Abstract

BACKGROUND: Few systematic reviews exist examining the relation between vegetables & fruits and esophageal cancer. The most recent publication came from the World Cancer Research Fund (WCRF). The report concluded that both non-starchy vegetables, and total fruits probably protect against esophageal cancer. However, at least half of the cohorts should not have been included in the results from which these conclusions were drawn.
OBJECTIVES:

• To review all prospective studies which provided data about total fruits or vegetables, their subgroups, and specific types of vegetables and fruits, in relationship with esophageal cancer risk, disease progression and mortality/survival.
• To define the amount of consumption found to be related with the described effects on cancer.
• To define possible effect modification by confounders.

DATA SOURCE: The Pubmed database was searched (No start date - Sep 11, 2009) for relevant articles using the keywords "cancer, neoplasm, or neoplasms" combined with "vegetables, vegetable, fruit, or fruits" and a fair amount of other keywords. The exact search term is described in the methods.
Prospective studies published in the English language were included. Reference lists were searched for additional articles.
RESULTS: 12 articles were found which provided information about 10 different cohorts. Of these, 0 articles were excluded.
Results are described when any evidence for an association - as defined in the Methods - was found. In addition, data about total vegetables or fruits is described.

Total disease risk.

• Data about total vegetables was provided by 5 cohorts, including a total of 2,865 cases. No evidence was found for an association between total vegetables and esophageal cancer risk (RR = 1.00).
• Data about total fruits was provided by 5 cohorts, including a total of 2,865 cases. Total fruit possibly protects against esophageal cancer risk (RR = 0.78). No level of consumption could be defined for this effect. Though the evidence was stronger for Asian populations, effect sizes were similar between Asian and European cohorts.
• Data about green (leafy) vegetables was provided by 3 cohorts, including a total of 497 cases. Suggestive evidence was found for a protective effect of green (leafy) vegetables against esophageal cancer risk (RR = 0.65). No level of consumption could be defined for this effect.

Advanced stage/metastatic disease risk or disease progression.
No data was found.
Mortality risk.

• No data was found about the relation between total vegetables and esophageal cancer death.
• Data about total fruits was provided by 1 cohort, including 80 cases. No evidence was found for an association between total fruits and esophageal cancer death.

CONCLUSION: Total fruit possibly protects against esophageal cancer risk (- 22%). Suggestive evidence was found for a protective effect of green (leafy) vegetables (- 35%), but no association was found with total vegetables. No level of consumption could be defined for the associations, because few articles presented risks in grams, servings, or cups. No evidence was found for an association with any specific vegetable or fruit item. In addition, little data is available about disease progression or mortality end points.

Introduction.

Conclusion from the WCRF: In 2007, the WCRF published a systematic review with information about a large amount of life style variables in relation to a large variety of cancer types (World Cancer Research Fund). The WCRF concluded that both nonstarchy vegetables and fruits probably protect against esophageal cancer.
Results from the WCRF: The 2 following figures show the cohort studies which were included in the WCRF results. Four cohorts were found for both the analysis with vegetables, and fruits.

The WCRF included debatable results:
-For total vegetables, 2 out of 4 studies (Yu Y; Guo W) should not have been included in the results, leaving 2 cohorts with a RR > 1. In addition, Hirayama T examined the relation with green-yellow vegetables, and not total (nonstarchy) vegetables.
-For total fruits, 3 out of 4 studies should not have been included in the analysis (Yu Y; Chyou PH; Guo W).

The reasons why these articles should not have been included, can be found in the following table:

Summary: The WCRF included studies which should not have been included, because of various reasons. Exclusion of these articles, leaves 2 cohorts for the analysis with total vegetables (both RR's > 1), and leaves only 1 cohort for the analysis with total fruits (significant protective effect). This means that the evidence for a protective effect of total vegetables is based on results from case-control studies only.

|References:
-Chyou PH, Nomura AMY, Stemmermann GN. Diet, alcohol, smoking and cancer of the upper aerodigestive tract: a prospective study among Hawaii Japanese men. Int J Cancer 1995;60:616- 21. Link.
-Guo W, Blot WJ, Li JY, et al. A nested case control study of oesophageal and stomach cancers in the Linxian nutrition intervention trial. Int J Epidemiol 1994;23:444-50. Link.
-World Cancer Research Fund / American Institute for Cancer Research. Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. Washington DC: AICR, 2007. Link.
-Tran GD, Sun XD, Abnet CC, et al. Prospective study of risk factors for esophageal and gastric cancers in the Linxian general population trial cohort in China. Int J Cancer 2005;113:456-63. Link.
-Yu Y, Taylor PR, Li JY, et al. Retrospective cohort study of risk-factors for esophageal cancer in Linxian, People's Republic of China. Cancer Causes Control 1993;4:195-202. Link.|

Total vegetables and esophageal cancer.

Esophageal cancer risk: Data about the relation with total vegetables was provided by 5 cohorts, including a total of 2,865 cases. Two articles about The EPIC Study were found, and since the articles published data about different histologic types of cancer, they were both included.
A nonsignificant protective effect was found in one cohort (Yamaji T [8]). No (non)significant associations were found in any other cohort. The average RR = 1.00 (excluding incomplete data from Boeing H [5]).

Inclusion of intermediate levels of consumption:
Data about results published in grams, servings, or cups was restricted to findings from 2 cohorts. A significantly increased risk among women was found at the level of consumption of 320-440 g/day in one cohort (George SM [7]). No other (non)significant effects were found in any cohort at any level of consumption.

Esophageal cancer mortality: No data was found.

Conclusion: A nonsignificant protective effect was found in one cohort, but no other (non)significant associations were found with esophageal cancer risk. No data was found about cancer mortality. No evidence was found for an association between total vegetables and esophageal cancer risk or mortality.

Total fruit and esophageal cancer.

Esophageal cancer risk: Data about the relation with total fruit was provided by 5 cohorts, including a total of 2,865 cases. Two articles about The EPIC Study were found, and since the articles published data about different histologic types of cancer, they were both included.
Significant protective effects were found in 2 cohorts, including 2,059 cases (Tran G [2], Fan Y [9]). In addition, nonsignificant protective effect were found in 1 cohort (Fan Y [9]), and in another cohort among men only (George SM [7]), including another 579 cases. The average RR = 0.78 (excluding incomplete data from Boeing H [5]).

Inclusion of intermediate levels of consumption:
Data about results published in grams, servings, or cups was restricted to findings from 2 cohorts. Though trends were nonsignificantly protective in both cohorts among men, no (non)significant effects were related to any level of consumption.

Effect modification: (Non)significant protective effects were found in all 3 Asian cohorts (2, 8, 9), but protective effects among the Western cohorts (5, 7) were restricted to a nonsignificant protective effect in only one of these cohorts, and among men only. Average RR's were 0.78, and 0.81 for Asian, and Western cohorts, respectively.

Esophageal cancer mortality: Data about the relation with total fruit was provided by 1 cohort, including 80 cases (4). A nonsignificant protective effect was found.

Conclusion: Significant protective effects of total fruit against esophageal cancer risk were found in 2 cohorts of moderate size, including 72% of all cases. In addition, nonsignificant protective effects were found in 2 other cohorts, including another 20% of all cases. Total fruit possibly protects against esophageal cancer risk (- 22%). No level of consumption could be defined for this effect. Though the evidence was restricted to findings among Asian cohorts - according to the methodology used - effect sizes were similar between Asian and Western cohorts. Data about cancer mortality was restricted to findings from one cohort. No evidence was found for an association between total fruits and esophageal cancer mortality.