|Home | About | Journals | Submit | Contact Us | Français|
Correspondence to: Natalie Ertz-Archambault, MD, Department of Internal Medicine, Mayo School of Graduate Medical Education, Mayo Clinic Arizona, 13400 E Shea Blvd, Scottsdale, AZ 85259, United States. email@example.com
Telephone: +1-480-3019824 Fax: +1-480-3014171
To review microbiome alterations associated with pancreatic cancer, its potential utility in diagnostics, risk assessment, and influence on disease outcomes.
A comprehensive literature review was conducted by all-inclusive topic review from PubMed, MEDLINE, and Web of Science. The last search was performed in October 2016.
Diverse microbiome alterations exist among several body sites including oral, gut, and pancreatic tissue, in patients with pancreatic cancer compared to healthy populations.
Pilot study successes in non-invasive screening strategies warrant further investigation for future translational application in early diagnostics and to learn modifiable risk factors relevant to disease prevention. Pre-clinical investigations exist in other tumor types that suggest microbiome manipulation provides opportunity to favorably transform cancer response to existing treatment protocols and improve survival.
Core tip: Recent literature reports influences of microbiome alterations contributing to carcinogenesis of pancreatic cancer. The poor prognostics of pancreatic cancer are related to late recognition and treatment resistance, thus warranting investigations for modifiable risk factors, early screening biomarkers, and microenvironment elements that affect outcomes. Learning the role of microbiome in carcinogenesis may lead to identifying reliable, non-invasive screening strategies, and additional modifiable risk factors. Microbiome studies in pancreatic cancer could offer therapeutic targets and an extraordinary opportunity to favorably transform cancer response to existing treatment protocols and improve survival by reduction of cancer-related cachexia by manipulating human gut microbiota.
A commensal microbiome, by definition maintains a symbiotic relationship in healthy individuals, offering protection from disease by nutritive, inflammatory-modulating activity, hormonal homeostasis, detoxification, and metabolic effects of bacterial metabolites[1-3]. Dysbiosis is the manifestation of a corrupt, imbalanced microbiome, which contributes to pathogenesis of several diseased states. Recently, there are literature reports on influences of microbiome alteration contributing to carcinogenesis of multiple malignancies[1,2,4-6]. A classic pathogen in the literature is Helicobacter pylori (H. pylori), which has revealed inconsistent and paradoxical associations pending the body site studied[7,8]. H. pylori has been extensively scrutinized as a risk factor for development of pancreatic cancer and an association is controversial[9-12]. Pancreatic cancer often denotes a poor clinical prognosis in part due to late recognition and treatment resistance, warranting investigations for modifiable risk factors, early screening biomarkers, and microenvironment elements that affect outcomes[13,14].
Search methods: PubMed, MEDLINE, and Web of Science for medical search terms: “pancreatic cancer” and “microbiome,” “carcinogenesis,” antibiotic,” “probiotic,” “microorganism,” “bacteria,” “colonization,” “cachexia,” or “infection.” The relevant articles reference lists were also searched manually for additional articles. The last search was performed in October 2016.
Selection criteria: Manuscripts and abstracts describing pre-clinical studies, animal models, epidemiological studies, case series, case-control, retrospective chart reviews, prospective studies, pilot, meta-analysis, and literature topic reviews were included. There were no randomized clinical trials identified from these search terms. Articles were limited to abstract and manuscript publications in the English written language.
Characterization of the healthy microbiome spectrum is ongoing. In 2012, the NIH Human Microbiome Project, demonstrated no microbial taxa were universally present across all humans in a single body site. The oral cavity contains an extensive reservoir of bacteria with more than 700 species observed, most of which have not been cultured in a laboratory[15,16]. Healthy oral habitats are dominated by Streptococcus, followed by Haemophilus in the buccal mucosa, Actinomyces in the supragingival plaque, and Prevotella in adjacent, low-oxygen subgingival region.
Alterations in the ecological balance of the microbiome exist during diseased oral cavity states including gingivitis and periodontal disease compared to a healthy oral cavity[16-20]. Periodontal disease, manifested by an inflamed oral activity, pathogenic oral flora, and tooth loss are well-established independent risk factors associated with development of pancreatic cancer[21-23]. Therefore, the shifts in taxa dominance and diversity of bacterial communities that deviate from an established healthy microbiome may be reflective of disease states[2,3]. Pilot studies have proposed a role in oral pathogenic bacteria in periodontal disease as an early screening test and as a biomarker of pancreatic cancer[12,24,25]. Several dedicated studies have aimed to define microbiome changes in the oral cavity associated with pancreatic cancer, results are summarized in Table Table11.
Oral flora alterations exist in pancreatic cancer patients compared to healthy populations. Salivary RNA studies reveal bacteroides genus and Granulicatella adiacens are more common in pancreatic cancer patients than healthy subjects[12,24]. However, Neisseria elongata, Streptococcus mitis, Corynebacterium genus, and the Aggregatibacter genus are present in lower concentrations in pancreatic cancer than healthy subjects[12,24]. Combining salivary RNA biomarkers for N. elongata and S. mitis yielded an ROC-plot AUC value of 0.90 with 96.4% sensitivity and 82.1% specificity in distinguishing patients with pancreatic cancer from healthy subjects. A cross-sectional study identified of a significantly higher Leptotrichia and lower Porphyromonas colonization in pancreatic cancer patient saliva, translating to an Leptotrichia:Porphyromonas (L:P) ratio of biomarker significance. In this same study, a patient classified with an unknown digestive disease presented with an elevated L:P ratio that led to dedicated workup revealing a new diagnosis of pancreatic cancer. Pilot successes deserve further exploration into utilizing salivary markers as potentially valuable non-invasive, economical screening strategies.
Interestingly, the highest concentration of plasma antibodies to Porphyromonas gingivalis (strain ATTC 53978), a pathogenic bacteria associated with periodontal disease, was linked with a 2-fold increased risk of pancreatic cancer. The association was amplified over time, with the addition of 5 or 7 year lag. Similar to case control studies of saliva samples revealing oral pathogens, P. gingivalis and A. actinomycetemcomitans are associated with increased risk for subsequent development of pancreatic cancer. This finding is consistent with epidemiologic data that periodontal disease is an independent risk factor for pancreatic cancer development[20,23,27]. Alternatively, high antibody titers against non-pathogenic, commensal bacteria were associated with 45% decreased risk of pancreatic cancer compared to those with a lower antibody level profile. Similarly Fusobacterium and Lepotrichia are protective and decreases risk, also in a dose dependent relationship. Lactobacillus is a commensal oral cavity bacterium that diminishes gingival inflammation and cariogenic periodontal pathogenic bacteria. Thus, with the clearly established role of periodontal disease and associated periodontal pathogens for pancreatic cancer risk profiles, any measures to prevent periodontal pathogens may serve protective role to prevent pancreatic cancer, but has not been studied on this topic specifically.
There is literature that illustrates a paradoxical nature of microorganisms relative to by site and tumor studied. For example, eradication of H. pylori causes regression of MALT lymphoma and decreases risk of metachronous gastric carcinoma after endoscopic resection for early stage gastric cancer[1,29]. However, H. pylori gastric colonization decreases the risk of oesophageal adenocarcinoma that does not involve the gastric cardia. H. pylori is a diverse bacteria with several virulent strain variations. Among the best studied are Cytotoxin-associated gene A (Cag-A) positive strains that express Cag-A virulence factor, which is linked to gastric inflammation, ulceration, and promoting malignant transformation in gastric cancer[31,32]. H. pylori and Cag-A dominate microbiome studies in pancreatic cancer. Study results are variable and complex, as is noted in Table Table22[9-11,33-42].
Results from H. pylori case studies in pancreatic cancer reveals complex mixed results pending virulence strain cag-A status. Consensus from recent meta-analysis is that there is a modestly significant increased risk associated with development of pancreatic cancer for cag-A-negative H. pylori strain[9-11,39], with positive correlated adjustment factors including non-O blood type[37,43] and active smoking status[34,36]. The general literature trend summarized in Table Table22 is cag-A-positive strains results in decreased risk or non-significant association with pancreatic cancer. Notable global population differences exist as the majority of studies highlighted in this review are mainly relevant to Western European or North American ethnic groups. The results of one meta-analysis addressing global studies and pancreatic cancer risk including two Eastern Asian population case-cohorts that suggest a decreased risk of pancreatic cancer risk for H. pylori seropositivity overall, including Cag-A-positive strains in Eastern Asian ethnic region.
In one case control study, enteric strains of Helicobacter DNA were demonstrated to colonize the pancreas in 75% of adenocarcinoma patients but not in pancreatic controls with benign disease. Among proposed mechanisms for dissemination may result from hepatobiliary translocation or hematogenous seeding[44,46]. However, DNA of different Helicobacter species is mutually exclusive by sampled site. For example, Helicobacter identified in the pancreas compared with Helicobacter of gastroduodenal tissue of the same patient were different Helicobacter subspecies. Thus, dissemination of H. pylori from the stomach to the pancreas is unlikely, instead a subspecies tissue tropism may exist.
Both direct microbe colonization and downstream proliferative metabolic affects may promote tumor-associated inflammation preserved by low-grade chronic inflammation[6,29,47] . Evidence of this effect in a pre-clinical study of human a pancreatic cell line showed H. pylori colonization of a human pancreatic cell line expressed increased factors for malignant potential including proliferative factors, NF-kappa-B, activator protein-1, proflammatory IL-8 activity, vascular endothelial growth factor secretion, and the growth factor promoter, serum response element. The overall result is activation of molecular pathways for tumor growth and progression in the setting of H. pylori infection.
Fusobacterium is an anaerobic, oral bacterium that has been identified in pancreatic abscesses and carries unfavorable prognostic implications in some gastrointestinal cancers. To explore a role for Fusobacterium in pancreatic cancer, surgical specimens of pancreatic adenocarcinoma were analyzed for presence of this bacterium. Only 8% of specimens in this cohort contained Fusobacterium colonization. However, pancreatic ductal adenocarcinoma surgical specimens with presence of Fusobacterium colonization was identified as an independent predictive factor for shorter survival compared to Fusobacterium negative tumors. The fusobacterium positive sample group also demonstrated 28% detection of paired normal tissue. The presence of Fusobacterium in normal tissue margin suggests it may contribute to malignant potential, but this theory requires further exploration.
The oral microbiome has a protective role against pancreatic cancer in a healthy, commensal state, but may promote malignancy in a pathologic state[1,2,4-6,12,18,24,25]. Shifts in taxa dominance and diversity of oral bacterial communities, especially those reflective of periodontal disease are associated with increased pancreatic cancer risk[12,18,24,25]. This correlates clinically with periodontal disease status, a validated independent risk factor for development of pancreatic cancer[21-23]. Bacterial markers of periodontal disease and shifts in microbial taxa diversity[12,24,25] have promising potential to serve as non-invasive screening biomarkers of pancreatic cancer. The evidence is strong enough to warrant targeted risk reduction strategies in patient education and modifiable lifestyle counseling regarding maintenance of oral hygiene.
A directly carcinogenic role for H. pylori has been explored after discovering enteric strains of Helicobacter DNA demonstrated to colonize the pancreas in a majority of sampled pancreatic adenocarcinoma but not in patients with benign disease. A preclinical study examined direct H. pylori colonization and associated activation of molecular pathways for tumor growth and progression. These downstream molecular effects highlight oncogenic potential with microbiome influence that promotes tumor-associated inflammation preserved by low-grade chronic inflammation[6,29,47]. Despite the existence of several proposed carcinogenic mechanisms of dysbiosis, inflammation is a central facilitator illustrated in pancreatic cancer murine models, human cell lines, and tumor translational expression profiles.
There have been studies that indicate the microbiome and antibiotics modulate tumor response to chemotherapy[48,49]. Germ-free and antibiotic treated murine models highlight the protective effect of commensal bacteria by shaping the inflammatory network required for favorable response to anti-tumor therapy. In murine models, platinum therapy eliminated most subcutaneous lymphoma tumors and prolonged survival in control mice. However, antibiotic-treated and germ free mice failed to respond to platinum-treatment, in part by decreasing reactive oxygen species. Similarly, CTLA-4 inhibitor treated murine models with sarcoma suggest that gut microbiota, specifically bacteroides subspecies, are required for the successful anti-tumor effects of CTLA-4 blockade. Notably, antibiotic and germ free mice with sarcomas do not respond to CTLA-4 inhibitor at baseline, but recover antitumor activity with recolonization of gut commensals by human fecal microbiota transplantation of specific bacteroides subspecies. Oral administration of Bifidobacterium in murine models with melanoma augments the immune response to tumor cells, in part by dendritic cell activation of the innate immune system. This effect was not observed with administration of lactobacillus species, suggesting a complex, species specific modulation of the immune system in vivo. The potential to utilize probiotics in humans to amplify antitumor response to existing chemotherapy and immunotherapy protocols requires further investigation.
Anti-tumor therapy and commensal flora collaborate in part, by loss of TNF-dependent early tumor necrosis response, down-regulation of inflammatory cytokines, phagocytosis, antigen presentation, and adaptive immune response gene expression controlling tissue development and cancer. The loss of commensal organisms by antibiotics and the possibility of carcinogenic promoting effects of antibiotics have been explored. The risk related to pancreatic cancer seems limited to the penicillin class, especially with more than five courses, but this risk diminishes over time. Macrolides, cephalosporins, tetracyclines, antivirals, and antifungals were not associated with increased risk of pancreatic cancer. The impact of antibiotics on commensal framework may explain the need for repeated antibiotic exposures, leading to an enduring change in bacterial community diversity. Murine models demonstrate lactobacillus was among quickest flora to recover in the gut after antibiotic therapy. However, the effect of antibiotics on the gut microbiome is enduring at four weeks after exposure; the population is deficient, and not reflective of its healthy, baseline, pre-antibiotic diversity.
Commensal bacteria offer protection from disease by inflammatory-modulating activity as above, but also by hormonal homeostasis, detoxification, and metabolic effects of bacterial metabolites. For example, murine models show lactobacilli are consistently reduced in cachectic mouse models. A lactobacilli cocktail combination with prebiotic substrate that supports growth of microorganisms, changes the dysbiotic populations of cecal microbiota composition in murine models, clinically resulting in improved survival and reduction of cachexia. These are highly important implications in pancreatic adenocarcinoma population since these patients carry the strongest burden of cancer cachexia among all malignancies, present in up to 80% of patients[54,55] resulting in reduced survival and progressive disease[55-57]. Weight stabilization alone significantly proven to improve survival in pancreatic adenocarcinoma patients with unresectable disease.
In conclusion, the initial motive to explore microbiome role in carcinogenesis may lead to identifying reliable non-invasive screening strategies and discern additional modifiable risk factors. With further investigation, potentially microbiome studies in pancreatic cancer could offer therapeutic targets. Perhaps the most extraordinary opportunity is to favorably transform cancer response to existing treatment protocols and improve survival by reduction of cancer-related cachexia by manipulating human gut microbiota.
Recently, there are literature reports on influences of microbiome alteration contributing to carcinogenesis of multiple malignancies. Among the most controversial is dysbiosis related to pancreatic cancer. Pancreatic cancer often denotes a poor clinical prognosis in part due to late recognition and treatment resistance, warranting investigations for modifiable risk factors, early screening biomarkers, and microenvironment elements that affect patient outcomes.
Murine models demonstrate commensal microbiome taxa modulates a favorable tumor response to chemotherapy in multiple tumor types In addition, manipulation of cecal microbiome composition with lactobacillus in murine models, have resulted in improved survival and reduction of cachexia a clinically significant burden in the majority of pancreatic cancer patients.
This review article serves to update literature on microbiome alterations associated with pancreatic cancer, its potential utility as an early screening biomarker, examine the influence of the microbiome in antitumor therapy, and the potential impact of microbiome manipulation to affect pancreatic cancer patient outcomes.
Exploring the microbiome role in carcinogenesis may lead to identifying reliable non-invasive screening strategies and discern additional modifiable risk factors. With further investigation, potentially microbiome studies in pancreatic cancer could offer therapeutic targets. Perhaps the most extraordinary opportunity is to favorably transform cancer response to existing treatment protocols and improve survival by reduction of cancer-related cachexia by manipulating human gut microbiota.
This review describes the relationships between microbiome and pancreatic cancer. The data in this report is of considerable importance in investigations for modifiable risk factors of pancreatic cancer.
Manuscript source: Unsolicited manuscript
Specialty type: Gastroenterology and hepatology
Country of origin: United States
Peer-review report classification
Grade A (Excellent): A, A
Grade B (Very good): 0
Grade C (Good): 0
Grade D (Fair): D
Grade E (Poor): 0
Conflict-of-interest statement: All the authors declare that they have no competing interests.
Data sharing statement: This manuscript represents comprehensive topic review from published manuscript on topic as indicated in methods section. Prior drafts and PDF versions of articles utilized as referenced in citation section are available with first author on request firstname.lastname@example.org. No additional data are available.
Peer-review started: October 7, 2016
First decision: October 28, 2016
Article in press: December 21, 2016
P- Reviewer: Kimura K, MatsudaY, Wei DY S- Editor: Qi Y L- Editor: A E- Editor: Wang CH