165 Outcomes of Esophagogastric Junction Adenocarcinomas Initially Treated by Endoscopic Resection

Abstract

Outcomes of Esophagogastric Junction Adenocarcinomas Initially Treated by Endoscopic Resection Hiromitsu Kanzaki, Ryu Ishihara, Noriya Uedo, Sachiko Yamamoto, Yoji Takeuchi, Koji Higashino, Hiroyasu Iishi Department of Gastrointestinal Oncololy, Osaka Medical Center for Cancer and Cardiaovascular Diseases, Osaka, Japan Introduction: Few studies have reported the outcome of Siewert type II adenocarcinomas treated by endoscopic resection (ER). We used ER as initial treatment for esophagogastric junction (EGJ) cancers 5cm in diameter and confined the mucosal layer in preoperative staging. Aims & Methods: A total of 1578 gastric or esophageal adenocarcinomas were treated by ER from April 2000 to December 2007. The data for the patients and lesions were stored consecutively in a database. Fifty-seven EGJ cancers in 53 patients from the database met our inclusion criteria. The inclusion criteria were; (1) histrogically proven adenocarcinoma; (2) tumor center within 1cm above and 2cm below the junction (Siewert type II cancer); (3) no lymph node or distant metastases; and (4) clinically diagnosed as mucosal cancer 5cm. Treatment outcomes of EGJ cancers after ER were compared with those of non-EGJ gastric cancers. And long-term survival of EGJ cancers after the treatment was investigated. Results: In the cases of EGJ cancers, ER was completed in all lesions. Thirteen of the 53 patients received additional surgical resection of the stomach with lymph node dissection because of the risk of lymph node metastasis, based on submucosal deep invasion revealed by histological examination of resected specimens. Two patients received photodynamic therapy. The other patients were followed-up with no additional treatment. The complete resection rate for EGJ cancers (58%: 33/57) was significantly lower than that for gastric cancers (76%: 1162/1521) (P 0.03). Diagnostic accuracy for the invasion depth of EGJ cancers (63%: 36/ 57) was significantly lower than that for gastric cancers (85%: 1275/1505) (P 0.001). Major complications of ER included perforation (2cases) and postoperative bleeding (2cases). All complications were successfully treated by endoscopy. All patients with EGJ cancer were followed-up until their death, or until January 2009. Local recurrences were observed during follow-up in two patients, but no lymph node or distant metastases were observed. All local recurrences were successfully treated by additional ER. Five deaths were identified after a mean observation period of 1396 775 days. However, no patient died of esophageal or gastric cancer. The overall 3-year and 5-year survival rates for all patients were 95.5% and 87.9%, respectively. The causespecific 3-year and 5-year survival rates for all patients were both 100%. Conclusion: Additional surgical resection was required in some patients because of the lower preoperative diagnostic accuracy for the depth of EGJ cancers. However, the overall and cause-specific survival rates for patients with EGJ cancer were favorable, and the complications were acceptable. Endoscopic resection is suitable for the initial treatment of patients with EGJ cancers confined to the mucosal layer at preoperative staging.