Long-Term Safety of Delayed Surgery After Upfront Endoscopic Resection for Early Gastric Cancer: A Propensity Matched Study

Background: When non-curative resection is confirmed after endoscopic resection (ER) of early gastric cancer (EGC), delayed surgery is recommended because it provides favorable survival outcomes. Long-term outcome after surgery of EGC with or without previous ER has not been evaluated. Objective: The aim of this study was to compare the long-term oncologic safety between primary surgery and delayed surgery after ER. Methods: Patients who had undergone curative surgery (R0) for EGC were included and were divided into primary and delayed surgery groups. Primary surgery was defined as gastrectomy without ER for EGC, whereas delayed surgery was defined as additional curative gastrectomy due to non-curative resection after ER; an average delay of 21.5 days (range 1–195) was observed. Propensity score matching was performed. The primary outcome was overall survival (OS) and the secondary outcomes were cancer-specific survival (CSS) and disease-free survival (DFS). Results: After propensity score matching, 1439 patients were included, of whom 1042 (72.4%) were in the primary surgery group and 397 (27.6%) were in the delayed surgery group. The OS (hazard ratio [HR] 0.87, 95% confidence interval [CI] 0.59–1.27; p = 0.459), CSS (HR 0.47, 95% CI 0.15–1.47; p = 0.196), and DFS (HR 0.54, 95% CI 0.15–1.90; p = 0.334) were not different. Conclusions: The long-term outcomes of delayed surgery after non-curative ER for EGC were non-inferior to primary surgery. Therefore, an attempt for ER of EGC that satisfies the absolute and expanded indication seems justified for preventing gastrectomy. In case of non-curative resection after ER, additional delayed surgery should be performed.