Cancer of the esophagus and gastric cardia : etiological aspects

Abstract

Esophageal cancer is one of the most lethal forms of cancer and the prognosis has not improved substantially during recent decades despite new treatments and diagnostic methods. In view of these discouraging results it is important to identify risk factors that might make primary prevention possible. The major risk factors for esophageal squamous cell carcinoma are tobacco smoking and alcohol use. Infection with human papillomavirus type 16 has also been claimed to be associated with the risk. While the incidence of squamous cell carcinoma of the esophagus is stable, that of adenocarcinoma of the esophagus and gastric cardia has increased dramatically in several Western countries. The reasons for this increase are unknown, and there is a paucity of known risk factors.

To identify risk factors for esophageal and gastric cardia cancer, and to estimate the magnitude of their influence in the Swedish population, a nationwide population-based case-control study was performed. The study base was the entire Swedish population born in Sweden and aged below 80 years in 1995 through 1997. Collaboration between all Swedish hospital departments involved in the diagnosis or treatment of esophageal or cardia tumors ensured that incident cases were uniformly classified and reported to the study. Cases and population-based control subjects were interviewed face to-face about various exposures. Blood samples were collected for serological analyses of human papillomavirus.

Gastroesophageal reflux symptoms were strongly associated with the risk of esophageal adenocarcinoma and moderately with the risk of gastric cardia adenocarcinoma. Increasing frequency, severity, and duration of symptoms increased the risk dose-dependently. Presence of Barrett s esophagus or obesity did not affect the association. Esophageal squamous cell carcinoma was not associated with reflux symptoms.

There was a strong and dose-dependent relation between increasing body mass index (BMI) and risk of esophageal adenocarcinoma through the entire range ol BMI values. The risk was independent of reflux symptoms. BMI showed a weak association with adenocarcinoma of the gastric cardia, but not with esophageal squamous cell carcinoma. The risk of esophageal adenocarcinoma was not importantly associated with smoking, snuff dipping, or alcohol use. Gastric cardia adenocarcinoma was dose-dependently associated with smoking, but not with alcohol or snuff use. Esophageal squamous cell carcinoma was strongly associated with tobacco, and moderately with alcohol, but not with snuff use. The combined use of tobacco and alcohol entailed a strongly increased risk of squamous cell carcinoma.

There was no serological evidence of a positive association between human papillomavirus infection types 16 or 18 and either form of esophageal cancer.

The striking male predominance in patients with esophageal adenocarcinoma is unexplained. We hypothesized that female sex hormones may be protective. In a retrospective population-based cohort study among all patients given a diagnosis of prostate cancer in Sweden between 1958 and 1992, and thus heavily exposed to estrogen, we tested this hypothesis. There was no evidence of a role of sex hormones in the etiology of esophageal adenocarcinoma.