A randomized phase II trial of systemic chemotherapy with or without trastuzumab followed by surgery in HER2 positive advanced gastric or esophagogastric junction adenocarcinoma with extensive lymph node metastasis: 3-year follow-up data of the Japan

393 Background: We previously reported better objective and pathological response with the addition of trastuzumab to preoperative chemotherapy in the randomized phase II trial comparing the efficacy of systemic chemotherapy with or without trastuzumab followed by surgery in HER2 positive advanced gastric or esophagogastric junction adenocarcinoma with extensive lymph node metastasis. Here, we report 3-year follow-up data. Methods: Eligible patients with HER2-positive gastric cancer having extensive lymph node metastasis were randomized to receive preoperative chemotherapy with S-1/cisplatin (SP: S-1 80–120 mg/body on days 1–14, and cisplatin 60 mg/m2 on day 1, q21 days, 3-4 courses) in arm A or SP plus trastuzumab (first course 8 mg/kg followed by 6 mg/kg, day 1 on each course) in arm B. After gastrectomy, adjuvant chemotherapy with S-1 was added for 1 year in both arms. The primary endpoint was overall survival (OS), and the sample size was planned to be 130 patients in total, expecting a 10% increase in the 3-year OS (70% vs 80%) with a 1-sided alpha of 0.2, a power of 0.75. Results: The study was terminated in March 2021 due to slow patient accrual. In total, 46 patients were randomized to either arm A (22 patients) or arm B (24 patients). Patient characteristics were well balanced between the arms. Planned preoperative chemotherapy was completed in 20 patients (90.9%) in arm A and in 23 patients (95.8%) in arm B. Objective response rate tended to be higher in arm B (66.7% [16/24]) than in arm A (36.4% [8/22], p=0.08), and R0 resection was achieved in 20 patients (90.9%) in arm A and 22 patients (91.7%) in arm B. The adjuvant chemotherapy was completed in 54.6% (12/22) in arm A and in 54.2% (13/24) in arm B. With the median observation period of 4.1 years, the 3-year OS was 68.2% in arm A and was 78.9% in arm B with a hazard ratio of 0.698 (95% confidence interval (CI), 0.221-2.209). The 3-year progression-free survival of arm A and arm B was 39.3% and 62.3%, respectively, with a hazard ration of 0.562 (95% CI, 0.229-1.377). Conclusions: Preoperative chemotherapy with SP plus trastuzumab was feasible, and tended to improve the objective response rate, OS and progression-free survival. An integrated analysis with other clinical trials evaluating perioperative trastuzumab is being planned. Clinical trial information: s031180006.