George Kouklakis

Helicobacter pylori Infection Might Contribute to Esophageal Adenocarcinoma Progress in Subpopulations With Gastroesophageal Reflux Disease and Barrett's Esophagus

To the Editor, Fischbach et al. [1] concluded that estimates for the effect of Helicobacter pylori (H. pylori) on Barrett’s esophagus (BE) were heterogenous across studies; although overall H. pylori, and particularly cagA cytotoxin, tended to be protective for BE in most studies, H. pylori effect on BE varied by geographic location. Barrett’s esophagus is a complication of long-standing gastroesophageal reflux disease (GERD) and wellrecognized premalignant condition playing a pivotal role in the development of esophageal adenocarcinoma (EA), the most common esophageal malignancy in Western countries with increasing faster incidence than any other cancer [2]; GERD plays a crucial role in the pathophysiology and the clinical identification of BE [2]. In this regard, our data show that H. pylori infection (H. pylori-I) is frequent in Greek patients with GERD and even with nonendoscopical reflux disease [2], and H. pylori eradication leads to better control of GERD symptoms and improves esophagitis [2,3]. Moreover, consistent associations with the Greek data were shown by others [3] also reporting improvement in reflux symptoms following H. pylori treatment. It is important to note that some other authors, usually prior supporters of the theory that H. pylori “protects” against GERD, relented their initial findings, claiming that H. pylori eradication does not cause or protect against GERD and, moreover, recommending H. pylori eradication in GERD [4]. Additionally, although epidemiologic studies do not suggest causality with H. pylori, however, such studies support our and others’ findings; for instance, a large study (~21,000 cases) showed that the decrease in H. pylori-I parallels the decrease in peptic ulcer prevalence, and the increase in GERD and reappearance of GERD after H. pylori eradication is rare. Moreover, contrary to expectation, patients hospitalized with duodenal ulcers (61,548 cases), obviously attributed to H. pylori-I, had a significant 70% excess risk of EA. Much evidence further potentiates the concern that H. pylori is not “protective” against GERD [5] and its complications including BE and EA. The interplay between H. pylori and host factors plays an important role in the pathogenesis of GERD. Specifically, H. pylori may contribute to GERD pathogenesis by several mechanisms including release of several mediators, cytokines, and nitric oxide, which may adversely affect the lower esophageal sphincter (LES); direct damage of the esophageal mucosa by bacterial products; increased production of prostaglandins that sensitize afferent nerves and reduce LES pressure; and augmented acidity (by gastrin release) that exacerbate GERD [3]. The authors considered some putative pathways involving H. pylori and a decreased risk of BE [1]. However, these pathways might represent the one BE pathogenic “coin’s” side. Regarding the other alternative side, gastrin, induced by H. pylori-I, is an oncogenic growth factor contributing to esophageal, gastric, and colon carcinogenesis and, in particular, playing a potential causal effect on neoplastic progression in BE; gastrin stimulates proliferation via JAK2and Akt-dependent NF-kappaB (NF-jB) activation in Barrett’s EA cells, shows antiapoptotic activity through upregulation of Bcl-2 and survivin, and upregulates cyclooxygenase (COX)-2 expression [2]. In this regard, H. pylori-I activates NF-jB, an oxidantsensitive transcription regulator of inducible expression of inflammatory genes such as COX-2, which regulates gastrointestinal cancer cell growth and proliferation. In particular, H. pylori-I induced NF-jB and COX-2 expression in esophageal epithelial cells, playing a role in inflammation associated with BE and tumorigenesis in the esophagus [2]; upon colonizing esophagus, H. pylori increases the severity of esophageal inflammation and the incidence of BE and EA [6]. Moreover, recent evidence indicates that 1, H. pylori-I prevalence is high in BE; 2, neither H. pylori-I nor H. pylori-I by CagA+ strains reduce the risk of BE in certain populations with high prevalence of H. pylori-I; 3, the expected incidence of EA with persistent H. pylori-I is higher than that of EA after eradication of infection [5]; H. pylori-I may affect specific molecular alterations (genetic instability, E-cadherin methylation, and monoclonal antibody Das-1) associated with the pathogenesis of BE; and 4, H. pylori induces

Chromogranin A as a biomarker of disease activity and biologic therapy in inflammatory bowel disease: a prospective observational study

AbstractObjective. To access the correlation of Chromogranin A (CgA) with inflammatory bowel disease (IBD) activity and responsiveness to medical therapy. Material and methods. A prospective observational study was conducted in 56 patients with moderate ulcerative colitis (UC) or Crohn’s disease (CD) (UC, n = 29, CD, n = 27), 17 patients with irritable bowel syndrome and predominant diarrhea (IBS-D) and 40 healthy volunteers. IBD patients were treated by biologics (infliximab or adalimumab) or conventional agents (aminosalicylates, thiopurines or methotrexate and steroids) and were classified according to their treatment in two groups. Serum CgA was measured at baseline and 4-week posttreatment period. Results. Serum CgA was significantly higher in IBD patients than in those with IBS-D or healthy volunteers (p < 0.01). Furthermore, serum CgA was markedly increased in CD patients than in UC patients (p < 0.01). CgA value was significantly reduced in ‘biologic’ group (24 IBD patients, UC, n = 15, CD, n = 9)...

Helicobacter pylori infection and colorectal carcinoma: pathologic aspects

To the Editor, Fleming et al. (1), by reviewing the pathologic aspects of colorectal neoplasms, summarized the pathogenesis and molecular classification of colorectal carcinoma (CRC) including mainly molecular pathways and environmental factors. However, they did not mention the potential pathologic aspects of environmental factors involved in colorectal oncogenesis, particularly in sporadic CRC. More than 95% of colorectal cancers are sporadic, also mentioned by the authors (1), without a significant hereditary risk. Geographic variation in the incidence of CRC is substantial with a higher incidence observed in the West. Environmental factors contribute considerably to this variation (2); the majority of the sporadic cancer is believed to be due to modification of mutation risk by other genetic and/or environmental factors. Dietary factors may influence the oncogenic process by modifying intestinal transit time, altering the flow and recycling of bile, or changing the intestinal bacterial flora composition. Numerous studies support a role for the gut microbiota in colorectal oncogenesis and the colonic microbiota drives the progression towards colorectal malignancy including generation of reactive metabolites and carcinogens, alterations in host carbohydrate expression and induction of chronic mucosal inflammation (3); long-term colonization of the colon by rogue commensal bacteria capable of inducing chronic DNA damage could contribute to sporadic CRC developement, thereby suggesting sporadic CRC as an infectious disease (4). In this regard, Helicobacter pylori (Hp), a curved spiral gram-negative bacterium found in the gastric mucosa of a large proportion of humans worldwide (>50%), has been evaluated as a possible etiologic agent for CRC and recent data indicate that there is a serological association between Hp infection (Hp-I) and the risk of CRC, especially for left-sided and early-stage cancers (5). Moreover, Hp seropositive subjects are associated with a modest increase in the risk for colorectal adenoma, and since Hp-I can increase the risk especially of advanced adenomas, the medical community should take into account that a preventive strategy is needed, and, furthermore, elucidating the pathophysiological role of Hp in the development of CRC is highly warranted (6). However, as mentioned by the authors (5,6), the serologic measurement of infection status is less than perfect, thereby representing a specific limitation of their studies. Indeed, the serological test does not discriminate between current and past infections and, apart from past infection that may even be more relevant for oncogenesis, such a distinction is essential because only current Hp-I induces humoral and cellular immune responses that produce or perpetuate chronic inflammatory processes in gastrointestinal tract with potential oncogenic sequelae; many neoplasms including colorectal adenomas and cancers arise at the sites of chronic inflammation and infection (7-10). Based on histology, the practical gold standard for Hp-I diagnosis, our own preliminary studies indicated Hp presence in malignant tissue in 34 of 41 (82.9%) patients with CRC (23 men, mean age 73.6±7.9 years) (11). Extending these preliminary data we currently included 50 patients (28 men, mean age 71.3±9.7 years) with CRC and 25 patients (13 men, mean age 72.8±10.1 years) with colonic polyps with the following results: Hp presence in malignant and polyp tissues of patients were observed in 84% and 64%, respectively, confirming our preliminary data (12). It is important to note that, apart from Cresyl fast violet staining mainly used to detect Hp, its presence was also documented by immunohistochemical method (using polyclonal rabbit anti-Hp antibody (dilution 1:50, DAKO, Athens, Greece) in adenoma and malignant colonic tissues. Specifically, in accordance with Hong et al. (6), Hp progressive increased presence was observed in our patients with adenomas associated with mild (50%) and moderate/high-grade (80%) dysplasia; the latter lesions are frequently described as advanced adenomas. However, contrary to the authors’ considerations (6), our series showed an increased Hp presence in left-sided (79%) than in the proximal colon (21%) adenomas; left-sided cancers were also observed in 70.7% of our patients, a finding also noticed by Zhang et al. (5), thereby suggesting that Hp-I might be associated with a rather relevant risk increase in the left CRC. The multistep model of gastric cancer postulates that there is initially an inflammation, caused mainly by Hp-I, which can lead to the development of chronic active gastritis. In a subset of these patients, this inflammatory process leads to the development of atrophic gastritis, followed by intestinal metaplasia, dysplasia, and, ultimately, early and advanced gastric cancer (13). It is considered that all stages prior to the development of high-grade dysplasia are potentially reversible, although this is still controversial (13). Because, Hp also induces inflammatory changes in colonic mucosa (14), it would be reasonable to further speculate, in view of our data, that chronic inflammatory process induced by Hp-I in colonic mucosa may lead to adenoma - mild-moderate/high grade dysplasia - CRC development sequence. These findings may emphasize the need for Hp eradication to prevent the development of both colon and gastric cancer (13). In addition, we found that presence of Hp in malignant colonic tissue was associated with Ki-67 oncogene increased expression in all tumor specimens and low expression in all adjacent tissue specimens (15). Moreover, p53 increased and low expression was observed in 72.5% and 100% of tumor and adjacent tissues specimens, respectively. Likewise, antiapoptotic Bcl-2 protein was observed in 60% and 9% of tissue specimens, respectively, whereas proapoptotic Bax protein was observed in 9% and 100% of tissue specimens, respectively (15). Therefore, Hp colonizing colonic neoplasm tissue seems to be associated with an increased cell proliferation and impaired apoptotic process in malignant tissue compared with normal adjacent colonic mucosa, thereby possibly contributing to colon normal mucosa-adenoma-cancer sequence (15). In this regard, Hp-induced gastrin as an oncogenic growth factor, shows antiapoptotic activity through the Bcl-2 upregulation and contributes to gastric and colon carcinogenesis through stimulation of mutagenic and tumorigenic cyclooxygenase-2 expression (16). Animal models suggested the mitogenic action of gastrin to be limited to the left colon, elevated gastrin levels are more pronounced in their associations with rectal than with colon cancer, and the relation between hypergastrinemia and colorectal adenomas confers an increased risk only for distal colon adenomas. These findings are consistent with and may explain our findings and Zhang et al. (5) findings of selective risk increase with respect to left-sided CRC and adenomas. Experimental data indicate that Hp-I leads to development of chronic inflammation, hyperplasia, metaplasia, dysplasia and recruitment and accumulation of bone marrow-derived cells (BMDCs) which may contribute to tumor formation in animal models with Hp-induced chronic gastric inflammatory process (9,13). Because Hp similarly induces the mentioned inflammatory changes in colonic mucosa (14), it would be reasonable to further speculate that chronic Hp-I in humans also induces repopulation of the colon with BMDCs that might facilitate colon adenoma and cancer development and progression (9,13). In this regard, our own preliminary studies indicated increased expression of CD44 [a marker of human hematopoietic stem and progenitor cells and cancer stem cells (CSCs)] in malignant colonic tissue in 75.6% patients with CRC (11). Extending these preliminary data, increased expression of CD44 was observed in 78% and 16% of patients with cancers and polyps, respectively (12). We also obtained comparable data with gastric cancer (9,13). Therefore, these findings suggest the possible BMDCs and/or CSCs involvement in Hp-associated gastric cancer development and colon adenoma and cancer growth and/or progression (9,13). However, larger-scale future studies are warranted to show that the BMDCs move into areas of the upper and lower gastrointestinal tract and/or CSCs might be induced in the context of Hp chronic injury or inflammation with potential long-term colon adenoma malignant consequences in Hp-positive subjects. Finally, it is important to know if the authors (1) considered relative pathologic aspects in their studies.

Helicobacter pylori induced cognitive dysfunction might be associated with falls and fractures in cirrhosis

1. Lampertico P, Vigano M, Cheroni C, Facchetti F, Invernizzi F, Valveri V, et al. IL28B polymorphisms predict interferon-related HBsAg seroclearance in genotype D HBeAg-negative patients with chronic hepatitis B. HEPATOLOGY 2013;57:890-896. 2. Sonneveld MJ, Wong VW, Woltman AM, Wong GL, Cakaloglu Y, Zeuzem S, et al. Polymorphisms near IL28B and serological response to peginterferon in HBeAg-positive patients with chronic hepatitis B. Gastroenterology 2012;142:513-520. 3. Fischer J, Bohm S, Scholz M, Muller T, Witt H, George J, et al. Combined effects of different interleukin-28B gene variants on the outcome of dual combination therapy in chronic hepatitis C virus type 1 Infection. HEPATOLOGY 2012 [Epub ahead of print].

Association between Helicobacter pylori burden and Alzheimer's disease

Keywords: Alzheimers disease; bacterial infections; cognitive disorders and dementia; infections; mild cognitive impairment; neurological disorders

Relationship between Helicobacter pylori infection and autoimmune disorders.

* Corresponding author: Jannis Kountouras , MD, PhD, Professor of Medicine, Department of Gastroenterology, Second Medical Clinic, Aristotle University of Thessaloniki, Ippokration Hospital, 8 Fanariou St, Byzantio, 551 33 Thessaloniki, Greece, Phone: + 3

Active Helicobacter pylori infection is associated with colorectal mucosa – adenomatous polyp – early and advanced adenocarcinoma sequence

To the Editor, Shmuely et al. [1] concluded that Helicobacter pylori infection (Hp-I) serology is associated with advanced colorectal neoplasia (ACRN) development. They postulated that (a) opposing histology, serology may remain positive even after curing the infection, and the latter is an advantage, as the past infectionmay be even more important for oncogenesis; (b) the specific neoplasia colorectal location was not associated with Hp-I presence, though their data were not shown; and (c) a few possible physiopathological mechanisms for Hp-I and ACRN association include bacterial direct oncogenic effect or hypergastrinemia indirect trophic effect on colorectal mucosa. However, the serological test does not discriminate between current and past infections and, apart from past infection that might even be more relevant for oncogenesis, such a distinction is crucial because only current Hp-I induces humoral and cellular immune responses that induce or perpetuate chronic inflammatory processes in gastrointestinal tract with potential oncogenicsequelae;manymalignancies, includingcolorectal carcinoma (CRC), arise at the sites of chronic inflammation and infection [2,3]. Likewise, two recent large-scale relative epidemiologic studies also suggest that serologic measurement of infection status is less than perfect, thereby representing a limitation of their estimations [4,5]; one study [5] was also mentioned by the authors [1]. Basedonhistology, thegoldstandardforcurrentHp-I diagnosis, ourdata in 50CRCpatients, 25patientswith colorectal adenomas (CRAs) and 10 controls, showed significantly higher presence ofHp-I in theCRA (68%) andCRC (84%) groups comparedwith controls (30%) [6,7]; Hp presence was documented by immunohistochemical stain in colonic tissues.Our series also showed a tendency of increasedHppresence in left-sided (79%) compared with the proximal colon (21%) adenomas; left-sidedcancers(LSCs)wereobservedin64.3%ofour patients, a finding also noticed by others [4], thereby suggesting a possibleHp-I-related risk for LSC in some subpopulations. PresenceofHp-Iwith immunohistochemical expression of CD44 [cancer stem cells (CSCs) and/or bone marrow-derived stem cells (BMDSCs) indicator] was found in a high proportion of CRA patients with moderate-severe dysplasia (88%) and CRC patients with moderate-severe degree of malignancy

Helicobacter pylori infection in a Greek cohort with biliary disease.

SIR, In a recent study, Helaly et al.1 showed that Helicobacter pylori infection might be an aetiological factor leading to cholecystitis; gastric colonisation with H. pylori could be a source for gall bladder infection, and the organism may act as a lithogenic component, especially in the context of pure pigmented gallstones. In this regard, using serological anti-H. pylori IgG antibodies, our own relative data indicate presence of past and/or current H. pylori infection in 63/123 (51.2%) patients (64 females, mean age: 63 years) with calcular biliary and pancreatic diseases (cholecystitis/cholangitis and pancreatitis, respectively).2 Moreover, histological presence of H. pylori infection was detected in gall bladder tissue (by cresyl fast violet staining) in 19.3% of Greek cholecystectomised patients (all female).3 Although Helaly et al. report that the pathways of H. pylori penetration into the bile have not been completely explained, they comment that “the possibility is the translocation from the duodenum via Oddi’s sphincter and, moreover, bacterial antigen penetration into the portal circulation and lymphatic vessels is also possible”.1 We also considered the possible pathways of H. pylori migration and colonisation in the biliary tract and its potential involvement in inflammatory biliary diseases.2 Among the risk factors involved in gall bladder stone development, the inflammatory process plays an essential role; chronic biliary inflammation can also contribute to gall bladder cancer.4 In this regard, H. pylori infection could affect the pathophysiology of gall bladder stone creation and its complications, including cholecystitis, cholangitis, pancreatitis and even biliary cancer by the following mechanisms: i) Releasing large amounts of proinflammatory and vasoactive substances, such as interleukin (IL)-1, IL-6, IL-8 and tumour necrosis factor-α (TNFα) or eicosanoids (leukotrienes, prostaglandins catalysed by cyclo-oxygenase enzymes), and acute-phase proteins (fibrinogen, C-reactive protein) involved in a number of inflammatory diseases,5 also including gall bladder disorders;6 H. pylori could indirectly affect extragastric tissues possibly including the gall bladder through the release of numerous cytokines such as TNFα acting at a distance;7 ii) Producing oxidative stress and circulating lipid peroxides5 also involved in gall bladder disease;6 for instance, increased production of oxygen and lipoperoxide free radicals and macrophage inflammatory cytokines (IL-6, IL-1α, TNFα) have been reported in blood in gallstone disease female patients, indicating the macrophages’ dominant role in the inflammatory and oxidative response during gall bladder stone disease in postmenopausal women;6 iii) Influencing the apoptotic process,5 also involved in chronic calculous cholecystitis and gall bladder oncogenesis;4 seropositivity to H. pylori proteins is associated with an increased risk of biliary tract cancer.8 In addition, H. pylori-induced vacA cytotoxin promotes intracellular survival of the bacterium, modulates host immune responses and induces autophagy.9 Subsequently, H. pylori, as an intracellular microorganism, invades and replicates in the cells. The autophagy induction by H. pylori is not only found in macrophages but also in dendritic cells (also expressed in acute and chronic cholecystitis)10 and gastric epithelial cells.9 The bacterium’s residence inside the infected cells will increase its resistance to antimicrobial treatment, avoid neutralisation by anti-H. pylori antibodies, impair antigen presentation, and alter the cellular immune response.9 In turn, the potential influx of activated monocytes infected with H. pylori in the gall bladder may lead to gall bladder-related pathologies; a comparable potential influx of activated monocytes infected with H. pylori through the disrupted blood-brain barrier, induced by several H. pylori-related mediators, in the brain (‘Trojan Horse’ pathway) might also lead to brain pathologies.9 Moreover, comparable to H. pylori translocation from the duodenum via Oddi’s sphincter to the biliary tract, mentioned by Helaly et al.,1 this bacterium may reach the brain through the oral cavity (which appears to act as a permanent reservoir for H. pylori)-nasal cavity-olfactory neuroepithelium pathway, causing central nervous system pathologies.11 Therefore, H. pylori eradication may have a positive impact on H. pylori-related gall bladder and other tissue pathologies. As there is lack of literature showing any demonstrable evidence to support the aforementioned H. pylori-related mechanisms involved in the pathophysiology of gall bladder diseases, large-scale studies are necessary to elucidate these fields. J. Kountouras, E. Tsiaousi, S. Trigonis, C. Zavos, G. Kouklakis Department of Gastroenterology Second Medical Clinic Aristotle University of Thessaloniki Ippokration Hospital Thessaloniki Macedonia Greece

Letter: low risk of colorectal cancer in a Greek cohort of inflammatory bowel disease patients.

SIRS, We thank Dr Ding and colleagues for the interest in our paper. 2 Their updated appraisal of the current literature comparing the effects of lactulose with probiotics in minimal hepatic encephalopathy (MHE) is a useful addition to the current knowledge of the efficacy of these agents. The authors conclude that probiotics may be a better therapeutic option compared with lactulose for the treatment of MHE, due to a lower adverse effect profile and similar efficacy. Although these data are useful, several aspects related to the nature of the probiotic agents need consideration. Probiotics may appear safer than lactulose. However, experience with these agents is limited compared with the longer duration of experience with lactulose. Furthermore, before probiotics can be recommended as the preferred treatment option for MHE, other aspects need consideration. Firstly, the types and dosages of probiotics used in trials have been variable and it is currently unclear which specific species, strain, dose and regimen is most efficacious. Secondly, once treatment with probiotics is initiated, the duration and end point of treatment remain unclear; most of the available probiotics are expensive and can add significantly to the treatment cost for patients. Thirdly, cirrhotics often receive antibiotics such as rifaximin or cephalosporins for prophylaxis of overt HE or spontaneous bacterial peritonitis. The interactions between these agents and probiotics are not known. Future research should focus on these aspects to better elucidate how probiotics best fit into the treatment algorithm for MHE.