Vitamins B2 and B6 and Genetic Polymorphisms Related to One-Carbon Metabolism as Risk Factors for Gastric Adenocarcinoma in the European Prospective Investigation into Cancer and Nutrition

Article (PDF Available)inCancer Epidemiology Biomarkers & Prevention 19(1):28-38 · January 2010with23 Reads
DOI: 10.1158/1055-9965.EPI-08-1096 · Source: PubMed
Abstract
B vitamins and polymorphisms in genes coding for enzymes involved in one-carbon metabolism may affect DNA synthesis and methylation and thereby be implicated in carcinogenesis. Previous data on vitamins B2 and B6 and genetic polymorphisms other than those involving MTHFR as risk factors for gastric cancer (GC) are sparse and inconsistent. In this case-control study nested within the European Prospective Investigation into Cancer and Nutrition cohort, cases (n = 235) and controls (n = 601) were matched for study center, age, sex, and time of blood sampling. B2 and B6 species were measured in plasma, and the sum of riboflavin and flavin mononucleotide was used as the main exposure variable for vitamin B2 status, whereas the sum of pyridoxal 5'-phosphate, pyridoxal, and 4-pyridoxic acid was used to define vitamin B6 status. In addition, we determined eight polymorphisms related to one-carbon metabolism. Relative risks for GC risk were calculated with conditional logistic regression, adjusted for Helicobacter pylori infection status and smoking status. Adjusted relative risks per quartile (95% confidence interval, P(trend)) were 0.85 (0.72-1.01, 0.06) for vitamin B2 and 0.78 (0.65-0.93, <0.01) for vitamin B6. Both relations were stronger in individuals with severe chronic atrophic gastritis. The polymorphisms were not associated with GC risk and did not modify the observed vitamin-cancer associations. In summary, results from this large European cohort study showed an inverse association between vitamin B2 and GC risk, which is borderline significant, and a significant inverse association between vitamin B6 and GC risk.
Research Article
Vitamins B2 and B6 and Genetic Polymorphisms Related to
One-Carbon Metabolism as Risk Factors for Gastric
Adenocarcinoma in the European Prospective
Investigation into Cancer and Nutrition
Simone J.P.M. Eussen
1
, Stein Emil Vollset
1,2
, Steinar Hustad
1,3
, Øivind Midttun
1,4
, Klaus Meyer
1
,
Åse Fredriksen
1
, Per Magne Ueland
1
, Mazda Jenab
5
, Nadia Slimani
5
, Pietro Ferrari
5
, Antonio Agudo
6
,
Núria Sala
7
, Gabriel Capellá
8
, Giuseppe Del Giudice
9
, Domenico Palli
10
, Heiner Boeing
11
, Cornelia Weikert
11
,
H. Bas Bueno-de-Mesquita
12
, Frederike L. Büchner
12
, Fátima Carneiro
13
, Franco Berrino
14
, Paolo Vineis
15,16
,
Rosario Tumino
17
, Salvatore Panico
18
, Göran Berglund
19
, Jonas Manjer
20
, Roger Stenling
21
,
Göran Hallmans
22
, Carmen Martínez
23,24
, Larraitz Arrizola
24,25
, Aurelio Barricarte
24,26
, Carmen Navarro
24,27
,
Laudina Rodriguez
28
, Sheila Bingham
29,30,40
, Jakob Linseisen
31,32
, Rudolf Kaaks
31
, Kim Overvad
33
,
Anne Tjønneland
34
, Petra H.M. Peeters
15,35
, Mattijs E. Numans
35
, Françoise Clavel-Chapelon
36
,
Marie-Christine Boutron-Ruault
36
, Sophie Morois
36
, Antonia Trichopoulou
37
, Eiliv Lund
38
,
Mario Plebani
39
, Elio Riboli
15
, and Carlos A. González
6
Abstract
B vitamins and polymorphisms in genes coding for enzymes involved in one-carbon metabolism may af-
fect DNA synthesis and methylation and thereby be implicated in carcinogenesis. Previous data on vitamins
B2 and B6 and genetic polymorphisms other than those involving MTHFR as risk factors for gastric cancer
(GC) are sparse and inconsistent. In this case-control study nested within the European Prospective Investi-
gation into Cancer and Nutrition cohort, cases (n = 235) and controls (n = 601) were matched for study center,
age, sex, and time of blood sampling. B2 and B6 species were measured in plasma, and the sum of riboflavin
and flavin mononucleotide was used as the main exposure variable for vitamin B2 status, whereas the sum of
pyridoxal 5-phosphate, pyridoxal, and 4-pyridoxic acid was used to define vitamin B6 status. In addition, we
determined eight polymorphisms related to one-carbon metabolism. Relative risks for GC risk were calculat-
ed with conditional logistic regression, adjusted for Helicobacter pylori infection status and smoking status.
Adjusted relative risks per quartile (95% confidence interval, P
trend
) were 0.85 (0.72-1.01, 0.06) for vitamin
B2 and 0.78 (0.65-0.93, <0.01) for vitamin B6. Both relations were stronger in individuals with severe chronic
atrophic gastritis. The polymorphisms were not associated with GC risk and did not modify the observed
vitamin-cancer associations. In summary, results from this large European cohort study showed an inverse
association between vitamin B2 and GC risk, which is borderline significant, and a significant inverse asso-
ciation between vitamin B6 and GC risk.
Cancer Epidemiol Biomarkers Prev; 19(1); 2838. ©2010 AACR.
Introduction
Although the incidence of gastric cancer (GC) has de -
clined in industrialized countries over the last century, it
is still the second most important cause of cancer mortality
in many developing countries (1). Helicobacter pylori (Hp)
infection of the gastric mucosa has been associated with
noncardia GC, whereas severe chronic atrophic gastritis
(SCAG) has been associated with cardia GC (2). The rela-
tively high prevalence of Hp infection in populations with
Cancer
Epidemiology,
Biomarkers
& Prevention
Authors' Affil iations:
1
LOCUS for homocysteine and related v itamins,
Department of Pharmacology, Institute of Medi cine, University of
Bergen, and Haukeland University Hospital;
2
Medical Birth Registry, and
Norwegian Institute of Public Health;
3
Hormone Laboratory, Haukeland
University Hospital;
4
Bevital A/S, Bergen, Norway;
5
IARC-WHO, Lyon,
France;
6
Unit of Nutrition, Environment and Cancer, IDIBELL-Catalan
Institute of Oncology;
7
Unit of Nutrition, Environment and Cancer and
Transitional Laboratory Research-ICO;
8
Transitional Laborator y
Research-ICO, Barcelona, Spain;
9
Research Center, Novartis Vaccines
and Diagnostics, Siena, Italy;
10
Molecular and Nutritional Epidemiology
Unit, Istituto per lo Studio e la Prevenzione Oncologica, Scientific
Institute of Tuscany, Florence, Italy;
11
Department of Epidemiology,
German Institute of Human Nutritio n, Potsdam-Rehbrücke, Germany;
12
National Institute for Public Health and the Environment, Bilthoven, the
Netherlands;
13
Institute of Pathology and Molecular Immunology of the
University of Porto and Medical Faculty/H.S. Joao, Porto, Portugal;
14
Etiologic Epidemiology and Prevention Unit, Fondazione IRCCS
Istituto Nazionale dei Tumori, Milan, Italy;
15
Department of Epidemiology
and Public Health, Imperial College, London, United Kingdom;
16
Department of Biomedical Science, University of Torino, Turin, Italy;
17
Cancer Registr y Azienda Osp edaliera Civi le-M.P. A rezzo, Ragu sa,
Italy;
18
Department of Clinical and Experimental Medicine, Federico II
University, N aples, Italy;
19
Department of Medicine, Lund University;
20
Department of Surgery, University Hospital, Malmö, Sweden;
21
Department of Medical Biosciences, Pathology, and
22
Department of
Public Health and Clinical Medicine, Nutritional Research, Umeå
University, Umeå, Sweden;
23
Andalusian School of Public Health,
Granada, Spain;
24
CIBER Epidemiología y Salud Pública, Madrid, Spain;
25
Department of Public Health of Gipuzkoa, San Sebastian, Spain;
26
Public Health Institute of N avarra, Pamplona, Spain;
27
Department of
Cancer Epidemiol Biomarkers Prev; 19(1) January 201028
low GC risk (3) suggests that also other factors, such as di-
et and genetic predisposition, may play an etiologic role.
One-carbon metabolism (Fig. 1) is of special interest
with respect to carcinogenesis. The activity of enzymes
involved can be affected by low concentrations of B vi-
tamin cofactors and by genetic polymorphisms. De-
rangement of one-carbon metabolism (4) might be
implicated in carcinogenesis via impaired synthesis
and methylation of DNA (5). Folate (6-9) and cobalamin
(9-13) are among the B vitamins that have been studied
most frequently in relation to GC risk and show incon-
sistent associations. Riboflavin (v itamin B2) and pyri-
doxine (vitamin B6) have been studied less frequently.
Only a few case-control studies have investigated asso-
ciations of vitamin B2 (10, 13-21) and B6 (10, 13, 15-18)
intake with GC risk. The results suggest either no asso-
ciation (10, 13-17), an increased (18, 19) or decreased (20,
21) risk with higher vitamin B2 intake, or a decreased
risk with higher vitamin B6 intake (10, 16-18). The vita-
mins B2 and B6 are interrelated because the interconver-
sion of some vitamin B6 species requires the vitamin B2
species flavin mononucleotide (FMN) and flavin dinu-
cleotide as cofactors (22, 23). In one-carbon metabolism,
vitamin B2 serves as a cofactor for the enzymes methyle-
netetrahydrofolate reductase (MTHFR) and methionine
synthase reductase (MTRR), whereas vitamin B6 serves
as a cofactor for the enzyme cystathionine β-synthase
(CBS) and cystathionine γ-lyase. The majority of previ-
ous research on effects of genetic variation and GC risk
has focused on MTHFR 677CT and 1298CA poly-
morphisms (9, 24), whereas few results have been pub-
lished on the CBS 844ins68 (25), MTRR 66AG (26, 27),
methionine synthase (MTR) 2756AG (25, 27), methy-
lenetetrahydrofolate dehydrogenase (MTHFD1)
1958GA (28), and reduced folate carrier [solute carrier
family 19 (SLC19A1)] 80GA (29) polymorphisms. An
association with GC risk has previously been observed
for variant genotypes of the MTHFR 677C T (24) ,
MTHFR 1298C A (9, 24), and the MTHFD1
1958GA (28) polymorphisms.
Most previous studies on vitamins B2 (10, 13-21) and
B6 (10, 13, 15-18) in GC have been relatively small, and
none measured plasma concentrations of vitamin B2 and
B6 species. Moreover, in all previous studies, intake of vi-
tamin B2 and B6 was reported after diagnosis of the dis-
ease, which may have affected the results. To obtain a
clearer understanding of the role of prediagnostic plasma
concentration s of vitamin B2 and B6, and their possible
interactions with eight one- carbon polymorphisms in
GC etiology, this case-control study nested within the Eu-
ropean Prospective Investigation into Cancer and Nutri-
tion (EPIC; refs. 30, 31) was conducted.
Materials and Methods
Study Population and Collection of Blood Samples
The design and methods of the EPIC study have pre-
viously been described in detail (30, 31). Briefly, the EPIC
cohort consists of 23 centers in 10 Euro pean countries
(Denmark, France, Greece, Germany, Italy, Netherlands,
Norway, Spain, Sweden, and United Kingdom). Between
1992 and 1998, country-specific die tary questionnaires,
standardized lifestyle and personal history question-
naires, anthropometric data, and blood samples were
collected from the majority of the participants.
The present study includes GC cases, which were diag-
nosed after blood collection, and matched control cohort
members free of cancer from all EPIC countries. In each of
the recruitment centers, fasting and nonfasting blood
samples of at least 30 mL were drawn from all participants
and stored at 5°C to 10°C while protected from light and
transported to local laboratories for processing and ali-
quoting, as previously described (30, 31), except for the
EPIC-Oxford center. The EPIC-Oxford center collected
blood s amples from a network of general practitioners
and health conscious individuals, and blood samples
were transp orted to a central laboratory in Oxford via
mail. While protected from light, the whole blood samples
from Oxford were exposed to ambient temperatures for
up to 48 h. As B vitamins are partly degraded by such han-
dling, all EPIC-Oxford (three cases, nine controls) samples
were excluded from the present analyses. GC cases were
not diagnosed among the Norwegian cohort members.
In all countries, except Denmark and Sweden, blood
was separated into 0.5 mL fractions (serum, plasma, red
cells, and buffy coat for DNA extraction). Each fraction
was placed into straws, which were heat sealed and
stored in liquid nitrogen (196°C). One half of all ali-
quots were stored at the local study center and the other
half in the central EPIC biorepository at the IARC
(Lyon, France). In Denmark, blo od fraction aliquots of
Epidemiology, Health Council of Murcia, Murcia, Spain;
28
Public Health
and Health Planning Directorate, Asturias, Spain (Dirección General de
Salud Pública, Consejería de Salud y Servicios Sanitarios Asturi as,
Asturias, Spain);
29
MRC Dunn Human Nutrition Unit;
30
MRC Centre for
Nutritional Epidemiology in Cancer Prevention and Survival, Department
of Public Health and Primary Care, University of Cambridge, Cambridge,
United Kingdom;
31
Division of Cancer Epidemiology, German Cancer
Research Center, Heidel berg, Germany;
32
Institute of Epidemiology,
Helmholtz Zentrum München, German Research Centre for Environmental
Health, Neuherberg, Germany;
33
Department of Clinical Epidemiology,
Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark;
34
Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen,
Denmark;
35
Julius Center for Healt h Scienc es and Pri mary Care ,
University Medical Center, Utrecht, the Netherlands;
36
Institut National de
la Santé et de la Recherche Médicale, ERI 20, EA 4045, and Insti tut
Gustave Roussy, Villejui f, France;
37
Department of Hygiene and
Epidemiology, Medical School University of Athens, Athens, Greece;
38
Institute of Community Medicine, University of Tromsø, Tromsø,
Norway;
39
Department of Laboratory Medicine, A zienda Ospedaliera,
Università di Padova, Italy;
40
Deceased.
Corresponding Author: Simone J.P.M. Eussen, Section for Pharmacol-
ogy, Department of I nternal Medicine , University of Bergen, 5021
Laboratory Building, 9th Floor, Bergen, Norway. Phone: 47-55975786;
Fax: 47-55974605. E-mail: Simone.Eussen@farm.uib.no
doi: 10.1158/1055-9965.EPI-08-1096
©2010 American Association for Cancer Research.
Plasma Vitamin B2 and B6, SNPs, and Gastric Cancer
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010www.aacrjournals.org 29
1.0 mL were stored locally in Nunc tubes at 150°C
under nitrogen vapor. In Sweden, samples were stored
in 70°C freezers.
Follow-up for Cancer Incidence
In EPIC, follow-up is based on population cancer reg-
istries (Denmark, Italy, Netherlands, Norway, Spain, Swe-
den, and the United Kingdom), health insurance records,
pathology registries, and active contact of study subjects
or next of kin (France, Germany, and Greece). The follow-
up period for the present study was for cases included in
reports received at IARC until the end of October 2002,
representing complete follow-ups until either December
2000 or December 2001 for all centers using cancer regis-
try data and until 2002 for France, Germany, and Greece.
Cancers of the stomach included cancers coded as C16
(10th Revision of the International Statistical Classifica-
tion of Diseases, Injury and Causes of Death). The diag-
nosis, tumor site classification, and morphology
(according to ICD02 and Lauren classifications) of each
identified cancer were confirmed and validated by an in-
dependent panel of pathologists with a repre sentative
from each EPIC country and a coordinator (32). The pa-
thologist panel reviewed original histologic slides and/or
recuts from the paraffin blocks and original histopathol-
ogy reports that were provided by each EPIC center.
Nested Case-Control Study Design and Selection of
Study Subjects
Incident GC cases were cohort members who deve l-
oped cancer after recruitment into EPIC. The present
study includes a total of 221 gastric adenocarcinomas
and 14 adenocarcinomas of the gastroesophageal junc-
tion (GEJ). In this study, these 235 cases are grouped to-
gether. For each i dentified cancer case, cont rol subjects
with available blood samples were randomly selected
from all cohort members who were alive and free of can-
cer (except nonmelanoma skin cancer) at the time of di-
agnosis of the case patient. Controls (n =601)were
matched by gender, age group (±2.5 y), study center,
and date of blood sample collection (±45 d). GC cas es
were divided into three groups according to anatomic
subsite: (a) tumors originating from the gastric cardia
(n = 62), combining tumors that reached the GEJ, either
crossing the GEJ or located below the GEJ (all 14 GEJ can-
cers)ornot;(b) noncardial tumors (n = 102) grouping
cases from other sites in the stomach; and (c) tumors from
unknown or mixed sites (n = 71). When divided by his-
tologic subtype, of the 235 cancer cases, 82 were classified
as diffuse and 8 1 as intestinal according to the Lauren
classification. The remaining cases (n = 72) were of un-
known or mixed histologic types. All gastric lymphomas,
gastric stump cancers, other gastric nonadenocarcinoma,
esophageal nonadenocarcinomas, and otherwise unspec-
ified malignant neoplasms of the stomach were excluded
from this analysis. This study was approved by the Eth-
ical Review Board of the IARC and those of all individual
EPIC centers.
Laboratory Measurements
Vitamin B2 measures included citrate plasma concen-
trations of riboflavin and FMN, and vitamin B6 measures
included pyridoxal 5-p hosphate (PLP), pyridoxal (PL),
and 4-pyridoxic acid (PA). All vitamin species were de-
termined by liquid chromatographytandem mass spec-
trometry assay in the same laboratory in Bergen, Norway
(33). Within- and between-day c oefficients of variation
(CV) of vitamin B2 and B6 species were 3% to 18% and
6% to 22%, respectively. The largest CVs are for FMN
(within CV = 18%; between CV = 22%), which is present
in low concentrations and has a relatively low signal (33).
Eight polymorphisms (reference, am ino acid change)
of genes coding for enzymes involved in one-carbon me-
tabolism were determined by matrix-assisted laser de-
sorption/ionization time-of-flight mass spectrometry
(34) in Bergen, Norwa y. These included CBS (699CT;
rs 234706, Tyr
233
Tyr) and, in additio n, the CBS 844ins68
insertion, MTHFD1 (1958GA; rs 2236225, Arg
653
Gln),
MTR (2756A G; rs 1805087, Asp
919
Gly),MTRR
(66AG; rs 18013 94, Ile
22
Met), SLC19A1 (80G A; rs
1051266, Arg
27
His), and transcobalamin II (TCN2
67A G; rs RsaI, Ile
23
Val and TCN2 776C G; rs
1801198, Pro
259
Arg). These single-nucleotide polymorph-
isms (SNP) are nonsynonymous and are often called
potentially functional (35). Quantification of anti-Hp
(Hp) antibodies in stored plasma samples was done by
an ELIS A technique using incubation with lysates of
the CCUG Hp strain (36). Pepsinogen A (PGA) was as-
sayed in plasma by a commercial microplate-based quan-
titative ELISA kit (Biohit). SCAG was serologically
defined as a PGA circulating concentration lower than
22 μg/L (36).
Statistical Methods
Because riboflavin and FMN are interconvertible (37-
40), as are PLP and PL (41, 42), and PA is formed from
PL, we calculated the sum of riboflavin and FMN as an
index for vitamin B2 status and the sum of PLP, PL, and
PA as an index of vitamin B6 status. In the present study,
these sums of B2 and B6 vitamers are therefore consid-
ered as the main exposure variables for vitamin B2 and
B6 status, respectively. In addition, results are also pre-
sented for the individual B2 and B6 species.
The Mann-Whitney U test was used to assess potential
differences in plasma concentrations of vitamin B2 an d
B6 species according to sex, age (<60 y versus >60 y), Eu-
ropean region (north versus central), Hp infection status,
and SCAG status, whereas the Kruskall-Wallis test was
used to assess differences in concentration among catego-
ries of smoking categories (nonsmokers, ex-smokers,
current smokers, and missing) in the 601 controls. Differ-
ences in plasma concentrations between cases and con-
trols were investigated by the Mann-Whitney U test.
Relative risks (RR) and 95% confidence intervals (95%
CI) for GC in relation to indices of vitamin B2 and B6
status were calculated by conditional logistic regression
using the SAS LOGISTIC procedure (SAS statistical
Eussen et al.
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010 Cancer Epidemiology, Biomarkers & Prevention30
software, version 9.1; SAS Institute) stratified by the case-
control set. In our study, the RR indicated the incidence
rate ratio, which is reflected by the odds ratio calculated
from conditional logistic regressi on (43). Risk esti mates
were adjusted for Hp infection status and smoking cate-
gories. The RRs on GC were examined by quartiles with
cutoff points based on the distribution of B2 and B6 indi-
ces in all 601 controls combined. Likelihood ratio tests
were used to assess linear trends in RRs across the catego-
ries using values for quartile categories as the quantitative
score of exposure. Models were also carried out separate-
ly for each anatomic subsite (cardia versus noncardia),
histologic subtype (diffuse versus intestinal), European
region [north-central Europe (Sweden, Denmark, Ger-
many, t he Netherlands, United Kingd om, an d France)
versus south ern Europe (Italy, Greece and Spain)], time
from blood donation to cancer diagnosis (<1 y versus
1 y), age at recrui tment (<60 y versus >60 y), sex, and
Hp status (infection or no infection). Heterogeneity was
tested by adding the product term of the vitamin indices
and pote ntial effect modifiers in the model. In addition,
subgroup analyses by SCAG (yes or no) were done for
cases with and without SCAG and their matched controls.
The as sociations between the polymorphisms and G C
risk were studied with conditional logistic regression but
by stratifying on country instead of the matched sets and
with age and sex as covariates. The risk estimates were cal-
culated with the wild-type as the referenc e category. An
ordinal level variable with equally spaced integer weights
(0, 1, 2) for the genotypes was used to test for trend to sum-
marize the effect of each polymorphism. Effect modifica-
tion of the SNP-GC associations by vitamin concentrations
was studied with conditional logistic regression.
Results
Characteristics of the Study Population
Selected characteristics of the 235 GC cases and 601
matched controls are summarized in Tabl e 1. Forty-one
percent of the cases were female, mean age at diagnosis
was 62 years, and the mean time from blood donation to
cancer diagnosis was 3.2 years. GC cases had a higher
prevalence of Hp infection, SCAG, and current smoking
than their matched controls.
Concentrations of Vitamins B2 and B6
The distributions of vitamin B2 and B6, and the indi-
vidual vitamers, in the control group were skewed, with
a longer tail at higher concentrations ( Table 2). Plasma
concentrations of all vitamin B6 species correlated strong-
ly with each other after adjustment for age, sex, and
study center (correlation coefficients ranged f rom 0.51
to 0.74; P for all correlations < 0.01), and the correlation
between plasma concentrations of riboflavin and FMN
was 0.32 (P < 0.01; data not shown).
Table 2 shows that median vitamin B2 concentrations
were higher in fe males, whereas vitamin B6 co ncentra-
tions were higher in males. Furthermore, concentrations
of vitamin B2 and B6 were lower in southern European
countries compared with northern European c ountries
(P
difference
< 0.01) and in smokers compared with former
and never smokers (P
trend
< 0.01). Concentrations did not
differ with respect to age (<60 y versus 60 y), Hp infec-
tion (yes versus no), and SCAG (yes versus no) in con-
trols (Table 2). A trend toward higher concentrations of
all B6 species was observed for the homozygote GG var-
iants of the MTRR (P
trend
< 0.05) and TCN2 (P
trend
< 0.05)
polymorphisms, whereas concentrations of the B2 and B6
species did not differ across o ther polymorphic genetic
variants (data not shown).
Median concentrations (nmol/L) of the vitamin B2
sum, riboflavin, FMN, t he vitamin B6 sum, PLP, PL,
and PA in the GC cases were 15.6, 12.1, 2.9, 51.9, 22.1,
13.4, and 15.6 nmol/L, respectively. Furthermore, con-
centrations did not differ significant ly among cases and
controls with Hp infection (P
difference
> 0.05) and SCAG
(P
difference
> 0.05; data not shown).
Associations between Plasma Indices of Vitamin B2
and B6 Status with GC
GC risk decreased borderline s ignificantly with in-
creasing vitamin B2 sum (RR per increase in quartile of
0.85; P
trend
= 0.06) and with strongest associations for
FMN (P
trend
= 0.01). The vitamin B6 sum was significant-
ly inversely associated with GC risk (RR per increase in
quartile of 0.78; P
trend
< 0.01), and among the separate B6
species, only high PA concentrations were significantly
associated with lower GC risk (P
trend
= 0.01; Table 3). Ex-
clusion of cohort members with extremely high concen-
trations of the B6 sp ecies, attributable to vitamin B6
supplements as indicated by the presence of pyridoxine
in plasma (n = 81; ref. 44), did no t alter any of the ob-
served associations (data not shown).
We also analyzed the associations between vitamin B2
and B6 indices with GC risk separately in cases with (n =
42) and without (n = 191) SCAG (Table 4). Generally, as-
sociations of the vitamins with GC risk were more pro-
nounced in those with SCAG compared with those
without SCAG. Associations were stronger, with RR/
quartile (P
trend
) of 0.38 (<0.01) for FMN and 0.51 (0.01)
for PA. Moreover, the difference in associations in those
with and without SCAG was significant for FMN (P
inter-
action
= 0.03; Table 4). Because gastritis was not a match-
ing criterion, 2% of controls matched to cases without
SCAG had SCAG, and 32% of controls matched to cases
with SCAG had SCAG, which may have biased these as-
sociations. Therefore, we additionally analyzed these as-
sociations stratified for gastritis comparing cases and
controls with or without SCAG. These analyses revealed
even stronger inverse associations between vitamin B2
and B6 and GC in individuals with SCAG (data not
shown).
We further studied whether the association s between
GC and vitamin B2 and B6 were modified by Hp infec-
tion status (yes or no), time between blood sampling
and cancer diagnosis (<1st year or 1st year), sex, age
Plasma Vitamin B2 and B6, SNPs, and Gastric Cancer
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010www.aacrjournals.org 31
(<60 y or 60 y), anatomic subsite (cardia or noncardia),
histologic subtype (diffuse or intestinal), or European re-
gion (north or south), and associations between vitamin
B2 and B6 and GC risk were not modified by these fac-
tors (data not shown).
Polymorphisms in Enzymes with Vitamin B2 and
Vitamin B6 as Cofactors and Their Association
with GC
All the SNPs in MTHFR, MTRR, MTR, and CBS genes
were in Hardy-Weinberg equilibrium (P >0.05,χ
2
test,
for all SNPs). In this EPIC cohort, the MTHFR 677CT
polymorphism was not associated with GC risk, whereas
the variant CC genotype of the MTHFR 1298AC poly-
morphism showed a 47% increased GC risk compared
with the AA genotype (P = 0.04), as presented elsewhere
(9). The genotype distributions did not differ among
cases and controls (Table 5). Overall, homozygote variant
genotypes of all polymorphisms were more prevalent
in northern European countries compared with southern
European countries (P < 0.03 for all SNPs), with most
pronounced differences for MTRR 66AG (31% in north
versus 20% in south). None of the MTRR, MTR,and
CBS polymorphisms was associated with GC risk (P
trend
for all > 0.05; Table 5). However, European region signif-
icantly modified overall associations for the CBS 844ins68
(P
interaction
= 0.03), showing opposite associations in
northern and southern European countries (Table 5).
We further assessed potential effect modification by vi-
tamin B2 and vitamin B6 status of t he associations be-
tween cancer and the MTHFR, MTRR, MTR,andCBS
polymorphisms. None of the associations between the
SNPs and GC risk was statistically significantly modified
by variable vitamin B2 and vitamin B6 status (P
interaction
>
0.06 for all relevant interactions; data not shown).
Polymorphisms in Enzymes with Folate and Vitamin
B12 as Cofactors and Their Association with GC
Overall, homozygote variant genotypes of the TCN2
776C G, TCN2 67A G, MTHFD1 1958G A,and
SLC19A1 180GA polymorphisms were more prevalent
in northern European countries than in southern Europe-
an countries (P < 0.03 for all SNPs), with most pro-
nounced differences for MTHFD1 1958GA (13% in
north versus 6% in south). None of these polymorphisms
was associated with GC risk (P
trend
for all > 0.05). Howev-
er, GG variant of the TCN2 776 gene showed an increased
risk in northern European countries and a decreased risk
in southern countries (P
interaction
=0.04),whereasthe
AA variant of the SLC19A1 gene showed an increased
risk in southern and decreased risk in northern countries
Figure 1. One-carbon metabolism and related enzymes and genetic
polymorphisms. BHMT, betaine homocysteine methyltransferase
(remethylation of homocysteine to methionine); Cysta, cystathionine;
Cys, cysteine; CH
2
THF, methylenetetrahydrofolate; CH
3
THF,
methyltetrahydrofolate; DMG, dimethylglycine; Hcy, homocysteine;
Hcy-tl, homocysteine thiolactone; Met, methionine; THF, tetrahydrofolate.
Table 1. Characteristics of cohort members
Cases (n = 235) Controls (n = 601)
Sex (female), n (%) 96 (41) 245 (41)
Age at recruitment, mean (range) 58.9 (31.8-76.3) 58.7 (28.5-76.6)
Age at diagnosis, mean (range) 62.1 (34.3-77.7) n.a.
Years between blood donation and diagnosis, mean (range) 3.2 (0.01-9.69) n.a.
Infection with Helicobacter pylori, n (%) 200 (84) 404 (67)
Severe atrophic chronic gastritis, n (%)* 42 (18) 45 (8)
Current/former smokers, n (%) 67 (28)/88 (37) 136 (23)/198 (33)
Abbreviation: n.a., not applicable.
*Defined as plasma PGA concentrations <22 μg/L.
Eussen et al.
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010 Cancer Epidemiology, Biomarkers & Prevention32
(P
interac tion
= 0.05). None of the associations between the
SNPs and GC risk was statistically significantly modified
by folate and vitamin B12 status (P
interaction
> 0.19 for all
relevant interactions; data not shown).
Discussion
The present nested case-control study investigated the
association of GC risk with indices of vitamin B2 and B6
status, and nine polymorphisms in genes encoding for
enzymes involved in one-carbon metabolism. Overall
analyses indicated a tendency for an inverse association
between vitamin B2 and GC risk and a significant inverse
association between vitamin B6 and GC risk. Further-
more, the associations were more pronounced in indivi-
duals with SCAG. None of the studied polymorphisms
was related to GC risk, nor did they modify the observed
associations for vitamin B2 and B6.
Table 2. Vitamins B2 and B6 [mean/median (5-95 percentile)] in relation to demographic characteristics
and risk factors of GC in control cohort members (n = 601)
n Vitamin B2 sum Riboflavin FMN Vitamin B6 sum PLP PL PA
Overall 601 23.9/16.6
(7.6-59.1)
18.8/12.8
(5.0-46.0)
5.1/3.2
(1.2-12.6)
89/56
(28-174)
32.4/23.5
(8.5-78.0)
23.6/14.5
(7.2-36.9)
33.0/16.4
(8.7-66.5)
Sex Male 356 24.1/15.3
(7.7-55.6)
18.7/12.2
(5.0-43.3)
5.4/2.9
(1.2-13.2)
98/59
(30-178)
34.2/26.2
(9.0-79.8)
26.2/14.7
(7.4-36.9)
37.4/17.1
(8.4-71.4)
Female 245 23.6/18.5
(7.6-61.6)
18.9/14.6
(5.3-50.8)
4.7/3.5
(1.2-12.2)
76/55
(26-164)
29.7/20.6
(7.3-71.3)
20.0/13.8
(7.2-37.0)
26.7/16.0
(8.9-60.9)
P
difference
* <0.01 <0.01 0.08 <0.01 <0.01 0.16 0.13
Age <60 y 289 23.9/15.6
(7.5-62.9)
18.8/12.0
(5.1-50.8)
5.1/3.2
(1.2-13.2)
97/54
(27-148)
33.8/23.9
(8.9-73.1)
27.6/14.2
(7.1-33.2)
35.9/15.1
(7.8-51.9)
60 y 312 23.8/17.6
(7.7-51.5)
18.8/13.5
(5.0-44.7)
5.1/3.0
(1.1-11.4)
81/58
(28-185)
31.1/23.4
(7.9-78.0)
20.0/14.6
(7.4-38.2)
30.3/18.7
(9.3-76.8)
P
difference
* 0.12 0.06 0.41 0.05 0.92 0.23 <0.01
European
region
North 338 27.5/19.0
(9.3-60.6)
22.1/14.9
(6.8-49.0)
5.4/3.2
(1.3-14.4)
113/64
(32-236)
37.8/27.2
(10.0-90.5)
30.0/15.7
(8.0-51.1)
44.9/20.8
(9.6-97.8)
South 263 19.3/14.4
(6.7-47.7)
14.6/10.7
(4.2-38.9)
4.6/3.2
(1.2-12.1)
59/48
(25-101)
25.5/20.0
(7.0-55.6)
15.5/12.9
(6.7-24.5)
17.8/14.1
(7.8-28.3)
P
difference
* <0.01 <0.01 0.87 <0.01 <0.01 <0.01 <0.01
Smoking
status
Never 256 25.5/18.2
(7.6-58.0)
20.7/13.8
(5.4-48.2)
4.7/3.4
(1.4-13.2)
103/59
(28-189)
33.4/24.7
(8.5-79.8)
29.0/15.1
(7.2-35.4)
40.5/17.2
(9.7-59.4)
Former 198 27.8/18.0
(7.8-69.5)
21.1/13.8
(5.8-53.2)
6.7/3.3
(1.2-17.0)
89/59
(33 -174)
35.4/26.7
(9.0-82.4)
21.5/14.8
(7.8-37.3)
31.8/17.6
(9.9-76.8)
Current 136 15.7/13.1
(6.3-39.0)
12.2/9.9
(4.3-30.2)
3.5/2.7
(0.9-8.0)
65/47
(24-148)
26.3/19.1
(7.3-62.5)
17.5/12.8
(6.6-38.2)
21.4/13.7
(6.5-52.4)
Unknown 11 15.7/14.2
(5.4-26.9)
12.1/11.7
(4.4-20.5)
3.6/2.4
(0.7-8.1)
69/51
(21-232)
29.9/20.1
(5.5-89.7)
15.3/11.6
(6.5-54.7)
23.5/16.1
(8.9-87.7)
P
trend
<0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01
Hp infection No 196 25.0/16.5
(7.4-69.5)
19.1/12.8
(5.1-53.5)
5.8/2.8
(1.3-16.3)
107/57
(31-168)
32.8/25.1
(9.4-78.0)
30.7/14.3
(7.4-35.3)
43.7/18.1
(9.6-71.4)
Yes 404 23.3/16.6
(7.7-51.5)
18.6/12.8
(5.0-43.8)
4.7/3.3
(1.1-12.1)
80/56
(27-185)
32.1/22.3
(7.6-78.0)
20.2/14.6
(7.2-38.2)
27.8/16.1
(8.2-66.5)
P
difference
* 0.71 0.57 0.18 0.36 0.30 0.70 0.13
SCAG No 548 24.5/16.9
(7.6-60.5)
19.3/12.9
(5.1-48.2)
5.2/3.2
(1.2-13.0)
91/56
(28-172)
32.7/23.9
(8.9-72.1)
24.3/14.5
(7.4-36.5)
34.2/16.5
(8.4-64.8)
Yes 45 17.2/15.0
(7.9-31.8)
13.3/11.0
(4.4-28.4)
3.7/3.0
(0.9-8.2)
68/48
(26-232)
29.2/18.8
(7.0-110.0)
16.6/13.2
(7.0-50.7)
22.3/15.0
(9.1-76.3)
P
difference
* 0.12 0.11 0.20 0.08 0.08 0.24 0.35
*P for difference based on median concentrations.
North: Scandinavia, United Kingdom, and Central Europe; south: Italy, Spain, and Greece.
Category unknown not included in the test for trend.
Plasma Vitamin B2 and B6, SNPs, and Gastric Cancer
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010www.aacrjournals.org 33
Study Design
An important strength of our study is the availability
of blood samples before cancer diagnosis. One could ar-
gue that a mean lag time of 3 years between blood dona-
tion and cancer diagnosis may be too short to o btain
unbiased associations between the vitamins and GC. Vi-
tamin concentrations measured shortly before diagnosis
should be considered with caution because low vitamin
status may be a consequence of malignancy rather than
causing it. In this regard, we previously observed that co-
balamin was inversely associated with GC risk, particu-
larly among individuals with severe atrophic chronic
gastritis (9). This agrees w ith a reduced bioavailabili ty
of cobalamin due to infection with H. pylori and atrophic
gastritis (45), which may precede GC (46). Notably, in the
present study, the inverse association between cancer risk
and concentrations of vitamin B2 and B6 was strongest in
cases with SCAG. This may suggest that impaired vita-
min B2 and B6 status contributes to the development of
GC or that atrophic gastritis may adversely affect the sta-
tus of these vitamins. In line with the l atter possibility,
redu ced PLP concentration s are consistently assoc iated
with inflammation (47-49). In the present study, vitamin
B2 and B6 concentrations did not differ significantly from
people with and without SCAG, and correction for low
pepsinogen concentrations did not confound the associa-
tion between the vitamins and GC. It should be noted
that low PGA concentrations are a marker for severe gas-
tritis (50). Results of the present study do therefore only
apply to severe gastritis and not for milder forms of gas-
tritis. Previous studies on the association of the vitamins
B2 (10, 13-21) and B6 (10, 13, 15-18) with GC investigated
the effects of dietary intake rather than plasma concentra-
tions. Blood sampling is less likely to result in bias due to
measurement errors that are inherent to food frequency
questionna ires. Within this multicenter study, all study
centers collected and stored blood samples according to
a standardized protocol (31), and all biochemical analy-
ses were carried out in one laboratory, thereby optimiz-
ing sample treatment during analysis. Moreover, the
present study included plasma concentrations of several
B2 and B6 species. Although individual vitamin B2 and
B6 species reflect vitamin B2 (51) and B6 (47) status, there
is no clear consensus as to which vi tamin specie is the
best marker. We also used the sum of the individual spe-
cies as indices of vitamin B2 and B6 status because the B2
species (37-40) and the B6 species (41, 42) are intercon-
vertible, and the plasma concentrations are positively
correlated, as shown here and by others (44).
Vitamin B2 and B6 Intake and GC
The most important dietary sources of vitamin B2 are
milk and dairy products (52). Contrary to our results
showing an inverse association of plasma v itamin B2
with GC risk, other studies o bserved that vitamin B2
(18, 19) and milk intake (15, 53) are associated wi th in-
creased risk. Although information on dairy intake was
not taken into account in the present study, it is possible
that milk intake is increased to reduce gastric pain during
disease progression. Vitamin B6 may be obtained from
various food groups, including fruit, vegetables, and
meat (41), and once ingested, all vitamin B6 species are
converted into PLP and PL (41, 42). No strong overall ev-
idence was obse rved for a protective role of fruit and
veget able intake in GC in our study population (5 4) or
in other cohort studies (55, 56). In contrast, meat intak e
has been found to be associated with increased risk of
the noncardia subsite of GC in the EPIC cohort (57).
Mechanism
Studies in ribofl avin-d eficient rats show an increased
GC risk, increased induction of DN A repair enzymes
(58), and increased carcinogen binding to DNA (59) com-
pared with rats on a riboflavin-replete diet. Mo reover,
Table 3. RRs (95% CI) for GC risk by quartiles of plasma vitamins B2 and B6 indices
Case/control Q1* Q2 Q3 Q4 RR/quartile P
trend
Vitamin B2 sum 231/590 1 0.78 (0.48-1.27) 0.71 (0.43-1.17) 0.61 (0.36-1.03) 0.85 (0.72-1.01) 0.06
Riboflavin 231/591 1 1.02 (0.64-1.64) 0.73 (0.44-1.23) 0.71 (0.42-1.19) 0.87 (0.74-1.03) 0.09
FMN 231/590 1 1.18 (0.76-1.85) 0.64 (0.38-1.08) 0.55 (0.32-0.96) 0.79 (0.67-0.95) 0.01
Vitamin B6 sum 233/596 1 0.70 (0.44-1.11) 0.59 (0.35-0.98) 0.46 (0.26-0.80) 0.78 (0.65-0.93) 0.006
PLP 235/600 1 0.82 (0.51-1.30) 0.61 (0.36-1.05) 0.64 (0.36-1.13) 0.85 (0.71-1.02) 0.08
PL 234/599 1 1.02 (0.65-1.60) 0.56 (0.33-0.95) 0.74 (0.44-1.26) 0.86 (0.72-1.02) 0.07
PA 234/599 1 0.87 (0.56-1.35) 0.81 (0.50-1.30) 0.48 (0.28-0.83) 0.81 (0.68-0.96) 0.01
NOTE: Calculated by conditional logistic regression, stratified by the case-control set and adjusted for Hp infection status and
smoking. Controls were matched to cases by age, sex, study center, and date of blood sample collection. The cutoff values for
the quartiles were as follows: vitamin B2 sum, 11.88, 16.63, and 25.15 μ mol/L; riboflavin, 8.52, 12.80, and 20.00 μmol/L; FMN, 2.13,
3.15, and 5.21 μmol/L; vitamin B6 sum, 40.8, 55.7, and 80.6 μmol/L; PLP, 14.90, 23.50, and 36.60 μmol/L; PL, 10.60, 14.50, and
18.97 μmol/L; and PA, 12.60, 16.40, and 24.90 μmol/L.
*Reference category.
Eussen et al.
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010 Cancer Epidemiology, Biomarkers & Prevention34
some carcinogens are metabolized by flav in-dependent
enzymes, and vitamin B2 status may modify effects of
the carcinogen (58). PLP catalyzes 100 essenti al enzy-
matic reactions in human metabolism (60). Alth ough it
is unclear whether low vitamin concentrations are a
cause or a consequence of the disease, low PLP concen-
trations have been associated with high concentrations of
inflammatory markers (47-49) , which may explain our
finding of stronger associations among those cases with
SCAG. Furthermore, low PLP concentrations cause a de-
creased enzyme activity of serine hydroxymethyltrans-
fer ase, which results in a lack of me thylene groups for
5,10-methylenetetrahydrofolate p roduction (61). This
may lead to increased chromosome s trand breaks (62),
impaired DNA repair (63, 64), and DNA hypomethyla-
tion (65, 66), which has been observed in different types
of tumors.
The majority of previous studies on genetic variations
related to one-carbon metabolism in GC focused on the
MTHFR gene and were mainly conducted in Asian po-
pulations (24). The overall result of a meta-analysis
shows that in East Asian populations, but not in Cauca-
sians, the risk of GC increases with the number of T al-
leles of the MTHFR 677C T polymorphism (24).
Publ ished results for this EPIC population showed no
association between the MTH FR 677C T polymor-
phism and GC (9), and the null findings in the present
report are in line with previous studies investigating
polymorphisms of the MTRR 66AG (26, 27), MTR
2756A G (27), and SLC19A1 180GA (29) gen es, but
not with one study showing increased GC risk with var-
iant genotypes of the MTHFD1 1958GA (28) polymor-
phism. Notably, we observed opposite trends for the
CBS 844ins68, the TCN2 776, and the SLC19A1 180 poly-
morphisms in northern and southern European
countries. Furthermore, none of the polymorphisms
modified the associations of the vitamins B2 and B6
with GC risk. However, sample sizes might have been
too small to detect genetic associations with GC risk, in-
teractions between genes and European region, and in-
teractions between genes and vitamins and might have
resulted in chance findings.
In summary, this large prospective European multicen-
ter study revealed that higher concentrations of vitamin
B2 suggest to decrease GC risk, whereas vitamin B6 were
associated with a decreased GC risk, with more pro-
nounced associations in a subgroup with SCAG. Further-
more, none of the polymorphisms related to one-carbon
metabolism was associated with GC risk, nor did they
modify the associations of vitamin B2 and B6 with this
type of cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Table 4. RRs (95% CI) for GC risk by quartiles of plasma concentrations of vitamin B2 and B6 indices in
cases with and without SCAG
SCAG* Case/control Q1
Q2 Q3 Q4 RR/quartile P
trend
P
interaction
Vitamin
B2 sum
Yes 42/106 1 1.10 (0.32-3.85) 0.46 (0.12-1.78) 0.31 (0.07-1.39) 0.64 (0.40-1.04) 0.07
No 187/474 1 0.76 (0.44-1.31) 0.76 (0.44-1.33) 0.77 (0.43-1.37) 0.92 (0.77-1.11) 0.40 0.23
Riboflavin Yes 42/106 1 1.49 (0.47-4.79) 0.61 (0.15-2.50) 0.33 (0.07-1.45) 0.65 (0.41-1.02) 0.06
No 187/475 1 0.91 (0.54-1.55) 0.74 (0.42-1.30) 0.89 (0.50-1.58) 0.94 (0.78-1.13) 0.53 0.21
FMN Yes 42/106 1 0.66 (0.19-2.33) 0.11 (0.02-0.61) 0.06 (0.01-0.44) 0.38 (0.20-0.73) <0.01
No 187/474 1 1.25 (0.77-2.02) 0.79 (0.45-1.39) 0.73 (0.41-1.32) 0.87 (0.73-1.05) 0.16 0.03
Vitamin
B6 sum
Yes 41/105 1 0.44 (0.14-1.42) 0.57 (0.15-2.09) 0.20 (0.04-0.94) 0.64 (0.40-1.03) 0.07
No 190/481 1 0.74 (0.44-1.25) 0.58 (0.32-1.05) 0.52 (0.28-0.96) 0.80 (0.66-0.98) 0.03 0.78
PLP Yes 42/106 1 0.84 (0.28-2.56) 0.55 (0.15-2.03) 0.50 (0.13-1.96) 0.78 (0.50-1.22) 0.27
No 191/484 1 0.82 (0.49-1.38) 0.63 (0.34-1.15) 0.65 (0.34-1.24) 0.86 (0.69-1.05) 0.14 0.95
PL Yes 42/105 1 0.80 (0.25-2.55) 0.97 (0.29-3.22) 0.50 (0.11-2.19) 0.85 (0.54-1.33) 0.47
No 190/484 1 1.09 (0.65-1.80) 0.47 (0.26-0.87) 0.81 (0.44-1.46) 0.86 (0.71-1.04) 0.12 0.59
PA Yes 42/105 1 0.46 (0.16-1.27) 0.35 (0.10-1.20) 0.10 (0.02-0.62) 0.51 (0.30-0.86) 0.01
No 190/484 1 1.04 (0.63-1.71) 1.02 (0.60-1.74) 0.63 (0.35-1.15) 0.88 (0.73-1.06) 0.19 0.15
NOTE: RR (95% CI) calculated by conditional logistic regression, stratified by the case-control set and adjusted for Hp infection
status and smoking. Of the 45 matched controls with SCAG, 32% (34 of 106) were matched to cases with SCAG and 2% (11 of
484) to cases without SCAG. The cutoff values for the quartiles were as follows: vitamin B2 sum, 11.88, 16.63, and 25.15 μmol/L;
riboflavin, 8.52, 12.80, and 20.00 μmol/L; FMN, 2.13, 3.15, and 5.21 μmol/L; vitamin B6 sum, 40.8, 55.7, and 80.6 μmol/L; PLP,
14.90, 23.50, and 36.60 μmol/L; PL, 10.60, 14.50, and 18.97 μmol/L; and PA, 12.60, 16.40, and 24.90 μmol/L.
*SCAG defined as PGA concentrations <22 μg/L.
Reference category.
Plasma Vitamin B2 and B6, SNPs, and Gastric Cancer
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010www.aacrjournals.org 35
Acknowledgments
We thank the members of the pathologist panel for their
valuable work: Dr. Johan Offerhaus (Amsterdam, the
Netherlands), Dr. Vicki Save (Cambridge, United
Kingdom), Dr. Julio Torrado (San S ebastian, Spain),
Dr. Gabriella Nesi (Firenze, Italy), Dr. U. Mahlke
(Potsdam, Germany), Dr. Hendrik Bläker (Heildelberg;
Germany),andDr.ClausFenger(Denmark).Wethank
Dr. Dimitrious Roukos (Ioannina, Greece) for his
contribution to the collection of pathologic material and
Catia Moutinho (Porto, Portugal) for her technical work in
the preparation of pathologic material.
Grant Support
European Commission FP5 project (QLG1-CT-2001-
01049). The EPIC study was funded by Europe Against
Cancer Programme of the European Commission; Ligue
contre le Cancer (France); Société 3M (France); Mutuelle
Générale de l'Education Nationale; Institut National de la
Santé et de la Recherche Médicale; German Cancer Aid;
German Cancer Research Center; German Federal Minis-
try of Education and Research; Danish Cancer Society;
Health Research Fund of the Spanish Ministry of Health
[RCESP-C03/09; RTICCC (C03/10)]; the participating
regional govern ments and inst itutions of Spain; Cancer
Research UK; Medical Research Council, UK; Stroke
Association, UK; British Heart Foundation ; Department
of Health, UK; Food Standards Agency, UK; Wellcome
Trust, UK; Greek Ministry of Health; Greek Ministry of
Education; Italian Association for Research on Cancer;
Dutch Ministry of Public Health, Welfare and Sports;
Dutch Ministry of Health; Dutch Prevention Funds; LK
Research Funds; Dutch Zorg Onderzoek Nederland;
World Cancer Research Fund; Swedish Cancer Society;
Swedish Scientific Council; Regional Government of
Skane, Sweden; Danish Cancer Society; Norwegian Can-
cer Society; and Foundation to promote research into
functional vitamin B12-deficiency, Norway. Some authors
are partners of ECNIS, a network of excellence of the EC
(6FP contract 513943).
The costs of publication of this article were defrayed in
part by the payment of page charges. This article must
therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 11/17/08; re vised 10/6/09; accepted 10/20/09;
published online 1/7/10.
References
1. Verdecchia A, Mariotto A, Gatta G, Bustamante-Teixeira MT, Ajiki W.
Comparison of stomach cancer incidence and survival in four conti-
nents. Eur J Cancer 2003;39:16039.
2. Webb PM, Law M, Varghese C, et al. Gastric cancer and Helicobac-
ter pylori: a combined analysis of 12 case control studies nested
within prospective cohorts. Gut 2001;49:34753.
3. Schlemper RJ, van der Werf SD, Vandenbroucke JP, Biemond I,
Lamers CB. Seroepidemiology of gastritis in Japanese and Dutch
Table 5. Distribution of genotypes by European region and their associations with GC risk
Gene and SNP Genotype GC (all) GC (northern Europe) GC (southern Europe)
Case/control OR (95% CI) Case/control OR (95% CI) Case/control OR (95% CI)
CBS 699 CC 103/282 1 62/147 1 41/135 1
CT 118/251 1.26 (0.91-1.74) 72/141 1.18 (0.77-1.81) 46/110 1.37 (0.84-2.24)
TT 16/69 0.58 (0.31-1.08) 11/37 0.62 (0.28-1.37) 5/32 0.52 (0.19-1.43)
MAF = 0.32 P
trend
= 0.61 MAF = 0.33 P
trend
= 0.65 MAF = 0.31 P
trend
= 0.81
CBS* 0 insertions 210/536 1 128/312 1 82/224 1
1 insertion 34/91 0.96 (0.62-1.49) 24/40 1.44 (0.82-2.53) 10/51 0.54 (0.26-1.12)
2 insertions 3/4 2.00 (0.43-9.35) 2/1 4.40 (0.36-53.5) 1/3 1.10 (0.11-10.8)
MAF = 0.08 P
trend
= 0.81 MAF = 0.07 P
trend
= 0.10 MAF = 0.09 P
trend
= 0.15
MTR 2756 AA 171/415 1 104/219 1 68/196 1
AG 64/188 0.80 (0.57-1.12) 42/108 0.77 (0.49-1.21) 22/80 0.82 (0.48-1.40)
GG 8/13 1.58 (0.62-3.99) 4/11 0.76 (0.23-2.47) 4/2 13.1 (1.43-120)
MAF = 0.17 P
trend
= 0.56 MAF = 0.18 P
trend
= 0.26 MAF = 0.15 P
trend
= 0.62
MTRR 66 AA 58/156 1 33/72 1 24/84 1
AG 100/286 0.96 (0.65-1.41) 55/149 0.82 (0.48-1.40) 45/137 1.13 (0.65-1.98)
GG 81/165 1.25 (0.82-1.91) 59/109 1.20 (0.70-2.07) 22/56 1.26 (0.64-2.48)
MAF = 0.52 P
trend
= 0.28 MAF = 0.57 P
trend
= 0.39 MAF = 0.46 P
trend
= 0.51
NOTE: RR (95% CI) calculated by conditional logistic regression, stratified by country and adjusted for age and sex.
Abbreviation: MAF, minor allele frequency (cases and controls combined).
*Significant interaction term (P = 0.03) for CBS 844ins68 with European region.
Eussen et al.
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010 Cancer Epidemiology, Biomarkers & Prevention36
working populations: evidence for the developme nt of atrophic
gastritis that is not related to Helicobacter pylori. Gut 1995;37:
199204.
4. Hustad S, Midttun Ø, Schneede J, et al. The methylenetetrahydrofo-
late reductase 677CT polymorphism as a modulator of a B vitamin
network with major effects on homocysteine metabolism. Am J Hum
Genet 2007;80:84655.
5. Bailey LB, Moyers S, Gregory JF. Present knowledge in nutrition. pp.
21429Washington (DC): International Life Sciences Institute; 2001.
6. Larsson SC, Giovannucci E, Wolk A. Folate intake, MTH FR poly-
morphisms, and risk of esophageal, gastric, and pancreatic cancer:
a meta-analysis. Gastroenterology 2006;131:127183.
7. Larsson SC, Giovannucci E, Wolk A. Folate intake and stomach can-
cer incidence in a prospective cohort of Swedish women. Cancer
Epidemiol Biomarkers Prev 2006;15:140912.
8. Botterweck AA, van den Brandt PA, Goldbohm RA. Vitamins, caro-
tenoids, dietary fiber, and the risk of gastric carcinoma: results from
a prospective study after 6.3 years of follow-up. Cancer 2000;88:
73748.
9. Vollset SE, Igland J, Jenab M, et al. The association of gastric cancer
risk with plasma folate , cobalamin, and methylenetetrahydrofolate
reductase polymorphisms in the European Prospective Investigation
into Cancer and Nutrition. Cancer Epidemiol Biomarkers Prev 2007;
16:241624.
10. Mayne ST, Risch HA, Dubrow R, et al. Nutrient intake and ris k of
subtypes of esophageal and gastric cancer. Cancer Epidemiol Bio-
markers Prev 2001;10:105562.
11. Hsing AW, Hansson LE, McLaughlin JK, et al. Pernicious anemia and
subsequent cancer. A population-based cohort study. Cancer 1993;
71:74550.
12. Karlson BM, Ekbom A, Wacholder S, McLaughlin JK, Hsing AW.
Cancer of the upper gastrointestinal tract among patients with per-
nicious anemia: a case-cohort study. Scand J Gastroenterol 2000;
35:84751.
13. Munoz N, Plummer M, Vivas J, et al. A case-control study of gastric
cancer in Venezuela. Int J Cancer 2001;93:41723.
14. Graham S, Haughey B, Marshall J, et al. Diet in the epidemiology of
gastric cancer. Nutr Cancer 1990;13:1934.
15. Cornee J, Pobel D, Riboli E, Guyader M, Hemon B. A case-control
study of gastric cancer and nutritional factors in Marseille, France.
Eur J Epidemiol 1995;11:5565.
16. Harrison LE, Zhang ZF, Karpeh MS, Sun M, Kurtz RC. The role of
dietary factors in the intestinal and diffuse histologic subtypes of
gastric adenocar cinoma: a case-control study in the U.S. Cancer
1997;80:10218.
17. Zhang ZF, Kurtz RC, Yu GP, e t al. Adenocarcinomas of the
esophagus and gastri c cardia: the role of diet. Nu tr Cancer
1997;27:298309.
18. Kaaks R, Tuyns AJ, Haelterman M, Riboli E. Nutrient intake patterns
and gastric cancer risk: a case-control study in Belgium. Int J Cancer
1998;78:41520.
19. Risch HA, Jain M, Choi NW, et al. Dietary factors and the incidence
of cancer of the stomach. Am J Epidemiol 1985;122:94759.
20. Ji BT, Chow WH, Yang G, et al. Dietary habits and stomach cancer in
Shanghai, China. Int J Cancer 1998;76:65964.
21. Kim HJ, Kim MK, Chang WK, et al . Effect of nutrient intake and
Helicobacter pylori infection on gastric cancer in Korea: a case-
control study. Nutr Cancer 2005;52:13846.
22. Lowik MR, van den Berg H, Kistemaker C, Brants HA, Brussaard JH.
Interrelationships between riboflavin and vitamin B6 among elderly
people (Dutch Nutriti on Surveillance System). Int J Vitam Nutr Res
1994;64:198203.
23. McCormick DB. Two interconnected B vitamins: riboflavin and pyri-
doxine. Physiol Rev 1989;69:117098.
24. Zintzaras E. Associati on of met hylenet etrahydr ofolat e reductas e
(MTHFR) polymorphisms with genetic susceptibility to gastric can-
cer: a meta-analysis. J Hum Genet 2006;51:61824.
25. Ott N, Geddert H, Sarbia M. Polymorphisms in methionine synthase
(A2756G) and cystathionine β-synthase (844ins68) and susceptibility
to carcinomas of the upper gastrointestinal tract. J Cancer Res Clin
Oncol 2008;134:40510.
26. Stolzenberg-Solomon RZ, Qiao YL, Abnet CC, et al. Esophageal and
gastric cardia cancer risk and folate- and vitamin B(12)-related poly-
morphisms in Linxian, China. Cancer Epidemiol Biomarkers Prev
2003;12:12226.
27. Zhang FF, Terry MB, Hou L, et al. Genetic polymorphisms in folate
metabolism and the risk of stomach cancer. Cancer Epidemiol Bio-
markers Prev 2007;16:11521.
28. Wang L, Ke Q, Chen W, et al. Polymorphisms of MTHFD, plasma
homocysteine levels, and risk of gastric cancer in a high-risk Chinese
population. Clin Cancer Res 2007;13:252632.
29. Gotze T , Rocken C, Rohl FW, e t al. Gene p olymor phisms of
folate metabolizing enzymes and the risk of gastric cancer.
Cancer Lett 2007.
30. Riboli E, Hunt KJ, Slimani N, et al. European Prospective Investiga-
tion into Cancer and Nutrition (EPIC): study populations and data
collection. Public Health Nutr 2002;5:111324.
31. Riboli E, Kaaks R, European Prospective Investigation into Cancer
and N utrition. The EPIC Project: rationale and study des ign. Int J
Epidemiol 1997;26 Suppl 1:S614.
32. Carneiro F, Moutinho C, Pera G, et al. Pathology findings and valida-
tion of gastric and esophageal cancer cases in a European cohort
(EPIC/EUR-GAST). Scand J Gastroenterol 2007;42:61827.
33. Midttun Ø, Hustad S, Solheim E, Schneede J, Ueland PM. Multiana-
lyte quantification of vitamin B6 and B2 species in the nanomolar
range in human plasma by liquid chro matography-tandem mass
spectrometry. Clin Chem 2005;51:120616.
34. Meyer K, Fredriksen A, Ueland PM. High-level multiplex genotyping
of polymorphisms involved in folate or homocysteine metabolism by
matrix-assisted laser desorption/ionization mass spectrometry. Clin
Chem 2004;50:391402.
35. Koushik A, Kraft P, Fuchs CS, et al. Nonsynonymous polymorphisms
in genes in the one-carbon metabolis m pathway and associations
with colorectal cancer. Cancer Epidemiol Biomarkers Prev 2006;15:
240817.
36. Palli D, Masa la G, Del Giudice G, et al. CagA+ Helicobacter pylori
infection and gastric cancer risk in the EPIC-EURGAST study. Int J
Cancer 2007;120:85967.
37. Foraker AB, Khantwal CM, Swaan PW. Current perspectives on the
cellular uptake and trafficking of riboflavin. Adv Drug Deliv Rev 2003;
55:146783.
38. Merrill AH, Jr., McCormick DB. Affinity chromatographic purification
and properties of flavokinase (ATP:riboflavin 5-phosphotransferase)
from rat liver. J Biol Chem 1980;255:13358.
39. Yamada Y, Merrill AH, Jr., McCormick DB. Probable reacti on
mechanisms of flavokinase and FAD synthetase from rat liver. Arch
Biochem Biophys 1990;278:12530.
40. Karthikeyan S, Zhou Q, Mseeh F, et al. Crystal structure of human
riboflavin kinase reveals a β barrel fold and a novel active site arch.
Structure 2003;11:26573.
41. L eklem J. Present knowledge in nutrition. pp. 17483Washington
(DC): ILSI Press; 1996.
42. Coburn SP. Modeling vitamin B6 metabolism. Adv Food Nutr Res
1996;40:10732.
43. Knol MJ, Vandenbroucke JP, Scott P, Egger M. What do case-
control studies estimate? Survey of methods and assumptions in
published case-control research. Am J Epidemiol 2008.
44. Midttun Ø, Hustad S, Schneede J, Vollset SE, Ueland PM. Plasma
vitamin B-6 forms and their relation to transsulfuration metabolites
in a large, population-based study. Am J Clin Nutr 2007;86:1318.
45. Carmel R. Cobalamin, the stomach, and aging. Am J Clin Nutr 1997;
66:7509.
46. Weck MN, Brenner H. Prevalence of chronic atrophic gastritis in dif-
ferent parts of the world. Cancer Epidemiol Biomarkers Prev 2006;
15:108394.
47. Bates CJ, Pentieva KD, Prentice A. An appraisal of vitamin B6 status
indices and associated confounders, in young people aged 4-18
years and in people aged 65 years and over, in two national British
surveys. Public Health Nutr 1999;2:52935.
48. Chiang EP, Smith DE, Selhub J, et al. Inflammation causes tis-
sue-specific depletion of vitamin B6. Arthritis Res Ther 2005;7:
R125462.
Plasma Vitamin B2 and B6, SNPs, and Gastric Cancer
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010www.aacrjournals.org 37
49. Gori AM, Sofi F, Marcucci R, et al. Association between homocys-
teine, vitamin B(6) concentrations and inflammation. Clin Chem Lab
Med 2007;45:172836.
50. Kekki M, Sam loff IM, Varis K, Ihamaki T. Serum pepsinogen I and
serum gastrin in the screen ing of severe atrophic corpus gastritis.
Scand J Gastroenterol Suppl 1991;186:10916.
51. Hustad S, McKinley MC, McNulty H, et al. Riboflavin, flavin mononu-
cleotide, and flavin adenine dinucleotide in human plasma and ery-
throcytes at baseline and after low-dose riboflavin supplementation.
Clin Chem 2002;48:15717.
52. Powers HJ. Riboflavin (vitamin B-2) and health. Am J Clin Nutr 2003;
77:135260.
53. Tuyns AJ, Kaaks R, Haelterman M, Riboli E. Diet and gastric cancer.
A case-control study in Belgium. Int J Cancer 1992;51:16.
54. Gonzalez CA, Pera G, Agudo A, et al. Fruit and vegetable intake and
the risk of stomach and oesophagus adenocarcinoma in the Europe-
an Prospective Investigation into Cancer and Nutrition (EPIC-EUR-
GAST). Int J Cancer 2006;118:255966.
55. IARC. Fruit and vegetables. IARC handbook of cancer prevention.
Lyon (France): IARC Press; 2003.
56. Riboli E, Norat T. Epidemiologic evidence of the protective effect
of fruit and vegetables on cancer risk. Am J Clin Nutr 2003;78:
55969S.
57. Gonzalez CA, Jakszyn P, Pera G, et al. Meat intake and risk of stom-
ach and esophageal adenocarcinoma within the European Prospec-
tive Investigation Into Cancer and Nutrition (EPIC). J Natl Cancer Inst
2006;98:34554.
58. Webster RP, Gawde MD, Bhattacharya RK. Modulation of carcino-
gen-induced DNA damage and repair enzyme activity by dietary ri-
boflavin. Cancer Lett 1996;98:12935.
59. Pangrekar J, Krishnaswamy K, Jagadeesan V. Effects of riboflavin
deficiency and riboflavin administration on carcinogen-DNA binding.
Food Chem Toxicol 1993;31:74550.
60. Komatsu S, Yanaka N, Matsubara K, Kato N. Antitumor effect of vi-
tamin B6 and its mechanisms. B iochim Biophys Act a 2003;1647:
12730.
61. Stabler SP, Lindenbaum J, Savage DG, Allen RH. Elevation of serum
cystathionine levels in patients with cobalamin and folate deficiency.
Blood 1993;81:340413.
62. Blount BC, Mack MM, Wehr CM, et al. Folate deficiency causes ura-
cil mi sincorporation into human DNA and chromosome breaka ge:
implications for cancer and neuronal damage. Proc Natl Acad Sci
U S A 1997;94:32905.
63. Choi SW , Kim YI, Weitzel JN, Mason JB. Folate depletion impairs
DNA excision repair in the colon of the rat. Gut 1998;43:939.
64. Huang YC, Chen W, Evans MA, Mitchell ME, Shultz TD. Vitamin B-6
requirement and status assessment of young women fed a high-pro-
tein diet with various levels of vitamin B-6. Am J Clin Nutr 1998;67:
20820.
65. Jarrard DF, Bova GS, Isaacs WB. DNA methylation, molecular ge-
netic, and linkage studies in prostate cancer. Prostate Suppl 1996;
6:3644.
66. Jones PA, Buckley JD. The role of DNA methylation in cancer. Adv
Cancer Res 1990;54:123.
Eussen et al.
Cancer Epidemiol Biomarkers Prev; 19(1) January 2010 Cancer Epidemiology, Biomarkers & Prevention38
    • "Therefore, measurement of plasma riboflavin can be used to assess vitamin B2 status in individuals at high risk for GC. Although riboflavin supplements can significantly reduce the risk of GC, different individual intervention effects were observed after dietary supplementation with riboflavin [19] . Therefore, RFT2 may be the key target of environmental and genetic factors because the RFT2 gene also has been reported as a susceptibility gene for GC using a genome-wide association study approach [10] . "
    [Show abstract] [Hide abstract] ABSTRACT: To investigate the relationship between blood riboflavin levels and riboflavin transporter 2 (RFT2) gene expression in gastric carcinoma (GC) development. High-performance liquid chromatography was used to detect blood riboflavin levels in patients with GC. Real-time fluorogenic quantitative polymerase chain reaction and immunohistochemistry were used to analyze the expression of RFT2 mRNA and protein in samples from 60 GC patients consisting of both tumor and normal tissue. A significant decrease in the RFT2 mRNA levels was detected in GC samples compared with those in the normal mucous membrane (0.398 ± 0.149 vs 1.479 ± 0.587; P = 0.040). Tumors exhibited low RFT2 protein expression (75%, 16.7%, 8.3% and 0% for no RFT2 staining, weak staining, medium staining and strong staining, respectively), which was significantly lower than that in the normal mucous membrane (10%, 16.7%, 26.7% and 46.7% for no RFT2 staining, weak staining, medium staining and strong staining, respectively; P < 0.05). Tumors with low RFT2 expression were significantly associated with tumor stage and histological grade. Moreover, a significantly decrease in Uyghur patients was observed compared with Han patients. However, other parameters-gender, tumor location and lymph node metastasis-showed no significant relationship with RFT2 expression. Blood riboflavin levels were reverse correlated with development of GC (1.2000 ± 0.97569 ng/mL in high tumor stage patients vs 2.5980 ± 1.31129 ng/mL in low tumor stage patients; P < 0.05). A positive correlation of plasma riboflavin levels with defective expression of RFT2 protein was found in GC patients (χ² = 2.619; P = 0.019). Defective expression of RFT2 is associated with the development of GC and this may represent a mechanism underlying the decreased plasma riboflavin levels in GC.
    Full-text · Article · Jun 2012
  • [Show abstract] [Hide abstract] ABSTRACT: Folate plays a key role in DNA synthesis and methylation. Limited evidence suggests high intake may reduce risks of esophageal cancer overall; however, associations with esophageal cancer subtypes and Barrett's esophagus (BE), a precancerous lesion, remain unexplored. We evaluated the relation between intake of folate, B vitamins, and methyl-group donors (methionine, choline, betaine) from foods and supplements, polymorphisms in key folate-metabolizing genes, and risk of BE, esophageal adenocarcinoma (EAC), and esophageal squamous cell carcinoma (ESCC) in 2 population-based case-control studies in Australia. BE patients without (n = 266) or with (n = 101) dysplasia were compared with population controls (n = 577); similarly, EAC (n = 636) or ESCC (n = 245) patients were compared with population controls (n = 1507) using multivariable adjusted logistic regression. Increasing intake of folate from foods was associated with reduced EAC risk (P-trend = 0.01) and mitigated the increased risks of ESCC associated with smoking and alcohol consumption. In contrast, high intake of folic acid from supplements was associated with a significantly elevated risk of BE with dysplasia. High intakes of riboflavin and methionine from food were associated with increased EAC risk, whereas increasing betaine intake was associated with reduced risks of BE without (P-trend = 0.004) or with dysplasia (P-trend = 0.02). Supplemental thiamin, riboflavin, niacin, and vitamin B-12 were associated with increased EAC risk. There were no consistent associations between genetic polymorphisms studied and BE or EAC risk. High intake of folate-containing foods may reduce risk of EAC, but our data raise the possibility that folic acid supplementation may increase risks of BE with dysplasia and EAC.
    Article · Feb 2011
  • [Show abstract] [Hide abstract] ABSTRACT: Nutritional status assessment is a critical tool for the identification of nutrient deficiencies or excesses in individual healthcare and epidemiologic screening. Because low but 'normal' status of folate, vitamin B-12, and vitamin B-6 have been associated with an increased risk of chronic diseases, research has focused on defining sensitive indicators of B-vitamin status and on the development and validation of analytical methods for their quantification. With the increasing availability and more user-friendly configuration of liquid chromatograph-tandem mass spectrometers (LC-MS/MS), numerous analytical methods for determination of B-vitamin indicators by LC-MS/MS have been developed over the last years. These methods include folate assays for simultaneous determination of numerous folate forms at their specific reduction level. The functional indicators for vitamin B-12 status are plasma methylmalonic acid and total homocysteine and can be measured, either individually or in combination, by high-throughput analysis using LC-MS/MS. Methods for vitamin B-6 status assessment are multianalyte platforms that determine vitamin B-6 forms and functional indicators by the same assay. The high sensitivity, selectivity, and specificity of isotope-dilution LC-MS/MS [and gas chromatography-mass spectrometry (GC-MS)] techniques have allowed the development of reference methods and the creation of multianalyte platforms. The additional convenience of automated sample preparation enables high sample throughput and makes those sensitive methods prospective analytical candidates for larger settings including clinical laboratories.
    Article · Sep 2011
Show more

  • undefined · undefined
  • undefined · undefined
  • undefined · undefined