Full Te ms & Condi ions o access and use can be ound a
h p://www. and online.com/ac ion/jou nalIn o ma ion?jou nalCode=igas20
Download by: [Tampe e Uni e si y] Da e: 19 Sep embe 2016, A : 03:31
Scandina ian Jou nal o Gas oen e ology
ISSN: 0036-5521 (P in ) 1502-7708 (Online) Jou nal homepage: h p://www. and online.com/loi/igas20
Risk o gas ic cance in Helicobac e pylo i
in ec ion in a 15-yea ollow-up
Ilkka Vohlonen, Ee o Pukkala, Nea Malila, Ma i Hä könen, Ma i Hakama,
Veli Kois inen & Pen i Sipponen
To ci e his a icle: Ilkka Vohlonen, Ee o Pukkala, Nea Malila, Ma i Hä könen, Ma i Hakama,
Veli Kois inen & Pen i Sipponen (2016) Risk o gas ic cance in Helicobac e pylo i in ec ion
in a 15-yea ollow-up, Scandina ian Jou nal o Gas oen e ology, 51:10, 1159-1164, DOI:
10.1080/00365521.2016.1183225
To link o his a icle: h p://dx.doi.o g/10.1080/00365521.2016.1183225
© 2016 Uni e si y o Eas e n Finland
Published by In o ma UK Limi ed, ading as
Taylo & F ancis G oup
Published online: 24 Jun 2016.
Submi you a icle o his jou nal
A icle iews: 564
View ela ed a icles
View C ossma k da a
ORIGINAL ARTICLE
Risk o gas ic cance in Helicobac e pylo i in ec ion in a 15-yea ollow-up
Ilkka Vohlonen
a
,Ee o Pukkala
b
,Nea Malila
c
,Ma i H€
a k€
onen
d
,Ma i Hakama
c
,Veli Kois inen
e
and
Pen i Sipponen
a
Depa men o Public Heal h, Uni e si y o Eas e n Finland, Kuopio, Finland;
b
Finnish Cance Regis y, Helsinki, Finland;
c
Depa men o
Epidemiology, Finnish Cance Regis y, Helsinki, Finland;
d
Depa men o Clinical Chemis y, Uni e si y o Helsinki, Helsinki, Finland;
e
Depa men o Bios a is ics, Finnish Consul ing G oup, Helsinki, Finland;
Depa men o Pa hology, Pa olab Oy, Espoo, Finland
ABSTRACT
Objec i e: We in es iga ed he isk o gas ic cance among men wi h Helicobac e pylo i (H. pylo i)
in ec ion o a ophic gas i is (AG) in a 15-yea ollow-up.
Ma e ials and me hods: S udy popula ion consis s o 12,016 men aged 50–65 yea s a he beginning
o he ollow-up in 1994–1996. Se um le els o pepsinogen I (SPGI) and an ibodies (IgG) o H. pylo i
(HpAb) we e assayed om se ums collec ed in 1994–1996. Incidence o gas ic cance in he s udy
popula ion was assessed in ollow-up om 1994 o 2011 by da a om he na ionwide cance egis y.
Based on SPGI and HpAb alues, s anda dized incidence a ios (SIRs) o gas ic cance we e calcula ed
in h ee subg oups, ha is, in hose wi h a heal hy s omach, hose wi h H. pylo i in ec ion bu wi hou
AG and hose wi h AG. Risk a ios (RR) o gas ic cance we e calcula ed using SIR o subg oups.
Resul s: Du ing 15 yea s, se en gas ic cance s appea ed pe 79,928 pe son yea s among men wi h
heal hy s omachs, 50 cance s pe 92,533 pe son yea s in men wi h H. pylo i in ec ion bu wi hou AG,
and 8 pe 8658 pe son yea s in men wi h AG. Risk a io (RR) o s omach cance in men wi h H. pylo i
in ec ion was 5.8 (95%CI: 2.7–15.3) compa ed o men wi h heal hy s omachs, and 9.1 (95%CI: 2.9–30.0)
in men wi h AG. The e we e no di e ences in cance isk be ween ca dia and dis al s omach.
Conclusions: Risk o gas ic cance is low in men wi h heal hy s omachs. I is signi ican ly inc eased in
hose wi h H. pylo i in ec ion and mo e in hose wi h AG.
ARTICLE HISTORY
Recei ed 4 Ma ch 2016
Re ised 20 Ap il 2016
Accep ed 23 Ap il 2016
KEYWORDS
A ophic gas i is; gas ic
cance ; Helicobac e pylo i
In oduc ion
A ophic gas i is (AG) and acid- ee s omach, which is ei he
au oimmune o a consequence o Helicobac e pylo i (H. pylo i)
in ec ion, a e isk condi ions o s omach cance , o he
cance o an in es inal ype in pa icula .[1–8] Less is known
abou he magni ude o cance isks in subjec s wi h heal hy
s omach mucosa o in hose wi h only nona ophic H. pylo i
in ec ion, al hough he H. pylo i in ec ion was classi ied as a
class 1 ca cinogen by WHO/IARC al eady in 1994.[1]
The H. pylo i in ec ion causes ch onic gas i is ha is
ini ially nona ophic, bu i may la e de elop in o a ious
o ms and s ages o a ophic gas i is and may end up as an
acid- ee s omach.[6–9] Like AG, he nona ophic o m o H.
pylo i gas i is is likely a p ecance ous condi ion, pa icula ly
o gas ic cance o he di use ype.[1,3,10]
In he p esen s udy, we in es iga ed he long- e m isk o
gas ic cance in a la ge popula ion-based sample o men
wi h o wi hou H. pylo i in ec ion o AG. The s udy popula-
ion consis ed o 12,016 men ep esen ing he gene al male
popula ion om wo Finnish ci ies and was collec ed om
men who pa icipa ed in a se um pepsinogen I (SPGI) sc een-
ing s udy in 1994–1996. The ea e , he men we e ollowed
o 15 yea s, and gas ic cance s in he s udy popula ion
du ing ollow-up we e iden i ied om he na ionwide cance
egis y. The s a us o gas ic mucosa in all 12,016 men was
assessed wi h bioma ke es s o SPGI and an ibodies (IgG)
o H. pylo i (HpAb) om se um samples collec ed in
1994–1996. Based on he bioma ke assays, we could classi y
he s udy popula ion in o h ee subg oups; ha is, hose wi h
a heal hy and no mal gas ic mucosa, hose wi h pu e H.
pylo i in ec ion (ch onic nona ophic gas i is) and hose wi h
mode a e o se e e a ophic gas i is.
The main objec i e o his s udy was o es ima e he isk
o gas ic cance due o only H. pylo i in ec ion; ha is, in
ch onic gas i is ha is caused by H. pylo i bu no ye p o-
g essed o he a ophic s age. The o he objec i es we e o
es ima e he isk o gas ic cance in men wi h heal hy s om-
ach mucosa, and o look a whe he he H. pylo i ela ed isk
o gas ic cance a ies be ween cance s in gas ic ca dia and
dis al s omach.
Me hods
S udy popula ion and he s udy coho s
Ini ially, 16,872 men (50–65 yea s old) om wo Finnish ci ies
we e iden i ied om he popula ion egis y and we e in i ed
CONTACT Ilkka Vohlonen [email p o ec ed] Uni e si y o Eas e n Finland, Public Heal h, BOX 1627, Kuopio 70100, Finland
ß2016 Uni e si y o Eas e n Finland Published by In o ma UK Limi ed, ading as Taylo & F ancis G oup
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i a i es License (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/),
which pe mi s non-comme cial e-use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, and is no al e ed, ans o med, o buil upon in any
way.
SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY, 2016
VOL. 51, NO. 10, 1159–1164
h p://dx.doi.o g/10.1080/00365521.2016.1183225
o gi e a blood sample o he SPGI es , hal in he au umn
o 1994 and hal in he au umn o 1996 [11] (Table 1).
Al oge he , 12,016 men (71%) pa icipa ed and o m he s udy
popula ion in he p esen in es iga ion. Among pa icipan s,
he se um le els o pepsinogen I (SPGI) and H. pylo i an ibod-
ies (IgG) (HpAb) we e assayed om blood samples collec ed
in 1994–1996 o de e mine he p esence o absence o an
ongoing H. pylo i in ec ion (ch onic gas i is) and he p esence
o absence o a mode a e o se e e s age o a ophic gas i is
in he s omach. The HpAb assays we e done in 2012–2013
om he same se um samples om which he SPGI es s
we e done al eady in 1994–1996.
Blood sampling and es s o SPGI and H. pylo i
an ibodies
Fas ing se a we e collec ed in ap o inin (T asylol, Baye
Ge many; 200 KIE/mL) con aining Venojec ubes and s o ed
a 70 C un il analyzed. SPGI was analyzed in 1994–1996
using he speci ic enzyme-linked immunoso ben assay
(ELISA) es s p o ided by Biohi Heal hca e Plc, Helsinki,
Finland. The assay has been calib a ed o co espond o
esul s ob ained by adioimmunoassay (RIA) used by Samlo
[12] using 238 se um samples wi h se um pepsinogen con-
cen a ions be ween 1.5 and 120 lg/L. The sensi i i y and
speci ici y o he SPGI es o ad anced (mode a e o se e e)
a ophic gas i is a e 92% and 90% a a cu o le el o 30 lg/
L, espec i ely, acco ding o he manu ac u e
s ki ins uc ions
o use. In clinical p ac ice and in endoscopy and biopsy his -
ology con olled ials, he SPGI cu o le el o 25 lg/L has
been demons a ed o be a eliable c i e ion in exclusion,
inclusion o delinea ion o cases ega ding he p esence o
absence o ad anced (mode a e o se e e) a ophic co pus
gas i is.[13–15] The e o e, he le el o 25 lg/L was also
selec ed as a cu o c i e ion o SPGI in he p esen s udy. In
addi ion, his cu o co esponds wi h he SPGI le els o
assays ini ially speci ied by Samlo .[12]
The HpAb assay was done in 2012–2013 by he ELISA es
p o ided by Biohi Heal hca e Plc, Helsinki, Finland, and was
pe o med acco ding o he ins uc ions o he manu ac u e .
The sensi i i y and speci ici y o he es o de ec he
ongoing H. pylo i in ec ion a e 93% and 96%, espec i ely,
acco ding o an independen analysis and epo o AFSSAPS
(Agence F ancaise de Secu i e Sani ai e des P odui s de
San e) on comme cial H. pylo i ELISA es ki s in 2008.
The se um le els o SPGI and HpAb we e used as c i e ia in
he classi ica ion o he s udy popula ion in o h ee subcoho s
(subg oups). The SPGI le els o 25 lg/L o mo e (25þ)we e
conside ed o exclude he cases wi h ad anced (mode a e o
se e e) a ophic gas i is in he s omach, i espec i e o he
cause o a ophic gas i is (au oimmuni y o H. pylo i o i-
gin).[14–16] The HpAb le els 30 EIU o mo e (30þ)we econ-
side ed o indica e an ongoing H. pylo i in ec ion (gas i is).[17]
Classi ica ion in o h ee subg oups was as ollows (Table 1):
Heal hy. Men wi h no mal and heal hy s omach mucosa: SPGI
25 lg/L o highe (SPGI 25þ) and HpAb lowe han 30 EIU (HpAb
<30).
H. pylo i in ec ion. Men wi h only H. pylo i in ec ion (gas i is)
wi hou mode a e o se e e a ophic gas i is (AG), and wi hou an
acid ee s omach: SPGI 25 lg/L o highe (SPGI 25þ) and HpAb
30 EIU o mo e (HpAb 30þ).
A ophic gas i is. Men wi h mode a e o se e e a ophic gas i is
in he gas ic co pus o undus (SPGI <25 lg/L), i espec i e o he
p esence o absence o an ongoing H. pylo i in ec ion. The
a ophic gas i is can be ei he au oimmune (HpAb nega i e) o o
H. pylo i o igin (HpAb posi i e).
Follow-up and eco ding o he cance incidence da a in
1994–2011
In o ma ion abou cance cases du ing he ollow-up om
1994 o 2011 was ecei ed in 2012 om he Finnish Cance
Regis y (FCR). Due o manda o y epo ing o all cance diag-
noses in Finland, he FCR has a na ional co e age o o e
99% o all new cance cases in Finland.[16] The ollow-up
began on 1 Decembe 1994 o he i s hal o he men and
on 1 No embe 1996 o he second hal o he men (one
Table 1. Follow-up da a o o al s udy popula ion and h ee subg oups, and on e e al ac ions ca ied ou in men om 1994–1996 o 2011.
Bioma ke es Pe sons Pe son Ac ions and es s
Subg oup SPG1 lg/L HpAb (IgG) Numbe Yea s Ca ied ou in each g oup
Heal hy 25þ<30 5232 79,928 SPG1 assayed in 1994–1996
HpAb s a us assayed in 2014 om
se um samples aken in 1994–1996
No ac i e clinical in e en ions in 1994–1996
H. pylo i in ec ion 25þ30þ6178 92,533 SPG1 assayed in 1996–1996
HpAb s a us assayed in 2014
om se um samples aken in 1994–1996
No ac i e clinical in e en ions in 1994–1996
A ophic gas i is <25 any 606 8658 SPG1 assayed in 1994–1996
HpAb s a us assayed in 2014 om se um samples aken in 1994–1996
All e e ed in 1994–1996 o consul a ion, ea men , endoscopies and su eillance
in local hospi als o heal h ca e cen e s
Whole sample any any 12,016 181,118 All 16,872 men bo n 1929–1949 in wo ci ies we e in i ed o he SPG1 es in
1994–1996
12,016 men could be es ed o SPG1 and HpAb (IgG) s a us om same se um sam-
ples aken in 1994–1996
SPG1 was assayed in 1994–1996 and HpAb s a us in 2014
Follow-up o he s udy coho om 1994 o 2011 by Finnish Cance Regis y
1160 I. VOHLONEN ET AL.
mon h a e inishing SPGI sc eening) and ended on 31
Decembe 2011).
The cance egis y includes da a on all in asi e cases o
cance . I does no include da a on cases classi ied as nonin-
asi e (p eneoplas ic lesions, dysplasia o in amucosal neo-
plasia). In o ma ion om he FCR p o ided da a on he
subsi e o cance in he s omach. The e o e, he analyses
could be pe o med sepa a ely o cases in which he cance
was loca ed in he dis al s omach (pylo us, an um, angulus,
co pus o undus) o in he gas ic ca dia.
In calcula ing he s anda dized incidence a ios (SIRs), he
age-s anda dized incidence o cance in he o al male popu-
la ion o Finland (expec ed a es) we e compa ed wi h he
obse ed incidence a es o cance in he subcoho s. The
expec ed numbe o cance s o each age g oup (5-yea age
ca ego ies) and 4-yea calenda pe iod we e es ima ed by
mul iplying he numbe o pe son yea s in he ca ego y accu-
mula ed in he subcoho s wi h he espec i e incidence a e
in he o al male popula ion a he same age in Finland. The
s anda dized incidence a es we e calcula ed acco ding no -
mal p ocedu es o age s anda diza ion.[17] The 95% con i-
dence in e als (CI) we e calcula ed on he basis o s anda d
e o o SIR [15] and hese we e es ima ed assuming a
Poisson dis ibu ion o he numbe o obse ed cance s. The
p alues o he di e ences be ween SIRs we e calcula ed on
he basis o con idence in e als.[18,19]
The au ho s had o icial pe mission o collec ion o he
da a and o ca ying ou he p esen in es iga ion by linking
he labo a o y es esul s wi h he in o ma ion om FCR
(Dn o THL/1349/5.05.00/2009).
Resul s
Al oge he , 65 gas ic cance cases appea ed in 15 yea s in
he p esen s udy popula ion. O hese, only se en (11%)
occu ed in men wi h a heal hy s omach (wi hou H. pylo i
in ec ion o AG) mucosa a he ime o d awing he blood
sample (1994–1996).
Table 2 p esen s he obse ed and expec ed cumula i e
incidence o gas ic cance in h ee subg oups du ing he 15-
yea ollow-up pe iod. Resul s on he incidence o cance sep-
a a ely in he dis al s omach and gas ic ca dia a e p o ided.
In addi ion, Table 2 p esen s he SIRs o gas ic cance and
he SIR o o al cance (all malignan diseases) in each sub-
g oup and in he whole-s udy coho .
We ound he SIR o gas ic cance among men wi h
heal hy s omachs o be signi ican ly lowe han he co e-
sponding SIR among men wi h H. pylo i in ec ion (SIR 0.21,
95% CI: 0.04–0.60) and lowe han among he whole s udy
coho (Table 3).
Table 3 p o ides in o ma ion on incidences o gas ic cance
and on SIRs acco ding o di e en leng hs o ollow-up ime
du ing he 15-yea s udy pe iod. A simila ly low SIR o gas ic
cance was e iden among men wi h heal hy s omachs o e
he whole 15 yea s o ollow-up. Among men wi h H. pylo i
in ec ion (gas i is), he incidence o gas ic cance ended o
inc ease wi h he leng h o ollow-up ime. The SIR o gas ic
cance was signi ican ly (p<0.05) inc eased among men in he
H. pylo i in ec ion g oup, ollowed up o 10 yea s o mo e.
The isk a io (RR) o gas ic cance be ween hose wi h H.
pylo i in ec ion and men wi h heal hy s omachs was 5.8. The
RR be ween hose wi h AG compa ed o hose wi h H. pylo i
in ec ion was 1.6 (0.6–3.3), and 9.1 (2.9–30.0) as compa ed o
men wi h heal hy gas ic mucosa. The isk a ios we e qui e
cons an ega dless o he sub-si e o gas ic cance (Table 4).
Table 2. Cance egis y based obse ed and expec ed incidence and SIR o
gas ic cance in h ee subg oups and in whole s udy popula ion du ing he ol-
low-up om 1994–1996 o 2011.
Subg oup, cance
ype and si e SPG1 lg/L
HpAb
(IgG) Obs. Exp. SIR 95%SD
Heal hy 25þ<30
Gas ic cance
Dis al s omach 5 25 0.20 0.01–0.39
Ca dia 2 7 0.29 0.01–0.73
S omach, all si es 7 32 0.22
a
0.04–0.40
To al cance , any si e 1403 1341 1.05 0.97–1.16
H. pylo i in ec ion 25þ30þ
Gas ic cance
Dis al s omach 37 31 1.19 0.62–1.76
Ca dia 13 8 1.63 0.19–3.06
S omach, all si es 50 39 1.28 0.75–1.82
To al cance , any si e 1692 1642 1.03 0.96–1.13
A ophic gas i is <25 any
Gas ic cance
Dis al s omach 6 3 2.00 0.01–4.77
Ca dia 2 1 2.00 0.01–6.80
S omach, all si es 8 4 2.00 0.01–4.40
To al cance , any si e 167 165 1.01 0.79–1.32
Whole sample any any
Gas ic cance
Dis al s omach 48 60 0.80 0.50–1.10
Ca dia 17 15 1.13 0.35–1.92
S omach, all si es 65 75 0.87 0.58–1.15
To al cance , any si e 3262 3146 1.04 0.99–1.10
a
p¼0.009.
Table 3. Cance egis y based obse ed and expec ed incidence and SIR o
gas ic cance in h ee subg oups in di e en ime pe iods o ollow-up om
1994–1996 o 2011.
Subg oup, cance
ype and si e SPG1 lg/L
HpAb
(IgG) Obs. Exp. SIR 95%SD
Heal hy 25þ<30
Gas ic cance
Follow-up 0–2 yea s 1 3 0.33 0.01–1.62
Follow-up 2–9 yea s 3 14 0.21
a
0.04–0.62
Follow-up 10 þyea s 3 14 0.21
a
0.04–0.62
H. pylo i in ec ion 25þ30þ
Gas ic cance
Follow-up 0–2 yea s 4 4 1.00 0.25–2.32
Follow-up 2–9 yea s 17 17 1.00 0.57–1.55
Follow-up 10 þyea s 29 17 1.71 0.68–2.72
A ophic gas i is <25 any
Gas ic cance
Follow-up 0–2 yea s 2 0.5 4.00 0.50–14.83
Follow-up 2–9 yea s 2 2 1.00 0.13–3.97
Follow-up 10 þyea s 4 2 2.37 0.65–6.06
Whole sample any any
Gas ic cance
Follow-up 0–2 yea s 7 8 0.88 0.32–1.73
Follow-up 2–9 yea s 22 33 0.67 0.41–1.00
Follow-up 10 þyea s 36 34 1.06 0.75–1.48
To al cance
Follow-up 0–2 yea s 215 214 1.00 0.87–1.14
Follow-up 2–9 yea s 1384 1325 1.04 0.99–1.08
Follow-up 10 þyea s 1663 1608 1.03 0.99–1.07
a
p¼0.025.
SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY 1161
Discussion
Al oge he , 65 gas ic cance s de eloped in he whole s udy
popula ion o 12,016 men du ing he 15-yea ollow-up
pe iod, o which only 7 (11%) cance s de eloped in men wi h
heal hy s omach mucosa (5232 men). The SIR o gas ic can-
ce in he subg oup o men wi h heal hy s omachs was nea ly
80% lowe han expec ed, and he RR o SIRs was app oxi-
ma ely six imes highe in hose wi h H. pylo i in ec ion com-
pa ed o men wi h a heal hy s omach. Thus, he people wi h
heal hy s omach mucosa a e clea ly hose who ha e he low-
es isk o gas ic cance known so a .
The p esen obse a ion also demons a es ha an H.
pylo i in ec ion alone (nona ophic H. pylo i gas i is) is by
i sel a clea isk condi ion o gas ic cance as was sugges ed
by he IARC/WHO s a emen in 1994.[1] The obse a ion indi-
ca es ha he simple in ec ion ma kedly inc eases he cance
isk when compa ed o a heal hy s omach. Suppo ing he
conclusions o H. pylo i in ec ion as a isk condi ion o gas ic
cance , he SIR o gas ic cance among men wi h H. pylo i
in ec ion ended o ise du ing ollow-up. P ac ically, a hal
(29 o 50) o gas ic cance s in he H. pylo i subg oup
appea ed du ing he las 5-yea pe iod (10þyea s).
We au ho s conside he es ima es o SIRs and RRs o gas-
ic cance in he p esen subg oups o ‘Heal hy s omach
mucosa’ o ‘H. pylo i in ec ion wi hou AG’ o be eliable and
epidemiologically alid. The delinea ion o men in o sub-
g oups was based on bioma ke es s ha a e highly eliable
in e lec ing he heal h o s omach mucosa and enable nonin-
asi e es s o la ge popula ions o asymp oma ic people wi h
a simple me hod.[17–20] Fu he mo e, he gas ic cance s
iden i ied du ing he ollow-up pe iod we e based on cance
cases eco ded by na ionwide cance egis y wi h co e age
o o e 99% o all cance cases in Finland.[16] In addi ion, he
subg oups ollowed up in he p esen in es iga ion we e
de i ed om he s udy popula ion ha ep esen s he gene al
popula ion o males who we e 50–65 yea s old a he begin-
ning o he ollow-up and we e li ing in wo ci ies in sou h-
e n Finland in 1994–1996.[11]
All men in he subg oups o heal hy gas ic mucosa o H.
pylo i in ec ion wi hou AG had no mal le els o SPGI in
1994–1996. No e e al o ea men , clinical su eillance o
o he ac ions we e ca ied ou by he in es iga o s p any-
one, and all men we e passi ely ollowed up by he cance
egis y du ing he 15-yea ollow-up pe iod om 1994 o
2011.[11] In addi ion, he delinea ion o he men o hose
wi h heal hy s omach mucosa and o hose wi h nona ophic
H pylo i gas i is was done wi h he HpAb es ( om se um
samples collec ed in 1994–1996) only a he end o he
ollow-up pe iod. Howe e , unknown con ounde s may s ill
exis ha may con ibu e o es ima es o cance isks in he
s udy subg oups. Fo example, di e ences in smoking, die a y
habi s, alcohol consump ion, socioeconomic s a us o e adica-
ion o H. pylo i du ing he 15-yea ollow-up among hou-
sands o men in each subg oup, could no be con olled o
in he p esen in es iga ion. Howe e , hese possible biases,
e en hough ce ainly exis ing, can ha dly explain he di e -
ences obse ed in he gas ic cance isk be ween men wi h
heal hy gas ic mucosa and hose wi h H. pylo i in ec ion.
On he o he hand, he obse ed SIRs and RRs o gas ic
cance a e likely se e ely biased in he p esen s udy in he
subg oup o men wi h AG; ha is, in he g oup o 606 men
who had mode a e o se e e a ophic gas i is in gas ic co -
pus and undus by he SPGI es in 1994–1996. All hese men
had a low (<30 lg/L) se um le el o pepsinogen I (SPGI) and
we e, he e o e, conside ed o ha e a hypochlo hyd ic o
e en achlo hyd ic s omach in 1994–1996; ha is, a se e ely
sick and a ophic s omach mucosa conside ed o be a p ema-
lignan condi ion ha needs special a en ion.[1–5] The e o e,
all hese men we e ac i ely e e ed in 1994–1996 o medical
consul a ion, ea men and clinical su eillance in specialized
hospi als o heal h ca e cen e s.[11]
Possible endoscopic, su gical, he apeu ic o p e en i e
in e en ions ca ied ou among he 606 men wi h AG may
dec ease he incidence o gas ic cance in he 15-yea ollow-
up.[11] The e o e, he obse ed incidences and SIRs o gas ic
cance in his subg oup o men a e likely unde es ima ions o
he eal cance isk. The au ho s we e no able o explo e he
p ocedu es ha we e done o hese 606 men wi h AG du ing
he 15-yea ollow-up pe iod. In spi e o his, he RR o gas ic
cance was app oxima ely 9 in he subg oup o men wi h AG
as compa ed o men wi h heal hy s omach mucosa.
In he p esen s udy, he RR o cance s in he gas ic ca dia
was 5.4 in men wi h H. pylo i se oposi i i y as compa ed o
ca dia cance s in subjec s wi h heal hy s omach mucosa. The
SIR o ca dia cance s was also qui e simila (1.6 s. 1.3) o ha
o dis al gas ic cance in men wi h H. pylo i se oposi i i y.
E en hough he numbe o cases is low, and e en hough
he isk es ima es a e insigni ican , he obse a ions suppo
he iew ha he H. pylo i in ec ion associa es wi h ca dia
cance s simila ly as wi h gas ic cance s in he dis al s omach.
The obse a ions o H. pylo i in ec ion as an e iological ac-
o o cance o he gas ic ca dia ha e been con adic o y.
In con as o he p esen obse a ions, some s udies indica e
ha ca dia adenoca cinomas do no associa e wi h H. pylo i
in ec ion and ha he pa hogenesis o hese cance s esem-
bles he pa hogenesis o lowe esophageal adenoca cinomas
in pa ien s wi h gas oesophageal e lux disease (GERD).[20]
Di e ences in pa hogenesis o ca dia and nonca dia gas ic
ca cinomas ha e been p oposed.[21–25]
In he pas , se e al p ospec i e in es iga ions wi h
case–con ol design ha e been published on he isk o
gas ic cance due o H. pylo i in ec ion. In a sys ema ic
e iew co e ing 10 s udies be o e 1998, including app oxi-
ma ely 800 gas ic cance cases, he analysis yielded a RR
o 2.5 (95%CI: 1.9–3.4) o gas ic cance in H. pylo i-se o-
posi i e people.[26] An Eu opean p ospec i e case–con ol
s udy o 233 gas ic cance s and 910 con ols, including
Table 4. The isk a ios (RR) and CI95% o isk a ios o gas ic cance be ween
he subg oups by si e o cance om 1994–1996 o 2011.
Si e o cance
Be ween
H. pylo i and
heal hy mucosa
Be ween a ophic
gas i is (AG) and
H. pylo i in ec ion
Be ween a ophic
gas i is (AG) and
heal hy mucosa
Dis al s omach 6.0 (2.3–19. 0) 1.7 (0.6–4.0) 10.0 (2.5–41.0)
p¼0.001 p¼0.276 p¼0.001
Ca dia 5.7 (1.3–52.0) 1.2 (0.1–5.4) 7.0 (0.5–97.0)
p¼0.064 p¼0.867 p¼0.148
S omach, all si es 5.8 (2.7–15.3) 1.6 (0.6–3.3) 9.1 (2.9–30.0)
p¼0.000 p¼0.283 p¼0.000
1162 I. VOHLONEN ET AL.

he SPGI assay in addi ion o he H. pylo i es , yielded a
RR o 6.5 (95%CI: 3.3–12.6) o nonca dia gas ic cance in
subjec s in ec ed wi h a cy o oxic (CagA) H. pylo i s ain.[27]
In a p ospec i e Finnish s udy, he RR o gas ic cance was
3.1 (95%CI: 1.97–4.95) be ween H. pylo i in ec ed and non-
in ec ed pe sons.[28] The RR, based on case–con ol s udy
designs, a ied be ween 1.6 and 7.9 in h ee published
pape s om wo ex ensi e p ospec i e nu i ional in e en-
ion ials o o e 29,000 males a age o 50–69 yea s in
Linxian, China and Finland.[23–25]
Ou es ima e o he isk o gas ic cance due o H. pylo i
in ec ion, 5.8 (95%CI: 2.7–15.3), is simila o somewha highe
han he published isks es ima ed in he abo e-men ioned
p ospec i e in es iga ions. In con as o he men ioned
case–con ol in es iga ions, he isk es ima es o gas ic can-
ce in he p esen s udy we e ecei ed om all new cases o
gas ic cance ha appea ed du ing he ollow-up ime in all
subg oups o men wi hou ongoing H. pylo i in ec ion o AG,
and hose wi h a e i ied ongoing H. pylo i in ec ion o AG a
beginning o he ollow-up.[11]
Di e ences in he magni ude o he obse ed isk o
gas ic cance be ween he p esen and ea lie published
in es iga ions may be due o di e ences in he s udy
design, in he applica ion o bioma ke es s o de ining H.
pylo i in ec ion o AG, o in he c i e ia o he subjec s who
a e classi ied as ha ing heal hy gas ic mucosa o AG. All
subjec s wi h a nega i e H. pylo i es do no ha e heal hy
gas ic mucosa and he e o e canno be classi ied as
heal hy con ols. Helicobac e pylo i nega i e cases a e, on
he o he hand, o en conside ed heal hy, e en hough
hei s omachs a e, in ac , se e ely sick. In an ea lie
Finnish s udy among subjec s wi h endoscopically e i ied
a ophic gas i is, an ongoing H. pylo i in ec ion was ound
only in 82% o cases wi h he se ological H. pylo i es .[29]
Thus, wi hou con olling o SPGI, he H. pylo i se onega i e
cases wi h AG a e easily misclassi ied in o he subg oup o
people wi h no mal and heal hy gas ic mucosa. On he
o he hand, he cases wi h H. pylo i-se oposi i e AG and
acid- ee s omach may be e oneously classi ied as cases
wi h a simple uncomplica ed H. pylo i in ec ion.
We conclude ha he p esen in es iga ion emphasizes
he ollowing: o all gas ic cance s ha occu ed du ing he
15-yea ollow-up among elde ly men, only 11% appea ed in
men wi h heal hy s omachs. The isk o s omach cance is
app oxima ely six imes highe among men wi h H. pylo i
in ec ion han among men wi h heal hy s omach mucosa,
and he H. pylo i in ec ion aises he gas ic cance isk simi-
la ly in he gas ic ca dia and in o he si es o he s omach.
Acknowledgemen s
The au ho s hank Biohi Plc o pe o ming o all SPGI and HpAb assays
in Biohi Se ice Labo a o y, Helsinki, Finland.
Re e ences
[1] IARC. Schis osomes, li e lukes and Helicobac e pylo i. IARC
Wo king G oup on he E alua ion o Ca cinogenic Risks o
Humans. Lyon, 7–14 June 1994. IARC Monog E al Ca cinog Risks
Hum 1994;61:1–241
[2] Co ea P, Haenszel W, Cuello C, e al. Gas ic p ecance ous p ocess
in a high isk popula ion: coho ollow-up. Cance Res. 1990;50:
4737–4740.
[3] Co ea P, Piazuelo MB. The gas ic p ecance ous cascade. J Dig
Dis. 2012;13:2.
[4] Haenszel W, Co ea P, Cuello C, e al. Gas ic cance in Colombia.
II. Case-con ol epidemiologic s udy o p ecu so lesions. J Na l
Cance Ins . 1976;57:1021–1026.
[5] Sipponen P, Kekki M, Haapakoski J, e al. Gas ic cance isk in
ch onic a ophic gas i is: s a is ical calcula ions o c oss-sec ional
da a. In J Cance . 1985;35:173–177.
[6] Siu ala M, Sipponen P, Kekki M. Ch onic gas i is: dynamic
and clinical aspec s. Scand J Gas oen e ol Suppl.
1985;109:69–76.
[7] Valle J, Kekki M, Sipponen P, e al. Long- e m cou se and conse-
quences o Helicobac e pylo i gas i is. Resul s o a 32-yea ollow-
up s udy. Scand J Gas oen e ol. 1996;31:546–550.
[8] Villako K, Kekki M, Maa oos HI, e al. Ch onic gas i is: p og ession
o in lamma ion and a ophy in a six-yea endoscopic ollow-up o
a andom sample o 142 Es onian u ban subjec s. Scand J
Gas oen e ol Suppl. 1991;186:135–141.
[9] Kekki M, Iham€
aki T, Saukkonen M, e al. P og ession o gas i is a
a popula ion le el. Compa ison o a long- e m ollow-up wi h s o-
chas ic analysis o c oss-sec ional da a. Scand J Gas oen e ol.
1980;15:651–655.
[10] Ma shall BJ, Wa en JR. Uniden i ied cu ed bacilli in he s omach
o pa ien s wi h gas i is and pep ic ulce a ion. Lance .
1984;323:1311–1315.
[11] Vohlonen I, H€
a k€
onen M, Malila N, e al. Bioma ke sc eening on
isk o s omach cance . Cance mo ali y in a 15-yea ollow-up o
men wi h low se um pepsinogen in 1994–1996. To be published
[12] Samlo IM. Pepsinogens I and II: pu i ica ion om gas ic
mucosa and adioimmunoassay in se um. Gas oen e ology.
1981;82:26–33.
[13] Va is K, Sipponen P, Laxen F, e al. Implica ions o se um pepsin-
ogen I in ea ly endoscopic diagnosis o gas ic cance and dyspla-
sia: Helsinki Gas i is S udy G oup. Scand J Gas oen e ol.
2000;35:950–956.
[14] S o sk ubb T, A o P, Ronkainen J, e al. Se um bioma ke s p o ide
an accu a e me hod o diagnosis o a ophic gas i is in a gene al
popula ion: he Kalixanda s udy. Scand J Gas oen e ol. 2008
43:1448–1455.
[15] Samlo IM, Va is K, Iham€
aki T, e al. Rela ionships among se um
pepsinogen I, se um pepsinogen II, and gas ic mucosal his ology.
A s udy in ela i es o pa ien s wi h pe nicious anemia.
Gas oen e ology. 1982;83:204–209.
[16] The Finnish Cance Regis y 2011. [In e ne ]. A ailable om: www.
cance egis y. i.
[17] Fleiss JL. S a is ical me hods o a es and p opo ions. New Yo k:
Wiley; 1981.
[18] Bland JM, Al man DG. S a is ics no es. Loga i hms. BMJ. 1996;312:700.
[19] Al man DG, Bland JM. How o ob ain he P alue om a con i-
dence in e al. BMJ. 2011;343:d2304.
[20] Ag 
eusL,Kuipe sEJ,KupcinskasL,e al.Ra ionaleindiagno-
sis and sc eening o a ophic gas i is wi h s omach-
speci ic plasma bioma ke s. Scand J Gas oen e ol.
2012;47:136–147.
[21] Hansen S, Vollse SE, De akhshan MH, e al. Two dis inc ae iolo-
gies o ca dia cance ; e idence om p emo bid se ological
ma ke s o gas ic a ophy and Helicobac e pylo i s a us. Gu .
2007;56:918–925.
[22] De akhshan MH, Malekzadeh R, Wa abe H, e al. Combina ion o
gas ic a ophy, e lux symp oms and his ological sub ype indi-
ca es wo dis inc ae iologies o gas ic ca dia cance . Gu .
2008;57:298–305.
[23] Kamanga F, Qiao YL, Blase MJ, e al. Helicobac e pylo i and
oesophageal and gas ic cance s in a p ospec i e s udy in China.
B J Cance . 2007;96:172–176.
SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY 1163
[24] Kamanga F, Dawsey SM, Blase MJ, e al. Opposing isks o gas ic
ca dia and nonca dia gas ic adenoca cinomas associa ed wi h
Helicobac e pylo i se oposi i i y. J Na l Cance Ins .
2006;98:1445–1452.
[25] Limbu g P, Qiao Y, Ma k S, e al. Helicobac e pylo i se oposi i i y
and subsi e-speci ic gas ic cance isks in Linxian, China. J Na l
Cance Ins . 2001;93:226–233.
[26] Danesh J. Helicobac e pylo i in ec ion and gas ic cance : sys em-
a ic e iew o he epidemiological s udies. Alimen Pha macol
The . 1999;13:851–856.
[27] Palli D, Masala G, Del Giudice G, e al. CagA þHelicobac e pylo i
in ec ion and gas ic cance isk in he EPIC-EURGAST s udy. In J
Cance . 2007;120:859–867.
[28] Knek P, Teppo L, A omaa A, e al. Helicobac e pylo i IgA and IgG
an ibodies, se um pepsinogen I and he isk o gas ic cance :
changes in he isk wi h ex ended ollow-up pe iod. In J Cance .
2006;119:702–705.
[29] Kokkola A, Rau elin H, Puolakkainen P, e al. Diagnosis o
Helicobac e pylo i in ec ion in pa ien s wi h a ophic gas i is:
compa ison o his ology, 13C-u ea b ea h es , and se ology.
Scand J Gas oen e ol. 2000;35:138–141.
1164 I. VOHLONEN ET AL.