Classification and key features of gastric polyps
To ensure a uniform understanding of the terminology used here, this section briefly outlines the main characteristics and the generally accepted views of the three entities considered. Potentially relevant diagnostic and management issues raised by more recent findings are discussed in the subsequent sections. For the complete classification of gastric polyps followed here the reader is referred to Odze's 2004 Textbook of Surgical Pathology of the GI Tract [1].

Hyperplastic polyps (Synonyms: Inflammatory, regenerative):
they represent 70% to 80% of all gastric polyps, and arise most frequently on the background of an inflamed gastric mucosa. They are believed to originate as hyperproliferative responses to tissue injury; foveolar hyperplasia may be the elemental lesion. They are composed by elongated, grossly distorted, branching and dilated hyperplastic foveolae lying in an edematous stroma rich in vasculature, and small, haphazardly distributed smooth muscle bundles; they contain varying degrees of chronic and active inflammation and granulation tissue. Hyperplastic polyps are usually small, measuring 0.5 - 1.5 cm in diameter, but may occasionally reach much larger dimensions. They are often multiple. Frequently, areas of the surface epithelium are eroded; this may result in chronic blood loss and iron-deficiency anemia, one of the most common clinical manifestations of hyperplastic polyps of the stomach. The prevalence of dysplasia has been reported to range from 1% to 20%, generally, but not in all series, more frequent in larger polyps [2]. The overall malignant potential of hyperplastic polyps has been estimated to be less than 2%. Polypoid foveolar hyperplasia, gastric foveolar polyps and gastritis cystica polyposa (characteristic of post-Billroth I and II gastric stumps) may be considered variants of the same basic hyperproliferative foveolar lesion.

Fundic gland polyps: often classified as hamartomatous lesions, fundic gland polyps (generally referred to as "fundic polyps") consist of smooth sessile circumscribed elevations (usually measuring < 0.5 cm) in the oxyntic mucosa. The basic lesion is one or more cystically dilated oxyntic gland. They may occur sporadically, in association with long-term use of proton-pump inhibitors (PPI), and in the familial adenomatous polyposis (FAP) syndrome. Although dissenting opinions have been published, sporadic and PPI-associated fundic polyps are traditionally believed to have neither malignant potential nor ominous associations; a "minimal" cancer risk is often mentioned for FAP-associated fundic polyps.

Adenomas
Gastric adenomas may occur sporadically and in association with FAP. Only the former are discussed here. Adenomas are defined here as circumscribed, pedunculated or sessile proliferations of dysplastic epithelium without detectable invasion of the lamina propria. For the controversies that have arisen between Japanese and Western pathologists with regards to the distinction between adenoma and adenocarcinoma, the reader is referred to the Padova [3] and Vienna Classifications [4] and to a historical accounts of the issue [5]. Sporadic gastric adenomas may be viewed as one of the possible steps in the development of adenocarcinoma: both arise most often on a background of chronic atrophic metaplastic gastritis; they share a common epidemiological pattern; the larger an adenomatous polyp the greater the probability that it contains foci of adenocarcinoma; and a synchronous adenocarcinoma in another area of the stomach has been found in up to 30% of patients with an adenoma. Most adenomas occur in the antrum and they are detected when they are solitary (~80%) and measure less than 2 cm.

Clinico-pathological correlations
To paraphrase somewhat blasphemously John Donne, no polyp is an island. Gastric polyps (like gastric cancer) rarely if ever arise in a normal stomach. Although most gastroenterologists are aware of this, a thorough search for the background on which gastric polyps arise is uncommonly performed at the time the polyp is discovered. A few years ago, when H. pylori was suspected to be the cause of almost every possible gastric and extra-gastric ailment, a colleague and I designed a simple retrospective study aimed at determining the type of gastritis found at the time a gastric polyp was removed. Being in a large medical center with a digestive diseases division that was very active in clinical research and had been the foster home to the Updated Sydney System (the real home being in Pathology!), we assumed that virtually every polyp would be accompanied by a set of representative biopsy specimens from the rest of the stomach. Although the exact figures are now forgotten, no more than one third of the one hundred-something polyps removed in the previous five years had any synchronous biopsy specimens, and in less than 10% of the cases was a topographically defined set of specimens available. Clearly, the study could not be carried out. I suspect that the situation is not very different in most other centers, and it is perhaps our responsibility as pathologists to deliver the message to our clinical colleagues: for an accurate and perhaps even predictive diagnosis of gastric polyps, let us have a good look at the rest of the gastric mucosa.

Hyperplastic polyps
H. pylori , and Chemical Injury
Foveolar hyperplasia is a compensatory response to increased cellular exfoliation from the surface epithelium that can be viewed as a visual surrogate for increased epithelial cell turnover [6, 7] . While greatly hyperplasic foveolae are easily recognized, lesser degrees of foveolar elongation and increased tortuosity may not be apparent. As an empirical tool, it has been suggested that if more that than four cross sections of the same pit are seen in a well-oriented gastric biopsy specimen, one can confidently diagnose foveolar hyperplasia [8]. The diagnosis of hyperplasia is also facilitated by the finding of hyperchromatic nuclei and mitotic activity reaching an increased height of the pit, and by signs of cellular immaturity (mucin depletion, a cuboidal shape, and a high nucleo-cytoplasmic ratio). Foveolar hyperplasia has long been recognized as a prominent feature of bile reflux gastritis [6]. More recently, hyperplasia it has been emphasized as a key feature of the gastropathy associated with long-term NSAIDs treatment [8]. Lesser degrees of hyperplasia are commonly seen in H. pylori gastritis, but when hyperplasia is marked and diffuse it suggests co-existent chemical injury [9].

These two broad categories of conditions (chronic active gastritis caused by H. pylori and chemical injury) are the ones most commonly associated with hyperplastic polyps. In H. pylori gastritis (or in patients who may have had a long-standing H. pylori gastritis that progressed to atrophy with neither activity nor detectable bacteria) extreme focal foveolar hyperplasia, perhaps in response to a localized more severe injury (erosion, superficial ulcer) may evolve into an agglomerate of highly hypertrophic foveolae, and the long-standing processes of inflammation, repair, and proliferation result into the features of a hyperplastic polyp: grossly distorted, long, dilated pits, inflammation, granulation tissue and erosions. Removal of the injury (i.e., eradication of H. pylori) has been shown to cause the regression of hyperplastic polyps in a high proportion of patients (up to 70% in one study) [10].

A histologically similar picture, but usually associated with larger and more expansive lesions, can be found in areas adjacent to gastrojejunostomy stomas. The initial step is likely foveolar hyperplasia in response to p ancreaticoduodenal secretions as well as acids, bile salts, and lysolecithin, the last being produced by the action of phospholipase in pancreatic juice on the lecithin in bile. Duodenogastric reflux results in disruption of the mucus barrier and the direct action of chemicals on the surface epithelium. Loss of the mucous barrier allows back-diffusion of hydrogen ions and secondary injurious effects [11] . The combined injury leads to accelerated exfoliation of surface epithelial cells and a histamine-mediated vascular response that manifests as edema and hyperemia. Repetitive injury may lead to the release of other pro-inflammatory agents, such as platelet-derived growth factor, which among its many actions stimulates smooth muscle and, later on, fibroblastic proliferation [7]. The first macroscopically visible lesion is polypoid foveolar hyperplasia, which may evolve into the large masses of hyperplastic polyps found in some patients at the gastrectomy site.

From hyperplasia to carcinoma
Both isolated hyperplastic polyps and the polypoid aggregates found at gastrectomy sites have a low but definite potential for the development of malignancy. Between 1% and 20% of hyperplastic polyps have been found to harbor foci of dysplasia; furthermore, mutations of the p53 gene, chromosomal aberrations, and microsatellite instability have been detected in these polyps [12, 13, 14, 15, 16] . In the presence of such a wide array of evidence supporting the biological plausibility of a neoplastic evolution, what should the pathologist's attitude be?

A practical approach to hyperplastic polyps



Two considerations may help put these potentially worrisome findings into perspective. First, the polyp under examination is no longer in the patient; if dysplasia, or even non-invasive carcinoma, is present, it has been removed and cured. Second, the molecular studies carried out so far were designed to acquire insights into possible neoplastic mechanisms, not to develop predictive tests: thus, in the clinical setting the performance of immunostains to detect p53 accumulation or microsatellite instability (both easily available and accurate) or gene arrays, would yield results that can neither be interpreted nor used.

More important, in my view, is to be able to evaluate the stomach in which the polyp has arisen. When a hyperplastic polyp of any size with or without dysplasia (by far the more common occurrence) is diagnosed, the gastroenterologist should be personally informed and, unless one has been done recently, a full set of topographically defined biopsy specimens ("gastric mapping") needs to be obtained. If H. pylori-gastritis is present, eradication is warranted, with a follow-up endoscopy after a few months to monitor not only the successful eradication of H. pylori, but also the possible recurrence of polyps.



If extensive atrophy and metaplasia are found, the patient should be considered at risk for gastric cancer, the polyp could be viewed as an alarming lesion, and an individualized surveillance plan (for which guidelines do not exist as yet) should be implemented [17]. If the polyp is obtained from a gastrectomy site, in the absence of dysplasia, the optimal management remains uncertain. Although the European literature has emphasized a high risk for gastric stump cancer, particularly after Billroth I operations [18], a landmark study conducted in the Netherlands revealed a 0.2% overall incidence of adenocarcinoma in post-gastrectomy patients followed for 15 to 46 years. In North America, although case reports and calls for surveillance are occasionally published, a low incidence of cancer has generally been found [19, 20] .

Fundic polyps: a distant connection
The dilated oxyntic glands that form the small mucosal bumps in the mucosa of the gastric corpus are unlikely candidates for the development of malignancy. The epithelium is that of a mixture of flattened oxyntic, chief, and mucous cells. Dysplastic changes are virtually never seen, although some authors have insistently disputed the entirely benign nature of these polyps [21] .

An association of FAP-associated fundic polyps with colonic neoplasia is not unexpected, since in that nosological context the two are expressions of the same inherited condition.

In the last few years, there have been several reports of patients with sporadic fundic polyps and a colorectal neoplasia (adenomas or carcinomas), and a number of investigators have conducted both retrospective and prospective studies to better define the strength of the association. The conclusions of the most recent of these studies [15] are astonishing: 29 of 64 patients (45.3%) with sporadic fundic polyps had colorectal neoplasia (including 8 adenocarcinomas, 3 adenomas with intraepithelial neoplasia, and 18 tubular adenomas). In the control group, similar in size and demographic characteristics, only 6 patients (9.3%) had adenomas. Although there may be considerable methodological problems with the study (judging from the information available in the article), the authors' conclusions ("Our results suggest that it is necessary to conduct a careful diagnostic work-up of the colon in patients with gastric fundic gland polyps") should be taken seriously. It is certainly worthwhile to carry out larger rigorously designed studies in different populations to explore this puzzling association.

Another interesting and potentially important finding is the recent discovery that a significant proportion of sporadic fundic polyps (even those arising in apparent response to long-tem PPI use) of have genetic alterations involving beta-catenin stabilization, as do those associated with FAP [22, 23] . Abnormal expression of the E-cadherin/catenin membrane complex is common in esophageal adenocarcinoma (as well as in other tumors) and occurs early in the dysplasia/carcinoma sequence in Barrett's esophagus, indicating that disturbances in this cell adhesion complex might be important in tumorigenesis and tumor progression in this disorder [24]. The significance of these findings in fundic polyps may not yet be apparent, but the view that these polyps are completely devoid of malignant potential may deserve a second look. The pathologist's role in after the diagnosis of a fundic polyp is limited. No specific studies are warranted on the polyps themselves. However, particularly in the case of multiple polyps, alerting clinicians about the possibility of FAP and reminding them about the possible connection with colorectal neoplasia may be appropriate. If other biopsy specimens from the corpus are available, the telltale signs of chronic PPI use (dilated oxyntic glands, protruding hyperplastic oxyntic cells, and H. pylori located in the lower portions of the glands as well as within the oxyntic cell canaliculi) should also be reported, suggesting that the fundic polyps may be related to pharmacological acid suppression.

Adenomas and the goblet of mucus
The approach to gastric adenomas has not been changed significantly by molecular studies that have confirmed its neoplastic nature. One of the most interesting recent studies reports that the distinction of the type ("intestinal" versus "gastric") may further define the cancer risk (greater in the former) [25]. In the presence of adenomas, however, irrespective of their cellular make up, a complete gastric bioptic mapping is necessary to determine the phenotype of gastritis on which the adenoma arises. As stated above, metaplastic atrophic gastritis with an adenoma is an indication for an aggressive surveillance program.

Summary
No published guidelines exist for the managements of most types of gastric polyps. Since cancer risk in these heterogeneous patients has not been quantified, the pathologist can influence clinicians' attitudes and ultimately benefit their patients by suggesting extensive biopsy protocols that elucidate the status of the gastric mucosa and help make informed decisions on the design of individualized follow-up protocols.

References

1. Turner JR, Odze RD. Polyps of the stomach. Odze RD, Goldblum JR, Crawford JM (eds). Surgical Pathology of the GI Tract, Liver, Biliary Tract, and Pancreas Philadelphia: Saunders, 2004:267-294.
2. Ginsberg GG, al Kawas FH, Fleischer DE et al. Gastric polyps: relationship of size and histology to cancer risk. Am J Gastroenterol 1996;91:714-717.
3. Rugge M, Correa P, Dixon MF et al. Gastric dysplasia: the Padova international classification. Am J Surg Pathol 2000;24:167-176.
4. Schlemper RJ, Riddell RH, Kato Y et al. The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47:251-255.
5. Genta, R. M. Gastric Dysplasia in the East and West. Current Gastroenterology Reports 2. 2001.
6. Dixon MF, O'Connor HJ, Axon AT et al. Reflux gastritis: distinct histopathological entity? J Clin Pathol 1986;39:524-530.
7. Dixon MF. The Components of Gastritis: Histology and Pathogenesis. Graham DY, Genta RM, Dixon MF (eds). Gastritis Philadelphia: Lippincott Williams & Wilkins, 1999:51-66.
8. Quinn CM, Bjarnason I, Price AB. Gastritis in patients on non-steroidal anti-inflammatory drugs. Histopathology 1993;23:341-348.
9. El Zimaity HM, Genta RM, Graham DY. Histological features do not define NSAID-induced gastritis. Hum Pathol 1996;27:1348-1354.
10. Ohkusa T, Takashimizu I, Fujiki K et al. Disappearance of hyperplastic polyps in the stomach after eradication of Helicobacter pylori. A randomized, clinical trial. Ann Intern Med 1998;129:712-715.
11. Hawkey CJ. Nonsteroidal anti-inflammatory drug gastropathy. Gastroenterology 2000;119:521-535.
12. Nogueira AM, Carneiro F, Seruca R et al. Microsatellite instability in hyperplastic and adenomatous polyps of the stomach. Cancer 1999;86:1649-1656.
13. Murakami K, Mitomi H, Yamashita K et al. p53, but not c-Ki-ras, mutation and down-regulation of p21WAF1/CIP1 and cyclin D1 are associated with malignant transformation in gastric hyperplastic polyps. Am J Clin Pathol 2001;115:224-234.
14. Abraham SC, Park SJ, Mugartegui L et al. Sporadic fundic gland polyps with epithelial dysplasia : evidence for preferential targeting for mutations in the adenomatous polyposis coli gene. Am J Pathol 2002;161:1735-1742.
15. Jung A, Vieth M, Maier O et al. Fundic gland polyps (Elster's cysts) of the gastric mucosa. A marker for colorectal epithelial neoplasia? Pathol Res Pract 2002;198:731-734.
16. Weiss MM, Kuipers EJ, Postma C et al. Genome wide array comparative genomic hybridisation analysis of premalignant lesions of the stomach. J Clin Pathol : Mol Pathol 2003;56:293-298.
17. Abraham SC, Singh VK, Yardley JH et al. Hyperplastic polyps of the stomach: associations with histologic patterns of gastritis and gastric atrophy. Am J Surg Pathol 2001;25:500-507.
18. Pointner R, Schwab G, Konigsrainer A et al. Gastric stump cancer: etiopathological and clinical aspects. Endoscopy 1989;21:115-119.
19. Safatle-Ribeiro AV, Ribeiro U, Jr., Reynolds JC. Gastric stump cancer: what is the risk? Dig Dis 1998;16:159-168.
20. Schuman BM, Waldbaum JR, Hiltz SW. Carcinoma of the gastric remnant in a U.S. population. Gastrointest Endosc 1984;30:71-73.
21. Declich P, Ambrosiani L, Grassini R et al. Fundic gland polyps: a still elusive entity on the eve of the year 2000. Pol J Pathol 2000;51:3-8.
22. Torbenson M, Lee JH, Cruz-Correa M et al. Sporadic fundic gland polyposis: a clinical, histological, and molecular analysis. Mod Pathol 2002;15:718-723.
23. Sekine S, Shibata T, Yamauchi Y et al. Beta-catenin mutations in sporadic fundic gland polyps. Virchows Arch 2002;440:381-386.
24. Washington K, Chiappori A, Hamilton K et al. Expression of beta-catenin, alpha-catenin, and E-cadherin in Barrett's esophagus and esophageal adenocarcinomas. Mod Pathol 1998;11:805-813.
25. Abraham SC, Montgomery EA, Singh VK et al. Gastric adenomas: intestinal-type and gastric-type adenomas differ in the risk of adenocarcinoma and presence of background mucosal pathology. Am J Surg Pathol 2002;26:1276-1285.