Clinical outcomes of endoscopic resection for remnant gastric cancer: a multicenter prospective cohort study.

BACKGROUND

While the incidence of early-stage remnant gastric cancer (RGC) is increasing, evidence for the validity of endoscopic resection (ER) for RGC remains limited. We aimed to clarify outcomes of ER for RGC.

METHODS

This is the second analysis of a nationwide prospective cohort study of ER for early gastric cancer (EGC) at 41 Japanese centers from July 2010 to June 2012. We compared short-term and long-term outcomes of 355 patients with 369 RGCs and 8460 patients with 9394 primary EGCs in the naïve stomach. We calculated 5-year overall survival (OS) and disease-specific survival (DSS) rates. Hazard ratios (HRs) for all-cause mortality were estimated using a Cox regression model; patients with curatively resected primary EGC were the reference group.

RESULTS

For RGCs and primary EGCs, median procedure times were 87 and 77 min (p < 0.001); en-bloc resection rates were 96.7 and 99.5% (p < 0.001); R0 resection rates were 83.5 and 93.3% (p < 0.001); intraoperative perforation rates were 4.2 and 1.9% (p = 0.002); and delayed perforation rates were 1.1 and 0.3% (p = 0.018), respectively. 5-year OS (83.1% vs 89.5%, p < 0.001) and DSS (98.5% vs 99.6%, p = 0.002) were lower in RGC cases than primary EGC cases. HRs for all-cause mortality against curatively resected primary EGC were 1.19 (95% CI 0.85-1.66) for curatively resected RGC, 1.37 (1.12-1.60) for non-curatively resected primary EGC, and 1.87 (1.21-2.90) for non-curatively resected RGC.

CONCLUSIONS

Although ER for RGC showed slightly inferior short-term outcomes to ER for primary EGC, long-term outcomes were comparable under the curative resection.