| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2006. Published by Oxford University Press.
ARTICLE |
Opposing Risks of Gastric Cardia and Noncardia Gastric Adenocarcinomas Associated With Helicobacter pylori Seropositivity
Affiliations of authors: Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD (FK, SMD, CCA, DA, PRT); Department of Epidemiology, The Johns Hopkins Bloomberg School of Public Health, Baltimore, MD (FK, CJN); Departments of Medicine and Microbiology, New York University School of Medicine, New York, NY (MJB, GIPP); Department of Epidemiology and Health Promotion, National Public Health Institute, Helsinki, Finland (PP, JV)
Correspondence to: Farin Kamangar, MD, PhD, Division of Cancer Epidemiology and Genetics, National Cancer Institute, 6120 Executive Blvd., Rm. 3034, Bethesda, MD 20892-7232 (e-mail: Kamangaf{at}mail.nih.gov).
 |
ABSTRACT |
|---|
 TopNotes Abstract IntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Background: Colonization with Helicobacter pylori is a risk factor for gastric adenocarcinoma, but the magnitude of this association and its relationship to anatomic location of the cancer, duration of follow-up, age at diagnosis, histologic subtype, and H. pylori strain differences are less clear. We conducted a prospective nested case–control study of H. pylori serology to address these questions. Methods: Case and control subjects were selected from the 29 133 50- to 69-year-old males recruited into the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. At baseline, detailed demographic data and a serum sample were collected. From 1985 to 1999, 243 incident cases of gastric adenocarcinoma were diagnosed in cohort members. Serum samples from 234 case subjects (173 with noncardia gastric cancers and 61 with gastric cardia cancers) and 234 age-matched control subjects were assayed for antibodies against H. pylori whole-cell and CagA antigens. We fit conditional logistic regression models to estimate the unadjusted and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for the association of H. pylori seropositivity, defined as seropositivity to either whole-cell or CagA antigens, with noncardia gastric and gastric cardia cancers. All statistical tests were two-sided. Results: H. pylori seropositivity was strongly associated with the risk of noncardia gastric cancer (adjusted OR = 7.9, 95% CI = 3.0 to 20.9) but was inversely associated with the risk of gastric cardia cancer (adjusted OR = 0.31, 95% CI = 0.11 to 0.89). H. pylori seropositivity rates did not vary statistically significantly by length of follow-up, age at diagnosis, or histologic subtype. A calculation of rates showed that the absolute risks of noncardia gastric and cardia gastric adenocarcinomas in the H. pylori–positive participants of this cohort would be 63 and 12 per 100 000 person-years, respectively, whereas corresponding rates in H. pylori–negative participants would be 8 and 37 per 100 000 person-years, respectively. Conclusion: H. pylori is a strong risk factor for noncardia gastric cancer but is inversely associated with the risk of gastric cardia cancer. These findings bolster the hypothesis that decreasing H. pylori prevalence during the past century may have contributed to lower rates of noncardia cancer and higher rates of cardia cancer in Western countries.
|
INTRODUCTION |
|---|
|
TopNotesAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Colonization of the human stomach with Helicobacter pylori is an important risk factor for gastric adenocarcinomas (1–13), but this association varies by anatomic subsite (5,8). A combined analysis of 12 prospective studies found a strong positive association between seropositivity against H. pylori antigens at study entry and subsequent risk of noncardia gastric adenocarcinoma (8). However, this analysis found no association between anti–H. pylori seropositivity and risk of gastric cardia adenocarcinoma (8).
Although the association between H. pylori colonization and gastric noncardia adenocarcinoma is strong (8), estimates of the magnitude of the association vary, presumably because of differences in study design (8), length of follow-up (8), patient age at diagnosis (1,8,11,14), histologic subtype of cancer (6), H. pylori strain (15,16), and host characteristics (16–18). Long-term prospective studies are probably more accurate than retrospective studies in assessing the association between anti–H. pylori seropositivity and risk of gastric cancer. H. pylori does not colonize areas of stomach with malignant or premalignant changes (19), so after the development of advanced gastric disease the H. pylori population may decline, resulting in lower levels of circulating anti–H. pylori antibodies. Such reductions in the levels of circulating antibodies could limit the accuracy of retrospective case–control studies and short-term prospective studies in estimating the association between H. pylori infection and the risk of noncardia gastric cancer, whereas prospective studies with long-term follow-up and in which serum collection was done before the development of premalignant disease would provide more valid estimates of risk (8).
Several other factors could affect the risk estimates. Results of several studies suggest that H. pylori infection is more strongly associated with the risk of noncardia gastric cancer risk in younger individuals than in older individuals (1,8,14). It remains unclear whether H. pylori is a stronger risk factor for gastric cancers with intestinal histology than for gastric cancers with diffuse-type histology (8,20–22). Risk variation may also reflect differences between CagA-positive and CagA-negative H. pylori strains (12,23,24). CagA-positive strains increase the turnover of gastric epithelium by delivering CagA protein into the epithelial cells (23) and may further increase risk of noncardia gastric cancer. Most, but not all, studies have observed a higher risk of noncardia gastric cancer associated with CagA-positive strains than with CagA-negative strains (25).
Although the overall association between anti–H. pylori seropositivity and gastric cardia cancer has been reported to be null (8), there is substantial geographic variation with respect to this association; most studies of East Asian populations have shown an increased risk, whereas most studies of Western populations have shown either no association or reduced risk (26). The direction of the association by geographic region may be important because, since the 1970s, both a substantial reduction in H. pylori prevalence and a substantial increase in the incidence of gastric cardia adenocarcinoma have been observed in the Western world (27–29) and in Japan (30,31) but not in other East Asian countries (32,33). Variations in risk estimates for gastric cardia adenocarcinomas, like those for noncardia gastric adenocarcinomas, may also reflect differences in study design, the length of follow-up, age at diagnosis, histology, and H. pylori strain. However, the small number of cardia cancers in prospective studies has made it difficult to examine subgroup-stratified associations (8,26).
We conducted a long-term prospective case–control study to estimate the magnitudes of the associations between H. pylori seropositivity and the risks of gastric cardia and noncardia gastric cancers among Finnish males who participated in the Alpha-Tocopherol, Beta-Carotene Cancer Prevention (ATBC) Study, a randomized cancer prevention trial conducted in Southern Finland. We also examined the magnitude of these associations by time from serum collection to cancer diagnosis, age at diagnosis, histologic subtype, and CagA positivity.
|
SUBJECTS AND METHODS |
|---|
|
TopNotesAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Study Population
The case and control subjects included in this study were identified from among the participants in the ATBC study, a randomized, double-blind, placebo-controlled, primary prevention trial that was conducted to determine whether daily supplementation with alpha-tocopherol, beta-carotene, or both would reduce the incidence of lung or other cancers among male smokers. Between 1985 and 1988, 29 133 eligible Finnish male smokers between the ages of 50 and 69 years were recruited to the ATBC study. The trial ended in 1993, but the participants are still being followed as a cohort. Details of the study design and methods have been published (34). This study was approved by the institutional review boards of both the National Public Health Institutes in Finland and the National Institutes of Health in the United States. All participants provided written informed consent.
All case subjects had incident gastric adenocarcinoma that was diagnosed through April 30, 1999. Cancer cases were identified primarily via the Finnish Cancer Registry, which provides nearly 100% coverage of all cancer cases in Finland (35). Diagnosis of gastric cancer cases defined according to the International Classification of Diseases, 9th Revision [ICD-9; (36)] code 151 was confirmed by review of hospital records and histopathologic specimens. Cases were classified as gastric cardia cancer if they involved the esophagogastric junction and as noncardia gastric cancer if they did not.
Of the 243 case subjects diagnosed with gastric adenocarcinoma, 234 (96%) had adequate serum samples for H. pylori analysis. We used a computer program written in SAS (SAS Institute Inc, Cary, NC) that randomly selected an equal number of control subjects (N = 234) from among those ATBC study participants who remained cancer free through April 30, 1999, so that each control subject was matched with one of the cases for age (±3 months).
Demographic and Dietary Data Collection
At ATBC study entry, all participants completed a questionnaire on demographic characteristics and lifestyle and provided a fasting serum sample that was stored in aliquots at –70 °C. Diet was assessed with the use of a self-administered food frequency questionnaire that contained 276 food items (37). The food frequency questionnaire was satisfactorily completed by 27 110 participants (93%) at ATBC study entry.
Serum Assays
Serum was evaluated for immunoglobulin G antibodies against H. pylori whole-cell and CagA antigens by enzyme-linked immunosorbent assays, as previously described (38). The results were expressed as optical density ratios relative to simultaneously analyzed laboratory standards. Results were considered positive if the optical density ratio was greater than or equal to 1.0 for the whole-cell antigen assay and greater than or equal to 0.35 for the CagA antigen assay. Individuals who were seronegative for both types of antibodies (whole cell and CagA) were classified as H. pylori negative, whereas individuals who were seropositive for either whole-cell or CagA antibodies were classified as H. pylori positive. This classification system was used because a study comparing serology with gastric biopsy culture has shown that the stomachs of individuals who are negative for H. pylori whole-cell antibodies but positive for CagA antibodies are in fact colonized with CagA-positive strains of H. pylori (39). Therefore, H. pylori–positive individuals were further classified according to whether they carried CagA-negative strains (whole-cell seropositive, CagA seronegative) or CagA-positive strains (whole-cell seropositive, CagA seropositive or whole-cell seronegative, CagA seropositive) (39).
Serum samples from each case subject and the matched control subject were assayed in duplicate in the same batch by experienced technicians who were unaware of the case–control status of the samples. When the duplicate samples provided indeterminate results (i.e., the values straddled the seropositivity threshold), additional aliquots were analyzed, and the average of all results (excluding obvious outliers) was used to determine serologic status. Thirty-five quality control serum samples, aliquoted from a single large pool, were equally distributed among different batches. On the basis of these samples, average within-batch coefficients of variation for the whole-cell and CagA antigen assays were 15% and 20%, respectively.
Statistical Analysis
All P values are two-sided, and P values less than .05 were considered statistically significant. We used the Wilcoxon matched-pairs test to compare demographic characteristics between case subjects and the matched control subjects, including education level (primary school or lower versus high school or higher) and residence (urban versus rural), and potential confounders, including smoking duration (years), body mass index ([weight in kg]/[height in m]2), and intakes of nitrates (mg/day) and of fruits, vegetables, starch, and sodium (all in g/day). All dietary intake data were adjusted for calorie intake using the method of residuals (40). We fit conditional logistic regression models to estimate unadjusted and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for cancer by anatomic subsite. Adjustment was done for demographic characteristics and the potential confounders mentioned above. We also estimated the crude and adjusted odds ratios by duration of case subject follow-up (0.0–6.9 versus 7.0 years or more), by age at diagnosis (younger than 65 years versus 65 years or older), by histologic subtype (diffuse, intestinal, or not specified), and by H. pylori strain (CagA positive versus CagA negative). Cut points for age and follow-up were chosen so that there would be sufficient numbers of noncardia gastric cancer case subjects and gastric cardia cancer case subjects (40%–60% of cancer subjects) on each side of the cut point. Odds ratio point estimates varied by strata of follow-up, age at diagnosis, histology, and H. pylori strain, but none of the differences between these strata were statistically significant; 95% confidence intervals for the estimate in each group included the point estimates in other groups. We also repeated all the analyses using unconditional logistic regression analysis. No qualitative differences were observed between conditional and unconditional models. Therefore, we report only the results from unconditional logistic regression models for the associations by anatomic location. Only four diffuse-type cardia adenocarcinomas were diagnosed, which did not allow a meaningful subgroup analysis by histology for cardia adenocarcinomas. However, all descriptive data were reported for these four cases.
|
RESULTS |
|---|
|
TopNotesAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
In total, 220 subjects (47%) were positive for antibodies to both H. pylori whole-cell and CagA antigens, 130 (28%) were positive for whole-cell antibodies only, 23 (5%) were positive for CagA antibodies only, and 95 (20%) were positive for neither. Thus, 80% of the studied population (197 [84%] of the gastric cancer case subjects and 176 [75%] of the control subjects) had serologic evidence of H. pylori infection. The median time from serum collection to cancer diagnosis was 5.8 years (interquartile range [IQR] = 3.4–8.6 years) for noncardia gastric cancer case subjects and 6.7 years (IQR = 4.4–9.0 years) for cardia cancer case subjects.
Demographic Characteristics and Potential Confounders
Table 1 shows the distributions of the case and control subjects by demographic characteristics and potential confounders. Fruit consumption was statistically significantly lower among the noncardia gastric cancer case subjects than among the matched control subjects (P = .03). All other variables showed non–statistically significant differences between the noncardia gastric cancer case subjects and the matched control subjects. Results for fruits, vegetables, and other potential confounders in our nested study were similar to those obtained for the full cohort (41).
|
Anatomic Subsite
The prevalence of whole-cell antibody seropositivity was nearly identical among the control subjects matched to the noncardia gastric cancer case subjects and those matched to the gastric cardia cancer case subjects (71% and 72%, respectively; Table 2). Whole-cell antibody seropositivity was more common in the noncardia cancer case subjects than in the matched control subjects (86% versus 71%; adjusted OR = 3.32, 95% CI = 1.72 to 6.42) but less common in cardia cancer case subjects than in the matched control subjects (57% versus 72%; adjusted OR = 0.28, 95% CI = 0.09 to 0.86). H. pylori seropositivity rates were identical (75%) in the two groups of control subjects when we used the broader criterion for H. pylori seropositivity (i.e., positive for either whole-cell or CagA antigens). H. pylori seropositivity was more common in noncardia gastric cancer case subjects (93%; adjusted OR = 7.92, 95% CI = 3.02 to 20.9) but less common in gastric cardia cancer case subjects (59%; adjusted OR = 0.31, 95% CI = 0.11 to 0.89) compared with the respective control subjects. Thus, we observed opposing associations between H. pylori seropositivity and either noncardia or cardia gastric cancers, regardless of which definition of seropositivity was used. On the basis of these adjusted odds ratios, a calculation of rates showed that the absolute risks of noncardia gastric and cardia gastric adenocarcinomas in the H. pylori–positive participants of this cohort would be 63 and 12 per 100 000 person-years, respectively, whereas corresponding rates in the H. pylori–negative participants would be 8 and 37 per 100 000 person-years, respectively.
|
We also observed opposing associations when we used unconditional logistic models. The crude and adjusted odds ratios for the association between the risk of noncardia gastric cancer and H. pylori seropositivity were 4.50 (95% CI = 2.34 to 8.65) and 5.11 (95% CI = 2.51 to 10.43), respectively. The crude and adjusted odds ratios for the association between the risk of gastric cardia cancer and H. pylori seropositivity were 0.49 (95% CI = 0.28 to 0.88) and 0.47 (95% CI = 0.25 to 0.89), respectively.
Time From Serum Collection to Diagnosis
Identical proportions (93%) of H. pylori–seropositive noncardia gastric cancer case subjects were observed among those diagnosed within the 7 years after serum collection and those diagnosed 7 or more years after serum collection (Table 3). Compared with the control subjects, noncardia gastric cancer case subjects diagnosed within the 7 years after serum collection and those diagnosed 7 years or more after serum collection had adjusted odds ratios of cancer in association with H. pylori seropositivity of 12.97 (95% CI = 2.79 to 60.23) and 6.15 (95% CI = 1.42 to 26.56), respectively. The H. pylori seropositivity rates among gastric cardia cancer case subjects diagnosed in these two periods were close (55% and 64%, respectively), and both rates were less than the average rate among the matched control subjects (75%).
|
Age at Diagnosis
We used the median age at diagnosis (65 years) to stratify case subjects (and the age-matched control subjects) into two groups. Overall, the H. pylori seropositivity rate among control subjects younger than 65 years was lower than that among control subjects aged 65 years or older (72% versus 79%), reflecting a decline in H. pylori carriage in younger cohorts. H. pylori seropositivity was strongly associated with the risk of noncardia gastric cancer in both age groups. The adjusted odds ratio was 14.54 (95% CI = 3.04 to 69.64) for the younger cohort and 8.57 (95% CI = 1.73 to 45.54) for the older cohort (Table 3). However, the 95% confidence interval for the odds ratio in each cohort included the odds ratio point estimate for the other cohort. Therefore, the difference in odds ratios between the two age groups was not statistically significant. H. pylori seropositivity rates were lower among the gastric cardia cancer case subjects than among the control subjects, and the matched odds ratio point estimates were less than 1.0 in both age groups (Table 3).
Histologic Subtype
Among the noncardia gastric cancer case subjects, 94% of those with diffuse histology and 94% of those with intestinal histology were seropositive for anti–H. pylori antibodies; the H. pylori seropositivity rates among the corresponding matched control subjects were 79% and 77%, respectively (Table 3). The adjusted odds ratios for diffuse- and intestinal-type cancers were 5.70 (95% CI = 0.85 to 38.86) and 9.77 (95% CI = 2.21 to 45.60), respectively.
Among the gastric cardia cancer case subjects, only four (6%) had diffuse histology and 40 (66%) had intestinal histology. Because of the small number of diffuse-type cases, meaningful statistical comparisons with respect to histology were not possible.
CagA-Positive Versus CagA-Negative H. pylori Strains
Both CagA-negative and CagA-positive H. pylori strains were more common in the noncardia gastric cancer case subjects than in the matched control subjects, with respective adjusted odds ratios of noncardia gastric cancer associated with H. pylori seropositivity of 6.55 (95% CI = 2.31 to 18.53) and 8.93 (95% CI = 3.27 to 24.40), respectively (Table 4). By contrast, both CagA-negative and CagA-positive H. pylori strains were less common in cardia gastric cancer case subjects than in the matched control subjects, but the difference was statistically significant only for CagA-negative strains (adjusted OR = 0.21, 95% CI = 0.06 to 0.81) (Table 4).
|
|
DISCUSSION |
|---|
|
TopNotesAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
The results of this study show opposing associations between H. pylori with risk of noncardia gastric and gastric cardia adenocarcinomas. Our results confirm that there is a strong association between H. pylori seropositivity and risk of noncardia gastric cancer. A combined analysis of prospective studies of the association between H. pylori seropositivity and the risk of gastric cancer produced summary odds ratios of 3.0 for noncardia gastric adenocarcinomas (8). However, because it has been suggested that H. pylori does not colonize premalignant lesions of the stomach (19), serum antibodies to H. pylori may decline by the time gastric cancer develops. Therefore, the authors of the combined analysis suggested that the best estimate of the magnitude of odds ratio for noncardia adenocarcinomas was close to 6.0, the estimate obtained from case subjects with more than 10 years of follow-up. The adjusted odds ratio of 7.92 that we observed is close to that predicted by the combined analysis. However, we found no evidence for changes in H. pylori antibody levels over time; about 93% of noncardia cancer case subjects were positive for H. pylori antibodies, regardless of the length of time from serum collection to diagnosis. Because the prevalence of H. pylori seropositivity remained constant among the case subjects over time, any non–statistically significant variation in odds ratios is likely to be due to the small variation in H. pylori seropositivity among the control subjects over time.
The strength of association between H. pylori seropositivity and the risk of gastric cancer did not vary statistically significantly by age at case subject diagnosis. However, our finding of a higher point estimate for the association between H. pylori seropositivity and the risk of noncardia gastric cancer in younger (<65 years) men versus older (
65 years) men was consistent with findings from previous studies (1,8,14). One explanation for this observation is that noncardia gastric cancer may occur in older people via mechanisms that are unrelated to H. pylori infection. An alternative explanation is that, in the younger cohort, there is lower prevalence of H. pylori seropositivity in control subjects (11) but cases subjects remain positive for H. pylori.
Some studies (20,21) have suggested that the association between H. pylori and intestinal-type noncardia gastric cancer is stronger than the association between H. pylori and diffuse-type noncardia gastric cancer. However, two meta-analyses examined the H. pylori–associated risk ratio for histopathologic subtypes of noncardia cancer and reported no difference in risk between histologic subtypes (5) or a slightly higher risk (OR = 1.14) for the intestinal type compared with diffuse type (6). The combined analysis of prospective studies (8) did not find a risk difference between intestinal and diffuse-type noncardia gastric cancers. Our finding of no difference in the prevalence of H. pylori antibodies between individuals with intestinal and diffuse noncardia cancers (94% in both types) is consistent with the results of most previous studies (8,11,42). Our results are also consistent with those of a Swedish study that reported a decline in the rates of both intestinal and diffuse types of noncardia adenocarcinoma from 1989 to 1994 (43).
Several studies have shown that, compared with CagA-negative strains, CagA-positive strains of H. pylori are associated with a higher risk of noncardia gastric cancer (25). The genomes of CagA-positive strain of H. pylori contain the cag pathogenicity island, which includes approximately 31 putative genes, including cagA (23). Once delivered into epithelial cells, CagA may increase the risk of gastric cancer by increasing the turnover of the gastric epithelium (23). CagA-positive strains are also more likely than CagA-negative strains to have the s1 allele of vacA, which encodes a molecule that impairs epithelial cell proliferation (24) and may increase gastric cancer risk (16). CagA-positive strains are also more likely to express the babA gene product, which controls bacterial adherence to Lewisb antigens on gastric epithelial cells (24). We also found that, compared with CagA-negative strains, CagA-positive strains had a stronger association with risk of noncardia gastric cancer, although this difference was not statistically significant.
CagA measurement enabled improved classification of H. pylori–positive subjects in our study. A study that compared seropositivity to H. pylori whole-cell and CagA antigens versus H. pylori culture of biopsy samples showed that individuals who were seronegative for whole-cell antibodies but seropositive for CagA antibodies were in fact H. pylori positive and carried CagA-positive strains (39). Approximately 5% of all tested individuals in our study were positive for CagA antibodies but negative for whole-cell antibodies, a frequency similar to that reported in previous studies (38,42). Had we defined H. pylori–positive cases solely on the basis of whole-cell antibody positivity rather than on the basis of either whole-cell or CagA antibody positivity, the adjusted odds ratio would have been 3.3 (rather than 7.9), probably because of nondifferential misclassification of individuals with respect to their prior exposure to H. pylori.
We found that H. pylori seropositivity was inversely associated with the risk of gastric cardia adenocarcinomas. Only one other prospective study (44) has shown a statistically significant inverse association between H. pylori seropositivity and the risk of gastric cardia cancer. Other studies have reported conflicting results. For example, most studies of Asian populations have found a positive association between H. pylori seropositivity and this cancer, whereas most studies of Western populations have found no association or an inverse association (8,26). The present study, conducted in Finland, yielded results similar to those observed in most other Western studies (44,45).
One explanation for the different associations reported by Asian and Western studies may be that the latter studies (26), as well as the present study, might have included a number of esophageal adenocarcinoma subjects among cardia cancer subjects. It is not always possible to distinguish between adenocarcinomas that arise in the gastric cardia and those that arise in the lower esophagus or the body of the stomach. This issue is further complicated by the ambiguities in the definitions of cardia (46) and cardia cancer (47). We defined gastric cardia cancer cases as those that involved the esophagogastric junction. However, in some parts of Asia where Barrett's esophagus and adenocarcinomas of the lower esophagus are very rare or nonexistent [e.g., Linxian, China; (38)], tumors that are classified as cardia gastric cancers do not include any esophageal adenocarcinoma cases. Nevertheless, the results of this and other studies (44,45,48–50) provide evidence that H. pylori colonization in the stomach is associated with decreased risks of some forms of adenocarcinoma arising near the junction of the esophagus and the stomach.
Several lines of evidence argue that our finding of an inverse association between H. pylori seropositivity and the risk of gastric cardia cancer is real and not due to chance. First, an association of similar direction was found when serum samples from a subgroup of these case subjects and a different set of control subjects from the ATBC study were tested in a different laboratory. The results of this smaller study have not been published as an independent paper, but summary statistics are shown in the combined analysis of the prospective studies (8). Second, using unconditional models, the point estimate for this association remained around 0.5, and the association remained statistically significant regardless of other variables included in the model. Third, inverse associations between H. pylori seropositivity and risks of gastroesophageal junctional cancers (i.e., esophageal or cardia cancers) have been shown in other Western studies (44,45,48–50). The reasons for the observed inverse association between H. pylori seropositivity and the increased risk of gastroesophageal junctional adenocarcinomas are not entirely clear (48). However, one hypothesis is that H. pylori colonization induces gastric atrophy, which results in reduced gastric acidity, less acid reflux into the esophagus, and a reduced risk of Barrett's esophagus and junctional cancers (51,52).
H. pylori seropositivity was inversely associated with the risk of gastric cardia cancer for both shorter (0.0–6.9 years) and longer (7.0 years or more) follow-ups and in both younger and older cohorts, although the modest number of cardia cancer cases made it difficult to obtain stable estimates of odds ratios by subgroup, especially when using conditional models. Similarly, in this and prior studies (8), there were insufficient numbers of diffuse-type cardia cancer cases to reach meaningful conclusions about the effect of tumor histologic subtype on the association. The inverse association observed was more pronounced and statistically significant for CagA-negative strains than for CagA-positive strains, which contrasts with the results of an earlier retrospective study (48) that suggested that CagA-positive strains were more strongly associated with a decreased risk of gastric cardia cancers.
The results of this study suggest that caution is warranted against mass treatment to eradicate H. pylori. Because H. pylori has been consistently associated with higher risk of gastric noncardia cancer, several large clinical trials are ongoing to examine the effect of H. pylori eradication on gastric cancer incidence. The inverse association of H. pylori with gastric cardia cancer or esophageal adenocarcinoma, shown in this study and several other studies on Western populations (44,45,48–50), collectively suggests that H. pylori eradication may increase risk for gastroesophageal junctional cancers, at least in some Western populations. This hypothesis is supported by the substantial increase in the incidence of gastroesophageal junctional cancers in Western countries in the past few decades, possibly because of the lack of acquisition of H. pylori due to improved sanitation or the coincidental elimination of H. pylori due to widespread antibiotic use. Therefore, mass treatment to eradicate H. pylori may be justified in parts of the world where the incidence of distal gastric cancers far exceeds that of junctional cancers. However, in Western countries, where rates of gastroesophageal junctional cancer are now high, a mass eradication policy should be considered with caution. Future studies with stratifications based on H. pylori strains and careful anatomic classification should help clarify these issues.
The strengths of this study include the large sample size, the prospective design, the long-term follow-up, the availability of data on the location and histopathologic classification of tumors, the availability of extensive data on potential confounders, and the measurement of both H. pylori whole-cell and CagA antibodies. The most important limitation of this study, the possible misclassification in distinguishing cardia cancers from adenocarcinomas of the proximal noncardia stomach or lower esophagus, is a limitation inherent to all studies of gastric cardia cancers and does not refute our finding that H. pylori seropositivity is associated with reduced risks of some forms of gastroesophageal junctional cancer. The relatively small number of gastric cardia cancer cases limited our ability to perform subgroup analyses.
In summary, in this prospective study, H. pylori seropositivity was a strong risk factor for noncardia gastric cancer but was inversely associated with the risk of gastric cardia cancer. Moreover, H. pylori was a strong risk factor for noncardia gastric cancer regardless of the length of the interval between serum collection and diagnosis or the tumor histologic subtype. Our findings bolster the hypothesis that the decreasing prevalence of H. pylori infection during the past century may have contributed to the lower rates of noncardia gastric cancer and to the higher rates of cardia/junctional gastric cancers observed in Western countries (53,54).
|
NOTES |
|---|
|
TopNotesAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
M. J. Blaser is a codiscoverer of CagA and could receive royalties from diagnostic tests, none of which are currently licensed.
This research was supported in part by the Intramural Research Program of the National Institutes of Health, National Cancer Institute. The study sponsor did not have any role in the design of the study; the collection, analysis, and interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication.
|
REFERENCES |
|---|
|
TopNotesAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
(1)Nomura A, Stemmermann GN, Chyou PH, Kato I, Perez-Perez GI, Blaser MJ. Helicobacter pylori infection and gastric carcinoma among Japanese Americans in Hawaii. N Engl J Med 1991;325:1132–6.[Abstract]
(2) Parsonnet J, Friedman GD, Vandersteen DP, Chang Y, Vogelman JH, Orentreich N, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991;325:1127–31.[Abstract]
(3) Forman D, Newell DG, Fullerton F, Yarnell JW, Stacey AR, Wald N, et al. Association between infection with Helicobacter pylori and risk of gastric cancer: evidence from a prospective investigation. BMJ 1991;302:1302–5.
(4) International Agency for Research on Cancer. Schistosomes, liver flukes and Helicobacter pylori. Lyon (France): IARC Scientific Publication; 1994.
(5) Huang JQ, Sridhar S, Chen Y, Hunt RH. Meta-analysis of the relationship between Helicobacter pylori seropositivity and gastric cancer. Gastroenterology 1998;114:1169–79.[CrossRef][Web of Science][Medline]
(6) Eslick GD, Lim LL, Byles JE, Xia HH, Talley NJ. Association of Helicobacter pylori infection with gastric carcinoma: a meta-analysis. Am J Gastroenterol 1999;94:2373–9.[CrossRef][Web of Science][Medline]
(7) Danesh J. Helicobacter pylori infection and gastric cancer: systematic review of the epidemiological studies. Aliment Pharmacol Ther 1999;13:851–6.[CrossRef][Web of Science][Medline]
(8) Helicobacter and Cancer Collaborative Group. Gastric cancer and Helicobacter pylori: a combined analysis of 12 case control studies nested within prospective cohorts. Gut 2001;49:347–53.
(9) Uemura N, Okamoto S, Yamamoto S, Matsumura N, Yamaguchi S, Yamakido M, et al. Helicobacter pylori infection and the development of gastric cancer. N Engl J Med 2001;345:784–9.
(10) Suerbaum S, Michetti P. Helicobacter pylori infection. N Engl J Med 2002;347:1175–86.
(11) Kato M, Asaka M, Shimizu Y, Nobuta A, Takeda H, Sugiyama T. Relationship between Helicobacter pylori infection and the prevalence, site and histological type of gastric cancer. Aliment Pharmacol Ther 2004;20(Suppl 1):85–9.[CrossRef][Web of Science][Medline]
(12) Brenner H, Arndt V, Stegmaier C, Ziegler H, Rothenbacher D. Is Helicobacter pylori infection a necessary condition for noncardia gastric cancer? Am J Epidemiol 2004;159:252–8.
(13) Knekt P, Teppo L, Aromaa A, Rissanen H, Kosunen TU. Helicobacter pylori IgA and IgG antibodies, serum pepsinogen I and the risk of gastric cancer: changes in the risk with extended follow-up period. Int J Cancer 2006;119:702–5.[CrossRef][Web of Science][Medline]
(14) Hansson LE, Engstrand L, Nyren O, Evans DJ Jr, Lindgren A, Bergstrom R, et al. Helicobacter pylori infection: independent risk indicator of gastric adenocarcinoma. Gastroenterology 1993;105:1098–103.[Web of Science][Medline]
(15) Blaser MJ, Perez-Perez GI, Kleanthous H, Cover TL, Peek RM, Chyou PH, et al. Infection with Helicobacter pylori strains possessing cagA is associated with an increased risk of developing adenocarcinoma of the stomach. Cancer Res 1995;55:2111–5.
(16) Figueiredo C, Machado JC, Pharoah P, Seruca R, Sousa S, Carvalho R, et al. Helicobacter pylori and interleukin 1 genotyping: an opportunity to identify high-risk individuals for gastric carcinoma. J Natl Cancer Inst 2002;94:1680–7.
(17) El Omar EM, Carrington M, Chow WH, McColl KE, Bream JH, Young HA, et al. Interleukin-1 polymorphisms associated with increased risk of gastric cancer. Nature 2000;404:398–402.[CrossRef][Medline]
(18) El Omar EM, Rabkin CS, Gammon MD, Vaughan TL, Risch HA, Schoenberg JB, et al. Increased risk of noncardia gastric cancer associated with proinflammatory cytokine gene polymorphisms. Gastroenterology 2003;124:1193–201.[CrossRef][Web of Science][Medline]
(19) Karnes WE Jr, Samloff IM, Siurala M, Kekki M, Sipponen P, Kim SW, et al. Positive serum antibody and negative tissue staining for Helicobacter pylori in subjects with atrophic body gastritis. Gastroenterology 1991;101:167–74.[Web of Science][Medline]
(20) Parsonnet J, Vandersteen D, Goates J, Sibley RK, Pritikin J, Chang Y. Helicobacter pylori infection in intestinal- and diffuse-type gastric adenocarcinomas. J Natl Cancer Inst 1991;83:640–3.
(21) Martin-de-Argila C, Boixeda D, Redondo C, Alvarez I, Gisbert JP, Garcia PA, et al. Relation between histologic subtypes and location of gastric cancer and Helicobacter pylori. Scand J Gastroenterol 1997;32:303–7.[Web of Science][Medline]
(22) Wu MS, Chen SY, Shun CT, Lee WJ, Wang HP, Wang TH, et al. Increased prevalence of Helicobacter pylori infection among patients affected with intestinal-type gastric cancer at non-cardiac locations. J Gastroenterol Hepatol 1997;12:425–8.[Web of Science][Medline]
(23) Hatakeyama M. Oncogenic mechanisms of the Helicobacter pylori CagA protein. Nat Rev Cancer 2004;4:688–94.[CrossRef][Web of Science][Medline]
(24) Blaser MJ, Berg DE. Helicobacter pylori genetic diversity and risk of human disease. J Clin Invest 2001;107:767–73.[Web of Science][Medline]
(25) Huang JQ, Zheng GF, Sumanac K, Irvine EJ, Hunt RH. Meta-analysis of the relationship between cagA seropositivity and gastric cancer. Gastroenterology 2003;125:1636–44.[CrossRef][Web of Science][Medline]
(26) Dawsey SM, Mark SD, Taylor PR, Limburg PJ. Gastric cancer and H pylori. Gut 2002;51:457–8.
(27) Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049–53.[CrossRef][Web of Science][Medline]
(28) Walther C, Zilling T, Perfekt R, Moller T. Increasing prevalence of adenocarcinoma of the oesophagus and gastro-oesophageal junction: a study of the Swedish population between 1970 and 1997. Eur J Surg 2001;167:748–57.[CrossRef][Web of Science][Medline]
(29) Wijnhoven BP, Louwman MW, Tilanus HW, Coebergh JW. Increased incidence of adenocarcinomas at the gastro-oesophageal junction in Dutch males since the 1990s. Eur J Gastroenterol Hepatol 2002;14:115–22.[CrossRef][Web of Science][Medline]
(30) Blaser MJ, Saito D. Trends in reported adenocarcinomas of the oesophagus and gastric cardia in Japan. Eur J Gastroenterol Hepatol 2002;14:107–13.[CrossRef][Web of Science][Medline]
(31) Shimatani T, Inoue M, Iwamoto K, Hyogo H, Yokozaki M, Saeki T, et al. Prevalence of Helicobacter pylori infection, endoscopic gastric findings and dyspeptic symptoms among a young Japanese population born in the 1970s. J Gastroenterol Hepatol 2005;20:1352–7.[CrossRef][Web of Science][Medline]
(32) Chang SS, Lu CL, Chao JY, Chao Y, Yen SH, Wang SS, et al. Unchanging trend of adenocarcinoma of the esophagus and gastric cardia in Taiwan: a 15-year experience in a single center. Dig Dis Sci 2002;47:735–40.[CrossRef][Web of Science][Medline]
(33) Lee JY, Kim HY, Kim KH, Jang HJ, Kim JB, Lee JH, et al. No changing trends in incidence of gastric cardia cancer in Korea. J Korean Med Sci 2003;18:53–7.[Web of Science][Medline]
(34) The ATBC Cancer Prevention Study Group. The alpha-tocopherol, beta-carotene lung cancer prevention study: design, methods, participant characteristics, and compliance. Ann Epidemiol 1994;4:1–10.[Medline]
(35) Korhonen P, Malila N, Pukkala E, Teppo L, Albanes D, Virtamo J. The Finnish Cancer Registry as follow-up source of a large trial cohort—accuracy and delay. Acta Oncol 2002;41:381–8.[CrossRef][Web of Science][Medline]
(36) World Health Organization. International statistical classification of diseases and related health problems. Geneva (Switzerland): World Health Organization; 1978.
(37) Pietinen P, Hartman AM, Haapa E, Rasanen L, Haapakoski J, Palmgren J, et al. Reproducibility and validity of dietary assessment instruments. I. A self-administered food use questionnaire with a portion size picture booklet. Am J Epidemiol 1988;128:655–66.
(38) Limburg P, Qiao Y, Mark S, Wang G, Perez-Perez G, Blaser M, et al. Helicobacter pylori seropositivity and subsite-specific gastric cancer risks in Linxian, China. J Natl Cancer Inst 2001;93:226–33.
(39) Romero-Gallo J, Perez-Perez GI, Novick RP, Kamath P, Norbu T, Blaser MJ. Responses of endoscopy patients in Ladakh, India, to Helicobacter pylori whole-cell and Cag A antigens. Clin Diagn Lab Immunol 2002;9:1313–7.[CrossRef][Medline]
(40) Willett W, Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol 1986;124:17–27.
(41) Nouraie M, Pietinen P, Kamangar F, Dawsey SM, Abnet CC, Albanes D, et al. Fruits, vegetables, and antioxidants and risk of gastric cancer among male smokers. Cancer Epidemiol Biomarkers Prev 2005;14:2087–92.
(42) Nomura AM, Lee J, Stemmermann GN, Nomura RY, Perez-Perez GI, Blaser MJ. Helicobacter pylori CagA seropositivity and gastric carcinoma risk in a Japanese American population. J Infect Dis 2002;186:1138–44.[CrossRef][Web of Science][Medline]
(43) Ekstrom AM, Hansson LE, Signorello LB, Lindgren A, Bergstrom R, Nyren O. Decreasing incidence of both major histologic subtypes of gastric adenocarcinoma—a population-based study in Sweden. Br J Cancer 2000;83:391–6.[CrossRef][Web of Science][Medline]
(44) Hansen S, Melby KK, Aase S, Jellum E, Vollset SE. Helicobacter pylori infection and risk of cardia cancer and non-cardia gastric cancer. A nested case-control study. Scand J Gastroenterol 1999;34:353–60.[CrossRef][Web of Science][Medline]
(45) Aromaa A, Kosunen TU, Knekt P, Maatela J, Teppo L, Heinonen OP, et al. Circulating anti-Helicobacter pylori immunoglobulin A antibodies and low serum pepsinogen I level are associated with increased risk of gastric cancer. Am J Epidemiol 1996;144:142–9.
(46) Derdoy JJ, Bergwerk A, Cohen H, Kline M, Monforte HL, Thomas DW. The gastric cardia: to be or not to be? Am J Surg Pathol 2003;27:499–504.[CrossRef][Web of Science][Medline]
(47) Taniere P, Martel-Planche G, Maurici D, Lombard-Bohas C, Scoazec JY, Montesano R, et al. Molecular and clinical differences between adenocarcinomas of the esophagus and of the gastric cardia. Am J Pathol 2001;158:33–40.
(48) Chow WH, Blaser MJ, Blot WJ, Gammon MD, Vaughan TL, Risch HA, et al. An inverse relation between cagA+ strains of Helicobacter pylori infection and risk of esophageal and gastric cardia adenocarcinoma. Cancer Res 1998;58:588–90.
(49) Ye W, Held M, Lagergren J, Engstrand L, Blot WJ, McLaughlin JK, et al. Helicobacter pylori infection and gastric atrophy: risk of adenocarcinoma and squamous-cell carcinoma of the esophagus and adenocarcinoma of the gastric cardia. J Natl Cancer Inst 2004;96:388–96.
(50) de Martel C, Llosa AE, Farr SM, Friedman GD, Vogelman JH, Orentreich N, et al. Helicobacter pylori infection and the risk of development of esophageal adenocarcinoma. J Infect Dis 2005;191:761–7.[CrossRef][Web of Science][Medline]
(51) Richter JE, Falk GW, Vaezi MF. Helicobacter pylori and gastroesophageal reflux disease: the bug may not be all bad. Am J Gastroenterol 1998;93:1800–2.[CrossRef][Web of Science][Medline]
(52) Lagergren J. Adenocarcinoma of oesophagus: what exactly is the size of the problem and who is at risk? Gut 2005;54(Suppl 1):i1–5.
(53) Blaser MJ. Hypothesis: the changing relationships of Helicobacter pylori and humans: implications for health and disease. J Infect Dis 1999;179:1523–30.[CrossRef][Web of Science][Medline]
(54) Peek RM Jr, Blaser MJ. Helicobacter pylori and gastrointestinal tract adenocarcinomas. Nat Rev Cancer 2002;2:28–37.[CrossRef][Web of Science][Medline]
Manuscript received February 7, 2006; revised August 4, 2006; accepted August 23, 2006.
Editorial about this Article
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2006 98: 1432-1434.
This article has been cited by other articles:
|
|
 |
|
  M. J. Blaser Disappearing Microbiota: Helicobacter pylori Protection against Esophageal Adenocarcinoma Cancer Prevention Research, October 1, 2008; 1(5): 308 - 311. [Full Text] [PDF]
|
|
|
|
|
|
F. Islami and F. Kamangar Helicobacter pylori and Esophageal Cancer Risk: A Meta-analysis Cancer Prevention Research, October 1, 2008; 1(5): 329 - 338. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Takaishi, T. Okumura, and T. C. Wang Gastric Cancer Stem Cells J. Clin. Oncol., June 10, 2008; 26(17): 2876 - 2882. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D A Corley, A Kubo, T R Levin, G Block, L Habel, W Zhao, P Leighton, G Rumore, C Quesenberry, P Buffler, et al. Helicobacter pylori infection and the risk of Barrett's oesophagus: a community-based study Gut, June 1, 2008; 57(6): 727 - 733. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M H Derakhshan, R Malekzadeh, H Watabe, A Yazdanbod, V Fyfe, A Kazemi, N Rakhshani, R Didevar, M Sotoudeh, A A Zolfeghari, et al. Combination of gastric atrophy, reflux symptoms and histological subtype indicates two distinct aetiologies of gastric cardia cancer Gut, March 1, 2008; 57(3): 298 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hansen, S. E. Vollset, M. H Derakhshan, V. Fyfe, K. K Melby, S. Aase, E. Jellum, and K. E L McColl Two distinct aetiologies of cardia cancer; evidence from premorbid serological markers of gastric atrophy and Helicobacter pylori status Gut, July 1, 2007; 56(7): 918 - 925. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Gastric Carcinomas Associated with H. pylori Infection Journal Watch Oncology and Hematology, December 4, 2006; 2006(1204): 1 - 1. [Full Text]
|
|
|
|
 |
|
 O. Nyren and W. J. Blot Helicobacter pylori Infection: Mainly Foe but Also Friend? J Natl Cancer Inst, October 18, 2006; 98(20): 1432 - 1434. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||