ENDOSCOPIC SURVEILLANCE ACCORDING TO THE STATE OF HELICOBACTER PYLORI INFECTION FOLLOWING ENDOSCOPIC SUBMUCOSAL DISSECTION FOR EARLY GASTRIC CANCER : A MULTICENTER RETROSPECTIVE STUDY BY THE ESD HP STUDY GROUP, JEGS HOKURIKU BRANCH

Abstract

Background and Aims : The appropriate endoscopic surveillance interval for early detection of metachronous gastric cancer (MGC) following endoscopic submucosal dissection (ESD) of early gastric cancer (EGC) is unclear. This retrospective study investigated the characteristics of MGC following ESD of EGC according to the state of Helicobacter pylori (HP) infection, and the appropriate endoscopic surveillance interval from the viewpoint of stomach preservation.

Methods : This multicenter retrospective cohort study comprised 15 hospitals belonging to the Hokuriku branch of the Japanese Gastroenterological Endoscopy Society (JGES). Patients with EGC who had undergone initial ESD with endoscopic curative resection between 2002 and 2012 and who were followed up for >12 months were divided into three groups : HP-eradicated (successful eradication ; n=455) ; HP-persistent (unsuccessful or no eradication ; n=556) ; and HP-negative (negative in more than two HP tests ; n=291). The cumulative incidence of MGC in these three groups was compared. The proportion of MGC cases that underwent second ESD but did not meet the criteria of endoscopic curative resection as a result of the second ESD was compared according to the endoscopy intervals of 6 months (<9 months), 12 months (9-14 months) and >12 months (>14 months).

Results : During a mean period of 36.3 months, 88 patients developed MGC : 24 (5.3%) in the HP-eradicated group, 47 (8.5%) in the HP-persistent group, and 17 (5.9%) in the HP-negative group. The cumulative incidence of MGC did not differ among the three groups (log rank test : P=0.28), nor did their clinicopathological characteristics. Ten cases of MGC did not meet the criteria of endoscopic curative resection as a result of the second ESD, and the incidence of MGC did not significantly differ according to the endoscopic interval (P=0.61). The endoscopic interval of the 10 cases was distributed around 1 year [median 12.2 (5.5-17.4, range) months].

Conclusion : For the detection of MGC, it is desirable to take the same approach regardless of HP infection. It is anticipated that there will be a certain number of MGCs diagnosed in annual surveillance that do not meet the criteria of endoscopic curative resection as a result of the second ESD. Implementing surveillance at intervals shorter than 12 months may be worthy of consideration from the viewpoint of stomach preservation. To determine the appropriate endoscopic surveillance interval, further prospective studies with greater numbers of cases are warranted.