Citrus Fruit Intake and Pancreatic Cancer Risk: A Quantitative Systematic Review
Bae, Jong-Myon MD, PhD*; Lee, Eun Ja MD, PhD†; Guyatt, Gordon MD, MSc‡
From the *Department of Preventive Medicine, Cheju National University College of Medicine, Jejudo; †Department of Radiology, Myongji Hospital, Kwandong University, College of Medicine, Koyang, Korea; and ‡CLARITY Research Group, Department of Clinical Epidemiology & Biostatistics, Health Sciences Centre, McMaster University, Hamilton, Ontario, Canada.
Received for publication April 15, 2008; accepted July 31, 2008.
Reprints: Jong-Myon Bae, MD, PhD, 1-Ara-1-dong Jeju, Jejudo, 690-756, Republic of Korea (e-mail: email@example.com).
This study was supported by the research grant from the Chuongbong Academic Research Fund of the Cheju National University in 2007.
Objectives: The purpose of this systematic review was to investigate the association between dietary intake of citrus fruits and pancreatic cancer risk.
Methods: Authors searched electronic databases and the reference lists of publications of studies addressing diet and pancreatic cancer up to December 2007. All of the epidemiological studies that obtained individual data on dietary intake of citrus fruits and presented risk estimates of the association between intake of citrus fruits and risk of pancreatic cancer were identified and included. Using general variance-based methods, study-specific odds ratios (ORs)/relative risk and associated confidence interval (CI)/SE for highest versus lowest intake of citrus fruits level were extracted from each article.
Results: Nine articles including 4 case-control studies and 5 cohort studies proved eligible. Overall summary OR using random effect model suggested an inverse association in risk of pancreatic caner with intake of citrus fruits (summary OR, 0.83; 95% CI, 0.70-0.98) with large heterogeneity across studies (I2 = 49.9%).
Conclusions: Pooled results from observational studies showed an inverse association between intake of citrus fruits and the risk of pancreatic cancer, although results vary substantially across studies, and the apparent effect is restricted to the weaker study design (case-control studies).
Because of the poor prognosis and increasing incidence in recent decades, investigation of pancreatic cancer has assumed a position of growing importance.1 Whereas the primary causes in pancreatic cancer have been poorly understood, the only risk factors consistently reported are age and cigarette smoking.2 Despite inconsistent findings across studies,3 the variability in incidence rates in different countries4 suggests that an important role of diet in the development of pancreatic cancer is apparent.5 Indeed, up to 30% to 50% of pancreatic cancers may be attributed to dietary factors.6
Several studies have reported inverse associations with consumption of fruits and vegetables.7-13 In addition, previous reports consistently showed the inverse relationship between vitamin C intake and the risk of pancreatic cancer.9,14-16 However, a systematic review of randomized trials did not support the hypothesis that the use of supplements of vitamin C or E in the doses tested helps prevent and/or treat cancer in the populations tested.17 In addition, the World Cancer Research Fund in association with the American Institute for Cancer Research has proposed the level of evidence of fruits protecting against pancreatic cancer to adjust downward from "probable"6 in 1997 into "limited"18 in 2007. As yet, there is not a specific evidence showing that addition of antioxidants or specific vitamins to the normal diet reduces the risk of pancreatic cancer for normal-risk or high-risk populations.19
Although a summary estimate per 50 g/d intake of citrus fruits for all cancer risk was 0.70 (95% confidence interval [CI], 0.56-0.88),18 citrus fruits are the main source of vitamin C as well as β-cryptoxanthin, which is another major carotenoid in human plasma.20 Recently, a pooled analysis of 7 cohort studies demonstrated an association between β-cryptoxanthin ingestion and lung cancer.21 The inhibition of developing carcinogen-induced cancers by citrus fruits has been investigated in a variety of different animal models, including models for stomach, lung, and skin cancers.22 In addition, a systematic review reported an inverse association with citrus fruit ingestion and stomach cancer risk.23 These considerations suggest a hypothesis that dietary intake of citrus fruits might be associated with a reduced risk of pancreas cancer. We conducted a systematic review to explore this hypothesis.
MATERIALS AND METHODS
A computerized search of the PubMed English language literature on citrus fruit and pancreas cancer yielded no relevant publication to December 2007. We therefore decided to use the key words "fruits" and "citrus." Thus, the search terms were ([pancreas] OR [pancreatic]) AND ([neoplasm] OR [cancer]) AND ([fruit] OR [citrus]). We limited the search to human adults without language restrictions. We searched the 3 major electronic databases: PubMed, Ovid Medline, and EMBASE. In addition, authors reviewed the references cited in the full-text articles and in the relevant review articles or meta-analyses identified in the search.
We chose the following inclusion criteria: (1) comparative epidemiological studies; (2) human adult participants; and (3) addressing the association between fruits intake and pancreas cancer. The full-text articles of all references selected after the applying inclusion criteria were collected. To full-text articles including potential references listed by hand-search, reviewers applied the exclusion criteria: (1) no original data, that is, reviews, meta-analysis; (2) studies not measuring the intake of citrus fruit or citrus juice at the individual level; and (3) studies not reporting the SE of the associated measure of association. The eligibility of each abstract or full-text article was assessed independently by 2 reviewers. Disagreements between reviewers were resolved by consensus.
The following information was extracted for all eligible studies: study design, country of origin, years of enrollment, sampling frame or kinds of control, number of participants, range of age, comparison of exposure level, and potential confounding variables adjusted. From the eligible studies that met the inclusion criteria, estimates of odds ratio (OR)/relative risk (RR), and their associated 95% CIs, were extracted.
Using general variance-based methods,24 study-specific OR/RR and 95% CIs for highest versus lowest intake of citrus fruits level were extracted from each article. For all studies, the reported OR/RR estimate was adjusted for age. The SE of the log OR/RR was calculated from the extracted OR/RR estimates and 95% CIs by using the following equation: SE = [ln(OR/RR_upper) − ln(OR/RR_lower)] / 3.92. For an article showing 90% CIs,25 we substituted 3.29 by 3.92. Where authors provided OR/RR only by sex26-28 or kinds of control (eg, population and hospital controls),25 we used a fixed effect model to obtain a pooled estimate from the individual study (Table 1).
We assessed heterogeneity with I2, which describes the percentage of total variation across studies because of study differences rather than chance.29 I2 with 0% means that there is not an observed heterogeneity. We used the random effect model to calculate the summary OR and its 95% CI24,30 with suspecting heterogeneity. We evaluated the impact of the changes on the pooled ORs by study design and kinds of control group as a prior hypothesis to explain heterogeneity through subgroup analyses.
In an attempt to detect publication bias, we visually examined asymmetry in funnel plot.31 Also, we tested the degree of asymmetry of the funnel plot using the Egger regression asymmetry test.30 We considered the funnel plot to be asymmetric if the intercept of the regression line deviated from 0 with P < 0.10.
We used the Cochrane Collaboration software RevMan 4.2 to analyze the extracted data with fixed or random effects model analysis.32 STATA was used to conduct the Egger regression asymmetry test by using the metabias command.33
The computerized search yielded 148 references of which 54 were included after abstract review. Citation search identified another 329 articles. Of the 383 articles that were obtained for full-text review, we excluded 374 articles based on the exclusion criteria (Fig. 1). Especially, 2 population-based case-control studies3,34 were excluded for not reporting the SE of OR (Table 2). Finally, 9 articles were included in the meta-analysis, including 5 cohort studies26,28,35-37 and 4 case-control studies.25,27,38,39
Some details of the selected studies were shown in Tables 2 and 3. All articles were published in English. Four studies were conducted among residents of United States,26,37-39 3 in Scandinavia,25,35,36 and the remaining 2 study in China26 and Japan,28 respectively. Five of the studies recruited participants in the 1980s and another 4 in the 1990s.
All of the studies matched or adjusted for age, and all except Chan et al39 adjusted for cigarette smoking. Studies differed in adjustment for other possible confounders (Tables 2 and 3). In all of the studies, intake of citrus fruits was part of a broader dietary assessment, and the relation between citrus fruits intake and pancreas cancer was not a primary hypothesis. The adjusted OR/RR for the highest category for citrus fruits intake varied considerably, with the ORs/RRs ranging from 0.3 to 1.12. Three case-control studies25,27,39 reached the usual threshold P = 0.05 in the association between citrus fruits and pancreatic cancer.
Heterogeneity, Pooled Results, and Publication Bias
The overall summary OR using random effect model showed the inverse association in risk of pancreatic cancer with high intake of citrus fruits (summary OR, 0.83; 95% CI, 0.70-0.98; Fig. 2). The I2 showed substantial heterogeneity among all 9 articles (I2 = 49.9%; P = 0.04). We estimated the summary OR based on random effect model and conducted sensitivity analyses by study design and kinds of control group (Table 4).
When we calculated a pooled estimate of risk separately for cohort and case-control studies, the inverse association between citrus fruits intake and pancreas cancer risk remained in 4 case-control studies (summary OR, 0.59; 95% CI, 0.48-0.73; I2 = 0.0%, P = 0.74) but was absent in 5 cohort studies (summary OR, 0.97; 95% CI, 0.86-1.10; I2 = 0.0%, P = 0.81). The subgroup analysis restricting to the population-based control group from the case-control studies suggested that the magnitude of the association decreased but the association remained significant (Table 4).
The visualization of the Begg's funnel plot does not suggest publication bias (Fig. 3). Formal testing using the Egger method does not support the notion of a publication bias (intercept = −1.48, P = 0.16).
The overall summary OR suggests an inverse association between citrus fruits and pancreatic cancer (OR, 0.83; 95% CI, 0.70-0.98; P = 0.04). This conclusion is supported by strengths in the systematic review-explicit eligibility criteria, a comprehensive search, and a rigorous approach to data analysis. Especially, the statistical power could be increased applying the systematic review in pancreatic cancer having a low incidence rate.
Whereas the main concern in the meta-analysis for observational studies is to pose a new hypothesis as an exploratory meta-analysis,40 the limitations of this systematic review are not only those of the primary studies but also the heterogeneity. Firstly, all data come from observational studies with a diversity of study designs, racial population, control groups, and definition of exposure.41 Using the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) working group system of rating quality of evidence, these trials would start as low quality.42 Secondarily, measurement errors of dietary intakes should be considered because the consumption pattern of fruits and vegetables may be different among participants. Finally, the inconsistency in results of studies also decreases the strength of inference. Within the GRADE system, this would decrease the quality of the evidence to very low quality.
Table 4 shows that the study design seems to explain the heterogeneity in study results. The summary OR based on 5 cohort studies suggested no effect of citrus fruit consumption on the incidence of pancreatic cancer. The findings in this meta-analysis accorded with some previous systematic reviews.43-45 Riboli and Norat43 explained these discrepancies by recall and selection biases in case-control studies and by imprecise dietary measurements and limited variability of dietary intakes in cohort studies. For the recall bias in case-control studies, the dominance of proxy interviews in case-control studies of pancreatic cancer with extremely low survival rate would result in the misclassification of dietary information.3 In this study, the effect of this bias would be less because 2 case-control studies25,39 took the information directly from cases, whereas the I2 of summary OR in 4 case-control studies was 0%. For the selection bias resulting from the inclusion of only survivors in case-control studies, controlling the bias would be difficult because of the low survival rate in pancreatic cancer.6 However, the fact that the summary OR with only population-based case-control studies was 0.68 (95% CI, 0.56-0.83) would let the survival effect be less.
Meanwhile, prospective studies assess diet before disease occurrence, avoiding both recall bias and the need for proxy interviews. However, because pancreatic cancer incidence is relatively low, most prospective studies have had too few cases and thus inadequate statistical power to detect the small relative risks expected with dietary exposures.46 The residual confounding or unmeasured confounding in diet measurement has also been challenging because of the difficulty in obtaining accurate dietary exposure information.47 In addition, most dietary studies are based on recent estimates of dietary intake,27 whereas it is well known that development of cancer after exposure to even a potent risk factor takes several decades.48
In summary, although there is limited evidence suggesting that fruits protect against pancreatic cancer,18 the results of this review provide only very low quality evidence supporting association between citrus intake and pancreatic cancer. Furthermore, the apparent association is restricted to the case-control studies. Nevertheless, the results do not exclude a true association, and authors hope that this meta-analysis will be used to shape future research questions by helping us understand the results from past studies.
1. Ghadirian P, Lynch HT, Kreswki D. Epidemiology of pancreatic cancer: an overview. Cancer Detect Prev
2. Li D, Xie K, Wolff R, et al. Pancreatic cancer. Lancet
3. Silverman DT, Swanson CA, Gridley G, et al. Dietary and nutritional factors and pancreatic cancer: a case-control study based on direct interviews. J Natl Cancer Inst
4. Lowenfels AB, Maisonneuve P. Epidemiology and risk factors for pancreatic cancer. Best Pract Res Clin Gastroenterol
5. Lin Y, Tamakoshi A, Hayakawa T, et al. Nutritional factors and risk of pancreatic cancer: a population-based case-control study based on direct interview in Japan. J Gastroenterol
6. World Cancer Research Fund in association with the American Institute of Cancer Research. Food, Nutrition, and Cancer Prevention: A Global Perspective
. Washington DC: AICR; 1997.
7. Gold EB, Gordis L, Diener MD, et al. Diet and other risk factors for cancer of the pancreas. Cancer
8. Mack TM, Yu MC,Hanisch R, et al. Pancreas cancer and smoking, beverage consumption, and past medical history. J NatI Cancer Inst
9. Howe GR, Jain M, Miller AB. Dietary factors and risk of pancreatic cancer: results of a Canadian population-based case-control study. Int J Cancer
10. Mills PK, Beeson WL, Abbey DE, et al. Dietary habits and past medical history as related to fatal pancreas cancer risk among adventists. Cancer
11. Falk RT, Pickle LW, Fontham ET, et al. Life-style risk factors for pancreatic cancer in Louisiana: a case-control study. Am J Epidemiol
12. La Vecchia C, Negri E, D'Avanzo B, et al. Medical history, diet and pancreatic cancer. Oncology
13. Shibata A. A prospective study of pancreatic cancer in the elderly. Int J Cancer
14. Kalapothaki V, Tzonou A, Hsieh CC, et al. Nutrient intake and cancer of the pancreas: a case-control study in Athens, Greece. Cancer Causes Control
15. Ghadirian P, Boyle P, Simard A, et al. Reported family aggregation of pancreatic cancer within a population-based case-control study in the Francophone community in Montreal, Canada. Int J Pancreatol
16. Howe GR, Burch JD. Nutrition and pancreatic cancer. Cancer Causes Control
17. Coulter ID, Hardy ML, Morton SC, et al. Antioxidants vitamin C and vitamin E for the prevention and treatment of cancer. J Gen Intern Med
18. World Cancer Research Fund in association with the American Institute for Cancer Research. Food, Nutrition, Physical Activity, and the Prevention of Cancer: A Global Perspective
. Washington DC: AICR;2007:271-274.
19. Lowenfels AB, Maisonneuve P. Epidemiologic and etiologic factors of pancreatic cancer. Hematol Oncol Clin North Am
20. Mangels AR, Holden JM, Beecher GR, et al. Carotenoid content of fruits and vegetables: an evaluation of analytic data. J Am Diet Assoc
21. Mannisto S, Smith-Warner SA, Spiegelman D, et al. Dietary carotenoids and risk of lung cancer in a pooled analysis of seven cohort studies. Cancer Epidemiol Biomarkers Prev
22. Manthey JA, Guthrie N. Antiproliferative activities of citrus flavonoids against six human cancer cell lines. J Agric Food Chem
23. Bae JM, Lee EJ, Guyatt G. Citrus fruits intake and stomach cancer risk: a quantitative systematic review. Gastric Cancer
24. Deeks J, Altman DG, Bradburn MJ. Statistical methods for examining heterogeneity and combining results from several studies in meta-analysis. In: Egger M, Smith DG, Altman DG, eds. Systematic Reviews in Health Care
. 2nd ed. London, England: BMJ Books; 2007:285-312.
25. Norell SE, Ahlbom A, Erwald R, et al. Diet and pancreatic cancer: a case-control study. Am J Epidemiol
26. Coughlin SS, Calle EE, Patel AV, et al. Predictors of pancreatic cancer mortality among a large cohort of United States Adults. Cancer Causes Control
27. Ji BT, Chow WH, Gridley G, et al. Dietary factors and the risk of pancreatic cancer: a case-control study in Shanghai, China. Cancer Epidemiol Biomarkers Prev
28. Lin Y, Kikuchi S, Tamakoshi A, et al. Dietary habits and pancreatic cancer risk in a cohort of middle-aged and elderly Japanese. Nutr Cancer
29. Higgins JPT, Thompson SG. Quantifying heterogeneity in a metaanalysis. Stat Med
30. Fleiss JL, Gross AJ. Meta-analysis in epidemiology, with special reference to studies of the association between exposure to environmental tobacco smoke and lung cancer: a critique. J Clin Epidemiol
31. Begg CB, Berlin JA. Publication bias and dissemination of clinical research. J Natl Cancer Inst
32. The Cochrane Collaboration. Review Manager Version 4.2 for Windows
. Copenhagen, Denmark: The Nordic Cochrane Centre; 2003.
33. Stata Corporation. Stata Statistical Software. Special Edition. 8.2 for Windows
. College Station, TX: Stata Corporation; 2004.
34. Bueno De Mesquita HB, Maisonneuve P, Runia S, et al. Intake of foods and nutrients and cancer of the exocrine pancreas: a population-based case-control study in The Netherlands. Int J Cancer
35. Stolzenberg-Solomon RZ, Pietinen P, Taylor PR, et al. Prospective study of diet and pancreatic cancer in male smokers. Am J Epidemiol
36. Larsson SC, Hakansson N, Naslund I, et al. Fruit and vegetable consumption in relation to pancreatic cancer risk: a prospective study. Cancer Epidemiol Biomarkers Prev
37. Nothlings U, Murphy SP, Wilkens LR, et al. Dietary glycemic load, added sugars, and carbohydrates as risk factors for pancreatic cancer: the Multiethnic Cohort Study. Am J Clin Nutr
38. Olsen GW, Mandel JS, Gibson RW, et al. Nutrients and pancreatic cancer: a population-based case-control study. Cancer Causes Control
39. Chan JM, Wang F, Holly EA. Vegetable and fruit intake and pancreatic cancer in a population-based case-control study in the San Francisco bay area. Cancer Epidemiol Biomarkers Prev
40. Anello C, Fleiss JL. Exploratory or analytic meta-analysis: should we distinguish between them? J Clin Epidemiol
41. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology. A proposal for reporting. JAMA
42. Schünemann HJ, Jaeschke R, Cook DJ, et al. An official ATS statement. Grading the quality of evidence and strength of recommendations in ATS guidelines and recommendations. Am J Respir Crit Care Med
43. Riboli E, Norat T. Epidemiologic evidence of the protective effect of fruit and vegetables on cancer risk. Am J Clin Nutr
. 2003;78(3 suppl):559S-569S.
44. Smith-Warner SA, Spiegelman D, Yaun SS, et al. Intake of fruits and vegetables and risk of breast cancer. A pooled analysis of cohort studies. JAMA
45. Lunet N, Lacerda-Vieira A, Barros H. Fruit and vegetables consumption and gastric cancer: a systematic review and meta-analysis of cohort studies. Nutr Cancer
46. Nothlings U, Wilkens LR, Murphy SP, et al. Meat and fat intake as risk factors for pancreatic cancer: the Multiethnic Cohort Study. J Natl Cancer Inst
47. Lowenfels AB, Maisonneuve P. Risk factors for pancreatic cancer. J Cell Biochem
48. Michaud DS, Skinner HG, Wu K, et al. Dietary patterns and pancreatic cancer risk in men and women. J Natl Cancer Inst
This article has been cited 8 time(s).
British Journal of CancerThe role of Mediterranean diet on the risk of pancreatic cancerBritish Journal of Cancer
Clinical Cancer ResearchDown-regulation of ZIP4 by RNA Interference Inhibits Pancreatic Cancer Growth and Increases the Survival of Nude Mice with Pancreatic Cancer XenograftsClinical Cancer Research
Cancer Causes & ControlCitrus fruit and cancer risk in a network of case-control studiesCancer Causes & Control
Cancer Causes & ControlDietary habits and risk of pancreatic cancer: an Italian case-control studyCancer Causes & Control
Cancer Causes & ControlIntake of folate, vitamins B-6, B-12 and methionine and risk of pancreatic cancer in a large population-based case-control studyCancer Causes & Control
Medizinische KlinikThe S3 Guideline Exocrine Pancreatic CancerMedizinische Klinik
Annals of EpidemiologyNutrient-based dietary patterns and pancreatic cancer riskAnnals of Epidemiology
Journal of Breast CancerCitrus Fruit Intake and Breast Cancer Risk: A Quantitative Systematic ReviewJournal of Breast Cancer
citrus fruit; pancreas neoplasm; meta-analysis; cancer risk; prevention
© 2009 Lippincott Williams & Wilkins, Inc.
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Highlight selected keywords in the article text.
Data is temporarily unavailable. Please try again soon.
Readers Of this Article Also Read