Su1251 DIFFERENT RISK FACTORS BETWEEN EARLY AND LATE CANCER RECURRENCES IN PATIENTS WITH NO ADDITIONAL TREATMENT AFTER NON-CURATIVE ENDOSCOPIC SUBMUCOSAL DISSECTION FOR EARLY GASTRIC CANCER
Shinya Yamada,Waku Hatta,Takuji Gotoda,Tsuneo Oyama,Noboru Kawata,Akiko Takahashi,Yoshikazu Yoshifuku,Shu Hoteya,Masahiro Nakagawa,Masaaki Hirano,Mitsuru Esaki,Mitsuru Matsuda,Ken Ohnita,Ryo Shimoda,Motoyuki Yoshida,Osamu Dohi,Jun Takada,Keiko Tanaka,Tsuyotoshi Tsuji,Hirotaka Ito,Yoshiaki Hayashi,Tooru Shimosegawa
DOI: https://doi.org/10.1016/j.gie.2018.04.1638
IF: 10.396
2018-01-01
Gastrointestinal Endoscopy
Abstract:Although additional surgery is recommended for all patients with non-curative endoscopic submucosal dissection (ESD) for early gastric cancer (EGC) in the Japanese guidelines because of the risk of lymph node metastasis (LNM), near half of them are followed up with no additional treatment after ESD in the clinical setting. Our prior study demonstrated that the risk factor for cancer recurrence in such patients was lymphatic invasion. However, it is still unknown whether risk factors for cancer recurrence is different between early and late ones. Thus, the aim of this study was to identify the predictors of early and late cancer recurrences in these patients. In this multi-center retrospective study collecting data from 19 Japanese institutions between January 2000 and August 2011, we enrolled patients with no additional treatment after non-curative ESD. Cancer recurrence after non-curative ESD was evaluated by using Kaplan-Meier method. In addition, we investigated the risk factors for early and late cancer recurrences, separately, after non-curative ESD by using Cox proportional hazard regression analysis. The estimated factors were tumor size of >30 mm, deep submucosal invasion, undifferentiated-type, lymphatic invasion, venous invasion, ulceration (scar), and positive vertical margin. In this study, cancer recurrence is defined as tumor relapse in the lymph node and/or other organs after ESD. In addition, early and late cancer recurrences were defined as those within and after 24 months, respectively, after non-curative ESD. A total of 905 patients, who were followed up with no additional treatment after non-curative ESD, with a median follow-up duration of 64 months were analyzed in this study. The 5-year cancer recurrence rate after ESD in such patients was 3.0%. The early and late cancer recurrences were shown in 17 and 10 patients, respectively. Multivariate Cox analysis, adjusted by age, gender, location, and so on, revealed that lymphatic invasion (hazard ratio (95% confidence interval) =6.02 (1.73-21.1), p =0.005) was the independent risk factor for the early cancer recurrence, whereas the other factors including venous invasion (hazard ratio (95% confidence interval) =1.31 (0.42-4.09), p =0.65) was not. On the other hand, the independent risk factor for late cancer recurrence was venous invasion (hazard ratio (95% confidence interval) =5.08 (1.03-25.0), p =0.046) in the multivariate analysis, and lymphatic invasion tended to be associated with late cancer recurrence (hazard ratio (95% confidence interval) =4.25 (0.90-20.2), p =0.069). In this multicenter study with the largest cohort, we found different risk factors between early and late cancer recurrences, which may contribute to the individualized follow-up strategies in patients with no additional treatment after non-curative ESD for EGC.