Fundic Gland Polyp Dysplasia Is Common in Familial Adenomatous Polyposis
2008; Elsevier BV; Volume: 6; Issue: 2 Linguagem: Inglês
10.1016/j.cgh.2007.11.018
ISSN1542-7714
AutoresLaura Bianchi, Carol A. Burke, Ana E. Bennett, Rocío López, Hennie Hasson, James M. Church,
Tópico(s)Gastric Cancer Management and Outcomes
ResumoBackground & Aims: Fundic gland polyps (FGPs) are common in familial adenomatous polyposis (FAP) but have been considered nonneoplastic. Gastric carcinoma arises from FGPs in FAP presumably from a dysplasia–carcinoma pathway. Our study examined the prevalence of FGPs and FGP dysplasia in FAP and identified endoscopic or demographic features associated with FGPs and dysplasia. Methods: Demographic and endoscopic information were obtained prospectively from 75 consecutive subjects undergoing upper-endoscopic surveillance for FAP. Systematic biopsy specimens of FGPs, normal-appearing fundic mucosa, and antral mucosa for Helicobacter pylori were obtained. Multivariable analysis assessed the association of demographic or endoscopic factors with the presence of FGP or FGP dysplasia. Results: FGPs were detected in 88% of subjects and were dysplastic in 41% (38% low grade, 3% high grade). H pylori infection was rare in subjects with vs without FGPs (1.5% vs 33.3%, P = .005). In the multivariable analysis larger FGP size (odds ratio [OR], 4.0; 95% confidence interval [CI], 1.1–14.4), higher stage of duodenal polyposis (OR, 2.3; 95% CI, 1.2–4.5), and antral gastritis (OR, 11.2; 95% CI, 1.2–103.9) were associated with FGP dysplasia. Exposure to acid-suppressive medications was associated with a marked decrease in dysplastic FGPs (OR, 0.14; 95% CI, 0.03–0.64). Conclusions: The majority of FAP patients have FGPs and nearly half will have dysplastic FGPs. There is an inverse relationship between H pylori and FGPs. FGP dysplasia is associated with larger polyp size, increased severity of duodenal polyposis, and antral gastritis. Acid-suppressive therapy use appears protective against dysplasia in FGPs. Background & Aims: Fundic gland polyps (FGPs) are common in familial adenomatous polyposis (FAP) but have been considered nonneoplastic. Gastric carcinoma arises from FGPs in FAP presumably from a dysplasia–carcinoma pathway. Our study examined the prevalence of FGPs and FGP dysplasia in FAP and identified endoscopic or demographic features associated with FGPs and dysplasia. Methods: Demographic and endoscopic information were obtained prospectively from 75 consecutive subjects undergoing upper-endoscopic surveillance for FAP. Systematic biopsy specimens of FGPs, normal-appearing fundic mucosa, and antral mucosa for Helicobacter pylori were obtained. Multivariable analysis assessed the association of demographic or endoscopic factors with the presence of FGP or FGP dysplasia. Results: FGPs were detected in 88% of subjects and were dysplastic in 41% (38% low grade, 3% high grade). H pylori infection was rare in subjects with vs without FGPs (1.5% vs 33.3%, P = .005). In the multivariable analysis larger FGP size (odds ratio [OR], 4.0; 95% confidence interval [CI], 1.1–14.4), higher stage of duodenal polyposis (OR, 2.3; 95% CI, 1.2–4.5), and antral gastritis (OR, 11.2; 95% CI, 1.2–103.9) were associated with FGP dysplasia. Exposure to acid-suppressive medications was associated with a marked decrease in dysplastic FGPs (OR, 0.14; 95% CI, 0.03–0.64). Conclusions: The majority of FAP patients have FGPs and nearly half will have dysplastic FGPs. There is an inverse relationship between H pylori and FGPs. FGP dysplasia is associated with larger polyp size, increased severity of duodenal polyposis, and antral gastritis. Acid-suppressive therapy use appears protective against dysplasia in FGPs. Familial adenomatous polyposis (FAP) is a dominantly inherited colon cancer syndrome. Nearly 100% of individuals with FAP will develop colorectal cancer, usually by the age of 40, unless colectomy is performed. Endoscopic surveillance and prophylactic colectomy in FAP subjects have reduced mortality from colorectal cancer and improved survival.1Belchetz L.A. Berk T. Bapat B.V. et al.Changing causes of mortality in patients with familial adenomatous polyposis.Dis Colon Rectum. 1996; 39: 384-387Crossref PubMed Scopus (147) Google Scholar, 2Heiskanen I. Luostarinen T. Jarvinen H.J. Impact of screening examinations on survival in familial adenomatous polyposis.Scand J Gastroenterol. 2000; 35: 1284-1287Crossref PubMed Scopus (105) Google Scholar With the improved survival from colon cancer, surveillance strategies for FAP patients also must include prevention of extracolonic malignancies. Benign duodenal and periampullary neoplasms are detected in nearly 100% of FAP subjects and are associated with a heightened risk of duodenal cancer.3Burke C.A. Beck G.J. Church J.M. et al.The natural history of untreated duodenal and ampullary adenomas in patients with familial adenomatous polyposis followed in an endoscopic surveillance program.Gastrointest Endosc. 1999; 49: 358-364Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 4Bulow S. Bjork J. Christensen I.J. et al.DAF Study GroupDuodenal adenomatosis in familial adenomatous polyposis.Gut. 2004; 53: 381-386Crossref PubMed Scopus (285) Google Scholar, 5Alexander J.R. Andrews J.M. Buchi K.N. et al.High prevalence of adenomatous polyps of the duodenal papilla in familial adenomatous polyposis.Dig Dis Sci. 1989; 34: 167-170Crossref PubMed Scopus (72) Google Scholar, 6Church J.M. McGannon E. Hull-Boiner S. et al.Gastroduodenal polyps in patients with familial adenomatous polyposis.Dis Colon Rectum. 1992; 35: 1170-1173Crossref PubMed Scopus (143) Google Scholar FAP-associated fundic gland polyps (FGPs) are reported to occur in 20% to 84% of patients.6Church J.M. McGannon E. Hull-Boiner S. et al.Gastroduodenal polyps in patients with familial adenomatous polyposis.Dis Colon Rectum. 1992; 35: 1170-1173Crossref PubMed Scopus (143) Google Scholar, 7Sarre R.G. Frost A.G. Jagelman D.G. et al.Gastric and duodenal polyps in familial adenomatous polyposis: a prospective study of the nature and prevalence of upper gastrointestinal polyps.Gut. 1987; 28: 306-314Crossref PubMed Scopus (166) Google Scholar FGPs have been thought of historically as nonneoplastic, however, several cases of high-grade dysplasia (HGD) and gastric carcinoma arising from FGPs in FAP have been reported.8Attard T.M. Giardiello F.M. Argani P. et al.Fundic gland polyposis with high-grade dysplasia in a child with attenuated familial adenomatous polyposis and familial gastric cancer.J Pediatr Gastroenterol Nutr. 2001; 32: 215-218Crossref PubMed Scopus (35) Google Scholar, 9Coffey Jr, R.J. Knight Jr, C.D. van Heerden J.A. et al.Gastric adenocarcinoma complicating Gardner's syndrome in a North American woman.Gastroenterology. 1985; 88: 1263-1266PubMed Scopus (38) Google Scholar, 10Goodman A.J. Dundas S.A. Scholefield J.H. et al.Gastric carcinoma and familial adenomatous polyposis (FAP).Int J Colorectal Dis. 1988; 3: 201-203Crossref PubMed Scopus (31) Google Scholar, 11Hofgartner W.T. Thorp M. Ramus M.W. et al.Gastric adenocarcinoma associated with fundic gland polyps in a patient with attenuated familial adenomatous polyposis.Am J Gastroenterol. 1999; 94: 2275-2281Crossref PubMed Google Scholar, 12Zwick A. Munir M. Ryan C.K. et al.Gastric adenocarcinoma and dysplasia in fundic gland polyps of a patient with attenuated adenomatous polyposis coli.Gastroenterology. 1997; 113: 659-663Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar These cases and recent pathogenetic studies suggest the presence of a dysplasia–carcinoma sequence.13Abraham S.C. Nobukawa B. Giardiello F.M. et al.Fundic gland polyps in familial adenomatous polyposis: neoplasms with frequent somatic adenomatous polyposis coli gene alterations.Am J Pathol. 2000; 157: 747-754Abstract Full Text Full Text PDF PubMed Scopus (181) Google Scholar, 14Wu T.T. Kornacki S. Rashid A. et al.Dysplasia and dysregulation of proliferation in foveolar and surface epithelia of fundic gland polyps from patients with familial adenomatous polyposis.Am J Surg Pathol. 1998; 22: 293-298Crossref PubMed Scopus (132) Google Scholar The prevalence of dysplasia in FGPs in FAP has not been studied widely, although 1 study of 37 Italian FAP patients reported FGP dysplasia in 44%.15Bertoni G. Sassatelli R. Nigrisoli E. et al.Dysplastic changes in gastric fundic gland polyps of patients with familial adenomatous polyposis.Ital J Gastroenterol Hepatol. 1999; 31: 192-197PubMed Google Scholar We undertook this prospective study to assess the prevalence of FGPs and dysplasia in FGPs in North American patients with FAP. In addition, we investigated whether any clinical, endoscopic, or demographic features may be associated with the presence of FGPs and/or FGP dysplasia. Consecutive patients with FAP undergoing surveillance esophagogastroduodenoscopy (EGD) from August 2005 through February 2007 at a single tertiary referral center (Cleveland Clinic) were enrolled. Subjects gave written informed consent and the study was approved by the Institutional Review Board. Subjects were excluded if they had prior gastric surgery or a personal history of gastric cancer. Medical and demographic data were collected from all subjects including sex, age, race, prior colonic surgery, family history of gastrointestinal malignancies, tobacco exposure, current use of nonsteroidal anti-inflammatory drugs (NSAIDs), and/or acid suppressive therapy. Tobacco exposure was defined as 1 or more pack-years of current or prior tobacco use. Exposure to NSAIDs was defined as current daily use for more than 2 months and acid-suppressive therapy exposure was defined as current daily use of either a proton pump inhibitor or a histamine antagonist for more than 2 months. Each subject underwent EGD with an Olympus endoscope (GIF 160; Olympus, Melville, NY) after conscious sedation. All endoscopic examinations were performed by 1 of 2 endoscopists (L.K.B. or C.A.B.). During endoscopy the presence or absence of FGPs was noted. If FGPs were present, the total number was documented: 1 to 20, 21 to 30, or more than 30. The area of the stomach affected by FGPs was divided visually into 3 equal segments (proximal, middle, and distal). The size of FGPs in each segment was estimated using an open 5-mm biopsy forceps for comparison and recorded as 1 to 4 mm, 5 to 10 mm, and/or greater than 10 mm. For statistical analysis the largest size of the FGP(s) was used. Directed biopsy specimens were taken from large FGPs (>10 mm) and FGPs with irregular shape or surface characteristics. The number of FGPs from which a biopsy specimen was taken was based on the total number of FGPs present: if there were 1 to 20 FGPs, a biopsy specimen from 3 polyps was taken from each segment; if there were 21 to 30 FGPs, a biopsy specimen of 5 polyps was taken from each segment; if there were more than 30 FGPs, a biopsy specimen of 7 polyps was taken from each segment. Random biopsy specimens were obtained from the gastric antrum for detection of Helicobacter pylori and from endoscopically normal–appearing fundic mucosa, except in those subjects whose fundus was carpeted with FGPs. The size and number of antral polyps, if present, were documented and biopsy specimens were obtained. The size (1–4 mm, 5–10 mm, >10 mm) and number (1–4, 5–20, >20) of duodenal polyps, if present, were recorded and duodenal polyp and papillary biopsy specimens were obtained. The stage of duodenal polyposis was graded according to the Spigelman et al16Spigelman A.D. Williams C.B. Talbot I.C. et al.Upper gastrointestinal cancer in patients with familial adenomatous polyposis.Lancet. 1989; 2: 783-785Abstract PubMed Scopus (602) Google Scholar classification, which takes into account duodenal polyp number (1–4, 5–10, >20), size (1–4 mm, 5–10 mm, >10 mm), histology (tubulous, tubulovillous, villous), and grade of dysplasia. All biopsy specimens were fixed in 10% neutral buffered formalin. Giemsa stain of the single antral biopsy specimen was performed and specimens were evaluated for the presence or absence of H pylori. The remainder of the biopsy specimens were stained with H&E. One gastrointestinal pathologist (A.B.) who was blinded to the location, size, and number of FGPs, interpreted all slides. Dysplasia in FGPs was evaluated on histologic criteria of nuclear enlargement, crowding, stratification, and hyperchromasia in the foveolar epithelium overlying the fundic gland polyp. FGPs were classified as follows: (1) negative for dysplasia, (2) indefinite for dysplasia, (3) low-grade dysplasia (LGD), or (4) HGD. An FGP with LGD was defined as one that showed nuclear enlargement, stratification, and hyperchromasia with overall preservation of the nuclear polarity (Figure 1A). HGD was defined as a loss of nuclear polarity and glandular crowding (Figure 1B). If there was a milder degree of nuclear enlargement and hyperchromatism insufficient to be classified as dysplasia, or if there was significant obscuring background inflammation, then these changes were deemed indefinite for dysplasia. To investigate the various demographic, clinical, and endoscopic features potentially associated with the presence of FGPs, the Pearson chi-square test or the Fisher exact test was used to study univariable associations of categoric demographic and endoscopic factors with the presence of dysplasia. The Student t test was used for continuous factors (if distributional assumptions were not met, the nonparametric Wilcoxon rank sum test was used). Similarly, to identify associations of demographic, clinical, and endoscopic features with the presence of FGP dysplasia in FAP subjects, the Pearson chi-square test or the Fisher exact test was used to study univariable associations of categoric demographic and endoscopic factors with the presence of dysplasia. Subjects with histologic changes considered indefinite for dysplasia were excluded from the dysplasia analysis. The Student t test was used for age and the nonparametric Wilcoxon rank sum test was used for duodenal polyposis stage, number, and size of FGPs. Multivariable logistic regression analysis was used to study the associations of demographic and/or endoscopic factors with presence of dysplasia in FGPs. A significance level of 0.05 was used for all analyses. Results were computed with SAS version 9.1 software (SAS Institute, Cary, NC). Seventy-five consecutive subjects with a mean age of 44 ± 12.2 years were enrolled. Eighty-eight percent (66 of 75) of subjects had FGPs. Dysplasia was detected in 41% (27 of 66) of subjects with FGPs. Dysplasia was low grade in 39% (25 of 66), high grade in 3% (2 of 66), and indefinite in 6% (4 of 66). Both subjects with HGD were identified based on targeted biopsy specimens of large (>10 mm) or irregular FGPs. No random biopsy specimens had evidence of HGD. Thirty-nine subjects (52%) were male and the majority of study participants were Caucasian (95%) (Table 1). Sixty-eight subjects (91%) had undergone previous colonic surgery and 7 subjects had an intact colon at the time of study enrollment. Six of the subjects reported a family history of gastric cancer, 4 subjects were adopted, and a fifth individual could not provide a family history. Approximately one third of subjects reported tobacco exposure, NSAID, or acid-suppression therapy use. Gastric adenomas were present in 11% (8 of 75) of subjects and the stage of duodenal polyposis in the majority of subjects was II and III. A similar percentage of subjects were found to have antral (16%; 12 of 75) or fundic (12%; 9 of 75) gastritis. The prevalence of H pylori in our subjects was 5% (4 of 75), and all subjects with H pylori had antral gastritis.Table 1Demographic and Endoscopic Characteristics of all SubjectsCharacteristicAll subjectsTotal subjects, N75Mean age, y (±SD)44 ± 12.2FGP present66 (88%)FGP dysplasia present27 (41%) Low grade25 (38%) High grade2 (3%) Indefinite4 (6%)Sex Female36 (48%) Male39 (52%)Race Caucasian71 (95%) Non-Caucasian4 (5%)Prior surgery IRA33 (44%) IPAA25 (33%) EI10 (13%) Intact colon7 (9%)Family history of gastric cancer6 (8%)Tobacco use24 (32%)NSAID use25 (33%)Acid-suppressive medication use21 (28%)Gastric adenoma8 (11%)Duodenal polyposis stage 08 (11%) I7 (9%) II27 (36%) III23 (31%) IV10 (13%)Histology Antrum GastritisaIncludes acute gastritis and active or inactive chronic gastritis.12 (16%) Intestinal metaplasia3 (4%) FundusbFour subjects had no fundic biopsy owing to carpeting of fundus with FGPs. GastritisaIncludes acute gastritis and active or inactive chronic gastritis.9 (12%) Intestinal metaplasia1 (1%)H pylori present4 (5%)EI, total colectomy and end ileostomy; IPAA, total proctocolectomy and ileal pouch–anal anastomosis; IRA, total colectomy and ileorectal anastomosis.a Includes acute gastritis and active or inactive chronic gastritis.b Four subjects had no fundic biopsy owing to carpeting of fundus with FGPs. Open table in a new tab EI, total colectomy and end ileostomy; IPAA, total proctocolectomy and ileal pouch–anal anastomosis; IRA, total colectomy and ileorectal anastomosis. There were no significant differences in sex, race, age, prior colonic surgery, tobacco, and NSAID use, or family history of gastric cancer between subjects with and without FGPs (Table 2). There was a trend for users of acid-suppressive medications to have FGPs, being reported in 32% (21 of 66) of subjects with FGPs versus 0% (0 of 9) of subjects without FGPs (P = .054). Subjects with FGPs were significantly less likely to have H pylori infection than those without FGPs (2% vs 33%; P = .005), but there was no association between the presence of antral gastritis and FGPs. Fundal gastritis was detected significantly less often in subjects with FGPs (5% vs 66%; P <.001). There was no association between the presence of intestinal metaplasia, gastric adenoma, more severe duodenal polyposis stage, and the presence of FGPs.Table 2Comparison of Subjects With and Without FGPsCharacteristicFGPs (N = 66)No FGPs (N = 9)P valueaP value from Student t test for age, Wilcoxon rank sum test for duodenal polyposis stage, or Fisher exact test otherwise.Mean age, y (±SD)43.6 (±12.5)47.2 (±9.4).32Sex.99 Female32 (48%)4 (44%) Male34 (52%)5 (56%)Race.99 Caucasian62 (94%)9 (100%) Non-Caucasian4 (6%)0Prior colonic surgery IRA29 (44%)4 (44%) IPAA23 (35%)2 (22%) EI8 (12%)2 (22%) Intact colon6 (9%)1 (11%)Family history of gastric cancer5 (8%)1 (11%).55Tobacco use19 (29%)5 (56%).14NSAID use23 (35%)2 (22%).71Acid-suppressive medication use21 (32%)0.054Gastric adenoma6 (9%)2 (22%).24Duodenal polyposis stage.43 07 (11%)1 (11%) I6 (9%)1 (11%) II23 (35%)4 (44%) III20 (30%)3 (33%) IV10 (15%)0Histology Antrum GastritisbIncludes acute gastritis and active or inactive chronic gastritis.9 (14%)3 (33%).11 Intestinal metaplasia3 (5%)0.99 FunduscFour subjects had no fundic biopsy owing to carpeting of fundus with FGPs. GastritisbIncludes acute gastritis and active or inactive chronic gastritis.3 (5%)6 (66%)<.001 Intestinal metaplasia01 (11%).12H pylori present1 (2%)3 (33%).005EI, total colectomy and end ileostomy; IPAA, total proctocolectomy and ileal pouch–anal anastomosis; IRA, total colectomy and ileorectal anastomosis.a P value from Student t test for age, Wilcoxon rank sum test for duodenal polyposis stage, or Fisher exact test otherwise.b Includes acute gastritis and active or inactive chronic gastritis.c Four subjects had no fundic biopsy owing to carpeting of fundus with FGPs. Open table in a new tab EI, total colectomy and end ileostomy; IPAA, total proctocolectomy and ileal pouch–anal anastomosis; IRA, total colectomy and ileorectal anastomosis. Subjects with and without dysplasia were similar in age, sex, race, family history of gastric cancer, and reported tobacco or NSAID exposure (Table 3). The use of acid-suppressive medications was reported significantly less often in subjects with versus without dysplasia (19% vs 43%, P = .042). No significant association was noted between the size or number of FGPs, H pylori infection, antral or fundic gastritis, intestinal metaplasia, or gastric adenoma and dysplasia. There was a trend toward a higher stage of duodenal polyposis being associated with dysplasia (P = .054).Table 3Comparison of Subjects With and Without FGP DysplasiaCharacteristicDysplasia N = 27No dysplasia N = 35P valueaP value from Student t test for age, Wilcoxon rank sum test for duodenal polyposis stage, or the Fisher exact test otherwise.Mean age, y (±SD)41.0 (±11)45 (±8).54Sex.63 Female14 (52%)16 (46%) Male13 (48%)19 (54%)Race.99 Caucasian25 (93%)33 (94%) Non-Caucasian2 (7%)2 (6%)Family history of gastric cancer1 (3%)3 (8%).62Tobacco use5 (19%)12 (34%).15NSAID use11 (41%)11 (31%).45Acid-suppressive medication use5 (19%)15 (43%).042Number of FGPs.25 1–202 (7%)8 (23%) 21–302 (7%)1 (3%) >3023 (85%)26 (74%)Size of FGP, mm.19 1–417 (63%)26 (35%) 5–105 (19%)9 (26%) >105 (19%)0Gastric adenoma4 (15%)2 (6%).39Duodenal polyposis stage.054 03 (11%)4 (11%) I1 (4%)5 (14%) II7 (26%)13 (37%) III9 (33%)11 (31%) IV7 (26%)2 (6%)Histology Antrum GastritisbIncludes acute gastritis and active or inactive chronic gastritis.6 (22%)2 (6%).069 Intestinal metaplasia3 (11%)0.077 FunduscFour subjects had no fundic biopsy owing to carpeting of fundus with FGPs. GastritisbIncludes acute gastritis and active or inactive chronic gastritis.01 (3%).99 Intestinal metaplasia00H pylori present01 (3%).99a P value from Student t test for age, Wilcoxon rank sum test for duodenal polyposis stage, or the Fisher exact test otherwise.b Includes acute gastritis and active or inactive chronic gastritis.c Four subjects had no fundic biopsy owing to carpeting of fundus with FGPs. Open table in a new tab In the multivariable analysis, tobacco consumption was not shown to be associated with dysplasia, however, several factors were (Table 4). The factors associated with dysplasia included an increased stage of duodenal polyposis (odds ratio [OR], 2.3; 95% confidence interval [CI], 1.2–4.5), larger FGP size (OR, 4.0; 95% CI, 1.1–14.4), and presence of antral gastritis (OR, 11.2; 95% CI, 1.2–103.9). Interestingly, the use of acid-suppressive medications had a significant negative association with the presence of dysplasia (OR, 0.14; 95% CI, 0.03–0.64).Table 4Multivariable Regression Analysis of Factors Associated With DysplasiaFactorOR (95% CI)P valueAcid-suppressive medications0.14 (0.03–0.64).01Duodenal polyposis stage Each increase in stage2.3 (1.2–4.5).01 Stage IV vs stage 030.2 (2.2–409.7)Presence of antral gastritis11.2 (1.2–103.9).03Size of largest FGP Each increase in size range4.0 (1.1–14.4).035 Size >1 cm vs size 1–4 mm15.92 (1.2–207.2)Tobacco consumption3.8 (0.79–18.6).096 Open table in a new tab FGPs are identified commonly in individuals with FAP during routine upper-endoscopic surveillance, but the prevalence has been reported to vary widely from 20% to 84% in the United States, Asia, and Europe.6Church J.M. McGannon E. Hull-Boiner S. et al.Gastroduodenal polyps in patients with familial adenomatous polyposis.Dis Colon Rectum. 1992; 35: 1170-1173Crossref PubMed Scopus (143) Google Scholar, 7Sarre R.G. Frost A.G. Jagelman D.G. et al.Gastric and duodenal polyps in familial adenomatous polyposis: a prospective study of the nature and prevalence of upper gastrointestinal polyps.Gut. 1987; 28: 306-314Crossref PubMed Scopus (166) Google Scholar, 15Bertoni G. Sassatelli R. Nigrisoli E. et al.Dysplastic changes in gastric fundic gland polyps of patients with familial adenomatous polyposis.Ital J Gastroenterol Hepatol. 1999; 31: 192-197PubMed Google Scholar, 17Domizio P. Talbot I.C. Spigelman A.D. et al.Upper gastrointestinal pathology in familial adenomatous polyposis: results from a prospective study of 102 patients.J Clin Pathol. 1990; 43: 738-743Crossref PubMed Scopus (141) Google Scholar, 18Watanabe H. Enjoji M. Yao T. et al.Gastric lesions in familial adenomatosis coli: their incidence and histologic analysis.Hum Pathol. 1978; 9: 269-283Abstract Full Text PDF PubMed Scopus (161) Google Scholar In our study, using a prospective systematic approach to documenting FGPs we found the prevalence to be 88%. We found striking inverse associations between FGPs and 2 factors: H pylori and fundal gastritis. Prior studies have reported the inverse relationship between H pylori and both FAP-associated and sporadic FGPs.19Dickey W. Kenny B.D. McConnell J.B. Prevalence of fundic gland polyps in a western European population.J Clin Gastroenterol. 1996; 23: 73-75Crossref PubMed Scopus (38) Google Scholar, 20Nakamura S. Matsumoto T. Kobori Y. et al.Impact of Helicobacter pylori infection and mucosal atrophy on gastric lesions in patients with familial adenomatous polyposis.Gut. 2002; 51: 485-489Crossref PubMed Scopus (53) Google Scholar, 21Watanabe N. Seno H. Nakajima T. et al.Regression of fundic gland polyps following acquisition of Helicobacter pylori.Gut. 2002; 51: 742-745Crossref PubMed Scopus (62) Google Scholar The pathophysiologic basis of this relationship is unclear, however, it has been shown that H pylori organisms were significantly less adherent to ex vivo antral biopsy specimens obtained from subjects with FGPs.22Sakai N. Tatsuta M. Hirasawa R. et al.Low prevalence of Helicobacter pylori infection in patients with hamartomatous fundic polyps.Dig Dis Sci. 1998; 43: 766-772Crossref PubMed Scopus (37) Google Scholar This finding suggests there is some host factor that inhibits H pylori colonization. The inverse relationship between fundic gastritis and FGPs also has been identified previously,22Sakai N. Tatsuta M. Hirasawa R. et al.Low prevalence of Helicobacter pylori infection in patients with hamartomatous fundic polyps.Dig Dis Sci. 1998; 43: 766-772Crossref PubMed Scopus (37) Google Scholar but further investigation is necessary to understand if FGPs are protective against fundal gastritis or if fundal inflammation acts to inhibit growth of FGPs. Previous studies have examined the relationship between acid-suppressive medications and FGPs, but the data are not consistent.23Choudhry U. Boyce Jr, H.W. Coppola D. Proton pump inhibitor-associated gastric polyps: a retrospective analysis of their frequency, and endoscopic, histologic, and ultrastructural characteristics.Am J Clin Pathol. 1998; 110: 615-621Crossref PubMed Scopus (147) Google Scholar, 24Vieth M. Stolte M. Fundic gland polyps are not induced by proton pump inhibitor therapy.Am J Clin Pathol. 2001; 116: 716-720Crossref PubMed Scopus (96) Google Scholar, 25Abraham S.C. Park S.J. Cruz-Correa M. et al.Frequent CpG island methylation in sporadic and syndromic gastric fundic gland polyps.Am J Clin Pathol. 2004; 122: 740-746Crossref PubMed Scopus (17) Google Scholar We identified a trend that supports the association of FGPs with acid suppression. All subjects in our study who reported acid-suppression exposure had FGPs. Dysplasia was detected in 42% of FAP patients with FGPs. This prevalence data, in a US population, were similar to the 44% found in an Italian FAP population.15Bertoni G. Sassatelli R. Nigrisoli E. et al.Dysplastic changes in gastric fundic gland polyps of patients with familial adenomatous polyposis.Ital J Gastroenterol Hepatol. 1999; 31: 192-197PubMed Google Scholar Our study found suggestive evidence that dysplasia is associated inversely with the use of acid-suppressive medications. The use of acid-suppressive agents for their chemopreventive properties has been investigated in Barrett's esophagus.26Ouatu-Lascar R. Fitzgerald R.C. Triadafilopoulos G. Differentiation and proliferation in Barrett's esophagus and the effects of acid suppression.Gastroenterology. 1999; 117: 327-335Abstract Full Text Full Text PDF PubMed Scopus (343) Google Scholar, 27Hillman L.C. Chiragakis L. Shadbolt B. et al.Proton-pump inhibitor therapy and the development of dysplasia in patients with Barrett's oesophagus.Med J Aust. 2004; 180: 387-391PubMed Google Scholar Acid suppression induced differentiation and decreased cellular proliferation in ex vivo biopsy specimens of Barrett's esophagus26Ouatu-Lascar R. Fitzgerald R.C. Triadafilopoulos G. Differentiation and proliferation in Barrett's esophagus and the effects of acid suppression.Gastroenterology. 1999; 117: 327-335Abstract Full Text Full Text PDF PubMed Scopus (343) Google Scholar and seemed beneficial in the prevention of dysplasia and adenocarcinoma in patients with Barrett's esophagus.27Hillman L.C. Chiragakis L. Shadbolt B. et al.Proton-pump inhibitor therapy and the development of dysplasia in patients with Barrett's oesophagus.Med J Aust. 2004; 180: 387-391PubMed Google Scholar Surface epithelial and foveolar cells of FGPs may respond similarly to acid-suppressive medications and therefore the chemopreventive effects of acid-suppressive medications on FGP dysplasia warrant further investigation. Based on our results, dysplasia is associated with increasing stage of duodenal (adenomatous) polyposis, increasing size of the largest FGP, and the presence of antral gastritis. Each increase in stage of duodenal polyposis (0→I/I→II/II→III/III→IV) was associated with a more than 2-fold increase in the odds of having dysplasia. This suggests that the carcinogenic stimulus responsible for the development of duodenal polyposis also may contribute to the development of dysplasia in FGPs. An increase in stage from 0 to IV resulted in a 30 times increase in the odds of having dysplasia. As the size of the largest FGP increased so did the odds of having dysplasia (OR, 4.0; 95% CI, 1.1–14.4). FGPs appear polypoid owing to cystic dilations, but as they grow increased numbers of surface epithelial and foveolar cells are present. FGP size may be associated with dysplasia because larger polyps will have increased numbers of cells exposed to a carcinogen. Finally, antral gastritis was associated with dysplasia (OR, 11.2; 95% CI, 1.2–103.9). It is plausible that the stimulus responsible for antral inflammation may initiate dysplastic changes in FGPs. Dysplasia prevention is important in FAP because multiple cases of gastric cancer arising in FGPs have been reported.9Coffey Jr, R.J. Knight Jr, C.D. van Heerden J.A. et al.Gastric adenocarcinoma complicating Gardner's syndrome in a North American woman.Gastroenterology. 1985; 88: 1263-1266PubMed Scopus (38) Google Scholar, 10Goodman A.J. Dundas S.A. Scholefield J.H. et al.Gastric carcinoma and familial adenomatous polyposis (FAP).Int J Colorectal Dis. 1988; 3: 201-203Crossref PubMed Scopus (31) Google Scholar, 11Hofgartner W.T. Thorp M. Ramus M.W. et al.Gastric adenocarcinoma associated with fundic gland polyps in a patient with attenuated familial adenomatous polyposis.Am J Gastroenterol. 1999; 94: 2275-2281Crossref PubMed Google Scholar, 12Zwick A. Munir M. Ryan C.K. et al.Gastric adenocarcinoma and dysplasia in fundic gland polyps of a patient with attenuated adenomatous polyposis coli.Gastroenterology. 1997; 113: 659-663Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar In addition to these case reports, 8% of our study subjects reported a family member with FAP and gastric cancer and 83% of those subjects had FGPs found on endoscopy. A prolonged life expectancy for FAP patients postcolectomy has led to an increased incidence of and mortality from upper gastrointestinal malignancies; European and Asian FAP registries have documented the incidence of gastric cancer to range from 0.6% to 4.2%.28Jagelman D.G. DeCosse J.J. Bussey H.J. Upper gastrointestinal cancer in familial adenomatous polyposis.Lancet. 1988; 1: 1149-1151Abstract PubMed Scopus (382) Google Scholar, 29Park J.G. Park K.J. Ahn Y.O. et al.Risk of gastric cancer among Korean familial adenomatous polyposis patients Report of three cases.Dis Colon Rectum. 1992; 35: 996-998Crossref PubMed Scopus (73) Google Scholar The reported incidence of gastric cancer among our subjects' family members was high and requires further confirmation, yet surveillance for upper-gastrointestinal cancers remains a priority for comprehensive care of patients with FAP because individuals with HGD detected in FGPs have been reported to undergo prophylactic gastrectomy.8Attard T.M. Giardiello F.M. Argani P. et al.Fundic gland polyposis with high-grade dysplasia in a child with attenuated familial adenomatous polyposis and familial gastric cancer.J Pediatr Gastroenterol Nutr. 2001; 32: 215-218Crossref PubMed Scopus (35) Google Scholar Identifying subjects with dysplasia is oftentimes difficult given the sheer number of FGPs that can be present (100s–1000s) and the lack of uniform distribution of dysplasia. Our study determined that FGP dysplasia is not diffuse because only 12% of subjects had dysplasia identified in biopsy specimens from all 3 gastric segments. Given the patchy nature of LGD, we recommend a systematic approach to biopsies, as performed in our study, in individuals with FAP and gastric polyposis. It is standard of care to offer individuals with FAP lifelong upper-endoscopic surveillance for gastric and duodenal neoplasia. Traditionally, the upper-endoscopic surveillance interval has been determined solely on the stage of duodenal polyposis. Based on our data, which highlight the high prevalence of dysplasia in fundic gland polyposis, we recommend that individuals with FAP and gastric polyposis undergo a systematic approach to biopsy that includes a biopsy of numerous polyps (based on the total number of polyps) from 3 segments of involved mucosa—proximal, middle, and distal. We also recommend targeted biopsies of polyps larger than 10 mm or with an irregular appearance. If no dysplasia is detected in FGPs, surveillance and intervention should continue according to the traditional schema for duodenal polyposis surveillance (Table 5). If gastric polyposis with LGD is found, the surveillance interval should shorten in those with stage 0 duodenal polyposis (Table 6). We recommend that patients with HGD, similar to our 2 study subjects, undergo targeted gastric polypectomy(ies) with repeat endoscopic surveillance in 3 to 6 months and consideration for treatment with daily proton pump inhibitors.Table 5Endoscopic Surveillance in the Absence of Dysplastic FGPsDuodenal polyposisSurveillance intervalMethodStage 05 yEGD/DStage I–II3 yEGD/DStage III1 yEGD/D3 yCEStage IVaPreferred strategy is pylorus-preserving prophylactic duodenectomy, but recommend aggressive surveillance if surgical resection is not pursued.3–6 moEGD/D3 yCED, duodenoscopy with biopsy of the papilla; CE, capsule endoscopy.a Preferred strategy is pylorus-preserving prophylactic duodenectomy, but recommend aggressive surveillance if surgical resection is not pursued. Open table in a new tab Table 6Endoscopic Surveillance in Individuals With Dysplastic Gastric PolyposisDuodenal polyposisFGP dysplasiaSurveillance intervalMethodGastric neoplasm interventionStage 0–IILGD3 yEGD/DNoneStage IIILGD1 y3 yEGD/DCENoneStage IVLGD3–6 mo3 yEGD/DCENoneStage 0–IVHGDaConsider prophylactic gastrectomy if HGD is detected in small (<10 mm) FGPs or normal mucosa and persists for 2–3 follow-up intervals.3–6 mo3 yEGD/DCETargeted polypectomy and consider chemopreventionD, duodenoscopy with biopsy of the papilla; CE, capsule endoscopy.a Consider prophylactic gastrectomy if HGD is detected in small (<10 mm) FGPs or normal mucosa and persists for 2–3 follow-up intervals. Open table in a new tab D, duodenoscopy with biopsy of the papilla; CE, capsule endoscopy. D, duodenoscopy with biopsy of the papilla; CE, capsule endoscopy. We have not encountered any FAP patient with HGD in normal-appearing fundic mucosa or small FGPs. Our recommendation for such a patient would include chemoprevention with an acid-suppressive medication with consideration of adding celecoxib, and repeat surveillance with numerous biopsies in 3 months. If HGD persists for 2 to 3 follow-up intervals, consideration should be given to gastrectomy. Future directions for research regarding dysplasia in FGPs are wide-ranging. More information needs to be gathered regarding the natural history of FGPs and dysplastic FGPs. In addition, future studies should examine a genotype–phenotype association, which may help to predict more definitively which patients will be at risk for developing dysplastic FGPs. Further investigation of chemopreventive agents, especially acid-suppressive medications, is warranted based on our convincing data that the regular use of either proton pump inhibitors or H2 receptor antagonists is protective against dysplasia.
Referência(s)