Abstract

The risk of gastric cancer (GC) associated with dietary intake of nitrosodimethylamine (NDMA) and endogenous formation of nitroso compounds (NOCs) was investigated in the European Prospective Investigation into Cancer and Nutrition (EPIC). The study included 521 457 individuals and 314 incident cases of GC that had occurred after 6.6 average years of follow-up. An index of endogenous NOC (ENOC) formation was estimated using data of the iron content from meat intake and faecal apparent total NOC formation according to previous published studies. Antibodies to Helicobacter pylori and vitamin C levels were measured in a sub-sample of cases and matched controls included in a nested case–control within the cohort. Exposure to NDMA was <1 μg on average compared with 93 μg on average from ENOC. There was no association between NDMA intake and GC risk (HR, 1.00; 95% CI, 0.7–1.43). ENOC was significantly associated with non-cardia cancer risk (HR, 1.42; 95% CI, 1.14–1.78 for an increase of 40 μg/day) but not with cardia cancer (HR, 0.96; 95% CI, 0.69–1.33). Although the number of not infected cases is low, our data suggest a possible interaction between ENOC and H.pylori infection ( P for interaction = 0.09). Moreover, we observed an interaction between plasma vitamin C and ENOC ( P < 0.02). ENOC formation may account for our previously reported association between red and processed meat consumption and gastric cancer risk.

Introduction

Humans are exposed to N -nitroso compounds (NOCs) from diet, tobacco smoke and other environmental sources, as well as from endogenous synthesis, which contributes to 45–75% of total exposure ( 1 ). Various NOCs have been found to be carcinogenic in multiple organs in at least 40 animal species including higher primates ( 1 ). Despite continuing concern that NOCs may be causally related to gastrointestinal cancer, the epidemiological literature has failed so far to support this link with any degree of conviction ( 2 ). Helicobacter pylori infection and other dietary and environmental factors are thought to have a major role in gastric carcinogenesis. Our previous results show a positive association between red and processed meat intake and non-cardia gastric cancer, especially in H.pylori- infected subjects ( 3 ).

Bingham et al . ( 4 ) have shown that red but not white meat diets markedly increase faecal NOCs [as apparent total NOCs (ATNC)], and that there is a clear dose response with increasing red meat intake, suggesting that endogenous N -nitrosation could explain the association between colorectal cancer and red and processed meat. Furthermore, there appears to be a specific role of haem on endogenous nitrosation whereas non-meat protein or inorganic iron has no effect ( 5 ). Under certain conditions, haems are known to be nitrosated, and act as nitrosating agents ( 6 ). Nitric oxide (NO) has been shown to react directly with haemoglobin and myoglobin to produce NOCs ( 7 ). The formation of N -nitrosoarginine by haem enzymes under anaerobic conditions has also been demonstrated ( 8 ). In addition, endogenous formation of NOCs is increased in the upper gastrointestinal tract following processed and red meat consumption ( 9 ). Endogenous production is substantiated by the markedly elevated levels of faecal output (in the order of 500 μg/day) compared with dietary intakes of only 13 μg/day ( 4 )

In this large prospective study of diet and cancer, we have estimated the dietary intake of nitrosodimethylamine (NDMA) and the endogenous formation of NOCs using an index of endogenous nitrosation (ENOC). Then, we assessed the effect of both on gastric cancer (GC) risk within the full cohort. Also, we have examined the effect of modification of ENOC by plasma vitamin C and H.pylori infection in relation to non-cardia tumours in a nested case–control study.

Materials and methods

Subjects

The rationale and methods of the European Prospective Investigation into Cancer and Nutrition (EPIC) study have been described in detail previously ( 10 ). Briefly, the EPIC cohort consists of 521 457 subjects (368 010 women and 153 447 men), from 10 European countries, aged 35–70 years, and recruited mostly between 1992 and 1998. Subjects from Norway (owing to a short follow-up period) and Greece (because the ENOC could not be calculated) were excluded. Finally, 314 incident GC cases confirmed as adenocarcinoma were included in the cohort analysis.

End points

The follow-up was based mostly on population cancer registries ( 10 ). GC includes cancers coded as C16 according to the 10th Revision of the International Statistical Classification of Diseases, Injuries and Causes of Death (ICD) ( 11 ). Validation and confirmation of the diagnosis and classification of tumour site was carried out by a panel of pathologists ( 3 ). Of the 314 GC incident cases, 92 were classified as cardia cancer (including 19 gastro–oesophageal junction cases) and 155 as non-cardia, and for 67 cases the site was unknown.

Diet and lifestyle questionnaires

Usual diet over the previous 12 months was measured at recruitment by country-specific validated questionnaires ( 10 ). Lifestyle questionnaires included questions on education, lifetime history of smoking and alcohol, occupation, reproductive history and use of hormones, history of previous illness and physical activity. Values for total energy and dietary iron were computed using country-specific food composition tables.

Nitrites and NDMA intakes

Dietary intake of nitrites and NDMA was estimated by matching food items on the country-specific questionnaires with a food database of potential carcinogens ( 12 ). Since considerably more information is available for NDMA than for other nitrosamines found in foods, only dietary intakes of NDMA were estimated. Given that food items could have several different source of information, we selected country-specific values when they were available. For food items for which analyzed values were not available, the value of the nearest comparable food was assigned. When there were several matches for a food item, the mean of all the suitable values was assigned (<5% of the food items).

Endogenous NOC (ENOC)

An index for ENOC exposure was determined using data of iron intake from meat and faecal ATNC formation from published studies ( 4 , 5 , 13 , 14 ). First, we estimated the amount of iron from meat for diets administered in these studies, using the UK Food Composition Database ( 15 ). Then, the correlation between iron content and faecal ATNC levels was assessed. These estimations showed a very high correlation between intake of iron from meat and ATNC formation ( r = 0.948) ( Figure 1 ), whereas that from meat was substantially lower ( r = 0.576). Therefore, we used linear regression to predict the ENOC for each subject of the cohort using their consumption of iron from meat as independent variable (ENOC μg/day = 40.52 + 18.92 * iron from meat) for intakes of iron less than 8.5 mg/day. For iron intakes > 8.5 mg/day, we applied a fixed value of 201.41 obtained from the model (0.7% of cases).

Fig. 1.

ENOC formation in relation to dietary iron from meat. Data gathered from feeding studies from Bingham et al . (4), Cross et al . (5), Hughes et al . (13), Silvester et al . (14). Iron content from UK Food Composition Database (15).

Fig. 1.

ENOC formation in relation to dietary iron from meat. Data gathered from feeding studies from Bingham et al . (4), Cross et al . (5), Hughes et al . (13), Silvester et al . (14). Iron content from UK Food Composition Database (15).

Nested case–control study: plasma vitamin c and H.pylori infection

Plasma vitamin C levels and antibodies against H.pylori were determined in cases and controls selected for the nested case–control study within the EPIC cohort. For each incident GC case with available blood sample, between two and four control subjects were randomly selected from the cohort, matched by sex, age group (±2.5 years), centre and data of blood sample collection (±45 days).

Statistical methods

The proportional hazard model (Cox regression) was used for analysis of the cohort data. It was stratified by age and by centre to control for potential confounding factors as differences in follow-up procedures and questionnaire design. Age was used as the timescale variable in all models. Entry time was defined as age at recruitment and final time as age of diagnosis (cases) or age at censoring (at risk subjects). All models were adjusted for sex, height, weight, educational level, alcohol intake (g/day), status of smoking (never, former and current), daily cigarette smoking (in current smokers only), work physical activity (no activity, sedentary, standing, manual and heavy manual), leisure physical activity (as continuous METS-hour/week), energy intake (kcal/day), consumption of total vegetables, fresh fruit and citrus fruit (g/day) and nitrites intake (mg/day). Intake was analyzed as continuous variables (increment of 1 and 40 μg/day for NDMA and ENOC, respectively), log-transformed and as categorical variables, by EPIC-wide sex-specific tertiles. The increment of 40 μg/day of ENOC was chosen because this corresponded with 100 g of meat. Categorical variables were scored from 1 to 3 and trend tests were calculated on these scores. The odds ratio (OR) for association of ENOC in H.pylori- infected and non-infected subjects and according to plasma levels of vitamin C were estimated by multiple unconditional logistic regression, including matching variables in the model. Interaction between predicted ENOC and H.pylori infection and plasma vitamin C was tested by likelihood ratio test.

Results

The results reported here are from one of the largest cohorts of men and women specifically designed to examine the relationships between diet and cancer. There were 2 916 642 person-years in 6.64 average years of follow-up. Baseline characteristics of the participants are given in Table I . The exposure to NOCs estimated by ENOC was far greater than the dietary exposure according to NDMA intake estimations. GC cases had higher H.pylori infection rates and lower plasma vitamin C than controls.

Table I.

Sample description


 
Mean
 
SD
 
Mean
 
SD
 
Full cohort GC cases  Controls  
N 314 — 439077 — 
Male (%) 56 — 31 — 
Age at recruitment 59.20 7.48 51.48 10.05 
Follow-up (year) 3.73 2.22 6.64 1.72 
ENOC (μg/day) 93.05 30.39 84.46 30.71 
NDMA (μg/day) 0.26 0.34 0.19 0.31 
Nested case–control     
N 229  1086  
H.pylori positive (%)  83 — 66% — 
Nested case–control     
N 200  491  
Plasma vitamin C (mmol/l) 39.85 25,63 41.12  

 
Mean
 
SD
 
Mean
 
SD
 
Full cohort GC cases  Controls  
N 314 — 439077 — 
Male (%) 56 — 31 — 
Age at recruitment 59.20 7.48 51.48 10.05 
Follow-up (year) 3.73 2.22 6.64 1.72 
ENOC (μg/day) 93.05 30.39 84.46 30.71 
NDMA (μg/day) 0.26 0.34 0.19 0.31 
Nested case–control     
N 229  1086  
H.pylori positive (%)  83 — 66% — 
Nested case–control     
N 200  491  
Plasma vitamin C (mmol/l) 39.85 25,63 41.12  

ENOC, endogenous nitroso compounds exposure index. NDMA: N-Nitrosodimethlyamine.

Table II shows the hazard ratio (HR) of GC associated with dietary NDMA and ENOC exposure. Dietary NDMA was not associated with increased risk of GC (HR, 1.00; 95% CI, 0.70–1.43 for an increment of 1 μg of NDMA). This lack of effect was observed in all the models performed and for both cardia and non-cardia tumours. ENOC was positively but not statistically significantly associated with GC (HR, 1.18; 95% CI, 0.99–1.39 for an increment of 40 μg of EN). When analyzed by tumour site, ENOC was not associated with cardia cancer risk (HR, 0.96; 95% CI, 0.69–1.33 for an increment of 40 μg of EN). However, it was significantly associated with non-cardia cancer risk (HR, 1.42; 95% CI, 1.14–1.78 for an increment of 40 μg of ENOC). Results were consistent in all models performed.

Table II.

Endogenous nitroso compounds exposure index (ENOC) and dietary NDMA exposure and the risk of stomach adenocarcinoma in the EPIC-EURGAST study a

Site Cases number  Tertiles b
 
     Continuous
 
   
  HR  2
 
  3
 
  Original c
 
  log 2
 
 

 

 

 
(95% CI)
 
HR
 
(95% CI)
 
P trend
 
HR
 
(95% CI)
 
HR
 
(95% CI)
 
NDMA N          
    Stomach 314 0.87 (0.64–1.2) 0.99 (0.69–1.41) 0.96 (0.7–1.43) 1.01 (0.9–1.12) 
    Cardia 92 0.74 (0.41–1.34) 0.68 (0.34–1.37) 0.29 0.73 (0.3–1.79) 1.01 (0.81–126) 
    Non-cardia 155 1.04 (0.66–1.63) 1.09 (0.65–1.81) 0.75 1.09 (0.69–1.73) 0.96 (0.83–1.12) 
ENOC c           
    Stomach 314 1.12 (0.83–1.51) 1.32 (0.94–1.84) 0.1 1.18 (0.99–1.39) 1.42 (1.06–1.92) 
    Cardia 92 1.43 (0.81–2.52) 1.29 (0.67–2.47) 0.5 0.96 (0.69–1.33) 1.04 (0.59–1.83) 
    Non-cardia 155 1.22 (0.79–1.88) 1.61 (1.01–2.58) 0.04 1.42 (1.14–1.78) 1.93 (1.28–2.91) 
Site Cases number  Tertiles b
 
     Continuous
 
   
  HR  2
 
  3
 
  Original c
 
  log 2
 
 

 

 

 
(95% CI)
 
HR
 
(95% CI)
 
P trend
 
HR
 
(95% CI)
 
HR
 
(95% CI)
 
NDMA N          
    Stomach 314 0.87 (0.64–1.2) 0.99 (0.69–1.41) 0.96 (0.7–1.43) 1.01 (0.9–1.12) 
    Cardia 92 0.74 (0.41–1.34) 0.68 (0.34–1.37) 0.29 0.73 (0.3–1.79) 1.01 (0.81–126) 
    Non-cardia 155 1.04 (0.66–1.63) 1.09 (0.65–1.81) 0.75 1.09 (0.69–1.73) 0.96 (0.83–1.12) 
ENOC c           
    Stomach 314 1.12 (0.83–1.51) 1.32 (0.94–1.84) 0.1 1.18 (0.99–1.39) 1.42 (1.06–1.92) 
    Cardia 92 1.43 (0.81–2.52) 1.29 (0.67–2.47) 0.5 0.96 (0.69–1.33) 1.04 (0.59–1.83) 
    Non-cardia 155 1.22 (0.79–1.88) 1.61 (1.01–2.58) 0.04 1.42 (1.14–1.78) 1.93 (1.28–2.91) 

a Full cohort analysis: stratified center and age; Adjusted by sex, height, weight, education level, tobacco smoking, cigarette smoking intensity, work and leisure physical activity, citrus and non-citrus fruits intake, vegetables intake, energy intake and nitrites.

b Tertiles are full cohort sex-specific. Cut points are: ENOC: Men (78 and 106), women (65 and 87); NDMA: men (0.12 and 0.28), women (0.06), and 0.11).

c Per 40 μg/day of ENOC and per 1 μg/day for NDMA.

The risk of non-cardia tumour regarding ENOC stratified by H.pylori infection and plasma vitamin C is shown in Table III . Among infected patients, ENOC was found to be associated with non-cardia cancer in all models (OR, 1.82; 95% CI, 1.32–2.51 for categorical trends), while in non-infected subjects we did not find any association. However, the test for interaction was not statistically significant ( P = 0.09). In relation to plasma vitamin C, the positive association with ENOC was present only among those with low serum levels of vitamin C (OR, 3.24; 95% CI, 1.77–5.93 for those with less than 40 micromol/l of plasma vitamin C). Among those with high levels of vitamin C, there was no association between ENOC and non-cardia GC risk. The test for interaction was statistically significant ( P = 0.02) and consistent in all models performed. Moreover, the stronger effect of ENOC at low levels of plasma vitamin C persisted even after restricting the analysis to H.pylori- infected subjects (OR, 3.52; 95% CI, 1.8–8.9). We analyzed also the interaction with tobacco smoking, but we did not observe any association ( P for interaction = 0.83).

Table III.

Risk of non-cardia adenocarcinoma regarding endogenous nitroso compounds exposure index (ENOC) levels, stratified by H.pylori infection and plasma vitamin C levels a in the EPIC-EURGAST study


 
Cases
 
Controls
 
Continuous OR 95% CI
 
Log2 OR 95% CI
 
Stratified by H.pylori infection      
    Infected 111 717 1.82 (1.32–2.51) 2.93 (1.63–5.29) 
    Non-infected 12 369 0.15 (0.01–4.06) 0.22 (0.003–15.3) 
     P for interaction    0.09 0.13 
Stratified by plasma vitamin C     
    <40 μmol 54 235 3.24 (1.77–5.93) 10.1 (3.25–31.1) 
    >40 μmol 55 256 1.10 (0.63–1.93) 1.26 (0.52–3.08) 
     P for interaction    0.02 0.01 

 
Cases
 
Controls
 
Continuous OR 95% CI
 
Log2 OR 95% CI
 
Stratified by H.pylori infection      
    Infected 111 717 1.82 (1.32–2.51) 2.93 (1.63–5.29) 
    Non-infected 12 369 0.15 (0.01–4.06) 0.22 (0.003–15.3) 
     P for interaction    0.09 0.13 
Stratified by plasma vitamin C     
    <40 μmol 54 235 3.24 (1.77–5.93) 10.1 (3.25–31.1) 
    >40 μmol 55 256 1.10 (0.63–1.93) 1.26 (0.52–3.08) 
     P for interaction    0.02 0.01 

a 40 μmol/l:median of plasmatic vitamin C.

Unmatched analysis: adjusted by sex, age, centre and date of blood extraction, height, weight, education level, tobacco smoking, cigarette smoking intensity, work and leisure physical activity; alcohol intake, energy intake and nitrites. H.pylori models were also adjusted by citrus and non-citrus fruits intake and vegetables intake. Vitamin C models were also adjusted by H.pylori infection.

Discussion

This is the first study reporting relationships between both endogenous and exogenous exposure to NOCs and GC risk. The exposure of NDMA from food was <1 μg/day, whereas that from ENOC was 93 μg/day ( Table I ). We found that non-cardia GC was positively associated with ENOC exposure but not with dietary NDMA. There is only one other cohort study that has investigated the association between NDMA and GC, and no association was found ( 16 ). In the present study, we estimated intake of NDMA, the volatile nitrosamines most commonly detected in foods. Although this compound only represents a small fraction of total NOC exposure, it is considered an indicator of dietary exposure ( 17 ). The limitations of assigning values of NDMA to foods based on published reports include the limited availability of data of these compounds in some food items, and the variation of levels found in similar items across countries. During the second half of the last century, reductions in the use of nitrates and nitrites for curing meat and the modifications of malting techniques in the brewing industry have resulted in significant reductions in the levels of NDMA in foods. Therefore, it may be possible that current NDMA exposure in our population is below the biological level required to increase the risk of cancer

The contribution of ENOC was estimated using content of iron from meat intake. The effect of red and processed meat on ENOC can be attributed to the haem content, but heam iron was not used to derive the index because standard databases estimate haem iron as 40% of iron in all meats, whereas newer methods of analysis indicate that the proportion should vary from 22 to 79% according to the type of meat ( 18 ). The use of an indirect measurement of endogenous exposure is a limitation but the correlation between iron and ENOC was very high ( Figure 1 ), and it supports the rationale of our methodology. On the other hand, this methodology could be a useful tool in epidemiological studies where it is not possible to obtain direct estimations of endogenous nitrosation. Our results support the hypothesis that endogenous exposure is probably the major contributor to the overall burden of human exposure to NOC.

There is sufficient evidence that tobacco smoking increases the risk of GC ( 19 ). Our previous results suggest an increase of risk of 79% for GC in current smokers ( 20 ). Although tobacco is a source of specific nitrosamines, we did not observe any interaction with ENOC. It might suggest that dietary and tobacco-related nitrosamines could have independent effects on the risk of GC.

It is well established that H.pylori infection increases the risk of developing distal non-cardia gastric cancer but is not associated with the cardia site ( 21 ). Infection with H.pylori is likely to increase NO production from macrophages in response to bacterial overgrowth, so that the availability of NO in H.pylori- infected individuals will be increased ( 22 ). Haem is readily nitrosated and can then nitrosate other substrates in the presence of NO. Table 2 shows that the effect of ENOC is only observed in non-cardia cancer and is restricted to infected individuals. This result is consistent with the previous reported association in the same population between red meat intake and non-cardia cancer in H.pylori- infected subjects. ( 3 ).

Vitamin C may be protective against H.pylori- associated gastric carcinogenesis by enhancing mucosal immune response, neutralizating free radicals, reducing the formation of gastric NOC, inhibiting cell proliferation and directly affecting H.pylori growth ( 23 ). Inflammation induced by H.pylori infection in the stomach not only causes significantly increased requirements for vitamin C but also reduces secretion of the vitamin C into gastric lumen ( 23 ). We observed a positive association of ENOC among the individuals with low levels of plasma vitamin C ( Table 3 ). This may reflect lower vitamin C levels associated with H.pylori infection. Given the potential capacity of H.pylori to modulate the effect of vitamin C (or vice versa) we analyzed the interaction between ENOC and vitamin C stratifying by H.pylori infection. The results show that the interaction was even stronger among infected subjects. It is plausible that there is synergy between vitamin C and H.pylori , which modifies the effect of NOCs on GC risk.

In conclusion, we have found a significant increase of non-cardia cancer risk associated with endogenous exposure to NOCs but not with pre-formed NDMA intake. Cardia cancer was not associated with any type of NOC exposure. Although the number of not infected cases in our study is low, the data suggest a possible interaction between ENOC and H.pylori infection. Moreover, we observed an interaction between plasma vitamin C and ENOC. Further cohort studies with more cases and years of follow-up are needed to confirm these findings. Clarifications of the mechanisms of action of these compounds and their potential interactions with H.pylori and vitamin C levels are needed given the high prevalence of this infection in the general population.

The EPIC study was funded by ‘Europe Against Cancer’ Programme of the European Commission (SANCO); Ligue contre le Cancer (France); Société 3M (France); Mutuelle Générale de l'Education Nationale; Institut National de la Santé et de la Recherche Médicale (INSERM); German Cancer Aid; German Cancer Research Center; German Federal Ministry of Education and Research; Danish Cancer Society; Health Research Fund (FIS) of the Spanish Ministry of Health (RCESP-C03/09); the participating regional governments and institutions of Spain; Cancer Research, UK; Medical Research Council, UK; the Stroke Association, UK; British Heart Foundation; Department of Health, UK; Food Standards Agency, UK; the Wellcome Trust, UK; Greek Ministry of Health; Greek Ministry of Education; Italian Association for Research on Cancer; Italian National Research Council; Dutch Ministry of Public Health, Welfare and Sports; Dutch Ministry of Health; Dutch Prevention Funds; LK Research Funds; Dutch ZON (Zorg Onderzoek Nederland); World Cancer Research Fund (WCRF); Swedish Cancer Society; Swedish Scientific Council; Regional Government of Skane, Sweden; Norwegian Cancer Society. We thank the members of the panel of pathologists for their valuable work: Dr R.S. Umea, Sweden; Dr U. Mahlke, Postdam, Germany; Dr Hendrik Bläker, Heildelberg, Germany; Dr Gabriella Nesi, Firenze, Italy; Dr Vicki Save, Cambridge, UK; Dr Claus Fenger, Copenhagen, Denmark; Dr Julio Torrado, San Sebastian, Spain; Dr Johan Offerhaus, Amsterdam, The Netherlands; Ioannina, Greece. We also thank Dr Dimitrious Roukos and Cátia Moutinho, Porto, Portugal, for her excellent technical support to the panel of pathologists. Specific study results of the nested case–control study within EPIC (EUR-GAST) were obtained with financial support from the FP5 of European Commission (QLG1-CT-2001-01049). The participation of P.J. in the project is supported by a grant of the ‘Fundació Privada Institut d'Investigació Biomèdica de Bellvitge (IDIBELL).

Conflict of Interest Statement : None declared.

References

1.
Tricker,A.R. (
1997
) N -nitroso compounds and man: sources of exposure, endogenous formation and occurrence in body fluids
Eur. J. Cancer Prev.
  ,
6
,
226
–268.
2.
World Cancer Research Fund. (
1997
)
Food, Nutrition and the Prevention of Cancer: A Global Perspective.
  American Institute for Cancer Research, Washington, DC.
3.
Gonzalez,C.A., Jakszyn,P., Pera,G. et al. (
2006
) Meat intake and risk of stomach and esophageal adenocarcinoma within the European Prospective Investigation into Cancer and Nutrition (EPIC).
J. Natl. Cancer Inst.
  ,
98
,
345
–354.
4.
Bingham,S.A., Pignatelli,B., Pollock,J.R., Ellul,A., Malaveille,C., Gross,G., Runswick,S., Cummings,J.H. and O'Neill,I.K. (
1996
) Does increased endogenous formation of N -nitroso compounds in the human colon explain the association between red meat and colon cancer?
Carcinogenesis
  ,
17
,
515
–523.
5.
Cross,A.J., Pollock,J.R. and Bingham,S.A. (
2003
) Haem, not protein or inorganic iron, is responsible for endogenous intestinal N -nitrosation arising from red meat.
Cancer Res.
  ,
63
,
2358
–2360.
6.
Bonnett,R., Holleyhead,R., Johnson,B.L. and Randall,E.W. (
1975
) Reaction of acidified nitrite solutions with peptide derivatives: evidence for nitrosamine and thionitrite formation from 15N NMR studies.
J. Chem. Soc. [Perkin 1],
  ,
22
,
2261
–2241.
7.
Wade,R.S. and Castro,C.E. (
1990
) Redox reactivity of iron(III) porphyrins and heme proteins with nitric oxide. Nitrosyl transfer to carbon, oxygen, nitrogen, and sulfur.
Chem. Res. Toxicol.
  ,
3
,
289
–291.
8.
Hirst,J. and Goodin,D.B. (
2000
) Unusual oxidative chemistry of N -(omega)-hydroxyarginine and N -hydroxyguanidine catalyzed at an engineered cavity in a heme peroxidase.
J. Biol. Chem.
  ,
275
,
8582
–8591.
9.
Lunn,J., Pollock,J. and Bingham,S. (
2004
) The effect of increased red and processed meat on endogenous formation of N -nitroso compounds and DNA strand breaks in ileostomists.
Cancer Epidemiol. Biomarkers Prev.
  ,
13
((Suppl. 11) Abstract),
A 95
–1852.
10.
Riboli,E., Hunt,K.J., Slimani,N. et al. (
2002
) European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection.
Public Health Nutr.
  ,
5
,
1113
–1124.
11.
World Health Organization. (
1992
)
International Statistical Classification of Diseases and Related Health Problems.
  10 th version, WHO, Geneva.
12.
Jakszyn,P., Agudo,A., Ibanez,R., Garcia-Closas,R., Pera,G., Amiano,P. and Gonzalez,C.A. (
2004
) Development of a food database of nitrosamines, heterocyclic amines, and polycyclic aromatic hydrocarbons.
J. Nutr
  ,
134
,
2011
–2014.
13.
Hughes,R., Cross,A.J., Pollock,J.R. and Bingham,S. (
1997
) Dose-dependent effect of dietary meat on endogenous colonic N -nitrosation [Erratum in: Carcinogenesis 2001 22 , 685.]
Carcinogenesis
  ,
22
,
199
–202.
14.
Silvester,K.R., Bingham,S.A., Pollock,J.R., Cummings,J.H. and O'Neill,I.K. (
1997
) Effect of meat and resistant starch on fecal excretion of apparent N -nitroso compounds and ammonia from the human large bowel.
Nutr. Cancer
  ,
29
,
13
–23.
15.
Food Standards Agency. (
2002
)
McCance and Widdowson's The Composition of Foods.
  Sixth summary edition Royal Society of Chemistry, Cambridge.
16.
Knekt, P., Jarvinen, R., Dich, J. and Hakulinen, T. (
1999
) Risk of colorectal and other gastro-intestinal cancers after exposure to nitrate, nitrite and N -nitroso compounds: a follow-up study
Int. J. Cancer
  ,
80
,
852
–856.
17.
Ministry of Agriculture, Fisheries and Food. (
1992
)
Nitrate, Nitrite and N-Nitrosocompounds in Foods.
  Second Report. Food surveillance Paper No. 23. HMSO, London.
18.
Carpenter,C.E. and Clark,E. (
1995
) Evaluation of methods used in meat iron analysis and iron content of raw and cooked meats.
J. Agric. Food Chem.
  ,
43
,
1824
–1827.
19.
International Agency for Research on Cancer. (
2002
)
IARC Monogaphs. Tobacco smoking and tobacco smoke.
  IARC, Lyon.
20.
Gonzalez,C.A., Pera,G., Agudo,A. et al. (
2003
) Smoking and the risk of gastric cancer in the European Prospective Investigation Into Cancer and Nutrition (EPIC).
Int. J. Cancer
  ,
107
,
629
–634.
21.
Helicobacter and Cancer Collaborative Group. (
2001
) Gastric cancer and Helicobacter pylori : a combined analysis of 12 case control studies nested within prospective cohorts.
Gut
  ,
49
,
347
–353.
22.
Pignatelli,B., Bancel,B., Esteve,J., Malaveille,C., Calmels,S., Correa,P., Patricot,L.M., Laval,M., Lyandrat,N. and Ohshima,H. (
1998
) Inducible nitric oxide synthase, anti-oxidant enzymes and Helicobacter pylori infection in gastritis and gastric precancerous lesions in humans.
Eur. J. Cancer Prev.
  ,
7
,
439
–447.
23.
Zhang,Z.W. and Farthing,M.J. (
2005
) The roles of vitamin C in Helicobacter pylori associated gastric carcinogenesis.
Chin. J. Dig. Dis.
  ,
6
,
53
–58.

Author notes

Department of Epidemiology, Catalan Institute of Oncology, (ICO-IDIBELL) Barcelona, Spain, 1Medical Research Council, Dunn Human Nutrition Unit, Wellcome Trust/MRC Building, Cambridge, UK, 2Strangeways Research Laboratory, Cambridge, UK, 3German Institute of Human Nutrition, Potsdam-Rehbücke, Germany, 4IRIS Research Center, Chiron-Vaccines, Siena, Italy, 5Molecular and Nutritional Epidemiology Unit, CSPO – Scientific Institute of Tuscany, Florence, Italy, 6Epidemiology Unit, Istituto Tumori, Milan, Italy, 7University of Torino, Italy, 8Cancer Registry, Azienda Ospedaliera ‘Civile M.P. Arezzo’, Ragusa, Italy, 9Dipartamento di Medicina Clinica e Sperimentale, Federico II University, Compagnia di San Paolo, Naples, Italy, 10Department of Medicine, University Hospital Malmo, Lund University Malmo, Sweden, 11Department of Nutritional Research, University of Umeå, Sweden, 12Andalusian School of Public Health, Granada, Spain, 13Department of Public Health of Guipuzkoa, San Sebastian, Spain, 14Public Health Institute of Navarra, Pamplona, Spain, 15Epidemiology Department, Health Council of Murcia, Spain, 16Public Health and Health Planning Directorate, Asturias, Spain, 17Cancer Epidemiology Unit, University of Oxford, UK, 18Institute of Community Medicine, University of Tromso, Norway, 19Institute of Molecular Pathology and Immunology of the University of Porto and Medical Faculty, Porto, Portugal, 20Division of Clinical Epidemiology, Deutsches Krebsforschungszentrum, Heidelberg, Germany, 21Department of Clinical Epidemiology, Aalborg Hospital, Aarhus University Hospital, Aalborg, Denmark, 22Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark, 23Center for Nutrition and Health, National Institute for Public Health and the Environment, Bilthoven, The Netherlands, 24Julius Center for Health Sciences and Primary Care, University Medical Center, Utrecht, The Netherlands, 25INSERM, Institut Gustave Roussy, Villejuif, France, 26Department of Hygiene and Epidemiology, Medical School, University of Athens, Greece, 27Det Sundhedsvidenskabelige Fakultet, Syddansk Universitet, Denmark, 28Department of Medical Biosciences, University of Umea, Sweden, 29Nutrition and Hormones Group, International Agency for Research on Cancer, Lyon, France and 30Department of Epidemiology and Public Health, Imperial College, London, UK