Parental Occupational Exposure to Organic Solvents and Testicular Germ Cell Tumors in their Offspring: NORD-TEST Study

Pa en al Occupa ional Exposu e o O ganic Sol en s and Tes icula Ge m Cell Tumo s in hei O sp ing: NORD-TEST S udy Cha lo e Le Co ne , 1,2 Béa ice Fe e s, 3,4 Ee o Pukkala, 5,6 To e Tynes, 7,8 Ma ia Feych ing, 9 Johnni Hansen, 10 Kayo Togawa, 1 Ka l-Ch is ian No dby, 8 Susanne Oksbje g Dal on, 10 Sanni Uuksulainen, 11 Pe nilla Wiebe , 12 To ill Woldbæk, 8 Niels E. Skakkebæk, 13 Ann Olsson, 1,12 and Joachim Schüz 1 1 Sec ion o En i onmen and Radia ion, In e na ional Agency o Resea ch on Cance (IARC), Lyon, F ance 2 Di ision o Cance Epidemiology, Deu sches K ebs o schungszen um (DKFZ), Heidelbe g, Ge many 3 Dépa emen Cance e En i onnemen , Cen e Léon Bé a d, Lyon, F ance 4 Uni e si é Claude Be na d–Lyon1, 43 Bl d. du 11 No emb e 1918, Villeu banne, F ance 5 Finnish Cance Regis y, Ins i u e o S a is ical and Epidemiological Cance Resea ch, Helsinki, Finland 6 Facul y o Social Sciences, Uni e si y o Tampe e, Finland 7 Cance Regis y o No way, Majo s uen, Oslo, No way 8 Na ional Ins i u e o Occupa ional Heal h, Oslo, No way 9 Uni o Epidemiology, Ins i u e o En i onmen al Medicine, Ka olinska Ins i u e , S ockholm, Sweden 10 Danish Cance Socie y Resea ch Cen e , Copenhagen, Denma k 11 Finnish Ins i u e o Occupa ional Heal h (FIOH), Helsinki, Finland 12 Uni o Occupa ional Medicine, Ins i u e o En i onmen al Medicine, Ka olinska Ins i u e , S ockholm, Sweden 13 In e na ional Resea ch and Resea ch T aining Cen e in Endoc ine Dis up ion o Male Rep oduc ion and Child Heal h (EDMaRC), Copenhagen, Denma k BACKGROUND:Tes icula ge m cell umo s (TGCT) we e sugges ed o ha e a p ena al en i onmen ally ela ed o igin. The po en ial endoc ine dis up - ing p ope ies o ce ain sol en s may in e e e wi h he male geni al de elopmen in u e o. OBJECTIVES:We aimed o assess he associa ion be ween ma e nal and pa e nal occupa ional exposu es o o ganic sol en s du ing he p ena al pe iod and TGCT isk in hei offsp ing. METHODS:This egis y-based case con ol s udy included TGCT cases aged 14–49 y (n= 8,112) diagnosed om 1978 o 2012 in Finland, No way, and Sweden. Con ols (n= 26,264) we e andomly selec ed om he cen al popula ion egis ies and we e indi idually ma ched o cases on yea and coun y o bi h. Occupa ional his o ies o pa en s p io o he child’s bi h we e ex ac ed om he na ional censuses. Job codes we e con e ed in o sol en exposu e using he No dic job-No dic Occupa ional Cance S udy Job-Exposu e Ma ix. Condi ional logis ic eg ession was used o es ima e odds a ios (OR) and 95% confidence in e als (CI). RESULTS:O e all, no associa ion was ound be ween p ena al ma e nal exposu e o sol en s and TGCT isk. In subse analyses using only mo he s o whom occupa ional in o ma ion was a ailable in he yea o o in he yea p io o he child’s bi h, he e was an associa ion wi h ma e nal exposu e o a oma ic hyd oca bon sol en s (ARHC) (OR = 1:53; CI: 1.08, 2.17), d i en by exposu e o oluene (OR = 1:67 ; CI: 1.02, 2.73). No associa ion was seen o any pa e nal occupa ional exposu e o sol en s wi h he excep ion o exposu e o pe chlo oe hylene in Finland (OR = 2:42; CI: 1.32, 4.41). CONCLUSIONS:This s udy sugges s a modes inc ease in TGCT isk associa ed wi h ma e nal p ena al exposu e o ARHC. h ps://doi.o g/10.1289/ EHP864 Backg ound Tes icula ge m cell umo s (TGCT) a e he mos common malig- nancy in young men aged 15 o 39 yea s old (Fe lay e al. 2013). The age-s anda dized incidence a e o es icula cance using he wo ld s anda d popula ion has inc eased d as ically in he pas 40 yea s, eco ding an 8- old inc ease in Finland ( om 1.8 o 15.0 pe 100,000 pe son-yea s), 3- old inc ease in Sweden ( om 4.6 o 15.4) and No way ( om 7.8 o 22.9), and a 1.5- old inc ease in Denma k ( om 12.2 o 18.2) be ween 1970 and 2013 (Engholm e al. 2016). This s eep inc ease, he la ge geog aphical a ia ion in incidence, and he dis inc pa e ns o incidence a es in mig an s (Schmiedel e al. 2010) suppo he ole o en i on- men al ac o s in he e iology o TGCT. In Eu ope, p edic ion o he u u e bu den o es icula cance up o 2025 sugges s con inuing incidence inc ease (Le Co ne e al. 2014). Al hough causes and isk ac o s o TGCT emain poo ly unde s ood, he uni o mly inc easing incidence bu den om es icula cance in Eu ope sugges s inc easing isk in men in mos popula ions, a he han ongoing changes in he na ional demog aphic p ofiles. While many epidemiological s udies ha e in es iga ed isk ac o s o TGCT, findings ha e been inconclusi e. The ea ly age a TGCT diagnosis, oge he wi h es ablished associa ions wi h congeni al mal o ma ions such as c yp o chidism (Cook e al. 2010;K a z e al. 2010) and wi h bi h cha ac e is ics such as bi h o de , pa i y, and low bi h weigh (Cook e al. 2009,2010), sugges a p ena al o igin o TGCT. Also, a a ecen in e na ional consensus mee ing, i was concluded ha he common p ecu so cells o TGCT (bo h seminomas and nonseminomas) ha e e al cha ac e is ics, including emb yonic ma ke s (Moch e al. 2016). Howe e , he majo i y o published epidemiological s udies ha e ocused on adul hood exposu es (Bé ange e al. 2013). Acco ding o he es icula dysgenesis synd ome (TDS) hy- po hesis, TGCT, c yp o chidism, hypospadias, and educed spe m coun o igina e om a common e al diso de hampe ing he no mal male es icula de elopmen (Skakkebaek e al. 2016). Exposu e o endoc ine-dis up ing chemicals (EDCs) has been sugges ed as a plausible ac o o ini ia ing TDS h ough Add ess co espondence o J. Schüz, Sec ion o En i onmen and Radia ion, In e na ional Agency o Resea ch on Cance , 150 Cou s Albe Thomas, 69372 Lyon CEDEX 08, F ance. Telephone: +33 (0)4 72 73 84 41. Email: [email p o ec ed] Supplemen al Ma e ialis a ailable online (h ps://doi.o g/10.1289/EHP864). The au ho s decla e hey ha e no ac ual o po en ial compe ing financial in e es s. Recei ed 25 July 2016; Re ised 16 Decembe 2016; Accep ed 22 Decembe 2016; Published 30 June 2017. No e o eade s wi h disabili ies: EHP s i es o ensu e ha all jou nal con en is accessible o all eade s. Howe e , some figu es and Supplemen al Ma e ial published in EHP a icles may no con o m o 508 s anda ds due o he complexi y o he in o ma ion being p esen ed. I you need assis ance accessing jou nal con en , please con ac [email protected] . Ou s aff will wo k wi h you o assess and mee you accessibili y needs wi hin 3 wo king days. En i onmen al Heal h Pe spec i es 067023-1 A Sec ion 508–con o man HTML e sion o his a icle is a ailable a h ps://doi.o g/10.1289/EHP864. Resea ch in e e ence wi h gonadal de elopmen by mimicking ho mones o in e e ing wi h ho mone p oduc ion (Sha pe 2006). One epi- demiological s udy ound he concen a ion o pe sis en o ganic pollu an s such as polychlo ina ed biphenyl (PCB), hexachlo o- benzene (HCB), and polyb omina ed diphenyl e he s (PBDEs) o be significan ly highe o mo he s o es icula cance cases com- pa ed o mo he s o con ols (Ha dell e al. 2003,2006). Widely used sol en s such as oluene, ichlo oe hylene, and pe chlo oe hylene ha e possible endoc ine dis up ing p ope ies (B ouwe s e al. 2009), and migh also in e ac wi h he masculin- iza ion p ocess in u e o (Sha pe 2006). Fu he mo e, benzene and ichlo oe hylene we e classified as ca cinogenic o humans (G oup 1) (IARC 2012,2014), and pe chlo oe hylene (also called e achlo oe hylene) was classified as p obably ca cinogenic o humans (G oup 2A) by he In e na ional Agency o Resea ch on Cance (IARC 2014). O ganic sol en s a e used o ex ac ing, dissol ing, o suspending ma e ials such as a s, waxes, and es- ins ha a e no soluble in wa e . O ganic sol en s can be ound in a ious commonly used p oduc s such as pain s, adhesi es, cos- me ics, glues, and plas ics (Ve ma and Rana 2009), esul ing in widesp ead exposu es. Sol en exposu e is pa icula ly high in specific occupa ions and indus ies, such as plas ic p oduc wo k- e s, chemical and ood indus y, well d ille s, laund y wo ke s, and ubbe p oduc wo ke s (B ouwe s e al. 2009;Kauppinen e al. 2009). Li le is known abou he ole o e al exposu es o o ganic sol- en s in TGCT de elopmen (Bé ange e al. 2013). Only one s udy, based on a small numbe o cases, in es iga edpa en al occupa ions ha we e po en ially exposed o hyd oca bons (ai c a mechanics, engine epai e s, machinis s, ligh uck d i e s, gas andse ice s a- ion a endan s, pain e s, and p in e s) and ound no associa ion wi h TGCT in he offsp ing (Ka daun e al. 1991). Gi en he TDS hypo hesis and he endoc ine p ope ies o some sol en s such as pe chlo oe hylene (Eu opean Commission DG ENV 2000), we assessed pa en al occupa ional exposu e o sol en s du ing he p ena al pe iod and he isk o de eloping TGCT in he offsp ing in a la ge-scale popula ion-based case– con ol s udy linking he egis ies o he No dic coun ies, he NORD-TEST s udy (Le Co ne e al. 2015). Ma e ial and Me hods S udy Popula ion The NORD-TEST s udy, which has been desc ibed p e iously in de ail (Le Co ne e al. 2015), is a egis y-based case–con ol s udy nes ed wi hin he popula ions o Denma k, Finland, No way, and Sweden. All esiden s o he No dic coun ies a e egis e ed wi h a unique pe sonal iden i y code, allowing linkage be ween egis ies o acking pe sonal medical in o ma ion and occupa ional s a us o esea ch pu poses. The occupa ionalin o ma ion is based on job codes de i ed om censuses in Finland, No way, and Sweden and on indus y codes de i ed om he Supplemen a y Pension Fund’s da abase in Denma k. As we expec ed ha he ha moniza ion o da a could esul in unce ain y in sol en exposu e es ima es o Denma k, we decided o exclude da a om Denma k om he p es- en pooled analysis. All es icula cance cases diagnosed be ween ages 14 and 49 y old om 1988–2012 in Finland, 1978–2010 in No way, and1979–2011 in Sweden we eex ac ed om he espec- i e na ionwide cance egis y, using he ollowing In e na ional Classifica ion o Diseases (ICD) codes: ICD-7: 178 (WHO 1955); ICD-8 and ICD-9: 186 (WHO 1968,1978); and ICD-O-3: C62 (WHO 2000). Fou con ols we e andomly selec ed om he cen- al popula ion egis ies and we e indi idually ma ched o cases by yea and coun y o bi h. Cases and hei ma ched con ols we e eli- gible when hey we e ee o any p e ious neoplasm (excep nonmelanoma skin cance ) and, o con ols, we e ali e on he da e o es icula cance diagnosis o hei ma ched case. The pa en al occupa ional codes (Finland, No way, and Sweden) p io o he cases’o he con ols’bi h had o be known o a leas one o he pa en s o be included in he analysis. The final sample included 8,112 cases and 26,264 con ols: 3,769 cases wi h ou ma ched con ols, 2,816 cases wi h h ee ma ched con ols, 1,213 cases wi h wo ma ched con ols, and 314 cases wi h one ma ched con ol. Exposu e Assessmen Job codes o bo h pa en s we e e ie ed om he las census conduc ed p io o he child’s bi h and he fi s census conduc ed a e he child's bi h. The No dic Occupa ional Cance S udy Job-Exposu e Ma ix (NOCCA-JEM), cons uc ed by a eam o expe s in 2009 (Kauppinen e al. 2009), was used o assign pa- en al exposu es. The NOCCA-JEM con e s he occupa ional codes in o coun y- and ime-specific quan i a i e exposu e es i- ma es o sol en s. The pa en al exposu e es ima es we e calcu- la ed as he p oduc o he p opo ion o exposed wo ke s (P) wi hin an occupa ion and he mean le el o exposu e o he agen among he exposed wo ke s (L) (Kauppinen e al. 2014). The sol- en exposu e is exp essed in pa s pe million (ppm) in he wo k- oom ai o he wo king day. Censuses we e manda o y o he whole popula ion in he No dic coun ies and ook place e e y 5 y in Sweden (1960–1990, excep o he 1965 census) and in Finland (1970–1990), and e e y 10 y in No way (1960–1990). The occupa ional exposu e o sol en s was es ima ed based on he occupa ional i les eco ded in he census in he yea o he child’s bi h o he las census be o e he child’s bi h. Pa en al exposu es a ailable in NOCCA-JEM include he ollowing six indi idual sol en s: benzene, oluene, pe chlo oe hy- lene, me hylene chlo ide, ichlo oe hylene, and 1,1,1- ichlo o- e hane. Addi ionally, h ee g oups o sol en s we e c ea ed, which include he ollowing: a oma ic hyd oca bon sol en s (ARHC, combining exposu e o benzene and oluene), chlo ina ed hyd oca - bon sol en s (CHC, combining exposu e o ichlo oe hanes, i- chlo oe hylene, pe chlo oe hylene, and me hylene chlo ide), and “any sol en ”exposu e, combining exposu e o he six indi idual sol en s. Addi ional da a ob ained a e as ollows: es icula cance am- ily his o y was e ie ed o a he s and b o he s o cases and con ols om he cance egis ies, and in o ma ion on diagnosis and/o su ge y o c yp o chidism, hypospadias, and inguinal he - nia in cases and con ols was ex ac ed om he Medical Bi h Regis y and he Hospi al Discha ge Regis y, as well as o Finland, om he egis y o congeni al mal o ma ions. Fu he de ails ha e been desc ibed p e iously (Le Co ne e al. 2015). S a is ical Me hods Pai wise co ela ions be ween bina y sol en s exposu e a iables we e calcula ed using he phi coefficien . Bi a ia e analysis was used o assess associa ions o TGCT wi h he ollowing ac o s: a he ’s his o y o es icula cance ; b o he ’s his o y o es icula cance ; pe sonal his o y o inguinal he nia, hypospadias, and c yp o chidism; age o a he a childbi h; and age o mo he a childbi h. Condi ional logis ic eg ession was used o es ima e he s eng h o he associa ions gi en by odds a ios (OR) and he co esponding 95% confidence in e als (CI). Each pa en al occupa ional exposu e was classified in o wo ca ego ies (exposed/unexposed) as well as, o he analyses o he indi idual sol en s, in o h ee ca ego ies (unexposed, low expo- su e, and high exposu e). The cu offs o exposu e we e specific o each sol en s’exposu e, we e based on hei dis ibu ion among pa en s o con ols, and we e se a he 90 h (pa e nal En i onmen al Heal h Pe spec i es 067023-2 exposu e) o 75 h pe cen ile (ma e nal exposu e) o ha e enough subjec s in he high exposu e ca ego y (see Table S1). I should be no ed ha he es ima ion o exposu e le els esul s in a semi- quan i a i e a iable a he han ully quan i a i e one, as illus- a ed wi h he example o me hylene chlo ide exposu e le els among mo he s and 1,1,1- ichlo oe hane exposu e le els among a he s in Supplemen a y Figu e S1. This explains why in he ca - ego iza ion by pe cen ile, he cu offis no always a he exac 90 h o 75 h cu off. I is also he eason why we decided in a o o ca ego ical analyses a he han mo e complex modeling o dose– esponse ends. Models we e buil using he unexposed g oup o each sol en indi idually as he e e ence ca ego y. Age o a he a bi h, age o mo he a bi h, and amily his o y o es- icula cance [ a he and/o b o he (s)] we e in oduced as po en- ial con ounde s in he models, bu did no change he OR es ima e wi h any sol en by 10% o g ea e . Analyses we e s a ified, and he esul s we e epo ed by TGCT sub ype and by coun y. Coun y-specific analyses we e no conduc ed o ma e nal exposu es because he numbe o exposed cases was oo small. The Wald es was used o assess he he e ogenei y o effec s ac oss s a a. An al e na i e analysis was pe o med whe e he e e ence g oup was edefined as hose ha we e no exposed o any o he sol en s e alua ed. Fu he mo e, since he e is no s ong biological a gumen sup- po ing a ole o pa en al exposu e be o e 1 y o p egnancy in he e iology o es icula cance , a sensi i i y analysis was pe o med o a ge he concep ion and p egnancy ime, he eby educing he po en ial misclassifica ion due o job changes. The e o e, he pop- ula ion was es ic ed o bo h pa en s o whom he census was a ailable he yea o o he yea p io o he child’s bi h as well as o pa en s o whom he census be o e and a e childbi h eco ded he same occupa ion, assuming ha he pa en s held his occupa ion also he yea o and he yea p io o he child’s bi h. Addi ionally, he Kappa coefficien was calcula ed o measu e he deg ee o ag eemen be ween p ena al and adul hood occupa- ional exposu e o sol en in cases and con ols. All analyses we e pe o med using SAS s a is ical package ( e sion 9.3; SAS Ins i u e, Inc.). NORD-TEST has been app o ed by he ele an da a p o ec- ion and e hical commi ees in Finland, No way, and Sweden, and by he IARC e hics commi ee. Resul s A o al o 8,112 TGCT cases (seminoma and nonseminoma) and 26,264 con ols we e included in he s udy. Coun y-specific dis- ibu ions o cases and con ols ha e been desc ibed p e iously (Le Co ne e al. 2015). Table 1 shows he dis ibu ions o diffe - en cha ac e is ics o cases and con ols. O e all, 94% o he cases we e diagnosed be ween 1990 and 2012, and 81% we e bo n be ween 1960 and 1980. The p opo ion o nonseminoma cases was 55% wi h he median age o 26.7 yea s, whe eas ha o seminoma was 45% wi h he median age o 32.2 yea s. Abou 87% o he mo he s and 97% o he a he s had an occupa ional s a us known be o e he child’s bi h, whe eo 2.3% o he mo h- e s and 13.1% o he a he s we e es ima ed as ha ing been exposed o sol en s in hei wo kplace. The p e alence o occu- pa ional exposu e o sol en s was ela i ely low in mo he s, ang- ing om 0.6% o pe chlo oe hylene o 1.1% o 1,1,1- ichlo oe hane. The p e alence was somewha highe in a he s, anging om 1.2% o pe chlo oe hylene o 10.7% o 1,1,1- i- chlo oe hane. Pai wise co ela ions be ween bina y sol en expo- su es anged om 0.00 o 0.73 o ma e nal exposu es and om −0:01 o 0.89 o pa e nal exposu es (Table S2). The lowes co - ela ion was ound be ween oluene and pe chlo oe hylene, whe eas he highes was ound o ichlo oe hylene and 1,1,1- ichlo oe hane in ma e nal exposu es and o oluene and ben- zene in pa e nal exposu es. While pe sonal his o y o inguinal he nia, age o a he , and age o mo he a childbi h we e no associa ed wi h TGCT isk, c yp o chidism, hypospadias, and b o he ’s and a he ’s his o y o es icula cance we e associa ed wi h an inc eased TGCT isk (Table 1). When he co a ia es conside ed as po en ial con- ounde s we e in oduced one by one in he models, he ORs co - esponding o sol en exposu es did no change by 10% o mo e. Table 1. Demog aphic cha ac e is ics o he s udy popula ion by case– con ol s a us and odds a ios (OR) o p ena al ac o s known o po en ially ela ed o es icula cance isk. Demog aphic cha ac e is ic Cases n(%) Con ols n(%) OR (95% CI) To al 8,112 26,264 Coun y Finland 1,034 (12.7) 4,030 (15.3) No way 3,163 (39.0) 9,089 (34.6) Sweden 3,915 (48.3) 13,145 (50.0) Bi h yea 1960–1964 1,651 (20.4) 5,352 (20.4) 1965–1969 1,489 (18.4) 4,490 (17.1) 1970–1974 1,926 (23.7) 6,398 (24.4) 1975–1979 1,518 (18.7) 4,848 (18.5) 1980–1995 1,528 (18.8) 5,176 (19.7) Diagnos ic pe iod 1978–1989 469 (5.7) 1990–1994 825 (10.2) 1995–1999 1,376 (17.0) 2000–2004 1,965 (24.2) 2005–2009 2,517 (31.0) 2010–2012 960 (11.8) His ological sub ype Seminoma 3,687 (45.4) Nonseminoma 4,425 (54.6) Age o mo he a childbi h 14–20 440 (5.4) 1,500 (5.7) 1.00 20–29 5,155 (63.6) 16,625 (63.3) 1.04 (0.93, 1.17) 30–39 2,314 (28.5) 7,484 (28.5) 1.04 (0.92, 1.17) 40 + 192 (2.4) 647 (2.5) 1.00 (0.82, 1.22) Missing 11 (0.1) 8 (0.0) Age o a he a childbi h 14–20 85 (1.1) 250 (1.0) 1.00 20–29 4,050 (49.9) 13,205 (50.3) 0.89 (0.69, 1.15) 30–39 3,168 (39.1) 10,353 (39.4) 0.89 (0.69, 1.14) 40 + 732 (9.0) 2,285 (8.7) 0.92 (0.71, 1.20) Missing 77 (1.0) 171 (0.7) Fa he ’s his o y o es icula cance No 8,043 (99.2) 26,192 (99.7) 1.00 Yes 69 (0.9) 72 (0.3) 2.95 (2.11, 4.12) B o he ’s his o y o es icula cance No 8,055 (99.3) 26,230 (99.9) 1.00 Yes 57 (0.7) 34 (0.1) 5.30 (3.45, 8.12) Pe sonal his o y o inguinal he nia No 6,991 (86.2) 22,820 (86.9) 1.00 Yes 108 (1.3) 414 (1.6) 0.93 (0.75, 1.16) Missing 1,013 (12.5) 3,030 (11.5) Pe sonal his o y o hypospadias No 7,062 (87.1) 23,187 (88.3) 1.00 Yes 37 (0.5) 47 (0.2) 2.52 (1.62, 3.90) Missing 1,013 (12.5) 3,030 (11.5) Pe sonal his o y o c yp o chidism No 7,008 (86.4) 23,115 (88.0) 1.00 Yes 91 (1.1) 119 (0.5) 2.56 (1.94, 3.38) Missing 1,013 (12.5) 3,030 (11.5) No e: CI, con idence in e al; n, numbe ; OR, odds a io. En i onmen al Heal h Pe spec i es 067023-3 Adjus ed ORs a e he e o e p esen ed in he supplemen a y file (Table S3). Figu e 1 shows he associa ion be ween p ena al pa- en al occupa ional exposu es o sol en s and he isk o TGCT in he offsp ing. Sligh ly ele a ed ORs we e ound among sons whose mo he s we e exposed o se e al indi idual sol en s such as benzene, oluene, me hylene chlo ide, and he g ouped sol en Figu e 1. Fo es plo o he isk o de elop es icula ge m cell umo a e pa e nal o ma e nal exposu e o sol en s p io o bi h. Odds a ios (ORs) and co e- sponding 95% confidence in e als (CIs) we e de i ed based on logis ic eg ession analysis condi ional on yea and coun y o child's bi h, he e e ence ca e- go y (unexposed) was he pa en s ha we e no occupa ionally exposed o he sol en o in e es . A oma ic hyd oca bon sol en s (ARHC) include benzene and/o oluene. Chlo ina ed hyd oca bon sol en s (CHC) include me hylene chlo ide, pe chlo oe hylene, ichlo oe hylene, and/o 1,1,1- ichlo oe hane. Any sol en ca ego y includes a leas one o he ollowing: benzene, oluene, me hylene chlo ide, pe chlo oe hylene, ichlo oe hylene, 1,1,1- ichlo oe hane. En i onmen al Heal h Pe spec i es 067023-4 ARHC, as illus a ed in Figu e 1, bu none o he associa ions we e s a is ically significan . When he analysis was epea ed wi h a e e ence g oup o mo he s occupa ionally no exposed o any sol en s (Table 3), he associa ions emained no significan . Simila ly, he analysis o low and high le els o ARHC exposu e (Table 2) did no show an inc eased TGCT isk in he offsp ing. Only he low-exposu e ca ego y o ARHC had a bo de line sig- nifican associa ion. Howe e , he sensi i i y analysis es ic ed o sons bo n in he yea o o he yea a e he census, o sons o pa en s who held he same occupa ion a censuses be o e and a - e he child’s bi h, showed a s onge s a is ically significan associa ion be ween ma e nal p ena al exposu e o ARHC and TGCT in offsp ing (OR = 1:53; 95% CI: 1.08, 2.17, Table 4). Al hough ORs emained sligh ly ele a ed o se e al indi idual sol en s, only oluene exposu e was ound significan ly associ- a ed wi h TGCT isk (OR = 1:67; 95% CI: 1.02, 2.73, Table 4)in he sensi i i y analysis. Mos ORs associa ed wi h pa e nal exposu es we e close o 1.0. When exposu e was subdi ided in o ca ego ies o low and high exposu es, none o he exposu es in a he s we e ound o be associa ed wi h TGCT isk (Table 2). In he sensi i i y analyses assessing he pa e nal occupa ional exposu e o sol en s in he yea o o he yea p io o he childbi h, no associa ions wi h TGCT isks in offsp ing we e ound, which is consis en wi h he esul s o he p ima y analysis. S a ifica ion by coun y (see Table S4) indica ed an inc eased TGCT isk in Finland wi h p e- na al pa e nal exposu e o pe chlo oe hylene (OR = 2:42; 95% CI: 1.32, 4.41), and he es o he e ogenei y be ween coun ies was s a is ically significan (p=0:02). No majo diffe ences in associa ions we e obse ed be ween seminomas and nonseminomas o any o he ma e nal o pa e - nal sol en exposu es (Table S5). Discussion In his la ge-scale nes ed case–con ol s udy o mo e han 8,000 TGCT cases in Finland, No way, and Sweden, we ound li le e idence o an associa ion be ween pa en al occupa ional expo- su e o sol en s and he isk o TGCT in hei offsp ing. The excep ion was a s a is ically significan mode a ely inc eased isk wi h ma e nal exposu e o ARHC and oluene when es ic ing he analysis o subjec s whose ma e nal exposu e in o ma ion nea he ime o p egnancy/childbi h was a ailable, possibly showing a s onge associa ion han he o e all analyses due o educed exposu e misclassifica ion. Coun y-specific esul s showed no addi ional associa ions besides he associa ion wi h pa e nal exposu e o pe chlo oe hylene, which was obse ed exclusi ely in Finland. The finding o a ole o ARHC and oluene exposu e has o be in e p e ed wi h cau ion, as he numbe o mo he s exposed was limi ed in he sensi i i y analysis, and he s onges associa- ion was ound in he low exposu e ca ego y. Job-exposu e ma i- ces (JEMs) alloca e he same le el and p obabili y o exposu e o all indi iduals wi hin one occupa ional code. The indi idual ex- posu e es ima es a e hus p one o nondiffe en ial misclassifica- ion, pa icula ly when he p obabili y o exposu e wi hin one occupa ional code is low. The exposu e ca ego iza ion (low, high) o he con inuous dis ibu ion o exposu e unde lying an- dom measu emen e o migh induce a sys ema ic nondiffe en ial o e - o unde es ima ion (B enne and Loomis 1994). The bo - de line significan OR obse ed o he low ca ego y o exposu e o ARHC migh be due o sys ema ic nondiffe en ial exposu e misclassifica ion. Howe e , i migh be no ed ha nonmono onic dose– esponse cu es ha e p e iously been desc ibed o expo- su es o EDCs in ela ion o a ious endpoin s (Vandenbe g e al. 2016). Also, he posi i e associa ion ound be ween ma e nal ex- posu e o ARHC and TGCT in offsp ing migh be d i en by olu- ene exposu e, since ARHC is he combina ion o benzene and oluene. Only one o he case–con ol s udy has in es iga ed occupa- ions o a he s and mo he s po en ially exposed o hyd oca bons, bu no excess isk o TGCT was obse ed in he offsp ing (Ka daun e al. 1991). Howe e , exposu e o indi idual chemical agen s we e no assessed, and he s udy, including 223 cases, has low s a is ical powe . Se e al agen s belonging o he g oup o ARHC sol en s, such as s y ene o agen s de i ing om benzene and oluene (pen achlo obenzene, ichlo obenzene, HCB, ni o- oluene), ha e been classified in he p io i y lis o EDCs by he Table 2. Pa en al occupa ional exposu e o sol en s be o e he child’s bi h ca ego ized in o low/high exposu e and he isk o es icula ge m cell umo in he o sp ing. Agen s Le el o exposu e a Ma e nal exposu e Pa e nal exposu e Cases n(%) Con ols n(%) OR (95% CI) b Cases n(%) Con ols n(%) OR (95% CI) b A oma ic hyd oca bon sol en s (ARHC) Benzene Unexposed 6,932 (98.8) 22,840 (99.0) 7,315 (93.1) 23,810 (93.4) Low 55 (0.8) 155 (0.7) 1.22 (0.89, 1.68) 267 (3.4) 862 (3.4) 1.01 (0.88, 1.16) High 31 (0.4) 86 (0.4) 1.10 (0.71, 1.71) 273 (3.5) 824 (3.2) 1.06 (0.92, 1.22) Toluene Unexposed 6,963 (99.2) 22,921 (99.3) 7,274 (92.6) 23,644 (92.7) Low 40 (0.6) 110 (0.5) 1.32 (0.90, 1.93) 505 (6.4) 1,615 (6.3) 1.02 (0.92, 1.13) High 15 (0.2) 50 (0.2) 1.01 (0.55, 1.84) 76 (1.0) 237 (0.9) 1.01 (0.77, 1.31) Chlo ina ed hyd oca bon sol en s (CHC) Me hylene chlo ide Unexposed 6,964 (99.2) 22,939 (99.4) 7,322 (93.2) 23,848 (93.5) Low 40 (0.6) 102 (0.4) 1.36 (0.94, 1.99) 423 (5.4) 1,311 (5.1) 1.06 (0.95, 1.19) High 14 (0.2) 40 (0.2) 1.28 (0.69, 2.40) 110 (1.4) 337 (1.3) 1.01 (0.81, 1.26) Pe chlo oe hylene Unexposed 6,977 (99.4) 22,956 (99.5) 7,763 (98.8) 25,220 (98.9) Low 28 (0.4) 92 (0.4) 1.03 (0.67, 1.58) 86 (1.1) 248 (1.0) 1.11 (0.87, 1.43) High 13 (0.2) 33 (0.1) 1.28 (0.67, 2.47) 6 (0.1) 28 (0.1) 0.67 (0.28, 1.63) T ichlo oe hylene Unexposed 6,958 (99.1) 22,866 (99.1) 7,396 (94.2) 24,073 (94.4) Low 38 (0.5) 148 (0.6) 0.85 (0.59, 1.23) 213 (2.7) 639 (2.5) 1.12 (0.96, 1.32) High 22 (0.3) 67 (0.3) 1.06 (0.65, 1.72) 246 (3.1) 784 (3.1) 1.05 (0.91, 1.22) 1,1,1, ichlo oe hane Unexposed 6,937 (98.8) 22,815 (98.8) 7,017 (89.3) 22,937 (90.0) Low 45 (0.6) 162 (0.7) 0.95 (0.68, 1.33) 380 (4.8) 1,147 (4.5) 1.10 (0.98, 1.25) High 36 (0.5) 104 (0.5) 1.14 (0.77, 1.67) 458 (5.8) 1,412 (5.5) 1.07 (0.95, 1.19) No e: CI, con idence in e al; n, numbe ; OR, odds a io. a The exposed g oup was di ided in o wo g oups, based on he 75 h pe cen ile o mo he s who we e exposed and on he 90 h pe cen ile o a he s who we e exposed. b ORs and he co esponding 95% CIs we e de i ed based on logis ic eg ession analysis condi ional on yea and coun y o child’s bi h. The e e ence ca ego y (unexposed) consis s o he pa en s who we e no occupa ionally exposed o he sol en o in e es . En i onmen al Heal h Pe spec i es 067023-5 Eu opean Commission (Eu opean Commission DG ENV 2000). Exposu e o EDCs has been sugges ed o con ibu e o he de el- opmen o TDS h ough in e e ence wi h ho mone syn hesis, sec e ion, and signaling (Cook e al. 2011;Gio dano e al. 2010; Sha pe 2006). In wo s udies, e al exposu e o oluene and s y- ene was epo ed o be associa ed wi h a educed syn hesis and sec e ion o es os e one in he e al es es and dec eased weigh o male ep oduc i e o gans (Ohyama e al. 2007;Tsukaha a e al. 2009). One s udy epo ed a posi i e associa ion be ween a he s wo king as pain e s and occupa ionally exposed o o ganic sol- en s wi hin he 3 mon hs be o e p egnancy and bi h de ec isk in child en (Hooi eld e al. 2006). Ou esul s on o e all expo- su e o sol en s do no suppo he associa ion obse ed in his s udy, bu he wo s udies a e difficul o compa e, since in he s udy by Hooi eld e al. (2006), he numbe o child en wi h bi h de ec s was e y small (n= 48) and in ol ed a wide ange o congeni al mal o ma ions and diso de s. The mechanisms unde - lying a po en ial effec o pa e nal exposu e o sol en s on TGCT in offsp ing is unclea , bu a plausible pa hway has been hypo he- sized; no ably an effec on spe m DNA, p oducing mu a ions o ch omosomal abno mali ies (Olshan e al. 1991). Al hough ex- pe imen al s udies ha e suppo ed a po en ial effec o sol en ex- posu e on he es es, none ha e in es iga ed he co esponding effec o pa e nal exposu e on he es es in he offsp ing (Lamb and Hen z 2006;NTP 2002;Ve ma and Rana 2009;Xu e al. 2004). The e o e, he associa ion ound o pa e nal exposu e o pe chlo oe hylene in ou s udy equi es u he suppo om mechanis ic s udies. Howe e , i should be no ed ha nei he he p e alence o exposu e no he exposu e le els among he exposed o pe chlo oe hylene we e subs an ially highe in Finland compa ed o No way o Sweden. While he diffe ences in ORs obse ed ac oss he coun ies migh be explained by some diffe ence in occupa ional p ac ices, we canno exclude he possibili y o an obse a ion by chance due o mul iple compa isons. The main ad an age o he NORD-TEST s udy is he linkage be ween he popula ion-based egis ies om he No dic coun- ies ep esen ing an excep ional se ing o design a e ospec- i e s udy, wi h access o p ospec i ely egis e ed exposu e in o ma ion. The ex ac ion o he cases diagnosed du ing he las 30 yea s in h ee coun ies p o ides a la ge sample size o analyze pa en al occupa ional exposu e o o ganic sol en s, e en hough he p e alence o exposu e emained limi ed o ce ain sol en s and exposu e ca ego ies. Ano he s eng h is he use o a coun y-specific JEM c ea ed o No dic coun ies (NOCCA-JEM) ha encompasses he exposu e es ima es o single subs ances ha a e likely o ha e hei own mechanisms Table 3. Pa en al occupa ional exposu e o di e en sol en s be o e childbi h and TGCT in he o sp ing wi h unexposed o any o he sol en s as e e ence ca ego y. Exposu e Ma e nal exposu e Pa e nal exposu e Cases n(%) Con ols n(%) OR (95% CI) a Cases n(%) Con ols n(%) OR (95% CI) a A oma ic hyd oca bon sol en s (ARHC) Any o ARHC sol en s b Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 104 (1.5) 287 (1.2) 1.23 (0.97, 1.55) 612 (7.8) 1,951 (7.7) 1.03 (0.93, 1.13) Exposed only o o he s 89 (1.3) 275 (1.2) 1.09 (0.86, 1.40) 545 (6.9) 1,666 (6.5) 1.10 (1.00, 1.22) Benzene Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 86 (1.2) 241 (1.0) 1.18 (0.91, 1.53) 540 (6.9) 1,686 (6.6) 1.04 (0.94, 1.15) Exposed only o o he s 107 (1.5) 321 (1.4) 1.15 (0.91, 1.44) 617 (7.9) 1,931 (7.6) 1.08 (0.98, 1.19) Toluene Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 55 (0.8) 160 (0.7) 1.22 (0.89, 1.69) 581 (7.4) 1,852 (7.3) 1.02 (0.93, 1.13) Exposed only o o he s 138 (2.0) 402 (1.7) 1.14 (0.93, 1.39) 576 (7.3) 1,765 (6.9) 1.10 (1.00, 1.22) Chlo ina ed hyd oca bon sol en s (CHC) Any o CHC sol en s c Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 108 (1.5) 344 (1.5) 1.06 (0.85, 1.32) 967 (12.3) 2,972 (11.7) 1.07 (0.99, 1.16) Exposed only o o he s 85 (1.2) 218 (0.9) 1.33 (1.03, 1.73) 190 (2.4) 645 (2.5) 1.00 (0.85, 1.18) Me hylene chlo ide Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 54 (0.8) 142 (0.6) 1.34 (0.97, 1.86) 533 (6.8) 1,648 (6.5) 1.06 (0.95, 1.17) Exposed only o o he s 139 (2.0) 420 (1.8) 1.10 (0.90, 1.35) 624 (7.9) 1,969 (7.7) 1.07 (0.97, 1.17) Pe chlo oe hylene Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 41 (0.6) 125 (0.5) 1.10 (0.77, 1.58) 92 (1.2) 276 (1.1) 1.08 (0.84, 1.37) Exposed only o o he s 152 (2.2) 437 (1.9) 1.18 (0.97, 1.43) 1,065 (13.6) 3,341 (13.1) 1.06 (0.98, 1.14) T ichlo oe hylene Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 60 (0.9) 215 (0.9) 0.93 (0.69, 1.24) 459 (5.8) 1,423 (5.6) 1.09 (0.98, 1.22) Exposed only o o he s 133 (1.9) 347 (1.5) 1.31 (1.06, 1.61) 698 (8.9) 2,194 (8.6) 1.04 (0.95, 1.14) 1,1,1, ichlo oe hane Unexposed 6,825 (97.2) 22,519 (97.6) 6,698 (85.3) 21,879 (85.8) Exposed 81 (1.2) 266 (1.2) 1.03 (0.80, 1.33) 838 (10.7) 2,559 (10.0) 1.08 (1.00, 1.18) Exposed only o o he s 112 (1.6) 296 (1.3) 1.28 (1.02, 1.61) 319 (4.1) 1,058 (4.1) 1.01 (0.88, 1.14) No e: CI, con idence in e al; n, numbe ; OR, odds a io; TGCT, es icula ge m cell umo isk; Exposed, exposed o he sol en s o in e es bu can also be exposed o o he sol en s; Exposed only o o he s, exposed o any o he sol en s bu he sol en o in e es ; Unexposed, unexposed o any o he indi idual o g oups o sol en s. a ORs and he co esponding 95% CIs we e de i ed based on logis ic eg ession analysis condi ional on yea and coun y o child’s bi h, he e e ence ca ego y (unexposed) being he pa en s ha we e occupa ionally unexposed o any o he indi idual o g oups o sol en s. b Benzene and/o oluene. c Me hylene chlo ide, pe chlo oe hylene, ichlo oe hylene, and/o 1,1,1- ichlo oe hane. En i onmen al Heal h Pe spec i es 067023-6 o ac ion in con as o g oups o subs ances (Pukkala e al. 2009). One o he ad an ages o using a JEM is ha i assigns occupa ional exposu es in a sys ema ic and objec i e way, he e o e a oiding diffe en ial misclassifica ion be ween cases and con ols; his is pa icula ly ue o he NORD-TEST s udy collec ing occupa ional da a ia censuses, which we e ca ied ou be o e he diagnosis o TGCT. One o he limi a ions o he NORD-TEST s udy is he lack o in o ma ion on he ime and du a ion o each occupa ion and he e- o e he inabili y o p ecisely a ge he c i ical ime window o ex- posu e and calcula e p econcep ion cumula i e exposu es. Because o he s udy design, we had o assume ha he job a he ime o cen- sus eflec ed he exposu e a he ime o concep ion and du ing in a- u e ine de elopmen , e en hough he census migh be se e al yea s be o e he child’s bi h. Ne e heless, he sensi i i y analyses, including pa en s o whom he census was a ailable wi hin he yea be o e o he yea o he child’s bi h o pa en s who held he same occupa ion in censusp io anda e he child’s bi h, yielded s onge associa ions. In addi ion, NOCCA-JEM assumes homogenei y o exposu e wi hin jobs. The ue a iance o sol en exposu e wi hin jobs ep esen ed by equal le els o exposu e migh ha e dilu ed he es ima ed s eng h o a possible associa ion. Ne e heless, ou app oach allowed us o compa e he pa en s wi h a highe isk o sol- en exposu e wi h hose wi h alowe isk o sol en exposu e. A ecen hypo hesis sugges s ha TGCT could a ise om a combined effec o p ena al and ea ly- and la e -li e exposu es (McGlynn and T abe 2012). The ge m cell neoplasia in si u (Be ney e al. 2016) assumed o be gene a ed in u e o by p ima y p ena al exposu e does no become in asi e un il pube y ollowing ac i a ion o he pi ui a y–gonadal axis (Rajpe -De 2006; Skakkebaek e al. 1987,2001;Sonne e al. 2009). The design o NORD-TEST s udy did no allow collec ion o indi idual li elong exposu es. Howe e , some o he subjec s had a known occupa- ional his o y. Wi h his in o ma ion, we checked whe he subjec s exposed o sol en s du ing he p ena al pe iod we e likely o be exposed la e in li e in hei occupa ions, bu ound his was no he case (as measu ed by he kappa s a is ics o ag eemen o ca ego i- cal a iables; j≤0:1). Conclusions This is he fi s la ge nes ed case–con ol s udy in es iga ing asso- cia ions be ween ma e nal and pa e nal exposu e o se e al indi- idual o ganic sol en s and ypes o hem, and he isk o TGCT in hei offsp ing. We ound no e idence o any associa ion wi h pa e nal occupa ional exposu e o sol en s, excep an ele a ed isk obse ed wi h pe chlo oe hylene in Finland only. Fo ma e nal ex- posu e, a mode a ely ele a ed isk was obse ed among sons o mo he s occupa ionally exposed o ARHC, pa icula ly oluene, when analyses we e es ic ed o mo he s o whom he occupa ion was known in he yea o o he yea be o e childbi h. Fu he s udies a e needed, including in es iga ions o molecula mecha- nisms, o be e unde s and he obse ed associa ion be ween ma e nal exposu es o oluene and TGCT isk in hei offsp ing. Acknowledgmen s This wo k was suppo ed by public unding om he Ly ic G an INCa-DGOS-4664 (Ins i u e o Cance Resea ch, F ance), he In e na ional Agency o Resea ch on Cance (IARC), and he Cancé opôle Lyon Au e gne Rhône-Alpes (CLARA). We would Table 4. Sensi i i y analysis: es icula ge m cell umo isk when pa en s ha e been exposed o sol en s wi hin he yea be o e childbi h. Exposu e Ma e nal exposu e Pa e nal exposu e Cases n(%) Con ols n(%) OR (95% CI) a Cases n(%) Con ols n(%) OR (95% CI) a A oma ic hyd oca bon sol en s (ARHC) Any o ARHC sol en s b Unexposed 4,007 (98.6) 13,567 (99.0) 4,034 (91.9) 13,482 (91.8) Exposed 56 (1.4) 138 (1.0) 1.53 (1.08, 2.17) 354 (8.1) 1,198 (8.2) 0.98 (0.85, 1.13) Benzene Unexposed 4,022 (99.0) 13,589 (99.2) 4,074 (92.8) 13,638 (92.9) Exposed 41 (1.0) 116 (0.8) 1.24 (0.83, 1.85) 314 (7.2) 1,042 (7.1) 0.99 (0.86, 1.15) Toluene Unexposed 4,037 (99.4) 13,642 (99.5) 4,047 (92.2) 13,542 (92.2) Exposed 26 (0.6) 63 (0.5) 1.67 (1.02, 2.73) 341 (7.8) 1,138 (7.8) 0.99 (0.86, 1.14) Chlo ina ed hyd oca bon sol en s (CHC) Any o CHC sol en s c Unexposed 4,016 (98.8) 13,568 (99.0) 3818 (87.0) 12,904 (87.9) Exposed 47 (1.2) 137 (1.0) 1.17 (0.83, 1.66) 570 (13.0) 1,776 (12.1) 1.07 (0.95, 1.20) Me hylene chlo ide Unexposed 4,037 (99.4) 13,646 (99.6) 4,064 (92.6) 13,636 (92.9) Exposed 26 (0.6) 59 (0.4) 1.54 (0.94, 2.51) 324 (7.4) 1,044 (7.1) 1.03 (0.89, 1.19) Pe chlo oe hylene Unexposed 4,043 (99.5) 13,658 (99.7) 4,338 (98.9) 14,532 (99.0) Exposed 20 (0.5) 47 (0.3) 1.43 (0.83, 2.45) 50 (1.1) 148 (1.0) 1.02 (0.71, 1.46) T ichlo oe hylene Unexposed 4,038 (99.4) 13,618 (99.4) 4,124 (94.0) 13,873 (94.5) Exposed 25 (0.6) 87 (0.6) 0.98 (0.62, 1.54) 264 (6.0) 807 (5.5) 1.10 (0.94, 1.28) 1,1,1, ichlo oe hane Unexposed 4,022 (99.0) 13,590 (99.2) 3,886 (88.6) 13,145 (89.5) Exposed 41 (1.0) 115 (0.8) 1.22 (0.84, 1.78) 502 (11.4) 1,535 (10.5) 1.09 (0.96, 1.22) Any sol en d Unexposed 5,076 (97.6) 16,899 (97.8) 4,732 (86.0) 15,975 (87.1) Exposed 127 (2.4) 378 (2.2) 1.15 (0.94, 1.42) 773 (14.0) 2,376 (12.9) 1.09 (0.99, 1.19) No e: CI, con idence in e al; n, numbe ; OR, odds a io. a ORs and he co esponding 95% CIs we e de i ed based on logis ic eg ession analysis condi ional on yea and coun y o child’s bi h, he e e ence ca ego y (unexposed) was he pa en s ha we e no occupa ionally exposed o he sol en o in e es . b Benzene and/o oluene. c Me hylene chlo ide, pe chlo oe hylene, ichlo oe hylene, and/o 1,1,1- ichlo oe hane. d A leas one o he ollowing sol en s: benzene, oluene, me hylene chlo ide, pe chlo oe hylene, ichlo oe hylene, 1,1,1- ichlo oe hane. En i onmen al Heal h Pe spec i es 067023-7 like o acknowledge M. S eding-Jessen and A. Mee sohn om he Danish Cance Socie y Resea ch Cen e , K. F emling om he Ins i u e o En i onmen al Medicine, and Ka olinska Ins i u e and V. Luzon om IARC o hei help wi h he da a managemen . The Family Cance Da abase was c ea ed by linking egis e s main ained a S a is ics Sweden and he Swedish Cance Regis y. Da a om he Finnish Cance Regis y was ex ac ed by he pe mission o he esea ch No.THL/1123/5.05.00/2012. We would also like o hank he NOCCA eam o app o ing ou use o he NOCCA-JEM and hei expe ise on he occupa ional exposu e in he No dic coun ies. Re e ences Bé ange R, Le Co ne C, Schüz J, Fe e s B. 2013. Occupa ional and en i onmen- al exposu es associa ed wi h es icula ge m cell umou s: sys ema ic e iew o p ena al and li e-long exposu es. PLoS One 8(10):e77130, PMID: 24155923, h ps://doi.o g/10.1371/jou nal.pone.0077130. Be ney DM, Looijenga LH, Id ees M, Oos e huis JW, Rajpe -De Mey s E, Ulb igh TM, e al. 2016. Ge m cell neoplasia in si u (GCNIS): e olu ion o he cu en nomencla u e o es icula p e-in asi e ge m cell malignancy. His opa hology 69(1):7–10, PMID: 26918959,h ps://doi.o g/10.1111/his.12958. B enne H, Loomis D. 1994. Va ied o ms o bias due o nondi e en ial e o in measu ing exposu e. Epidemiology 5(5):510–517, PMID: 7986865. B ouwe s MM, an Tonge en M, Hi s AA, B e eld RW, Roele eld N. 2009. Occupa ional exposu e o po en ial endoc ine dis up o s: u he de elopmen o a job exposu e ma ix. Occup En i on Med 66(9):607–614, PMID: 19286684, h ps://doi.o g/10.1136/oem.2008.042184. Cook MB, Ak e O, Fo man D, Madigan MP, Richia di L, McGlynn KA. 2009. A sys- ema ic e iew and me a-analysis o pe ina al a iables in ela ion o he isk o es icula cance –expe iences o he mo he . In J Epidemiol 38(6):1532– 1542, h ps://doi.o g/10.1093/ije/dyp287. Cook MB, Ak e O, Fo man D, Madigan MP, Richia di L, McGlynn KA. 2010. A sys- ema ic e iew and me a-analysis o pe ina al a iables in ela ion o he isk o es icula cance –expe iences o he son. In J Epidemiol 39(6):1605–1618, PMID: 20660640,h ps://doi.o g/10.1093/ije/dyq120. Cook MB, T abe B, McGlynn KA. 2011. O ganochlo ine compounds and es icula dysgenesis synd ome: human da a. In J And ol 34(4 P 2):e68–e84; discussion: e84–e85, PMID: 21668838,h ps://doi.o g/10.1111/j.1365-2605.2011.01171.x. Engholm G, Fe lay J, Ch is ensen N, Kejs A, Johannesen T, Khan S, e al. 2016. NORDCAN: Cance Incidence, Mo ali y, P e alence and Su i al in he No dic Coun ies, Ve sion 7.2 (16.12.2015). Copenhagen, Denma k:Associa ion o he No dic Cance Regis ies, Danish Cance Socie y. Eu opean Commission DG ENV. 2000. Towa ds he Es ablishmen o a P io i y Lis o Subs ances o Fu he E alua ion o hei Role in Endoc ine Dis up ion- P epa a ion o a Candida e Lis o Subs ances as a Basis o P io i y Se ing- BKH Repo . Annex 13.h p://ec.eu opa.eu/en i onmen /a chi es/docum/01262_ en.h m#Eu opean Wo kshop on Endoc ine Dis up o s [accessed 15 June 2015]. Fe lay J, Soe joma a am I, E ik M, Dikshi R, Ese S, Ma he s C, e al. 2013. GLOBOCAN 2012 1.0, Cance Incidence and Mo ali y Wo ldwide: IARC Cance Base No. 11. In e na ional Agency o Resea ch on Cance , Lyon, F ance. Gio dano F, Abballe A, De Felip E, di Domenico A, Fe o F, G amma ico P, e al. 2010. Ma e nal exposu es o endoc ine dis up ing chemicals and hypospadias in o sp ing. Bi h De ec s Pa A Clin Mol Te a ol 88(4):241–250, PMID: 20196143,h ps://doi.o g/10.1002/bd a.20657. Ha dell L, Ba el B, Linds öm G, E iksson M, Ca lbe g M. 2006. In u e o exposu e o pe sis en o ganic pollu an s in ela ion o es icula cance isk. In J And ol 29(1):228–234, PMID: 16371110,h ps://doi.o g/10.1111/j.1365-2605. 2005.00622.x. Ha dell L, an Ba el B, Linds öm G, Ca lbe g M, D ei ald AC, Wijks öm H, e al. 2003. Inc eased concen a ions o polychlo ina ed biphenyls, hexachlo oben- zene, and chlo danes in mo he s o men wi h es icula cance . En i on Heal h Pe spec 111(7):930–934, h ps://doi.o g/10.1289/ehp.5816. Hooi eld M, Ha eman W, Roskes K, B e eld R, Bu s yn I, Roele eld N. 2006. Ad e se ep oduc i e ou comes among male pain e s wi h occupa ional expo- su e o o ganic sol en s. Occup En i on Med 63(8):538–544, PMID: 16757511, h ps://doi.o g/10.1136/oem.2005.026013. IARC monog aphs. 2012. Chemical Agen s and Rela ed Occupa ions. Lyon, F ance: IARC, 9–562. IARC monog aphs. 2014. T ichlo oe hylene, Te achlo oe hylene, and Some O he Chlo ina ed Agen s. Lyon, F ance:IARC, 1–512. Ka daun JW, Hayes RB, Po e n LM, B own LM, Hoo e RN. 1991. Tes icula cance in young men and pa en al occupa ional exposu e. Am J Ind Med 20(2):219–227, PMID: 1951369,h ps://doi.o g/10.1002/ajim.4700200208. Kauppinen T, Heikkilä P, Pla o N, Woldbaek T, Len ik K, Hansen J, e al. 2009. Cons uc ion o job-exposu e ma ices o he No dic Occupa ional Cance S udy (NOCCA). Ac a Oncol 48(5):791–800, PMID: 19225948,h ps://doi.o g/10. 1080/02841860902718747. Kauppinen T, Uuksulainen S, Saalo A, Mäkinen I, Pukkala E. 2014. Use o he Finnish In o ma ion Sys em on Occupa ional Exposu e (FINJEM) in epidemio- logic, su eillance, and o he applica ions. Ann Occup Hyg 58(3):380–396, PMID: 24401793,h ps://doi.o g/10.1093/annhyg/me 074. K a z CP, Mai PL, G eene MH. 2010. Familial es icula ge m cell umou s. Bes P ac Res Clin Endoc inol Me ab 24(3):503–513, PMID: 20833340,h ps://doi.o g/ 10.1016/j.beem.2010.01.005. Lamb JC, Hen z KL. 2006. Toxicological e iew o male ep oduc i e e ec s and i- chlo oe hylene exposu e: assessing he ele ance o human male ep oduc i e heal h. Rep od Toxicol 22(4):557–563, PMID: 16938429,h ps://doi.o g/10.1016/j. ep o ox.2006.07.001. Le Co ne C, Fe e s B, Dal on SO, Feych ing M, Pukkala E, Tynes T, e al. 2015. Tes icula ge m cell umou s and pa en al occupa ional exposu e o pes i- cides: a egis e -based case-con ol s udy in he No dic coun ies (NORD- TEST s udy). Occup. En i on. Med 72(11):805–811, h ps://doi.o g/10.1136/ oemed-2015-102860. Le Co ne C, Lo e -Tieulen J, Fo man D, Bé ange R, Flechon A, Fe e s B, e al. 2014. Tes icula cance incidence o ise by 25% by 2025 in Eu ope? Model- based p edic ions in 40 coun ies using popula ion-based egis y da a. Eu J Cance 50(4):831–839, PMID: 24369860,h ps://doi.o g/10.1016/j.ejca.2013.11. 035. McGlynn KA, T abe B. 2012. Adolescen and adul isk ac o s o es icula can- ce . Na Re U ol 9(6):339–349, PMID: 22508459,h ps://doi.o g/10.1038/n u ol. 2012.61. Moch H, Humph ey PA, Ulb igh TM, Reu e VE. 2016. WHO Classi ica ion o Tumou s. Volume 8, Lyons, F ance:IARC, 189. NTP (Na ional Toxicology P og am). 2002. Toxicology and ca cinogenesis s udies o p-ni o oluene (CAS no. 99-99-0) in F344/N a s and B6C3F(1) mice ( eed s udies). Na l Toxicol P og am Tech Rep Se 498:1–277, h ps://doi.o g/10.1038/ n u ol.2012.61. Ohyama K, Sa oh K, Sakamo o Y, Oga a A, Nagai F. 2007. E ec s o p ena al expo- su e o s y ene ime s on geni al o gans and ho mones in male a s. Exp Biol Med (Maywood) 232(2):301–308, PMID: 17259338. Olshan AF, Teschke K, Bai d PA. 1991. Pa e nal occupa ion and congeni al anoma- lies in o sp ing. Am J Ind Med 20(4):447–475, PMID: 1785611. Pukkala E, Ma insen JI, Lynge E, Gunna sdo i HK, Spa én P, T ygg ado i L, e al. 2009. Occupa ion and cance - ollow-up o 15 million people in i e No dic coun ies. Ac a Oncol 48(5):646–790. Rajpe -De ME. 2006. De elopmen al model o he pa hogenesis o es icula ca cinoma in si u: gene ic and en i onmen al aspec s. Hum Rep od. Upda e 12 (3):303–323. Schmiedel S, Schüz J, Skakkebaek NE, Johansen C. 2010. Tes icula ge m cell cance incidence in an immig a ion pe spec i e, Denma k, 1978 o 2003. J U ol 183(4):1378–1382, PMID: 20171682,h ps://doi.o g/10.1016/j.ju o.2009.12.058. Sha pe RM. 2006. Pa hways o endoc ine dis up ion du ing male sexual di e en ia- ion and masculiniza ion. Bes P ac Res Clin Endoc inol Me ab 20(1):91–110, PMID: 16522522,h ps://doi.o g/10.1016/j.beem.2005.09.005. Skakkebaek NE, Rajpe -De Mey s E, Buck Louis GM, Toppa i J, Ande sson AM, Eisenbe g ML, e al. 2016. Male ep oduc i e diso de s and e ili y ends: in luences o en i onmen and gene ic suscep ibili y. Physiol Re 96(1):55–97, PMID: 26582516,h ps://doi.o g/10.1152/phys e .00017.2015. Skakkebaek NE, Rajpe -De Mey s E, Main KM. 2001. Tes icula dysgenesis syn- d ome: an inc easingly common de elopmen al diso de wi h en i onmen al aspec s. Hum Rep od 16:972–978, h ps://doi.o g/10.1093/hum ep/16.5.972. Skakkebaek NE, Be helsen JG, Giwe cman A, Mülle J. 1987. Ca cinoma-in-si u o he es is: possible o igin om gonocy es and p ecu so o all ypes o ge m cell umou s excep spe ma ocy oma. In J And ol 10(1):19–28, PMID: 3034791. Sonne SB, Alms up K, Dalgaa d M, Juncke AS, Edsga d D, Ruban L, e al. 2009. Analysis o gene exp ession p o iles o mic odissec ed cell popula ions indi- ca es ha es icula ca cinoma in si u is an a es ed gonocy e. Cance Res 69(12):5241–5250, PMID: 19491264,h ps://doi.o g/10.1158/0008-5472.CAN-08- 4554. Tsukaha a S, Nakajima D, Ku oda Y, Hojo R, Kageyama S, Fujimaki H. 2009. E ec s o ma e nal oluene exposu e on es os e one le els in e al a s. Toxicol Le 185(2):79–84, PMID: 19110042,h ps://doi.o g/10.1016/j. oxle .2008.12.001. Vandenbe g LN, Colbo n T, Hayes TB, Heindel JJ, Jacobs DR J , Lee DH, e al. 2016. Ho mones and endoc ine-dis up ing chemicals: low-dose e ec s and nonmono onic dose esponses. Endoc Re 33(3):378–455, PMID: 22419778, h ps://doi.o g/10.1210/e .2011-1050. Ve ma Y, Rana SV. 2009. Endoc inal oxici y o indus ial sol en s–a mini e iew. Indian J Exp Biol 47(7):537–549, PMID: 19761037. En i onmen al Heal h Pe spec i es 067023-8 WHO (Wo ld Heal h O ganiza ion). 1955. In e na ional Classi ica ion o Diseases, 7 h Edi ion. h p://www.wol bane.com/icd/icd7h.h m [accessed 16 June 2017]. WHO. 1968. In e na ional Classi ica ion o Diseases, 8 h Edi ion. h p://www. wol bane.com/icd/icd8h.h m [accessed 16 June 2017]. WHO. 1978. In e na ional Classi ica ion o Diseases, 9 h Edi ion. h p://www. wol bane.com/icd/icd9h.h m [accessed 16 June 2017]. WHO. 2000. In e na ional Classi ica ion o Diseases o Oncology, 3 d Edi ion. h p://www. who.in /classi ica ions/icd/adap a ions/oncology/en/ [accessed 16 June 2017]. Xu H, Tanphaichi N, Fo ke PG, Anup iwan A, Wee acha yanukul W, Vincen R, e al. 2004. Exposu e o ichlo oe hylene and i s me aboli es causes impai - men o spe m e ilizing abili y in mice. Toxicol Sci 82(2):590–597, PMID: 15375293,h ps://doi.o g/10.1093/ oxsci/k h277. En i onmen al Heal h Pe spec i es 067023-9