Blackwell Science, LtdOxford, UK JGHJournal of Gastroenterology and Hepatology0815-93192002 Blackwell Science Asia Pty Ltd 174April 2002 2770

Gastric cancer J Houghton 10.1046/j.0815-9319.2002.02770.x Review Article495502BEES SGML

Journal of Gastroenterology and Hepatology (2002) 17, 495502

HELICOBACTER PYLORI AND OTHER CAUSES OF GASTRIC ULCERATION

Gastric cancer: Laboratory bench to clinic

JEANMARIE HOUGHTON,* JAMES G FOX AND TIMOTHY C WANG

*Department of Medicine, Division of Gastroenterology, University of Massachusetts Medical Center, Worcester and Division of Bioengineering and Environmental Health, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America

Abstract Gastric cancer is the second most common cause of cancer-related mortality worldwide and the 14th overall cause of death. Detection of disease usually occurs at an advanced stage and overall survival rates for gastric cancer are poor. Our current model for gastric cancer progression clearly maintains Helicobacter infection as the primary inducer of gastric metaplastic and neoplastic disease. Helicobacter pylori is a ubiquitous organism, infecting more than half the worlds population. It has been suggested that this infection directly contributes to the formation of gastric cancer in up to 80% of cases; however, gastric malignancy develops in only a subset (< 1%) of infected patients. Therefore, predisposition to Helicobacter-associated gastric cancer is most likely multifactorial, including the interaction of bacterial, host and environmental components. Our understanding of how the organism interacts with the gastric mucosa and synergizes with dietary and other environmental factors to induce malignant mucosal changes is evolving. Indeed, H. pylori has direct effects on the gastric mucosa, but the major factor in disease progression appears to be a robust host Th1 immune response in the setting of a permissive environment. In combination, these factors predispose to the formation of atrophy, metaplasia and gastric cancer. Understanding the interaction of the bacterium with the host and the environment can potentially identify patients most at risk. Identifying potentially removable factors (in addition to H. pylori infection) in the acquisition and progression of neoplastic disease may provide targets for early intervention and prevention strategies. 2002 Blackwell Publishing Asia Pty Ltd Key words: dietary factors, epidemiology, gastric atrophy, gastric cancer, gastrin, Helicobacter pylori, pathogenesis, Th1 immune response.

EPIDEMIOLOGY
Gastric cancer is the second most common cause of cancer-related mortality worldwide and the 14th overall cause of death. There are marked geographic variations in gastric cancer incidence, with the highest rates in Japan, China and South America and much lower rates in Western countries, including the USA.1 Overall survival rates for gastric cancer are poor and depend on the stage of disease at the time of presentation. In Japan, where small intramucosal lesions can be detected endoscopically during screening examinations, transendoscopic gastric mucosal resections result in improved survival2 (see also the review by Inoue in this issue of the Journal). Unfortunately, these types of aggressive screening and prevention programs are not universal and the disease typically presents at later

stages, resulting in a poor prognosis with overall 5-year survival rates of less than 25%.3

HISTOLOGICAL SUBTYPES OF GASTRIC CANCER

There are two major histological subtypes of gastric cancer: (i) intestinal or well-differentiated adenocarcinoma; and (ii) diffuse gastric cancer. Histologically, intestinal-type gastric cancer consists of gland-like structures that mimic intestinal glands. Epidemiologically, intestinal-type gastric carcinoma has been more closely linked to environmental and dietary factors than has the diffuse type. In contrast, diffuse-type gastric carcinoma is more poorly differentiated than the intestinal

Correspondence: JeanMarie Houghton, Assistant Professor of Medicine, University of Massachusetts Medical Center, NRB Second Floor, 364 Plantation Street, Worcester, MA 01605-2324, USA. Email: jeanmarie.houghton@umassmed.edu

496 type and lacks any glandular structure. In addition, diffuse gastric cancer is less strongly associated with environmental factors and is often familial in distribution. The diffuse form of gastric cancer has been linked to mutations in the E-cadherin gene. E-cadherin is a calcium-dependent protein involved in cellcell adhesion at the level of the adherent junctions within the epithelium, and also connects to the actin cytoskeleton via - and -catenin. Continued expression and functional activity of E-cadherin is required for cells to maintain tight association in the epithelium layer. Because Ecadherin acts to maintain the state of adhesion between epithelial cells, it has an important role in tissue differentiation and maintenance of mucosal integrity. Furthermore, E-cadherin is thought to act as an important suppressor of epithelial tumor cell invasion and metastatic spread. Indeed, kindreds have been identied that carry a germline E-cadherin mutation that leads to the formation of non-functional truncated protein products.4 These kindreds have a markedly increased incidence of diffuse-type gastric carcinoma. However, the majority of cases of gastric cancer do not demonstrate germline mutations in E-cadherin. Investigation into mutations of known tumor suppressor genes show that p53 is most commonly mutated (60 70% of cancers),5 while mutations in Ras and Myc are rare.6 Other genetic abnormalities found at high frequency in gastric adenocarcinoma include deletions or suppression of the fragile histadine triad gene (FHIT; 60%), adenomatous polyposis coli gene (APC; 50%) and deleted in colorectal cancer gene (DCC; 50%), while overexpression/amplication of cyclooxygenase 2 (COX-2; 70%), hepatocyte growth factor (HGF)/ scatter factor (SF) (60%), vascular endothelial growth factor (VEGF; 50%), c-met (45%), AIB-1 (40%), catenin (25%), microsatellite stability (2540%) and DNA anneuploidy (6075%) have also been demonstrated.7 Most mutations studied to date appear to accumulate once the cell has undergone malignant transformation,8 so the precise role, if any, they play in initiating malignant transformation is not clear. In addition to the loss of classic tumor suppressor genes and the gain of growth-promoting genes, factors that alter the immune-mediated removal of tumor cells may be important in tumor survival. Fas ligand (Fas L), required for triggering Fas-mediated apoptosis in Fas antigen-bearing cells, is expressed at high levels on the surface of intestinal-type tumors, but at much lower abundance on diffuse-type gastric carcinoma.9 Lymphocytes inltrating Fas L-bearing tumors undergo apoptosis at a high rate, as evidenced by terminal deoxyribonucleotidyl transferase-mediated dUTPdigoxigenin nick end-labeling (TUNEL) analysis; this indicates that Fas L may be involved in a tumor-initiated immune counterattack.6 The differential expression of Fas L suggests that intestinal and diffuse types of gastric carcinoma may arise through distinct mechanisms. In addition to potential modulation of the immune response, there are a number of growth factors and growth receptors, including Cox-2, which are upregulated in both types of gastric cancer; these represent attractive targets for interventional strategies.

J Houghton et al.

FACTORS INVOLVED IN MEDIATING GASTRIC CANCER

Helicobacter pylori
Most gastric cancers are sporadic in nature. They arise secondary to multiple interacting environmental factors. Helicobacter pylori is the single most common and has been classied by the World Health Organization (WHO) as a class 1 carcinogen, although the precise mechanism by which this bacterium causes gastric cancer is not clear.10 Candidate tumor promoters include direct H. pylori bacterial factors, host immune response, dietary cofactors, including high salt and decreased ascorbate, gastrin hormonal responses and decreased acid secretion occurring in the setting of host genetic susceptibility. While diffuse gastric cancers can be familial, all noncardia gastric cancers (both diffuse and intestinal types) are strongly linked to H. pylori infection. This Gramnegative spiral organism infects the gastric mucosa of over half the worlds population; it is the second most common chronic bacterial infection in humans, with an extremely variable natural history (Fig. 1). Most infected people remain asymptomatic, whereas approximately 15% will develop gastric or duodenal ulcer disease and a minority (< 1%) will develop gastric cancer.11 Interestingly, the development of duodenal ulcer is associated with a lower risk of gastric cancer, through mechanisms that are not fully understood. Predisposition to Helicobacter-associated gastric cancer is most likely multifactorial, including the interaction of bacterial, host and environmental components. The development of intestinal-type gastric cancer is a multistep process.12 It involves temporal progression from chronic gastritis to gastric atrophy, intestinal metaplasia, dysplasia and nally gastric cancer. While a number of factors contribute to this transition, it has become clear that H. pylori is the primary trigger for neoplastic progression. The association between H. pylori and gastric cancer has been known for over a decade.13 The early studies underestimated the prevalence of H. pylori in patients with gastric cancer and recent work by Ekstrom et al.14 and Uemura et al.15 using a combination of two or three methods to diagnose H. pylori infection have shown a much stronger association. Indeed, Uemura et al. demonstrated that gastric cancer

Chronic, active gastritis Early adulthood Asymptomatic Body Menetrier's (5060%) gastritis Hyperplastic Duodenal Late adulthood polyps (<1%) ulcer (10%) Atrophic gastritis/ MALT gastric atrophy (25%) lymphoma (<0.01%) Gastric Gastric ulcer (5%) cancer (1%)

Figure 1 The natural history of Helicobacter pylori infection in the USA. MALT, mucosa-associated lymphoid tissue.

Gastric cancer developed in 3% of infected patients, but not in uninfected patients.15 The risk of gastric cancer was highest in patients with atrophy, corpus-predominant gastritis and intestinal metaplasia, further supporting the hypothesis that these lesions represent premalignant mucosal changes. There was also an increased risk of gastric cancer in infected patients with non-ulcer dyspepsia, gastric ulcer and hyperplastic polyps, but not in those with duodenal ulcer. These studies unambiguously demonstrate a signicant association between gastric cancer and Helicobacter infection. Helicobacter has been postulated to induce growth alterations and malignant transformation of the gastric mucosa through mechanisms involving direct bacterial host cell contact and/or bacterial secreted products.16 Helicobacter pylori possesses a type IV secretion system. This enables intimate interaction with gastric epithelial cells and injection of bacterial proteins, such as CagA, directly into host epithelial cells. Once inside the cell, CagA undergoes host cell kinase-mediated phosphorylation that leads to activation of intracellular signaling pathways.17 A similar process is likely involved in the activation of nuclear factor (NF)-B and interleukin (IL)-8 transcription by the organism.11 In addition, other bacterial virulence factors, such as cagE and picB, may play a role in the modulation of apoptosis and the host inammatory response,18 thereby contributing to disease manifestations.

497 gression of premalignant lesions by increasing Helicobacter colonization.

Gastric acid secretion

The level of acid output may also inuence the outcome of Helicobacter infection. Studies by Hansson et al.21 and Uemura et al.15 have conrmed that duodenal ulcer patients have a markedly reduced risk of gastric cancer. This reduced risk may be secondary to the effect of acid hypersecretion on the distribution of H. pylori within the stomach.22 With high acid output, H. pylori remains conned largely to the gastric antrum and the combination of high acid along with H. pylori within the antrum gives rise to duodenal ulcer disease. In contrast, with a low acid output, H. pylori is able to spread to the body and corpus of the stomach, resulting in further inhibition of acid secretion, destruction of parietal and chief cells and progression to atrophy and cancer (Fig. 2).

Gastrin
Gastrin is a peptide hormone produced primarily by G cells located in the gastric antrum that regulates gastric acid secretion and oxyntic gland proliferation. It may synergize with Helicobacter to induce cell growth alterations and, ultimately, gastric tissue damage, although the relationship is not straightforward. Gastrin exists in both amidated (G34 and G17) and non-amidated or incompletely processed forms (progastrin and the glycine-extended gastrins (G34-GLY and G17-GLY)), all of which show biological activity (Fig. 3). Amidated gastrins are the predominant form of gastrin in endocrine cells, while the incompletely processed forms are found in non-endocrine cells (i.e. cancer cells). It

High-salt diet
Non-bacterial factors linked to increased gastric cancer risk function, at least in part, by synergizing with Helicobacter to induce gastric growth alterations. Diets that are high in salt, such as with pickled and smoked foods, soy sauce, dried and salted sh and meats, are associated with a 5080% increased risk of gastric cancer.19 Work from our group has shown that C57BL/6 mice infected with the SS1 strain of H. pylori and fed a highsalt diet (7.5% sodium) develop more pronounced gastric atrophy, hyperproliferation, foveolar hyperplasia and maintain higher bacterial colonization rates compared with mice maintained on a normal diet.20 Therefore, excessive salt may exacerbate Helicobacter-related gastritis, as well as contribute to the histological proHelicobacter pylori infection

Elevated G-gly (or progastrin) Increased G-17

Helicobacter pylori infection

Low G-gly (or progastrin) Increased G-17)

Higher acid output

Low acid output

Lower acid output IL-1b IFNg, TNFa

High acid output

Antral colonization
Body/corpus colonization

Body/corpus colonization

Antral colonization Duodenal ulcer

Duodenal ulcer
Atrophy/cancer

Atrophy/cancer

Figure 2 Impact of acid secretion on Helicobacter pylori disease presentation. Duodenal ulcer patients have a markedly reduced risk (odds ratio 0.6) of gastric cancer.15

Figure 3 Level of incompletely processed gastrins could inuence the outcome from Helicobacter pylori infection. IFN-, interferon gamma; IL-1b, interleukin 1 beta; TNF-, tumor necrosis factor alpha.

498 appears that the predominant form of gastrin present, as well as the ratios of the different forms of gastrin, dictate predilection to disease.23 It now seems clear that the level of incompletely processed gastrins functions to regulate acid secretion. For example, patients with ZollingerEllison syndrome demonstrate an increase in non-amidated gastrins and have severe acid hypersecretion.24,25 Indeed, G-17 and G-gly have been shown in mice to synergize to stimulate acid secretion.26 Similarly, in some studies, duodenal ulcer patients show an increased proportion of incompletely processed gastrins, analogous to patients with ZollingerEllison syndrome.25 In this setting of increased G-gly, there may be a higher acid output leading to the increased duodenal ulcer risk. In contrast, patients who are at increased risk for gastric cancer, such as those with non-ulcer dyspepsia, appear to have a decreased proportion of incompletely processed gastrins, potentially resulting in lower rates of gastric acid secretion. Thus, gastrin processing, and the ratio of non-amidated to amidated gastrins, could inuence the long-term response to H. pylori infection. However, further studies are needed to dene the precise role of gastrin processing in the progression of Helicobacterinduced disease and the role this gut hormone may play in protection against cancer seen in duodenal ulcer patients. In addition to alterations in incompletely processed forms of gastrin, serum G-17 levels are elevated 1.5 twofold in the majority of H. pylori-infected patients.27 However, G-17 levels themselves are not predictive of ulcer disease28 because elevated G-17 is observed in both ulcer and non-ulcer patients. There does appear to be a relationship between G-17 and atrophy and several studies indicate that increased G-17 levels may not be just a marker of atrophy but, rather, may be associated with increased risk of progression to atrophy.29

J Houghton et al. T-helper (Th1) mucosal immune response) progresses to gastric atrophy, while BALB/c mice, which respond to infection with a polarized Th2 response, appear protected from mucosal damage. Strains such as the C3H, which have a mixed Th1/Th2 cytokine prole show intermediate disease, suggesting that cytokines within an immune response interact to form a continuum of disease rather than discrete disease states.32 Although the composite immune milieu most likely dictates disease manifestations, there may be a role for individual cytokines in both the predisposition to and protection from disease. Indeed, the importance of interferon gamma (IFN-), a Th1 cytokine, is illustrated in the IFN--knockout mouse, where lack of IFN- protects infected mice from atrophy whereas the knockout of IL-10, a Th2 cytokine, leads to severe atrophic gastritis.33,34 In addition, a shift in the Th1/Th2 immune response may explain, in part, the African enigma, namely the observation that, in Africa, there is an extremely high rate of Helicobacter infection but a paradoxically low rate of gastric cancer.35 We postulated that this protection from gastric cancer may be related to modulation of the immune response by environmental factors, such as intestinal parasite infestations, which are known to cause a strong Th2 bias. In support of this, we have shown in a recent study that concurrent infection of Helicobacter-infected C57BL/6 mice with an intestinal helminth, Heligmosomoides polygyrus, is able to shift the mucosal immune response from the usual Th1 response to a polarized Th2 response and, in doing so, protect against atrophy and preneoplastic lesions.36 Because the progression to atrophy and cancer is so dependent on a Th1 immune response, it was natural to look for a link between these immune factors and host genetic susceptibility. There has been some evidence for genetic factors in gastric cancer, including a family history in 1015% of cases, an elevated risk in rst-degree relatives, reports of clustering of cases across several generations and, most recently, an increased prevalence of atrophy (a premalignant lesion) in relatives of gastric cancer patients.37,38 Thus, investigators have turned to studies of host cytokine genes and the potential role of single nucleotide polymorphisms (SNP). Attention focused rst on interleukin (IL)-1, a pro-inammatory cytokine shown to induce gastrin release, inhibit acid secretion and promote apoptosis. Specic genotypes are believed to enhance the production of IL-1. Recent studies by El-Omar et al. in patients from Scotland and Poland suggest that high-expressing IL-1 genotypes increase the risk of developing both atrophy and gastric cancer secondary to Helicobacter infection.39 These ndings have been conrmed by another group.40 Furthermore, recent investigations by El-Omar et al. indicate that a combination of IL-1, tumor necrosis factor alpha (TNF-) and IL-10 SNP, which would potentially result in a phenotype of elevated IL-1 and TNF and decreased IL-10, confers a 50-fold increased risk of gastric cancer.41 Cytokine-mediated cell signaling, either directly through receptorligand interaction or via upregulation of growth-modulating signal cascades, may possibly explain the deleterious effects of Th1 cytokines on

Inammation: Role of cytokines

While it is likely that gastrin plays some role in the development of atrophy and cancer, the evidence to date suggests that the most important cofactor in the induction of Helicobacter-related disease is the host immune response. Inammation is a key factor in progression to atrophy and gastric cancer and CD4 lymphocytes appear to be the most important inammatory cell type within the mucosa. Infection of recombinaseactivating gene (RAG)-decient mice, severe combined immunodeciency (SCID) and T-cell-decient mice with Helicobacter results in high levels of gastric bacterial colonization, with minimal tissue damage.30,31 Furthermore, H. pylori infection of B-cell-decient mice results in severe atrophy and metaplasia, identical to what is seen in infected wild-type mice.31 These results clearly demonstrate a crucial role for CD4 T lymphocytes in orchestrating disease. In addition to studies in immune-decient mice, different inbred strains of wild-type mice show differential susceptibility to Helicobacter-induced disease. Once infected, the C57BL/6 strain (which mounts a strong

Gastric cancer gastric mucosal cell growth. In fact, previous data from our laboratory demonstrate increased expression of surface Fas antigen (a member of the TNF superfamily that transduces both apoptotic and proliferative signaling) by cytokines prominent within the Th1-type response: IL-1, TNF- and IFN-. In contrast, IL-4, prominent in the Th2 immune response, interferes with this upregulation and protects against cell death, offering a plausible mechanism through which the cytokine milieu inuences disease in the Helicobacter-infected host.42

499 key differentiation decisions in the oxyntic glands. Thus, ablation of parietal cells leads to the emergence and expansion of undifferentiated precursor cells. This dysregulated proliferation of undifferentiated precursor cells is often associated with the appearance of a mucus neck cell lineage expressing trefoil factor 2 (TFF2) or spasmolytic polypeptide (SP). This peptide is a member of the trefoil factor family. An increase in SP- or TFF2expressing cells is typically observed in regeneration and ulcer healing in the stomach.44 The function of TFF2 is unclear. However, it may play a role in cytoprotection and maintenance of mucosal integrity and repair.44,45 More importantly, it appears that TFF2 is a marker for a metaplastic cell lineage, which we have termed the spasmolytic polypeptide-expressing metaplastic (SPEM) cell lineage. This is strongly associated with gastric dysplasia and appears to be a precursor for gastric cancer.46 In addition to expansion of the proliferative zone and resultant metaplastic cell changes, loss of parietal cells results in achlorhydria, hypergastrinemia and bacterial overgrowth; the latter may also contribute to tissue damage and abnormal cell signaling. Because gastric atrophy increases the risk of gastric cancer, it would be advantageous to have a non-invasive method to detect its presence. Altered pepsinogen I/II ratios and measurement of increased serum gastrin levels have been used as markers of atrophy. However, they are only of modest usefulness; the gold standard remains histological evaluation.

Apoptosis and proliferation

In the normal stomach, apoptosis and proliferation occur at very low levels and are tightly regulated to achieve controlled cell turnover and maintain tissue homeostasis. Apoptosis occurs mostly in the gastric pits, whereas proliferation is conned to the upper-neck region of the gastric isthmus. In response to H. pylori gastritis, there is rst an increase in apoptosis, followed later by an increase in proliferation. The increase in apoptosis occurs, at least in part, through upregulation of the Fas/Fas L pathway.43 In response to this increased cell death (or as a result of other growth signals) proliferation is increased and homeostasis is maintained at the cost of increased cell turnover. Over time, the rate of apoptosis falls, while increased levels of proliferation are maintained. The proliferative zone expands and extends down throughout the gastric glands, resulting in increased cell numbers and the potential for accumulation of genetic defects and progression to cancer (Fig. 4).

ANIMAL MODELS FOR THE STUDY OF HELICOBACTER -RELATED GASTRIC CANCER

The relationship between gastric Helicobacter infection and cancer has been strengthened through the development of a number of animal models. The rst animal model developed was the ferret, in which infection with Helicobacter mustelae in the presence of a nitrosamine was shown to induce gastric cancer.47 Helicobacter pylori infection of Mongolian gerbils was shown in some, but

Spasmolytic peptide (trefoil factor 2)

The increase in apoptosis occurs in a non-random fashion, with parietal cells and, to a lesser extent, chief cells appearing particularly susceptible to apoptotic signaling and removal; resulting in gastric mucosal atrophy. In addition to acid production, the parietal cell regulates

Normal stomach
Apoptosis regulated

Helicobacter pylori gastritis

Fas/FasL Apoptosis IL-1b, TNF-a

Gastric cancer
Apoptosis

Proliferation regulated

Proliferation Proliferation

Figure 4 Cell growth alterations secondary to Helicobacter infection. Fas L, Fas ligand; IL, interleukin; TNF-, tumor necrosis factor-.

Homeostasis Normal cell turnover

Dysequilibrium Increased Cell number Homeostasis Increased cell turnover

500 not all laboratories, to result in intestinal metaplasia and gastric cancer.48 Because of their small size, ease of infection, reproducibility of results and the power of murine transgenic techniques, the mouse model has become the preferred model for Helicobacter infection. Our laboratory has shown that Helibobacter felis infection in the C57BL/6 and insulingastrin (INS-GAS) mice induces atrophy, metaplasia and gastric cancer,29 making these strains ideal models of Helicobacterinduced gastric cancer. Through the use of inbred strains, transgenics and knockouts, the mouse model can be used to assess the importance of host genetic factors and the impact of various aspects of the immune response on disease presentation. For example, the C57BL/6 mouse model responds to H. felis infection with a robust Th1 immune response, thereby closely approximating the human response to H. pylori infection. In this model, initial colonization of the antrum later spreads to the body/ corpus. An increase in apoptosis is followed by an increase in proliferation with resultant loss of parietal and chief cells, an increase in the SPEM cell lineage, intestinal metaplasia and dysplasia followed by invasive gastric cancer. Genetic manipulation of growth promoters and signaling molecules within the C57BL/6 permissive background has enabled detailed study and facilitated a deeper understanding of genetic factors that promote gastric cancer. To address the role of increased gastrin-17 in gastric carcinogenesis, the INS-GAS transgenic mouse was generated on the FVB/N background.29 In this model, gastrin is transcribed from the rat insulin promoter, resulting in pancreatic islet secretion of gastrin and twofold increases in serum G-17. Importantly, there is no change in the serum levels of G-gly, creating a model that mimics the serum G-17/G-gly prole seen in patients who develop gastric atrophy and cancer. As predicted, the INS-GAS mice show an initial increase in parietal cell number and acid secretion but, over time, they progress to atrophy, achlorhydria, foveolar hyperplasia, metaplasia, dysplasia and, eventually, by 20 months of age, invasive gastric cancer.29 This neoplastic progression appears to proceed through intermediate stages of mucosal alteration similar to that observed in human gastric carcinogenesis. As such, it constitutes one of the most authentic and potentially useful transgeneic mouse models of gastric cancer. In INS-GAS mice, progression from preneoplastic lesions to gastric cancer is a slow process, requiring nearly an entire lifetime. This suggests that cofactors, or genetic alterations in addition to hypergastrinemia, may be required to induce cancer. Importantly, as is seen in human disease, Helicobacter infection provides this synergy. The combination of hypergastrinemia and Helicobacter infection results in atrophy and gastric cancer at approximately 8 months of age, supporting the hypothesis that direct effects of the organism or effects of inammatory mediators (Th1 cytokines) within the infected mucosa accelerate the progression to gastric cancer. In contrast with the INS-GAS mouse, which demonstrates accelerated Helicobacter-associated gastric growth abnormalities, the B.6MRL-FASlpr (lpr), a Fas

J Houghton et al. antigen knockout on the C57BL/6 background, is resistant to Helicobacter-induced damage.27 We have shown that the Th1 immune response (specically TNF-, IL-1 and IFN-) upregulates Fas antigen growthregulating signaling to result in elevated rates of apoptosis and proliferation.26 The lpr mice maintain Helicobacter infection, demonstrate similar inammatory inltrates to the wild-type controls, but fail to alter mucosal apoptosis or proliferation levels. Most importantly, the infected lpr mice fail to demonstrate parietal cell loss, mucous cell metaplasia or atrophy, conrming the central role for Fas signaling in the pathogenesis of Helicobacter-induced disease.27 Further studies examining the interplay between gastrin and the Fas pathway are ongoing and will further clarify the complex gastric signaling pathways.

CURRENT MODEL FOR GASTRIC CANCER AND TREATMENT OPTIONS

Our current model for gastric cancer progression clearly maintains Helicobacter infection as the primary inducer of gastric metaplastic and neoplastic disease (Fig. 5). However, our understanding of how the organism induces these changes is evolving. Thus, H. pylori has direct effects on the gastric mucosa, but the major factor in disease progression appears to be a robust Th1 immune response along with other hormonal, dietary and host genetic factors. In combination, these alterations lead to loss of parietal cells, resulting in either compensatory or independent stimulation of proliferation, atrophy and metaplasia. In addition, achlorhydria, bacterial overgrowth and changes in ascorbate and nitrate metabolism can lead to further progression of mucosal growth alterations. As a consequence of these series of interrelated events, there appears to be opportunity for early intervention and/or prevention of preneoplastic and neoplastic gastric mucosal disease. Currently, treatment of gastric cancer is suboptimal. The overall prognosis for gastric cancer is poor, with a 5-year survival rate < 25%. Surgical resection is the only curative approach for advanced gastric cancer, although
Th1 immune responses Increased G-17 IL-1, TNF, IFN (decreased G-gly) Th2 immune response Dietary cofactors IL-4, IL-10 (high salt) Other host genetic factors Helicobacter pylori infection

Normal gastric mucosa

Gastric atrophy

Gastric cancer
Achlorhydria bacterial overgrowth
Ascorbate Nitrosamines

Figure 5 Current model for Helicobacter-mediated gastric cancer. IFN-, -interferon; IL, interleukin; Th, T helper cell.

Gastric cancer endoscopic mucosal resection appears to be promising for early lesions conned within the mucosa (see review by Inoue in this issue of the Journal). A recent study suggests that adjuvant chemoradiotherapy (uorouracil plus leucovorin followed by 4500 cGy radiation) is benecial in those patients who undergo a curative resection.49 However, the median survival for this select group of patients still remains less than 4 years. Thus, it is clear that prevention is the best overall approach to the treatment of gastric malignancy. Because Helicobacter infection appears to be the most consistent factor in gastric cancer induction, the effects of Helicobacter treatment on modulation of gastric cancer risk has been evaluated extensively. Eradication of H. pylori has been demonstrated to be cost-effective as a preventive therapy for gastric cancer if it prevented at least 30% of cancers.50 In limited studies, H. pylori eradication has been suggested to prevent progression and may lead to regression of precursor lesions, such as atrophy.51 It has also been suggested that Helicobacter eradication in patients who have had endoscopic mucosal resection for endothelial gastric cancer can reduce recurrent cancers.2 In addition, a recent study shows that Helicobacter eradication prevents patients from progressing to cancer up to 4.8 1.2 years after eradication therapy. However, the relatively short follow-up time limits the conclusions that can be drawn from the results of this study.52 Conrmation in other populations and longer follow up will be needed to verify these studies. In addition to bacterial eradication, work is ongoing addressing appropriate vaccine strategies as well as dietary modications to include decreased sodium, increased anti-oxidants and increased ascorbate. Ultimately, the goal would be to identify high-risk individuals through mechanisms that identify at-risk genetic alterations, such as the use of single nucleotide polymorphisms. Once identied, these individuals can be targeted for investigation and treatment based on individual characteristics detected.

501
12, 13 and 61 of the ras gene in a high-incidence area for esophageal and gastric cancers. Cancer Res. 1990; 50: 491114. Stemmermann G, Heffelnger SC, Noffsinger A et al. The molecular biology of esophageal and gastric cancer and their precursors: oncogenes, tumor suppressor genes and growth factors. Hum. Pathol. 1994; 25: 968 81. Farinati FC. Genetic, dietary and environmental factors in the pathogenesis of gastric cancer: Study of a high incidence family. Ital. J. Gastroenterol. 1987; 19: 3214. Houghton J, Harrison LE, Hameed M, Choe JK. The Fas pathway is associated with intestinal but not diffuse type gastric adenocarcinoma. In: Brennan MF, Karpeh MS, eds. Proceedings from the 4th International Gastric Cancer Congress, April 30 to May 2, 2001. New York: Monduzzi Editore, 2001; 54751. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Schistosomes, liver ukes and Helicobacter pylori. Lyon, June 7-14, 1994. IARC Monogr. Eval. Carcinog. Risks Hum. 1994; 61: 1241. Kuipers EJ, Uyterlinde AM, Pena AS et al. Long-term sequelae of Helicobacter pylori gastritis. Lancet 1995; 345: 15258. Correa P. A human model of gastric carcinogenesis. Cancer Res. 1998; 48: 355460. Parsonnet J. Helicobacter pylori and gastric cancer. Gastroenterol. Clin. North Am. 1993; 22: 89104. Ekstrom AM, Held M, Hanson LE, Engstrand L, Nyren O. Helicobacter pylori in gastric cancer established by CagA immunoblot as a marker of past infection. Gastroenterology 2001; 121: 78491. Uemura N, Okamoto S, Yamamoto S et al. Helicobacter pylori infection and the development of gastric cancer. N. Engl. J. Med. 2001; 345: 7849. Mobley HLT. Helicobacter pylori factors associated with disease development. Gastroenterology 1997; 113: S21 S28. Segal ED, Cha J, Lo J, Falkow S, Tompkins LS. Altered states: Involvement of phosphorylated CagA in the induction of host cellular growth changes by Helicobacter pylori. Proc. Natl Acad. Sci. USA 1999; 96: 14 55964. Tummurau MK, Sharma SA, Blaser MJ. Helicobacter pylori picB, a homologue of the Bordetella pertussis toxin secretion protein, is required for induction of IL-8 in gastric epithelial cells. Mol. Microbiol. 1995; 18: 86776. Howson CP, Hiyama T, Wynder EL. The decline in gastric cancer: Epidemiology of an unplanned triumph. Epidemiol. Rev. 1986; 8: 1127. Fox JG, Dangler CA, Taylor NS, King A, Koh TJ, Wang TC. High-salt diet induces gastric epithelial hyperplasia and parietal cell loss, and enhances Helicobacter pylori colonization in C57BL/6 mice. Cancer Res. 1999; 59: 48238. Hansson LE, Nyren O, Hsing AW et al. The risk of stomach cancer in patients with gastric or duodenal ulcer disease. N. Engl. J. Med. 1996; 335: 2429. Dixon M. Acid, ulcers and H. pylori. Lancet 1993; 342: 3845. Dockrey GJ, Varro A, Dimaline R, Wang T. The gastrins: Their biological activities. Annu. Rev. Physiol. 2001; 63: 11939.

10

11

12 13 14

15

16

17

REFERENCES
1 Parkin DM, Laara E, Muir CS. Estimates of the worldwide frequency of sixteen major cancers in 1980. Int. J. Cancer 1988; 41: 18497. 2 Uemura N, Mukai T, Okamoto S et al. Effect of Helicobacter pylori eradication on subsequent development of cancer after endoscopic resection of early gastric cancer. Cancer Epidemiol. Biomark. Prevent. 1997; 6: 63942. 3 Chan AO, Wong BC, Lam SK. Gastric cancer: Past, present and future. Can. J. Gastroenterol. 2001; 15: 469 74. 4 Guilford P, Hopkins J, Harraway J et al. E-Cadherin germline mutations in familial gastric cancer. Nature 1998; 392: 4025. 5 Kim JH, Takahashi T, Chiba I et al. Occurrence of p53 gene abnormalities I gastric carcinoma tumors and cell lines. J. Natl Cancer Inst. 1991; 51: 292631. 6 Victor T, Du Toit R, Jordaan AM, Bester AJ, van Helden PD. No evidence for point mutations in codons

18

19

20

21

22 23

502
24 Calam J. Circulating hormones in peptic ulcer disease. In: Walsh JH, Dockray GJ, eds. Gut peptides: Biochemistry and Physiology. New York: Raven, 1994; 65573. 25 Azuma T, Magami Y, Habu Y et al. Carboxyl terminal glycine extended progastrin (gastrin-G) in gastric antral mucoasa of patients with gastric or duodenal ulcer and in gastrinomas. J. Gastroenterol. Hepatol. 1990; 5: 5259. 26 Chen D, Zhao CM, Dockray GJ et al. Glycine extended gastrin synergizes with gastrin 17 to stimulate acid secretion in gastrin-decient mice. Gastroenterology 2000; 119: 75665. 27 Russo F, Jiillo E, Clemente C et al. Circulating cytokines and gastrin levels in asymptomatic subjects infected by Helicobacter pylori. Immunopharmacol. Immunotoxicol. 2001; 23: 1324. 28 Graham DY, Go MF, Lew GM et al. Helicobacter pylori infection and exaggerated gastrin release: Effects of inammation and progastrin processing. Scand. J. Gastroenterol. 1993; 28: 6904. 29 Wang TC, Dangler CA, Chen D et al. Synergistic interaction between hypergastrinemia and Helicobacter infection in a mouse model of gastric cancer. Gastroenterology 2000; 118: 3647. 30 Eaton KA, Mefford M, Thevenot T. The role of T cell subsets and cytokines in the pathogenesis of Helicobacter pylori gastritis in mice. J. Immunol. 2001; 166: 745661. 31 Roth KA, Kapadia SB, Martin SM, Lorenz RG. Cellular immune responses are essential for the development of Helicobacter felis-associated gastric pathology. J. Immunol. 1999; 163: 14907. 32 Wang TC, Goldenring JR, Dangler C et al. Mice lacking secretory phospholipase A2 show altered apoptosis and differentiation with Helicobacter felis infection. Gastroenterology 1998; 114: 67589. 33 Smythies LE, Waites KB, Lindsey JR et al. Helicobacter pylori-induced mucosal inammation is Th1 mediated and exacerbated in IL-4 but not IFN-gamma, gene-decient mice. J. Immunol. 2000; 165: 10229. 34 Sutton P, Kolesnikov T, Danon S et al. Dominant nonresponsiveness to Helicobacter pylori infection is associated with production of interleukin 10 but not gamma interferon. Infect. Immun. 2000; 68: 48024. 35 Holcombe C. Helicobactor pylori: The African enigma. Gut 1992; 33: 42931. 36 Fox JG, Beck P, Dangler CA et al. Concurrent enteric helminth infection modulates inammation, gastric immune responses and reduces Helicobacter induced gastric atrophy. Nat. Med. 2000; 6: 53642. 37 El-Omar EM, Oien K, Murray LS et al. Increased prevalence of precancerous changes in relatives of gastric cancer patients: critical role of H. pylori. Gastroenterology 2000; 1189: 2230. 38 Carneiro F, Taveira-Gomes A, Cabral-Correia A et al. Characteristics of the gastric mucosa of direct relatives of

J Houghton et al.
patients with sporadic gastric carcinoma. Eur. J. Cancer Prev. 1993; 2: 23946. El-Omar EM, Carrington M, Chow WH et al. Interleukin 1 polymorphisms associated with increased risk of gastric cancer. Nature 2000; 404: 398402. Keller G, Rudelius M, Vogelsang H et al. Microsatellite instability and loss of heterozygosity in gastric carcinoma in comparison to family history. Am. J. Pathol. 1998; 152: 12819. El-Omar EM, Chow WH, Gammon MD et al. Proinammatory genotypes of IL-1, TNF- and IL-10 increase risk of distal gastric cancer but not of cardia or oesophageal adenocarcinomas. Gastroenterology 2001; 120 (Suppl. 1): A12086. Houghton J, Macera-Bloch LS, Harrison L et al. Tumor necrosis factor alpha and interleukin 1 up-regulate gastric mucosal Fas antigen expression in Helicobacter pylori infection. Infect. Immun. 2000; 68: 118995. Houghton J, Macera-Bloch LS, Goldstein M et al. In vivo disruption of the Fas pathway abrogates gastric growth alterations secondary to Helicobacter infection. J. Infect. Dis. 2000; 182: 85664. Mashimo H, Wu DC, Podolsky DK, Fishman MC. Impaired defense of intestinal mucosa in mice lacking intestinal trefoil factor. Science 1996; 274: 2625. Farrell JJ, Taupin D, Koh TJ et al. TFF2/SP-decient mice show decreased gastric proliferation, increased acid secretion, and increased susceptibility to NSAID injury. J. Clin. Invest. 2002; 109: 193204. Wang TC, Goldenring JR, Dangler C et al. Mice lacking secretory phospholipase A2 show altered apoptosis differentiation with Helicobacter felis infection. Gastroenterology 1998; 114: 67589. Fox JG, Wishnok JS, Murphy JC et al. MNNG-induced gastric carcinoma in ferrets infected with Helicobacter mustelae. Carcinogenesis 1993; 14: 195761. Watanabe T, Tada M, Nagai H et al. Helicobacter pylori infection induces gastric cancer in Mongolian gerbils. Gastroenterology 1998; 115: 6428. Baeza MR, Giannini TO, Rivera SR et al. Adjuvant radiochemotherapy in the treatment of completely resected, locally advanced gastric cancer. Int. J. Radiat. Oncol. Biol. Phys. 2001; 50: 64550. Parsonnet J, Harris RA, Hack HM et al. Modeling cost effectiveness of H. pylori screening to prevent gastric cancer: A mandate for clinical trials. Lancet 1996; 348: 1504. Ohkusa T, Funiki K, Takashimizu I et al. Improvement in atrophic gastritis and intestinal metaplasia in patients in whom Helicobacter pylori was eradicated. Ann. Intern. Med. 2001; 134: 3806. Uemura N, Okamoto S, Yamamoto S et al. Helicobacter pylori infection and the development of gastric cancer. N. Engl. J. Med. 2001; 345: 7849.

39

40

41

42

43

44

45

46

47

48

49

50

51

52