Association of Adenocarcinomas of the Distal Esophagus, “Gastroesophageal Junction,” and “Gastric Cardia” With Gastric Pathology

Controversy exists as to whether adenocarcinomas occurring in the gastroesophageal junctional region and gastric cardia originate in the esophagus or the stomach. Esophageal adenocarcinoma is known to be strongly associated with gastroesophageal reflux disease; gastric adenocarcinoma with Helicobacter pylori gastritis, and gastric intestinal metaplasia. This study evaluates the association of these tumors with pathologic findings in the biopsies of the gastric body and the antrum. It is hypothesized that if these malignancies are esophageal, they should have little or no significant association with gastric pathology; if they are gastric, these patients should have a high prevalence of gastric pathology. Between 2004 and 2008, 234 patients were diagnosed with high-grade dysplasia (HGD) and/or adenocarcinoma; 107 were distal esophageal, 79 straddled the distal end of the tubular esophagus, and 48 were in the “gastric cardia.” Simultaneous biopsies of the distal body and antrum were present in 185 patients; 49 had biopsy of either antrum or body. Gastric biopsies were assessed for inflammation, H. pylori infection, and intestinal metaplasia. During this period, 2146 patients had nonmalignant columnar epithelia in the esophagus with similar assessment of the stomach; these acted as a control group. The gastric biopsy was normal in 201/234 (85.9%) patients and showed significant inflammation, H. pylori infection, and/or gastric intestinal metaplasia in 33/234 (14.1%) patients. There was no gastritis, H. pylori infection, or intestinal metaplasia in 88/107 (82.2%) of the patients with distal esophageal HGD and/or adenocarcinoma, 70/79 (88.6%) with junctional HGD and/or adenocarcinoma, and 43/48 (85.9%) with “gastric cardiac” HGD and/or adenocarcinoma. The incidence of gastritis was significantly higher in the patients with HGD and/or adenocarcinoma (33/234 or 14.1%) than in the control population (146/2146 or 9.0%; P=0.01). This difference was largely the result of a higher incidence of gastritis in patients with HGD and/or adenocarcinoma in the distal third of the esophagus (19/107 or 17.8%) versus the control population (146/2146 or 9.0%; P=0.01). The incidence of H. pylori positivity was also significantly higher in the patients with HGD and/or adenocarcinoma in the distal third of the esophagus (13/107 or 12.2%) than in the control population (117/2146 or 5.5%; P=0.01). There was no significant difference between the control group and the patients with junctional and gastric cardiac HGD and/or adenocarcinoma for gastritis, H. pylori infection, or the gastric intestinal metaplasia. The absence of gastritis, H. pylori, and the gastric intestinal metaplasia in 85.9% of the patients with HGD and/or adenocarcinoma of the gastroesophageal junctional region strongly suggest that most of these originate in the esophagus. In the small minority of patients whose HGD and/or adenocarcinoma were associated with gastric pathology, the incidence of gastritis and H. pylori infection was significantly higher in patients with HGD and/or adenocarcinoma in the distal third of the esophagus and not in the junctional and “gastric cardiac” tumors. This suggests that the reflux of the gastric juice whose composition has been altered by gastritis and H. pylori infection may be associated with an increased tendency to HGD and/or adenocarcinoma in the distal third of the esophagus.