Abstract
We used Helicobacter pylori sero‐positivity and mucosal atrophy as detected by the serum pepsinogen method to identify H. pylori infection‐negative gastric cancer patients with or without atrophy. One hundred and six of 748 (14.2%) primary gastric cancer patients were infection‐negative by a serum antibody detection system. Further, 121 (16.2%) of the 748 were negative for gastric mucosal atrophy by the pepsinogen method, of whom 15/748 (2.0%) were H. pylori‐negative by pepsinogen I level (>70 ng/mL) and pepsinogen I/II ratio (>3.0). Twenty‐seven of 782 (3.6%) gastric cancer patients were H. pylori‐negative by antibodies and severe atrophy as determined by pepsinogen I level (<30 ng/mL) and pepsinogen I/II ratio (<2.0). H. pylori‐negative gastric cancer patients with severe atrophy likely had a previous infection. These results indicate that the actual number of H. pylori‐negative patients is 2.0% at minimum and 10.6% (14.2% minus 3.6%) at maximum in the general Japanese population. Five of 15 (33%) cases displaying neither anti‐H. pylori antibodies nor atrophy were intestinal‐type and 10 (67%) were diffuse‐type adenocarcinomas. Thirteen surgical patients with primary gastric cancer displaying neither antibodies nor mucosal atrophy were further analyzed for pathological and phenotypic characteristics. The mucin phenotype was divided into four gastric, five gastric and intestinal, two intestinal and two null types, independent of histological classification. Intestinal phenotype elements were detected by Cdx2 immunohistochemical methods in nine of 13 (70%) cases examined. We conclude that a small fraction of gastric cancer patients displayed multifactorial carcinogenesis without H. pylori infection, indicating that gastric cancer risk still exists in the absence of H. pylori infection, at an incidence of 2.0% at minimum and 10.6% at maximum in the general Japanese population. (Cancer Sci 2007; 98: 790–794)
The prevalence of Helicobacter pylori (H. pylori) infection, which varies by ethnicity, location and race, is closely associated with the incidence of gastric cancer. One prospective study found that gastric cancer develops only in patients with H. pylori infection.( 1 , 2 , 3 ) The gastric mucosa in infected persons is marked by severe atrophic changes associated with intestinal metaplasia. These pathological changes are reported as precursory lesions of gastric cancer,( 4 ) and have raised the possibility of active prevention of the disease.( 5 , 6 ) However, because H. pylori infection is diminished in severely atrophic mucosa, detection systems for H. pylori infection‐negative status have not been fully established. Any conclusive finding of negativity is therefore problematic. Pathological analysis of gastric mucosa after gastrectomy in a series of Japanese patients found that only 2% of gastric cancers occurred in H. pylori‐negative patients.( 7 ) Other studies using multiple detection systems including pathological and serological diagnosis similarly found that only 2% of patients with early gastric cancer were H. pylori‐negative.( 8 , 9 ) The difficulty of distinguishing past H. pylori infection positivity in gastric mucosa with stronger atrophic changes is well known, even with the combined use of culture, pathological diagnosis, anti‐H. pylori antibody and the 13C‐urea breath test.( 10 , 11 ) Further, false‐negative cases independently detected by the anti‐H. pylori antibody and other methods have occasionally been erroneously categorized as H. pylori infection‐negative.( 12 , 13 ) These findings highlight the need for care in classifying infection‐negative patients.( 14 , 15 , 16 , 17 )
Human gastric carcinoma is classified into two major groups, well‐ and poorly‐differentiated cancer. These closely correspond to the intestinal and diffuse types of gastric cancer classified by Lauren, respectively.( 18 ) Both types are considered to be H. pylori infection‐related.( 4 ) Recent advances in immunohistochemistry using gastric and intestinal epithelial cell markers such as MUC5AC, MUC6, MUC2, villin and Cdx2 (caudal‐related homeobox gene) have allowed the phenotypic expression of gastric cancer cells of each histological type to be clearly classified into gastric and intestinal epithelial types,( 19 ) and Cdx2 gene expression has facilitated prognostic profiling in gastric cancer.( 20 ) Human gastric cancers in advanced cancers tend to progress to having more intestinal‐phenotype malignant cells.( 20 ) This shift in phenotypic expression has not been detailed, however, particularly with regard to H. pylori‐related status and atrophic mucosal changes in the stomach.
Here, to determine the ratio of gastric cancer patients who are negative for H. pylori infection, we focus on H. pylori sero‐positivity and mucosal atrophy detected by the serum pepsinogen method to identify H. pylori infection‐negative patients with or without atrophy. We also assessed gastric and intestinal phenotype by immunohistochemistry to investigate the histological characteristics of phenotypic expression in H. pylori infection‐negative gastric cancers without a background of mucosal atrophy.
Materials and Methods
Sydney System Scores detected by three‐point biopsy specimens. A total of 4125 patients with a benign disease or a gastric cancer that was diagnosed grossly with a gastrointestinal fiberscope and histologically with a biopsy specimen were eligible for the study. A large number of patients (1058) were excluded due to H. pylori eradication therapy or gastrectomy. We examined the correlation between the degrees of atrophy and intestinal metaplasia (IM) detected by Updated Sydney System Scores in triple‐site biopsy specimens and the degrees of atrophy detected by pepsinogen methods from 4125, this limited number of patients (Table 1) have been analysed by both PG method and Sydney method. Biopsy specimens collected using the triple‐site biopsy technique were used to determine scores for inflammation, activity, atrophy and intestinal metaplasia in accordance with the Updated Sydney System Score, and four levels of severity (from 0 to 3) were assigned by a pathologist. The triple sites were as follows: greater curvature of the lower corpus (1: antrum), greater curvature of the upper corpus (2: corpus) and lesser curvature of the lower body (3: angulus), as reported previously.( 5 ) H. pylori detection was carried out by either an anti‐H. pylori IgG antibody, or by a histological finding of biopsy specimens, as mentioned previously. The Updated Sydney System Scores by biopsies were classified as 1 (no atrophic changes [atrophy and IM scores were – or ±]), 2 (severe atrophic changes [atrophy score +++ or IM ++, +++]), and 3 mild or moderate atrophic changes (exceptions for 1 and 2).
Table 1.
Comparing with the degree of atrophy detected by pepsinogen test method and by the Updated Sydney System Scores in the angle of the stomach with Helicobacter pylori‐negative status
| Atrophy detected by the pepsinogen method | Atrophy detected by Updated Sydney System Scores | ||
|---|---|---|---|
| No atrophy ¶ | Mild and moderate atrophy †† | Severe atrophy ‡‡ | |
| No atrophy † | 34 | 1 | 4 |
| Mild and moderate atrophy ‡ | 153 | 10 | 14 |
| Severe atrophy § | 3 | 3 | 8 |
No atrophy, pepsinogen (PG) I > 70 & PG I/II > 3.0;
‡ Mild and moderate atrophy: exclude no atrophy and severe atrophy cases;
Severe atrophy, PG I ≤ 30 & PG I/II < = 2.0;
No atrophic changes (atrophy and intestinal metaplasia [IM] scores were – or ±);
†† Mild or moderate atrophic changes (exception for 4 and 6);
Severe atrophic changes (atrophy score +++ or IM ++, +++).
Patient characteristics. A total of 748 gastric cancer patients among 3088 patients with upper GI diseases undergoing endoscopic examination and simultaneous serum anti‐H. pylori antibody testing and pepsinogen assay at our department between 1993 and 2004 were enrolled. The group comprised 519 men (64.9 ± 10.8) and 263 women (62.9 ± 12.6) diagnosed at either an early or advanced stage of gastric cancer according to the Japanese Classification of Gastric Carcinoma.( 21 ) Average age was 64.2 years (±12.6 years).
All cancer cases were histologically confirmed as primary gastric cancer. Serum samples were collected before operating. Histological typing was carried out according to the Japanese Classification of Gastric Carcinoma,( 21 ) with papillary and tubular adenocarcinoma classified as well‐differentiated adenocarcinoma, and signet ring cell adenocarcinoma and mucinous cell adenocarcinoma as poorly‐differentiated adenocarcinoma.
Serum samples were obtained at our hospital (Nippon Medical School Hospital, Tokyo, Japan) between 1993 and 2004. This study was carried out in accordance with the principals embodied in the Declaration of Helsinki 1975. Informed consent was obtained from all subjects.
Enzyme‐linked immunosorbent assay for anti‐H. pylori IgG antibody. Serum obtained by centrifugation was stored at –80°C. Serum IgG antibody against H. pylori was assayed by enzyme‐linked immunosorbent assay (ELISA) (Cobas Core, Anti‐H. pylori EIA, Roche, Basel, Switzerland). Briefly, after termination of the enzyme reaction, absorbance at 450 nm was measured, and the results were expressed quantitatively as an index according to the standard curve. Seropositivity was set as a value of 8 IU/mL according to preliminary experiments conducted to optimize specificity and sensitivity, and was used to classify patients as negative or positive.
Detection of serum pepsinogen I and II levels. Pepsinogen (PG) I and II levels were assayed using a RIA kit (DAINABOT, Japan) according to a previous method.( 22 ) Detection limits with this kit were 0.1–160.0 for PG I and 0.7–100.0 ng/mL for PG II. The degree of gastric mucosal atrophy was determined from these pepsinogen levels and classified as 1, no atrophic change (PG I > 70 ng/mL and pepsinogen I/pepsinogen II (PG I/II) ratio > 3.0); 2, severe atrophy (PG I 30 ng/mL and PG I/II ratio 2.0) according to the previous reports,( 12 , 23 ) and 3, mild or moderate atrophic changes (30 ng/mL < PG I < 70 ng/mL or 2.0 < PG I/II ratio < 3.0), as shown in Table 1.
Immunohistochemistry. Expression of MUC5AC, MUC6, MUC2, and villin in carcinoma cells was examined by immunohistochemistry as previously described.( 19 , 20 ) Briefly, 4‐µm consecutive tumor sections were deparaffinized and hydrated through a graded series of alcohol. After inhibition of endogenous peroxidase activity by immersion in a 3% H2O2/methanol solution, antigen retrieval was carried out with 10 mmol/L citrate buffer (pH 6.0) in a microwave oven for 10 min at 120°C. The sections were then incubated with the primary antibodies. After thorough washing in phosphate‐buffered solution (PBS), they were incubated with biotinylated secondary antibody, followed by avidin‐biotinylated horseradish peroxidase complex (Vectastain Elite ABC kit, Vector Laboratories 30, Burlingame, CA, USA). Finally, immune complexes were visualized by incubation with 0.01% H2O2 and 0.05% 3,3′‐diaminobenzidine tetrachloride (DAB). Nuclear counterstaining was accomplished with Mayer's hematoxylin. Expression of Cdx2 using anti‐Cdx2 monoclonal antibody (BioGenex, CA, USA) was carried out using the same immunohistochemical approach.( 24 , 25 ) Phenotype was detected by markers for MUC5AC, MUC6, MUC2, villin and Cdx2, with 0% field positivity for each degree of immunohistochemical staining set as –, 1–9% as ±, 10–49% as +, 50–79% as ++, and 80–100% as +++, respectively.
Statistical analysis. Associations among location and pathological phenotype were analyzed by standard methods using the χ2 test, Fisher's exact test and Pearson's correlation coefficiency test. All analyses were carried out using the SPSS statistical analysis program v. 14.0 (SPSS, Tokyo, Japan).
Results
Correlation for the degree of atrophy detected by Updated Sydney System Scores or by the pepsinogen method. Among the 1158 patients, there were 828 H. pylori‐positive patients and 230 negative patients. Correlation coefficiencies between the degree of atrophy detected by the Updated Sydney System Scores and the degrees of atrophy detected by pepsinogen methods were 0.151 in 1, 0.361 in 2 and 0.346 in 3, with statistical significances (P < 0.001) in all 1158 cases. Cases with H. pylori‐negative status detected by the biopsy at the point of angle in the stomach are shown in Table 1. In both the χ2 test and Pearson's test, there were statistically significantly correlations in the degree of atrophy detected by the updated Sydney System scores and the degrees of atrophy detected by pepsinogen methods (r = 0.243, P < 0.001).
Helicobacter pylori sero‐positivity and mucosal atrophy status in gastric cancer. Helicobacter pylori positivity among the 748 gastric cancer patients as determined by serum detection is summarized in Table 2. Among them, atrophy using the pepsinogen method was positive in 642 (85.8%), while 106 (14.2%) patients were anti‐H. pylori antibody‐negative. The degree of atrophy was classified into three categories by serum pepsinogen level, (i.e. no atrophy, severe atrophy and mild‐moderate atrophy).( 23 , 24 ) There were no statistical differences in the population of H. pylori‐positive and ‐negative cases among these categories.
Table 2.
Helicobacter pylori status and degree of atrophy in 782 patients with gastric cancer. Degree of atrophy was divided into three categories according to the results of the pepsinogen (PG) method
| H. pylori‐positive | H. pylori‐negative | Total | H. pylori‐negative as a percentage of all cases (748 cases) | |
|---|---|---|---|---|
| No atrophy † | 106 (87.6%) | 15 (12.4%) | 121 | 2.0% |
| Mild and moderate atrophy ‡ | 377 (85.5%) | 64 (14.5%) | 441 | 8.6% |
| Severe atrophy § | 159 (85.5%) | 27 (14.5%) | 186 | 3.6% |
| Total | 642 (85.8%) | 106 (14.2%) | 748 | 14.2% |
No atrophy (–), PG I > 70 & PG I/II > 3.0;
‡ Mild and moderate atrophy, exclude no atrophy and severe atrophy cases;
Severe atrophy, PG I ≤ 30 & PG I/II ≤ 2.0.
Fifteen of the 106 patients (12.4%) without atrophy as detected by the pepsinogen method (PG I > 70 and PG I/II ratio > 3.0) were H. pylori‐negative. As the overall rate of gastric cancer without atrophy was 16.1% (121/748), these findings show that only 2.0% of the overall population (15/748) were both H. pylori‐negative and without atrophy in gastric mucosa.
Severe mucosal atrophy was defined in accordance with previous reports( 16 ) as a serum pepsinogen level of 30 and a PGI/II ratio of 2.0. This population comprised 24.9% of all gastric cancer cases (186/748), of which 14.5% (27/186) were H. pylori‐negative. Given their severe mucosal atrophy, these patients were considered to have undergone spontaneous elimination of H. pylori infection,( 26 ) giving an incidence of gastric cancer cases with severe atrophy and a H. pylori‐negative status of 3.6% (27/748) (Table 2).
Patients having pepsinogen method values between these two conditions (i.e. between no atrophy [PG1 > 70 plus PGI/II ratio > 3.0] and severe atrophy [PG 30 plus a PGI/II ratio ≤ 2.0]) were defined in this study as having a mild or moderate degree of atrophy. A total of 441 gastric cancer patients, or almost 60% (59.0%; 441/748), were included in this category (Table 2). H. pylori positivity in this group was 377/441 (85.5%), a ratio similar to that of overall positivity in all 748 patients. A total of 64 of all 748 patients were negative for H. pylori (8.6%).
Thus, taken together with the finding that 2.0% of the overall population (15/748) was both H. pylori‐negative and without atrophy, these results indicate that the actual ratio of H. pylori infection‐negative gastric cancer among Japanese patients is 2.0% at minimum and 10.6% (14.2% minus 3.6%) at maximum.
Clinical and pathological characteristics of H. pylori infection‐negative gastric cancer. Gender differences, type of pathology and gastric cancer staging in the H. pylori infection‐negative gastric cancer patients are summarized in Table 3. There were 66 men and 40 women, with an average age of 64.4 ± 12.5 and 63.8 ± 12.8 years, respectively. By cancer type, intestinal and diffuse cancers accounted for 46% and 54% of cases, respectively, and 38% were early stage cancers.
Table 3.
Clinical and pathological characteristics of Helicobacter pylori infection‐negative gastric cancer
| Mucosal atrophy by pepsinogen (PG) method | n | Sex male/female | Pathology % Well/poor | Stage % Early/advanced |
|---|---|---|---|---|
| No Atrophy † | 15 | 10/5 | 33/67 | 37/63 |
| Mild and moderate atrophy ‡ | 64 | 40/24 | 53/47 | 38/62 |
| Severe Atrophy § | 27 | 16/11 | 44/56 | 40/60 |
| Total | 106 | 66/40 | 46/54 | 38/62 |
No atrophy (–), PG I > 70 & PG I/II > 3.0;
‡ Mild and moderate atrophy: exclude no atrophy and severe atrophy cases;
Severe atrophy, PG I ≤ 30 & PG I/II ≤ 2.0.
Clinical characteristics of the 15 H. pylori‐negative subjects without gastric atrophy are summarized in Table 4. Ten patients were men and five were women. Five cases were well‐differentiated adenocarcinoma (33%)( 21 ) and 10 were poorly‐differentiated (six signet‐ring cell carcinoma, two poorly‐differentiated carcinoma and two mucinous adnenocarcinoma). By location, six cases were in the upper third (U), five were in the middle third (M) and two were in the lower third (L). One case was located in the upper and middle third, with the upper location dominant.
Table 4.
Clinical characteristics and serum pepsinogen (PG) levels of 15 Helicobacter pylori‐ and atrophy‐negative gastric cancer cases
| Case | Age/sex | Pathology and TNM † | Stage † | Location ‡ | Surgery | PG I (ng/mL) | PG II (ng/mL) | PG I/II ratio | H. pylori IgG |
|---|---|---|---|---|---|---|---|---|---|
| Case 1 | 56/M | T3N1M0 | II | L | Total gastrectomy | 105 | 32.8 | 3.2 | – |
| Case 2 | 64/F | T1N0M0 | I a | U | Total gastrectomy | 98.6 | 9.6 | 10.3 | _ |
| Case 3 | 34/F | T2N0M0 | I b | M | Total gastrectomy | 93.4 | 25.6 | 3.6 | _ |
| Case 4 | 59/F | T2N0M0 | I b | M | Total gastrectomy | 80.4 | 18.8 | 4.3 | _ |
| Case 5 | 63/F | T4NxM1 | IV | MU | Without operation | 88.1 | 11.6 | 7.6 | _ |
| Case 6 | 67/M | T2N3M1 | IV | M | Total gastrectomy | 136.3 | 35.7 | 3.8 | _ |
| Case 7 | 59/M | T1N0M0 | I a | L | Partial gastrectomy | 303.3 | 20.3 | 14.9 | _ |
| Case 8 | 80/M | Unclear | – | – | Without operation | 73.7 | 10.5 | 7 | _ |
| Case 9 | 80/M | T1N0M0 | I a | U | Proximal gastrectomy | 115.8 | 17 | 6.8 | _ |
| Case 10 | 72/M | T2N0M0 | I b | U | Proximal gastrectomy | 103 | 20.5 | 5 | _ |
| Case 11 | 77/M | T2N2M0 | III a | U | Total gastrectomy | 122.1 | 17.6 | 6.9 | _ |
| Case 12 | 46/M | T3N3M0 | IV | U | Total gastrectomy | 85.2 | 19.4 | 4.4 | _ |
| Case 13 | 70/M | T1N0M0 | I a | U | Proximal gastrectomy | 103.9 | 29.2 | 3.6 | _ |
| Case 14 | 67/M | T2N0M0 | I b | M | Total gastrectomy | 107 | 15.6 | 6.9 | _ |
| Case 15 | 59/F | T4N2M0 | IV | M | Total gastrectomy | 80.5 | 26.3 | 3.1 | – |
TNM classifications are summarized according to the Japanese Classification of Gastric Carcinoma (http://www.jgca.jp/PDFfiles/JCGC‐2E.PDF).
‡ Location, described as U (upper third), M (middle third) and L (lower third) and pathology.
Expression of Cdx2 and gastric/intestinal phenotype in H. pylori‐negative gastric cancer. Thirteen of 15 patients who were H. pylori‐negative, with neither anti‐H. pylori antibodies nor atrophy in the gastric mucosa, were further analyzed for mucin and phenotypic characteristics( 19 , 20 ) (Table 5). Five of the 15 (33%) had intestinal‐type adenocarcinomas and 10 (67%) had the diffuse type. The mucin phenotype was divided into four gastric (G), five gastric and intestinal (GI), two intestinal (I) and two null (N) types, independent of histological classification. Intestinal phenotype elements as detected by Cdx2 immunohistochemical methods were seen in nine of 13 (70%) cases examined.
Table 5.
Pathological and phenotypic characteristics in 15 Helicobacter pylori‐ and atrophy‐negative patients
| Pathology † | T factor ‡ | MUC5AC | MUC6 | MUC2 | Villin | Cdx2 | Phenotype in parts § | |
|---|---|---|---|---|---|---|---|---|
| Case 1 | sig | se | +++ | ++ | ± | – | ++ | G type |
| Case 2 | por | sm | ++ | + | + | – | + | GI type |
| Case 3 | sig | mp | ++ | ± | – | – | – | G type |
| Case 4 | sig | si | + | – | – | – | ± | G type |
| Case 5 | por | OP (–) | ||||||
| Case 6 | tub2 | ss | ± | ± | – | ± | +++ | N type |
| Case 7 | tub1 | m | – | – | – | – | – | N type |
| Case 8 | tub2 | OP (–) | ||||||
| Case 9 | sig | sm | ++ | – | ++ | ± | +++ | GI type |
| Case 10 | muc | ss | ++ | + | ++ | – | ++ | GI type |
| Case 11 | sig | sm | ++ | + | + | – | ++ | GI type |
| Case 12 | muc | sm | ++ | ++ | ++ | – | + | I type |
| Case 13 | tub2 | m | ++ | ++ | – | ++ | +++ | GI type |
| Case 14 | tub2 | ss | ± | – | + | + | +++ | I type |
| Case 15 | sig | si | + | ± | ± | – | – | G type |
tub1 and tub2 indicate well differentiated‐type and por, muc and sig indicate poorly differentiated‐type gastric cancer according to the Japanese Classification of Gastric Carcinoma.
Depth of invasion indicate mucosa (m), submucosa (sm), muscularis propria (mp), subserosa (ss), serosa‐exposed (se), si (serosa‐infiltrating) and OP (–) indicating no operation.
+Cases with more than 10% of the section area consisting of at least one gastric or intestinal epithelial cell phenotype were classified as gastric (G type) or intestinal (I type) phenotype cancer, respectively. Other cases were classified as either the gastric or intestinal mixed phenotype (GI type), while those with no expression phenotype were grouped as unclassified (N type).
Discussion
This study shows that a small fraction of gastric cancer patients have multifactorial carcinogenesis without H. pylori infection. The risk of gastric cancer without H. pylori infection in Japanese hospital patients is between 2.0% at minimum and 10.6% at maximum.
Helicobacter pylori‐infected persons show severe atrophic changes in the gastric mucosa associated with intestinal metaplasia, and these characteristics in turn lead to gastrocarcinogenesis.( 24 ) In vivo investigation in animal models suggests that H. pylori infection has a promoter effect on tumor development and progression, while epidemiological and experimental evidence indicates that excessive salt intake, which promotes cell replication, increases gastric cancer risk.
Although epidemiological evidence has confirmed the definitive role of H. pylori infection in stomach carcinogenesis, the difficulty of diagnosing a past infection by serum and other detection methods complicates the establishment of clear cause in H. pylori‐negative gastric cancer. A past infection is difficult to identify owing to the spontaneous elimination of H. pylori, which may occur when the mucosa changes from atrophic to metaplasic, given that H. pylori cannot survive in the metaplastic mucosa. For this reason, we used both serum anti‐H. pylori antibody to detect H. pylori infection and serum pepsinogen markers to check mucosal atrophy, in accordance with previously identified associations between serum pepsinogen levels and the degree of atrophy.( 12 )
According to our results as shown in Table 1, the degrees of no, mild and moderate or severe atrophy determined by pepsinogen methods were significantly associated with the pathological degrees of atrophy or intestinal metaplastic changes detected by Updated Sydney System Scores. In the case of H. pylori‐negative status, almost all populations that were classified as having no or mild and moderate atrophy determined by pepsinogen methods, were classified as having no atrophy in pathological findings. There results were also consistent with the aim of our studies to find the true number of H. pylori negative‐gastric cancers excluding a H. pylori false‐negative‐gastric cancer.
Table 2 shows that 106 of 748 (14.2%) of the present gastric cancer patients were H. pylori‐negative to serum anti‐H. pylori antibodies. Twenty‐seven of the 748 (3.6%) were H. pylori‐negative but had severe atrophy. This pattern has been reported to indicate the spontaneous elimination of H. pylori. Excluding the 3.6% (27/748) of H. pylori‐negative patients with the possibility of previous infection from the 14.2% (106/748) who were H. pylori‐negative as determined by serum markers, these findings indicate that the percentage of H. pylori‐negative Japanese gastric cancer patients ranges from 2.0% (15/748) with no atrophy to a maximum of 10.6% (14.2% minus 3.6%) with a mild or moderate degree of atrophy.
We further analyzed the subpopulation of H. pylori‐negative patients without atrophy. Five of 15 (33%) cases were well‐differentiated adenocarcinoma, identical to the intestinal type according to Lauren's criterion,( 17 ) while the remaining 10 (67%) had poorly‐differentiated or signet ring cell adenocancinoma; in other words a diffuse type of gastric cancer, the dominant pathological type in this study. Among the 13 of 15 cases that could be examined for pathological and phenotypic characteristics,( 19 , 20 ) phenotypic elements showed three gastric (G), five gastric and intestinal (GI), two intestinal (I) and two null (N) types. When H. pylori‐negative gastric cancer was subclassified according to the phenotype element, the resulting classification was Lauren's histological classification, as was also the case for H. pylori‐positive gastric cancers.( 19 )
Regarding Cdx2, nine of 13 (70%) cases examined were positive by immunohistochemical methods, indicating that Cdx2 is an intestinal phenotypic element.( 20 ) Immunohistochemical analysis in advanced gastric cancer has suggested that CDX2 indicates a better prognosis in H. pylori‐positive than ‐negative cases.( 20 ) A recent prospective study found that a negative H. pylori infection status in gastric cancer patients treated with surgery was associated with a poor prognosis.( 27 ) This result appears inconsistent with our present finding that a higher positivity rate for Cdx2 in H. pylori‐negative patients may indicate a better prognosis, and warrants further investigation of this question. The intestinal phenotype of mucin elements (I and GI type) was observed in seven of 13 H. pylori‐negative gastric cancer cases, despite the absence of a background of atrophic or intestinal metaplastic mucosal change, indicating a shift from gastric to intestinal phenotypic expression as previously described by Tatematsu,( 28 ) even in H. pylori‐negative gastric cancer.
In conclusion, a small fraction of gastric cancer patients displayed multifactorial carcinogenesis without H. pylori infection. The risk of gastric cancer in the absence of H. pylori infection in Japanese hospital patients is between 2.0% and 10.6%. We consider that this range may be extrapolated to the general Japanese population.
Acknowledgment
This work was supported in part by Grants‐in‐Aid for Cancer Research (H18‐3rd‐G‐005) from the Ministry of Health, Labor and Welfare, Japan.
References
- 1. Uemura N, Okamoto S, Yamamoto S et al. Helicobacter pylori infection and the development of gastric cancer. N Engl J Med 2001; 345: 829–32. [DOI] [PubMed] [Google Scholar]
- 2. Take S, Mizuno M, Ishiki K et al. The effect of eradicating helicobacter pylori on the development of gastric cancer in patients with peptic ulcer disease. Am J Gastroenterol 2005; 100: 1037–42. [DOI] [PubMed] [Google Scholar]
- 3. Asaka M, Sugiyama T, Nobuta A et al. Atrophic gastritis and intestinal metaplasia in Japan: results of a large multicenter study. Helicobacter 2001; 6: 294–9. [DOI] [PubMed] [Google Scholar]
- 4. Correa P, Shiao YH. Phenotypic and genotypic events in gastric carcinogenesis. Cancer Res 1994; 54 (7 Suppl): S1941–3. [PubMed] [Google Scholar]
- 5. Kelley JR, Duggan JM. Gastric cancer epidemiology and risk factors. J Clin Epidemiol 2003; 56: 1–9. [DOI] [PubMed] [Google Scholar]
- 6. Matsuhisa T, Yamada N, Kato S, Matsukura N. Helicobacter pylori infection, mucosal atrophy and intestinal metaplasia in Asian populations. a comparative study in age‐, gender‐ and endoscopic diagnosis‐matched subjects. Helicobacter 2003; 8: 29–35. [DOI] [PubMed] [Google Scholar]
- 7. Enomoto H, Watanabe H, Nishikura K et al. Topographic distribution of Helicobacter pylori in the resected stomach. Eur J Gastroenterol Hepatol 1998; 10: 473–8. [DOI] [PubMed] [Google Scholar]
- 8. Uemura N. H. pylori infection and the development of gastric cancer in Japan. Nippon Rinsho 2003; 61: 25–9. Review. In Japanese. [PubMed] [Google Scholar]
- 9. Asaka M, Kato M, Kudo M et al. Relationship between Helicobacter pylori infection, atrophic gastritis and gastric carcinoma in a Japanese population. Eur J Gastroenterol Hepatol 1995; 7 (Suppl 1): S7–10. [PubMed] [Google Scholar]
- 10. Furuta T, Kaneko E, Baba et al. Percentage changes in serum pepsinogens are useful as indices of eradication of Helicobacter pylori . Am J Gastroenterol 1997; 92: 84–8. [PubMed] [Google Scholar]
- 11. Matsukura N, Tajiri T, Kato S et al. Diagnostic value of culture, histology and PCR for Helicobacter pylori in the remnant stomach after surgery. Aliment Pharmacol Ther 2004; 20 (Suppl 1): 33–8. [DOI] [PubMed] [Google Scholar]
- 12. Miki K, Ichinose M, Kawamura N et al. The significance of low serum pepsinogen levels to detect stomach cancer associated with extensive chronic gastritis in Japanese subjects. Jpn J Cancer Res 1989; 80: 111–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tabata H, Fuchigami T, Kobayashi H. Helicobacter pylori and mucosal atrophy in patients with gastric cancer: a special study regarding the methods for detecting Helicobacter pylori . Dig Dis Sci 1999; 44: 2027–34. [DOI] [PubMed] [Google Scholar]
- 14. Takahashi S. Long‐term Helicobacter pylori infection and the development of atrophic gastritis and gastric cancer in Japan. J Gastroenterol. 2002; 37 (Suppl 13): 24–7. [DOI] [PubMed] [Google Scholar]
- 15. Kiyohira K, Yoshihara M, Ito M et al. Serum pepsinogen concentration as a marker of Helicobacter pylori infection and the histologic grade of gastritis; evaluation of gastric mucosa by serum pepsinogen levels. J Gastroenterol 2003; 38: 332–8. [DOI] [PubMed] [Google Scholar]
- 16. Fukuda H, Saito D, Hayashi S. Helicobacter pylori infection, serum pepsinogen level and gastric cancer: a case‐control study in Japan. Jpn J Cancer Res 1995; 86: 64–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Yamagata H, Kiyohara Y, Aoyagi K et al. Impact of Helicobacter pylori infection on gastric cancer incidence in a general Japanese population: the Hisayama study. Arch Intern Med 2000; 160: 1962–8. [DOI] [PubMed] [Google Scholar]
- 18. Lauren P. The two histological main types of gastric carcinoma: diffuse and so‐called intestinal‐type carcinoma. An attempt at a histo‐clinical classification. Acta Pathol Microbiol Scand 1965; 64: 31–49. [DOI] [PubMed] [Google Scholar]
- 19. Mizoshita T, Tsukamoto T, Inada K et al. Immunohistochemically detectable Cdx2 is present in intestinal phenotypic elements in early gastric cancers of both differentiated and undifferentiated types, with no correlation to non‐neoplastic surrounding mucosa. Pathol Int 2004; 54: 392–400. [DOI] [PubMed] [Google Scholar]
- 20. Mizoshita T, Tsukamoto T, Nakanishi H et al. Expression of Cdx2 and the phenotype of advanced gastric cancers: relationship with prognosis. J Cancer Res Clin Oncol 2003; 129: 727–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Japanese Gastric Cancer Association . Japanese classification of gastric carcinoma, 2nd edn. Gastric Cancer 2001; 1: 10–24. [DOI] [PubMed] [Google Scholar]
- 22. Matsukura N, Onda M, Tokunaga A et al. Significant of serum markers Pepsinogen I and II for chronic atrophic gastritis, peptic ulcer and gastric cancer. J Clin Gastroenterol 1993; 17: s146–50. [DOI] [PubMed] [Google Scholar]
- 23. Miki K, Ichinose M, Shimizu A et al. Serum pepsinogens as a screening test of extensive chronic gastritis. Gastroenterol Jpn 1987; 22: 133–41. [DOI] [PubMed] [Google Scholar]
- 24. Seno H, Oshima M, Taniguchi MA et al. Cyclooxygenase 2‐ and prostaglandin E(2) receptor EP(2)‐dependent angiogenesis in Apc (Delta716) mouse intestinal polyps. Cancer Res 2002; 62: 506–11. [PubMed] [Google Scholar]
- 25. Almeida R, Silva E, Santos‐Silva F et al. Expression of intestine‐specific transcription factors, CDX1 and CDX2, in intestinal metaplasia and gastric carcinomas. J Pathol 2003; 199: 36–40. [DOI] [PubMed] [Google Scholar]
- 26. Sipponen P, Kosunen TU, Valle J, Riihela M, Seppala K. Helicobacter pylori infection and chronic gastritis in gastric cancer. J Clin Pathol 1992; 45: 319–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Meimarakis G, Winter H, Assmann I et al. Helicobacter pylori as a prognostic indicator after curative resection of gastric carcinoma: a prospective study. Lancet Oncol 2006; 7: 211–22. [DOI] [PubMed] [Google Scholar]
- 28. Tatematsu M, Hasegawa R, Ogawa K et al. Histogenesis of human stomach cancers based on assessment of differentiation. J Clin Gastroenterol 1992; 14 (Suppl 1): S1–7. [DOI] [PubMed] [Google Scholar]