点击显示 收起
1 From the Department of Cell Biology and Human Anatomy, University of California, Davis, School of Medicine.
See corresponding article on page 137.
The roles that some foods and dietary components may play in the etiology and pathogenesis of cancer of the esophagus and stomach have been recognized for several years. Most of the evidence for those roles has come from ecologic, descriptive studies, as well as from case-control and cohort-type studies. Although both tumor sites are part of the digestive system, the pathogenesis of the tumors and their relation to specific dietary components may differ.
In the past 40 y, the type of tumor most commonly found in the esophagus has changed. Although 40 y ago 90% of all esophageal tumors were of the squamous cell type, the incidence of esophageal adenocarcinoma has risen such that today it is more prevalent than the squamous cell type in the United States and Europe (1). The epidemiology of these types of tumors varies markedly, and thus, one may predict that dietary components have different effects on the risks of these cancers. However, many dietary factors, such as the type and amount of food, preparation method, and energy intake, can affect risk. Some of these factors may be considered in the design of a human study whereas others may not. Other factors to consider include sex and race. Deaths from cancer of the esophagus are >3 times more common in men than in women (2). Moreover, squamous cell cancer is 6 times more likely to occur in black men than in white men. Risk factors associated with the development of squamous cell cancer are well documented in medical texts, and several of those risk factors are dietary. For example, squamous cell cancer of the esophagus is strongly associated with excess alcohol consumption. In addition, the combined use of tobacco, in any of its forms, tends to greatly enhance the risk of developing this cancer. In contrast, diets high in fruit and vegetables are associated with a decreased risk of squamous cell cancer of the esophagus. What might be considered a high level of consumption? Probably 5 servings of fruit and vegetables/d would be considered high because only 23.9% of US adults eat that amount. In Nebraska, only 17.6% of adults consume 5 servings/d (2). Although several health organizations have recommended that level of daily consumption and two-thirds of US adults believe that a diet containing fruit and vegetables is very important, consumption has increased only slightly in the past 30 y (2). What specific components of fruit and vegetables are associated with the reduction in the risk of squamous cell cancer of the esophagus? The components may be vitamin C and carotenoids because high intakes of those compounds may decrease the risk. At the same time, the evidence concerning the relation between meat and dairy products and the risk of esophageal cancer appears to be inconsistent (3).
The risk factors, particularly dietary components, for adenocarcinoma of the esophagus are certainly less well defined. Therefore, epidemiologic studies such as the one by Chen et al (4) in this issue of the Journal are important in defining which nutrients may play a role. One of the most important, documented differences between the 2 types of tumors is the association between gastroesophageal reflux disease and adenocarcinoma of the esophagus. The severity, frequency, and duration of reflux are apparently positively associated with an increased risk of esophageal adenocarcinoma (5). Whether gastroesophageal reflux can be altered by diet and whether such alterations in turn can modify the risk of esophageal cancer remain to be determined. Related to this reflux disease is a predisposition to develop Barrett esophagus, a condition in which an abnormal epithelium replaces that which normally lines the distal esophagus. Another important relation exists between adenocarcinoma of the esophagus and body mass index because the risk of this cancer increases with obesity; however, there is no such relation for squamous cell cancer of the esophagus. The reason for that difference is not known. Body mass index was controlled for by Chen et al in their study (4).
Although the number of deaths due to stomach cancer has been decreasing globally, particularly in the United States, stomach cancer ranks second in cancer incidence worldwide. For one site in the stomach, the cardia, incidence rates appear to be increasing, particularly among white men. Most stomach cancers in the United States are at an advanced stage at diagnosis, and the 5-y survival rates for both stomach and esophageal cancer are quite low (2). Thus, an effective means of preventing the disease, such as alteration of dietary or other habits, could have an important effect on public health worldwide.
Excessive salt intake and the consumption of salted foods have been identified as possible risk factors for stomach cancer. The intake of salt, however, has decreased in many countries as the result of recommendations to reduce hypertensive diseases. Infection with Helicobacter pylori is also associated with an increased risk of stomach cancer. Both of those factors seem to be implicated in the development of superficial gastritis and chronic atrophic gastritis. Stomach cancer is considered to be a multistage disease (6), and dietary factors may be involved in some, but not all, of the different stages.
Although several epidemiologic studies investigated the relation of diet to esophageal adenocarcinoma and to stomach cancer, the study by Chen et al incorporated pattern analysis of food intake by a specific population instead of focusing on individual nutrients or foods. Thus, food-intake patterns and statistical analysis defined the clusters. This approach provided a fresh look at patterns, including ones that investigators might not normally consider, for example, a diet high in white bread or high in desserts as associated with cancers of the esophagus and stomach. Although there did not appear to be any specific link, analysis of those dietary patterns and foods makes an important contribution. As with other epidemiologic studies, criticism of this study can be raised regarding the identification of cases and the potential time lag until telephone interview of the next of kin (61–80% of the interviewees) or the questionnaire used to measure food frequency. However, several of the findings are consistent with those previously reported by other investigators who used different methods of analysis.
A study of dietary patterns and tumors of the esophagus and stomach presents special problems that may not occur when considering cancer of other organs. For example, a diagnosis at a relatively late stage of the disease would be associated with extensive pathology of the esophagus or stomach and possible metastasis to regional lymph nodes. This alteration could certainly influence the types of food tolerated and consumed. For those patients with diagnosed stomach cancer (stage I and II), partial or total gastrectomy and lymphadenectomy are standard treatment operations (7). Adjuvant chemotherapy and radiation therapy may also be used in more advanced stages. Thus, either the presence of the tumor or the therapy associated with it could have profound effects not only on the foods selected but also on their metabolism.
The conclusions of Chen et al generally support the findings of other investigators that diets high in fruit and vegetables may decrease the risk of esophageal adenocarcinoma and that diets high in meat may increase the risk of distal stomach cancer. This latter finding raises future questions about the characteristics of meat; for example, is it the protein, the fat, the method of cooking, or other factors that are associated with the increased risk? From other studies, there does not appear to be any relation between animal or plant protein and stomach cancer (3). However, there does appear to be an inverse relation between fish consumption and esophageal adenocarcinoma. As discussed by Chen et al, this relation could be due to the n-3 fatty acid composition of the fish fat and alteration in eicosanoid metabolism. Tumorigenesis might then be changed through altered eicosanoid metabolites. Other components of meat that may affect cancer risk are the heterocyclic amines, which are carcinogenic chemicals formed from the cooking of muscle meats such as beef, pork, chicken, and fish. Other sources of protein have few or no heterocyclic amines naturally or when cooked. The temperature at which meats are cooked appears to be important. In one study (8), subjects who ate beef medium-well or well done had a 3-fold greater risk of stomach cancer than did those who ate their beef rare or medium-rare. Interestingly, gravy, which was analyzed by Chen et al, can contain substantial amounts of heterocyclic amines. Currently, there are neither good measures of the quantity of heterocyclic amines that would have to be eaten to increase cancer risk nor guidelines concerning consumption of foods containing heterocyclic amines. These public health issues need further examination. There also appeared to be an inverse relation between dairy products and esophageal adenocarcinoma. Whether this was related to components such as conjugated linoleic acid or probiotic bacteria found in those dairy products remains to be determined.
Despite relatively large odds ratios, dietary pattern analysis did not identify an association between 1 of the 6 dietary patterns and esophageal or stomach adenocarcinoma. One of the reasons may have been the statistical analysis. Generally, Bonferroni inequality becomes very conservative with 3 or more comparisons and makes it difficult for many data sets to achieve statistical significance. Alternatively, as identified by the investigators, a larger sample size may show significant differences between the dietary pattern groups.
Because this study was conducted in a population from Nebraska, an interesting follow-up analysis would be to compare the incidence of stomach cancer in farmers with that in nonfarmers by using dietary pattern analysis. The Nebraska Rural Health and Safety Coalition (9) notes the higher incidence of stomach cancer in farmers and believes that nitrate contamination of ground water and exposure to organophosphate pesticides may be important risk factors.
Stomach cancer has been one of the most extensively studied cancers with respect to fruit and vegetable consumption. Evidence gathered from more than 38 cohort and case-control studies generally indicates that vegetables and citrus fruit, in particular, exert a protective effect and reduce the incidence of stomach cancer. The ß-carotene and vitamin C found in foods also appear to have a protective effect (10). A recent clinical trial reinforces some of these findings (11). The progression of a precancerous marker, nonmetaplastic atrophy, was assessed in that 6-y study. Subjects receiving anti-H. pylori treatment were 4.8 times more likely to have regression of the precancerous marker than were the control subjects. Subjects receiving vitamin C and those receiving ß-carotene were 5.0 and 5.1 times more likely, respectively, to have the regression of the precancerous marker. Thus, it was surprising that Chen et al found no correlation between consumption of fruit and vegetables and stomach cancer. Alternatively, fruit and vegetables may play a less important role in reducing the risk of stomach cancer than they do in reducing the risk of esophageal cancer. That possibility, however, will need to be verified by other investigators. Nevertheless, by using a different analytic method, cluster analysis, Chen et al provide additional evidence that some foods such as fish, dairy products, and fruit and vegetables may play a role in reducing the risk of esophageal adenocarcinoma.
REFERENCES