Skip Navigation


American Journal of Epidemiology Advance Access originally published online on October 26, 2006
American Journal of Epidemiology 2007 165(2):138-147; doi:10.1093/aje/kwj366
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
165/2/138    most recent
kwj366v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Nöthlings, U.
Right arrow Articles by Kolonel, L. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nöthlings, U.
Right arrow Articles by Kolonel, L. N.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

ORIGINAL CONTRIBUTIONS

Vegetable Intake and Pancreatic Cancer Risk: The Multiethnic Cohort Study

Ute Nöthlings1, Lynne R. Wilkens1, Suzanne P. Murphy1, Jean H. Hankin1, Brian E. Henderson2 and Laurence N. Kolonel1

1 Cancer Research Center of Hawaii, University of Hawaii, Honolulu, HI
2 Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA

Reprint requests to Dr. Ute Nöthlings, Department of Epidemiology, German Institute of Human Nutrition Potsdam-Rehbruecke (DIfE), Arthur-Scheunert Allee 114–116, 14558 Nuthetal, Germany (e-mail: ute.noethlings{at}dife.de).

Received for publication November 30, 2005. Accepted for publication June 8, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Investigators studying associations between vegetable intake and pancreatic cancer risk have reported inconsistent findings to date. To further explore these associations, the authors analyzed data on 183,522 participants enrolled in the Hawaii–Los Angeles Multiethnic Cohort Study in 1993–1996. Intakes of total vegetables, light green, dark green, yellow-orange, and cruciferous vegetables, tomato products, and legumes were estimated from a quantitative food frequency questionnaire. After an average of 8.3 years of follow-up, 529 pancreatic cancer cases were identified. Multivariate-adjusted Cox proportional hazards models were created. All statistical tests were two-sided. Overall, total vegetable intake was not associated with pancreatic cancer risk, nor was intake of vegetable subgroups. Current smokers, who were at increased risk of pancreatic cancer (relative risk = 1.78, 95% confidence interval: 1.40, 2.27), had a decreased risk with higher intake of dark green vegetables (for comparison of extreme quartiles, relative risk = 0.50, 95% confidence interval: 0.27, 0.92; p-trend = 0.029). The inverse association for dark green vegetables was also seen in African Americans (p-trend = 0.043). In stratified analyses, inverse associations with total vegetables, light green vegetables, and legumes were significant in overweight/obese subjects. In conclusion, the authors found no evidence for an inverse association between vegetable intake and pancreatic cancer overall, but inverse associations in high-risk persons suggest the need for further investigation.

cohort studies; ethnic groups; pancreatic neoplasms; vegetables

Abbreviations: BMI, body mass index; CI, confidence interval; QFFQ, quantitative food frequency questionnaire; RR, relative risk


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Pancreatic cancer is the most fatal cancer in adults. It is generally diagnosed at a late stage and is poorly responsive to treatment. It ranks fourth among US cancer deaths, and the 5-year survival rate is less than 5 percent (1). More than 32,000 new cases of pancreatic cancer were estimated to have occurred in the United States in 2005 (2).

Because of its poor prognosis and the minimal impact of conventional treatment methods (3), it is important to focus on prevention for this disease. So far, only a few risk factors have been established. Cigarette smoking is the most important etiologic factor yet identified, increasing the risk approximately twofold for current smokers as compared with nonsmokers (4). The attributable risk for smoking has been estimated at 25 percent (5). A family history of pancreatic cancer, a diagnosis of diabetes mellitus, and obesity have also been associated with the disease (3, 6–14). Other risk factors include increasing age, male sex, and Native Hawaiian or African-American ethnicity (15).

It has been hypothesized that vegetable consumption offers protective effects against several cancers. Various constituents of these foods have been proposed as sources of possible beneficial effects, including dietary fiber, folic acid, selenium, and phytochemicals such as carotenoids, flavonoids, and isoflavones (16). Several investigators have studied the associations of vegetable intake with pancreatic cancer and have suggested that vegetable intake may reduce risk (17–19). However, the findings have been inconsistent, rendering further investigations necessary before firm conclusions can be drawn.

We studied the relation between vegetable intake and pancreatic cancer risk among participants in the Multiethnic Cohort Study, using data from an average of 8.3 years of follow-up.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study design
The Multiethnic Cohort Study was established in Hawaii and Los Angeles, California, to investigate lifestyle exposures in relation to disease outcomes, especially the relation between diet and cancer. The institutional review boards of the University of Hawaii and the University of Southern California approved the study proposal. Recruitment procedures, study design, and baseline characteristics have been reported elsewhere (20). In brief, men and women aged 45–75 years at cohort creation in 1993, primarily of five major ethnicities (African-American, Latino, Japanese-American, Native Hawaiian, and Caucasian), were enrolled in the cohort between 1993 and 1996. All study participants initially completed a self-administered comprehensive questionnaire that included a detailed dietary assessment, as well as sections on demographic factors; body weight and height; lifestyle factors other than diet, including smoking history; history of prior medical conditions, including diabetes mellitus; and family history of cancer. Follow-up of the cohort for cancer incidence and mortality entails active contact with the subjects, as well as passive computerized linkages to cancer registries and death certificate files in Hawaii and California and to the National Death Index.

Study population
At baseline, over 215,000 men and women were recruited. Because the quantitative food frequency questionnaire (QFFQ) was designed to cover the usual diet of the five major ethnic groups, participants not belonging to one of these ethnic groups were excluded from this analysis (n = 13,994), as were persons with diets deemed implausible on the basis of reported energy and macronutrient intakes (n = 8,265). To make the latter determination, we excluded the top and bottom 10 percent tails of the log energy distribution and computed a robust standard deviation, assuming a truncated normal distribution. Then we excluded all energy values outside of the range (mean ± 3 robust standard deviations). We used a similar procedure to exclude persons with extreme fat, protein, or carbohydrate intakes (outside the range of mean ± 3.5 robust standard deviations). Subjects with a pancreatic cancer diagnosis prior to baseline that was either self-reported in the questionnaire or identified by registry linkages (n = 59) and subjects with missing information at baseline on smoking status (never, former, current; n = 2,967), intensity and duration of smoking (n = 4,043), history of diabetes mellitus (n = 1), or self-reported body weight and height (used to compute body mass index (BMI; weight (kg)/height (m)2); n = 2,281) were excluded. To eliminate unreasonable combinations of self-reported weights and heights, we also excluded subjects with a BMI outside the range of 15–50 kg/m2 (n = 698). As a result, data on 183,522 participants were available.

Dietary assessment
Dietary intake was assessed at baseline using a comprehensive questionnaire especially designed and validated for use in this multiethnic population. The development of the self-administered QFFQ has been described elsewhere (20, 21). In brief, 3-day measured dietary records from approximately 60 men and women of each ethnic group were used to identify a minimum set of food items that contributed at least 85 percent of the intake of a specific list of nutrients. Food items that were common in the diet of a particular ethnic group were also included in the QFFQ, irrespective of their nutrient contributions. The QFFQ asked about the consumption of over 180 food items, including more than 20 individual vegetable items as well as mixed dishes containing vegetables. The QFFQ inquired about the usual frequency (based on 8–9 categories) and amount of each food consumed, based on three portion sizes per food item. The reference portion sizes were also derived from the 3-day measured dietary records.

Prior to calculation of food group intake, the food mixtures from the QFFQ were disaggregated to the ingredient level using a customized recipe database. For example, while the salami on a pizza was counted towards the processed meat group, the tomatoes on that pizza were counted towards the vegetable group. Food group intake was calculated as grams per day of the basic food commodities. Foods and food groups used in this analysis were: total vegetables; light green, dark green, yellow-orange, and cruciferous vegetables; tomatoes and tomato products; and legumes.

Identification of pancreatic cancer cases
Incident exocrine pancreatic cancer cases were identified through record linkages to the Hawaii Tumor Registry, the Cancer Surveillance Program for Los Angeles County, and the California State Cancer Registry. All three registries are part of the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute and have been operating since 1960, 1972, and 1988, respectively. Case ascertainment was complete through December 31, 2002, for this analysis. International Classification of Diseases for Oncology, Second Edition (22), codes C25.0–C25.3 and C25.7–C25.9 were counted as diagnoses of exocrine pancreatic cancer. Endocrine pancreatic cancers were not counted as cases, but follow-up for subjects with these tumors was censored at the date of diagnosis.

Statistical analysis
We applied Cox proportional hazards models using age as the time metric to calculate relative risks. Person-times were calculated beginning at the date of cohort entry, defined as the date of questionnaire completion or the date the participant turned 45 years of age, for the few persons who were younger than 45 when they completed the baseline questionnaire (n = 1,104). Person-times ended at the earliest of three dates: the date of pancreatic cancer diagnosis, the date of death, or December 31, 2002, the closure date for this analysis. Tests based on Schoenfeld residuals showed no evidence that the proportional hazards assumption was violated for any analysis. Separate models for men and women showed similar patterns. Therefore, we present models including all persons, stratified by sex to allow for different baseline hazard rates. All models were stratified by follow-up time, categorized as ≤2 years, >2–5 years, and >5 years. A separate analysis hinted at some differences between relative risks across the three follow-up time strata, although the differences were not statistically significant (data not shown).

Foods and food group exposures were investigated in disease models in terms of percentiles. For analyses using quintiles, four indicator variables were created to represent the quintiles, based on the distribution in the entire cohort (men and women). Trend variables, used to test dose-response, were assigned the sex- and ethnicity-specific median values for overall quintiles or quartiles. Age at cohort entry, ethnicity, history of diabetes mellitus, family history of pancreatic cancer, smoking status (never, former, or current), pack-years of smoking, intake of energy (logarithmically transformed), intakes of red meat and processed meat, and BMI were used as adjustment factors in multivariate models for the overall cohort. We included energy in order to investigate associations for foods independently of their relation to overall energy intake. Intakes of red and processed meat were previously associated with an increased risk of pancreatic cancer in the Multiethnic Cohort Study (23). In models additionally adjusted for educational attainment and alcohol consumption, the risk estimates were changed only marginally (data not shown), so these variables were not included in the final models. All statistical tests were two-sided.

The likelihood ratio test was used to determine the significance of the interactions of smoking status (current smoking vs. never smoking) and BMI (<25 kg/m2 vs. ≥25 kg/m2) with vegetable intake with respect to pancreatic cancer. The test compares a main-effects, no-interaction model with a fully parameterized model containing all possible interaction terms for the variables of interest.

To reduce measurement error in the dietary assessments, we chose to energy-adjust food group intakes by calculating densities, that is, grams per 1,000 kcal of energy intake per day. In a validation study, we found that energy-adjusted intakes produced substantially higher correlation coefficients with the reference instrument than did absolute intakes (21). This has also been reported in other studies (24). Densities measure the contribution of the food group to overall diet and are therefore interpreted differently from absolute measures.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
The analytic cohort comprised 1,520,886 person-years of follow-up and included 529 incident cases of exocrine pancreatic cancer. Table 1 gives study participants' characteristics by quintile of total vegetable intake. Forty-five percent of study participants were men. Sex was associated with intake of vegetables, with the proportion of men decreasing across quintiles. The mean age of study participants at cohort entry was 60 years, but age was not strongly associated with vegetable intake. The proportion of African Americans decreased and the proportion of Latinos increased across vegetable intake quintiles. The mean BMI decreased from 26.2 kg/m2 in the lowest quintile of total vegetable intake to 25.7 kg/m2 in the highest quintile. Energy intake and the mean number of pack-years of smoking were inversely associated with vegetable intake.


View this table:
[in this window]
[in a new window]

  		TABLE 1 Characteristics of participants in the Hawaii–Los Angeles Multiethnic Cohort Study, by quintile of total vegetable consumption (n = 183,522), 1993–2002

 
Because of the well-established relation between smoking and pancreatic cancer risk, we first examined this association in our cohort. Current smokers had a statistically significant 78 percent higher risk of pancreatic cancer than never smokers (relative risk (RR) = 1.78, 95 percent confidence interval (CI): 1.40, 2.27). Former smokers in general were not at higher risk than never smokers (RR = 1.10, 95 percent CI: 0.90, 1.34), but persons who had recently quit smoking (within 5 years of questionnaire administration) had significantly increased risks of pancreatic cancer. Pack-years of smoking also were positively associated with pancreatic cancer risk. Participants in the fifth quintile, reflecting more than 31.7 pack-years of smoking, had a 46 percent higher risk of pancreatic cancer than never smokers (RR = 1.46, 95 percent CI: 1.12, 1.89; p-trend < 0.001).

In order to explore the association between vegetable intake and pancreatic cancer risk, we analyzed total intake as well as various groupings of vegetables as exposures in disease risk models. After adjustment for confounding variables, no statistically significant association between total vegetable intake or vegetable subgroups and pancreatic cancer risk was detected (table 2). The multivariate-adjusted models suggested inverse associations for dark green vegetables and legumes (p-trend < 0.10), but the relative risk did not reach statistical significance. Comparing the fifth quintile with the first, the relative risks of pancreatic cancer in men and women, respectively, were 0.91 (95 percent CI: 0.59, 1.40; p-trend = 0.545) and 0.69 (95 percent CI: 0.46, 1.04; p-trend = 0.069) for dark green vegetables and 0.75 (95 percent CI: 0.50, 1.13; p-trend = 0.235) and 0.96 (95 percent CI: 0.61, 1.51; p-trend = 0.278) for legumes. The exclusion of participants with a history of diabetes mellitus did not change the null findings.


View this table:
[in this window]
[in a new window]

  		TABLE 2 Relative risk of exocrine pancreatic cancer according to quintile of vegetable consumption in the Hawaii–Los Angeles Multiethnic Cohort Study (n = 183,522), 1993–2002

 
Because of the suggestive findings for dark green vegetables and legumes in table 2, we further examined the associations between vegetable intake and pancreatic cancer risk in selected high-risk subgroups based on smoking status, ethnicity, and BMI. The results are shown in table 3. The table includes four vegetable groups for which there was a significant result in any of the analyses. In the analysis by smoking status (as noted above, we found an approximately twofold increase in pancreatic cancer risk among current smokers), current smokers showed a statistically significant inverse association with intake of dark green vegetables (p-trend = 0.029), whereas never smokers showed a statistically significant inverse association with total vegetables. However, there was no evidence of an interaction between smoking (current smoking vs. never smoking) and intake of dark green (p = 0.575) or total (p = 0.914) vegetables based on tertiles (data not shown). No smoking subgroup showed any significant inverse trends for legumes or light green vegetables (table 3) or for any other vegetable subgroup (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 3 Multivariate-adjusted* relative risk of exocrine pancreatic cancer according to category of vegetable consumption, by smoking status, body mass index, and ethnicity, in the Hawaii–Los Angeles Multiethnic Cohort Study (n = 183,522), 1993–2002

 
In the analysis by ethnicity (African Americans are at higher risk than Caucasians (15)), a statistically significant inverse association was seen for dark green vegetables among African Americans (p-trend = 0.043) but not among Japanese Americans or Caucasians. (The numbers of cases in Latinos (n = 85) and Native Hawaiians (n = 43) were too small for meaningful analyses.) None of the other vegetable groups included in the table showed any significant trend, and no significant trends were found for any other vegetable group (data not shown).

In the analysis by BMI (a high BMI has been reported to be a risk factor for pancreatic cancer (14)), we stratified the subjects as normal-weight (BMI <25 kg/m2) versus overweight/obese (BMI ≥25 kg/m2) (table 3). Inverse associations were found only in the overweight/obese group, where statistically significant inverse trends were seen for intakes of total vegetables (p = 0.026), light green vegetables (p = 0.042), and legumes (p = 0.045), though not for dark green vegetables. However, analyses showed no evidence of interaction between BMI and vegetable intake based on tertiles (p for interaction = 0.148, 0.137, and 0.791 for total vegetables, light green vegetables, and legumes, respectively). No other vegetable group showed any statistically significant inverse association (data not shown).

In a previous analysis (23), we identified processed meat as a risk factor for pancreatic cancer in our cohort. However, a stratified analysis of the associations between vegetables and pancreatic cancer by tertiles of processed meat intake did not show any differences in risk across strata.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
After 8 years of follow-up in the Multiethnic Cohort Study, we did not find a statistically significant inverse association between total vegetable intake and pancreatic cancer risk overall, though the results suggested a decreased risk associated with consumption of dark green vegetables and legumes. When we limited the analysis to current smokers, we found a significantly decreased risk with dark green vegetable intake, but results were not statistically different from those for never smokers. The inverse association with dark green vegetables was also shown in African Americans. After stratification by BMI, inverse associations with total vegetables, light green vegetables, and legumes were found in overweight/obese persons.

To date, 10 other prospective studies have investigated vegetable intake as a risk factor for pancreatic cancer (table 4) (10, 25–33). Only three of them found a statistically significant inverse association, one with intake of raw vegetables (26), one with intake of cabbage (33), and one with intake of beans, lentils, and peas (28). The 10 studies varied considerably in size (from 25 cases (32) to more than 3,500 cases (25)), in length of follow-up (from 7 years (28) to 20 years (31)), and in location (Japan (26, 29, 32), the United States (10, 25, 28, 31), and Scandinavia (27, 30, 33)). The dietary assessment methods used were heterogeneous, and most of the published articles included only a limited amount of detail on vegetable intake, that is, the number of items or portion sizes. Besides total vegetable intake or total fruit and vegetable intake, some studies investigated raw or cooked green/yellow vegetables, white/pale vegetables, cruciferous vegetables, dark green vegetables, green leafy vegetables, or legumes (10, 26, 30, 32, 33). None of these analyses showed statistically significant associations with any vegetable category. Those studies covering dark green vegetables and legumes (10, 30) had only small numbers of cases (65 and 163, respectively), and the latter was limited to male smokers in Finland (30), who reported considerably less vegetable intake than participants in our study. Our analysis was based on over 500 cases, and the dietary assessment instrument was quantitative and detailed. However, inverse associations between dark green vegetables and legumes were only suggestive overall.


View this table:
[in this window]
[in a new window]

  		TABLE 4 Results from prospective studies of the relation between vegetable consumption and pancreatic cancer risk

 
Because of the high fatality rate of pancreatic cancer, case-control studies have often necessarily relied on proxy interviews, which, in addition to possible recall bias in dietary reporting, is a major limitation of this study design for studying pancreatic cancer. This might be the reason why the findings of case-control studies of vegetable intake and pancreatic cancer risk conducted so far have been much more inconsistent. Of 15 such studies, a statistically significant inverse association was reported for one type of vegetable or another in nine (34–42), and no association was reported in six (43–49). With regard to particular vegetable groups, the findings were similarly inconsistent for dark green or dark leafy vegetables (34, 38, 41) and legumes or beans (34, 36, 38, 41). The findings for cruciferous vegetables (34–36, 38, 47, 50) showed somewhat greater agreement among studies, although statistically significant results were seen in only three (35, 36, 41). Our findings only suggested an inverse association for dark green vegetables and legumes. Dark green vegetables in our study included dark green cruciferous vegetables; however, we found no evidence of an association with cruciferous vegetables alone.

To our knowledge, only one case-control study (41) has reported associations between vegetable intake and pancreatic cancer risk by ethnicity. A larger pancreatic cancer risk reduction in the highest quartile of cruciferous vegetable intake was found for Whites than for Blacks. In contrast, in our study a decreased risk with higher intake of dark green vegetables (which included dark green cruciferous vegetables) was seen predominantly in African Americans. Nevertheless, both studies suggested the possibility of ethnic differences in the strength of the association.

We are not aware of any other study that has investigated associations between vegetable intake and pancreatic cancer risk by BMI. In our study, obesity (BMI ≥30 kg/m2) was associated with increased risk in men but not in women (unpublished data). We found that after stratification, the risk reductions with higher intakes of total and light green vegetables and with legumes were evident in overweight/obese participants only. This association was slightly stronger in men.

In an earlier analysis, we found a positive association between consumption of red and processed meat and pancreatic cancer in the Multiethnic Cohort Study (23). Therefore, we included these variables in the regression models. However, these adjustments changed the relative risks only slightly, and stratification by processed meat intake showed no differences across strata.

Various constituents of vegetables have been proposed to have anticancer effects, some of which are specific to a particular vegetable group, like the isothiocyanates in cruciferous vegetables, or are common in vegetables generally, like dietary fiber (17). Antioxidation, effects on cell differentiation, altered estrogen metabolism, effects on DNA methylation, and increased apoptosis are among the mechanisms that may underlie the anticancer effects. Our dark green vegetable group included green leafy vegetables, which contain lutein, an antioxidant, and are rich sources of folic acid, which might influence cancer development through its role in DNA methylation (17). The hypothesis of a role of folate metabolism in pancreatic cancer etiology has been proposed by other researchers (51, 52). Dark cruciferous vegetables like broccoli were also included in our dark green vegetable group. Evidence for cancer-preventive effects of isothiocyanates in experimental animal studies is convincing (53). Isothiocyanates are released from glucosinolates in the plant through enzymatic reactions that occur when plant cells are destroyed (54, 55). Different pathways for chemopreventive effects of a variety of isothiocyanates have been investigated for pancreatic cancer and other cancers (56, 57). Phytoestrogens, especially isoflavones in soy, are characteristic compounds of legumes. Isoflavones have been linked to cancer via their estrogenic properties, being able to bind to estrogen receptors (55). Interestingly, a role of estrogen exposure in pancreatic cancer etiology has been hypothesized, and studies of parity and reproductive factors and pancreatic cancer risk in women have given some (though so far inconclusive) support to the hypothesis (58–60). Protease inhibitors, saponins, phytosterols, and inositol hexaphosphate are further components of legumes which have been shown to have anticancer effects in animals or in vitro (17).

This analysis had some potential limitations. First is the applicability of the results to the US population in general. Although the study sample was only from Hawaii and California, the cohort was designed to be population-based in order to maximize the generalizability of findings (20). A second issue relates to measurement error. Self-reported dietary data are measured with error (61), and measurement error certainly is present in our QFFQ data. We attempted to minimize this limitation through the rigorous design of the questionnaire and by emphasizing densities (intake measured as g/1,000 kcal/day) for the analyses, which we (21) and other researchers (24) have shown to be better correlated with more accurate measurements of dietary intake.

On the other hand, our study had several strengths. First was the large sample size, which resulted in considerable statistical power. Second was the comprehensive dietary assessment instrument, which, although prone to measurement error as noted above, provided a thorough assessment of vegetable intake. The exhaustive food list, in combination with the breakdown of mixed recipes to the ingredient level and the availability of a food composition table developed and maintained specifically for this multiethnic population, resulted in detailed and clearly defined vegetable groups. Third, the QFFQ assessed the whole diet, which permitted adjustment for energy intake. Fourth, because of the multiethnic background of the participants, the cohort included considerable dietary heterogeneity, which facilitated the ability to identify meaningful associations. Fifth, the prospective study design ruled out the problem of recall bias, which can influence findings from case-control studies of diet and pancreatic cancer.

In conclusion, results from this analysis of the Multiethnic Cohort Study show no evidence of an inverse association between vegetable intake and pancreatic cancer overall. However, they do suggest the possibility that vegetables may afford some protection against the disease in persons at higher risk, namely current smokers and overweight or obese persons. These findings should be confirmed in other studies.


   ACKNOWLEDGMENTS
 
This work was supported in part by grant R37 CA054281 from the National Cancer Institute, US Department of Health and Human Services. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the agency.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Jemal A, Clegg LX, Ward E, et al. (2004) Annual report to the nation on the status of cancer, 1975 –2001, with a special feature regarding survival. Cancer 101:3–27.[CrossRef][Web of Science][Medline]
  2. American Cancer Society. (2005) Cancer facts and figures 2005. (American Cancer Society, Atlanta, GA).
  3. Li D, Xie K, Wolff R, et al. (2004) Pancreatic cancer. Lancet 363:1049–57.[CrossRef][Web of Science][Medline]
  4. Nilsen TI and Vatten LJ. (2000) A prospective study of lifestyle factors and the risk of pancreatic cancer in Nord-Trøndelag, Norway. Cancer Causes Control 11:645–52.[CrossRef][Web of Science][Medline]
  5. Lowenfels AB and Maisonneuve P. (2004) Epidemiology and prevention of pancreatic cancer. Jpn J Clin Oncol 34:238–44.[Abstract/Free Full Text]
  6. Klein AP, Brune KA, Petersen GM, et al. (2004) Prospective risk of pancreatic cancer in familial pancreatic cancer kindreds. Cancer Res 64:2634–8.[Abstract/Free Full Text]
  7. Vimalachandran D, Ghaneh P, Costello E, et al. (2004) Genetics and prevention of pancreatic cancer. Cancer Control 11:6–14.[Medline]
  8. Calle EE, Murphy TK, Rodriguez C, et al. (1998) Diabetes mellitus and pancreatic cancer mortality in a prospective cohort of United States adults. Cancer Causes Control 9:403–10.[CrossRef][Web of Science][Medline]
  9. Everhart J and Wright D. (1995) Diabetes mellitus as a risk factor for pancreatic cancer. A meta-analysis. JAMA 273:1605–9.[Abstract/Free Full Text]
  10. Shibata A, Mack TM, Paganini-Hill A, et al. (1994) A prospective study of pancreatic cancer in the elderly. Int J Cancer 58:46–9.[Web of Science][Medline]
  11. Stolzenberg-Solomon RZ, Pietinen P, Taylor PR, et al. (2002) A prospective study of medical conditions, anthropometry, physical activity, and pancreatic cancer in male smokers (Finland). Cancer Causes Control 13:417–26.[CrossRef][Web of Science][Medline]
  12. Batty GD, Shipley MJ, Marmot M, et al. (2004) Diabetes status and post-load plasma glucose concentration in relation to site-specific cancer mortality: findings from the original Whitehall study. Cancer Causes Control 15:873–81.[CrossRef][Web of Science][Medline]
  13. Huxley R, Ansary-Moghaddam A, de Gonzales A Berrington, et al. (2005) Type-II diabetes and pancreatic cancer: a meta-analysis of 36 studies. Br J Cancer 92:2076–83.[CrossRef][Web of Science][Medline]
  14. de Gonzales A Berrington, Sweetland S, Spencer E. (2003) A meta-analysis of obesity and the risk of pancreatic cancer. Br J Cancer 89:519–23.[CrossRef][Web of Science][Medline]
  15. Ghadirian P, Lynch HT, Krewski D. (2003) Epidemiology of pancreatic cancer: an overview. Cancer Detect Prev 27:87–93.[CrossRef][Web of Science][Medline]
  16. Liu RH. (2004) Potential synergy of phytochemicals in cancer prevention: mechanism of action. J Nutr 134:suppl, 3479S–85S.[Abstract/Free Full Text]
  17. Steinmetz KA and Potter JD. (1996) Vegetables, fruit, and cancer prevention: a review. J Am Diet Assoc 96:1027–39.[CrossRef][Web of Science][Medline]
  18. Key TJ, Schatzkin A, Willett WC, et al. (2004) Diet, nutrition and the prevention of cancer. Public Health Nutr 7:187–200.[CrossRef][Web of Science][Medline]
  19. World Cancer Research Fund, American Institute for Cancer Research. (1997) Food, nutrition and the prevention of cancer: a global perspective. (American Institute for Cancer Research, Washington, DC).
  20. Kolonel LN, Henderson BE, Hankin JH, et al. (2000) A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol 151:346–57.[Abstract/Free Full Text]
  21. Stram DO, Hankin JH, Wilkens LR, et al. (2000) Calibration of the dietary questionnaire for a multiethnic cohort in Hawaii and Los Angeles. Am J Epidemiol 151:358–70.[Abstract/Free Full Text]
  22. Health World Organization. (1990) International classification of diseases for oncology. Second Edition (World Health Organization, Geneva, Switzerland).
  23. Nöthlings U, Wilkens LR, Murphy SP, et al. (2005) Meat and fat intake as risk factors for pancreatic cancer: The Multiethnic Cohort Study. J Natl Cancer Inst 97:1458–65.[Abstract/Free Full Text]
  24. Kipnis V, Subar AF, Midthune D, et al. (2003) Structure of dietary measurement error: results of the OPEN biomarker study. Am J Epidemiol 158:14–21.[Abstract/Free Full Text]
  25. Coughlin SS, Calle EE, Patel AV, et al. (2000) Predictors of pancreatic cancer mortality among a large cohort of United States adults. Cancer Causes Control 11:915–23.[CrossRef][Web of Science][Medline]
  26. Inoue M, Tajima K, Takezaki T, et al. (2003) Epidemiology of pancreatic cancer in Japan: a nested case-control study from the Hospital-based Epidemiologic Research Program at Aichi Cancer Center (HERPACC). Int J Epidemiol 32:257–62.[Abstract/Free Full Text]
  27. Isaksson B, Jonsson F, Pedersen NL, et al. (2002) Lifestyle factors and pancreatic cancer risk: a cohort study from the Swedish Twin Registry. Int J Cancer 98:480–2.[CrossRef][Web of Science][Medline]
  28. Mills PK, Beeson WL, Abbey DE, et al. (1988) Dietary habits and past medical history as related to fatal pancreas cancer risk among Adventists. Cancer 61:2578–85.[CrossRef][Web of Science][Medline]
  29. Sauvaget C, Nagano J, Hayashi M, et al. (2003) Vegetables and fruit intake and cancer mortality in the Hiroshima/Nagasaki Life Span Study. Br J Cancer 88:689–94.[CrossRef][Web of Science][Medline]
  30. Stolzenberg-Solomon RZ, Pietinen P, Taylor PR, et al. (2002) Prospective study of diet and pancreatic cancer in male smokers. Am J Epidemiol 155:783–92.[Abstract/Free Full Text]
  31. Zheng W, McLaughlin JK, Gridley G, et al. (1993) A cohort study of smoking, alcohol consumption, and dietary factors for pancreatic cancer (United States). Cancer Causes Control 4:477–82.[CrossRef][Web of Science][Medline]
  32. Khan MM, Goto R, Kobayashi K, et al. (2004) Dietary habits and cancer mortality among middle aged and older Japanese living in Hokkaido, Japan by cancer site and sex. Asian Pac J Cancer Prev 5:58–65.[Medline]
  33. Larsson SC, Hakansson N, Naslund I, et al. (2006) Fruit and vegetable consumption in relation to pancreatic cancer risk: a prospective study. Cancer Epidemiol Biomarkers Prev 15:301–5.[Abstract/Free Full Text]
  34. Chan JM, Wang F, Holly EA. (2005) Vegetable and fruit intake and pancreatic cancer in a population-based case-control study in the San Francisco Bay Area. Cancer Epidemiol Biomarkers Prev 14:2093–7.[Abstract/Free Full Text]
  35. Baghurst PA, McMichael AJ, Slavotinek AH, et al. (1991) A case-control study of diet and cancer of the pancreas. Am J Epidemiol 134:167–79.[Abstract/Free Full Text]
  36. de Mesquita HB Bueno, Maisonneuve P, Runia S, et al. (1991) Intake of foods and nutrients and cancer of the exocrine pancreas: a population-based case-control study in the Netherlands. Int J Cancer 48:540–9.[Web of Science][Medline]
  37. Gold EB, Gordis L, Diener MD, et al. (1985) Diet and other risk factors for cancer of the pancreas. Cancer 55:460–7.[CrossRef][Web of Science][Medline]
  38. Ji BT, Chow WH, Gridley G, et al. (1995) Dietary factors and the risk of pancreatic cancer: a case-control study in Shanghai, China. Cancer Epidemiol Biomarkers Prev 4:885–93.[Abstract]
  39. Lyon JL, Slattery ML, Mahoney AW, et al. (1993) Dietary intake as a risk factor for cancer of the exocrine pancreas. Cancer Epidemiol Biomarkers Prev 2:513–18.[Abstract/Free Full Text]
  40. Mack TM, Yu MC, Hanisch R, et al. (1986) Pancreas cancer and smoking, beverage consumption, and past medical history. J Natl Cancer Inst 76:49–60.[Web of Science][Medline]
  41. Silverman DT, Swanson CA, Gridley G, et al. (1998) Dietary and nutritional factors and pancreatic cancer: a case-control study based on direct interviews. J Natl Cancer Inst 90:1710–19.[Abstract/Free Full Text]
  42. Ghadirian P, Baillargeon J, Simard A, et al. (1995) Food habits and pancreatic cancer: a case-control study of the Francophone community in Montreal, Canada. Cancer Epidemiol Biomarkers Prev 4:895–9.[Abstract]
  43. Soler M, Chatenoud L, La Vecchia C, et al. (1998) Diet, alcohol, coffee and pancreatic cancer: final results from an Italian study. Eur J Cancer Prev 7:455–60.[CrossRef][Web of Science][Medline]
  44. La Vecchia C, Negri E, D'Avanzo B, et al. (1990) Medical history, diet and pancreatic cancer. Oncology 47:463–6.[Web of Science][Medline]
  45. Mizuno S, Watanabe S, Nakamura K, et al. (1992) A multi-institute case-control study on the risk factors of developing pancreatic cancer. Jpn J Clin Oncol 22:286–91.[Abstract/Free Full Text]
  46. Howe GR, Jain M, Miller AB. (1990) Dietary factors and risk of pancreatic cancer: results of a Canadian population-based case-control study. Int J Cancer 45:604–8.[Web of Science][Medline]
  47. Olsen GW, Mandel JS, Gibson RW, et al. (1989) A case-control study of pancreatic cancer and cigarettes, alcohol, coffee and diet. Am J Public Health 79:1016–19.[Abstract/Free Full Text]
  48. Falk RT, Pickle LW, Fontham ET, et al. (1988) Life-style risk factors for pancreatic cancer in Louisiana: a case-control study. Am J Epidemiol 128:324–36.[Abstract/Free Full Text]
  49. Norell SE, Ahlbom A, Erwald R, et al. (1986) Diet and pancreatic cancer: a case-control study. Am J Epidemiol 124:894–902.[Abstract/Free Full Text]
  50. Silverman DT. (2001) Risk factors for pancreatic cancer: a case-control study based on direct interviews. Teratog Carcinog Mutagen 21:7–25.[CrossRef][Web of Science][Medline]
  51. Li D, Ahmed M, Li Y, et al. (2005) 5,10-Methylenetetrahydrofolate reductase polymorphisms and the risk of pancreatic cancer. Cancer Epidemiol Biomarkers Prev 14:1470–6.[Abstract/Free Full Text]
  52. Stolzenberg-Solomon RZ, Pietinen P, Barrett MJ, et al. (2001) Dietary and other methyl-group availability factors and pancreatic cancer risk in a cohort of male smokers. Am J Epidemiol 153:680–7.[Abstract/Free Full Text]
  53. International Agency for Research on Cancer. (2004) Cruciferous vegetables, isothiocyanates and indoles. (IARC handbooks of cancer prevention, vol 9). (IARC Press, Lyon, France).
  54. Bianchini F and Vainio H. (2004) Isothiocyanates in cancer prevention. Drug Metab Rev 36:655–67.[CrossRef][Web of Science][Medline]
  55. Ray A. (2005) Cancer preventive role of selected dietary factors. Indian J Cancer 42:15–24.[Medline]
  56. Kim YS and Milner JA. (2005) Targets for indole-3-carbinol in cancer prevention. J Nutr Biochem 16:65–73.[CrossRef][Web of Science][Medline]
  57. Nishikawa A, Furukawa F, Lee IS, et al. (2004) Potent chemopreventive agents against pancreatic cancer. Curr Cancer Drug Targets 4:373–84.[CrossRef][Web of Science][Medline]
  58. Kreiger N, Lacroix J, Sloan M. (2001) Hormonal factors and pancreatic cancer in women. Ann Epidemiol 11:563–7.[CrossRef][Web of Science][Medline]
  59. Skinner HG, Michaud DS, Colditz GA, et al. (2003) Parity, reproductive factors, and the risk of pancreatic cancer in women. Cancer Epidemiol Biomarkers Prev 12:433–8.[Abstract/Free Full Text]
  60. Teras LR, Patel AV, Rodriguez C, et al. (2005) Parity, other reproductive factors, and risk of pancreatic cancer mortality in a large cohort of U.S. women (United States). Cancer Causes Control 16:1035–40.[CrossRef][Web of Science][Medline]
  61. Subar AF, Kipnis V, Troiano RP, et al. (2003) Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. Am J Epidemiol 158:1–13.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am. J. Clin. Nutr.Home page
U. Nothlings, S. P Murphy, L. R Wilkens, H. Boeing, M. B Schulze, H B. Bueno-de-Mesquita, D. S Michaud, A. Roddam, S. Rohrmann, A. Tjonneland, et al.
A food pattern that is predictive of flavonol intake and risk of pancreatic cancer
Am. J. Clinical Nutrition, December 1, 2008; 88(6): 1653 - 1662.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
G. Bobe, S. J. Weinstein, D. Albanes, T. Hirvonen, J. Ashby, P. R. Taylor, J. Virtamo, and R. Z. Stolzenberg-Solomon
Flavonoid Intake and Risk of Pancreatic Cancer in Male Smokers (Finland)
Cancer Epidemiol. Biomarkers Prev., March 1, 2008; 17(3): 553 - 562.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
U. Nothlings, S. P. Murphy, L. R. Wilkens, B. E. Henderson, and L. N. Kolonel
Flavonols and Pancreatic Cancer Risk: The Multiethnic Cohort Study
Am. J. Epidemiol., October 15, 2007; 166(8): 924 - 931.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
165/2/138    most recent
kwj366v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Nöthlings, U.
Right arrow Articles by Kolonel, L. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nöthlings, U.
Right arrow Articles by Kolonel, L. N.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?