Etiological aspects of gastroesophageal cancers : an epidemiological approach

Abstract

Aims: The aims of this thesis were to study the association of overall dietary habits, duodenal ulcer and gastric ulcer diseases, as two models of Helicobacter pylori (H. pylori) infection and the development of gastroesophageal cancers to further explore the etiology of these malignancies. Moreover, the association between parity, a proxy for high level of female hormones, and stomach cancer risk was studied to shed light on the enigma of male predominance in this cancer.

Methods: In a nationwide population-based case-control study in Sweden with 165 esophageal squamous cell carcinoma, 185 esophageal adenocarcinoma and 258 cardia cancer cases and 815 randomly selected controls we estimated relative risks associated with dietary patterns. Furthermore, cohorts of 61,548 and 81,379 unoperated patients with duodenal ulcer and gastric ulcer, respectively, recorded in the Swedish Inpatient Register since 1965, were followed and standardized incidence ratios were estimated for esophageal cancer by histology, compared with the Swedish general population. We also followed cohorts of 59,550 and 79,412 unoperated patients with duodenal ulcer and gastric ulcer, respectively, plus 12,840 patients with partial gastric resection and 8,105 with vagotomy, recorded since 1970. We estimated relative risks for stomach cancer by anatomical subtype compared to the Swedish general population for unoperated cohorts, whereas relative risks were estimated among operated patients relative to unoperated ones with the same ulcer type. Finally, in a case-control study, nested within a cohort of Swedish women born in 1925 or later, with 286 cardia cancer and 2,498 non-cardia stomach cancer as well as corresponding 1,430 and 12,490 controls, we investigated the relationship between parity and risk of stomach cancer by anatomic subsite.

Results: A “healthy diet” tended to moderately decrease the risk of esophageal and cardia cancers, “Western diet” increased risks of cardia cancer and esophageal adenocarcinoma, whereas a dietary pattern characterized by high beer and liquor intake significantly increased the risk of squamous-cell carcinoma of the esophagus. We observed that patients with duodenal ulcer had a significant 70% excess risk of esophageal adenocarcinoma, a non- significant small excess risk of esophageal squamous cell carcinoma, a halved risk of non-cardia cancer, and a risk of cardia cancer slightly above expectation. Gastric ulcer was unrelated to esophageal adenocarcinoma but linked to 80% increased risk of esophageal squamous cell carcinoma, and doubled risks for both anatomical types of stomach cancer. Duodenal ulcer patients who underwent gastric resection had a 60% risk elevation for noncardia cancer compared to unoperated ones. Vagotomy was associated with a greater risk in the first 10 years, but this excess disappeared with further follow-up. Resected gastric ulcer patients had a 40% risk reduction for non-cardia cancer relative to their unoperated peers. We found no association between parity and risk of non-cardia stomach cancer comparing everparous women with nulliparous, whereas a statistically significant 30% risk reduction for postmenopausal cardia cancer was noted among everparous women relative to nulliparous.

Conclusions: Overall dietary habits seem to play an important role in the carcinogenesis of esophageal and cardia cancer. The well-established strong inverse association of H. pylori seropositivity and risk of esophageal adenocarcinoma does not pertain to all infections. The duodenal ulcer related protection against stomach cancer does not seem to affect cardia cancer. It seems that the pattern of stomach colonization and/or the clinical consequences in the stomach might play a pivotal role in the long term outcome of H. pylori infection. With gastric resection, risks are shifted toward normality, regardless of underlying ulcer type. Exposure to female sex hormones is not associated with protection against stomach cancer and does not seem to explain the male predominance for such cancer. However the observed moderate inverse relationship between parity and cardia cancer might be mediated by other factors than hormonal.