Chemotherapy is the standard of care for incurable advanced gastric cancer. Whether the addition of gastrectomy to chemotherapy improves survival for patients with advanced gastric cancer with a single non-curable factor remains controversial. We aimed to investigate the superiority of gastrectomy followed by chemotherapy versus chemotherapy alone with respect to overall survival in these patients.

We did an open-label, randomised, phase 3 trial at 44 centres or hospitals in Japan, South Korea, and Singapore. Patients aged 20–75 years with advanced gastric cancer with a single non-curable factor confined to either the liver (H1), peritoneum (P1), or para-aortic lymph nodes (16a1/b2) were randomly assigned (1:1) in each country to chemotherapy alone or gastrectomy followed by chemotherapy by a minimisation method with biased-coin assignment to balance the groups according to institution, clinical nodal status, and non-curable factor. Patients, treating physicians, and individuals who assessed outcomes and analysed data were not masked to treatment assignment. Chemotherapy consisted of oral S-1 80 mg/m² per day on days 1–21 and cisplatin 60 mg/m² on day 8 of every 5-week cycle. Gastrectomy was restricted to D1 lymphadenectomy without any resection of metastatic lesions. The primary endpoint was overall survival, analysed by intention to treat. This study is registered with UMIN-CTR, number UMIN000001012.

Between Feb 4, 2008, and Sept 17, 2013, 175 patients were randomly assigned to chemotherapy alone (86 patients) or gastrectomy followed by chemotherapy (89 patients). After the first interim analysis on Sept 14, 2013, the predictive probability of overall survival being significantly higher in the gastrectomy plus chemotherapy group than in the chemotherapy alone group at the final analysis was only 13·2%, so the study was closed on the basis of futility. Overall survival at 2 years for all randomly assigned patients was 31·7% (95% CI 21·7–42·2) for patients assigned to chemotherapy alone compared with 25·1% (16·2–34·9) for those assigned to gastrectomy plus chemotherapy. Median overall survival was 16·6 months (95% CI 13·7–19·8) for patients assigned to chemotherapy alone and 14·3 months (11·8–16·3) for those assigned to gastrectomy plus chemotherapy (hazard ratio 1·09, 95% CI 0·78–1·52; one-sided p=0·70). The incidence of the following grade 3 or 4 chemotherapy-associated adverse events was higher in patients assigned to gastrectomy plus chemotherapy than in those assigned to chemotherapy alone: leucopenia (14 patients [18%] vs two [3%]), anorexia (22 [29%] vs nine [12%]), nausea (11 [15%] vs four [5%]), and hyponatraemia (seven [9%] vs four [5%]). One treatment-related death occurred in a patient assigned to chemotherapy alone (sudden cardiopulmonary arrest of unknown cause during the second cycle of chemotherapy) and one occurred in a patient assigned to chemotherapy plus gastrectomy (rapid growth of peritoneal metastasis after discharge 12 days after surgery).

Since gastrectomy followed by chemotherapy did not show any survival benefit compared with chemotherapy alone in advanced gastric cancer with a single non-curable factor, gastrectomy cannot be justified for treatment of patients with these tumours.

The Ministry of Health, Labour and Welfare of Japan and the Korean Gastric Cancer Association.