Clinical validity of the expanded criteria for endoscopic resection of undifferentiated-type early gastric cancer based on long-term outcomes

Background According to the Japanese Gastric Cancer Treatment Guidelines, the expanded criteria for endoscopic resection (ER) of undifferentiated-type early gastric cancer (UEGC) is ulcer-negative, intramucosal cancer 20 mm or less in diameter without lymphovascular invasion. The aim of this study was to confirm validity of the expanded criteria for curative ER of UEGC. Methods Subjects were 125 patients from whom 125 UEGCs were resected endoscopically between April 1990 and March 2011. Endoscopic mucosal resection (EMR) (28 lesions, 28 patients) or endoscopic submucosal dissection (ESD) (97 lesions, 97 patients) was performed. We determined the complete resection rate, post-ER bleeding rate, perforation rate, and outcome in both groups, and analyzed the survival outcomes of 84 patients who were followed for more than 5 years (mean, 101.9 months) according to the indication for ER [expanded criteria for curative (n = 52) vs. criteria for noncurative (n = 32)]. Results Complete resection rates for EMR and ESD were 54 % (15 of 28) and 89 % (86 of 97), respectively, with that for ESD being significantly higher (p < 0.01). Outcomes after ER were as follows: among 52 cases of UEGC meeting the expanded criteria, additional surgical resection was performed in 11 cases of incomplete resection. No local recurrence or lymph node metastasis was observed. Forty-eight patients who were simply surveyed clinically (93.6 +/- 38.4 months) after ER survived without recurrence; the remaining 7 patients died of other causes. Among the 32 cases of UEGC meeting the criteria for noncurative resection, additional surgical resection was performed in 13 cases. Among the 19 follow-up cases (108.3 +/- 38.7 months), death due to metastasis of the primary disease occurred in 3 cases, death from other causes occurred in 5 cases, and local residual submucosal recurrence occurred in 1 case. Conclusions ESD is a useful technique for complete resection as a total excisional biopsy compared with EMR and radical cure of UEGCs meeting the expanded criteria.