Phase II study of systemic chemotherapy with S-1 plus oxaliplatin followed by surgery in patients with cT3-T4a and/or node-positive advanced adenocarcinoma of the esophagogastric junction: Primary endpoint results of the ESOX trial.

Abstract

214 Background: Perioperative chemotherapy has been suggested to be beneficial in patients with advanced gastric cancer. Based on the German FLOT-4 study, the FLOT regimen is considered as the new standard perioperative chemotherapy regimen for resectable gastric cancer in Europe. However, most clinical trials have included few or no cases of esophagogastric junction (EGJ) cancer, because of the difference in surgical procedures. The benefit of neoadjuvant chemotherapy in patients with advanced adenocarcinoma of the EGJ thus remains controversial in Japan. Methods: We conducted a phase II study in 13 Japanese institutions. Eligible patients had histopathologically confirmed adenocarcinoma of the EGJ (Siewert type I or II with invasion of the esophagus ≥30 mm) with clinical T3/4a and/or node-positive on imaging findings, who required thoracic surgery to achieve R0 resection. Patients received three cycles of S-1 (80 mg/m2) twice a day on days 1–14 and oxaliplatin (130 mg/m2) (SOX) on day 1 of a 21-day cycle before surgery. The primary endpoint was R0 resection rate and the secondary endpoints were overall response rate, pathological complete response (pCR) rate, 2-year and 3-year disease-free survival, overall survival, and toxicity. The planned sample size was 50 patients based on an expected R0 resection rate of 85% and the threshold was 70%, with a one-sided alpha of 0.1 and power of 80%. Results: Fifty patients were enrolled in this study between June 2016 and April 2020. Totals of 21/29 and 7/4/22/8/8/1 had Siewert type I/II and clinical stage IIA/IIB/IIIA/IIIB/IIIC/IV disease, respectively. The completion rates for preoperative chemotherapy and surgery were 92% and 88%, respectively. Neoadjuvant therapy resulted in downstaging in 46% of patients (95% confidence interval (CI) 31.8%-60.7%). The pCR rate was 18% (95% CI 8.6%-31.4%) and the R0 resection rate was 82.0% (95% CI 68.6%-91.4%, 80% CI 73.1%-88.9%). Adverse events ≥ grade 3 during chemotherapy included thrombocytopenia (10%), neutropenia (8%), anemia (4%), anorexia (4%), nausea(2%), hypocalcemia(2%), hyponatremia(2%) and diarrhea (2%). Surgical morbidity was acceptable (Clavien-Dindo Grade IIIa surgical complications included anastomotic leakage (6.3%), pleural effusion (4.2%), thromboembolism (2.1%), and anastomotic infection (2.1%)). Conclusions: Neoadjuvant SOX met the primary endpoint of R0 resection rate 82% ( > 70%), with acceptable adverse effects and no impression on surgeries, suggesting that neoadjuvant SOX might be a new treatment strategy for patients with EGJ adenocarcinoma in Japan. Clinical trial information: 000020815.