Endoscopic gastric mucosal atrophy distinguishes the characteristics of superficial esophagogastric junction adenocarcinoma.

Background And Aim: Western studies have suggested two distinct etiologies of esophagogastric junction (EGJ) cancer: Helicobacter pylori-associated atrophic gastritis and non-atrophic gastric mucosa resembling esophageal adenocarcinoma. The present study investigated whether endoscopic gastric mucosal atrophy can distinguish between these two types of EGJ adenocarcinoma.

Methods: Data were collected from patients with Siewert type II, T1 EGJ adenocarcinoma who underwent endoscopic or surgical resection at eight Japanese institutions in 2010-2015. Clinicopathological characteristics of EGJ cancers with and without endoscopic gastric mucosal atrophy were compared. EGJ was defined as the lower end of the palisade vein and/or the top of the gastric folds.

Results: Of the 229 patients identified, 161 had endoscopic gastric mucosal atrophy and 68 did not. The latter group was younger (64 vs 70 years, P = 0.000); had a higher proportion of patients negative for H. pylori (90% vs 47%, P < 0.0001); and had higher rates of gastroesophageal reflux disease symptoms (43% vs 12%, P = 0.017), mucosal breaks (25% vs 15%, P = 0.009), Barrett's esophagus (BE, 78% vs 42%, P < 0.0001), and tumors above the EGJ (81% vs 19%, P < 0.0001) and on the upper-right side (74% vs 38%, P < 0.0001) than the former group. Multivariate analysis showed that H. pylori positivity (odds ratio [OR] = 13.0, P < 0.001), long-segment BE (OR = 0.025, P = 0.033), and longitudinal (OR = 8.6, P = 0.001) and circumferential (OR = 4.7, P = 0.006) tumor locations were independently associated with gastric mucosal atrophy.

Conclusion: Two distinct types of EGJ cancer were identified, with and without endoscopic gastric mucosal atrophy. These types were associated with different tumor locations.