Fundic Gland Polyps

Given the relative frequency of fundic gland polyps (FGPs) in upper endoscopic examinations, nearly every gastroenterologist encounters patients with one (or more) of these lesions. In most patients, FGPs are an incidental finding of little clinical significance—indeed, they are more common in gastric mucosa without gastritis, Helicobacter pylori infection, or glandular atrophy.¹ This is in sharp contrast to other gastric polyps such as adenomas and hyperplastic polyps, which are linked to a background of mucosal injury, antral and/or fundic gland atrophy, and intestinal metaplasia.² As noted by Spiegel and colleagues, most FGPs are small (<5 mm) and their histologic appearance is innocuous.³ Composed of cystically dilated fundic glands beneath a normal gastric foveolar epithelium, FGPs do not appear to be neoplastic and have at times been regarded as retention cysts or hamartomas.⁴

However, there are still occasional cases in which FGPs pose significant clinical and management issues. This scenario is aptly illustrated by the patient treated by Spiegel and associates, a middle-aged woman on chronic proton pump inhibitor therapy who underwent upper endoscopy for esophageal reflux.³ During the procedure, she was found, incidentally, to have fundic gland polyposis carpeting the gastric body, fundus, and cardia, with the largest polyps measuring up to 3 cm. In this type of patient, three main questions arise: What is the etiology of the polyposis? Do these FGPs have neoplastic potential? Finally, and, most importantly, how should they be managed? Although it may not be possible to reach a definitive answer regarding management, some insight into this issue can be gained by addressing the first two questions.

FGPs arise in two settings: sporadic (possibly linked to proton pump inhibitor use) and familial adenomatous polyposis syndrome (FAP)-associated settings. FGPs are the most common gastric polyps in both groups, as they are found in up to 5.9% of adults undergoing upper endoscopy and 53–84% of adults with FAP¹ As FAP is estimated to occur in only 1 per 6,000–18,000 births, it follows that most FGPs are nonsyndromic. In comparison to sporadic FGPs, FAP-associated FGPs are more likely to be multiple and to occur at a younger age. In an 18-year review of pediatric gastric polyps, Attard and coworkers found that FAP accounted for 81% of FGPs in children and the polyps were multifocal in 85% of children with FAP-associated FGPs.⁵ Sporadic fundic gland polyposis is occasionally also observed in the non-FAP population.⁶ Two recent reports of giant FGPs in the sporadic setting described a 63-year-old man with an 8-cm FGP and a 67-year-old man with a FGP that covered large areas of the gastric body.^7,8 The latter case mimicked gastric carcinoma, with endosonography suggesting irregular thickening of the first three layers of the stomach.⁸

FAP or attenuated FAP, therefore, needs to be excluded in patients with numerous large or dysplastic FGPs. The patient reported by Spiegel and colleagues had a negative family history and had undergone a normal colonoscopy the previous year, supporting the final impression of sporadic giant fundic gland polyposis. Her age was also more typical of the age associated with sporadic FGPs, which has a median of 59 years of age.⁹ However, family history and patient age are not entirely reliable factors in excluding FAP. Approximately 25% of patients with FAP have no relevant family history and presumably represent de novo mutations in the adenomatous polyposis coli (APC) gene. In patients with FGPs plus colonic adenomas (who have fewer than 100), attenuated FAP is a consideration. In these cases, genetic testing for germline mutations in the APC gene can be undertaken. Genetic testing using a combination of DNA sequencing and protein truncation assay will identify most, but not all, such mutations. In negative cases, further testing can be performed to evaluate for germline mutations in the MYH gene.

Understanding the genetics of sporadic and FAP-associated FGPs also opens an alternative method for ruling out FAP or attenuated FAP in patients with fundic gland polyposis. In the setting of FAP, FGPs arise through “second-hit” alterations (somatic mutations or allelic loss on chromosome 5q) in the APC tumor suppressor gene, the same mechanism responsible for colorectal polyps and periampullary adenomas in these patients. Using a combination of loss of heterozygosity assays and direct DNA sequencing of the mutation cluster region in exon 15 of the APC gene, second-hit APC alterations were demonstrated in 51% of FAP-associated FGPs.¹⁰ In contrast, APC alterations are unusual in sporadic FGPs.¹⁰ Instead, most sporadic FGPs contain activating mutations on or near several phosphorylation sites in exon 3 of the Β-catenin oncogene. Β-catenin mutations have been found in 91%,¹¹ 76%,⁶ and 64%¹² of sporadic FGPs, but in none of the FAP-associated FGPs analyzed to date.^6,11 Both types of mutations—inactivation of the APC tumor suppressor gene and activation of the Β-catenin onco-gene—result in stabilization of Β-catenin protein and its abnormal accumulation in affected cells. In a study that specifically examined 8 non-FAP patients with fundic gland polyposis, Torbenson and colleagues showed that at least 2 FGPs from all patients contained independent Β-catenin mutations; however, Β-catenin mutations were never present in the nonpolypoid gastric mucosa of these patients nor were they present in the FGPs of a patient with clinical attenuated FAP⁶ Overall, these results suggest an alternative means for excluding FAP in the setting of fundic gland polyposis: multiple FGPs are biopsied and analyzed for Β-catenin mutations. If a majority contain Β-catenin mutations, FAP/attenuated FAP is essentially excluded.

The presence of clonal APC or Β-catenin mutations in FGPs indicates that they are neoplastic growths, albeit ones that have only very limited potential for malignant transformation. Low-grade dysplasia involving the foveolar epithelium is common in FAP-associated FGPs, with incidence rates of 25%,¹³ 41%,¹⁴ and 44%¹⁵ in three studies from the United States and Italy. Even in the pediatric population, FAP-associated FGPs commonly show low-grade dysplasia. In one study that included 13 children with FAP, 31% of FGPs were dysplastic and 19% had epithelial atypia indefinite for dysplasia.⁵ High-grade dysplasia occasionally arises in FAP-associated ^14,16,17 including one report of an 11-year-old child with high-grade dysplasia, fundic gland polyposis, and a family history of gastric cancer.¹⁸ The risk of gastric adenocarcinoma in FAP is markedly elevated in Japanese and Korean patients, but it has not been shown to be statistically increased in Western populations where the background rate of gastric cancer is low. Nevertheless, several well-documented cases of invasive adenocarcinoma arising from FGPs or fundic gland polyposis have been reported.^19-21

In contrast, sporadic FGPs only rarely show neoplastic progression. Low-grade foveolar dysplasia is diagnosed in only approximately 1% of sporadic FGPs (3 of 270 in one study).¹³ To date, there is only a single case report of high-grade dysplasia arising in sporadic FGPs²² and no cases of adenocarcinoma arising in this setting. The patient discussed by Spiegel and colleagues—as with the two previously reported patients with giant sporadic FGPs—lacked evidence of even low-grade dysplasia despite the high number and large size of her polyps.³

The management of sporadic fundic gland polyposis, therefore, centers more on the concern for FAP and exclusion of colorectal neoplasia than on concern for neoplastic progression in the gastric mucosa. Several studies have claimed an association between sporadic FGPs and the presence of colonic adenomas or even adenocarcinomas^23-25 prompting the suggestion that every patient with a sporadic FGP should have surveillance for colorectal neoplasia.²⁶ However, in the most recent study of 25,687 adults who underwent both upper endoscopy and colonoscopy, there was only a slightly increased prevalence of colonic adenomas and no increased prevalence of colonic adenocarcinoma among patients with sporadic FGPs as compared to those without FGPs.¹ In women, the odds ratio for concomitant adenomas was 1.43, whereas in men it was only 1.15 (and not statistically significant). An interesting (but most likely not clinically significant) finding in that study was a slightly higher prevalence of colonic hyperplastic polyps in men with FGPs. Despite these associations, there was no difference in the rate of colonic adenocarcinomas in women with or without FGPs, and in men, the rate of colorectal cancer was statistically lower in those with FGPs.¹ Therefore, it is difficult to justify colonoscopic surveillance on the basis of sporadic FGPs alone.

There are little data on the best management—or indeed, whether any management is needed at all—for sporadic fundic gland polyposis and giant sporadic FGPs, both of which were found in the patient treated by Spiegel and colleagues. The doctors elected to discontinue the patient's proton pump inhibitor therapy and recommended a surveillance upper endoscopy in 1 year's time.³ These decisions were entirely reasonable given the patient's polyp burden. Unfortunately, there is conflicting evidence regarding the role of acid suppression in the genesis of FGPs, and there is no empirical evidence regarding the role of gastric surveillance. Most studies suggest an association between proton pump inhibitor use and the increasing prevalence of FGPs in the general population.²⁷ However, this has been disputed by some researchers²⁸ and in some patients, the need for acid suppression may outweigh any concerns regarding gastric polyps. Given the overwhelmingly benign nature of sporadic FGPs, therapeutic decisions should be tailored to the clinical circumstances and the patient's wishes.