Diet and Upper Gastrointestinal Malignancies

Accepted Manuscript Diet and Upper Gastrointestinal Malignancies Christian C. Abnet, Douglas A. Corley, Neal D. Freedman, Farin Kamangar

PII: DOI: Reference:

S0016-5085(15)00198-5 10.1053/j.gastro.2015.02.007 YGAST 59605

To appear in: Gastroenterology Accepted Date: 5 February 2015 Please cite this article as: Abnet CC, Corley DA, Freedman ND, Kamangar F, Diet and Upper Gastrointestinal Malignancies, Gastroenterology (2015), doi: 10.1053/j.gastro.2015.02.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. All studies published in Gastroenterology are embargoed until 3PM ET of the day they are published as corrected proofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs.

ACCEPTED MANUSCRIPT

Title: Diet and Upper Gastrointestinal Malignancies Author: Christian C. Abnet1*, Douglas A. Corley2, Neal D. Freedman1, and Farin Kamangar3,4 Affiliations: Nutritional Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD 2 Division of Research, Kaiser Permanente of Northern California, Oakland, CA 3 Department of Public Health Analysis, School of Community Health and Policy, Morgan State University, Baltimore, MD 4 Digestive Oncology Research Center, Digestive Disease Research Institute, Tehran University of Medical Sciences, Tehran, Iran The authors have no competing interests to declare.

SC

RI PT

1

M AN U

This work was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute.

EP

Office: (240) 276-7213 Mobile: (240) 505-6299

TE D

* Address for Correspondence: Christian Abnet, PhD, MPH Senior Investigator Nutritional Epidemiology Branch Division of Cancer Epidemiology and Genetics National Cancer Institute 9609 Medical Center Dr Rm 6e344 MSC 9768 Bethesda MD 20892

AC C

[email protected]

1

ACCEPTED MANUSCRIPT

Abstract Diet is believed to modulate cancer risk and this relationship has been widely studied in the

RI PT

gastrointestinal tract. Observational epidemiologic studies have provided most of the evidence for the effects of diet on cancer risk, because clinical trials to determine nutritional exposures are often impossible, impractical, or unaffordable. Although a few foods or nutrients are

SC

thought to protect against specific types of cancer, it seems clear that the strength and even direction of dietary associations (increasing or decreasing risk) is organ site- and even histology-

M AN U

specific, along the gastrointestinal tract. Although some hypotheses are supported by a substantial body of observational data (drinking hot maté contributes to esophageal cancer), there is not much data to support others. We discuss some highly touted hypotheses and draw interim conclusions about what is known, and what could be done to improve the level of

TE D

evidence. The complex nature of diet and its associations can be productively investigated with disease-specific studies. However, public health recommendations for normal-risk individuals regarding diet and gastrointestinal cancer should probably emphasize the importance of eating

AC C

EP

for overall health, rather than eating specific foods to reduce risk for specific cancers.

2

ACCEPTED MANUSCRIPT

Many dietary exposures have been proposed to protect against or increase risk for gastrointestinal (GI) cancers, including diets, foods, individual nutrients, methods of

RI PT

preparation, and habits of consumption. Rather than a comprehensive review the vast literature published on these theories, we focus on the most critical, interesting, and popular hypotheses about the relationship between food and upper GI cancer (Table 1).

SC

Epidemiologic analysis of risk factors is useful for building a case for the role of a certain exposure in cancer risk, but additional types of research are required to determine mechanisms

M AN U

of action. We discuss results from randomized, controlled trials when possible—this study design provides the strongest evidence for or against a specific hypothesis (data on whether supplementation with a specific food or nutrient, for a specific length of time, at a certain phase of life, alters a specific outcome). Most high-quality data on diet and upper GI malignancies

TE D

come from prospective cohort studies, which assess the association between a measure of typical exposure (through questionnaire or biomarker) and future risk of disease. For uncommon diseases, such as esophageal adenocarcinoma, cohorts must be very large, to allow

EP

prospective assessment of these associations, so in some instances the best available data comes from case–control studies. There have been many publications on some exposures,

AC C

reporting findings from multiple clinical trials, whereas other exposures have only been evaluated in observational studies, of varying quality.

Fruits and Vegetables Esophageal adenocarcinoma, esophageal squamous cell carcinoma, gastric cancer, and some premalignant conditions of the upper GI tract (Barrett’s esophagus) are all associated with fruit

3

ACCEPTED MANUSCRIPT

and vegetable intake, although there may be differences with location and histology. Systematic review of all studies of diet and cancer risk concluded that higher intake of fruits and vegetables probably decreases the risk of esophageal and stomach cancers.1 And some authors

RI PT

recommended a diet that includes at least 400 g/day of total fruit and vegetables.2 A more recent systematic review of 24 cohort studies of gastric cancer suggested there may be some

SC

differences between fruits and vegetables in affecting risk for gastric cancer.3 Given the

limitations of data pooling, they contrasted high vs low intakes, and found a significantly

M AN U

reduced gastric cancer risk with fruit intake (summary relative risk (SRR), 0.90) but not vegetable intake (SRR, 0.96). However, an evaluation of dose–response found similar levels of association for fruits and vegetables: each increase of 100 g of intake/day of fruit was associated with a significant decrease in risk of gastric cancer (SRR, 0.95), and vegetables had a

TE D

comparable, but not statistically significant association with reduced risk (SRR, 0.96).3 Some of the effects for gastric cancer may be specific to tumor type or location, or may be restricted to persons with other risk factors. For example, in a large, prospective European

EP

study of more than 450,000 people, those with higher fruit and vegetable intakes were less likely to develop any gastric cancer (hazard ratio [HR], 0.77; P for trend=.02).4 However, the

AC C

associations were mainly significant for fresh fruit and diffuse-type gastric cancer (HR, 0.59; P for trend=.03), for citrus and gastric cardia cancers (HR, 0.61; P for trend=.01), and for smokers and persons residing in the Northern European countries.4 Similar reductions in risk are described for esophageal carcinomas. A meta-analysis of 12 studies of esophageal adenocarcinoma associated reduced risk with higher intakes of vegetables (SRR, 0.76) and fruit (SRRs, 0.73) .5 A meta-analysis of esophageal squamous cell

4

ACCEPTED MANUSCRIPT

carcinoma studies also associated decreased risk with higher-level consumption of vegetables (SRR, 0.56;) and fruits (SRR, 0.53).6 Patients with esophageal adenocarcinoma have been found to be less likely to consume fruits and vegetables, even after adjustment for intake of other

RI PT

food types (eg, meat).7

In addition to the quantity, the variety of intake may be important. In a prospective

SC

analysis of a large European cohort, within which 475 people developed gastric or esophageal cancer, increasing variety of the types of fruits and vegetables consumed, independent of total

particularly among smokers.8

M AN U

consumption, was inversely associated with the risk of esophageal squamous cell carcinoma—

The mechanisms by which fruits and vegetables reduce cancer risk might involve their ability to prevent development of precancerous conditions. In a cohort study performed in the

TE D

Netherlands, men in the highest vs the lowest quintiles of total vegetable intake had a lower risk for Barrett’s esophagus (HR, 0.66;); the same was true for raw vegetable intake (HR 0.63) However, no associations were found for fruits and no significant associations were found in

EP

women.9 Somewhat similar results were found in 2 community-based case–control studies. In one study, persons in the highest quartile of fruit and vegetable intake were also less likely to

AC C

have Barrett’s esophagus (odds ratio [OR], 0.27). There was evidence for increased effects with increasing intake, even after adjusting for total energy intake.10 In another study, subjects in the third tertile of fruit and vegetable intake also had a lower risk of Barrett’s esophagus (OR, 0.39).11 Similar effects for vegetables were found within a veterans population, in which persons in the highest tertile of intake of dark green vegetables were less likely to have Barrett’s esophagus (OR, 0.46).12 It is not clear whether these associations are directly related

5

ACCEPTED MANUSCRIPT

to chemoprotective effects of these foods, independent from other influences such as decreased acid reflux (itself a strongly associated risk factor for esophageal adenocarcinoma). An effect independent from reflux was observed in 1 study, which found a trend towards a

RI PT

lower risk even among persons with gastroesophageal reflux disease (OR, 0.64).10

The possible chemopreventive effects of fruits and vegetables might involve their high

SC

levels of micronutrients (including antioxidants), which can decrease DNA damage by

scavenging for oxygen radicals. Foods that are high in anti-oxidants tend to be inversely

M AN U

associated with the risk of upper GI malignancies and telomere length. Telomere length is associated with cumulative oxidative stress from environmental factors, and telomeres are shorter in persons with gastric cancer.13, 14 Fruits and vegetables might also have chemopreventive effects because they contain flavones, which inhibit cell process associated

TE D

with carcinogenesis (adhesion, invasion, and migration)—possibly through their effects on focal adhesion kinase and metalloproteinases.15 Fiber may also partially mediate the associations

AC C

Meat

EP

found for fruits and vegetables.

Consumption of red or processed meat has been linked to increased risk of GI cancers. There is a convincing link with colorectal cancers and limited or suggestive evidence for links between some types of meat and other types of GI cancer.1, 16 This long-standing hypothesis has been thoroughly investigated for colorectal cancer, and numerous hypotheses have been put forth to explain this apparent association. Relevant exposures include heterocyclic amines, N-nitroso

6

ACCEPTED MANUSCRIPT

compounds, and polycyclic aromatic hydrocarbons generated during specific cooking methods17, as well as the high heme–iron content of red meat.18 Processed meats have a high

RI PT

nitrate content, which could be particularly relevant to cancer risk. Studies of potential mechanisms have benefited from dietary intake databases, which provide intake estimates of carcinogens from meat consumption. The CHARRED database

SC

(http://dceg.cancer.gov/tools/design/charred) provides such estimates for carcinogenic and mutagenic exposures in meat cooked under different conditions.

M AN U

Studies of red meat intake must account for dietary substitution to fully model the risk associated with meat consumption. Commonly, people who reduce red meat consumption increase consumption of white meat, rather than other protein sources, so simultaneous modelling of these types of meat is important. This is particularly important given the potential

TE D

for white meat to have an independent association with lower risk of cancer.19 Red meat has been frequently linked to esophageal cancer risk of both histologic types in case–control studies, whereas cohort studies have provided weaker evidence.20, 21 Studies

EP

that assess meat components have suggested that heme–iron could be the most relevant factor in this potential association for esophageal adenocarcinoma.22 A similar dichotomy exists for

AC C

gastric cancer; case–control studies have associated red meat with gastric cancer, whereas cohort studies have not.23

There is evidence that interactions between diet and the GI microbiome could affect

cancer risk in response to different food products. Intestinal bacteria can reduce nitrate to nitrite, which can then non-enzymatically form nitrosamines, some of which are organ-specific carcinogens. Furthermore, diet strongly and quickly changes the microbial populations in the

7

ACCEPTED MANUSCRIPT

colon.24 The production of trimethylamine-N-oxide by intestinal microbes from choline in red meat has been linked to atherosclerosis25, 26. However, further studies are needed to establish a

RI PT

link between specific intestinal microbes or their products and cancer risk.

Hot beverages

SC

The relationship between hot drinks, including black and green tea, coffee, and maté (an infusion of the herb Ilex Paraguarensis), and GI cancers have been studied extensively.

M AN U

Epidemiologic studies have been conducted since at least the 1960s and clinical studies date even further back. For example, in 1939, WL Watson reviewed clinical records from 771 esophageal cancer patients and concluded that: “thermal irritation is probably the most constant factor predisposing to the cancer of the esophagus”.27 Despite this long history of

TE D

clinical and epidemiologic studies, the only established association thus far seems to be that of maté drinking and esophageal squamous cell carcinoma.28, 29 Other widely discussed associations, such as that between hot tea drinking and esophageal squamous cell carcinoma,

EP

remain uncertain.29

Hot tea and coffee drinking were proposed to be factors that could cause esophageal

AC C

cancer, mostly squamous type, based on the hypothesis that recurrent thermal injury could cause cancer. Support for this hypothesis came from studies by De Jong and colleagues, which showed that drinking hot beverages could substantially increase the intra-esophageal temperature, and this increase was a function of the initial drinking temperature and more importantly, the size of the sip.30 Drinking 65°C coffee increased the intra-esophageal temperature by 6–12 °C, depending on the sip size.

8

ACCEPTED MANUSCRIPT

However, epidemiologic studies have not consistently shown associations between hot tea or coffee consumption and risk of esophageal cancer.29 There is almost no evidence that coffee increases the risk of esophageal cancer. In fact, most new studies have suggested that

RI PT

coffee may protect against esophageal cancer.31, 32 Whereas several epidemiologic studies have shown an increased risk of esophageal cancer with drinking hot tea, others have shown no

SC

association or even a reduced risk.29 Studies in northern parts of Iran – where esophageal

squamous cell carcinoma and hot tea drinking are common – have shown a relatively strong

M AN U

positive association between the two.33 Similarly, studies performed in Kashmir, India reported a positive association between hot tea drinking and risk of esophageal squamous cell carcinoma.34 However, many other studies, including several large-scale studies from Europe and China, have shown no or even protective associations.32, 35, 36 Therefore, it appears that if

TE D

there is any risk, it is only for those who drink large amounts of burning-hot tea. There are several reasons that extensive research over the past few decades has not produced consistent findings. One reason involves the heterogeneity in the chemical

EP

constituents of tea, coffee, and maté. Although some studies have reported mutagenic effects of tea, coffee, and unprocessed maté herb extracts37, 38, others have reported that they may

AC C

prevent cancer.39, 40 Maté, in particular, can have substantial amounts of mutagenic polycyclic aromatic hydrocarbons.41 Conversely, many studies have reported that some green tea polyphenols, most notably i(-)-epigallocatechin-3-gallate, have anti-neoplastic properties, suppressing cell proliferation, inhibiting angiogenesis, and increasing apoptosis.42 Tea polyphenols have also been reported to act on the immune system, affecting T-cell functions.43 However, most of these studies were performed in cell lines or using animal models. Human

9

ACCEPTED MANUSCRIPT

epidemiologic studies have not consistently associated decreased risk of esophageal cancer with green intake; a recent meta-analysis found essentially no effect on risk.44

RI PT

There have been some troubling issues with study designs and methods (retrospective design, questionnaires with subjective questions and incomplete responses, and lack of information on histologic type of esophageal cancer), incomplete analyses, and data

SC

reporting.29 For example, few studies measured the actual temperature of tea, coffee, or maté that were consumed by participants, or obtained data on amount or frequency of drinking per

M AN U

day, total duration of drinking, or sip size (or an indicator of this). New, well-designed studies are underway with consideration of these factors.45

Other studies have investigated associations of tea, coffee, or maté with other types of GI cancers, such as gastric and colorectal cancer.46, 47 However, those associations are not well

TE D

established. One of the more interesting and perhaps most-established hypotheses is a role for coffee in preventing liver cancer.

EP

Micronutrients, Anti-oxidants, and Multi-vitamins Considering the difficulty of increasing fruit and vegetable intake, at a population level, studies

AC C

have tried to identify which components of fruits and vegetables are beneficial.48 Of the many constituents in fruits and vegetables, antioxidant micronutrients and fiber have likely received the most attention. Many carcinogens are thought to damage DNA by generation of reactive oxygen species. Anti-oxidants such as vitamin C and E, β-carotene, selenium, and others, may prevent this oxidative damage and help maintain a normal redox cellular environment.49 Supplementation or fortification of the diet with micronutrients, either individually or in

10

ACCEPTED MANUSCRIPT

combination, could ensure that people have sufficient intakes of micronutrients and help prevent cancer.

RI PT

Half of adults in the United States (US)50 regularly use micronutrient supplements. Although there is established evidence for the positive effects of micronutrient supplements in subgroups, such as pregnant women51 and patients with GI conditions such as pernicious

anemia,52 there is surprisingly little evidence that the supplements prevent cancers of the GI

SC

tract or other tissues.53, 54 The effects of micronutrient anti-oxidants have been examined in

M AN U

laboratory, observational, and randomized controlled trials.54-57 There is evidence from observational studies for an inverse association between serum levels of β-carotene, retinol, selenium, vitamin E, and other micronutrients and risk of upper GI cancer.58-61 However, foods typically contain multiple micronutrients, so findings for one micronutrient could be

TE D

confounded by another, or by other features of diet and lifestyle. Furthermore, findings in populations, such as central China, that have numerous nutrient deficiencies might not apply to well-nourished populations in other parts of the world.

EP

Prospective, controlled trials are needed to study the effects of micronutrients. However, due to considerable design challenge, few such trials have been conducted for less

AC C

common cancers. Key considerations62 include whether studies should examine single micronutrients, such as vitamin D or a multi-vitamin. How should the vitamins be formulated and what concentrations should be given? Should studies be conducted among nutritionally deficient populations, whom supplementation is most likely to show a benefit, or among wellnourished populations, which already consume the bulk of micronutrient supplements worldwide?

11

ACCEPTED MANUSCRIPT

Perhaps the strongest data in support of micronutrient supplementation and upper GI malignancies comes from a trial of 29,584 participants in Linxian, China, where many people

RI PT

had nutritionally deficencies.63 Participants who received factor D (15 mg β-carotene, 50 µg selenium, and 30 mg α-tocopherol) daily for 5.25 years had a lower risk of gastric cancer at the end of the trial and after an additional 10 -year post-supplementation follow-up period (HR,

SC

0.89). However, whether such findings apply to well-nourished populations is unclear.

Another important issue is the complexity and nearly always prohibitive cost of

M AN U

powering trials large enough to examine effects on particular cancer types. For example, the Physician Health Study II included 14,641 men and observed a beneficial effect of 13.8 ys of multi-vitamin supplementation on overall cancer incidence (HR, 0.92). However, even with this relatively large size, the study had only modest power to examine individual cancers.

TE D

Importantly, results from the few other trials that examined multi-vitamins did not find evidence for an effect, even among the nutritionally deficient Linxian population.64 Alternatively, trials focused on intermediate endpoints can be smaller and cost much

EP

less. However, findings for intermediate endpoints may not necessarily apply to cancer.65 Data from the Pioglitazone or Vitamin E for NASH Study (PIVENS) found that vitamin E

AC C

supplementation improved histologic features of patients with non-alcoholic steatohepatitis over 96 weeks.66 However, the effects of vitamin E supplementation on long term-outcomes, including liver cancer, are not clear.67 Data from observational studies, including prospective cohort studies, might also be evaluated. However, results from these types of studies can differ from those of randomized, controlled trials. Perhaps the most striking example of this occurred in studies of the effects of β-carotene supplements on lung cancer. Observational studies found

12

ACCEPTED MANUSCRIPT

that β-carotene supplementation might protect smokers from lung cancer,68 whereas later trials actually found that β-carotene increased cancer risk.69, 70 Studies of selenium and vitamin

RI PT

E and prostate cancer made analogous findings. The Selenium and Vitamin E Cancer Prevention Trial (SELECT) reported that randomization to these nutrients was associated with nonsignificantly higher prostate cancer risk,71 in contrast to previous findings.

SC

So, despite interesting hypotheses, few large-scale randomized trials have examined the effects of micronutrient supplementation on chronic disease, and even fewer focused on upper

M AN U

GI malignancies. As reflected in current guidelines72 and consensus reports,73 data do not support the widespread use of micronutrient supplements by much of the adult US population, and there is even some evidence that multi-vitamin supplementation might be harmful.

TE D

Vitamin D

Vitamin D is a hormone that in most people is primarily formed by the skin during exposure to ultraviolet light. Diet, including fortified milk in the US, also provides vitamin D.

EP

Vitamin D was discovered because of its activities in maintaining bone health and preventing rickets, but in recent years there has been an explosion of hypotheses about its effects.

AC C

Reduced levels of vitamin D have been proposed to increase risk for cancers of the colorectum, esophagus, pancreas, and other organs. Recently, the US Institute of Medicine comprehensively reviewed our knowledge about vitamin D and concluded that bone health is the only proven benefit of sufficient vitamin D intake.79 Nevertheless, the epidemiological evidence evaluating the association between vitamin and cancer in humans is growing rapidly. Prospective studies

13

ACCEPTED MANUSCRIPT

of serum levels of 25(OH) vitamin D have mostly associated lower concentrations with increased risk for colorectal cancer.80 Vitamin D binds to the vitamin D receptor—a ligand-activated transcription factor that

RI PT

regulates a number of biochemical processes and can reduce cell proliferation.81 Vitamin D is also thought to have substantial effects on the immune system, including stimulating antigenpresenting and T-regulatory cells, increasing expression of anti-microbial peptides, and

SC

regulating T-helper 1 and 2 cells.82 Serum 25(OH)D is the most widely used biomarker of

M AN U

vitamin D exposure. However, researchers have recently started to measure and assess the role of the vitamin D binding protein, which may have an independent or interactive effects on cancer risk.83, 84

Cross sectional case–control studies have examined the role of vitamin D in cancer

TE D

incidence, but the utility of these studies is suspect. Notable differences in findings from crosssectional and prospective studies of vitamin D and breast cancer suggest that lower vitamin D concentrations could be an effect, rather than a cause, of cancer development.85 The discovery

EP

of at least 4 common polymorphisms that are strongly associated with circulating vitamin D concentrations86, 87 has allowed for instrumental variable studies to evaluate vitamin D’s effect

AC C

on cancer risk 88.

Studies of the association between vitamin D levels and cancers of the upper GI tract

have produced mixed results; some studies have associated high, or higher, serum levels of vitamin D with increased risk of esophageal squamous cell carcinoma 89, its precursor lesion esophageal squamous dysplasia90, and pancreatic cancer91. However, no effects were observed on risk for gastric cancer.89 A number of less-common cancers were studied in the Vitamin D

14

ACCEPTED MANUSCRIPT

Pooling Project (VDPP),92 including upper GI cancers93 and cancers of the pancreas.94 The VDPP showed no evidence that higher concentrations of vitamin D protect against either of these

RI PT

cancers. In fact, there was some evidence that vitamin D was associated with increased risk of upper GI cancer in some subgroups. Findings from studies of pancreatic cancer have produced mixed results, reporting that high levels of vitamin D increase risk 91, have no effect on risk 95, or

SC

reduce risk.96 Given the complex and contradictory evidence from observational studies, it seems that large, well-powered clinical trials may be required to resolve the controversy

M AN U

regarding vitamin D and risk of these cancers.

Pickled vegetables

Asian pickled vegetables were suspected to have a role in the etiology of esophageal squamous

TE D

cell carcinoma, due to their high intake in areas of China where this cancer is prevalent. Analyses of cancer registries from the 1960s revealed an extremely high incidence of esophageal cancer in Taihang Mountain regions of China.97 Well-established causes of

EP

esophageal squamous cell carcinoma, such as tobacco and alcohol consumption, could not account for this high incidence98, so other factors,99 particularly intake of pickled vegetables,

AC C

were investigated.

In these areas of China, pickled vegetables were an integral part of the diet of many

families and were eaten 9–12 months of the year.98 In the traditional method of preparation, moist vegetables are kept tightly packed in a jar for a few months, allowing fermentation and growth of fungi and yeast,99, 100 which can generate potentially carcinogenic mycotoxins and N-

15

ACCEPTED MANUSCRIPT

nitroso compounds.99, 101, 102 Pickled vegetables have shown mutagenic and carcinogenic effects in some animal and in vitro studies.99, 103, 104 Nonetheless, a 1993 report from the International Agency for Research on Cancer (IARC)

RI PT

Working Group100 concluded that there was limited evidence from humans and inadequate evidence from animals to support the carcinogenicity of pickled vegetables. Whereas some

SC

human observational studies, including case–control and cohort studies, conducted before 1993 reported an increased risk of esophageal squamous cell carcinoma, others had not.

M AN U

Two meta-analyses summarized the results of human studies of pickled vegetables, one in relation to esophageal cancer105 and the other in relation to gastric cancer.106 The first metaanalysis105, which included 34 studies, showed a statistically significant 2-fold increase in risk of esophageal squamous cell carcinoma (OR, 2.08). Several sensitivity analyses confirmed the

TE D

increased risk, and most subgroup analyses showed a statistically significant association increased risk. However, most data were obtained from retrospective studies, so well-designed prospective studies are warranted. The second meta-analysis106 included 60 studies (49 case–

EP

control and 11 cohort). Thirty of the case–control studies found that pickled vegetables significantly increased risk for gastric cancer, and 1 found that it significantly reduced risk.

AC C

Among prospective studies, 2 found that the pickled vegetables significantly increased risk of gastric cancer, and none found a significant decrease in risk. OR values were 1.52 for an overall association, 1.56 from the case–control studies, and 1.32 from the cohort studies. Since 1993, there has been increasing overall evidence that pickled vegetables might cause esophageal and gastric cancers so it seems reasonable that people be advised to avoid these products or use methods of preparation that avoid production of carcinogens.

16

ACCEPTED MANUSCRIPT

Salt The World Cancer Research Fund/American institute for Cancer Research (WCRF/AICR) said

RI PT

that: “Salt, and also salt-preserved foods, are probably causes of [gastric cancer]”.107 Likewise, a World Health Organization/Food and Agriculture Organization (FAO) expert panel concluded

SC

that salt is a probable cause of gastric cancer.108

Salt intake was initially associated with gastric cancer in ecological studies that

M AN U

compared areas at high and low risk for gastric cancer. A 1965 study by Joossens et al, correlated mortality from stroke and gastric cancer across geographic areas and over time, and proposed that salt might be a shared risk factor for both diseases.109 A 1996 study110 showed a more direct link between salt and gastric cancer. In this study, urine samples collected over 24-

TE D

hr periods, from 39 populations in 24 countries, correlated gastric cancer mortality rates with salt excretion in urine (R values of .70 in men and 0.74 in women; both p < 0.001). It is difficult to measure salt intake in human studies. However, people can report their

EP

preference for salty vs non-salty foods. So, some epidemiologic studies ask the participants about their preference for salty food,111 in addition to, or in lieu of, measuring total salt intake.

AC C

Overall, findings from epidemiologic studies support the association between higher salt intake and increased risk of gastric cancer.112 A meta-analysis of 10 cohort studies, including nearly 269,000 individuals and 1474 diagnosed cases, showed an incrementally increasing risk of gastric cancer with higher salt intake.113Compared to people who consumed low levels of salt, those with moderate and high salt intake were at increased risk for gastric cancer; RR values were 1.41 and 1.68, respectively.113

17

ACCEPTED MANUSCRIPT

There are several mechanisms by which salt might contribute to gastric carcinogenesis. Ingestion of salt causes gastritis in rats, and salt increases the effects of gastric carcinogens such

RI PT

as N-methyl-N-nitro-N-nitrosoguanidine.114, 115 High salt concentration in the stomach can disrupt the mucosal barrier and lead to inflammation and atrophy. Such mucosal damage has been to shown to increase Helicobacter pylori colonization in mice 116 and perhaps humans.117

SC

Mongolian gerbils with cagA-positive H pylori infections placed on a high-salt diet developed more severe gastric inflammation than those on a regulate diet, and also had higher gastric pH

M AN U

and increased loss of parietal cells.118 Furthermore, animals fed a high-salt diet had increased transcription of cagA, compared to those on a regular diet. These findings indicate that a saltrich diet can potentiate the carcinogenic effects of cagA-positive strains of H pylori.118

TE D

Coffee

Despite its widespread popularity, coffee drinking is not typically considered among the pantheon of healthy behaviors. Many people even consider coffee drinking to be their guilty

EP

pleasure. The origins of such feelings likely stem from the presence of the drug caffeine in coffee; side effects such as insomnia, anxiety, and heartburn; withdrawal symptoms including

AC C

headaches and fatigue; and concerns about long-term health effects. However, there is much evidence that coffee drinking has beneficial effects for patients with chronic diseases,119-121 including GI malignancies.

There is most evidence for the effects of coffee on liver cancer.122 However, some studies have found inverse associations with colorectal123, 124 and esophageal cancers125, 126, but generally not gastric cancer.127 The most recent meta-analysis of liver cancer studies found that

18

ACCEPTED MANUSCRIPT

the RR for low-level consumption of coffee was 0.72, and for high-level consumption was 0.44, compared with not drinking coffee.122 Consistent findings have been reported from case–

RI PT

control and prospective cohort studies, in populations worldwide, including regions with distinct spectrums of liver cancer risk factors (China, Japan, and Europe).

In addition to liver cancer, inverse associations have been observed between coffee intake and a range of liver disease markers and outcomes in cross-sectional studies, including

SC

fibrosis,128 levels of liver enzymes129-131, and fatty liver132. In patients with hepatitis C virus

M AN U

infection, coffee was reported to affect response to treatment with peg-interferon and ribavirin133 and progression of advanced-stage disease.134 Coffee drinking has also been associated with incidence and mortality from chronic liver disease and cirrhosis in several cohort studies.135-140

TE D

Such findings are intriguing.141-144 However, a number of important questions must be answered before the association can be considered to be definitive. First, little is known about the mechanisms145 by which coffee protects the liver. Coffee contains more than 1000

EP

compounds, and we don’t know which ones are important. Few studies have evaluated differences in effects of caffeinated vs decaffeinated coffee130, let alone examined the possible

AC C

effects of coffee roasting or brewing methods139 on the compounds in coffee and associations with disease.

Coffee might affect the immune system; coffee drinking was inversely associated with

markers of systemic inflammation in several studies.146-149 Coffee, and compounds within, have also been shown to affect important immune signaling pathways, such as the JNK to STAT and transforming growth factor-β signaling.150-152 Coffee might also affect liver cancer development

19

ACCEPTED MANUSCRIPT

by modulating energy metabolism, because coffee drinking has been inversely associated with diabetes in a large number of epidemiology studies.153 Coffee is also one of the leading sources

RI PT

of micronutrients in the US diet, although, as described elsewhere, the effects of micronutrients on liver cancer are unclear.

Second, nearly all studies have been observational in design, making it difficult to prove

SC

that coffee itself affects liver disease development. Confounding, reverse causality, and bias could account for observed associations, so this hypothesis could be strengthened by

M AN U

randomized trials. Yet, it will be challenging to determine the appropriate setting and design for such studies. For example, most people have a firm opinion as to whether they like drinking coffee or not, so it would difficult to assign people to groups that do or do not drink coffee— particularly for a long-term study. Nonetheless, in spite of increasing numbers of observational

TE D

studies finding inverse associations between coffee drinking and liver cancer, further research is needed to demonstrate that coffee itself affects liver cancer development, and to identify its

Alcohol

EP

potential mechanisms.

AC C

Alcoholic beverages are widely consumed and serve many important roles in economics, society, and social rituals. However, alcohol also poses a major conundrum to public health. Abusive consumption of alcoholic beverages leads to public health problems that include direct effects on the consumer (liver disease, motor vehicle accidents, increased cancer risk) and indirect effects on those around them (social and familial instability, motor vehicle accidents,

20

ACCEPTED MANUSCRIPT

etc). Yet modest consumption of alcoholic beverages has been consistently linked to lower over mortality154, mainly via reduced risk of heart disease.155

RI PT

The association between alcohol consumption and cancer risk is complex. Alcoholic beverages are classified as class 1 carcinogens by the IARC; this was previously construed as the beverages in total, not as ethanol. But the most recent update has expanded the group 1 classification from alcohol consumption to the ethanol in alcoholic beverages and the

SC

acetaldehyde associated with consumption of alcoholic beverages.156 It seems likely that

M AN U

alcoholic beverages have multiple carcinogenic mechanisms, given the wide range of associations with cancer. For example, there seems to be a monotonic, possibly linear, association between higher consumption of alcohol and risk of breast cancer, such that even moderate intake could increase risk.157 On possible mechanism is that alcohol can influence sex

TE D

hormone metabolism.158 In contrast, greater consumption of alcohol has been associated with reduced risk of kidney cancer, even at high levels of alcohol intake.159 The diuretic effect of alcohol may keep urine dilute, thereby reducing exposure of kidney tissues to urinary

EP

mutagens.

In contrast to breast and kidney cancers, alcoholic beverages have a complex set of

AC C

associations with GI cancers (Figure 1). Risk of esophageal squamous cell carcinoma is closely tied to heavy alcohol consumption in Western countries, with a population-attributable fraction of about 72%.160 But in many studies this association does not appear to be monotonic. For example, in the NIH-AARP161 and Million Women cohorts162, abstaining subjects did not have the lowest risk for upper GI squamous cell carcinomas, and subjects who consumed larger amounts of alcohol had higher risks. This association has been reported in other studies, but

21

ACCEPTED MANUSCRIPT

some studies have not classified exposure in metrics that lend themselves to testing for this effect, collapsing modest and non-consuming subjects into a single referent category. It could

RI PT

be possible that the non-drinker category includes subjects who were formerly heavy drinkers, which alters the estimate of risk in this group. Or, modest alcohol consumption could be

confounded by the fact that many people in this category have otherwise healthy diets and

SC

lifestyles. Or the true association could be nonlinear, with protection afforded by an unknown mechanism.

M AN U

Many people of East Asian ancestry have polymorphisms that affect metabolism of ethanol, commonly known as the flushing response.163 Individuals with a single copy of a specific polymorphism in ALDH2 develop symptoms after consumption of ethanol, so they eliminate it from their diet. Subjects that continue to drink alcoholic beverages are at

TE D

extraordinarily high risk of developing esophageal squamous cell carcinoma, with estimated ORs >60.163 Subjects that carry 2 alleles of this polymorphism generally cannot tolerate any ethanol consumption and are therefore protected from cancer. This difference may reflect

EP

differences in the effects of modest alcohol consumption in the general population.164 Alcohol seems to have little or no role in the etiology of esophageal adenocarcinoma. An

AC C

analysis from the Barrett's and Esophageal Adenocarcinoma Consortium (BEACON) found no association between total alcoholic beverage consumption, even high doses (7 drinks/day) and esophageal adenocarcinoma.161 In most studies, adenocarcinomas of the stomach appear not to be associated with alcoholic beverage consumption, but there is some evidence for an adverse effect in subjects with genetic polymorphisms that alter metabolism of ethanol, similar to the effects of ALDH2

22

ACCEPTED MANUSCRIPT

variants on esophageal cancer risk, although the current level of evidence is modest.165, 166 Heavier consumption of alcohol has been causally linked to cancers of colorectum, but with a

RI PT

more modest magnitude than for other types of GI cancers, such as esophageal squamous cell carcinoma. There is some evidence that the increase in risk could be linked to the depletion of

SC

folate with heavy alcohol consumption.

Dietary Patterns

M AN U

Separate from individual foods or food groups, dietary patterns are also associated with cancer risk and may offer a more powerful alternative to food-by-food studies. One approach for evaluating such patterns is the use of principal components analyses, which can identify independent associations for groups of dietary elements that are associated with each other,

TE D

but not associated with other dietary elements. This method assesses shared dietary patterns that occur in the general population. A population-based case–control study in performed in Connecticut, New Jersey, and western Washington State used this type of analysis to evaluate

EP

subtypes of esophageal and gastric cancers, patterns of food intake, and other risk factors (eg, body mass index).167 The study found that dietary patterns with higher intakes of fruits and

AC C

vegetables, as scored by the principal components analysis, were associated with decreased risks for esophageal squamous cell carcinoma (OR, 0.36), esophageal adenocarcinoma (OR, 0.43), and gastric adenocarcinoma (OR, 0.63). The dietary pattern higher in meat and nitrates was strongly associated with an increased risk of esophageal adenocarcinoma (OR, 5.61), and somewhat more weakly with increased risks of esophageal squamous cell carcinoma and noncardia gastric adenocarcinomas (OR, 2.40 for fourth vs first quartiles).

23

ACCEPTED MANUSCRIPT

A study that used these methods to study Barrett’s esophagus found similar results. In that study, a more health-conscious dietary pattern, with higher intakes of fruits, vegetables,

RI PT

and non-fried fish, was associated with a decreased risk of Barrett’s esophagus (OR, 0.35).168 In contrast, subjects with Western-style dietary patterns, including fast-foods and meat, were more than 2-fold more likely to have Barrett’s esophagus (OR, 2.30), although without a strong

SC

dose effect and with non-significant results for the fourth quartile of intake.168

Alternatively, dietary pattern analysis can use a priori dietary patterns that are thought

M AN U

to optimize health outcomes (often defined using knowledge of diet and heart disease), such as the healthy eating index169 or the Mediterranean diet.170 Several recent studies have assessed the association between such patterns and risk of GI cancer. These generally find that pre-

TE D

specified healthy diets are associated with lower risk of GI cancer.

Implications and Future Directions

Diet seems to have an important role in risk for GI cancer, but it is difficult to draw firm

EP

conclusions, except for some strongly adverse exposures, such as some forms of meat and colorectal cancer risk, or the consumption of extremely hot beverages and esophageal cancer

AC C

risk. Future observational studies must be primarily prospective, due to the difficulty of excluding bias when examining evidence derived from cross–sectional case–control studies. Trials can be helpful, but are not always possible, and negative results do not always quell the perception that modulating intake of certain foods or nutrients will lower risk for GI malignancies.

24

ACCEPTED MANUSCRIPT

The potentially protective effect of coffee on liver cancer seems ready to test in a randomized trial, if a suitable population can be identified. Several mechanisms of action are

RI PT

likely to be involved, including effects on the immune system. Future studies of coffee and liver cancer may be more informative if mechanistic analyses are involved. The causes of upper GI cancers are relatively under-studied, compared to cancer types that are more common in high-

SC

income countries. Finally, analyses of dietary patterns are intriguing and might be used to educate the public to eat for overall health, rather than focusing on individual foods or

M AN U

nutrients and their effects on specific cancers, at individual organ sites.

Reference List (First 50 in print; rest online-only in supplemental file)

TE D

1. World Cancer Research Fund / American Institute for Cancer Research. Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. Washington,D.C.: AICR, 2007.

EP

2. Key TJ, Schatzkin A, Willett WC et al. Diet, nutrition and the prevention of cancer. Public Health Nutr 2004;7:187-200.

AC C

3. Wang Q, Chen Y, Wang X et al. Consumption of fruit, but not vegetables, may reduce risk of gastric cancer: results from a meta-analysis of cohort studies. Eur J Cancer 2014;50:1498-1509. 4. Gonzalez CA, Lujan-Barroso L, Bueno-de-Mesquita HB et al. Fruit and vegetable intake and the risk of gastric adenocarcinoma: a reanalysis of the European Prospective Investigation into Cancer and Nutrition (EPIC-EURGAST) study after a longer follow-up. Int J Cancer 2012;131:2910-2919. 5. Li B, Jiang G, Zhang G et al. Intake of vegetables and fruit and risk of esophageal adenocarcinoma: a meta-analysis of observational studies. Eur J Nutr 2014;53:15111521.

25

ACCEPTED MANUSCRIPT

6. Liu J, Wang J, Leng Y et al. Intake of fruit and vegetables and risk of esophageal squamous cell carcinoma: a meta-analysis of observational studies. Int J Cancer 2013;133:473-485.

RI PT

7. Navarro Silvera SA, Mayne ST, Risch H et al. Food group intake and risk of subtypes of esophageal and gastric cancer. Int J Cancer 2008;123:852-860. 8. Jeurnink SM, Buchner FL, Bueno-de-Mesquita HB et al. Variety in vegetable and fruit consumption and the risk of gastric and esophageal cancer in the European Prospective Investigation into Cancer and Nutrition. Int J Cancer 2012;131:E963-E973.

SC

9. Keszei AP, Schouten LJ, Driessen AL et al. Vegetable, fruit and nitrate intake in relation to the risk of Barrett's oesophagus in a large Dutch cohort. Br J Nutr 2014;111:14521462.

M AN U

10. Kubo A, Levin TR, Block G et al. Dietary antioxidants, fruits, and vegetables and the risk of Barrett's esophagus. Am J Gastroenterol 2008;103:1614-1623. 11. Thompson OM, Beresford SA, Kirk EA et al. Vegetable and fruit intakes and risk of Barrett's esophagus in men and women. Am J Clin Nutr 2009;89:890-896. 12. Jiao L, Kramer JR, Rugge M et al. Dietary intake of vegetables, folate, and antioxidants and the risk of Barrett's esophagus. Cancer Causes Control 2013;24:1005-1014.

TE D

13. Hou L, Savage SA, Blaser MJ et al. Telomere length in peripheral leukocyte DNA and gastric cancer risk. Cancer Epidemiol Biomarkers Prev 2009;18:3103-3109.

EP

14. Chung MY, Lim TG, Lee KW. Molecular mechanisms of chemopreventive phytochemicals against gastroenterological cancer development. World J Gastroenterol 2013;19:984993.

AC C

15. Lin SH, Shih YW. Antitumor effects of the flavone chalcone: inhibition of invasion and migration through the FAK/JNK signaling pathway in human gastric adenocarcinoma AGS cells. Mol Cell Biochem 2014;391:47-58. 16. Abid Z, Cross AJ, Sinha R. Meat, dairy, and cancer. Am J Clin Nutr 2014;100:386S-393S. 17. Kim E, Coelho D, Blachier F. Review of the association between meat consumption and risk of colorectal cancer. Nutr Res 2013;33:983-994. 18. Cross AJ, Ferrucci LM, Risch A et al. A large prospective study of meat consumption and colorectal cancer risk: an investigation of potential mechanisms underlying this association. Cancer Res 2010;70:2406-2414. 19. Daniel CR, Cross AJ, Graubard BI et al. Prospective investigation of poultry and fish intake in relation to cancer risk. Cancer Prev Res (Phila) 2011;4:1903-1911. 26

ACCEPTED MANUSCRIPT

20. Choi Y, Song S, Song Y et al. Consumption of red and processed meat and esophageal cancer risk: meta-analysis. World J Gastroenterol 2013;19:1020-1029. 21. Kubo A, Corley DA, Jensen CD et al. Dietary factors and the risks of oesophageal adenocarcinoma and Barrett's oesophagus. Nutr Res Rev 2010;23:230-246.

RI PT

22. O'Doherty MG, Cantwell MM, Murray LJ et al. Dietary fat and meat intakes and risk of reflux esophagitis, Barrett's esophagus and esophageal adenocarcinoma. Int J Cancer 2011;129:1493-1502.

SC

23. Song P, Lu M, Yin Q et al. Red meat consumption and stomach cancer risk: a metaanalysis. J Cancer Res Clin Oncol 2014;140:979-992.

24. Wu GD, Chen J, Hoffmann C et al. Linking long-term dietary patterns with gut microbial enterotypes. Science 2011;334:105-108.

M AN U

25. Koeth RA, Wang Z, Levison BS et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med 2013;19:576-585. 26. Wang Z, Klipfell E, Bennett BJ et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 2011;472:57-63. 27. Watson WL. Cancer of the esophagus: some etiological considerations. Am J Roentgenology 1939;41:420-424.

TE D

28. Andrici J, Eslick GD. Mate consumption and the risk of esophageal squamous cell carcinoma: a meta-analysis. Dis Esophagus 2013;26:807-816.

EP

29. Islami F, Boffetta P, Ren JS et al. High-temperature beverages and foods and esophageal cancer risk--a systematic review. Int J Cancer 2009;125:491-524. 30. De Jong UW, Day NE, Mounier-Kuhn PL et al. The relationship between the ingestion of hot coffee and intraoesophageal temperature. Gut 1972;13:24-30.

AC C

31. Oze I, Matsuo K, Kawakita D et al. Coffee and green tea consumption is associated with upper aerodigestive tract cancer in Japan. Int J Cancer 2014;135:391-400. 32. Zamora-Ros R, Lujan-Barroso L, Bueno-de-Mesquita HB et al. Tea and coffee consumption and risk of esophageal cancer: the European prospective investigation into cancer and nutrition study. Int J Cancer 2014;135:1470-1479. 33. Islami F, Pourshams A, Nasrollahzadeh D et al. Tea drinking habits and oesophageal cancer in a high risk area in northern Iran: population based case-control study. BMJ 2009;338:b929.

27

ACCEPTED MANUSCRIPT

34. Dar NA, Bhat GA, Shah IA et al. Salt tea consumption and esophageal cancer, a possible role of alkaline beverages in esophageal carcinogenesis. Int J Cancer 2014;10.

RI PT

35. Nechuta S, Shu XO, Li HL et al. Prospective cohort study of tea consumption and risk of digestive system cancers: results from the Shanghai Women's Health Study. Am J Clin Nutr 2012;96:1056-1063. 36. Sang LX, Chang B, Li XH et al. Green tea consumption and risk of esophageal cancer: a meta-analysis of published epidemiological studies. Nutr Cancer 2013;65:802-812.

SC

37. Tewes FJ, Koo LC, Meisgen TJ et al. Lung cancer risk and mutagenicity of tea. Environ Res 1990;52:23-33. 38. Dorado G, Barbancho M, Pueyo C. Coffee is highly mutagenic in the L-arabinose resistance test in Salmonella typhimurium. Environ Mutagen 1987;9:251-260.

M AN U

39. Lambert JD, Yang CS. Mechanisms of cancer prevention by tea constituents. J Nutr 2003;133:3262S-3267S. 40. Yang CS, Lambert JD, Ju J et al. Tea and cancer prevention: molecular mechanisms and human relevance. Toxicol Appl Pharmacol 2007;224:265-273.

TE D

41. Kamangar F, Strickland PT, Pourshams A et al. High exposure to polycyclic aromatic hydrocarbons may contribute to high risk of esophageal cancer in northeastern Iran. Anticancer Res 2005;25:425-428. 42. Yang CS, Wang H, Li GX et al. Cancer prevention by tea: Evidence from laboratory studies. Pharmacol Res 2011;64:113-122.

EP

43. Pae M, Wu D. Immunomodulating effects of epigallocatechin-3-gallate from green tea: mechanisms and applications. Food Funct 2013;4:1287-1303.

AC C

44. Sang LX, Chang B, Li XH et al. Green tea consumption and risk of esophageal cancer: a meta-analysis of published epidemiological studies. Nutr Cancer 2013;65:802-812. 45. Pourshams A, Khademi H, Malekshah AF et al. Cohort Profile: The Golestan Cohort Study--a prospective study of oesophageal cancer in northern Iran. Int J Epidemiol 2010;39:52-59. 46. Myung SK, Bae WK, Oh SM et al. Green tea consumption and risk of stomach cancer: a meta-analysis of epidemiologic studies. Int J Cancer 2009;124:670-677. 47. Sun CL, Yuan JM, Koh WP et al. Green tea, black tea and colorectal cancer risk: a metaanalysis of epidemiologic studies. Carcinogenesis 2006;27:1301-1309.

28

ACCEPTED MANUSCRIPT

48. Lampe JW. Health effects of vegetables and fruit: assessing mechanisms of action in human experimental studies. Am J Clin Nutr 1999;70:475S-490S. 49. Stanner SA, Hughes J, Kelly CN et al. A review of the epidemiological evidence for the 'antioxidant hypothesis'. Public Health Nutr 2004;7:407-422.

SC

RI PT

50. Bailey RL, Gahche JJ, Miller PE et al. Why US adults use dietary supplements. JAMA Intern Med 2013;173:355-361.

Table 1. Gastrointestinal cancer sites, hypotheses examined in this review, and apparent risk difference associated with greater exposure, if any.

Fruits vegetables

Esophagus

&

Gastric

Squamous

Adenocarcinoma

Cardia

Noncardia

↓

↓

↓

↓

Meat

↑

Hot beverages

↑

Micronutrients

↓

↔

↔ or ↑

↔ or ↑

Pickles

↑

Salt Coffee

↑

↓

↓

↔ or ↑

↔ or ↑

↑

↑

↔

↓

↓

Liver

↑

↓

↓

↔

↔

↔

↔

AC C

Dietary patterns

↑

↑

EP

Alcohol

↑

TE D

Vitamin D

M AN U

Exposure

↑

Figure 1. Associations between categories of alcoholic beverage intake and risk of upper GI Cancers Note: Data from the NIH-AARP Diet and Health Study cohort. Point estimates marked with an asterisk are significantly different from unity. Adapted from Freedman et al. American Journal of Epidemiology 2007.

29

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

SUPPLEMENTAL (ONLINE-ONLY) REFERENCES:

RI PT

51. Wolff T, Witkop CT, Miller T et al. Folic acid supplementation for the prevention of neural tube defects: an update of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2009;150:632-639. 52. Stabler SP. Clinical practice. Vitamin B12 deficiency. N Engl J Med 2013;368:149-160.

53. Guallar E, Stranges S, Mulrow C et al. Enough is enough: Stop wasting money on vitamin and mineral supplements. Ann Intern Med 2013;159:850-851.

SC

54. Mayne ST, Ferrucci LM, Cartmel B. Lessons learned from randomized clinical trials of micronutrient supplementation for cancer prevention. Annu Rev Nutr 2012;32:369-90.:369-390.

M AN U

55. Seifried HE, McDonald SS, Anderson DE et al. The antioxidant conundrum in cancer. Cancer Res 2003;63:4295-4298. 56. Byers T, Perry G. Dietary carotenes, vitamin C, and vitamin E as protective antioxidants in human cancers. Annu Rev Nutr 1992;12:139-59.:139-159. 57. Kohen R, Nyska A. Oxidation of biological systems: oxidative stress phenomena, antioxidants, redox reactions, and methods for their quantification. Toxicol Pathol 2002;30:620-650.

TE D

58. van PG, Goldbohm RA. Epidemiologic evidence for beta-carotene and cancer prevention. Am J Clin Nutr 1995;62:1393S-1402S.

EP

59. Wei WQ, Abnet CC, Qiao YL et al. Prospective study of serum selenium concentrations and esophageal and gastric cardia cancer, heart disease, stroke, and total death. Am J Clin Nutr 2004;79:80-85. 60. Mark SD, Qiao YL, Dawsey SM et al. Prospective study of serum selenium levels and incident esophageal and gastric cancers. J Natl Cancer Inst 2000;92:1753-1763.

AC C

61. Taylor PR, Qiao YL, Abnet CC et al. Prospective study of serum vitamin E levels and esophageal and gastric cancers. J Natl Cancer Inst 2003;95:1414-1416. 62. Taylor PR, Greenwald P. Nutritional interventions in cancer prevention. J Clin Oncol 2005;23:333-345. 63. Qiao YL, Dawsey SM, Kamangar F et al. Total and cancer mortality after supplementation with vitamins and minerals: follow-up of the Linxian General Population Nutrition Intervention Trial. J Natl Cancer Inst 2009;101:507-518.

ACCEPTED MANUSCRIPT

64. Wang JB, Abnet CC, Fan JH et al. The randomized Linxian Dysplasia Nutrition Intervention Trial after 26 years of follow-up: no effect of multivitamin supplementation on mortality. JAMA Intern Med 2013;173:1259-1261.

RI PT

65. Schatzkin A, Freedman LS, Dorgan J et al. Surrogate end points in cancer research: a critique. Cancer Epidemiol Biomarkers Prev 1996;5:947-953. 66. Sanyal AJ, Chalasani N, Kowdley KV et al. Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl J Med 2010;362:1675-1685.

SC

67. Bjelakovic G, Gluud LL, Nikolova D et al. Antioxidant supplements for liver diseases. Cochrane Database Syst Rev 2011;CD007749. 68. van Poppel G, Goldbohm RA. Epidemiologic evidence for beta-carotene and cancer prevention. Am J Clin Nutr 1995;62:1393S-1402S. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. N Engl J Med 1994;330:1029-1035.

M AN U

69.

70. Omenn GS, Goodman GE, Thornquist MD et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996;334:1150-1155.

TE D

71. Lippman SM, Klein EA, Goodman PJ et al. Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA 2009;301:39-51. 72. Moyer VA. Vitamin, mineral, and multivitamin supplements for the primary prevention of cardiovascular disease and cancer: U.S. Preventive services Task Force recommendation statement. Ann Intern Med 2014;160:558-564. National Institutes of Health State-of-the-science conference statement: multivitamin/mineral supplements and chronic disease prevention. Ann Intern Med 2006;145:364-371.

EP

73.

AC C

74. Wu AH, Tseng CC, Hankin J et al. Fiber intake and risk of adenocarcinomas of the esophagus and stomach. Cancer Causes Control 2007;18:713-722. 75. Zhang Z, Xu G, Ma M et al. Dietary fiber intake reduces risk for gastric cancer: a meta-analysis. Gastroenterology 2013;145:113-120. 76. Coleman HG, Murray LJ, Hicks B et al. Dietary fiber and the risk of precancerous lesions and cancer of the esophagus: a systematic review and meta-analysis. Nutr Rev 2013;71:474-482. 77. Bobe G, Peterson JJ, Gridley G et al. Flavonoid consumption and esophageal cancer among black and white men in the United States. Int J Cancer 2009;125:1147-1154.

ACCEPTED MANUSCRIPT

78. Kubo A, Block G, Quesenberry CP, Jr. et al. Effects of dietary fiber, fats, and meat intakes on the risk of Barrett's esophagus. Nutr Cancer 2009;61:607-616. 79.

Dietary reference intakes for calcium and vitamin D. Washington, D.C.: Institute of Medicine (IOM), 2011.

RI PT

80. Giovannucci E. The epidemiology of vitamin D and cancer incidence and mortality: a review (United States). Cancer Causes Control 2005;16:83-95. 81. Zhang X, Giovannucci E. Calcium, vitamin D and colorectal cancer chemoprevention. Best Pract Res Clin Gastroenterol 2011;25:485-494.

SC

82. Han YP, Kong M, Zheng S et al. Vitamin D in liver diseases: from mechanisms to clinical trials. J Gastroenterol Hepatol 2013;28 Suppl 1:49-55. doi: 10.1111/jgh.12016.:49-55.

M AN U

83. Anic GM, Weinstein SJ, Mondul AM et al. Serum vitamin D, vitamin D binding protein, and risk of colorectal cancer. PLoS One 2014;9:e102966. 84. Weinstein SJ, Purdue MP, Smith-Warner SA et al. Serum 25-hydroxyvitamin D, vitamin D binding protein and risk of colorectal cancer in the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial. Int J Cancer 2014;10. 85. Chlebowski RT. Vitamin D and breast cancer: interpreting current evidence. Breast Cancer Res 2011;13:217.

TE D

86. Ahn J, Yu K, Stolzenberg-Solomon R et al. Genome-wide association study of circulating vitamin D levels. Hum Mol Genet 2010;19:2739-2745. 87. Wang TJ, Zhang F, Richards JB et al. Common genetic determinants of vitamin D insufficiency: a genome-wide association study. Lancet 2010;376:180-188.

EP

88. Schatzkin A, Abnet CC, Cross AJ et al. Mendelian randomization: how it can--and cannot--help confirm causal relations between nutrition and cancer. Cancer Prev Res (Phila) 2009;2:104-113.

AC C

89. Chen W, Dawsey SM, Qiao YL et al. Prospective study of serum 25(OH)-vitamin D concentration and risk of oesophageal and gastric cancers. Br J Cancer 2007;97:123-128. 90. Abnet CC, Chen W, Dawsey SM et al. Serum 25(OH)-vitamin D concentration and risk of esophageal squamous dysplasia. Cancer Epidemiol Biomarkers Prev 2007;16:1889-1893. 91. Stolzenberg-Solomon RZ, Vieth R, Azad A et al. A prospective nested case-control study of vitamin D status and pancreatic cancer risk in male smokers. Cancer Res 2006;66:10213-10219. 92. Helzlsouer KJ. Overview of the Cohort Consortium Vitamin D Pooling Project of Rarer Cancers. Am J Epidemiol 2010;172:4-9.

ACCEPTED MANUSCRIPT

93. Abnet CC, Chen Y, Chow WH et al. Circulating 25-hydroxyvitamin D and risk of esophageal and gastric cancer: Cohort Consortium Vitamin D Pooling Project of Rarer Cancers. Am J Epidemiol 2010;172:94-106.

RI PT

94. Stolzenberg-Solomon RZ, Jacobs EJ, Arslan AA et al. Circulating 25-hydroxyvitamin D and risk of pancreatic cancer: Cohort Consortium Vitamin D Pooling Project of Rarer Cancers. Am J Epidemiol 2010;172:81-93. 95. Stolzenberg-Solomon RZ, Hayes RB, Horst RL et al. Serum vitamin D and risk of pancreatic cancer in the prostate, lung, colorectal, and ovarian screening trial. Cancer Res 2009;69:1439-1447.

The epidemiology and etiology of esophageal cancer in north China. A preliminary report. Chin Med J (Engl ) 1975;1:167-183.

M AN U

97.

SC

96. Wolpin BM, Ng K, Bao Y et al. Plasma 25-hydroxyvitamin D and risk of pancreatic cancer. Cancer Epidemiol Biomarkers Prev 2012;21:82-91.

98. Kamangar F, Chow WH, Abnet CC et al. Environmental causes of esophageal cancer. Gastroenterol Clin North Am 2009;38:27-57. 99. Yang CS. Research on esophageal cancer in China: a review. Cancer Res 1980;40:2633-2644. 100. International Agency for Research on Cancer. Pickled vegetables. In: IARC Monogr Eval Carcinog Risks Hum., ed. 56 ed. 1993:83-113.

TE D

101. Cheng SJ, Sala M, Li MH et al. Promoting effect of Roussin's red identified in pickled vegetables from Linxian China. Carcinogenesis 1981;2:313-319. 102. Zhang WX, Xu MS, Wang GH et al. Quantitative analysis of Roussin red methyl ester in pickled vegetables. Cancer Res 1983;43:339-341.

EP

103. Cheng SJ, Sala M, Li MH et al. Mutagenic, transforming and promoting effect of pickled vegetables from Linxian county, China. Carcinogenesis 1980;1:685-692.

AC C

104. Lu SH, Camus AM, Tomatis L et al. Mutagenicity of extracts of pickled vegetables collected in Linhsien County, a high-incidence area for esophageal cancer in Northern China. J Natl Cancer Inst 1981;66:33-36. 105. Islami F, Ren JS, Taylor PR et al. Pickled vegetables and the risk of oesophageal cancer: a metaanalysis. Br J Cancer 2009;101:1641-1647. 106. Ren JS, Kamangar F, Forman D et al. Pickled food and risk of gastric cancer--a systematic review and meta-analysis of English and Chinese literature. Cancer Epidemiol Biomarkers Prev 2012;21:905-915.

ACCEPTED MANUSCRIPT

107. Wiseman M. The second World Cancer Research Fund/American Institute for Cancer Research expert report. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Proc Nutr Soc 2008;67:253-256.

RI PT

108. World Health Organization/Food and Agriculure Oraganization (WHO/FAO). Diet, nutrition and the prevention of chronic diseases: Report of a joint WHO/FAO expert consultation. World Health Organ Technical Report Series 916 ed. 2003. 109. Joossens JV. La relation entre l'épidémiologie des accidents cérébro-vasculaires et du cancer de I'estomac. L' evolution Medicale 1968;12:381-385.

SC

110. Joossens JV, Hill MJ, Elliott P et al. Dietary salt, nitrate and stomach cancer mortality in 24 countries. European Cancer Prevention (ECP) and the INTERSALT Cooperative Research Group. Int J Epidemiol 1996;25:494-504.

M AN U

111. Kim J, Park S, Nam BH. Gastric cancer and salt preference: a population-based cohort study in Korea. Am J Clin Nutr 2010;91:1289-1293. 112. D'Elia L, Galletti F, Strazzullo P. Dietary salt intake and risk of gastric cancer. Cancer Treat Res 2014;159:83-95. 113. D'Elia L, Rossi G, Ippolito R et al. Habitual salt intake and risk of gastric cancer: a meta-analysis of prospective studies. Clin Nutr 2012;31:489-498.

TE D

114. Tatematsu M, Takahashi M, Fukushima S et al. Effects in rats of sodium chloride on experimental gastric cancers induced by N-methyl-N-nitro-N-nitrosoguanidine or 4-nitroquinoline-1-oxide. J Natl Cancer Inst 1975;55:101-106. 115. Takahashi M, Hasegawa R. Enhancing effects of dietary salt on both initiation and promotion stages of rat gastric carcinogenesis. Princess Takamatsu Symp 1985;16:169-82.:169-182.

AC C

EP

116. Fox JG, Dangler CA, Taylor NS et al. High-salt diet induces gastric epithelial hyperplasia and parietal cell loss, and enhances Helicobacter pylori colonization in C57BL/6 mice. Cancer Res 1999;59:4823-4828. 117. Tsugane S, Tsuda M, Gey F et al. Cross-sectional study with multiple measurements of biological markers for assessing stomach cancer risks at the population level. Environ Health Perspect 1992;98:207-10.:207-210. 118. Gaddy JA, Radin JN, Loh JT et al. High dietary salt intake exacerbates Helicobacter pylori-induced gastric carcinogenesis. Infect Immun 2013;81:2258-2267. 119. Lopez-Garcia E, van Dam RM, Li TY et al. The relationship of coffee consumption with mortality. Ann Intern Med 2008;148:904-914. 120. Freedman ND, Park Y, Abnet CC et al. Association of coffee drinking with total and cause-specific mortality. N Engl J Med 2012;366:1891-1904.

ACCEPTED MANUSCRIPT

121. Lopez-Garcia E. Coffee consumption and risk of chronic diseases: changing our views. Am J Clin Nutr 2012;95:787-788. 122. Bravi F, Bosetti C, Tavani A et al. Coffee reduces risk for hepatocellular carcinoma: an updated meta-analysis. Clin Gastroenterol Hepatol 2013;11:1413-1421.

RI PT

123. Li G, Ma D, Zhang Y et al. Coffee consumption and risk of colorectal cancer: a meta-analysis of observational studies. Public Health Nutr 2013;16:346-357. 124. Tian C, Wang W, Hong Z et al. Coffee consumption and risk of colorectal cancer: a dose-response analysis of observational studies. Cancer Causes Control 2013;24:1265-1268.

SC

125. Turati F, Galeone C, La VC et al. Coffee and cancers of the upper digestive and respiratory tracts: meta-analyses of observational studies. Ann Oncol 2011;22:536-544.

M AN U

126. Zheng JS, Yang J, Fu YQ et al. Effects of green tea, black tea, and coffee consumption on the risk of esophageal cancer: a systematic review and meta-analysis of observational studies. Nutr Cancer 2013;65:1-16. 127. Xie F, Wang D, Huang Z et al. Coffee consumption and risk of gastric cancer: a large updated meta-analysis of prospective studies. Nutrients 2014;6:3734-3746. 128. Modi AA, Feld JJ, Park Y et al. Increased caffeine consumption is associated with reduced hepatic fibrosis. Hepatology 2010;51:201-209.

TE D

129. Ruhl CE, Everhart JE. Coffee and caffeine consumption reduce the risk of elevated serum alanine aminotransferase activity in the United States. Gastroenterology 2005;128:24-32. 130. Xiao Q, Sinha R, Graubard BI et al. Inverse associations of total and decaffeinated coffee with liver enzyme levels in NHANES 1999-2010. Hepatology 2014;60:2091-2098.

EP

131. Casiglia E, Spolaore P, Ginocchio G et al. Unexpected effects of coffee consumption on liver enzymes. Eur J Epidemiol 1993;9:293-297.

AC C

132. Molloy JW, Calcagno CJ, Williams CD et al. Association of coffee and caffeine consumption with fatty liver disease, nonalcoholic steatohepatitis, and degree of hepatic fibrosis. Hepatology 2012;55:429-436. 133. Freedman ND, Curto TM, Lindsay KL et al. Coffee consumption is associated with response to peginterferon and ribavirin therapy in patients with chronic hepatitis C. Gastroenterology 2011;140:1961-1969. 134. Freedman ND, Everhart JE, Lindsay KL et al. Coffee intake is associated with lower rates of liver disease progression in chronic hepatitis C. Hepatology 2009;50:1360-1369. 135. Corrao G, Zambon A, Bagnardi V et al. Coffee, caffeine, and the risk of liver cirrhosis. Ann Epidemiol 2001;11:458-465.

ACCEPTED MANUSCRIPT

136. Goh GB, Chow WC, Wang R et al. Coffee, alcohol and other beverages in relation to cirrhosis mortality: the Singapore Chinese Health Study. Hepatology 2014;60:661-669. 137. Klatsky AL, Armstrong MA. Alcohol, smoking, coffee, and cirrhosis. Am J Epidemiol 1992;136:1248-1257.

RI PT

138. Klatsky AL, Morton C, Udaltsova N et al. Coffee, cirrhosis, and transaminase enzymes. Arch Intern Med 2006;166:1190-1195. 139. Lai GY, Weinstein SJ, Albanes D et al. The association of coffee intake with liver cancer incidence and chronic liver disease mortality in male smokers. Br J Cancer 2013;109:1344-1351.

SC

140. Ruhl CE, Everhart JE. Coffee and tea consumption are associated with a lower incidence of chronic liver disease in the United States. Gastroenterology 2005;129:1928-1936.

M AN U

141. Chen S, Teoh NC, Chitturi S et al. Coffee and non-alcoholic fatty liver disease: brewing evidence for hepatoprotection? J Gastroenterol Hepatol 2014;29:435-441. 142. Kumari R, Kim WR. Coffee: a panacea or snake oil for the liver? Clin Gastroenterol Hepatol 2014;12:1569-1571. 143. Torres DM, Harrison SA. Is it time to write a prescription for coffee? Coffee and liver disease. Gastroenterology 2013;144:670-672.

TE D

144. Trovato GM, Martines GF, Trovato FM et al. Regular coffee: a magic bullet or a naked gun? Regular coffee but not espresso drinking is protective against fibrosis in NAFLD. J Hepatol 2013;58:1264-1265. 145. Dranoff JA, Feld JJ, Lavoie EG et al. How does coffee prevent liver fibrosis? Biological plausibility for recent epidemiological observations. Hepatology 2014;60:464-467.

AC C

EP

146. Jacobs S, Kroger J, Floegel A et al. Evaluation of various biomarkers as potential mediators of the association between coffee consumption and incident type 2 diabetes in the EPIC-Potsdam Study. Am J Clin Nutr 2014;100:891-900. 147. Kempf K, Herder C, Erlund I et al. Effects of coffee consumption on subclinical inflammation and other risk factors for type 2 diabetes: a clinical trial. Am J Clin Nutr 2010;91:950-957. 148. Lopez-Garcia E, van Dam RM, Qi L et al. Coffee consumption and markers of inflammation and endothelial dysfunction in healthy and diabetic women. Am J Clin Nutr 2006;84:888-893. 149. Zampelas A, Panagiotakos DB, Pitsavos C et al. Associations between coffee consumption and inflammatory markers in healthy persons: the ATTICA study. Am J Clin Nutr 2004;80:862-867. 150. Shen T, Park YC, Kim SH et al. Nuclear factor-kappaB/signal transducers and activators of transcription-1-mediated inflammatory responses in lipopolysaccharide-activated macrophages are a major inhibitory target of kahweol, a coffee diterpene. Biol Pharm Bull 2010;33:1159-1164.

ACCEPTED MANUSCRIPT

151. Shim SG, Jun DW, Kim EK et al. Caffeine attenuates liver fibrosis via defective adhesion of hepatic stellate cells in cirrhotic model. J Gastroenterol Hepatol 2013;28:1877-1884.

RI PT

152. Arauz J, Moreno MG, Cortes-Reynosa P et al. Coffee attenuates fibrosis by decreasing the expression of TGF-beta and CTGF in a murine model of liver damage. J Appl Toxicol 2013;33:970979. 153. Ding M, Bhupathiraju SN, Chen M et al. Caffeinated and decaffeinated coffee consumption and risk of type 2 diabetes: a systematic review and a dose-response meta-analysis. Diabetes Care 2014;37:569-586.

SC

154. Jayasekara H, English DR, Room R et al. Alcohol consumption over time and risk of death: a systematic review and meta-analysis. Am J Epidemiol 2014;179:1049-1059.

M AN U

155. Ronksley PE, Brien SE, Turner BJ et al. Association of alcohol consumption with selected cardiovascular disease outcomes: a systematic review and meta-analysis. BMJ 2011;342:d671. 156. International Agency for Research on Cancer. Consumption of alcoholic beverages. Volume 100e ed. Lyon, France: IARC Press, 2012. 157. Brooks PJ, Zakhari S. Moderate alcohol consumption and breast cancer in women: from epidemiology to mechanisms and interventions. Alcohol Clin Exp Res 2013;37:23-30.

TE D

158. Dorgan JF, Baer DJ, Albert PS et al. Serum hormones and the alcohol-breast cancer association in postmenopausal women. J Natl Cancer Inst 2001;93:710-715. 159. Mahabir S, Leitzmann MF, Virtanen MJ et al. Prospective study of alcohol drinking and renal cell cancer risk in a cohort of finnish male smokers. Cancer Epidemiol Biomarkers Prev 2005;14:170175.

EP

160. Engel LS, Chow WH, Vaughan TL et al. Population attributable risks of esophageal and gastric cancers. J Natl Cancer Inst 2003;95:1404-1413.

AC C

161. Freedman ND, Murray LJ, Kamangar F et al. Alcohol intake and risk of oesophageal adenocarcinoma: a pooled analysis from the BEACON Consortium. Gut 2011;60:1029-1037. 162. Allen NE, Beral V, Casabonne D et al. Moderate alcohol intake and cancer incidence in women. J Natl Cancer Inst 2009;101:296-305. 163. Brooks PJ, Enoch MA, Goldman D et al. The alcohol flushing response: an unrecognized risk factor for esophageal cancer from alcohol consumption. PLoS Med 2009;6:e50. 164. Islami F, Fedirko V, Tramacere I et al. Alcohol drinking and esophageal squamous cell carcinoma with focus on light-drinkers and never-smokers: a systematic review and meta-analysis. Int J Cancer 2011;129:2473-2484.

ACCEPTED MANUSCRIPT

165. Terry MB, Gammon MD, Zhang FF et al. Alcohol dehydrogenase 3 and risk of esophageal and gastric adenocarcinomas. Cancer Causes Control 2007;18:1039-1046. 166. Zhang FF, Hou L, Terry MB et al. Genetic polymorphisms in alcohol metabolism, alcohol intake and the risk of stomach cancer in Warsaw, Poland. Int J Cancer 2007;121:2060-2064.

RI PT

167. Navarro Silvera SA, Mayne ST, Risch HA et al. Principal component analysis of dietary and lifestyle patterns in relation to risk of subtypes of esophageal and gastric cancer. Ann Epidemiol 2011;21:543-550.

SC

168. Kubo A, Levin TR, Block G et al. Dietary patterns and the risk of Barrett's esophagus. Am J Epidemiol 2008;167:839-846.

169. Guenther PM, Reedy J, Krebs-Smith SM. Development of the Healthy Eating Index-2005. J Am Diet Assoc 2008;108:1896-1901.

AC C

EP

TE D

M AN U

170. Trichopoulou A, Costacou T, Bamia C et al. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003;348:2599-2608.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT