Fiber and esophageal cancer prevention: Is there a role for the microbiome?

Epidemiological studies suggest that dietary fiber may decrease the risk for development of colorectal cancer. There appear to be four mechanisms by which fiber is protective in the colon and rectum: (a) increasing bulk of the stool; (b) binding to colorectal carcinogens; (c) decreasing transit time of waste through the bowel; and (d) altering the microbial composition of the colon leading to reduced risk for colon cancer (Kritchevsky, 1997; Zeng, 2014). Both (b) and (c) reduce the interaction of carcinogens with the lining of the colon and rectum. Dietary fiber may also play a role in reducing risk for the development of esophageal cancer, perhaps through mechanisms that differ from those in colorectal cancer. Epidemiological studies have identified protective effects of fiber against the precancerous lesion, Barrett's esophagus (BE), and its conversion to esophageal adenocarcinoma (EAC), presumably by reducing acid reflux from the stomach to the esophagus (Coleman et al., 2013). Similar associations were observed in studies of fiber intake and risk of conversion of esophageal dysplasia to esophageal squamous cell carcinoma (ESCC), but the results were not significant (Coleman et al., 2013). However, a recent cross-sectional study in China found that subjects were at increased risk for ESCC when they consumed diets low in vegetables and fruit and this was attributed in part to low fiber intake (Zang et al., 2022).

In any discussion of the role of fiber in disease occurrence, it is necessary to define what is meant by dietary “fiber.” Briefly, there are two types of dietary fiber, soluble and insoluble (Papandreou et al., 2015). Both types are carbohydrates found in most plant foods. Soluble fiber (largely pectin and inulin) dissolves in water and is digested by enteric bacteria in the large intestine. Dietary sources of soluble fiber include oats, legumes, and vegetables such as carrots, cabbage, Brussels sprouts, squash, and broccoli. Soluble fiber reduces low-density lipoprotein (LDL) cholesterol levels in blood and helps control blood sugar by preventing rapid rises in blood sugar levels after a meal (Kritchevsky, 1997). Insoluble fiber (cellulose, lignin) does not dissolve in water and passes directly through the gastrointestinal tract. Because it remains intact, it provides “bulk” with stool formation and speeds the movement of waste through the digestive system. Dietary sources of insoluble fiber include whole grains, rye, and fruits and vegetables. Bacterial degradation of fiber in the colon produces short-chain fatty acids (SCFA) such as butyric acid which may affect colonic and fecal pH. Butyric acid also exhibits both antiproliferative and proapoptotic activities (Vanamala et al., 2008). Given these various functions of soluble and insoluble fiber, it seems readily apparent how fiber might reduce risk for colon cancer but not for esophageal cancer, especially given the rapid transit of dietary fiber through the esophagus and the observation that protective compounds such as butyric acid are produced mainly by bacteria in the intestine.

Extensive studies indicate that the microbial composition (microbiome) of the esophagus is linked to the development of several esophageal diseases including eosinophilic esophagitis (EoE) (Harris et al., 2015), gastroesophageal reflux disease (GERD), BE and EAC (Gall et al., 2015; Snider et al., 2018; Yang et al., 2009). Microbial analysis of secretions from the esophageal mucosa of 70 subjects with either EoE or GERD indicated that the bacterial load was increased in both EoE and GERD relative to mucosa from normal subjects (Harris et al., 2015). Haemophilus was significantly increased in untreated EoE subjects as compared with normal subjects. Streptocossus was decreased in GERD subjects on protein pump inhibitors. In another study using biopsies of the distal esophagus obtained from 34 normal, EoE and BE patients, individuals with BE or EoE were more likely to harbor gram-negative anaerobic or microaerophilic organisms, such as Haemophilus, Neisseria, Veillonella, and Prevotella (Yang et al., 2009). In contrast, healthy individuals were more likely to be dominated by Streptococcus species. Gall et al. reported that in a BE cohort, Streptococcus and Prevotella species dominated the upper gastrointestinal tract and the ratio of these two species was associated with waist-to-hip ratio and hiatal hernia length, two known EAC risk factors in BE patients (Gall et al., 2015). A more recent study using saliva samples from 49 endoscopy patients without and without BE found that, at the phylum level, the oral microbiome in BE patients had significantly increased relative abundance of Firmicutes and decreased Proteobacteria (Snider et al., 2018). At the genus level, a model including the relative abundance of Lautropia, Streptococcus, and a genus in the order of Bacteroidales distinguished BE from controls. The authors concluded that BE is associated with a distinct oral microbiome which may represent a potential screening marker for BE. Because BE is a precursor lesion for EAC, changes in the microbiome from normal to BE could well be important for the eventual development of EAC. The microbiome associated with EAC itself remains poorly described. Blackett et al. used 16S rRNA sequencing of prespecified bacterial species, comparing controls with patients with GERD, BE, and EAC (Blackett et al., 2013). In many ways, the microbiome of EAC was more similar to the normal esophageal microbiome than to BE. Control and EAC samples had increased relative abundance of Bifidobacteria, Bacteroides, Fusobacteria, Veillonella, Lactobacilli, and Staphylocossus and decreased Campylobacter when compared to BE samples. Elliott et al. compared the microbiome of esophageal tissues from patients with BE and EAC and normal controls (Eliott et al., 2017). They reported a decrease in microbial diversity in EAC tumor samples and increased relative abundance of Lactobacillus fermentum.

ESCC is the most prevalent esophageal cancer worldwide and about 50% of the cases occur in China. Numerous factors have been associated with development of the disease including poor nutritional status, low intake of fruits and vegetables, consumption of carcinogenic food products, drinking temperature hot beverages, tobacco and alcohol use, and other factors (Stoner & Gupta, 2001). The precursor lesion for the development of ESCC is esophageal squamous dysplasia (ESD) and screening for the presence of dysplastic lesions is important for identifying individuals at risk for development of the disease. The microbiome of the esophagus in patients with ESD or ESCC has not been investigated as extensively as that in patients with BE or EAC. An early study in which a Human Oral Microbe Identification Microarray was used to test for the presence of 272 bacterial species in normal individuals and in patients with ESD indicated that the number of bacteria per sample was lower in ESD patients than in normal controls (Yu et al., 2014). However, the specific bacterial types in the samples were not determined. Another study of 325 resected esophagus cancer specimens, of which 92% were ESCC, revealed increases in the quantity of Fusobacterium nucleatum in cancer specimens when compared to normal esophageal mucosa (Yamamura et al., 2016). Moreover, the presence of F. nucleatum DNA in tumors was associated with shorter survival. These observations suggest that specific bacteria such as F. nucleatum may play a role in development of ESCC and its progression.

May and Abrams concluded in a recent review article (May & Abrams, 2018) that the esophageal microbiome has a potentially important role in the development of esophageal EAC and ESCC, but our understanding of the role of specific bacteria in the pathogenesis of these diseases remains limited. They emphasized the need to conduct investigations directed toward modifying the bacterial composition of the esophageal microbiome and assessing effects of these modifications on disease occurrence and progression. This led Nobel et al. to test the hypothesis that dietary fiber may reduce risk for esophageal cancer by producing alterations in the esophageal microbiome (Nobel et al., 2018). In their study, microbiome samples were collected from the squamous esophagus of 47 endoscopy patients who had completed a food frequency questionnaire quantifying fiber intake. The most abundant bacterial phyla among all samples were Firmicutes, Proteobacteria, Actinobacteria, and Fusobacteria. All other phyla had relative abundance <1%. At the genus level, Streptococcus was the most abundant bacterium. As indicated above, Streptococcus in the phylum Firmicutes is associated with a healthy, phenotypically normal esophagus. gram-negative bacteria comprised a mean of 49% of bacteria in the esophagus and as described above they are associated with an abnormal esophagus, including GERD and BE (Yang et al., 2009). Samples from patients with a higher fiber intake had an increased abundance of Streptococcus and other members of the phylum Firmicutes and a decreased abundance of gram-negative bacteria. Low fiber intake was associated with increased abundance of several gram-negative bacteria, including Prevotella, Neisseria, and Eikenella. The authors concluded that the inverse association between dietary fiber and prevalence of gram-negative bacteria provides one mechanism by which increased fiber consumption might protect against inflammation-related esophageal diseases including GERD and BE. Interestingly, samples from only 10 of the 47 patients were found to contain bacteria that produce SCFA such as butyrate and the relative abundance of these bacteria in all positive samples was less than 1%. This suggests that the esophageal and intestinal microbiomes differ significantly in their ability to produce SCFA. Clearly, additional studies are required to further evaluate the effects of fiber on the composition of the esophageal microbiome and its role in the development of esophageal cancer.

Animal models of esophageal carcinogenesis could play an important role in assessing the effects of fiber and other dietary constituents on the esophageal microbiome and the development of esophageal cancer. A rat model of ESCC involves the induction of tumors following subcutaneous injection of rats with the carcinogen, N-nitrosomethylbenzylamine (NMBA) (Daniel & Stoner, 1991). This model has been used extensively to evaluate the ability of multiple dietary factors including ellagic and cis-retinoic acids (Daniel & Stoner, 1991), isothiocyanates (Stoner et al., 1991), lyophilized black raspberries (Kresty et al., 2001), anthocyanins (Wang et al., 2009), green and black teas and green tea polyphenols (Morse et al., 1997; Wang et al., 1995), and others to reduce NMBA-induced esophageal tumor development when provided in a synthetic diet. Interestingly, the fiber constituent of black raspberries was nearly as chemopreventive as whole black raspberries in reducing NMBA-induced tumors in the rat esophagus (Wang et al., 2009). However, it is not known whether the fiber produced changes in the bacterial composition of the esophageal microbiome. In addition, no attempts were made in this study to identify potentially bioactive components in the fiber such as pectin, cellulose and lignin to inhibit esophageal tumorigenesis and their potential effects on the esophageal microbiome.

In 1989, Pera et al. (1989) introduced the first rat surgical model of EAC by inducing chronic duodenogastroesophageal reflux and treatment with the carcinogen 2,6-dimethylnitrosomorpholine. Since then, several models using different surgical procedures have been developed. Recently, an improved surgical procedure led to the establishment of a gastroesophageal reflux model with a high incidence of BE that would appear to be ideal for studies of the effects of fiber and other dietary factors on the development of BE and EAC. Correlations could be made between the effects of these factors on BE and EAC and their effects on the esophageal microbiome.

In summary, additional investigations are needed to further establish the role of the esophageal microbiome in the development of esophageal cancer and the mechanism(s) by which it influences the development of the disease. Because nearly one-half of ESCC worldwide occurs in China, it would seem prudent to develop long-term intervention studies in high-risk regions of China to determine if high fiber diets (e.g., rich in wheat bran) effect the composition of the oral and esophageal microbiome and the risk for ESCC. One cohort for such studies could be subjects found by endoscopy to have hyperplastic or early-stage dysplastic lesions in the esophagus. Subjects who have undergone an esophagectomy for ESCC could be another cohort to determine If high fiber diets reduce risk for recurrent disease. Fiber constituents found to be protective in the above-described rat model of ESCC could also be tested individually or in combination to evaluate whether nutraceutical food additives might be protective in individuals at high risk for the disease. In the Western world, such studies could be conducted in smokers, especially those who consume beverages containing alcohol. Similarly, intervention studies in BE patients could further determine the effects of fiber constituents on the lesion itself and its potential progression to EAC. Correlations could be made between these effects and the bacterial composition of the oral and esophageal microbiome. Results from investigations in the rat surgical model of esophageal BE and EAC could provide valuable information for designing the human studies.