- Email: [email protected]
Helicobacter pylori infection and gastric neoplasia: correlations with histological gastritis and tumor histology
THE AMERICAN JOURNAL OF GASTROENTEROLOGY Copyright © 1998 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.
Vol. 93, No. 8, 1998 ISSN 0002-9270/98/$19.00 PII S0002-9270(98)00289-5
Helicobacter pylori Infection and Gastric Neoplasia: Correlations With Histological Gastritis and Tumor Histology Kunihiko Komoto, M.D., Ken Haruma, M.D., Tomoari Kamada, M.D., Shinji Tanaka, M.D., Masaharu Yoshihara, M.D., Koji Sumii, M.D., Goro Kajiyama, M.D., and Nicholas J. Talley, M.D. First Department of Internal Medicine, Hiroshima University School of Medicine, Hiroshima, Japan and Department of Medicine, Nepean Hospital, University of Sydney, Sydney, Australia
Objective: Several authors have reported an association between Helicobacter pylori (H. pylori) and gastric carcinoma, but the data are conflicting. Atrophic gastritis and intestinal metaplasia (IM) have also been linked to gastric carcinoma, especially the intestinal tumor type. We investigated the relationship between H. pylori infection, gastric neoplasms, and histological gastritis. Methods: A total of 105 patients with gastric carcinoma, 36 patients with gastric adenoma, and 105 age- and sex-matched control subjects were examined for H. pylori infection and histological gastritis. H. pylori status was evaluated by Giemsa staining and IgG serology. Mucosal inflammation, atrophy, and IM were evaluated in biopsy specimens from antrum and corpus. Results: H. pylori seroprevalence was higher in patients with gastric carcinoma (98 of 105, 93%) and adenoma (34 of 36, 94%) than in control subjects (82 of 105, 71%, p < 0.05). H. pylori was more prevalent in patients with noncardia (OR, 5.67; 95% CI, 2.25–14.44) than cardia (OR, 5.20; 95% CI, 0.65– 41.68) tumors. Histologic types and tumor stage (early; OR, 6.60; 95% CI, 2.23–19.69, advanced; OR, 4.27; 95% CI, 1.21–15.03) showed no difference in H. pylori prevalence. Atrophy and IM scores were higher in patients with the intestinal- but not diffuse-type of carcinoma and adenoma than in H. pylori-positive control subjects. Smoking was associated with gastric carcinoma (OR, 3.05; 95% CI, 1.58 –5.93) but not alcohol or coffee use, blood group A, or a family history of gastric cancer. Conclusions: Our results confirm a strong association between H. pylori and gastric carcinoma and adenoma. The intestinal-type gastric carcinoma is associated with atrophic gastritis and IM. (Am J Gastroenterol 1998;93:1271–1276. © 1998 by Am. Coll. of Gastroenterology)
(1– 6). Several studies have also implicated H. pylori infection in the pathophysiology of gastric carcinoma (7–15); but recently, negative studies have been reported (16, 17). Before the discovery of H. pylori, researchers focused on the role that atrophic gastritis and intestinal metaplasia (IM) play in the pathogenesis of gastric carcinoma. Although the incidence of gastric carcinoma in western countries has declined dramatically, it remains the most frequent malignancy and the leading cause of death in Japan (18). Moreover, it is also well known that the pathogenesis of intestinal-type gastric carcinoma may be linked to environmental factors, whereas the diffuse type may have a genetic basis (7). Previous works have reported that the high incidence of blood group A and family history of gastric carcinoma in diffuse-type gastric carcinoma would suggest a genetic origin of this tumor type (19, 20). The aim of this study was to investigate the relationship between H. pylori infection and gastric neoplastic diseases (carcinoma and adenoma) and histological gastritis, and also the relationship between gastric neoplastic diseases and potential risk factors (alcohol, smoking, coffee, blood group A, and family history of gastric carcinoma). MATERIALS AND METHODS Subjects The study included consecutive 141 patients with histologically confirmed gastric carcinoma or adenoma who had undergone endoscopy at Hiroshima University Hospital between 1991 and 1996. There were 105 cases of carcinoma and 36 cases of adenoma. The study also included 105 sexand age-matched control subjects with no localized lesions in the upper GI tract who were selected from the population of patients who visited the hospital because of dyspepsia or for a mass survey for gastric carcinoma. There were 45 patients with dyspepsia and 60 subjects for a mass survey; no significant difference was found about H. pylori infection between these two groups. The case and control patients were interviewed about consumption of alcohol and coffee, history of smoking, blood groups, and family history of gastric carcinoma.
INTRODUCTION Helicobacter pylori (H. pylori) has been identified as a major cause of chronic gastritis and peptic ulcer disease Received Jan. 29, 1998; accepted April 1, 1998. 1271
1272
KOMOTO et al.
Serologic detection of antibodies to H. pylori Serum samples were independently tested for the presence of antibodies against H. pylori using the highly sensitive and specific enzyme-linked immunosorbent assay (Pylori stat kit, Whittaker Bioproducts, Walkersville, MD) described previously (21–28). In brief, serum specimens from patients were diluted to 1:20 for use in the kit, and the IgG concentration of the sera was determined according to the manufacturer’s instructions. The serum specimens were coded so that the laboratory technician could not distinguish patients from control subjects. A ratio $1.00 was considered positive and a ratio ,1.00 was considered negative, calibrated according to the kit. The test has a reported sensitivity of 96% and specificity of 94% (28). Identification of H. pylori and histological assessment Gastric biopsy specimens were obtained endoscopically under direct vision in all subjects. Two specimens were taken from the lesser curvature of the antrum and two specimens were obtained from the anterior wall and the posterior wall of the middle body. Biopsy specimens were fixed overnight in buffer formalin, enbedded in paraffin, cut to 4-mm thickness, and stained with hematoxylin and eosin and Giemsa. Identification of H. pylori was performed using the Giemsa stain. In accordance with the Sydney system, the degree of mucosal inflammation, atrophy, and IM were classified into four grades as follows: 0 5 none; 1 5 mild; 2 5 moderate; and 3 5 severe (29, 30). Gastric carcinoma was classified into intestinal and diffuse types according to Lauren’s criteria (31). Cardia was defined as the area in the stomach within 20-mm distance from the esophagogastric junction. Gastric adenoma was diagnosed according to the World Health Organization (WHO) criteria (32). Gastric adenoma is typically a sessile, broad-based lesion that is only slightly elevated macroscopically. Histologically, this is composed of an upper compartment of dysplastic crypts and a lower compartment of normal or cystic pyloric glands. The dysplastic crypts are usually straight and unbranched and are lined by crowded, darkly staining cells with elongated, pseudostratified nuclei and small amounts of apical mucin. Statistical analysis Results are given as mean 6 SD. Statistical analyses were performed using the Mann-Whitney U test. A p value ,0.05 was considered significant. Within the gastric carcinoma group, x2 tests were used to evaluate the association between positive antibodies to H. pylori and either carcinoma stage (early versus advanced), site (cardia versus noncardia), macroscopic type (polypoid, ulceration versus others), or histological type of the carcinoma (intestinal versus diffuse). Odds ratios (OR) obtained from 2 3 2 tables. The logistic regression models was used in multivariate analysis of the relationship between gastric neoplasia risk and H. pylori and potential risk factors such as alcohol, smoking, coffee, blood group A, and family history of gastric carci-
AJG – Vol. 93, No. 8, 1998 TABLE 1 Comparison of the Measured Parameters Among Patients With Gastric Carcinoma, Adenoma, and Control Subjects Diseases
Carcinoma
Adenoma
Control
No. M/F Mean age Hp-Antibody positive rate (%) Odds ratio 95% CI Hp-Giemsa stain positive rate (%)
105 82/23 64.9 6 1.2 93.3*
36 28/8 65.5 6 1.8 94.4*
105 82/23 62.4 6 1.1 71.4
5.60 2.33–13.45 96.2*†
6.80 1.54–30.10 77.8
1.00 73.3
* p , 0.05 as compared with control subjects. † p , 0.05 as compared with adenoma.
noma. ORs with confidence intervals (CI) were computed. The OR is an estimation of relative risk. The confidence intervals (CIs) were set at 0.05. RESULTS There were 105 patients with gastric carcinoma, 36 patients with adenoma, and 105 control subjects, of whom 82 (78%), 28 (78%), and 82 (78%), respectively, were men. The mean age of the patients with carcinoma and adenoma and the control subjects was 65 yr, 66 yr, and 62 yr, respectively. H. pylori seroprevalence The prevalence of H. pylori seropositivity was significantly higher in the patients with gastric carcinoma (93%) and adenoma (94%) than in the control subjects (71%), giving ORs of 5.60 (carcinoma) and 6.80 (adenoma), respectively (Table 1). H. pylori was identified by Giemsa stain in 96% of the patients with gastric carcinoma, 78% of those with gastric adenoma, and 73% of the control subjects. H. pylori infection in gastric carcinoma The prevalence of H. pylori seropositivity in the gastric carcinoma patients was 98% in the men and 78% in the women. Patients ,60 yr old had an OR of 19.38; patients 60 – 69 yr had an OR of 4.49; and patients .69 yr had an OR of 2.14 (Table 2). The decreasing OR with age was attributed to increasing prevalence of H. pylori seropositivity with age in the control subjects and a decreasing prevalence of seropositivity with age among the patients. A total of 94% of the patients with early gastric carcinoma showed positive H. pylori antibody (OR, 6.60; 95% CI, 2.23–19.69), whereas 91% of the patients with advanced gastric carcinoma had serologic evidence of H. pylori infection (OR, 4.27; 95% CI, 1.21–15.03). The OR of H. pylori seropositivity for the noncardia (OR, 5.67; 95% CI, 2.25–14.44) gastric carcinoma patients was significantly higher than the odds for control subjects. In an analysis stratified with respect to tumor location, only the OR for a noncardia location was found to be significantly increased
AJG – August 1998
H. PYLORI AND GASTRIC NEOPLASIA
TABLE 2 Risk of H. pylori Infection Among Patients With Gastric Carcinoma in Relation to Control Subjects
Patient Group Analyzed Sex Male (n 5 82) Female (n 5 23) Age #59 (n 5 32) 60 # 69 (n 5 34) $70 (n 5 39) Stage Early (n 5 70) Advanced (n 5 35) Location Noncardia (n 5 91) Cardia (n 5 14) Macroscopic type Polypoid (n 5 26) Ulceration (n 5 58) Others (n 5 21) Histologic type Intestinal (n 5 84) Diffuse (n 5 21) Smoking Noncardia (n 5 91) Cardia (n 5 14)
Positive H. pylori Serology (%)
OR for H. pylori Infection
95% CI
97.6 78.3
12.90 2.77
2.92–57.40 0.74–9.78
96.9 91.2 92.3
19.38 4.49 2.14
2.36–157.59 1.11–18.17 0.47–9.68
94.3 91.4
6.60 4.27
2.23–19.69 1.21–15.03
93.4 92.9
5.67 5.20
2.25–14.44 0.65–41.68
96.2 93.1 90.5
10.00 5.40 3.80
1.30–76.71 1.80–16.28 0.84–17.46
92.9 95.2
5.20 8.00
2.05–13.20 1.02–62.80
93.4 92.9
2.00 4.32
1.04–3.85 0.90–20.86
(OR, 5.67; 95% CI, 2.25–14.44). There was no difference in the prevalence of H. pylori infection in gastric carcinoma according to the histologic type: 93% (78 of 84) in intestinal-type gastric carcinoma and 95% (20 of 21) in diffusetype gastric carcinoma (Table 2). There was also no difference in the prevalence of H. pylori seropositivity according to the stage and macroscopic type of cancer. Antibody levels in both the intestinal- and diffuse-type gastric carcinoma were higher than in the control subjects and they were significantly higher in patients with the intestinal-type gastric carcinoma than in those with adenoma (Table 4). Corpus mucosal examination Among patients with gastric carcinoma, 14 (13.3%) had mild atrophic gastritis, 26 (24.8%) had moderate atrophic gastritis, and 65 (61.9%) had severe atrophic gastritis. In patients with adenoma, five (13.9%) had moderate atrophic gastritis and 31 (86.1%) had severe atrophic gastritis. Among the control subjects, 24 (22.9%) had normal mucosa, seven (6.6%) had superficial gastritis, 36 (34.3%) had mild atrophic gastritis, 11 (10.5%) had moderate atrophic gastritis, and 27 (25.7%) had severe atrophic gastritis. Among seronegative subjects with gastric carcinoma (n 5 7), two (1.9%) had mild atrophic gastritis, one (1.0%) had moderate atrophic gastritis, and four (3.8%) had severe atrophic gastritis. In adenoma patients (n 5 2), 1 (2.8%) had moderate atrophic gastritis and one (2.8%) had severe atrophic gastritis. Among seronegative control subjects (n 5 30), 22 (21.0%) had normal mucosa, four (3.8%) had mild
1273
TABLE 3 Comparison of Histological Corpus Gastritis Among Patients With Gastric Carcinoma, Adenoma, and Control Subjects Mucosal Status
Carcinoma (n 5 105)
Adenoma (n 5 36)
Control (n 5 105)
Normal Superficial gastritis Mild atrophic gastritis Moderate atrophic gastritis Severe atrophic gastritis
0 (0.0%) 0 (0.0%) 14 (13.3%) 26 (24.8%) 65 (61.9%)
0 (0.0%) 0 (0.0%) 0 (0.0%) 5 (13.9%) 31 (86.1%)
24 (22.9%) 7 (6.6%) 36 (34.3%) 11 (10.5%) 27 (25.7%)
atrophic gastritis, three (2.9%) had moderate atrophic gastritis, and one (1.0%) had severe atrophic gastritis (Table 3). No significant differences in the gastritis score were found among intestinal-type gastric carcinoma, diffuse-type gastric carcinoma, adenoma, and H. pylori-positive control subjects in either pyloric or fundic mucosa. Atrophy and IM scores in both pyloric and fundic mucosa were significantly higher in patients with intestinal-type gastric carcinoma and adenoma than in those with diffuse-type gastric carcinoma and H. pylori-positive control subjects. Both atrophy and IM scores were similar between the patients with diffuse-type gastric carcinoma and H. pylori-positive control subjects (Table 4). With respect to the baseline characteristics of the patients, only smoking was related to intestinal-type gastric carcinoma and adenoma (OR, 6.56; 95% CI, 2.86 –15.03, OR, 7.50; 95% CI, 2.05–27.51) in male patients (Table 5). No significant difference was found between patients with diffuse-type gastric carcinoma and control subjects except for H. pylori infection (OR, 8.00; 95% CI, 1.02– 62.80). When the tumors were divided into noncardia and cardia lesions, only noncardia carcinoma was related to smoking (OR, 2.00; 95% CI, 1.04 –3.85). Gastric adenoma was associated with H. pylori infection (OR, 6.80; 95% CI, 1.54 –30.10) and smoking and male gender (OR, 7.50; 95% CI, 2.05–27.51), but not with smoking and female gender (OR, 2.00; 95% CI, 0.11–37.83), alcohol, coffee, blood group A, or a family history of gastric cancer (Table 6). DISCUSSION The main finding in this study was the significantly increased prevalence of H. pylori seropositivity among patients with gastric carcinoma and adenoma compared with age- and sex-matched controls. The association was linked to noncardia tumors and was significant in male patients and in younger cases ,69 years of age. In addition, atrophic gastritis and IM were closely related to the intestinal-type gastric carcinoma and adenoma, after controlling for H. pylori infection. Smoking was also found to be associated with the intestinal-type gastric carcinoma and adenoma in men but not in women. The finding of a high incidence of chronic gastritis in patients with gastric carcinoma and adenoma supports pre-
1274
KOMOTO et al.
AJG – Vol. 93, No. 8, 1998
TABLE 4 Comparison of the Measured Parameters Among Patients With Gastric Carcinoma, Adenoma, and Control Subjects Carcinoma
Diseases Optical density Gastritis score Pyloric mucosa Fundic mucosa Atrophy score Pyloric mucosa Fundic mucosa Intestinal metaplasia score Pyloric mucosa Fundic mucosa
Intestinal (n 5 77)
Diffuse (n 5 20)
Adenoma (n 5 34)
Control (n 5 75)
2.36 6 0.14‡
2.36 6 0.14‡
1.81 6 0.14‡
2.26 6 0.09
1.51 6 0.07 1.82 6 0.07
1.85 6 0.13 2.00 6 0.18
1.38 6 0.09 1.65 6 0.12
1.55 6 0.08 1.81 6 0.09
2.54 6 0.08*† 2.23 6 0.09*†
1.80 6 0.20 1.75 6 0.22
2.76 6 0.09*† 2.74 6 0.08*†
1.59 6 0.11 1.51 6 0.12
1.62 6 0.07* 1.36 6 0.09*
1.15 6 0.17 1.00 6 0.19
1.76 6 0.09*† 1.68 6 0.10*†
1.01 6 0.10 0.87 6 0.10
* p , 0.05 as compared with control subjects. † p , 0.05 as compared with diffuse-type carcinoma. ‡ p , 0.05 as compared with adenoma. TABLE 5 Odds Ratios of Gastric Carcinoma in a Multiple Logistic Regression Model Intestinal
H. pylori Alcohol Smoking Coffee Blood group A Family history of gastric carcinoma
Diffuse
Odds Ratio
95% CI
Odds Ratio
95% CI
5.20 0.92 3.05 0.50 1.04 1.41
2.05–13.20 0.49–1.73 1.58–5.93 0.24–1.01 0.59–1.85 0.64–3.00
8.00 0.71 2.18 1.98 0.63 0.83
1.02–62.80 0.26–1.94 0.77–6.16 0.71–5.46 0.24–1.64 0.17–4.01
TABLE 6 Odds Ratios of Gastric Adenoma in a Multiple Logistic Regression Model
H. pylori Alcohol Smoking Coffee Blood group A Family history of gastric carcinoma
Odds Ratio
95% CI
6.80 1.27 3.05 0.99 0.91 2.04
1.54–30.10 0.55–2.91 0.70–5.30 0.42–2.26 0.45–1.84 0.86–1.02
vious studies (33, 34). All of the gastric carcinomas and adenomas were found in the setting of atrophic gastritis. Similarly, Sipponen et al. (33) have reported a study of 54 patients with gastric carcinoma, among whom 38 (70%) had H. pylori infection. Only five patients (16%) had normal mucosa, but had no evidence of H. pylori infection by serology or histology. Craanen et al. (35) have reported that atrophic mucosal changes were present in 90.3% (28 of 31) of patients with intestinal-type early gastric carcinoma and in 63.6% (14 of 22) of patients with diffuse-type early gastric carcinoma. H. pylori infection was found in the antral mucosa of 61.3% (18 of 31) of patients with intestinal-type early gastric carcinoma and in 54.5% (12 of 22) of
patients with diffuse-type early gastric carcinoma. In western countries, the prevalence of H. pylori infection is 70 – 80% in gastric carcinoma patients, but 10 –20% of gastric carcinoma cases develop in an apparently H. pylori-negative stomach. In our study, seven patients with gastric carcinoma and two patients with gastric adenoma had negative H. pylori serology, but all nine had moderate to severe atrophic gastritis by histology. It is now accepted that prolonged atrophy causes the spontaneous disappearance of H. pylori from the gastric mucosa and a fall in the antibody titer (26). As atrophy and IM become severe, H. pylori colonization is reduced because of hypoacidity and overgrowth with other bacteria (36). Positive serology, however, often occurs in such patients (37– 40). In our study, seven patients with gastric carcinoma and two patients with adenoma were seronegative, but H. pylori was detected by Giemsa in stain, and all had moderate to severe atrophic gastritis. It seems that extended atrophy may decrease colonisation by H. pylori and induce a fall antibody levels (26). Therefore, even our apparently seronegative patients with gastric carcinoma and adenoma probably had previous H. pylori infection. It is well known that the prevalence of both H. pylori and gastric carcinoma is higher in Japan than in western countries. The prevalence of H. pylori infection in this study was also higher than that reported in western countries. Asaka et al. (41), reported that the prevalence of H. pylori infection was significantly higher in patients with gastric carcinoma (88.2%) than in control patients (74.6%), but the frequency of high titer IgG antibodies was lower in patients with advanced gastric carcinoma (84.7%) than early gastric carcinoma (93.0%). Our result are similar to the findings reported by Asaka et al. Kato et al. (17), on the other hand, have indicated that H. pylori IgG antibody was positive in only 51 of 72 in gastric carcinoma cases (70.8%) and similarly in 98 of 143 control subjects (68.5%). This may be explained by selection bias or measurement bias. Parsonett et al. (7) have reported that the increased risk of
AJG – August 1998 gastric carcinoma was confined largely to women (OR, 18.0). This result is not consist with ours, but others have failed to confirm a higher risk in women (16, 42). From this and previous studies (7, 8, 15, 40), it is evident that tumors located in the gastric cardia are less associated with H. pylori infection compared with noncardia cancer. Carcinoma at these two sites is likely to be due to different etiologies. Our study indicates that H. pylori infection may play a major role in the etiology of noncardia cancer, whereas smoking is associated with noncardia cancer. Earlier studies have indicated that there are important clinical, histological, and demographic differences between intestinal and diffuse-type gastric cancer. It is generally accepted that intestinal-type gastric carcinoma is related to atrophic gastritis (8). The prevalence of H. pylori infection was significantly higher in patients with carcinoma or adenoma than in control subjects in the present study (p , 0.05). Atrophic gastritis and IM are important factors, along with H. pylori infection, in the pathogenesis of intestinaltype gastric carcinoma. Although we carefully examined the pathology to search for other risk factors for gastric carcinoma in addition to H. pylori infection, only atrophic gastritis and IM were related to intestinal-type gastric carcinoma. Diffuse type gastric carcinoma was associated with H. pylori infection, but there were no significant differences compared with H. pylori-positive control subjects in the degree of mucosal atrophy, IM, and inflammation. Previous work has reported that the pathogenesis of intestinal-type gastric carcinoma may be linked to environmental factors, whereas the diffuse type may have a genetic basis (7). The high incidence of blood group A and family history of gastric carcinoma in diffuse-type gastric carcinoma would suggest a genetic origin of this tumor type (19, 20). However, we could not detect that any factor other than H. pylori infection was associated with diffuse-type gastric carcinoma. Gastric adenoma is common in Japan, and is thought to be a precancerous lesion. It is recognized that intestinal-type gastric carcinoma may develop from gastric adenoma when the lesion is .2.0 cm in diameter (34). The prevalence of seropositivity in adenoma was similar to that of gastric carcinoma and significantly higher than in control subjects, but identification of H. pylori by Giemsa stain was significantly lower in gastric adenoma than in gastric carcinoma. Atrophy and IM scores were significantly higher in gastric adenoma than in H. pylori-positive control subjects and diffuse-type gastric carcinoma, and were similar between intestinal-type gastric carcinoma and adenoma. Gastric adenoma may develop from atrophic mucosa and may have a pathogenesis similar to intestinal-type gastric carcinoma. In conclusion, our results confirm the association between H. pylori infection and gastric carcinoma. We have also identified that H. pylori is a risk factor for gastric adenoma. Although intestinal-type gastric carcinoma and adenoma appear to be closely related to atrophic gastritis and IM,
H. PYLORI AND GASTRIC NEOPLASIA
1275
diffuse-type carcinoma is not. Only intestinal-type gastric carcinoma is related to smoking. Reprint requests and correspondence: Dr. Ken Haruma, First Department of Internal Medicine, Hiroshima University School of Medicine, 1-2-3- Kasumi, Minami-ku, Hiroshima, 734, Japan.
REFERENCES 1. Warren JR, Marshall BJ. Unidentified curved bacilli on gastric epithelium in active gastritis. Lancet 1983;1273–5. 2. Marshall BJ, Warren JR. Unidentified curved bacilli in the stomach of patients with gastritis and peptic ulceration. Lancet 1984;1311–5. 3. Graham DY. Helicobacter pylori: Its epidemiology and its role in duodenal ulcer disease. J Gastroenterol Hepatol 1991;6:105–13. 4. Graham DY, Klein PD, Evans DG, et al. Helicobacter pylori: Epidemiology, relationship to gastric cancer and the role of infants in transmission. Eur J Gastroenterol Hepatol 1992;4(suppl 1):S1– 6. 5. Marshall BJ. Campylobacter pylori: Its link to gastritis and peptic ulcer disease. Rev Infect Dis 1990;12(suppl 1):S87–93. 6. Siurala M, Sipponen P, Kekki M. Campylobacter pylori in a sample of Finish population: Relations to morphology and functions of the gastric mucosa. Gut 1988;29:909 –16. 7. Parsonnet J, Friedman GD, Wandersteen DP, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991;325: 1127–31. 8. Talley NJ, Zinsmeister AR, Weaver A, et al. Gastric carcinoma and Helicobacter pylori infection. J Natl Cancer Inst 1991;83:1734 –9. 9. Parsonett J, Vandersteen D, Goates J, et al. Helicobacter pylori infection in intestinal and diffuse type gastric carcinoma. J Natl Cancer Inst 1991;83:640 –3. 10. Loffeld RJLF, Willems I, Flendrig JA, et al. Helicobacter pylori and gastric carcinoma. Histopathology 1990;17:537– 41. 11. Forman D, Sitas F, Newell DG, et al. Geographic association of Helicobacter pylori antibody prevalence and gastric cancer mortality in rural China. Int J Cancer 1990;46:608 –11. 12. Correa P, Fox J, Fontham E, et al. Helicobacter pylori and gastric carcinoma: Serum antibody prevalence in populations with contrasting cancer risks. Cancer 1990;66:2569 –74. 13. Fox JG, Correa P, Taylor NS, et al. Campylobacter pylori associated gastritis and immune response in a population at increased risk of gastric carcinoma. Am J Gastroenterol 1989;84:775– 81. 14. The Eurogust Study Group. An international association between Helicobacter pylori infection and gastric cancer. Lancet 1993;341:1359 – 62. 15. Nomura A, Stemmermann GN, Chyou PH, et al. Helicobacter pylori infection and gastric carcinoma among Japanese American in Hawaii. N Engl J Med 1991;325:1132– 6. 16. Kuipers EJ, Gracia-Casanova M, Pena AS, et al. Helicobacter pylori serology in patients with gastric carcinoma. Scand J Gastroenterol 1993;28:433–7. 17. Kato S, Onda M, Matsukura N, et al. Genetic polymorphisms of the cancer related gene and Helicobacter pylori infection in Japanese gastric cancer patients. Cancer (suppl) 1996;77:1654 – 61. 18. Watanabe S, Arimoto H. Long term trends in cancer mortality rates and cumulative rates from 1955 to 1987 in Japan (II). Jpn J Clin Oncol 1989;19:413–25. 19. Wu MS, Sung CT, Wang HP, et al. Genetic alterations in gastric cancer: Relation to histological subtypes, tumor stage, and Helicobacter pylori infection. Gastroenterology 1997;112:1457– 65. 20. La-Vecchia C, D’Avanzo B, Negri E, et al. Attributable risks for stomach cancer in northern Italy. Int J Cancer 1995;60:748 –52. 21. Perez-Perez GI, Dworkin BM, Chodos JE, et al. Campylobacter pylori antibodies in humans. Ann Intern Med 1988;109:11–7. 22. Dooley CP, Cohen H, Fitzgibbons PL, et al. Prevalence of Helicobacter pylori infection and histologic gastritis in asymptomatic persons. N Engl J Med 1989;321:1562– 6. 23. Drumm B, Perez-Perez GI, Blaser MJ, et al. Intrafamilial clustering of Helicobacter pylori infection. N Engl J Med 1990;322:359 – 63. 24. Glassman MS, Dallal S, Berezin SH, et al. Helicobacter pylori-related gastroduodenal disease in children. Diagnostic utility of enzymelinked immunosorbent assay. Dig Dis Sci 1990;35:993–7. 25. Strauss RM, Wang TC, Kelsey PB, et al. Association of Helicobacter
1276
26. 27. 28. 29. 30. 31. 32. 33.
KOMOTO et al.
pylori infection with dyspeptic symptoms in patients undergoing gastroduodenoscopy. Am J Med 1990;89:464 –9. Karnes WE, Samloff IM Jr, Siurala M, et al. Positive serum antibody and negative tissue staining for Helicobacter pylori in subjects with atrophic body gastritis. Gastroenterology 1991;101:167–74. Talley NJ, Kost L, Haddad A, et al. Comparison of commercial serological tests for detect of Helicobacter pylori antibodies. J Clin Microbiol 1992;30:3146 –50. Talley NJ, Newell DG, Ormand JE, et al. Serodiagnosis of Helicobacter pylori: Comparison of enzyme linked immunosorbent assay. J Clin Microbiol 1991;29:1635–9. Whitehead R. The histological diagnosis of chronic gastritis in fibreoptic gastroscope biopsy specimens. J Clin Pathol 1972;25:1–11. Siurala M, Lehtola J, Ihama¨ki T. Atrophic gastritis its sequelae: Results of 19 –23 years’ follow up examinations. Scand J Gastroenterol 1974; 9:441– 6. Lauren P. The two histological main types of gastric carcinoma: Diffuse and so-called intestinal type carcinoma. Acta Path Microbiol Scand 1965;64:31– 49. Ohta K. Histological typing of gastric and esophageal tumors. International Histological Classification of Tumors, No. 18, WHO, Geneva, 1977. Sipponen P, Kosunen TU, Valle J, et al. Helicobacter pylori infection and chronic gastritis in gastric cancer. J Clin Pathol 1992;45:319 –23.
AJG – Vol. 93, No. 8, 1998 34. Todo H, Haruma K, Yoshihara M, et al. Clinical management of group III lesion (in Japanese). Gastroenterol Endosc 1990;32:1133–9. 35. Craanen ME, Blok P, Dekker W, et al. Helicobacter pylori and early gastric cancer. Gut 1994;35:1372– 4. 36. Wee A, Kang JY, Teh M. Helicobacter pylori and gastric cancer: Correlation with gastritis, intestinal metaplasia, and tumour histology. Gut 1992;33:1029 –32. 37. Newell DG, Hawtin PR, Stacey AR, et al. Estimation of prevalence of Helicobacter pylori infection in an asymptomatic elderly population comparing [14C] urea breath test and serology. J Clin Pathol 1991;44: 385–7. 38. Wyatt JI, Shallcross TM, Crabtree JE, et al. Helicobacter pylori, gastritis, and peptic ulceration in the elderly. J Clin Pathol 1992;45: 1070 – 4. 39. DeLuca VA Jr. Is pernicious anemia caused by Campylobacter pylori gastritis (letter). J Clin Gastroenterol 1989;11:584 –5. 40. Fiocca G, Villani L, Turpini F, et al. High incidence of Campylobacterlike organisms in endoscopic biopsies from patients with gastritis, with or without peptic ulcer. Digestion 1987;38:234 – 44. 41. Asaka M, Kimura T, Kato M, et al. Possible role of Helicobacter pylori infection in early gastric cancer development. Cancer 1994;73:2691– 4. 42. Hansson LE, Engstrand L, Nyren O, et al. Prevalence of Helicobacter pylori infection in subtypes of gastric cancer. Gastroenterology 1995; 109:885– 8.
Vol. 93, No. 8, 1998 ISSN 0002-9270/98/$19.00 PII S0002-9270(98)00289-5
Helicobacter pylori Infection and Gastric Neoplasia: Correlations With Histological Gastritis and Tumor Histology Kunihiko Komoto, M.D., Ken Haruma, M.D., Tomoari Kamada, M.D., Shinji Tanaka, M.D., Masaharu Yoshihara, M.D., Koji Sumii, M.D., Goro Kajiyama, M.D., and Nicholas J. Talley, M.D. First Department of Internal Medicine, Hiroshima University School of Medicine, Hiroshima, Japan and Department of Medicine, Nepean Hospital, University of Sydney, Sydney, Australia
Objective: Several authors have reported an association between Helicobacter pylori (H. pylori) and gastric carcinoma, but the data are conflicting. Atrophic gastritis and intestinal metaplasia (IM) have also been linked to gastric carcinoma, especially the intestinal tumor type. We investigated the relationship between H. pylori infection, gastric neoplasms, and histological gastritis. Methods: A total of 105 patients with gastric carcinoma, 36 patients with gastric adenoma, and 105 age- and sex-matched control subjects were examined for H. pylori infection and histological gastritis. H. pylori status was evaluated by Giemsa staining and IgG serology. Mucosal inflammation, atrophy, and IM were evaluated in biopsy specimens from antrum and corpus. Results: H. pylori seroprevalence was higher in patients with gastric carcinoma (98 of 105, 93%) and adenoma (34 of 36, 94%) than in control subjects (82 of 105, 71%, p < 0.05). H. pylori was more prevalent in patients with noncardia (OR, 5.67; 95% CI, 2.25–14.44) than cardia (OR, 5.20; 95% CI, 0.65– 41.68) tumors. Histologic types and tumor stage (early; OR, 6.60; 95% CI, 2.23–19.69, advanced; OR, 4.27; 95% CI, 1.21–15.03) showed no difference in H. pylori prevalence. Atrophy and IM scores were higher in patients with the intestinal- but not diffuse-type of carcinoma and adenoma than in H. pylori-positive control subjects. Smoking was associated with gastric carcinoma (OR, 3.05; 95% CI, 1.58 –5.93) but not alcohol or coffee use, blood group A, or a family history of gastric cancer. Conclusions: Our results confirm a strong association between H. pylori and gastric carcinoma and adenoma. The intestinal-type gastric carcinoma is associated with atrophic gastritis and IM. (Am J Gastroenterol 1998;93:1271–1276. © 1998 by Am. Coll. of Gastroenterology)
(1– 6). Several studies have also implicated H. pylori infection in the pathophysiology of gastric carcinoma (7–15); but recently, negative studies have been reported (16, 17). Before the discovery of H. pylori, researchers focused on the role that atrophic gastritis and intestinal metaplasia (IM) play in the pathogenesis of gastric carcinoma. Although the incidence of gastric carcinoma in western countries has declined dramatically, it remains the most frequent malignancy and the leading cause of death in Japan (18). Moreover, it is also well known that the pathogenesis of intestinal-type gastric carcinoma may be linked to environmental factors, whereas the diffuse type may have a genetic basis (7). Previous works have reported that the high incidence of blood group A and family history of gastric carcinoma in diffuse-type gastric carcinoma would suggest a genetic origin of this tumor type (19, 20). The aim of this study was to investigate the relationship between H. pylori infection and gastric neoplastic diseases (carcinoma and adenoma) and histological gastritis, and also the relationship between gastric neoplastic diseases and potential risk factors (alcohol, smoking, coffee, blood group A, and family history of gastric carcinoma). MATERIALS AND METHODS Subjects The study included consecutive 141 patients with histologically confirmed gastric carcinoma or adenoma who had undergone endoscopy at Hiroshima University Hospital between 1991 and 1996. There were 105 cases of carcinoma and 36 cases of adenoma. The study also included 105 sexand age-matched control subjects with no localized lesions in the upper GI tract who were selected from the population of patients who visited the hospital because of dyspepsia or for a mass survey for gastric carcinoma. There were 45 patients with dyspepsia and 60 subjects for a mass survey; no significant difference was found about H. pylori infection between these two groups. The case and control patients were interviewed about consumption of alcohol and coffee, history of smoking, blood groups, and family history of gastric carcinoma.
INTRODUCTION Helicobacter pylori (H. pylori) has been identified as a major cause of chronic gastritis and peptic ulcer disease Received Jan. 29, 1998; accepted April 1, 1998. 1271
1272
KOMOTO et al.
Serologic detection of antibodies to H. pylori Serum samples were independently tested for the presence of antibodies against H. pylori using the highly sensitive and specific enzyme-linked immunosorbent assay (Pylori stat kit, Whittaker Bioproducts, Walkersville, MD) described previously (21–28). In brief, serum specimens from patients were diluted to 1:20 for use in the kit, and the IgG concentration of the sera was determined according to the manufacturer’s instructions. The serum specimens were coded so that the laboratory technician could not distinguish patients from control subjects. A ratio $1.00 was considered positive and a ratio ,1.00 was considered negative, calibrated according to the kit. The test has a reported sensitivity of 96% and specificity of 94% (28). Identification of H. pylori and histological assessment Gastric biopsy specimens were obtained endoscopically under direct vision in all subjects. Two specimens were taken from the lesser curvature of the antrum and two specimens were obtained from the anterior wall and the posterior wall of the middle body. Biopsy specimens were fixed overnight in buffer formalin, enbedded in paraffin, cut to 4-mm thickness, and stained with hematoxylin and eosin and Giemsa. Identification of H. pylori was performed using the Giemsa stain. In accordance with the Sydney system, the degree of mucosal inflammation, atrophy, and IM were classified into four grades as follows: 0 5 none; 1 5 mild; 2 5 moderate; and 3 5 severe (29, 30). Gastric carcinoma was classified into intestinal and diffuse types according to Lauren’s criteria (31). Cardia was defined as the area in the stomach within 20-mm distance from the esophagogastric junction. Gastric adenoma was diagnosed according to the World Health Organization (WHO) criteria (32). Gastric adenoma is typically a sessile, broad-based lesion that is only slightly elevated macroscopically. Histologically, this is composed of an upper compartment of dysplastic crypts and a lower compartment of normal or cystic pyloric glands. The dysplastic crypts are usually straight and unbranched and are lined by crowded, darkly staining cells with elongated, pseudostratified nuclei and small amounts of apical mucin. Statistical analysis Results are given as mean 6 SD. Statistical analyses were performed using the Mann-Whitney U test. A p value ,0.05 was considered significant. Within the gastric carcinoma group, x2 tests were used to evaluate the association between positive antibodies to H. pylori and either carcinoma stage (early versus advanced), site (cardia versus noncardia), macroscopic type (polypoid, ulceration versus others), or histological type of the carcinoma (intestinal versus diffuse). Odds ratios (OR) obtained from 2 3 2 tables. The logistic regression models was used in multivariate analysis of the relationship between gastric neoplasia risk and H. pylori and potential risk factors such as alcohol, smoking, coffee, blood group A, and family history of gastric carci-
AJG – Vol. 93, No. 8, 1998 TABLE 1 Comparison of the Measured Parameters Among Patients With Gastric Carcinoma, Adenoma, and Control Subjects Diseases
Carcinoma
Adenoma
Control
No. M/F Mean age Hp-Antibody positive rate (%) Odds ratio 95% CI Hp-Giemsa stain positive rate (%)
105 82/23 64.9 6 1.2 93.3*
36 28/8 65.5 6 1.8 94.4*
105 82/23 62.4 6 1.1 71.4
5.60 2.33–13.45 96.2*†
6.80 1.54–30.10 77.8
1.00 73.3
* p , 0.05 as compared with control subjects. † p , 0.05 as compared with adenoma.
noma. ORs with confidence intervals (CI) were computed. The OR is an estimation of relative risk. The confidence intervals (CIs) were set at 0.05. RESULTS There were 105 patients with gastric carcinoma, 36 patients with adenoma, and 105 control subjects, of whom 82 (78%), 28 (78%), and 82 (78%), respectively, were men. The mean age of the patients with carcinoma and adenoma and the control subjects was 65 yr, 66 yr, and 62 yr, respectively. H. pylori seroprevalence The prevalence of H. pylori seropositivity was significantly higher in the patients with gastric carcinoma (93%) and adenoma (94%) than in the control subjects (71%), giving ORs of 5.60 (carcinoma) and 6.80 (adenoma), respectively (Table 1). H. pylori was identified by Giemsa stain in 96% of the patients with gastric carcinoma, 78% of those with gastric adenoma, and 73% of the control subjects. H. pylori infection in gastric carcinoma The prevalence of H. pylori seropositivity in the gastric carcinoma patients was 98% in the men and 78% in the women. Patients ,60 yr old had an OR of 19.38; patients 60 – 69 yr had an OR of 4.49; and patients .69 yr had an OR of 2.14 (Table 2). The decreasing OR with age was attributed to increasing prevalence of H. pylori seropositivity with age in the control subjects and a decreasing prevalence of seropositivity with age among the patients. A total of 94% of the patients with early gastric carcinoma showed positive H. pylori antibody (OR, 6.60; 95% CI, 2.23–19.69), whereas 91% of the patients with advanced gastric carcinoma had serologic evidence of H. pylori infection (OR, 4.27; 95% CI, 1.21–15.03). The OR of H. pylori seropositivity for the noncardia (OR, 5.67; 95% CI, 2.25–14.44) gastric carcinoma patients was significantly higher than the odds for control subjects. In an analysis stratified with respect to tumor location, only the OR for a noncardia location was found to be significantly increased
AJG – August 1998
H. PYLORI AND GASTRIC NEOPLASIA
TABLE 2 Risk of H. pylori Infection Among Patients With Gastric Carcinoma in Relation to Control Subjects
Patient Group Analyzed Sex Male (n 5 82) Female (n 5 23) Age #59 (n 5 32) 60 # 69 (n 5 34) $70 (n 5 39) Stage Early (n 5 70) Advanced (n 5 35) Location Noncardia (n 5 91) Cardia (n 5 14) Macroscopic type Polypoid (n 5 26) Ulceration (n 5 58) Others (n 5 21) Histologic type Intestinal (n 5 84) Diffuse (n 5 21) Smoking Noncardia (n 5 91) Cardia (n 5 14)
Positive H. pylori Serology (%)
OR for H. pylori Infection
95% CI
97.6 78.3
12.90 2.77
2.92–57.40 0.74–9.78
96.9 91.2 92.3
19.38 4.49 2.14
2.36–157.59 1.11–18.17 0.47–9.68
94.3 91.4
6.60 4.27
2.23–19.69 1.21–15.03
93.4 92.9
5.67 5.20
2.25–14.44 0.65–41.68
96.2 93.1 90.5
10.00 5.40 3.80
1.30–76.71 1.80–16.28 0.84–17.46
92.9 95.2
5.20 8.00
2.05–13.20 1.02–62.80
93.4 92.9
2.00 4.32
1.04–3.85 0.90–20.86
(OR, 5.67; 95% CI, 2.25–14.44). There was no difference in the prevalence of H. pylori infection in gastric carcinoma according to the histologic type: 93% (78 of 84) in intestinal-type gastric carcinoma and 95% (20 of 21) in diffusetype gastric carcinoma (Table 2). There was also no difference in the prevalence of H. pylori seropositivity according to the stage and macroscopic type of cancer. Antibody levels in both the intestinal- and diffuse-type gastric carcinoma were higher than in the control subjects and they were significantly higher in patients with the intestinal-type gastric carcinoma than in those with adenoma (Table 4). Corpus mucosal examination Among patients with gastric carcinoma, 14 (13.3%) had mild atrophic gastritis, 26 (24.8%) had moderate atrophic gastritis, and 65 (61.9%) had severe atrophic gastritis. In patients with adenoma, five (13.9%) had moderate atrophic gastritis and 31 (86.1%) had severe atrophic gastritis. Among the control subjects, 24 (22.9%) had normal mucosa, seven (6.6%) had superficial gastritis, 36 (34.3%) had mild atrophic gastritis, 11 (10.5%) had moderate atrophic gastritis, and 27 (25.7%) had severe atrophic gastritis. Among seronegative subjects with gastric carcinoma (n 5 7), two (1.9%) had mild atrophic gastritis, one (1.0%) had moderate atrophic gastritis, and four (3.8%) had severe atrophic gastritis. In adenoma patients (n 5 2), 1 (2.8%) had moderate atrophic gastritis and one (2.8%) had severe atrophic gastritis. Among seronegative control subjects (n 5 30), 22 (21.0%) had normal mucosa, four (3.8%) had mild
1273
TABLE 3 Comparison of Histological Corpus Gastritis Among Patients With Gastric Carcinoma, Adenoma, and Control Subjects Mucosal Status
Carcinoma (n 5 105)
Adenoma (n 5 36)
Control (n 5 105)
Normal Superficial gastritis Mild atrophic gastritis Moderate atrophic gastritis Severe atrophic gastritis
0 (0.0%) 0 (0.0%) 14 (13.3%) 26 (24.8%) 65 (61.9%)
0 (0.0%) 0 (0.0%) 0 (0.0%) 5 (13.9%) 31 (86.1%)
24 (22.9%) 7 (6.6%) 36 (34.3%) 11 (10.5%) 27 (25.7%)
atrophic gastritis, three (2.9%) had moderate atrophic gastritis, and one (1.0%) had severe atrophic gastritis (Table 3). No significant differences in the gastritis score were found among intestinal-type gastric carcinoma, diffuse-type gastric carcinoma, adenoma, and H. pylori-positive control subjects in either pyloric or fundic mucosa. Atrophy and IM scores in both pyloric and fundic mucosa were significantly higher in patients with intestinal-type gastric carcinoma and adenoma than in those with diffuse-type gastric carcinoma and H. pylori-positive control subjects. Both atrophy and IM scores were similar between the patients with diffuse-type gastric carcinoma and H. pylori-positive control subjects (Table 4). With respect to the baseline characteristics of the patients, only smoking was related to intestinal-type gastric carcinoma and adenoma (OR, 6.56; 95% CI, 2.86 –15.03, OR, 7.50; 95% CI, 2.05–27.51) in male patients (Table 5). No significant difference was found between patients with diffuse-type gastric carcinoma and control subjects except for H. pylori infection (OR, 8.00; 95% CI, 1.02– 62.80). When the tumors were divided into noncardia and cardia lesions, only noncardia carcinoma was related to smoking (OR, 2.00; 95% CI, 1.04 –3.85). Gastric adenoma was associated with H. pylori infection (OR, 6.80; 95% CI, 1.54 –30.10) and smoking and male gender (OR, 7.50; 95% CI, 2.05–27.51), but not with smoking and female gender (OR, 2.00; 95% CI, 0.11–37.83), alcohol, coffee, blood group A, or a family history of gastric cancer (Table 6). DISCUSSION The main finding in this study was the significantly increased prevalence of H. pylori seropositivity among patients with gastric carcinoma and adenoma compared with age- and sex-matched controls. The association was linked to noncardia tumors and was significant in male patients and in younger cases ,69 years of age. In addition, atrophic gastritis and IM were closely related to the intestinal-type gastric carcinoma and adenoma, after controlling for H. pylori infection. Smoking was also found to be associated with the intestinal-type gastric carcinoma and adenoma in men but not in women. The finding of a high incidence of chronic gastritis in patients with gastric carcinoma and adenoma supports pre-
1274
KOMOTO et al.
AJG – Vol. 93, No. 8, 1998
TABLE 4 Comparison of the Measured Parameters Among Patients With Gastric Carcinoma, Adenoma, and Control Subjects Carcinoma
Diseases Optical density Gastritis score Pyloric mucosa Fundic mucosa Atrophy score Pyloric mucosa Fundic mucosa Intestinal metaplasia score Pyloric mucosa Fundic mucosa
Intestinal (n 5 77)
Diffuse (n 5 20)
Adenoma (n 5 34)
Control (n 5 75)
2.36 6 0.14‡
2.36 6 0.14‡
1.81 6 0.14‡
2.26 6 0.09
1.51 6 0.07 1.82 6 0.07
1.85 6 0.13 2.00 6 0.18
1.38 6 0.09 1.65 6 0.12
1.55 6 0.08 1.81 6 0.09
2.54 6 0.08*† 2.23 6 0.09*†
1.80 6 0.20 1.75 6 0.22
2.76 6 0.09*† 2.74 6 0.08*†
1.59 6 0.11 1.51 6 0.12
1.62 6 0.07* 1.36 6 0.09*
1.15 6 0.17 1.00 6 0.19
1.76 6 0.09*† 1.68 6 0.10*†
1.01 6 0.10 0.87 6 0.10
* p , 0.05 as compared with control subjects. † p , 0.05 as compared with diffuse-type carcinoma. ‡ p , 0.05 as compared with adenoma. TABLE 5 Odds Ratios of Gastric Carcinoma in a Multiple Logistic Regression Model Intestinal
H. pylori Alcohol Smoking Coffee Blood group A Family history of gastric carcinoma
Diffuse
Odds Ratio
95% CI
Odds Ratio
95% CI
5.20 0.92 3.05 0.50 1.04 1.41
2.05–13.20 0.49–1.73 1.58–5.93 0.24–1.01 0.59–1.85 0.64–3.00
8.00 0.71 2.18 1.98 0.63 0.83
1.02–62.80 0.26–1.94 0.77–6.16 0.71–5.46 0.24–1.64 0.17–4.01
TABLE 6 Odds Ratios of Gastric Adenoma in a Multiple Logistic Regression Model
H. pylori Alcohol Smoking Coffee Blood group A Family history of gastric carcinoma
Odds Ratio
95% CI
6.80 1.27 3.05 0.99 0.91 2.04
1.54–30.10 0.55–2.91 0.70–5.30 0.42–2.26 0.45–1.84 0.86–1.02
vious studies (33, 34). All of the gastric carcinomas and adenomas were found in the setting of atrophic gastritis. Similarly, Sipponen et al. (33) have reported a study of 54 patients with gastric carcinoma, among whom 38 (70%) had H. pylori infection. Only five patients (16%) had normal mucosa, but had no evidence of H. pylori infection by serology or histology. Craanen et al. (35) have reported that atrophic mucosal changes were present in 90.3% (28 of 31) of patients with intestinal-type early gastric carcinoma and in 63.6% (14 of 22) of patients with diffuse-type early gastric carcinoma. H. pylori infection was found in the antral mucosa of 61.3% (18 of 31) of patients with intestinal-type early gastric carcinoma and in 54.5% (12 of 22) of
patients with diffuse-type early gastric carcinoma. In western countries, the prevalence of H. pylori infection is 70 – 80% in gastric carcinoma patients, but 10 –20% of gastric carcinoma cases develop in an apparently H. pylori-negative stomach. In our study, seven patients with gastric carcinoma and two patients with gastric adenoma had negative H. pylori serology, but all nine had moderate to severe atrophic gastritis by histology. It is now accepted that prolonged atrophy causes the spontaneous disappearance of H. pylori from the gastric mucosa and a fall in the antibody titer (26). As atrophy and IM become severe, H. pylori colonization is reduced because of hypoacidity and overgrowth with other bacteria (36). Positive serology, however, often occurs in such patients (37– 40). In our study, seven patients with gastric carcinoma and two patients with adenoma were seronegative, but H. pylori was detected by Giemsa in stain, and all had moderate to severe atrophic gastritis. It seems that extended atrophy may decrease colonisation by H. pylori and induce a fall antibody levels (26). Therefore, even our apparently seronegative patients with gastric carcinoma and adenoma probably had previous H. pylori infection. It is well known that the prevalence of both H. pylori and gastric carcinoma is higher in Japan than in western countries. The prevalence of H. pylori infection in this study was also higher than that reported in western countries. Asaka et al. (41), reported that the prevalence of H. pylori infection was significantly higher in patients with gastric carcinoma (88.2%) than in control patients (74.6%), but the frequency of high titer IgG antibodies was lower in patients with advanced gastric carcinoma (84.7%) than early gastric carcinoma (93.0%). Our result are similar to the findings reported by Asaka et al. Kato et al. (17), on the other hand, have indicated that H. pylori IgG antibody was positive in only 51 of 72 in gastric carcinoma cases (70.8%) and similarly in 98 of 143 control subjects (68.5%). This may be explained by selection bias or measurement bias. Parsonett et al. (7) have reported that the increased risk of
AJG – August 1998 gastric carcinoma was confined largely to women (OR, 18.0). This result is not consist with ours, but others have failed to confirm a higher risk in women (16, 42). From this and previous studies (7, 8, 15, 40), it is evident that tumors located in the gastric cardia are less associated with H. pylori infection compared with noncardia cancer. Carcinoma at these two sites is likely to be due to different etiologies. Our study indicates that H. pylori infection may play a major role in the etiology of noncardia cancer, whereas smoking is associated with noncardia cancer. Earlier studies have indicated that there are important clinical, histological, and demographic differences between intestinal and diffuse-type gastric cancer. It is generally accepted that intestinal-type gastric carcinoma is related to atrophic gastritis (8). The prevalence of H. pylori infection was significantly higher in patients with carcinoma or adenoma than in control subjects in the present study (p , 0.05). Atrophic gastritis and IM are important factors, along with H. pylori infection, in the pathogenesis of intestinaltype gastric carcinoma. Although we carefully examined the pathology to search for other risk factors for gastric carcinoma in addition to H. pylori infection, only atrophic gastritis and IM were related to intestinal-type gastric carcinoma. Diffuse type gastric carcinoma was associated with H. pylori infection, but there were no significant differences compared with H. pylori-positive control subjects in the degree of mucosal atrophy, IM, and inflammation. Previous work has reported that the pathogenesis of intestinal-type gastric carcinoma may be linked to environmental factors, whereas the diffuse type may have a genetic basis (7). The high incidence of blood group A and family history of gastric carcinoma in diffuse-type gastric carcinoma would suggest a genetic origin of this tumor type (19, 20). However, we could not detect that any factor other than H. pylori infection was associated with diffuse-type gastric carcinoma. Gastric adenoma is common in Japan, and is thought to be a precancerous lesion. It is recognized that intestinal-type gastric carcinoma may develop from gastric adenoma when the lesion is .2.0 cm in diameter (34). The prevalence of seropositivity in adenoma was similar to that of gastric carcinoma and significantly higher than in control subjects, but identification of H. pylori by Giemsa stain was significantly lower in gastric adenoma than in gastric carcinoma. Atrophy and IM scores were significantly higher in gastric adenoma than in H. pylori-positive control subjects and diffuse-type gastric carcinoma, and were similar between intestinal-type gastric carcinoma and adenoma. Gastric adenoma may develop from atrophic mucosa and may have a pathogenesis similar to intestinal-type gastric carcinoma. In conclusion, our results confirm the association between H. pylori infection and gastric carcinoma. We have also identified that H. pylori is a risk factor for gastric adenoma. Although intestinal-type gastric carcinoma and adenoma appear to be closely related to atrophic gastritis and IM,
H. PYLORI AND GASTRIC NEOPLASIA
1275
diffuse-type carcinoma is not. Only intestinal-type gastric carcinoma is related to smoking. Reprint requests and correspondence: Dr. Ken Haruma, First Department of Internal Medicine, Hiroshima University School of Medicine, 1-2-3- Kasumi, Minami-ku, Hiroshima, 734, Japan.
REFERENCES 1. Warren JR, Marshall BJ. Unidentified curved bacilli on gastric epithelium in active gastritis. Lancet 1983;1273–5. 2. Marshall BJ, Warren JR. Unidentified curved bacilli in the stomach of patients with gastritis and peptic ulceration. Lancet 1984;1311–5. 3. Graham DY. Helicobacter pylori: Its epidemiology and its role in duodenal ulcer disease. J Gastroenterol Hepatol 1991;6:105–13. 4. Graham DY, Klein PD, Evans DG, et al. Helicobacter pylori: Epidemiology, relationship to gastric cancer and the role of infants in transmission. Eur J Gastroenterol Hepatol 1992;4(suppl 1):S1– 6. 5. Marshall BJ. Campylobacter pylori: Its link to gastritis and peptic ulcer disease. Rev Infect Dis 1990;12(suppl 1):S87–93. 6. Siurala M, Sipponen P, Kekki M. Campylobacter pylori in a sample of Finish population: Relations to morphology and functions of the gastric mucosa. Gut 1988;29:909 –16. 7. Parsonnet J, Friedman GD, Wandersteen DP, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991;325: 1127–31. 8. Talley NJ, Zinsmeister AR, Weaver A, et al. Gastric carcinoma and Helicobacter pylori infection. J Natl Cancer Inst 1991;83:1734 –9. 9. Parsonett J, Vandersteen D, Goates J, et al. Helicobacter pylori infection in intestinal and diffuse type gastric carcinoma. J Natl Cancer Inst 1991;83:640 –3. 10. Loffeld RJLF, Willems I, Flendrig JA, et al. Helicobacter pylori and gastric carcinoma. Histopathology 1990;17:537– 41. 11. Forman D, Sitas F, Newell DG, et al. Geographic association of Helicobacter pylori antibody prevalence and gastric cancer mortality in rural China. Int J Cancer 1990;46:608 –11. 12. Correa P, Fox J, Fontham E, et al. Helicobacter pylori and gastric carcinoma: Serum antibody prevalence in populations with contrasting cancer risks. Cancer 1990;66:2569 –74. 13. Fox JG, Correa P, Taylor NS, et al. Campylobacter pylori associated gastritis and immune response in a population at increased risk of gastric carcinoma. Am J Gastroenterol 1989;84:775– 81. 14. The Eurogust Study Group. An international association between Helicobacter pylori infection and gastric cancer. Lancet 1993;341:1359 – 62. 15. Nomura A, Stemmermann GN, Chyou PH, et al. Helicobacter pylori infection and gastric carcinoma among Japanese American in Hawaii. N Engl J Med 1991;325:1132– 6. 16. Kuipers EJ, Gracia-Casanova M, Pena AS, et al. Helicobacter pylori serology in patients with gastric carcinoma. Scand J Gastroenterol 1993;28:433–7. 17. Kato S, Onda M, Matsukura N, et al. Genetic polymorphisms of the cancer related gene and Helicobacter pylori infection in Japanese gastric cancer patients. Cancer (suppl) 1996;77:1654 – 61. 18. Watanabe S, Arimoto H. Long term trends in cancer mortality rates and cumulative rates from 1955 to 1987 in Japan (II). Jpn J Clin Oncol 1989;19:413–25. 19. Wu MS, Sung CT, Wang HP, et al. Genetic alterations in gastric cancer: Relation to histological subtypes, tumor stage, and Helicobacter pylori infection. Gastroenterology 1997;112:1457– 65. 20. La-Vecchia C, D’Avanzo B, Negri E, et al. Attributable risks for stomach cancer in northern Italy. Int J Cancer 1995;60:748 –52. 21. Perez-Perez GI, Dworkin BM, Chodos JE, et al. Campylobacter pylori antibodies in humans. Ann Intern Med 1988;109:11–7. 22. Dooley CP, Cohen H, Fitzgibbons PL, et al. Prevalence of Helicobacter pylori infection and histologic gastritis in asymptomatic persons. N Engl J Med 1989;321:1562– 6. 23. Drumm B, Perez-Perez GI, Blaser MJ, et al. Intrafamilial clustering of Helicobacter pylori infection. N Engl J Med 1990;322:359 – 63. 24. Glassman MS, Dallal S, Berezin SH, et al. Helicobacter pylori-related gastroduodenal disease in children. Diagnostic utility of enzymelinked immunosorbent assay. Dig Dis Sci 1990;35:993–7. 25. Strauss RM, Wang TC, Kelsey PB, et al. Association of Helicobacter
1276
26. 27. 28. 29. 30. 31. 32. 33.
KOMOTO et al.
pylori infection with dyspeptic symptoms in patients undergoing gastroduodenoscopy. Am J Med 1990;89:464 –9. Karnes WE, Samloff IM Jr, Siurala M, et al. Positive serum antibody and negative tissue staining for Helicobacter pylori in subjects with atrophic body gastritis. Gastroenterology 1991;101:167–74. Talley NJ, Kost L, Haddad A, et al. Comparison of commercial serological tests for detect of Helicobacter pylori antibodies. J Clin Microbiol 1992;30:3146 –50. Talley NJ, Newell DG, Ormand JE, et al. Serodiagnosis of Helicobacter pylori: Comparison of enzyme linked immunosorbent assay. J Clin Microbiol 1991;29:1635–9. Whitehead R. The histological diagnosis of chronic gastritis in fibreoptic gastroscope biopsy specimens. J Clin Pathol 1972;25:1–11. Siurala M, Lehtola J, Ihama¨ki T. Atrophic gastritis its sequelae: Results of 19 –23 years’ follow up examinations. Scand J Gastroenterol 1974; 9:441– 6. Lauren P. The two histological main types of gastric carcinoma: Diffuse and so-called intestinal type carcinoma. Acta Path Microbiol Scand 1965;64:31– 49. Ohta K. Histological typing of gastric and esophageal tumors. International Histological Classification of Tumors, No. 18, WHO, Geneva, 1977. Sipponen P, Kosunen TU, Valle J, et al. Helicobacter pylori infection and chronic gastritis in gastric cancer. J Clin Pathol 1992;45:319 –23.
AJG – Vol. 93, No. 8, 1998 34. Todo H, Haruma K, Yoshihara M, et al. Clinical management of group III lesion (in Japanese). Gastroenterol Endosc 1990;32:1133–9. 35. Craanen ME, Blok P, Dekker W, et al. Helicobacter pylori and early gastric cancer. Gut 1994;35:1372– 4. 36. Wee A, Kang JY, Teh M. Helicobacter pylori and gastric cancer: Correlation with gastritis, intestinal metaplasia, and tumour histology. Gut 1992;33:1029 –32. 37. Newell DG, Hawtin PR, Stacey AR, et al. Estimation of prevalence of Helicobacter pylori infection in an asymptomatic elderly population comparing [14C] urea breath test and serology. J Clin Pathol 1991;44: 385–7. 38. Wyatt JI, Shallcross TM, Crabtree JE, et al. Helicobacter pylori, gastritis, and peptic ulceration in the elderly. J Clin Pathol 1992;45: 1070 – 4. 39. DeLuca VA Jr. Is pernicious anemia caused by Campylobacter pylori gastritis (letter). J Clin Gastroenterol 1989;11:584 –5. 40. Fiocca G, Villani L, Turpini F, et al. High incidence of Campylobacterlike organisms in endoscopic biopsies from patients with gastritis, with or without peptic ulcer. Digestion 1987;38:234 – 44. 41. Asaka M, Kimura T, Kato M, et al. Possible role of Helicobacter pylori infection in early gastric cancer development. Cancer 1994;73:2691– 4. 42. Hansson LE, Engstrand L, Nyren O, et al. Prevalence of Helicobacter pylori infection in subtypes of gastric cancer. Gastroenterology 1995; 109:885– 8.