Japanese Journal of Clinical Oncology 32:277-283 (2002)
© 2002 Foundation for Promotion of Cancer Research
Risk Factors for Stomach Cancer in Brazil (I): a Case-control Study among Non-Japanese Brazilians in São Paulo
1 Research Center and 3 Department of Head and Neck Surgery, Hospital do Cancer A.C. Camargo, São Paulo, 2 Nikkei Disease Prevention Center, Santa Cruz Hospital, São Paulo, 4 Department of Gastroenterology and 5 Department of Pathology, School of Medicine, University of São Paulo, São Paulo, Brazil and 6 Epidemiology and Biostatistics Division, National Cancer Center Research Institute East, Kashiwa, Chiba, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
Background: Stomach cancer is an important health problem in Brazil, with an estimated 20 000 new cases per year and it was the most frequent cancer site in men and the third most frequent site in women in São Paulo in 1993. Nevertheless, there are no reports of analytical epidemiological study on stomach cancer in Brazil.
Methods: A hospital-based case-control study was conducted among Brazilian residents with a non-Japanese background in the city; 236 consecutive cases of histologically confirmed stomach cancer were matched to the same number of controls admitted for non-neoplastic diseases by age (±5 years) and gender. The socio-demographic characteristics, personal and family medical history, lifetime history of tobacco use and dietary habits were determined by interview using a structured questionnaire.
Results: Non-white race, lower educational background and lower family income were more frequent in stomach cancer patients. After adjustment for race and educational background, cigarette smoking, frequent use of oil and frequent consumption of egg were significantly associated with increased risk of stomach cancer, while frequent consumption of fruit and vegetables decreased the risk. These associations did not change substantially after mutual adjustment of the other variables.
Conclusions: The present study confirmed that low socio-economic status, cigarette smoking and low consumption of fruit and vegetables were risk factors of stomach cancer in São Paulo, Brazil.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
According to recent estimates, approximately 20 350 new stomach cancer cases occurred in Brazil in 1999, ranking it as the third most frequent cancer following breast and uterine cervix cancers (Ministry of Health, Brazil, 1999). The world age-standardized incidence rates for stomach cancer in the city of São Paulo was 39.3 (per 100 000) for men and 15.5 for women in 1993 (1), classifying it as one of the high-risk areas in the world (2). Geographic variations within Brazil are described, ranging from 42.0 (men) and 16.0 (women) in Belem 1989–91 to 19.2 (men) and 10.6 (women) in Goiania 1990–1993 (2). However, to our knowledge, there is no report on risk factors of this cancer in Brazil.
Migrant studies, especially on Japanese immigrants to Brazil, have shown only a slight decrease in incidence of stomach cancer compared with Japanese living in Japan, which is in contrast to the decreasing incidence rate for stomach cancer found among Japanese Americans (3). Such findings are consistent with the occurrence of this type of cancer in the host countries, Brazil and USA, respectively, with high and low incidence rates.
In this study, we investigated the association between stomach cancer and potential risk factors through a hospital-based case-control study in the city of São Paulo. We conducted this study among Brazilian patients of non-Japanese origin, using the same methodology as carried out among Japanese Brazilians (see pages 284–290).
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
This hospital-based case-control study was conducted in the city of São Paulo from June 1991 to June 1994, in order to investigate potential risk factors for stomach cancer in non-Japanese Brazilians. Eligible case and control patients were living in São Paulo city without any Asian background.
Cases
Of the 250 Brazilians patients from 13 collaborating hospitals, of non-Japanese origin, 14 cases were excluded because histological diagnosis did not confirm malignant carcinoma. A total of 236 patients, comprising 170 men and 66 women, aged from 40 to 79 years were eligible as stomach cancer cases. There were no refusals to be interviewed among the cases.
Controls
We selected one control for each case, matching for gender and age within 5 years and cancer-free subjects. Control patients were preferentially selected in the same hospitals as the index cases. In only one hospital, a specialized cancer unit, the controls were selected from a neighboring public hospital. Brazilian controls had a refusal rate to interview of 8.4% (23 subjects), the main reason being rejection of donating blood samples to the study. A total of 236 non-Japanese Brazilian controls, comprising 235 inpatients and one outpatient, were eligible as matched controls. The main reason for hospital admission included infectious disease (10), metabolic disease (3), neurological disease (7), cardiovascular disease (67), respiratory disease (15), gastritis (3), other digestive and abdominal diseases (75), urinary disease (4), dermatological disease (1), musculo-skeletal disease (17), trauma (31) and others (2).
Interviews
Prior to the interview, a signed informed consent was obtained from cases and controls, assuring their voluntary participation in this study and the confidentiality of all information obtained. All subjects underwent a structured interview by a trained nurse and the same nurse interviewed each matched pair. A structured questionnaire included personal, social and demographic characteristics, personal and family medical history, lifetime history of tobacco and alcohol use and past dietary habits based on frequency of consumption for about 30 common food items. All participants were interviewed for ~50 min and one of the authors visited the interview site and monitored regularly to assure uniformity in the inquiry, especially between cases and controls. At the end of the interview, 15 ml of blood were collected for further analysis of selected biomarkers in order to assess pepsinogen I and II serum levels and Helicobactor pylori antibody and genetic polymorphism of metabolizing enzyme for nitrosamine (e.g. cytochrome P450IIE1). Parts of these analyses have been published elsewhere (4–6).
Statistical Analysis
Crude and adjusted odds ratios (OR) and their respective 95% confidence intervals (95% CI) were estimated for each exposure using conditional logistic regression analysis (7). Cumulative exposure to tobacco smoking was expressed in pack-years defined as cumulative exposure equivalent to packs smoked per day times the number of years of smoking. We considered the equivalence of 20 commercially manufactured cigarettes (one pack) with four hand-rolled black tobacco cigarettes or four cigars or five pipefuls of regular pipe tobacco (8). The lifetime alcohol consumption were calculated in terms of grams of ethanol consumed assuming the following alcoholic concentration: beer 5%, wine 10%, hard liquor 50% and cachaça (pinga, a spirit distilled from sugar cane) 50%. The daily amount of ethanol (g) was combined with frequency and years of drinking into a synthetic index. Diet was assessed with food frequency questions which were categorized into four categories: less than once a week, 1–2 days a week, 3–4 days a week and daily.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
The distribution of socio-economic and demographic factors of patients among stomach cancer cases and controls is shown in Table 1. Non-white race was more frequent among stomach cancer patients. Cases had significantly lower education level and lower family income than controls. Since family income was correlated with educational level (Spearman correlation = 0.41) and there were substantial percentages of unknown answers, educational level (illiterate, primary or secondary/university) as well as race (white or non-white) was included in the subsequent analyses for statistical adjustment.
|
Table 2 shows the association with smoking and alcohol drinking. Overall risk was increased among current cigarette smokers (OR = 1.7; 95% CI = 1.1–2.7 after adjustment for race and education, OR = 1.6; 95% CI = 1.0–2.6 after further adjustment for fruit and other vegetable intake) with no excess risk for ex-smokers. The associations were similar when the analysis was restricted to men. The increased lifetime consumption (pack-years) was associated with increased risk (P for trend = 0.004 and 0.008, respectively). All types of tobacco products as a whole were associated with stomach cancer, showing a significant increase in risk among those smoking more than 30 pack-years. The use of manufactured cigarettes was the most common form of tobacco consumption, with few subjects smoking hand-rolled black tobacco cigarettes, cigars or pipe. Frequency of alcohol drinking or the lifetime amounts of alcohol consumption were both not associated with stomach cancer risk, as shown in Table 2. The associations did not change when the analysis was restricted to men.
|
Diet was assessed using a food frequency questionnaire, with the main results presented in Table 3. A reduction in risk was associated with daily consumption of fresh fruit compared with those eating less than once a week with an OR = 0.4 (95% CI = 0.2–0.8), with a dose–response relationship (P for trend = 0.005). Although the association was slightly attenuated after further adjustment for smoking and consumption of other vegetables consisting of those except green and yellow vegetables, a dose–response reduction in risk was observed (P = 0.08) with increase in fruit consumption. Daily yellow vegetables consumption and the group of other vegetables were protective with the same magnitude of risk reduction (OR = 0.4, 95% CI = 0.2–0.8 for both vegetable groups) with a statistically significant trend (P = 0.04 for yellow vegetables and 0.004 for other vegetables). The observed associations were slightly attenuated after further adjustment for smoking and fruit; however, daily consumptions were still significantly associated with reduced risk of stomach cancer (OR = 0.5, 95% CI = 0.3–0.99 for yellow vegetables, OR = 0.5, 95% CI = 0.3–0.97 for other vegetables).
|
We explored the habit of adding salt to the food or a preference for salty food and the results showed there was no association with stomach cancer (data not shown) and a positive relation was obtained for daily oil use compared with less than once per week and the OR was 1.9 (95% CI = 1.0–3.5) (P for trend = 0.008) and 2.7 (95% CI = 1.5–4.9) (P for trend <0.001) for daily egg consumption. The positive associations of these two factors became more evident after further adjustment for smoking and consumption of fruit and other vegetables (OR = 2.3, 95% CI = 1.2–4.5 for a daily oil use, OR = 3.2, 95% CI = 1.7–6.0 for daily egg consumption.
The OR for family history of stomach/duodenal ulcer was 2.1 (95% CI = 1.2–3.5) and 3.1 (95% CI = 0.95–10.0) for the family history of stomach cancer. The observed associations were both slightly attenuated after further adjustment for smoking and consumption of fruit and other vegetables (OR = 1.8, 95% CI = 1.0–3.2 and OR = 2.7 and 95% CI = 0.8–9.0, respectively).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
The aim of this study was to investigate the effects of selected risk factors for stomach cancer with emphasis on diet, smoking habits and alcohol consumption. Our target population shows an intermediate risk for stomach cancer, with a downward trend in mortality rates during last two decades (1).
The epidemiological results of the effect of smoking habits on stomach cancer risk have been conflicting. Whereas most recent cohort studies have indicated a positive association between smoking and stomach cancer (9–12), evidence from case-control studies is abundant although less consistent, some showing positive results with a significant dose–response relationship (13,14) and others not (15–17). In this study an excess risk of stomach cancer was observed among current smokers, with a significant dose–response relationship regarding lifetime consumption. This suggests that a positive, although not strong, association may exist between cigarette smoking and stomach cancer risk.
In our study, in which cachaça was the most frequently consumed, intake of alcohol beverages was not related to any increased risk of stomach cancer, in line with many recent studies (9,13,18,19). Only a few studies have evaluated a significant association between alcohol drinking and stomach cancer risk (15,16,20). In most of these studies, the type of alcoholic beverage responsible for the increased or decreased risk was different: wine and liquor (16) or vodka (20). Thus far, the causal role of alcoholic beverages in stomach carcinogenesis is not yet established.
We examined the effect of certain groups of food rather than nutrients and we were able to detect the protective effects of fruit and three (green, yellow and other) types of vegetable consumption, although that of green vegetables was not statistically significant. These findings are consistent with a large number of epidemiological studies (21–23) and prevention of stomach cancer by an increase in fruit and vegetable consumption must be plausible in Brazil with relatively high incidence and mortality (2). Unexpected findings were increased risk in oil use and egg consumption and these associations did not change even after adjustment for smoking, fruit and other vegetable intake. One possibility is chance finding and this aspect needs further studies.
The reverse association was found with two indices of socio-economic factor: educational level and family income. Although low socio-economic status itself would not increase the risk of stomach cancer, it can be associated with various presumed risk factors such as salted food intake (24) or infection with Helicobacter pylori (25). Therefore, we included a variable of educational background as an index of socio-economic factor as well as race in the model of conditional logistic analysis for adjusting possible confounding factors. Moreover, Brazilians with very low family incomes may not be patients with non-fatal diseases and therefore less likely to be controls in a study such as this.
Selection bias is always of critical importance in case-control studies, especially when hospital-based. A large number of controls in this study were patients admitted with cardiovascular disease (n = 67) and abdominal disease (n = 75). Although exclusion of each disease group from controls did not substantially change the observed findings, the limitations of case-control studies should be acknowledged and the findings treated with caution.
In summary, the present case-control study of stomach cancer confirmed that low socio-economic status, cigarette smoking and low consumption of fruit and vegetables were risk factors of stomach cancer in São Paulo, Brazil. Egg and oil consumption were found to be associated with increased risk.
|
Acknowledgments |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology and for the Second Term Comprehensive 10-Year Strategy for Cancer Control and for Cancer Research from the Ministry of Health and Welfare, Japan. We thank Ms Raimunda N. Pereira, Júlia M. F. Toyota and Arlete B. Rubbo for assistance with the field work.
|
Study Group Members |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
The investigators and participating institutions in the São Paulo–Japan Gastric Cancer Study Group were as follows: M. Asai, R. A. Possik, R. G. Bevilacqua, M. S. Oyafuso, R. S. O. Filho and D. Gastanheira (Hospital do Cancer, A.C. Camargo); B. S. Lessa, Y. Menezes, C. A. M. Fonseca and A. J. T. Nigro (Hospital Heliópolis); C. A. Malheiros and G. C. Santo (Santa Casa de Misericórdia de São Paulo); C. Kobata (Hospital São Paulo, Universidade Federal de São Paulo/EPM); A. C. de Godoi (Hospital do Servidor Público Estadual de São Paulo); J. P. A. Birindelli and L. A. Soares (Hospital Ipiranga); J. J. G. Rodrigues, K. Iriya and Y. Iriya (Hospital das Clinicas – Faculdade de Medicina da USP); V. N. Silveira and I. T. Soubhia (Hospital do Servidor Público Municipal de São Paulo); V. N. Silveira and L. T. Horita (Hospital Barro Branco); J. K. Nakauchi and M. Akamine (Hospital Nipo-Brasileiro); M. Matsuda and J. P. A. Freitas (Hospital Santa Cruz); T. Tomishige (Centro Medico ENKYO); and P. Kassab (OSEC).
|
FOOTNOTES |
|---|
+ *Study group members are listed at the end of the paper
For reprints and all correspondence: Shoichiro Tsugane, 6–5–1 Kashiwanoha, Kashiwa, Chiba 277–8566, Japan. E-mail: stsugane@east.ncc.go.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsStudy Group MembersREFERENCES |
|---|
1 Mirra AP, editor. Cancer Incidence in São Paulo County, Brazil: 1983–1988–1993: Trends in the Period 1969–1993. Brazilia: Ministry of Health 1999.
2 Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J, editors. Cancer Incidence in Five Continents Vol. VII. IARC Scientific Publications No. 143. Lyon: International Agency for Research on Cancer 1997.
3 Tsugane S, de Souza JMP, Costa ML Jr, Mirra AP, Gotlieb SLD, Laurenti R, et al. Cancer incidence rates among Japanese immigrants in the city of Sao Paulo, Brazil, 1969–78. Cancer Causes Control 1990;1:189–93.[Web of Science][Medline]
4 Fahey MT, Hamada GS, Nishimoto IN, Kowalski LP, Iriya K, Gama Rodrigues JJ, et al. Ethnic differences in serum pepsinogen level among Japanese and non-Japanese Brazilian gastric cancer cases and controls. Cancer Detect Prev 2000;24:564–71.[Web of Science][Medline]
5 Fujioka N, Fahey MT, Hamada GS, Nishimoto IN, Kowalski LP, Iriya K, et al. Serological immunoglobulin G antibody titers to Helicobacter pylori in Japanese Brazilian and non-Japanese Brazilian gastric cancer patients and controls in São Paulo. Jpn J Cancer Res 2001;92:829–35.[Web of Science]
6 Nishimoto IN, Hanaoka T, Sugimura H, Nagura K, Ihara M, Li XJ, et al. Cytochrome P450 2E1 (Cyp2E1) polymorphism in gastric cancer in Brazil. Case-control studies of Japanese Brazilians and non-Japanese Brazilians. Cancer Epidemiol Biomarkers Prev 2000;9:675–80.
7 Breslow NE, Day NE. Statistical Methods in Cancer Research. Vol. I. The Analysis of Case-control Studies. IARC Scientific Publications No. 32. Lyon: International Agency for Research on Cancer 1980.
8 Franco E, Kowalski L, Oliveira B, Curadoso M, Pereira R, Silva M, et al. Risk factors for oral cancer in Brazil: A case-control study. Int J Cancer 1989;43:992–1000.[Web of Science][Medline]
9 Kneller RW, McLaughlin JK, Bjelke E, Schuman LM, Blot WJ, Wacholder S, et al. A cohort study of stomach cancer in a high-risk American population. Cancer 1991;68:672–8.[Web of Science][Medline]
10 Kato I, Tominaga S, Matsumoto K. A prospective study of stomach cancer among a rural Japanese population: a 6-year survey. Jpn J Cancer Res 1992;83:568–75.[Web of Science]
11 McLaughlin JK, Hrubec Z, Blot WJ, Fraumeni JF Jr. Smoking and cancer mortality among U.S. veterans: a 26-year follow-up. Int J Cancer 1995;60:190–3.[Web of Science][Medline]
12 Mizoue T, Tokui N, Nishisaka K, Nishisaka S, Ogimoto I, Ikeda M, et al. Prospective study on the relation of cigarette smoking with cancer of the liver and stomach in an endemic region. Int J Epidemiol 2000;29:232–7.
13 Hansson LE, Baron J, Nyren O, Bergstrom R, Wolk A, Adami HO. Tobacco, alcohol and the risk of gastric cancer. A population-based case-control study in Sweden. Int J Cancer 1994;57:26–31.[Web of Science][Medline]
14 Ji BT, Chow WH, Yang G, McLaughlin JK, Gao RN, Zheng W, et al. The influence of cigarette smoking, alcohol and green tea consumption on the risk of carcinoma of the cardia and distal stomach in Shanghai, China. Cancer 1996;77:2449–57.[Web of Science][Medline]
15 Boeing H, Beyme RF, Berger MM, Berndt V, Gores W, Korner M, et al. Case-control study on stomach cancer in Germany. Int J Cancer 1991;47:858–64.[Web of Science][Medline]
16 Agudo A, Gonzalez CA, Marcos G, Sanz M, Saigi E, Verge J, et al. Consumption of alcohol, coffee and tobacco and gastric cancer in Spain. Cancer Causes Control 1992;3:137–43.[Web of Science][Medline]
17 Guo W, Blot WJ, Li JY, Taylor PR, Liu BQ, Wang W, et al. A nested case-control study of oesphageal and stomach cancers in the Linxian nutrition intervention trial. Int J Epidemiol 1994;23:444–50.
18 Hoshiyama Y, Sasaba T. A case-control study and multiple stomach cancers in Saitama prefecture, Japan. Jpn J Cancer Res 1992;83:937–43.[Web of Science]
19 You WC, Zhang L, Gail MH, Chang YS, Liu WD, Ma JL, et al. Gastric dysplasia and gastric cancer: Helicobacter pylori, serum vitamin C and other risk factors. J Natl Cancer Inst 2000;92:1607–12.
20 Jedrychowski W, Boeing H, Wahrendorf J, Popiela T, Adamczyk BT, Kulig J. Vodka consumption, tobacco smoking and risk of gastric cancer in Poland. Int J Epidemiol 1993;22:606–13.
21 Palli D. Epidemiology of gastric cancer: an evaluation of available evidence. J Gastroenterol 2000;35(Suppl 12):84–9.[Medline]
22 Kono S, Hirohata T. Nutrition and stomach cancer. Cancer Causes Control 1996;7:41–55.[Web of Science][Medline]
23 Neugut AI, Hayek M, Howe G. Epidemiology of gastric cancer. Semin Oncol 1996;23:281–91.[Web of Science][Medline]
24 Gerber AM, James SA, Ammerman AS, Keenan NL, Garrett JM, Strogatz DS, et al. Socioeconomic status and electrolyte intake in black adults: the Pitt County Study. Am J Publ Hlth 1991;81:1608–12.
25 Brown LM. Helicobacter pylori: epidemiology and routes of transmission. Epidemiol Rev 2000;22:283–97.
Received October 2, 2001; accepted April 30, 2002
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. T. Ryan, S. L. Aquino, and R. L. Kradin Case 29-2007 -- A 51-Year-Old Man with Gastric Cancer and Lung Nodules N. Engl. J. Med., September 20, 2007; 357(12): 1239 - 1246. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hashibe, P. Boffetta, D. Zaridze, O. Shangina, N. Szeszenia-Dabrowska, D. Mates, V. Janout, E. Fabianova, V. Bencko, N. Moullan, et al. Evidence for an important role of alcohol- and aldehyde-metabolizing genes in cancers of the upper aerodigestive tract. Cancer Epidemiol. Biomarkers Prev., April 1, 2006; 15(4): 696 - 703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Tajima Challenging Epidemiological Strategy for Paradoxical Evidence on the Risk of Gastric Cancer from Helicobacter pylori Infection Jpn. J. Clin. Oncol., August 1, 2002; 32(8): 275 - 276. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||