eloped he model, pe o med he da a col- lec ion, ca ied ou he analyses, and in e p e ed he esul s. OD conduc ed he in e iews wi h clinical expe s, pe o med he cos calcula ions, and d a ed he manusc ip . JH p og ammed and calib a ed he model. AMK and YD helped o selec he model inpu alues and p o ided clinical expe ad ice. MEEK con ib- u ed o he s udy design and p o ided me hodological expe ad ice. RTM and WG con ibu ed o he s udy design, pa icipa ed in he in e p e a ion o he esul s, and supe ised he p ojec . All au ho s c i ically e iewed he ini ial d a . All au ho s ead and app o ed he inal manusc ip . Au ho de ails 1 Depa men o Heal h Economics and Heal h Ca e Managemen , School o Public Heal h, Biele eld Uni e si y, Uni e si ä ss aße 25, 33615 Biele eld, Ge many. 2 Epidemiological and S a is ical Me hods Resea ch G oup, Helm- hol z Cen e o In ec ion Resea ch, B aunschweig, Ge many. 3 Hanno e Medi- cal School, Hanno e , Ge many. 4 Ge man Cen e o In ec ion Resea ch, Si e Hanno e -B aunschweig, Hanno e /B aunschweig, Ge many. 5 Julius Cen e o Heal h Sciences and P ima y Ca e, Uni e si y Medical Cen e U ech , U e- ch , The Ne he lands. 6 Cen e o In ec ious Disease Con ol, RIVM, Bil ho en, The Ne he lands. 7 Gynecologic Tumo Immunology, Clinic o Gynecology, Cha i é-Uni e si ä smedizin Be lin, Be lin, Ge many. 8 P axis Löse /Kaden/ Dele é/Knappe, Be lin, Ge many. 9 Immunisa ion Uni , Robe Koch Ins i u e, Be lin, Ge many. 10 Depa men o Public Heal h Medicine, School o Public Heal h, Biele eld Uni e si y, Biele eld, Ge many. Acknowledgemen s We hank Ma hias W. Beckmann (Uni e si y Hospi al E langen, E langen, Ge many), Ch is ian Dannecke (Hospi al o he Ludwig-Maximilians-Uni e si y München, München, Ge many), Ingke Hagemann (ab s + pa ne , Kiel, Ge - many), Pe e Hillemanns (Hanno e Medical School, Hanno e , Ge many), Ka l U. Pe y (Wol sbu g Clinic, Wol sbu g, Ge many), Achim Schneide (Cha i é- Uni e si ä smedizin Be lin, Be lin, Ge many) and he membe s o he Ge man wo king g oup on ce ical pa hology and colposcopy o pa icipa ing in he expe in e iews. We also hank Julian Wi e (Biele eld Uni e si y, Biele eld, Ge many) o his assis ance in collec ing uni cos da a. We acknowledge sup- po o he A icle P ocessing Cha ge by he Deu sche Fo schungsgemein- scha and he Open Access Publica ion Fund o Biele eld Uni e si y. Compe ing in e es s OD and WG ha e conduc ed s udies o Sano i Pas eu MSD and GlaxoSmi h- Kline un ela ed o he cu en s udy. All o he au ho s ha e no compe ing in e es s o decla e. A ailabili y o da a and ma e ials No applicable. Consen o publica ion No applicable. E hics app o al and consen o pa icipa e No applicable. Funding The s udy was ully unded by Robe Koch Ins i u e (P ojec Numbe : 1362/1-927). Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in pub- lished maps and ins i u ional a ilia ions. Recei ed: 8 Sep embe 2016 Accep ed: 23 Augus 2017 Re e ences 1. Bosch FX, Lo incz A, Muñoz N, Meije CJLM, Shah KV. The causal ela- ion be ween human papilloma i us and ce ical cance . J Clin Pa hol. 2002;55:244–65. 2. Gillison ML, Cha u edi AK, Lowy DR. HPV p ophylac ic accines and he po en ial p e en ion o nonce ical cance s in bo h men and women. Cance . 2008;113:3036–46. 3. G ulich AE, Jin F, Conway EL, S ein AN, Hocking J. Cance s a ibu able o human papilloma i us in ec ion. Sex Heal h. 2010;7:244–52. 4. Dunne EF, Ma kowi z LE. Geni al human papilloma i us in ec ion. Clin In ec Dis. 2006;43:624–9. 5. Dele é Y, Wichmann O, Klug SJ, an de Sande M, Te ha d M, Zepp F, Ha de T. The e icacy and du a ion o accine p o ec ion agains human papilloma i us: a sys ema ic e iew and me a-analysis. D sch A z ebl In . 2014;111:584–91. 6. Kash N, Lee MA, Kollipa a R, Downing C, Guid y J, Ty ing SK. Sa e y and e icacy da a on accines and immuniza ion o human papilloma i us. J Clin Med. 2015;4:614–33. 7. Lu B, Kuma A, Cas ellsagué X, Giuliano AR. E icacy and sa e y o p o- phylac ic accines agains ce ical HPV in ec ion and diseases among women: a sys ema ic e iew & me a-analysis. BMC In ec Dis. 2011;11:13. 8. Muñoz N, Bosch FX, Cas ellsagué X, Díaz M, de Sanjose S, Hammouda D, Shah KV, Meije CJ. Agains which human papilloma i us ypes shall we accina e and sc een? The in e na ional pe spec i e. In J Cance . 2004;111:278–85. 9. G ee CE, Wheele CM, Ladne MB, Beu ne K, Coyne MY, Liang H, Langen- be g A, Yen TS, Rals on R. Human papilloma i us (HPV) ype dis ibu ion and se ological esponse o HPV ype 6 i us-like pa icles in pa ien s wi h geni al wa s. J Clin Mic obiol. 1995;33:2058–63. 10. S ändige Imp kommission. Emp ehlungen de S ändigen Imp kommis- sion (STIKO) am Robe Koch-Ins i u /S and: Augus 2014. Epidemiol Bull. 2014;34(2014):305–40. 11. S ändige Imp kommission. Wissenscha liche Beg ündung ü die Ände ung de Emp ehlung zu Imp ung gegen humane Papillom i en. Epidemiol Bull. 2014;35(2014):343–7. 12. Ho n J, Damm O, K e zschma MEE, Dele é Y, Wichmann O, Kau mann AM, Ga be E, K äme A, G eine W, Mikolajczyk RT. Es ima ing he long- e m e ec s o HPV accina ion in Ge many. Vaccine. 2013;31:2372–80. 13. Choi YH, Ji M, Gay N, Cox A, Ga ne GP, Edmunds WJ. T ansmission dynamic modelling o he impac o human papilloma i us accina ion in he Uni ed Kingdom. Vaccine. 2010;28:4091–102. 14. Elbasha EH, Dasbach EJ, Insinga RP. Model o assessing human papil- loma i us accina ion s a egies. Eme g In ec Dis. 2007;13:28–41. 15. Zechmeis e I, de Blasio BF, Ga ne G, Neilson AR, Siebe U. Cos - e ec i eness analysis o human papilloma i us- accina ion p og ams o p e en ce ical cance in Aus ia. Vaccine. 2009;27:5133–41. Page 18 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 16. Can ell K, Ba nabas R, Pa nick J, Be al V. The p edic ed e ec o changes in ce ical sc eening p ac ice in he UK: esul s om a modelling s udy. B J Cance . 2004;91:530–6. 17. Elbasha EH, Dasbach EJ, Insinga RP. A mul i- ype HPV ansmission model. Bull Ma h Biol. 2008;70:2126–76. 18. Kohli M, Fe ko N, Ma in A, F anco EL, Jenkins D, Galli an S, She law-John- son C, D ummond M. Es ima ing he long- e m impac o a p ophylac ic human papilloma i us 16/18 accine on he bu den o ce ical cance in he UK. B J Cance . 2007;96:143–50. 19. Schneide A, Hamme schmid T, Schwa z TF, Rogoza RM, Fe ko N, Siebe U. Long- e m public heal h impac o accina ion agains ce ical cance in Ge many. Gebu shil e F auenheilkd. 2007;67:850–8. 20. S oczynski G, Schnell-Inde s P, Mühlbe ge N, Lang K, Aidelsbu ge P, Wasem J, Mi endo T, Engel J, Hillemanns P, Pe y K, K äme A, Siebe U. En scheidungsanaly ische Modellie ung zu E alua ion de Langzei - E ek i i ä und Kos en-E ek i i ä des Einsa zes de HPV-DNA-Diagnos ik im Rahmen de Ze ixka zinom ühe kennung in Deu schland. Köln: Deu sches Ins i u ü Medizinische Dokumen a ion und In o ma ion; 2010. 21. Mikolajczyk RT, Ho n J, Damm O, Kau mann AM, K e zschma MEE. Epide- miological s udy o an i-HPV-16/18 se oposi i i y and subsequen isk o HPV-16 and -18 in ec ions. J Na l Cance Ins . 2012;104:163. 22. Ca e JJ, Kou sky LA, Hughes JP, Lee SK, Kuype s J, Ki ia N, Galloway DA. Compa ison o human papilloma i us ypes 16, 18, and 6 capsid an ibody esponses ollowing inciden in ec ion. J In ec Dis. 2000;181:1911–9. 23. Sa aeian M, Po as C, Schi man M, Rod iguez AC, Wacholde S, Gonzalez P, Quin W, an Doo n L, She man ME, Xhense al V, He e o R, Hildesheim A. Epidemiological s udy o an i-HPV16/18 se oposi i i y and subsequen isk o HPV16 and -18 in ec ions. J Na l Cance Ins . 2010;102:1653–62. 24. Insinga RP, Dasbach EJ, Elbasha EH. Epidemiologic na u al his o y and clinical managemen o human papilloma i us (HPV) disease: a c i ical and sys ema ic e iew o he li e a u e in he de elopmen o an HPV dynamic ansmission model. BMC In ec Dis. 2009;9:119. 25. Ke ek-Bodden H, Al enho en L, B enne G. Du ch üh ung eine e si- che enbezogenen Un e suchung zu Inansp uchnahme de F ühe ken- nung au Ze ixka zinom in den Jah en 2002, 2003 und 2004 au de Basis on Ab echnungsda en. Be lin: Zen alins i u ü die kassenä z li- che Ve so gung in de Bundes epublik Deu schland; 2009. 26. Robe Koch-Ins i u . Gesundhei in Deu schland. Be lin: Robe Koch- Ins i u ; 2006. 27. Dele é Y, Böhme MM, Wal e D, Wichmann O. HPV accina ion co e age among women aged 18–20 yea s in Ge many h ee yea s a e ecom- menda ion o HPV accina ion o adolescen gi ls: esul s om a c oss- sec ional su ey. Hum Vaccines Immuno he . 2013;9:1706–11. 28. Poe hko-Mülle C, Bu mann-Schwelge N, KiGGS S udy G oup. Imp s a- us und De e minan en de Imp ung gegen humane Papillom i en (HPV) bei Mädchen in Deu schland. Bundesgesundhei sbla Gesundhei s- o schung Gesundhei sschu z. 2014;57:869–77. 29. Remschmid C, Fesen eld M, Kau mann AM, Dele é Y. Sexual beha io and ac o s associa ed wi h young age a i s in e cou se and HPV accine up ake among young women in Ge many: implica ions o HPV accina- ion policies. BMC Public Heal h. 2014;14:1248. 30. Rieck T, Feig M, Dele é Y, Wichmann O. U iliza ion o adminis a i e da a o assess he associa ion o an adolescen heal h check-up wi h human papilloma i us accine up ake in Ge many. Vaccine. 2014;32:5564–9. 31. Roggendo H. E s e E ah ungen zu Akzep anz de HPV-Imp ung. Mona ssch Kinde heilkd. 2009;157:982–5. 32. S öcke P, Dehne M, Schus e M, Wichmann O, Dele é Y. Human papillo- ma i us accine up ake, knowledge and a i ude among 10 h g ade s u- den s in Be lin, Ge many, 2010. Hum Vaccines Immuno he . 2013;9:74–82. 33. de Vincenzo R, Con e C, Ricci C, Scambia G, Capelli G. Long- e m e icacy and sa e y o human papilloma i us accina ion. In J Womens Heal h. 2014;6:999–1010. 34. Da id M, an He ck K, Ha d K, Tibaldi F, Dubin G, Descamps D, an Damme P. Long- e m pe sis ence o an i-HPV-16 and -18 an ibod- ies induced by accina ion wi h he AS04-adju an ed ce ical cance accine: modeling o sus ained an ibody esponses. Gynecol Oncol. 2009;115:S1–6. 35. F ase C, Tomassini JE, Xi L, Golm G, Wa son M, Giuliano AR, Ba E, Aul KA. Modeling he long- e m an ibody esponse o a human papilloma i- us (HPV) i us-like pa icle (VLP) ype 16 p ophylac ic accine. Vaccine. 2007;25:4324–33. 36. Beckmann MW, Mehlko n G, Thiel F, B euel C, Fasching PA, Acke mann S. The apie o sch i e beim p imä en Ze ixka zinom. D sch A z ebl. 2005;102:A979–86. 37. Dannecke C, Han schmann P, F iese K. S adienbezogene The apie des Ze ixka zinoms. Gynäkologe. 2008;41:349. 38. Deu sche K ebsgesellscha , Deu sche Gesellscha ü Gynäkologie und Gebu shil e, A bei sgemeinscha Gynäkologische Onkologie. Diagnos ik und The apie des Ze ixka zinoms, AWMF 032/033 (S2k). Be lin: Deu sche Gesellscha ü Gynäkologie und Gebu shil e; 2008. 39. Hillemanns P. Manual Ze ixka zinom. 3 d ed. München: Tumo zen um München und W. Zuckschwe d Ve lag GmbH; 2004. 40. LAUER-TAXE. LAUER-FISCHER GmbH. h p://www2.laue - ische .de. Accessed 17 Sep 2012. 41. Pe y KU, B eugelmans JG, Béna d S, Lamu e E, Li lewood KJ, Hillemanns P. Cos o sc eening and ea men o ce ical dyska yosis in Ge many. Eu J Gynaecol Oncol. 2008;29:345–9. 42. Hillemanns P, B eugelmans JG, Gieseking F, Béna d S, Lamu e E, Li lewood KJ, Pe y KU. Es ima ion o he incidence o geni al wa s and he cos o illness in Ge many: a c oss-sec ional s udy. BMC In ec Dis. 2008;8:76. 43. K ankhei sa ens a is ik (Ve siche e de Allgemeinen O sk ankenkas- sen). AOK-Bundes e band. 2008. h p://www.gbe-bund.de. Accessed 17 Sep 2012. 44. Bundesagen u ü A bei . A bei sma k 2009. Nü nbe g: Bundesagen u ü A bei ; 2010. 45. S a is isches Bundesam . Deu sche Wi scha 2010. Wiesbaden: S a is- isches Bundesam ; 2011. 46. B isson M, Van de Velde N, de Wals P, Boily M. The po en ial cos -e ec i e- ness o p ophylac ic human papilloma i us accines in Canada. Vaccine. 2007;25:5399–408. 47. Insinga RP, Glass AG, Mye s ER, Rush BB. Abno mal ou comes ollowing ce ical cance sc eening: e en du a ion and heal h u ili y loss. Med Decis Making. 2007;27:414–22. 48. Ins i u e o Medicine (U.S.). Commi ee o S udy P io i ies o Vaccine De elopmen . Vaccines o he 21s cen u y: a ool o decisionmaking. Washing on, D.C.: Na ional Academy P ess; 2000. 49. Ins i u ü Quali ä und Wi scha lichkei im Gesundhei swesen. Allge- meine Me hoden, Ve sion 4.2. Köln: Ins i u ü Quali ä und Wi scha li- chkei im Gesundhei swesen; 2015. 50. on de Schulenbu g JMG, G eine W, Jos F, Klusen N, Kubin M, Leidl R, Mi endo T, Rebsche H, Schoe ski O, Vau h C, Volme T, Wahle S, Wasem J, Webe C. Ge man ecommenda ions on heal h economic e alua ion: hi d and upda ed e sion o he Hano e consensus. Value Heal h. 2008;11:539–44. 51. Damm O, Ul sch B. Heal h economic e alua ion o accines. Gesundh ökon Qual Manag. 2015;20:163–72. 52. Dobson SRM, McNeil S, Dionne M, Dawa M, Ogil ie G, K ajden M, Sau ageau C, Schei ele DW, Kollmann TR, Halpe in SA, Langley JM, Be inge JA, Singe J, Money D, Mille D, Naus M, Ma a F, Young E. Immunogenici y o 2 doses o HPV accine in younge adoles- cen s s 3 doses in young women: a andomized clinical ial. JAMA. 2013;309:1793–802. 53. K eime AR, S uy F, Rosa io-Raymundo Del, Rowena Ma ia, Hildesheim A, Skinne SR, Wacholde S, Ga land SM, He e o R, Da id M, Wheele CM, González P, Jiménez S, Lowy DR, Pin o LA, Po as C, Rod iguez AC, Sa aeian M, Schi man M, Schille JT, Schussle J, She man ME, Bosch FX, Cas ellsague X, Cha e jee A, Chow S, Descamps D, Diaz-Mi oma F, Dubin G, Ge ma MJ, Ha pe DM, e al. E icacy o ewe han h ee doses o an HPV-16/18 AS04-adju an ed accine: combined analysis o da a om he Cos a Rica accine and PATRICIA ials. Lance Oncol. 2015;16:775–86. 54. Romanowski B, Schwa z TF, Fe guson LM, Fe guson M, Pe e s K, Dionne M, Schulze K, Ramja an B, Hillemanns P, Beh e U, Su yaki an P, Thomas F, S uy F. Immune esponse o he HPV-16/18 AS04-adju an ed accine adminis e ed as a 2-dose o 3-dose schedule up o 4 yea s a e ac- cina ion: esul s om a andomized s udy. Hum Vaccines Immuno he . 2014;10:1155–65. 55. Fo han SE, Go lieb SL, S e nbe g MR, Xu F, Da a SD, McQuillan GM, Be man SM, Ma kowi z LE. P e alence o sexually ansmi ed in ec ions among emale adolescen s aged 14 o 19 in he Uni ed S a es. Pedia ics. 2009;124:1505–12. 56. I ne T, Ebe le S, I ne A, Holz B, Banik N, Quin W, S aube A. P e alence o low- isk and high- isk ypes o human papilloma i us and o he isk ac o s o HPV in ec ion in Ge many wi hin di e en age g oups in Page 19 o 19 Damm e al. Cos E Resou Alloc (2017) 15:18 women up o 30 yea s o age: an epidemiological obse a ional s udy. J Med Vi ol. 2010;82:1928–39. 57. Klug SJ, Hukelmann M, Hollwi z B, Düzenli N, Schopp B, Pe y K, I ne T. P e alence o human papilloma i us ypes in women sc eened by cy ol- ogy in Ge many. J Med Vi ol. 2007;79:616–25. 58. de Sanjosé S, Diaz M, Cas ellsagué X, Cli o d G, B uni L, Muñoz N, Bosch FX. Wo ldwide p e alence and geno ype dis ibu ion o ce ical human papilloma i us DNA in women wi h no mal cy ology: a me a-analysis. Lance In ec Dis. 2007;7:453–9. 59. Schneide A, Hoye H, Lo z B, Leis i za S, Kühne-Heid R, Nindl I, Mülle B, Hae ing J, Dü s M. Sc eening o high-g ade ce ical in a-epi helial neoplasia and cance by es ing o high- isk HPV, ou ine cy ology o colposcopy. In J Cance . 2000;89:529–34. 60. Pe o J, Gilham C, Deacon J, Taylo C, E ans C, Binns W, Haywood M, Elanko N, Coleman D, Yule R, Desai M. Ce ical HPV in ec ion and neoplasia in a la ge popula ion-based p ospec i e s udy: he Manches e coho . B J Cance . 2004;91:942–53. 61. GEKID-A las. Gesellscha de epidemiologischen K ebs egis e in Deu schland. h p://www.gekid.de. Accessed 17 Sep 2012. 62. Cas ellsagué X, de Sanjose S, Aguado T, Louie KS, B uni L, Muñoz N, Diaz M, I win K, Gacic M, Beau ais O, Albe o G, Fe e E, By ne S, Bosch FX. HPV and ce ical cance in he 2007 epo . Vaccine. 2007;25(Suppl 3):C1–230. 63. Cli o d GM, Gallus S, He e o R, Muñoz N, Snijde s PJF, Vacca ella S, Anh PTH, Fe eccio C, Hieu NT, Ma os E, Molano M, Rajkuma R, Ronco G, de Sanjosé S, Shin HR, Suk i ach S, Thomas JO, Tunsakul S, Meije CJLM, F anceschi S. Wo ldwide dis ibu ion o human papilloma i us ypes in cy ologically no mal women in he In e na ional Agency o Resea ch on Cance HPV p e alence su eys: a pooled analysis. Lance . 2005;366:991–8. 64. Cli o d GM, Rana RK, F anceschi S, Smi h JS, Gough G, Pimen a JM. Human papilloma i us geno ype dis ibu ion in low-g ade ce ical lesions: compa ison by geog aphic egion and wi h ce ical cance . Cance Epidemiol Bioma k P e . 2005;14:1157–64. 65. Smi h JS, Lindsay L, Hoo s B, Keys J, F anceschi S, Wine R, Cli o d GM. Human papilloma i us ype dis ibu ion in in asi e ce ical cance and high-g ade ce ical lesions: a me a-analysis upda e. In J Cance . 2007;121:621–32. 66. Ko sopoulos N, Connolly MP, Remy V. Quan i ying he b oade economic consequences o quad i alen human papilloma i us (HPV) accina ion in Ge many applying a go e nmen pe spec i e amewo k. Heal h Econ Re . 2015;5:54. 67. Schobe D, Remy V, Schoe ski O. Cos -e ec i eness o accina ion wi h a quad i alen HPV accine in Ge many using a dynamic ansmission model. Heal h Econ Re . 2012;2:19. 68. Soe gel P, Makowski L, Schippe C, S aboulidou I, Hille U, Hillemanns P. The cos e iciency o HPV accines is signi ican ly unde es ima ed due o omission o conisa ion-associa ed p ema u i y wi h neona al mo ali y and mo bidi y. Hum Vaccines Immuno he . 2012;8:243–51. 69. Hillemanns P, Pe y KU, La ge on N, McAllis e R, Tolley K, Büsch K. Cos - e ec i eness o a e a alen human papilloma i us accine in Ge many. J Public Heal h. 2009;17:77–86. 70. Za nke KB, Le ine MA, O’B ien BJ. Cos -bene i analyses in he heal h- ca e li e a u e: don’ judge a s udy by i s label. J Clin Epidemiol. 1997;50:813–22. 71. Yahia MBBH, Jouin-Bo olo i A, De aux B. Ex ending he human papil- loma i us accina ion p og amme o include males in high-income coun ies: a sys ema ic e iew o he cos -e ec i eness s udies. Clin D ug In es ig. 2015;35:471–85. 72. Kim JJ. Ta ge ed human papilloma i us accina ion o men who ha e sex wi h men in he USA: a cos -e ec i eness modelling analysis. Lance In ec Dis. 2010;10:845–52. 73. Deshmukh AA, Chiao EY, Das P, Can o SB. Clinical e ec i eness and cos - e ec i eness o quad i alen human papilloma i us accina ion in HIV- nega i e men who ha e sex wi h men o p e en ecu en high-g ade anal in aepi helial neoplasia. Vaccine. 2014;32:6941–7. 74. Lin A, Ong KJ, Hobbelen P, King E, Meshe D, Edmunds WJ, Sonnenbe g P, Gilson R, Bains I, Choi YH, Tan on C, Soldan K, Ji M. Impac and cos - e ec i eness o selec i e human papilloma i us accina ion o men who ha e sex wi h men. Clin In ec Dis. 2017;64:580–8. 75. Ul sch B, Damm O, Beu els P, Bilcke J, B üggenjü gen B, Ge be -G o e A, G eine W, Hanque G, Hu ubessy R, Ji M, Knol M, on K ies R, Kuhlmann A, Le y-B uhl D, Pe le h M, Pos ma M, Salo H, Siebe U, Wasem J, Wich- mann O. Me hods o heal h economic e alua ion o accines and immu- niza ion decision amewo ks: a consensus amewo k om a Eu opean accine economics communi y. Pha macoeconomics. 2016;34:227–44. 76. de Kok IMCM, Habbema JDF, an Rosmalen J, an Ballegooijen M. Would he e ec o HPV accina ion on non-ce ical HPV-posi i e cance s make he di e ence o i s cos -e ec i eness? Eu J Cance . 2011;47:428–35. 77. Olsen J, Jø gensen TR. Re isi ing he cos -e ec i eness o uni e sal HPV- accina ion in Denma k accoun ing o all po en ially accine p e en able HPV- ela ed diseases in males and emales. Cos E Resou Alloc. 2015;13:4. 78. Can in M, Sinka K, Hughes G, Meshe D. Decline in geni al wa s diagno- ses among young women and young men since he in oduc ion o he bi alen HPV (16/18) accina ion p og amme in England: an ecological analysis. Sex T ansm In ec . 2016. doi:10.1136/sex ans-2016-052626. 79. Sza ewski A, Skinne SR, Ga land SM, Romanowski B, Schwa z TF, Ap e D, Chow SN, Paa onen J, Del Rosa io-Raymundo MR, Teixei a JC, De Ca alho NS, Cas o-Sanchez M, Cas ellsagué X, Poppe WAJ, De Su e P, Huh W, Cha e jee A, Tjalma WA, Acke man RT, Ma ens M, Papp KA, Bajo- A enas J, Ha pe DM, To né A, Da id MP, S uy F, Leh inen M, Dubin G. E icacy o he HPV-16/18 AS04-adju an ed accine agains low- isk HPV ypes (PATRICIA andomized ial): an unexpec ed obse a ion. J In ec Dis. 2013;208:1391–6. 80. Luy en A, Bu mann-Schweige N, Luy en K, Mau i z C, Reinecke-Lü hge A, Pie alla M, Meije CJLM, Pe y KU. Ea ly de ec ion o CIN3 and ce ical cance du ing long- e m ollow-up using HPV/Pap smea co- es ing and isk-adap ed ollow-up in a locally o ganised sc eening p og amme. In J Cance . 2014;135:1408–16. 81. Paa onen J, Naud P, Salme ón J, Wheele CM, Chow S, Ap e D, Ki chene H, Cas ellsague X, Teixei a JC, Skinne SR, Hed ick J, Jaisam a n U, Limson G, Ga land S, Sza ewski A, Romanowski B, Aoki FY, Schwa z TF, Poppe WAJ, Bosch FX, Jenkins D, Ha d K, Zaha T, Descamps D, S uy F, Leh inen M, Dubin G. E icacy o human papilloma i us (HPV)-16/18 AS04-adju an ed accine agains ce ical in ec ion and p ecance caused by oncogenic HPV ypes (PATRICIA): inal analysis o a double-blind, andomised s udy in young women. Lance . 2009;374:301–14. 82. The FUTURE II S udy G oup. Quad i alen accine agains human papilloma i us o p e en high-g ade ce ical lesions. N Engl J Med. 2007;356:1915–27. 83. Ga land SM, He nandez-A ila M, Wheele CM, Pe ez G, Ha pe DM, Leodol e ST, G ace WK, Fe is DG, S eben M, B yan J, Taddeo FJ, Railka R, Esse MT, Sings HL, Nelson M, Boslego J, Sa le C, Ba E, Kou sky LA. Quad i alen accine agains human papilloma i us o p e en anogeni- al diseases. N Engl J Med. 2007;356:1928–43. 84. Giuliano AR, Pale sky JM, Golds one S, Mo ei a ED, Penny ME, A anda C, Va das E, Moi H, Jessen H, Hillman R, Chang Y, Fe is D, Rouleau D, B yan J, Ma shall JB, Vuocolo S, Ba E, Radley D, Haup RM, Gu is D. E icacy o quad i alen HPV accine agains HPV In ec ion and disease in males. N Engl J Med. 2011;364:401–11. 85. B own DR, Kjae SK, Sigu dsson K, I e sen O, He nandez-A ila M, Wheele CM, Pe ez G, Kou sky LA, Tay EH, Ga cia P, Aul KA, Ga land SM, Leodol e S, Olsson S, Tang GWK, Fe is DG, Paa onen J, S eben M, Bosch FX, Dillne J, Jou a EA, Ku man RJ, Majewski S, Muñoz N, Mye s ER, Villa LL, Taddeo FJ, Robe s C, Tadesse A, B yan J, e al. The impac o quad i alen human papilloma i us (HPV; ypes 6, 11, 16, and 18) L1 i us-like pa icle accine on in ec ion and disease due o oncogenic non accine HPV ypes in gen- e ally HPV-nai e women aged 16–26 yea s. J In ec Dis. 2009;199:926–35. 86. Sza ewski A. HPV accine: ce a ix. Expe Opin Biol The . 2010;10:477–87. 87. Goldie SJ, Kohli M, G ima D, Weins ein MC, W igh TC, Bosch FX, F anco E. P ojec ed clinical bene i s and cos -e ec i eness o a human papilloma- i us 16/18 accine. J Na l Cance Ins . 2004;96:604–15. 88. Gold MR, F anks P, McCoy KI, F yback DG. Towa d consis ency in cos - u ili y analyses: using na ional measu es o c ea e condi ion-speci ic alues. Med Ca e. 1998;36:778–92. 89. de Kok IMCM, an Ballegooijen M, Habbema JDF. Cos -e ec i eness analysis o human papilloma i us accina ion in he Ne he lands. J Na l Cance Ins . 2009;101:1083–92.