Heigh , selec ed gene ic ma ke s and
p os a e cance isk: esul s om he
PRACTICAL conso ium
A i aya Lopha ananon
1,2
, Sa ah S ewa -B own
1
, Zso ia Ko e-Ja ai
3
, Ali Amin Al Olama
4
,
Sa a Benlloch Ga cia
4
, Da id E Neal
5,6
, F eddie C Hamdy
7
, Jenny L Dono an
8
, G aham G Giles
9,10
,
Liesel M Fi zge ald
9
, Melissa C Sou hey
11
, Paul Pha oah
12
, No a Pashayan
12,13
, Hen ik G onbe g
14
,
F ed ik Wiklund
14
, Ma kus Aly
14,15
, Jane L S an o d
16,17
, He mann B enne
18,19
, Aida K Die enbach
20,21
,
Volke A nd
20
, Jong Y Pa k
21
, Hui-Yi Lin
22
, Thomas Selle s
21
, Cha da Sla o
23
, Radka Kane a
24
,
Vanio Mi e
24
, Jyo sna Ba a
25
, Amanda Spu dle
26
, Judi h A Clemen s
25
, APCB BioResou ce
60
,
The PRACTICAL conso ium
27,60
, Douglas Eas on
4
, Rosalind A Eeles
3,27
and Kenne h Mui *
,1,2
Backg ound: E idence on heigh and p os a e cance isk is mixed, howe e , ecen s udies wi h la ge da a se s suppo a
possible ole o i s associa ion wi h he isk o agg essi e p os a e cance .
Me hods: We analysed da a om he PRACTICAL conso ium consis ing o 6207 p os a e cance cases and 6016 con ols and a
subse o high g ade cases (2480 cases). We explo ed heigh , polymo phisms in genes ela ed o g ow h p ocesses as main e ec s
and hei possible in e ac ions.
Resul s: The esul s sugges ha heigh is associa ed wi h high-g ade p os a e cance isk. Men wi h heigh 4180 cm a e a a 22%
inc eased isk as compa ed o men wi h heigh o173 cm (OR 1.22, 95% CI 1.01–1.48). Gene ic a ian s in he g ow h pa hway gene
showed an associa ion wi h p os a e cance isk. The agg ega e sco es o he selec ed a ian s iden i ied a signi ican ly inc eased
isk o o e all p os a e cance and high-g ade p os a e cance by 13% and 15%, espec i ely, in he highes sco e g oup as
compa ed o lowes sco e g oup.
Conclusions: The e was no e idence o gene-en i onmen in e ac ion be ween heigh and he selec ed candida e SNPs. Ou
indings sugges a ole o heigh in high-g ade p os a e cance . The e ec o gene ic a ian s in he genes ela ed o g ow h is
seen in all cases and high-g ade p os a e cance . The e is no in e ac ion be ween hese wo exposu es.
P os a e cance is he second mos common cance in men
wo ldwide. App oxima ely 1.1 million men we e diagnosed wi h
p os a e cance in 2012 and almos 70% o he cases occu in mo e
de eloped egions (IARC, 2014). The es ablished isk ac o s
include a e age, e hnici y, amily his o y, and o e 100 common
gene ic a ian s. The e a e howe e o he isk ac o s wi h less
conclusi e e idence including heigh (Key e al, 1997; Hayes e al,
1999; Villeneu e e al, 1999; Hsing e al, 2000; No ish e al, 2000;
S a in e al, 2000). Heigh is a pheno ypic ai de e mined
by a combina ion o gene ics and en i onmen al ac o s. The
ela ionship be ween heigh and p os a e cance isk has been
p oposed o ac h ough possible ac o s including p e-adul
nu i ional s a us, and ogen and insulin-like g ow h ac o -I
(IGF-I; Gio annucci e al, 1997; Calle, 2000; Wille , 2000;
F eeman e al, 2001; Eme ging Risk Fac o s Collabo a ion, 2012;
T a is e al, 2016).
Heigh pe se is no a cause o cance bu i is a ma ke o o he
exposu es. I has also been sugges ed ha alle s a u e may
*Co espondence: P o esso K Mui ; E-mail: kenne h.mui @manches e .ac.uk
60
Addi ional membe s om he P os a e Cance Associa ion G oup o In es iga e Cance Associa ed Al e a ions in he Genome (PRACTICAL)
conso ium can be ound abo e he e e ence lis .
Recei ed 11 Ap il 2016; e ised 7 June 2017; accep ed 23 June 2017;
published online 1 Augus 2017
The Au ho (s) named abo e
FULL PAPER
Keywo ds: heigh ; SNPs; gene and en i onmen in e ac ion; p os a e cance
B i ish Jou nal o Cance (2017) 117, 734–743 | doi: 10.1038/bjc.2017.231
734 Published by Sp inge Na u e on behal o Cance Resea ch UK.
indica e inc eased isk o a numbe o cance s. The mos consis en
e idence has been ound in ela ion o b eas cance (Wille , 2000;
Gunnell e al, 2001).
In 2008, indings om a la ge nes ed case–con ol s udy
(P o ecT) and me a-analysis (58 s udies) sugges ed a posi i e
associa ion o heigh wi h high-g ade p os a e cance (OR: 1.23;
95% CI: 1.06–1.43; Zuccolo e al, 2008). In his a icle, we p esen
esul s om he in e na ional collabo a ion, he P os a e Cance
Associa ion G oup o In es iga e Cance Associa ed Al e a ions in
he Genome conso ium (PRACTCAL; h p://p ac ical.ccge.-
medschl.cam.ac.uk/). The aim was o explo e he e ec s o heigh
on p os a e cance isk. We we e also in e es ed o see i selec ed
candida e SNPs ela ed o heigh we e associa ed wi h p os a e
cance isk. Finally, we explo ed possible in e ac ions be ween he
selec ed SNPs and heigh .
MATERIALS AND METHODS
PRACTICAL conso ium. The PRACTICAL conso ium
consis s o 78 s udy g oups a ound he wo ld. The conso ium
was es ablished in Sep embe 2008. The co-o dina ion o
PRACTICAL is unded by Cance Resea ch UK and da a
ha e been con ibu ed o he Collabo a i e Oncology Gene-
en i onmen S udy (COGS), a p ojec unded by he Eu opean
Commission and 7 h F amewo k P og amme and he NIH g an .
Each s udy wi h ele an da a con ibu ed an epidemiological da a
se and blood samples. Da a on epidemiological ac o s o each
s udy we e p o ided in acco dance wi h an assembled da a
dic iona y. We pe o med quali y con ol checks o each s udy
be o e me ging he da a in o one combined da abase. The majo i y
o he samples a e o Eu opean ances y (95%). Since we
in es iga ed heigh as ou main exposu e we only analysed s udies
ha con ained subjec s wi h Eu opean ances y in o de o
minimise a ia ion o heigh po en ially in luenced by di e en
e hnic g oups.
Blood de i ed DNA samples we e geno yped o 211,155 SNPs
on a cus om Illumina a ay (iCOGS) in 25 074 p os a e cance
cases and 24 272 con ols. De ails o geno yping and quali y con ol
analysis can be ound in p e ious publica ion (Eeles e al, 2013).
Analysis o heigh exposu e. Du ing he QC p ocess, any
subjec s wi h ou lie alues we e checked di ec ly wi h he
indi idual s udy g oup and subsequen ly ei he co ec ed o
excluded. Heigh da a we e a ailable in 10 ou o 15 s udies ha
submi ed da a on epidemiological ac o s. The inclusion
c i e ion o his pa icula analysis is subjec s wi h Eu opean
ances y. The o al numbe o p os a e cance cases and con ols
we e 6207 cases and 6016 con ols. The lis o s udies included in
he heigh exposu e analyses a e lis ed in Supplemen a y Table 1.
Me a-analysis was pe o med using Me a-Analys so wa e
(Wallace e al, 2009). We pe o med analysis in all PCA cases
and high g ade cases as compa ed o con ols. The la e is
de ined by Gleason g ade X7. Ou o 6207 cases, 2480 cases a e
high g ade cases. Me a-analysis was ca ied ou in 9 s udies as
one o he s udies had no con ols. Heigh was i ed as a
con inuous a iable and s udy he e ogenei y was explo ed. We
also pe o med analysis whe eby heigh was ca ego ised in o
qua iles using con ol heigh alues o de e mine he anges.
Resul s sugges s udy homogenei y hence esul s om a ixed
e ec model a e epo ed. Pooled analysis was also pe o med.
Tes s o end we e ca ied ou o assess possible dose- esponse
ela ionships. Analyses we e pe o med using IBM SPSS
S a is ics e sion 20.0. All analyses we e adjus ed o age, amily
his o y o p os a e cance , and s udy si es. As he da a we e
de i ed om a ious s udies wi h di e ing sample sizes, he
analyses we e he e o e adjus ed o s udy si e o a oid possible
con ounding e ec s.
SNPs analyses. We explo ed he e ec s o candida e SNPs ela ed
o g ow h ac o s on p os a e cance isk. We iden i ied 168
candida e SNPs in IGF-I,GH-1,SHOX,FMR1,GHITM, and
GHRHR genes ela ed o human g ow h based on e idence om
he li e a u e and hese SNPs we e geno yped wi hin a cus om
Illumina a ay (iCOGS). The ull lis o 168 candida e SNPs and
associa ed ela i e isk es ima es a e shown in Supplemen a y
Table 2. To e alua e e ec sizes o hese SNPs, we c ea ed a da a se
consis ing o indi idual subjec s whose IDs appea ed in bo h he
geno ype and epidemiological da a se s by ma ching he IDs
be ween he wo se s. We included only Caucasian subjec s. This
esul ed in 13 123 con ols and 9424 cases. PLINK so wa e was
used o explo e mino allele equency (MAF) and Ha dy-
Weinbe g equilib ium (HWE; Pu cell e al, 2007). MAF anges
we e om 0.017 o 0.496. Ou o 155 SNPs, 168 SNPs me HWE
(P40.05). STATA ( e sion14) was used o ob ain isk es ima es
and R-squa e (LDsco e; Cheng e al, 2006). To quan i y isk, he
log-addi i e model was used by including a single a iable coded as
0, 1, o 2 based addi i ely on he numbe o mino alleles. Mul iple
logis ic eg ession analyses we e ca ied ou o ob ain he odd
a ios o all 168 SNPs. Va iables included in he model we e age,
amily his o y o p os a e cance , s udy si es, p incipal componen s
o Eu opean ances y, and SNPs. Twel e SNPs showed signi ican
associa ions (P- alue o0.05). We hen compu ed he R-squa e
alue o hese 12 SNPs (Table 1). The esul s showed ha
hese SNPs ell in o 4 egions. SNPs we e excluded i
2
alue was
40.8 among hem and we kep he mos in o ma i e SNP based
on associa ion and P- alue in each egion. R-squa ed alues o
hese 8 SNPs we e less han 0.26. A e his p ocess, eigh SNPs
we e selec ed o u he analysis. Among hese signi ican SNPs,
only wo yielded odds a ios (ORs) abo e 1.15.
Gene and en i onmen in e ac ion analyses. We ca ied ou gene
and en i onmen (GE) analyses in 6207 cases and 6016 con ols.
These a e subjec s wi h da a on geno ype and heigh . We applied
wo ype o analyses based on he e ec sizes o he SNP analyses.
1. Fo he 8 SNPs ha we e signi ican ly associa ed wi h p os a e
cance isk, indi idual s anda dised gene ic sco e was com-
pu ed. Fi s , we mul iplied coe icien o each SNP de i ed
om mul iple logis ic eg ession (as explained abo e) wi h
indi idual isk allele o ha pa icula SNPs. To ob ain o al
gene ic isk sco e, we summed esul s om each SNP. To
compu e s anda dised sco e, he o al sco e was di ided wi h s.d.
alue om con ol g oup. Fi s , gene ic isk sco es we e
analysed as o main e ec by compa ing subjec s in he second
and hi d e ile o he e e en ca ego y. Fo GE analysis, bo h
heigh and gene ic isk sco e we e hen compa ed as bina y
a iables. We classi ied bo h a iables in o e iles wi h lowes
e ile as e e ence g oup and highes e iles as exposed g oup.
We applied empi ical-Bayes (EB) me hod p oposed by
Mukhe jee e al (Mukhe jee e al, 2008). Resul s o all PCA
and high g ade cases a e p esen ed.
2. We also employed he gene al mul i ac o dimensionali y
educ ion (GMDR) me hod (Chen e al, 2011). Fo his we
included he op 2 SNPs wi h e ec sizes 41.15 and i ed hese
in o he model a he same ime. This p ocedu e is no possible
in he con en ional GE me hods. Heigh was i ed as a bina y
a iable. We included subjec s wi h heigh in he e e ence
(lowes e ile) and op hi d e ile. Analyses we e ca ied ou
o all PCA and high g ade cases. Age and amily his o y o PCA
we e i ed as co a ia es.
Heigh , selec ed gene ic ma ke s and p os a e cance isk BRITISH JOURNAL OF CANCER
www.bjcance .com | DOI:10.1038/bjc.2017.231 735
RESULTS
Subjec cha ac e is ics a e displayed in Table 2. Family his o y o
p os a e cance is associa ed wi h p os a e cance isk. Subjec s
wi h a posi i e amily his o y o p os a e cance had a 12% inc ease
in p os a e cance isk. Mean heigh o cases and con ols was
176.3 and 176.8 cms, espec i ely. The S uden ’s - es sugges s a
signi ican di e ence in he means be ween he wo g oups
(P- alue o0.05). Resul s om a me a-analysis o heigh a e
p esen ed in Figu es 1 and 2. ORs we e adjus ed o age, amily
his o y o p os a e cance , and s udy si e. In all cases and high
g ade cases, poin isk es ima es o each s udy a e e y simila and
a e close o 1. None o he es ima ed ela i e isks is s a is ically
signi ican . The he e ogenei y P- alue o 0.467 in all cases and
0.634 in high g ade cases sugges s ha s udies a e homogenous.
ORs o ixed e ec model in all cases and high g ade cases a e 1.002
(95% CI 0.996–1.009) and 1.003 (95% CI 0.996–1.011) espec i ely.
Resul s om pooled analysis yielded simila isk es ima es wi h OR
1.004, 95% CI 0.996–1.012 in all cases and OR 1.007, 95% CI 0.999–
1.015 in high g ade cases. We also analysed heigh as a ca ego ical
a iable. Resul s a e p esen ed in Table 3. Resul s also sugges no o e all
associa ion be ween heigh and p os a e cance isk compa ing all cases
wi h con ols. In he high-g ade case g oup, howe e , signi ican esul s
we e obse ed in he ou h qua ile as compa ed o he i s qua ile
(OR 1.22, 95% CI 1.014–1.477).
Table 4 shows he ORs o candida e SNPs wi h s a is ically signi ican
esul s. ORs ange om 0.90 o 1.32 wi h P- alue om 10
2
o 10
3
.
OneSNPsin heIGF-Igenehad hehighes ORs(1.32).
Table 5 shows he ORs o gene ic isk sco es and p os a e cance
isk. A signi ican esul was obse ed in he hi d e ile as compa ed
o e e ence e ile (OR 1.13 wi h 95% CI 1.03–1.23) when all p os a e
cance cases we e included. The P- alue o end is also s a is ically
signi ican . In he high g ade cases, simila esul s we e obse ed.
The e is also a end o inc easing isk wi h inc easing gene ic isk
sco es in all p os a e cance cases and in high g ade cases.
The in e ac ion esul s be ween heigh and gene ic isk sco es
sugges ha he e is no GE in e ac ion be ween heigh and gene ic
isk sco e (Table 6) ega dless o ype o cases.
Resul s o he GE analyses by GMDR me hod a e depic ed in
Table 7. We i ed 2 SNPs wi h e ec sizes 41.15 in o he model
and adjus ed o co a ia es (age and amily his o y o PCA). None
o he models yield signi ican ORs ega dless o case ype. This is
con i med by c oss- alida ion consis ency. Bo h all and high g ade
cases, he ex ended models show consis ency ac oss es ing se s.
DISCUSSION
This s udy in es iga ed he e ec o heigh and i s possible
in e ac ion wi h selec ed SNPs om he PRACTICAL conso ium
in 6207 cases and 6016 con ols. The conso ium is an
in e na ional collabo a ion on PCA and i has had no able
successes o example in iden i ying 100 new gene ic loci (Eeles
e al, 2008, 2009, 2013; Al Olama e al, 2009, 2012, 2014). These
loci con e small o medium isks wi h highly signi ican P- alues
o p10
7
(GWAS signi icance).
The e a e, howe e , many polymo phisms wi h es ima ed isks
less s a is ically signi ican which could s ill play an impo an ole,
pa icula ly in he p esence o en i onmen al exposu e. We
he e o e c ea ed a da a se (subjec s wi h epidemiological da a
and geno ype da a) which allowed us o in es iga e such a
hypo hesis.
Ou o he 6207 cases, 2480 cases (40%) a e high g ade cases
de ined by Gleason g ade X7. One o he limi a ions o de ining
high g ade cases is ha we did no ha e da a on Gleason g ade
b eakdown hence we ha e o use combined sco e da a o 7 a he
han (4 þ3o 3þ4). Age and amily his o y o PCA a e con i med
isk ac o s in ou s udy (Table 2). We in es iga ed heigh in 3
ways. Fi s , we explo ed heigh pheno ype as a main exposu e.
Second, we in es iga ed gene ic p o ile (candida e SNPs) ela ed o
heigh , and hi d, we de e mined i he e a e any po en ial
in e ac ions be ween he selec ed SNPs and heigh . SNPs we e
deemed ‘ ela ed o heigh ’ because hey a e ound in candida e
genes o heigh bu hey ha e no necessa ily been iden i ied in
GWAS as unde lying he a iabili y o he heigh pheno ype. We
p esen esul s o all PCA cases and high g ade cases as compa ed
o con ols. Al hough mean heigh alues we e e y simila
be ween cases and con ols he mean di e ence was s a is ically
signi ican and is in he opposi e di ec ion o ha expec ed. In a
mul i a ia e analysis adjus ed o age, amily his o y o PCA and
Table 1. R-squa e o 8 SNPs
SNPs s11630647 s11831436 s13317803 s2229765 s2871864 s35767 s5742612 s6503691
s11630647 1
s11831436 0.0000 1
s13317803 0.0001 0.0000 1
s2229765 0.0000 0.0001 0.0000 1
s2871864 0.2013 0.0001 0.0000 0.0030 1
s35767 0.0000 0.0484 0.0000 0.0000 0.0001 1
s5742612 0.0000 0.2629 0.0000 0.0000 0.0001 0.2080 1
s6503691 0.0002 0.0001 0.0001 0.0001 0.0000 0.0000 0.0001 1
Abb e ia ion: SNP ¼single-nucleo ide polymo phism.
Table 2. Demog aphic da a
95% CI
Va iables Case Con ol OR Lowe Uppe P- alue
Age (yea s)
Numbe 6207 6016
Mean±s.d. 63±760
±7o0.001
a
Family his o y o PCA
b
No 4051 3594 1.00
Yes 904 831 1.12 1.00 1.24 o0.05
Heigh (cm)- all cases
Mean±s.d. 176.3±7.0 176.8±7.1 o0.001
a
Numbe 2480 6016
Heigh (cm)- agg essi e cases
Mean±s.d. 176.3±7.0 176.8±7.1 o0.05
a
Abb e ia ions: CI ¼con idence in e al; OR ¼odds a io; PCA=p os a e cance .
a
P- alue o S uden - es .
b
Adjus ed o age.
BRITISH JOURNAL OF CANCER Heigh , selec ed gene ic ma ke s and p os a e cance isk
736 www.bjcance .com | DOI:10.1038/bjc.2017.231
s udy si es, heigh as a con inuous a iable did no show
associa ions wi h PCA isk in ei he all PCA cases o high g ade
PCA cases. Howe e , heigh ca ego ised in qua iles did show
signi ican ly inc eased isk in high g ade cases. Subjec s wi h a
heigh 4180 cm a e a 22% inc eased isk compa ed wi h subjec s
wi h heigh o173 cm. We did no obse e any associa ion be ween
heigh and low g ade cases ((Gleason g ade o7) esul s a e no
p esen ed in he pape ). Ou indings sugges alle subjec s a e a
inc eased isk o high g ade PCA isk. A p e ious epo om a
la ge nes ed case–con ol s udy (P o ecT) epo ed he OR o
p os a e-speci ic an igen–de ec ed high-g ade PCA pe 10 cm
inc ease in heigh was 1.23; 95% CI: 1.06–1.43. In a me a-analysis
o 58 s udies, a smalle e ec was epo ed ( andom-e ec s OR:
1.12; 95% CI: 1.05–1.19) (Zuccolo e al, 2008). Findings om The
Ea ly S age P os a e Cance Coho S udy which looked a he
ela ionship be ween heigh and p os a e cance g ade in a ious
0.95 1.0411
Odds a io (lo
g
scale)
S udies Es ima e (95% C. I. )
1. 012 (0.996, 1.028)
1. 017 (0.994, 1.041)
1. 007 (0.990, 1.024)
1. 005 (0.990, 1.020)
1. 001 (0.959, 1.045)
1. 002 (0.996, 1.009)
0. 990 (0.962, 1.019)
0. 987 (0.960, 1.015)
0. 991 (0.947, 1.037)
0. 991 (0.976, 1.006)
CAPS
ESTHER
FHCRC
MOFFITT
P o ecT
SEARCH
UKGPCS
PCMUS
QLD
O e all (I^2=0% , P=0.467)
Figu e 1. Fo es plo (all p os a e cance cases).
S udies
CAPS
ESTHER
FHCRC
MOFFITT
P o ecT
SEARCH
UKGPCS
PCMUS
QLD
O e all (I^2=0% , P=0.634)
Es ima e (95% C. I. )
1. 012 (0.993, 1.031)
1. 017 (0.986, 1.049)
1. 010 (0.988, 1.033)
1. 015 (0.995, 1.035)
0. 993 (0.949, 1.039)
0. 994 (0.961, 1.028)
0. 998 (0.956, 1.042)
1. 004 (0.954, 1.057)
1. 003 (0.996, 1.011)
0. 994 (0.982, 1.006)
0.95 1.06
11
Odds a io (log scale)
Figu e 2. Fo es plo (high g ade cases).
Table 3. Heigh as qua iles and p os a e cance isk
All cases High g ade cases
95% CI 95% CI
Heigh (cm)
Numbe o
subjec s
(all
cases þcon ols) OR
a
Lowe Uppe P- alue
Numbe o
subjec s
(High g ade
cases þcon ols) OR
a
Lowe Uppe P- alue
Q1 (p173.0) 2949 1.00 2000 1.00
Q2 (173.1–177.9) 3210 1.16 0.99 1.35 0.064 2196 1.20 0.99 1.46 0.069
Q3 (178.0–180.0) 1865 1.07 0.89 1.28 0.467 1331 1.19 0.94 1.49 0.150
Q4 (4180.0) 4199 1.11 0.96 1.28 0.173 2969 1.22 1.01 1.48 0.035
Abb e ia ions: CI ¼con idence in e al; OR ¼odds a io. All cases P o end 0.407. High-g ade cases P o end 0.075.
a
Adjus ed o age, amily his o y and s udy si es.
Heigh , selec ed gene ic ma ke s and p os a e cance isk BRITISH JOURNAL OF CANCER
www.bjcance .com | DOI:10.1038/bjc.2017.231 737
subpopula ions o men wi h po en ially di e en isk o high-g ade
PCA also sugges ed ha pa icipan s in he highes qua ile o
heigh we e mo e han wice as likely o ha e a Gleason sco e X7
(4 þ3) a biopsy han pa icipan s in he lowes qua ile o heigh
(OR 2.14 (95% CI 1.11, 4.14); Fa well e al, 2011). Two o he
s udies p esen ed esul s exclusi ely on cases wi h ad anced s age
PCA and bo h suppo ed a posi i e associa ion be ween heigh and
PCA isk (Hayes e al, 1999; No ish e al, 2000). Hayes and
colleagues obse ed a wo- old inc eased isk in whi e men wi h
heigh 41.75 me es compa ed o heigh o1.67 me es. The asso-
cia ion was absen among black men (Hayes e al, 1999). No ish
and colleagues in es iga ed he ole o heigh and PCA isk in bo h
Table 4. Candida e SNPs wi h signi ican associa ions
95% CI
SNP Mino allele Genes Odds a ios
a
Lowe Uppe P- alue
s6503691 A GHDC:STAT5B:STAT5A 0.90 0.82 0.99 0.036
s13317803 G GHSR:TNFSF10 1.08 1.01 1.14 0.016
s11831436 A IGF1 1.19 1.01 1.41 0.040
s35767 A IGF1 1.12 1.03 1.22 0.006
s5742612 G IGF1 1.32 1.13 1.55 0.001
s11630647 A IGF1R 1.08 1.01 1.15 0.035
s2871864 C IGF1R 1.11 1.01 1.22 0.031
s2229765 A IGF1R:PGPEP1L 1.08 1.02 1.15 0.013
Abb e ia ions: CI ¼con idence in e al; SNP ¼single-nucleo ide polymo phism.
a
Mul iple logis ic eg ession adjus ed o age, amily his o y o p os a e cance , s udy si e and P incipal Componen s o EU ances y.
Table 5. Es ima ed isk o gene ic isk sco es (s anda dised sco e) and p os a e cance isk
All PCA High g ade PCA
95% CI 95% CI
Gene ic isk sco e Odds a io
a
Lowe Uppe P- alue Odds a io
a
Lowe Uppe P- alue
Re e ence 1.00 1.00
2nd e ile 1.06 0.97 1.16 0.186 1.03 0.92 1.16 40.05
3 d e ile 1.13 1.04 1.23 0.006 1.55 1.03 1.29 o0.05
Abb e ia ions: CI ¼con idence in e al; PCA=p os a e cance . All cases P o end 0.006, high-g ade cases P o end 0.014.
a
Adjus ed o heigh .
Table 6. GE in e ac ion esul (Bayesian me hod)–in e ac ion be ween Heigh and gene ic isk sco es
G¼0G¼1 95% CI
G oup E¼0E¼1E¼0E¼1 To al Es ima ed
in e ac ion OR Lowe Uppe
Con ol 666 743 670 725 2804
All PCA case 738 629 766 743 2876 1.14 0.98 1.33
High-g ade PCA case 293 248 305 305 1151 1.18 0.94 1.49
Abb e ia ions: CI ¼con idence in e al; PCA=p os a e cance . G¼0-subjec s wi h gene ic isk sco e in he i s e ile, G¼1-subjec s wi h gene ic isk sco e in he hi d e ile. E¼0-subjec s
wi h heigh in he i s e ile, E¼1-subjec s wi h heigh in he hi d e ile.
Table 7. GE wi h 2 IGF-I pa hway SNPs by GMDR me hod
G oup Bes model Tes ing
accu acy
Tes ing
sensi i i y Tes ing odds a io
a
Tes ing
2
C oss- alida ion
consis ency
All PCA cases Heigh 0.51 0.51 1.07 (95% CI 0.72–1.59) 0.69 (P¼0.408) 10/10
Heigh , s5742612 0.51 0.51 1.12 (95% CI 0.75–1.66) 0.84 (P¼0.358) 10/10
Heigh , s5742612, s11831436 0.51 0.51 1.09 (95% CI 0.73–1.61) 0.75 (P¼0.387) 10/10
High g ade cases Heigh 0.51 0.54 1.16 (95% CI 0.63–2.11) 0.63 (P¼0.429) 10/10
Heigh , s11831436 0.51 0.52 1.06 (95% CI 0.58–1.94) 0.22 (P¼0.636) 10/10
Heigh , s5742612, s11831436 0.51 0.53 1.12 (95% CI 0.62–2.05) 0.44 (P¼0.507) 10/10
Abb e ia ions: GE ¼gene and en i onmen ; GMDR ¼gene al mul i ac o dimensionali y educ ion; IGF ¼insulin-like g ow h ac o ; PCA=p os a e cance ; SNP ¼single-nucleo ide
polymo phism.
a
Tes ing odds a ios adjus ed o age and amily his o y o p os a e cance .
BRITISH JOURNAL OF CANCER Heigh , selec ed gene ic ma ke s and p os a e cance isk
738 www.bjcance .com | DOI:10.1038/bjc.2017.231

spo adic cance cases and amilial cance cases. The s udy used he
Gleason g ading sco e o cha ac e ise he cases. Ad anced PCA
cases we e de ined by combined Gleason sco e X7 and localised
PCA cases by combined Gleason sco e p6. Resul s on spo adic
ad anced cance showed an indica ion o isk inc easing ac oss he
quin iles (p o end ¼0.07) which is simila o ou high g ade
cases. Mo eo e he isk was g ea e among hose wi h a posi i e
amily his o y o PCA (OR o heigh 4179 cm compa ed o
o170 cm ¼7.41, 95% CI 1.68–32.67, p o end ¼0.02). A null
associa ion was epo ed in localised cases. No only is heigh
po en ially associa ed wi h PCA isk bu i also shows associa ion
wi h PCA mo ali y. A ecen publica ion including mo e han 1
million subjec s in es iga ed adul heigh and he isk o cause-
speci ic dea h and ascula mo bidi y sugges ed ha haza d a ios
pe 6.5 cm g ea e heigh we e 1.04 (1.03–1.06) o dea h om
cance s and 1.07 (1.02–1.11) o dea h om PCA (Eme ging Risk
Fac o s Collabo a ion, 2012). In con as , he esul s o m a la ge
coho o 10 501 PCA cases and 10 831 con ols wi hin he NCI
B eas and P os a e Cance Coho Conso ium (BPC3) sugges ed
ha heigh was no associa ed wi h PCA isk bo h as a con inuous
a iable (OR: 1.001, 95% CI: 1.000–1.002 pe cm inc ease,
P¼0.12) o as in e iles (OR: 1.02, 95% CI: 0.99–1.06, P¼0.24)
(Linds om e al, 2011). A null associa ion was epo ed in he
s udy also using PRACTICAL geno ype da a se and in es iga ed
he e ec o heigh and p os a e cance incidence and mo ali y
using Mendelian andomisa ion app oach (Da ies e al, 2015). The
au ho s analysed gene ic a ian s associa ed wi h heigh om
published genome-wide associa ion s udies and epo ed ha hese
gene ic a ian s a e s ong ins umen o he a iable. The e a e
some limi a ions in ha GWA s udies will no explain a majo i y o
he es ima ed 80% con ibu ion o gene ic ac o s o a ia ion in
heigh (Lango Allen e al, 2010).
Human heigh is well known as a polygenic ai wi h a numbe
o genes ha con ibu e o heigh (Chial, 2008). Recen GWAS
s udies ha e iden i ied s ong and mode a e e ec s o genes ela ed
o human heigh (Weedon and F ayling, 2008; McE oy and
Vissche , 2009). Single SNPs wi h small e ec s in agg ega e o m
can be applied o assign indi iduals o hei heigh dis ibu ion
(Le e, 2009). We applied a candida e SNPs app oach and
iden i ied SNPs in genes ha had been geno yped in ou
conso ium ha we e ela ed o g ow h p ocesses. These SNPs
we e in he genes IGF-I,GH-1,SHOX,FMR1,GHITM, and
GHRHR (Gunnell, 2000; Ellis e al, 2001; Gunnell e al, 2001).
Twel e SNPs in hese genes show signi ican associa ions. We
compu ed
2
and kep he 8 SNP based on associa ion and P- alue
in each egion. Only one SNP ( s6503691) showed a small
p o ec i e e ec . This SNP is epo ed o associa e wi h
signi ican ly dec eased isk o b eas cance (Johansson e al,
2007; Zhao e al, 2015). Polymo phisms in he IGF signalling
pa hway ha e been shown o associa e wi h PCA mo ali y (Cao
e al, 2014). O he s udies epo ed null associa ions (Gu e al,
2010; Tsilidis e al, 2013). We also explo ed associa ion be ween
agg ega ed SNPs sco e as main e ec ; esul s suppo ha
indi iduals wi h gene ic isk sco es in he hi d e iles a e a
inc eased isk o high g ade PCA a 15% and o all PCA cases a
13% as compa ed wi h he lowes e ile. A es o end also
suppo s a dose- esponse ela ionship (P- alue o0.05 in bo h case
g oups). These indings suppo ha a gene ic isk sco e in he
g ow h pa hway a e associa ed wi h high g ade PCA. IGF genes
ha e been p e iously linked wi h PCA (Cheng e al, 2006;
Johansson e al, 2007; Cao e al, 2014; Gan e al, 2014; Qian
e al, 2014; Takeuchi e al, 2014; T a is e al, 2016). GHSR genes
a e also p e iously epo ed o associa e wi h p os a e cance isk
(D essen, 2007). We also in es iga ed possible gene-en i onmen
in e ac ions using wo app oaches. The i s app oach uses
combined gene ic isk sco es and a bina y a iable o heigh wi h
he i s e ile as he e e ence g oup and he hi d e ile as he
‘exposed’ g oup. Analyses we e done in bo h PCA and high g ade
case g oup using he Bayesian me hod p oposed by (Mukhe jee
e al, 2008). Resul s o he GE analyses howe e sugges ed no
in e ac ion be ween gene ic isk sco es and heigh . In he second
app oach, we selec ed he op 2 SNPs wi h he s onges e ec sizes
and i ed a model using he GMDR me hod (Chen e al, 2011).
The GMDR me hod allows adjus men o disc e e and quan i a-
i e co a ia es and is applicable o bo h dicho omous and
con inuous pheno ypes. The GMDR wi h co a ia e adjus men
had a powe o 480% in a case–con ol design wi h a sample size
o X2000. We applied he GMDR me hod because i di e s om
he adi ional GE me hod in ha i allows mo e han 1 SNP in he
model ( adi ional me hod-based on he concep o single- ac o –
based app oaches; Lou e al, 2007). The esul s also showed no
in e ac ions. None o he main e ec (heigh ) and ex ended models
showed any signi ican esul s.
In summa y, ou indings sugges ha heigh and
gene ic a ian s ela ed o he human g ow h pa hway a e
associa ed wi h high g ade PCA isk. Talle men o 41.80 m a e
a inc eased isk o high g ade PCA. Gene ic a ian s in genes ha
ela e o g ow h pa hways a e associa ed wi h p os a e cance isk.
The es ima ed isk is e iden amongs subjec s in he highes sco e
g oup when combined gene ic isk sco es we e used. The e is,
howe e , no GE in e ac ion be ween selec ed gene ic a ian s and
heigh .
ACKNOWLEDGEMENTS
This s udy would no ha e been possible wi hou he con ibu ions
o he ollowing: Pe Hall (COGS); Douglas F Eas on, Paul
Pha oah, Ky iaki Michailidou, Manjee K Bolla, Qin Wang
(BCAC), And ew Be chuck (OCAC), Rosalind A Eeles, Douglas
F Eas on, Ali Amin Al Olama, Zso ia Ko e-Ja ai, Sa a Benlloch
(PRACTICAL), Geo gia Chene ix-T ench, An onis An oniou,
Lesley McGu og, Fe gus Couch and Ken O i (CIMBA), Joe
Dennis, Alison M Dunning, And ew Lee, and Ed Dicks, C aig
Lucca ini and he s a o he Cen e o Gene ic Epidemiology
Labo a o y, Ja ie Beni ez, Anna Gonzalez-Nei a, and he s a o
he CNIO geno yping uni , Jacques Sima d and Daniel C Tessie ,
F ancois Baco , Daniel Vincen , Syl ie LaBoissie
` e and F ede ic
Robidoux and he s a o he McGill Uni e si y and Ge
´nome
Que
´bec Inno a ion Cen e, S ig E Bojesen, Sune F Nielsen, Bo ge
G No des gaa d, and he s a o he Copenhagen DNA labo a o y,
and Julie M Cunningham, Sha on A Windebank, Ch is ophe A
Hilke , Je ey Meye and he s a o Mayo Clinic Geno yping Co e
Facili y. Funding o he iCOGS in as uc u e came om: he
Eu opean Communi y’s Se en h F amewo k P og amme unde
g an ag eemen no 223175 (HEALTH-F2-2009-223175) (COGS),
Cance Resea ch UK (C1287/A10118, C1287/A 10710, C12292/
A11174, C1281/A12014, C5047/A8384, C5047/A15007, C5047/
A10692, and C8197/A16565), he Na ional Ins i u es o Heal h
(CA128978) and Pos -Cance GWAS ini ia i e (1U19 CA148537,
1U19 CA148065 and 1U19 CA148112– he GAME-ON ini ia i e),
he Depa men o De ence (W81XWH-10-1-0341), he Canadian
Ins i u es o Heal h Resea ch (CIHR) o he CIHR Team in
Familial Risks o B eas Cance , Komen Founda ion o he Cu e,
he B eas Cance Resea ch Founda ion, and he O a ian Cance
Resea ch Fund. Funding o ICEP (‘This wo k was also suppo ed
by CRUK (g an numbe C18281/A19169)’). Funding o he
CRUK s udy and PRACTICAL conso ium: his wo k was
suppo ed by he Canadian Ins i u es o Heal h Resea ch,
Eu opean Commission’s Se en h F amewo k P og amme g an
ag eemen no 223175 (HEALTH-F2-2009-223175), Cance
Resea ch UK G an s C5047/A7357, C1287/A10118, C5047/
A3354, C5047/A10692, C16913/A6135, and The Na ional Ins i u e
Heigh , selec ed gene ic ma ke s and p os a e cance isk BRITISH JOURNAL OF CANCER
www.bjcance .com | DOI:10.1038/bjc.2017.231 739
o Heal h (NIH) Cance Pos -Cance GWAS ini ia i e g an : No. 1
U19 CA 148537-01 ( he GAME-ON ini ia i e). We acknowledge
suppo om he NIHR o he Biomedical Resea ch Cen e a The
Ins i u e o Cance Resea ch and Royal Ma sden NHS Founda ion
T us .
CONFLICT OF INTEREST
The au ho s decla e no con lic o in e es .
THE PRACTICAL CONSORTIUM
(In addi ion o hose named in he au ho lis ) In o ma ion on
he conso ium can be ound a h p://p ac ical.ccge.medschl.
cam.ac.uk/.
Johanna Schleu ke
28,29
, Bø ge G No des gaa d
30,31
,
F ed ik Wiklund
32
, Ru h C T a is
33
, Ch is ophe A Haiman
34
,
S ephen N Thibodeau
35
,Ch is iane Maie
36
, Vogel Wal he
37
,
William J Blo
38,39
, Adam S Kibel
40
, Ceza y Cybulski
41
, Lisa
Cannon-Alb igh
42,43
, Ha de Pandha
44
, Manuel R Teixei a
45,46
,
Ma ga e Cook
47
, Ko eela Go indasami
48
, Michelle Guy
49
, Daniel
Leongamo nle
48
, Emma J Sawye
48
, Rosema y Wilkinson
48
,
Angela Mo gan
48
, Cy il Fishe
48
, Edwa d J. Saunde s
48
, Malgo za a
Tym akiewicz
48
, Naomi Li ni
48
, S e e Hazel
48
, Tokhi Dadae
48
,
Angela Cox
49
, Anne Geo ge
49
, A hene Lane
49
, Gemma Ma sden
49
,
Michael Da is
49
, Paul B own
49
, John Pede sen
50
, John L Hoppe
51
,
Ami Ka lsson
52
, Ca in Ca alli-Bjoe kman
52
, Jan Adol son
52
, Jan-
E ik Johansson
52
, Michael B oms
52
, Pae S a in
52
, Suzanne Kolb
53
,
Ch is a S egmaie
54
, Babu Zacha iah
55
, Hyun Pa k
55
, James
Haley
55
, Julio Pow-Sang
55
, Ma ia Rincon
55
, Selina Radlein
55
,
Aleksand ina Vlaho a
56
, A anaska Mi ko a
57
, Da ina Kacha-
ko a
57
, Elenko Popo
58
, S e lana Ch is o a
56
, Tihomi Diko
56
,
Allison Ecke
59
, Angus Collins
59
, Glenn Wood
59
, G eg Malone
59
,
Kimbe ly Alexande
59
, K is Ke
59
, Ma y-Anne Kedda
59
, Megan
Tu ne
59
, Pamela Saunde s
59
, Pe e Hea hco e
59
, S ilakshmi
S ini asan
59
, T acy Oma a
59
, T ina Yeadon
59
, Felici y Lose
59
28
Depa men o Medical Biochemis y and Gene ics Ins i u e o
Biomedicine Kiinamyllynka u 10, Uni e si y o Tu ku, FI-20014
Tu ku, Finland;
29
BioMediTech, Uni e si y o Tampe e and
FimLab Labo a o ies, Tampe e 33520, Finland;
30
Depa men o
Clinical Biochemis y, He le Hospi al, Copenhagen Uni e si y
Hospi al, He le Ring ej 75, DK-2730 He le , Denma k;
31
Facul y
o Heal h and Medical Sciences, Uni e si y o Copenhagen,
Blegdams ej 3B, Copenhagen 2200, Denma k;
32
Depa men o
Medical Epidemiology and Bios a is ics, Ka olinska Ins i u e,
S ockholm 10435, Sweden;
33
Cance Epidemiology Uni , Nu ield
Depa men o Popula ion Heal h Uni e si y o Ox o d, Ox o d
CB2 0RE, UK;
34
Depa men o P e en i e Medicine, Keck School
o Medicine, Uni e si y o Sou he n Cali o nia/No is Comp e-
hensi e Cance Cen e , Los Angeles, CA 90033, USA;
35
Depa -
men o Labo a o y Medicine and Pa hology, Mayo Clinic,
Roches e , MN 55905, USA;
36
Depa men o U ology, Uni e si y
Hospi al Ulm, Ulm 89081, Ge many;
37
Ins i u e o Human
Gene ics, Uni e si y Hospi al Ulm, 89081 Ulm, Ge many;
38
In e -
na ional Epidemiology Ins i u e, 1555 Resea ch Bl d., Sui e 550,
Rock ille, MD 20850, USA;
39
Di ision o Epidemiology, Depa -
men o Medicine, Vande bil Epidemiology Cen e , Vande bil -
Ing am Cance Cen e , Vande bil Uni e si y School o Medicine,
Nash ille, TN 37232, USA;
40
Di ision o U ologic Su ge y,
B igham and Womens Hospi al, Dana-Fa be Cance Ins i u e,
75 F ancis S ee , Bos on, MA 02115, USA;
41
In e na ional
He edi a y Cance Cen e , Depa men o Gene ics and Pa hology,
Pome anian Medical Uni e si y, Szczecin 70-115, Poland;
42
Di ision o Gene ic Epidemiology, Depa men o Medicine,
Uni e si y o U ah School o Medicine, Sal Lake Ci y, UT 84132,
USA;
43
Geo ge E Wahlen Depa men o Ve e ans A ai s Medical
Cen e , Sal Lake Ci y, UT 84148, USA;
44
Oncology G oup, Facul y
o Heal h and Medical Sciences, The Uni e si y o Su ey,
Guild o d, Su ey GU2 7WG, UK;
45
Depa men o Gene ics,
Po uguese Oncology Ins i u e, Po o 4200-072, Po ugal;
46
Bio-
medical Sciences Ins i u e (ICBAS), Uni e si y o Po o, Po o
4050-313, Po ugal;
47
Cen e o Cance Gene ic Epidemiology,
Depa men o Public Heal h and P ima y Ca e, Uni e si y o
Camb idge, S angeways Resea ch Labo a o y, Camb idge CB1
8RN, UK;
48
The Ins i u e o Cance Resea ch, Su on SM2 5PT,
UK;
49
CR-UK/YCR She ield Cance Resea ch Cen e, Uni e si y
o She ield, She ield S10 2TN, UK;
50
Tissupa h P y L d.,
Melbou ne, Vic o ia 3122, Aus alia;
51
Cen e o Epidemiology
and Bios a is ics, Melbou ne School o Popula ion and Global
Heal h, The Uni e si y o Melbou ne, Vic o ia 3010, Aus alia;
52
Depa men o Medical Epidemiology and Bios a is ics, Ka o-
linska Ins i u e, S ockholm 10435, Sweden;
53
Di ision o Public
Heal h Sciences, F ed Hu chinson Cance Resea ch Cen e , Sea le,
WA 98109-1024, USA;
54
Saa land Cance Regis y, Saa b u¨cken
66119, Ge many;
55
Depa men o Cance Epidemiology, Mo i
Cance Cen e , 12902 Magnolia D i e, Tampa, FL 33612, USA;
56
Depa men o Gene al and Clinical Pa hology, Medical
Uni e si y, So ia 1431, Bulga ia;
57
Depa men o Medical
Chemis y and Biochemis y, Molecula Medicine Cen e , Medical
Uni e si y, So ia, 2 Zd a e S ., So ia 1431, Bulga ia;
58
Depa men
o U ology and Alexand o ska Uni e si y Hospi al, Medical
Uni e si y, So ia 1431, Bulga ia;
59
Molecula Cance Epidemiology
Labo a o y, QIMR Be gho e Medical Resea ch Ins i u e, B isbane
4029, Aus alia
REFERENCES
Al Olama AA, Ko e-Ja ai Z, Be nd SI, Con i DV, Schumache F, Han Y,
Benlloch S, Hazele DJ, Wang Z, Saunde s E, Leongamo nle D,
Linds om S, Jugu nau h-Li le S, Dadae T, Tym akiewicz M,
S am DO, Rand K, Wan P, S am A, Sheng X, Poole LC, Pa k K, Xia L,
Ty e J, Kolonel LN, Le Ma chand L, Hoo e RN, Machiela MJ, Yeage M,
Bu de e L, Chung CC, Hu chinson A, Yu K, Goh C, Ahmed M,
Go indasami K, Guy M, Tammela TL, Au inen A, Wahl o s T,
Schleu ke J, Visako pi T, Leinonen KA, Xu J, Aly M, Dono an J,
T a is RC, Key TJ, Siddiq A, Canzian F, Khaw KT, Takahashi A, Kubo M,
Pha oah P, Pashayan N, Weische M, No des gaa d BG, Nielsen SF,
Kla sko P, Rode MA, I e sen P, Thibodeau SN, McDonnell SK,
Schaid DJ, S an o d JL, Kolb S, Hol S, Knudsen B, Coll AH, Gaps u SM,
Di e WR, S e ens VL, Maie C, Luedeke M, He komme K, Rinckleb AE,
S om SS, Pe away C, Yeboah ED, Te ey Y, Bi i wum RB, Adjei AA,
Tay E, T uelo e A, Niwa S, Chokkalingam AP, Cannon-Alb igh L,
Cybulski C, Wokolo czyk D, Kluzniak W, Pa k J, Selle s T, Lin HY,
Isaacs WB, Pa in AW, B enne H, Die enbach AK, S egmaie C, Chen C,
Gio annucci EL, Ma J, S amp e M, Penney KL, Mucci L, John EM,
Ingles SA, Ki les RA, Mu phy AB, Pandha H, Michael A, Kie zek AM,
Blo W, Signo ello LB, Zheng W, Albanes D, Vi amo J, Weins ein S,
Nemesu e B, Ca p en J, Leske C, Wu SY, Hennis A, Kibel AS, Rybicki BA,
Neslund-Dudas C, Hsing AW, Chu L, Goodman PJ, Klein EA, Zheng SL,
Ba a J, Clemen s J, Spu dle A, Teixei a MR, Paulo P, Maia S, Sla o C,
Kane a R, Mi e V, Wi e JS, Casey G, Gillande s EM, Semina a D,
Riboli E, Hamdy FC, Coe zee GA, Li Q, F eedman ML, Hun e DJ,
Mui K, G onbe g H, Neal DE, Sou hey M, Giles GG, Se e i G, Cook MB,
Nakagawa H, Wiklund F, K a P, Chanock SJ, Hende son BE, Eas on DF,
Eeles RA, Haiman CA (2014) A me a-analysis o 87,040 indi iduals
iden i ies 23 new suscep ibili y loci o p os a e cance . Na Gene 46:
1103–1109.
Al Olama AA, Ko e-Ja ai Z, Giles GG, Guy M, Mo ison J, Se e i G,
Leongamo nle DA, Tym akiewicz M, Jha a S, Saunde s E, Hoppe JL,
BRITISH JOURNAL OF CANCER Heigh , selec ed gene ic ma ke s and p os a e cance isk
740 www.bjcance .com | DOI:10.1038/bjc.2017.231
Sou hey MC, Mui KR, English DR, Dea naley DP, A de n-Jones AT,
Hall AL, O’B ien LT, Wilkinson RA, Sawye E, Lopha ananon A,
Ho wich A, Hudda RA, Khoo VS, Pa ke CC, Woodhouse CJ,
Thompson A, Ch is mas T, Ogden C, Coope C, Dono an JL, Hamdy FC,
Neal DE, Eeles RA, Eas on DF (2009) Mul iple loci on 8q24 associa ed
wi h p os a e cance suscep ibili y. Na Gene 41: 1058–1060.
Al Olama AA, Ko e-Ja ai Z, Schumache FR, Wiklund F, Be nd SI,
Benlloch S, Giles GG, Se e i G, Neal DE, Hamdy FC, Dono an JL,
Hun e DJ, Hende son BE, Thun MJ, Gaziano M, Gio annucci EL,
Siddiq A, T a is RC, Cox DG, Canzian F, Riboli E, Key TJ, And iole G,
Albanes D, Hayes RB, Schleu ke J, Au inen A, Tammela TL, Weische M,
S an o d JL, Os ande EA, Cybulski C, Lubinski J, Thibodeau SN,
Schaid DJ, So ensen KD, Ba a J, Clemen s JA, Chambe s S, Ai ken J,
Ga dine RA, Maie C, Vogel W, Do k T, B enne H, Habuchi T, Ingles S,
John EM, Dickinson JL, Cannon-Alb igh L, Teixei a MR, Kane a R,
Zhang HW, Lu YJ, Pa k JY, Cooney KA, Mui KR, Leongamo nle DA,
Saunde s E, Tym akiewicz M, Mahmud N, Guy M, Go indasami K,
O’B ien LT, Wilkinson RA, Hall AL, Sawye EJ, Dadae T, Mo ison J,
Dea naley DP, Ho wich A, Hudda RA, Khoo VS, Pa ke CC, Van As N,
Woodhouse CJ, Thompson A, Dudde idge T, Ogden C, Coope CS,
Lopha onanon A, Sou hey MC, Hoppe JL, English D, Vi amo J,
Le Ma chand L, Campa D, Kaaks R, Linds om S, Di e WR, Gaps u S,
Yeage M, Cox A, S e n MC, Co al R, Aly M, Isaacs W, Adol sson J, Xu J,
Zheng SL, Wahl o s T, Taa i K, Kujala P, Kla sko P, No des gaa d BG,
Rode MA, F ikke-Schmid R, Bojesen SE, Fi zge ald LM, Kolb S,
Kwon EM, Ka yadi DM, O n o TF, Bo e M, Rinckleb A, Luedeke M,
He komme K, Meye A, Se h J, Ma hick JR, Pa e son B,
Wokolo czyk D, Spu dle A, Lose F, McDonnell SK, Joshi AD, Shahabi A,
Pin o P, San os J, Ray A, Selle s TA, Lin HY, S ephenson RA, Tee link C,
Mulle H, Ro henbache D, Tsuchiya N, Na i a S, Cao GW, Sla o C,
Mi e V, Chanock S, G onbe g H, Haiman CA, K a P, Eas on DF, Eeles
RA (2012) A me a-analysis o genome-wide associa ion s udies o iden i y
p os a e cance suscep ibili y loci associa ed wi h agg essi e and
non-agg essi e disease. Hum Mol Gene 22(2): 408–415.
Calle EE (2000) In i ed Commen a y: Do An h opome ic Measu es P edic
isk o p os a e cance ? Am J Epidemiol 151: 550–553.
Cao Y, Linds om S, Schumache F, S e ens VL, Albanes D, Be nd S,
Boeing H, Bueno-de-Mesqui a HB, Canzian F, Chamosa S, Chanock SJ,
Di e WR, Gaps u SM, Gaziano JM, Gio annucci EL, Haiman CA,
Hende son B, Johansson M, Le Ma chand L, Palli D, Rosne B, Siddiq A,
S amp e M, S am DO, Tamimi R, T a is RC, T ichopoulos D,
Wille WC, Yeage M, K a P, Hsing AW, Pollak M, Lin X, Ma J (2014)
Insulin-like g ow h ac o pa hway gene ic polymo phisms, ci cula ing
IGF1 and IGFBP3, and p os a e cance su i al. J Na l Cance Ins 106:
dju085.
Chen GB, Xu Y, Xu HM, Li MD, Zhu J, Lou XY (2011) P ac ical and
heo e ical conside a ions in s udy design o de ec ing gene-gene
in e ac ions using MDR and GMDR app oaches. PLoS One 6: e16981.
Cheng I, S am DO, Penney KL, Pike M, Le Ma chand L, Kolonel LN,
Hi schho n J, Al shule D, Hende son BE, F eedman ML (2006) Common
gene ic a ia ion in IGF1 and p os a e cance isk in he Mul ie hnic
Coho . J Na l Cance Ins 98: 123–134.
Chial H (2008) Polygenic inhe i ance and gene mapping. Na Educ 1: 17.
Da ies NM, Gaun TR, Lewis SJ, Holly J, Dono an JL, Hamdy FC, Kemp JP,
Eeles R, Eas on D, Ko e-Ja ai Z, Al Olama AA, Benlloch S, Mui K,
Giles GG, Wiklund F, G onbe g H, Haiman CA, Schleu ke J,
No des gaa d BG, T a is RC, Neal D, Pashayan N, Khaw KT, S an o d JL,
Blo WJ, Thibodeau S, Maie C, Kibel AS, Cybulski C, Cannon-Alb igh L,
B enne H, Pa k J, Kane a R, Ba a J, Teixei a MR, Pandha H, La h op M,
Smi h GD, Ma in RM (2015) The e ec s o heigh and BMI on p os a e
cance incidence and mo ali y: a Mendelian andomiza ion s udy in
20 848 cases and 20 214 con ols om he PRACTICAL conso ium.
Cance Causes Con ol 26: 1603–1616.
D essen AS (2007) G ow h Ho mone Genes And P os a e Cance Risk. Mas e
o Science, Uni e si y o Pi sbu gh: Pi sbu gh.
Eeles RA, Ko e-Ja ai Z, Al Olama AA, Giles GG, Guy M, Se e i G, Mui K,
Hoppe JL, Hende son BE, Haiman CA, Schleu ke J, Hamdy FC,
Neal DE, Dono an JL, S an o d JL, Os ande EA, Ingles SA, John EM,
Thibodeau SN, Schaid D, Pa k JY, Spu dle A, Clemen s J, Dickinson JL,
Maie C, Vogel W, Do k T, Rebbeck TR, Cooney KA, Cannon-Alb igh L,
Chappuis PO, Hu e P, Zeege s M, Kane a R, Zhang HW, Lu YJ,
Foulkes WD, English DR, Leongamo nle DA, Tym akiewicz M,
Mo ison J, A de n-Jones AT, Hall AL, O’B ien LT, Wilkinson RA,
Saunde s EJ, Page EC, Sawye EJ, Edwa ds SM, Dea naley DP, Ho wich A,
Hudda RA, Khoo VS, Pa ke CC, Van As N, Woodhouse CJ,
Thompson A, Ch is mas T, Ogden C, Coope CS, Sou hey MC,
Lopha ananon A, Liu JF, Kolonel LN, Le Ma chand L, Wahl o s T,
Tammela TL, Au inen A, Lewis SJ, Cox A, Fi zGe ald LM,
Koopmeine s JS, Ka yadi DM, Kwon EM, S e n MC, Co al R, Joshi AD,
Shahabi A, McDonnell SK, Selle s TA, Pow-Sang J, Chambe s S, Ai ken J,
Ga dine RA, Ba a J, Kedda MA, Lose F, Polanowski A, Pa e son B,
Se h J, Meye A, Luedeke M, S e lo a K, Ray AM, Lange EM, Fa nham J,
Khan H, Sla o C, Mi ko a A, Cao G, Eas on DF (2009) Iden i ica ion o
se en new p os a e cance suscep ibili y loci h ough a genome-wide
associa ion s udy. Na Gene 41: 1116–1121.
Eeles RA, Ko e-Ja ai Z, Giles GG, Olama AA, Guy M, Jugu nau h SK,
Mulholland S, Leongamo nle DA, Edwa ds SM, Mo ison J, Field HI,
Sou hey MC, Se e i G, Dono an JL, Hamdy FC, Dea naley DP, Mui KR,
Smi h C, Bagna o M, A de n-Jones AT, Hall AL, O’B ien LT,
Geh -Swain BN, Wilkinson RA, Cox A, Lewis S, B own PM, Jha a SG,
Tym akiewicz M, Lopha ananon A, B yan SL, Ho wich A, Hudda RA,
Khoo VS, Pa ke CC, Woodhouse CJ, Thompson A, Ch is mas T,
Ogden C, Fishe C, Jamieson C, Coope CS, English DR, Hoppe JL,
Neal DE, Eas on DF (2008) Mul iple newly iden i ied loci associa ed wi h
p os a e cance suscep ibili y. Na Gene 40: 316–321.
Eeles RA, Olama AA, Benlloch S, Saunde s EJ, Leongamo nle DA,
Tym akiewicz M, Ghoussaini M, Lucca ini C, Dennis J,
Jugu nau h-Li le S, Dadae T, Neal DE, Hamdy FC, Dono an JL, Mui K,
Giles GG, Se e i G, Wiklund F, G onbe g H, Haiman CA, Schumache F,
Hende son BE, Le Ma chand L, Linds om S, K a P, Hun e DJ,
Gaps u S, Chanock SJ, Be nd SI, Albanes D, And iole G, Schleu ke J,
Weische M, Canzian F, Riboli E, Key TJ, T a is RC, Campa D, Ingles SA,
John EM, Hayes RB, Pha oah PD, Pashayan N, Khaw KT, S an o d JL,
Os ande EA, Signo ello LB, Thibodeau SN, Schaid D, Maie C, Vogel W,
Kibel AS, Cybulski C, Lubinski J, Cannon-Alb igh L, B enne H, Pa k JY,
Kane a R, Ba a J, Spu dle AB, Clemen s JA, Teixei a MR, Dicks E, Lee A,
Dunning AM, Baynes C, Con oy D, Ma anian MJ, Ahmed S,
Go indasami K, Guy M, Wilkinson RA, Sawye EJ, Mo gan A,
Dea naley DP, Ho wich A, Hudda RA, Khoo VS, Pa ke CC, Van As NJ,
Woodhouse CJ, Thompson A, Dudde idge T, Ogden C, Coope CS,
Lopha ananon A, Cox A, Sou hey MC, Hoppe JL, English DR, Aly M,
Adol sson J, Xu J, Zheng SL, Yeage M, Kaaks R, Di e WR, Gaude MM,
S e n MC, Co al R, Joshi AD, Shahabi A, Wahl o s T, Tammela TL,
Au inen A, Vi amo J, Kla sko P, No des gaa d BG, Rode MA,
Nielsen SF, Bojesen SE, Siddiq A, Fi zge ald LM, Kolb S, Kwon EM,
Ka yadi DM, Blo WJ, Zheng W, Cai Q, McDonnell SK, Rinckleb AE,
D ake B, Coldi z G, Wokolo czyk D, S ephenson RA, Tee link C,
Mulle H, Ro henbache D, Selle s TA, Lin HY, Sla o C, Mi e V, Lose F,
S ini asan S, Maia S, Paulo P, Lange E, Cooney KA, An oniou AC,
Vincen D, Baco F, Tessie DC, Ko e-Ja ai Z, Eas on DF (2013)
Iden i ica ion o 23 new p os a e cance suscep ibili y loci using he
iCOGS cus om geno yping a ay. Na Gene 45: 385–391. e1-2.
Ellis JA, S ebbing M, Ha ap SB (2001) Polymo phism o he and ogen
ecep o gene is associa ed wi h male pa e n baldness. J In es De ma ol
116: 452–455.
Eme ging Risk Fac o s Collabo a ion (2012) Adul heigh and he isk o
cause-speci ic dea h and ascula mo bidi y in 1 million people: indi idual
pa icipan me a-analysis. In J Epidemiol 41: 1419–1433.
Fa well WR, Lou enco C, Holmbe g E, Hall RB, D’A olio L, Lawle EV,
Michael Gaziano J (2011) The associa ion be ween heigh and p os a e
cance g ade in he Ea ly S age P os a e Cance Coho S udy. Cance
Causes Con ol 22: 1453–1459.
Fe lay J, Soe joma a am I, E ik M, Dikshi R, Ese S, Ma he s C, Rebelo M,
Pa kin DM, Fo man D, B ay F (2013) GLOBOCAN 2012 1.0, Cance
Incidence and Mo ali y Wo ldwide: IARC Cance Base No. 11 [In e ne ].
Lyon, F ance: In e na ional Agency o Resea ch on Cance . A ailable
om: h p://globocan.ia c. (accessed on 12 June 2016).
F eeman VL, Liao YL, Du azo-A izu R, Coope RS (2001) Heigh and isk o
a al p os a e cance : indings om he Na ional Heal h In e iew Su ey
(1986 o 1994). Ann Epidemiol 11: 22–27.
Gan Y, Buckels A, Liu Y, Zhang Y, Pa e son AJ, Jiang J, Zinn KR, F ank SJ
(2014) Human GH ecep o -IGF-1 ecep o in e ac ion: implica ions o
GH signaling. Mol Endoc inol 28(11): 1841–1854.
Gio annucci E, Rimm EB, S amp e MJ, Coldi z GA, Wille WC (1997)
Heigh , body weigh , and isk o p os a e cance . Cance Epidemiol
Bioma ke s P e 6: 557–563.
Heigh , selec ed gene ic ma ke s and p os a e cance isk BRITISH JOURNAL OF CANCER
www.bjcance .com | DOI:10.1038/bjc.2017.231 741
Gu F, Schumache FR, Canzian F, Allen NE, Albanes D, Be g CD, Be nd SI,
Boeing H, Bueno-de-Mesqui a HB, Bu ing JE, Chabbe -Bu e N,
Chanock SJ, Cla el-Chapelon F, Dumeaux V, Gaziano JM,
Gio annucci EL, Haiman CA, Hankinson SE, Hayes RB, Hende son BE,
Hun e DJ, Hoo e RN, Johansson M, Key TJ, Khaw KT, Kolonel LN,
Lagiou P, Lee IM, LeMa chand L, Lund E, Ma J, Onland-Mo e NC,
O e ad K, Rod iguez L, Sace do e C, Sanchez MJ, S amp e MJ, S a in P,
S am DO, Thomas G, Thun MJ, Tjonneland A, T ichopoulos D,
Tumino R, Vi amo J, Weins ein SJ, Wille WC, Yeage M, Zhang SM,
Kaaks R, Riboli E, Ziegle RG, K a P (2010) Eigh een insulin-like g ow h
ac o pa hway genes, ci cula ing le els o IGF-I and i s binding p o ein,
and isk o p os a e and b eas cance . Cance Epidemiol Bioma ke s P e
19: 2877–2887.
Gunnell D (2000) Heigh , insulin-like g ow h ac o s and cance isk. G ow h
Ho m IGF Res 10: S39–S40.
Gunnell D, Okasha M, Smi h GD, Oli e SE, Sandhu J, Holly JMP (2001)
Heigh , leg leng h, and cance isk: a sys ema ic e iew. Epidemiol Re 23:
313–342.
Hayes RB, Ziegle RG, G idley G, Swanson C, G eenbe g RS, Swanson GM,
Schoenbe g JB, Sil e man DT, B own LM, Po e n LM, Li J,
Schwa z AG, F aumeni J JF, Hoo e RN (1999) Die a y ac o s and isks
o p os a e cance among blacks and whi es in he Uni ed S a es. Cance
Epidemiol Bioma ke s P e 8: 25–34.
Hsing AW, Deng J, Ses e henn IA, Mos o i FK, S anczyk FZ, Benichou J,
Xie T, Gao YT (2000) Body size and p os a e cance : a popula ion-based
case-con ol s udy in China. Cance Epidemiol Bioma ke s P e 9:
1335–1341.
Johansson M, McKay JD, S a in P, Canzian F, Boillo C, Wiklund F,
Adami HO, Bal e K, G onbe g H, Kaaks R (2007) Comp ehensi e
e alua ion o gene ic a ia ion in he IGF1 gene and isk o p os a e
cance . In J Cance 120: 539–542.
Key TJ, Silcocks PB, Da ey GK, Appleby PN, Bishop DT (1997) A case-
con ol s udy o die and p os a e cance . B J Cance 76: 678–687.
Lango Allen H, Es ada K, Le e G, Be nd SI, Weedon MN, Ri adenei a F,
Wille CJ, Jackson AU, Vedan am S, Raychaudhu i S, Fe ei a T,
Wood AR, Weyan RJ, Seg e AV, Spelio es EK, Wheele E, So anzo N,
Pa k JH, Yang J, Gudbja sson D, Hea d-Cos a NL, Randall JC, Qi L,
Ve non Smi h A, Magi R, Pas inen T, Liang L, Heid IM, Luan J,
Tho lei sson G, Winkle TW, Godda d ME, Sin Lo K, Palme C,
Wo kalemahu T, Aulchenko YS, Johansson A, Zillikens MC, Fei osa MF,
Esko T, Johnson T, Ke ka S, K a P, Mangino M, P okopenko I,
Abshe D, Alb ech E, E ns F, Glaze NL, Haywa d C, Ho enga JJ,
Jacobs KB, Knowles JW, Ku alik Z, Monda KL, Polasek O, P euss M,
Rayne NW, Robe son NR, S ein ho sdo i V, Ty e JP, Voigh BF,
Wiklund F, Xu J, Zhao JH, Nyhol DR, Pellikka N, Pe ola M, Pe y JR,
Su akka I, Tammesoo ML, Al maie EL, Amin N, Aspelund T, Bhangale T,
Bouche G, Chasman DI, Chen C, Coin L, Coope MN, Dixon AL,
Gibson Q, G undbe g E, Hao K, Juhani Jun ila M, Kaplan LM,
Ke unen J, Konig IR, Kwan T, Law ence RW, Le inson DF, Lo en zon M,
McKnigh B, Mo is AP, Mulle M, Suh Ngwa J, Pu cell S, Ra el S,
Salem RM, Sal i E, Sanna S, Shi J, So io U, Thompson JR, Tu chin MC,
Vandenpu L, Ve laan DJ, Vi a V, Whi e CC, Ziegle A, Almg en P,
Balm o h AJ, Campbell H, Ci e io L, De G andi A, Dominiczak A,
Duan J, Ellio P, Elosua R, E iksson JG, F eime NB, Geus EJ, Glo ioso N,
Haiqing S, Ha ikainen AL, Ha ulinna AS, Hicks AA, Hui J, Igl W, Illig T,
Jula A, Kajan ie E, Kilpelainen TO, Koi anen M, Kolcic I, Koskinen S,
Ko acs P, Lai inen J, Liu J, Lokki ML, Ma usic A, Maschio A, Mei inge T,
Mulas A, Pa e G, Pa ke AN, Peden JF, Pe e smann A, Pichle I,
Pie ilainen KH, Pou a A, Ridde s ale M, Ro e JI, Samb ook JG,
Sande s AR, Schmid CO, Sinisalo J, Smi JH, S ingham HM,
B agi Wal e s G, Widen E, Wild SH, Willemsen G, Zaga o L, Zgaga L,
Zi ing P, Ala e e H, Fa all M, McA dle WL, Nelis M, Pe e s MJ,
Ripa i S, an Meu s JB, Aben KK, A dlie KG, Beckmann JS, Beilby JP,
Be gman RN, Be gmann S, Collins FS, Cusi D, den Heije M,
Ei iksdo i G, Gejman PV, Hall AS, Hams en A, Huiku i HV, I iba en C,
Kahonen M, Kap io J, Ka hi esan S, Kiemeney L, Koche T, Laune LJ,
Leh imaki T, Melande O, Mosley J TH, Musk AW, Nieminen MS,
O’Donnell CJ, Ohlsson C, Oos a B, Palme LJ, Rai aka i O, Ridke PM,
Rioux JD, Rissanen A, Ri ol a C, Schunke H, Shuldine AR,
Sisco ick DS, S um oll M, Tonjes A, Tuomileh o J, an Ommen GJ,
Viika i J, Hea h AC, Ma in NG, Mon gome y GW, P o ince MA,
Kayse M, A nold AM, A wood LD, Boe winkle E, Chanock SJ,
Deloukas P, Giege C, G onbe g H, Hall P, Ha e sley AT,
Hengs enbe g C, Ho man W, La h op GM, Salomaa V, Sch eibe S,
Uda M, Wa e wo h D, W igh AF, Assimes TL, Ba oso I, Ho man A,
Mohlke KL, Boomsma DI, Caul ield MJ, Cupples LA, E dmann J,
Fox CS, Gudnason V, Gyllens en U, Ha is TB, Hayes RB, Ja elin MR,
Moose V, Mun oe PB, Ouwehand WH, Penninx BW, P ams alle PP,
Que e mous T, Rudan I, Samani NJ, Spec o TD, Volzke H,
Wa kins H, Wilson JF, G oop LC, Ha i unians T, Hu FB, Kaplan RC,
Me spalu A, No h KE, Schlessinge D, Wa eham NJ, Hun e DJ,
O’Connell JR, S achan DP, Wichmann HE, Bo ecki IB,
an Duijn CM, Schad EE, Tho s einsdo i U, Pel onen L, Ui e linden
AG, Vissche PM, Cha e jee N, Loos RJ, Boehnke M, McCa hy MI,
Ingelsson E, Lindg en CM, Abecasis GR, S e ansson K, F ayling TM,
Hi schho n JN (2010) Hund eds o a ian s clus e ed in genomic
loci and biological pa hways a ec human heigh . Na u e 467:
832–838.
Le e G (2009) Gene ic egula ion o adul s a u e. Cu Opin Pedia 21:
515–522.
Linds om S, Schumache F, Siddiq A, T a is RC, Campa D, Be nd SI,
Di e WR, Se e i G, Allen N, And iole G, Bueno-de-Mesqui a B,
Chanock SJ, C aw o d D, Gaziano JM, Giles GG, Gio annucci E, Guo C,
Haiman CA, Hayes RB, Halkjae J, Hun e DJ, Johansson M, Kaaks R,
Kolonel LN, Na a o C, Riboli E, Sace do e C, S amp e M, S am DO,
Thun MJ, T ichopoulos D, Vi amo J, Weins ein SJ, Yeage M,
Hende son B, Ma J, Le Ma chand L, Albanes D, K a P (2011)
Cha ac e izing associa ions and SNP-en i onmen in e ac ions o
GWAS-iden i ied p os a e cance isk ma ke s– esul s om BPC3. PLoS
One 6: e17142.
Lou XY, Chen GB, Yan L, Ma JZ, Zhu J, Els on RC, Li MD (2007) A
gene alized combina o ial app oach o de ec ing gene-by-gene and
gene-by-en i onmen in e ac ions wi h applica ion o nico ine
dependence. Am J Hum Gene 80: 1125–1137.
McE oy BP, Vissche PM (2009) Gene ics o human heigh . Econ Hum Biol 7:
294–306.
Mukhe jee B, Ahn J, G ube SB, Renne G, Mo eno V, Cha e jee N (2008)
Tes s o gene-en i onmen in e ac ion om case-con ol da a: a
no el s udy o ype I e o , powe and designs. Gene Epidemiol 32:
615–626.
No ish AE, McRae CU, Holdaway IM, Jackson RT (2000) Heigh - ela ed isk
ac o s o p os a e cance . B J Cance 82: 241–245.
Pu cell S, Neale B, Todd-B own K, Thomas L, Fe ei a MA, Bende D,
Malle J, Skla P, de Bakke PI, Daly MJ, Sham PC (2007) PLINK: a ool
se o whole-genome associa ion and popula ion-based linkage analyses.
Am J Hum Gene 81: 559–575.
Qian J, Zhou H, Chen J, Ding Q, Cao Q, Qin C, Shao P, Li P, Cai H, Meng X,
Ju X, Wang M, Zhang Z, Li J, Hua L, Yin C (2014) Gene ic polymo phisms
in IGF-I and IGFBP-3 a e associa ed wi h p os a e cance in he Chinese
popula ion. PLoS One 9: e85609.
S a in P, Bylund A, Rinaldi S, Biessy C, Dechaud H, S enman UH, Ege ad L,
Riboli E, Hallmans G, Kaaks R (2000) Plasma insulin-like g ow h ac o -I,
insulin-like g ow h ac o -binding p o eins, and p os a e cance isk: a
p ospec i e s udy. J Na l Cance Ins 92: 1910–1917.
Takeuchi A, Shio a M, Be aldi E, Thape D, Takaha a K, Ibuki N, Pollak M,
Cox ME, Nai o S, Glea e ME, Zoubeidi A (2014) Insulin-like g ow h
ac o -I induces CLU exp ession h ough Twis 1 o p omo e p os a e
cance g ow h. Mol Cell Endoc inol 384: 117–125.
T a is RC, Appleby PN, Ma in RM, Holly JM, Albanes D, Black A,
Bueno-de-Mesqui a HB, Chan JM, Chen C, Chi laque MD, Cook MB,
Deschasaux M, Dono an JL, Fe ucci L, Galan P, Giles GG,
Gio annucci EL, Gun e MJ, Habel LA, Hamdy FC, Helzlsoue KJ,
He cbe g S, Hoo e RN, Janssen JA, Kaaks R, Kubo T, Le Ma chand L,
Me e EJ, Mikami K, Mo is JK, Neal DE, Neuhouse ML, Ozasa K,
Palli D, Pla z EA, Pollak MN, P ice AJ, Roobol MJ, Schae e C, Schenk JM,
Se e i G, S amp e MJ, S a in P, Tamakoshi A, Tangen CM, Tou ie M,
Wald NJ, Weiss NS, Ziegle RG, Key TJ, Allen NE (2016) A Me a-analysis
o indi idual pa icipan da a e eals an associa ion be ween
ci cula ing le els o IGF-I and p os a e cance isk. Cance Res 76:
2288–2300.
Tsilidis KK, T a is RC, Appleby PN, Allen NE, Linds om S, Albanes D,
Ziegle RG, McCullough ML, Siddiq A, Ba ica e A, Be nd SI,
Bueno-de-Mesqui a HB, Chanock SJ, C aw o d ED, Di e WR,
Gaps u SM, Gio annucci E, Gu F, Haiman CA, Hayes RB, Hun e DJ,
Johansson M, Kaaks R, Kolonel LN, K a P, Le Ma chand L, O e ad K,
Polido o S, Riboli E, Schumache FR, S e ens VL, T ichopoulos D,
BRITISH JOURNAL OF CANCER Heigh , selec ed gene ic ma ke s and p os a e cance isk
742 www.bjcance .com | DOI:10.1038/bjc.2017.231