Cancer Registry follow-up for 17 million person-years of a nationwide maternity cohort

3060 In oduc ion In he hie a chy o epidemiological e idence on causes o cance , longi udinal s udies (ei he nes ed case–con ol o case–coho s udies) a e second only o andomized ial e idence [1, 2]. The Finnish Ma e ni y Coho (FMC) was es ablished in 1983 o p o ide a esou ce o such s udies. All women in Finland a e a hei 12 h week o p egnancy o e ed sc eening o congeni al in ec ions (hepa- i is B, HIV, and syphilis). The esidual se um samples a e s o ed in he FMC [3]. The e is a sepa a e legal basis o he collec ion and scien i ic use o FMC (The law o he Na ional Ins i u e o Heal h and Wel a e 668/2008). Since 2001, a na ionwide in o med consen sys em based on he op - ou p inciple wi h negligible d op ou has been in ope a ion. An example o pionee ing cance esea ch based on he FMC a e longi udinal s udies documen ing ha exposu e o human papilloma i us (HPV) ypes 16 and 18 causes an excess isk o la e de elopmen o bo h ce ical, o he anogeni al and o opha yngeal cance s [4–6]. FMC’s se ial samples ha e been use ul o dis- en angling he empo al o de o e en s in ca cinogenesis, o example, an agonis ic in e ac ion o di e en HPV ypes [7], and disclosing smoking as an independen isk ac o o ce ical cance [8]. The p ediagnos ic se ial samples also enable s udies on he sensi i i y and speci- ici y o bioma ke s o u u e cance diagnosis and sc eening [9]. High- quali y da a on he ges a ional da e o he sample collec ion enable s udies on ho mones and cance [10]. The FMC is posi ioned as an in e - na ional Open Access esou ce. Linkage o pe sonal iden i ie s wi h he na ionwide, popula ion- based Finnish Cance Regis y and o he popula ion- based heal h and ial egis ies es ablishes a unique s udy base o lon- gi udinal s udies [11]. METHODS Cance Regis y ollow- up o 17 million pe son- yea s o a na ionwide ma e ni y coho Ma i Leh inen1,2,3 , Heljä-Ma ja Su cel3,4,5, Ka i Na unen1,3, Ee o Pukkala1,6 & Joakim Dillne 2,3 1Uni e si y o Tampe e, Tampe e, Finland 2Ka olinska Ins i u e, S ockholm, Sweden 3Eu opean Science In as uc u e Se ices EEIG, S ockholm, Sweden 4Na ional Ins i u e o Heal h & Wel a e, Helsinki, Finland 5Biobank Bo ealis o No he n Finland, Oulu Uni e si y Hospi al, Oulu, Finland 6Finnish Cance Regis y, Ins i u e o S a is ical and Epidemiological Cance Resea ch, Helsinki, Finland © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d. This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s use, dis ibu ion and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Keywo ds Biobank, cance causes, coho , epidemiology, umo ma ke s Co espondence Ma i Leh inen, Uni e si y o Tampe e, 33014 Tampe e, Finland. Tel: +358405437862; Fax +358 3 2100004; E-mail: [email p o ec ed] Funding In o ma ion NCI con ac : HHSN261201300016C Recei ed: 4 Sep embe 2017; Accep ed: 10 Sep embe 2017 Cance Medicine 2017; 6(12):3060–3064 doi: 10.1002/cam4.1222 Abs ac Popula ion- based Finnish Ma e ni y Coho (FMC) comp ises 2M i s imes e se a collec ed om 1M women du ing 33 yea s. In o med consen is by he op - ou p inciple, and linkages wi h cance and popula ion egis ies p o ide a base o o e ime and o e gene a ion s udies. Follow- up o 17M pe son- yea s by he end o 2014 can iden i y 39,700 cases o in asi e cance and 18,900 cases o p emalignan b eas and ce ix lesions, and basal cell ca cinoma diagnosed a e se um sampling. Fo women wi h mul iple p egnancies, se ial samples aken be o e cance diagnosis a e a ailable. O sp ing o he women ha e de eloped mo e han 4000 cance s. Fo 100,000 indi iduals, samples aken du - ing he p egnancies o bo h hei mo he s and g andmo he s enable amilial cance s udies. FMC con inues o collec samples, and su eillance o exposu es o in e en ions like accina ion p og ams is easible. In summa y, he FMC is a unique, accessible biobank o epidemiological, bioma ke , and su eillance s udies on cance . Cance Medicine Open Access 3061 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d. Cance Cases in Ma e ni y Coho M. Leh inen e al. Table 1. Es ima ed1 numbe s o inciden in asi e cance cases in he Finnish Ma e ni y Coho 1983–2014. P ima y si e ICD- 10 code A e i s sample (N = 953,000) A e second sample (N = 604,000) A e hi d sample (N = 240,000) All si es C00–96, D32–33, D42–43, D45–47, D76 39,700 19,900 6800 Mou h, pha ynx C00–14 490 220 85 Lip C00 10 5 1 Tongue C02 110 50 5 Mou h, o he C03–06 110 50 15 Sali a y glands C07–08 110 50 20 Pha ynx C01, C09–14 130 55 20 Diges i e o gans C15–26 3900 1900 700 Esophagus C15 65 30 10 S omach C16 550 280 95 Small in es ine C17 110 55 20 Colon C18 1400 700 250 Rec um, ec osigmoid, anus C19–20 790 380 130 Li e C22 135 50 20 Gallbladde , bile duc s C23–24 140 60 20 Panc eas C25 530 230 85 O he diges i e o gans C26 30 10 0 Respi a o y o gans C30–39 1050 420 130 Nose, sinuses C30–31 35 15 5 La ynx, epiglo is C32 15 5 2 Lung, achea C33–34 980 350 110 Medias inum, pleu a C38 10 8 4 Bone C40–41 90 40 10 Melanoma o he skin C43 2200 1200 430 Skin, nonmelanoma C44 400 180 60 Meso helioma C45 20 9 3 Au onomic ne ous sys em C47 10 4 0 So issues C48–49 280 140 60 B eas C50 18,400 9200 3000 Female geni al o gans C51–58 3800 1700 550 Ce ix u e i C53 1130 630 260 Co pus u e i C54 1100 400 130 U e us, o he C55, C58 20 5 1 O a y C56 1100 500 150 O he emale geni al C51–52, C57 280 110 25 Placen a 15 4 2 U ina y o gans C64–68 900 400 160 Kidney C64 620 290 120 Bladde and u ina y ac C65–68 270 110 40 Eye C69 70 30 10 B ain, cen al ne ous sys em C70–72, D32–33, D42–43 2500 1400 500 Thy oid gland C73 2300 1300 520 O he endoc ine glands C74–75 50 20 5 Ill de ined o unknown C76, C80 300 130 40 Lymphoid and hema opoi- e ic issue C81–96, D45–47, D76 2600 1300 430 Hodgkin lymphoma C81 350 180 55 Non- Hodgkin lymphoma C82–86, C96, D76 1200 600 200 Myeloma C90 220 100 25 Leukemia C91–95 600 300 90 196% o he Finnish p egnan women be ween 1983 and 2016 ha e consen ed o pa icipa e in he FMC. As he women can wi hd aw consen a will he numbe s should be conside ed as es ima es. 3062 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d. M. Leh inen e al.Cance Cases in Ma e ni y Coho Ma e ial and Me hods A he end o 2016 he FMC comp ised 2.0 million se um samples om abou 1 million women. As he congeni al sc eening is epea ed on each p egnancy, se ial samples a e a ailable o abou 50% o he women. Mo eo e , he e a e app oxima ely 100,000 pai s o mo he s and daugh e s, who bo h ha e dona ed se a o he FMC o scien i ic esea ch. We es ima ed he numbe s o inciden cance cases in he FMC (Table 1). Cance incidence in he FMC is simila o ha in he gene al emale popula ion o all cance ypes [11], wi h he excep ion o endome ial cance which is dec eased (43%) due o p o ec ion om p egnancy. Thus, i was possible o es ima e he numbe o cance cases by mul iplying he pe son- yea s gene a ed by women in FMC by he end o yea 2014 wi h cance incidence a e in he Finnish emale popula ion in each 5- yea age ca ego y (www.cance egis y. i). Resul s The e a e app oxima ely 40,000 p ospec i ely occu ing cases o cance wi hin 17.2 million pe son- yea s o ollow- up up o 2014. Du ing he i s 5 yea s a e p egnancy, he numbe o new cance cases is mode a e bu inc eases when he women become olde . B eas cance comp ised almos hal o all inciden in asi e cance cases iden i ied (18,400 cases, Table 1). The nex 10 mos common cance ypes we e as ol- lows: hy oid cance (2300 cases), melanoma (2200 cases), colo ec al cance s (2190 cases), lymphomas (1550 cases), ce ical cance (1130 cases), endome ial cance (1100 Table 2. Es ima ed1 numbe s o inciden in asi e cance cases in he o sp ing o he Finnish Ma e ni y Coho pa icipan s (FMC) 1983–2014. Cance ype ICD10 Boys Gi ls All 0–4 5–9 1 ± 14 0–14 0–4 5–9 10–14 0–14 0–4 5–9 10–14 0–14 All si es C00–96, D32–33, D42–43, D45–47, D76 1180 535 445 2160 1020 420 410 1850 2200 955 855 4010 Mou h, pha ynx C00–14 <5 <5 10 10 0 <5 5 10 <5 5 15 20 Diges i e o gans C15–26 20 10 25 60 15 10 45 70 35 25 70 130 Respi a o y o gans C30–39 10 <5 5 15 10 5 0 15 15 10 5 30 Female geni al o gans C51–58 10 5 20 35 10 5 20 35 Male geni al o gans C60–63 30 5 5 40 30 5 5 40 U ina y o gans C64–68 110 20 5 135 105 20 5 130 220 40 10 270 Melanoma o he skin C43 5 5 15 20 0 5 10 15 5 10 20 35 Skin, nonmelanoma C44 0 5 5 10 5 0 5 5 5 5 5 15 Eye C69 55 5 5 65 45 5 5 50 105 5 5 115 Thy oid gland C73 5 5 5 15 0 5 15 20 5 10 20 35 O he endoc ine glands C74–75 90 10 10 110 90 5 5 100 180 15 15 210 Bone C40–41 5 15 25 45 15 10 25 50 15 20 55 90 So issues C48–49 40 20 15 75 45 20 10 75 90 35 30 155 Ill de ined o unknown C76, C80 5 5 5 15 5 5 5 15 10 10 5 25 Au onomic ne ous sys em C47 40 5 5 50 40 5 5 50 80 5 5 90 B ain, CNS C70–72, D32–33, 42–43 265 185 140 590 195 145 115 455 460 330 260 1050 Lymphoid, hema opoe ic issue C81–96, D45–47, D76 500 235 175 910 445 175 155 770 940 410 325 1675 196% o all Finnish p egnan women be ween 1983 and 2014 ha e consen ed o pa icipa e in he FMC. As he women can la e wi hd aw consen a will he numbe s should be conside ed as es ima es. 3063 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d. Cance Cases in Ma e ni y Coho M. Leh inen e al. cases), o a ian cance (1100 cases), lung cance (980 cases), kidney cance (620 cases), and leukemia (600 cases). Nonin asi e cance s include 10,000 cases o in si u ce i- cal cance , 7400 cases o basal cell ca cinomas, and 1500 cases o in si u b eas ca cinomas. The numbe o childhood cance cases in he o sp ing o he FMC dono s (4000 cases) is sizeable (Table 2). Especially, he numbe s o childhood leukemias and lym- phomas (al oge he mo e han 1600 cases) a e high (Table 2). Discussion Cance incidence in he FMC is simila o ha in he gene al emale popula ion o all cance ypes [11], wi h he excep ion o endome ial cance which is dec eased (43%) due o p o ec ion om p egnancy. Wi h only 20 women op ing ou in 2017, he popula ion- based na u e o FMC emains i ually in ac . In addi ion o longi udinal s udies on cance e iology and sc eening s udies, he esou ce is also po en ially use- ul o s udies on gene ic cance isk. Howe e , he i s imes e se um samples con ain measu able amoun s o e al DNA, which needs o be con olled o in gene ic s udies [12]. Because he FMC is na ionwide and con ains samples om 94% o all Finnish p egnan women [11], e y la ge- scale o e - gene a ion s udies can be designed by linking he iden i ies o he mo he s o hei o sp ing and/o ela i es [13]. Due o he long his o y o he FMC, p ospec s o s udying congeni al causes o cance in he o sp ing o he women in FMC a e also good. Ongoing FMC p ojec s include s udies o he possible ole o he same mic obiome, ha is, simila se ological signa u e o mo he s and daugh e s wi h he same cance o he same ypes o Chlamydia achoma is [14], HPV [7], Helicobac e pylo i, o S ep ococcus galoly icus in amilial ce ical, colo ec al, and s omach cance s, espec i ely. The FMC sample collec ion can also suppo clinical ials and su eillance pu poses, including cance con ol. Fo example, we a e cu en ly compa ing he long- e m s abili y o an ibody esponses induced by ei he quad- i alen o bi alen accines agains HPV among women who pa icipa ed as adolescen s in la ge andomized albei popula ion- based ials o hese accines [15]. By linking he clinical ial iles o he FMC iles, i has been pos- sible o iden i y se um samples collec ed up o 14 yea s pos accina ion. E alua ion o accine- induced an ibody esponse in ye - o- be- iden i ied b eak h ough cases makes sea ch o co ela es o p o ec i e immuni y easible. This coho p o ile desc ibes he basic cha ac e is ics o he FMC and p o ides some examples o i s po en ial use. The FMC is also a ole model in s ic sa e gua d- ing o in eg i y and compliance wi h da a p o ec ion laws. While linkages o cou se need o be pe o med wi h iden i iable da a, once he samples a e e ie ed hey a e pseudonymized and i is no possible o link he esea ch da a o an iden i iable indi idual. To p o- mo e he use o he FMC as an in e na ional esou ce o epidemiological cance esea ch, an in e na ional nonp o i company (ESIS EEIG) specializing in assis ing in e na ional esea che s wi h he da a and samples hey need has been ounded and is a ailable o acili a e he o mal and logis ic p ocess o ob ain access o he sam- ples and da a ha in e na ional cance esea ch may need (www.esis. i). Acknowledgmen s We hank P. Koskela, who o iginally es ablished he FMC Biobank. Con lic s o In e es None decla ed. Re e ences 1. Hill, A. B. 1965. The en i onmen and disease: associa ion o causa ion. P oc. R. Soc. Med. 58:295–300. 2. Doll, R., and R. Pe o. 1981. Causes o cance . J. Na l. Cance Ins . 166:1191–1308. 3. Koskela, P., T. An ila, T. Bjo ge, A. B uns ig, J. Dillne , M. Hakama, e al. 2000. Chlamydia achoma is in ec ion is a isk ac o o ce ical cance . In . J. Cance 85:35–39. 4. Dillne , J., M. Leh inen, T. Bjo ge, T. Luos a inen, L. Youngman, E. Jellum, e al. 1997. A p ospec i e se oepidemiological s udy o human papilloma i us in ec ion as a isk ac o o in asi e ce ical cance . J. Na l. Cance Ins . 89:1293–1299. 5. Bjo ge, T., J. Dillne , T. An ila, V. Abele , A. Engeland, T. Hakulinen, e al. 1997. A p ospec i e se oepidemiological s udy o human papilloma i us and non- ce ical anogeni al cance s. BMJ 315:646–649. 6. Mo k, J., A.-K. Lie, E. Gla e, S. Cla k, G. Hallmans, E. Jellum, e al. 2001. A p ospec i e s udy on human papilloma i us as a isk ac o o head and neck cance . N. Engl. J. Med. 344:1125–1231. 7. Luos a inen, T., P. Namujju, M. Me ikukka, H. M. Su cel, T. Hakulinen, J. Dillne , e al. 2013. O de o p e alen /inciden sexually ansmi ed in ec ions and he isk o CIN g ade 3. In . J. Cance 133:1756–1760. 8. Kapeu, A., L. Youngman, E. Jellum, J. Dillne , M. Hakama, P. Koskela, e al. 2009. Smoking is an independen isk ac o o ce ical cance . Am. J. Epidemiol. 169:480–488. 9. Leh inen, M., P. Koskela, H. Ögmundsdo i , A. Bloigu, J. Dillne , M. Gudnado i , e al. 2003. Ma e nal 3064 © 2017 The Au ho s. Cance Medicine published by John Wiley & Sons L d. M. Leh inen e al.Cance Cases in Ma e ni y Coho he pes i us in ec ions and isk o acu e lymphoblas ic leukaemia in he o sp ing. Am. J. Epidemiol. 158:207–213. 10. Fo ne , R., H. Schock, R. Kaaks, M. Leh inen, E. Pukkala, H. Å. Lakso, e al. 2017. Human ch onionic gonado opin and ma e nal b eas cance : a nes ed case- con ol s udy in he Finnish Ma e ni y Coho . Can. Res. 77:134–141. 11. Pukkala, E., A. Ande ssen, G. Be glund, R. Gisle oss, V. Gudnason, G. Hallmans, e al. 2007. No dic biological specimen banks as basis o s udies o cance causes and con ol – mo e han 2 million sample dono s, 25 million pe son yea s and 100,000 p ospec i e cance s. Ac a Oncol. 46:286–307. 12. Sjöholm, M. I., G. Ho man, S. Lindg en, J. Dillne , and J. Ca lson. 2005. Compa ison o a chi al plasma and o malin- ixed pa a in- embedded issue o geno yping in hepa ocellula ca cinoma. Cance Epidemiol. Bioma ke s P e . 14:251–255. 13. Leh inen, M., M. Pawli a, K. Zumbach, M. Hakama, E. Jellum, P. Koskela, e al. 2003. E alua ion o an ibody esponse o human pilloma i us ea ly p o eins in women who de eloped ce ical cance 1- 20 yea s la e . Am. J. Obs e . Gynecol. 88:49–55. 14. An ila, T., P. Saikku, P. Koskela, A. Bloigu, J. Dillne , I. Ikäheimo, e al. 2001. Se o ypes o Chlamydia achoma is and isk o ce ical squamous cell ca cinoma. JAMA 285:46–51. 15. Leh inen, M., and J. Dillne . 2013. Clinical HPV accina ion ials and beyond. Na . Re . Clin. Oncol. 10:400–410.