Information on the study funded by Swedish Cancer Society
Until the mid 1970’s, adenocarcinoma was a rare cancer type in the esophagus. But since then, the incidence of this cancer has increased dramatically in Western populations (1, 2), including Sweden (3). In the US, it has been the cancer with the steepest rising incidence, and it surpassed squamous-cell carcinoma as the most common cancer in the esophagus around 1990. In Sweden, esophageal adenocarcinoma was more common than squamous cell carcinoma since year 2000. Given the almost simultaneous upward patterns in Western countries, some ubiquitous exposure introduced in the late 50’s or 60’s (assuming a 10-15 year induction and latency period) is likely to be responsible. Multidisciplinary efforts have so far failed to identify plausible candidate factors, except possibly for gastroesophageal reflux and relative body weight (4-7).
An inverse association between presence of antibodies to H. pylori and risk of esophago-cardial adenocarcinoma was first reported by Chow et al, based on analyses of a case-control study (8). In a nationwide case-control study, we found similarly that H. pylori infection, assayed by serum antibodies against whole H. pylori cell-surface antigens (HP-CSA) or CagA antibodies, was statistically significantly associated with a reduced risk for esophageal adenocarcinoma (for HP-CSA antibodies, odds ratio [OR] = 0.3, 95% confidence interval [CI] = 0.2 to 0.6; for CagA antibodies, OR = 0.5, 95% CI = 0.3 to 0.8; for both, OR = 0.2, 95% CI = 0.1 to 0.5) (9). The apparent protective effect of H. pylori against esophageal adenocarcinoma may be through its capacity to induce gastric mucosal atrophy, which in turn reduces acid secretion and therefore diminishes acid damage to the esophageal mucosa due to gastroesophageal reflux. We evaluated whether gastric atrophy, with its associated reduction of the gastric acid output, is in the causal pathway between H. pylori infection and esophageal adenocarcinoma development. We also reasoned that if H. pylori and gastric atrophy are both in the causal chain, then 1) their associations with esophageal adenocarcinoma should be in the same direction, 2) the inverse association between H. pylori infection and the risk of esophageal adenocarcinoma should be confined to subjects with gastric atrophy, and 3) the inverse association between H. pylori infection and risk of esophageal adenocarcinoma should be confined to subjects with gastroesophageal reflux symptoms. However, the odds ratio for esophageal adenocarcinoma among subjects with stomach atrophy, as indicated by low Pepsinogen I level, was 1.1 (95%CI 0.4-2.7), as compared with those with normal Pepsinogen I levels. Stratified analyses by presence of gastric atrophy or symptomatic gastroesophageal reflux did not reveal dramatic differences. This finding, along with no decreased risk for esophageal adenocarcinoma in pernicious anemia patients who have had long-term achlorhydria (10), seems to challenge the atrophy-induced hypochlorhydria hypothesis. However, the statistical power for our study is limited, especially for stratified analyses. Further, in our recent study a 70% excess risk of esophageal adenocarcimoa was noted in patients with duodenal ulcer patients as compared with the general population, which seems to support a role of gastric acid (11).
Recent studies have shown that H. pylori infection may influence levels of leptin and ghrelin, hormones affecting appetite and satiety (12, 13). A recent case-control study supports this hypothesis and shows that lack of H pylori infection, especially during childhood, might enhance the risk of the development of morbid obesity (14). Since overweight and obesity is a well-established risk factor for esophageal adenocarcinoma, an alternative explanation of the inverse association between H. pylori infection and this malignancy is through reduced body mass with the infection. However, none of previous studies has ever tested this hypothesis.
In patients with H. pylori infection, colonization in the cardia is common. Although evidence is accumulating that H. pylori colonization may be linked to carditis and intestinal metaplasia in the cardia, previous studies among Western populations do not support an important role of H. pylori infection in the development of gastric cardia adenocarcinoma (15). Of particular interest is the hypothesis that there are two etiological pathways, distinguished by presence of gastric atrophy, for the association between H. pylori infection and cardia adenocarcinoma (16).
1) To test whether the putative protective effect of H. pylori seropositivity against esophageal adenocarcinoma is explained by the induced gastric atrophy.
2) To test whether the putative protective effect of H. pylori seropositivity against esophageal adenocarcinoma is explained by differences in anthropometric measures, particularly body mass index.
3) To test whether the association between H. pylori infection and esophageal adenocarcinoma is modified by gender.
4) To test whether the association between H. pylori seropositivity and gastric cardia adenocarcinoma is modified by gastric atrophy.