249 Background: Thoracoscopic esophagectomy (TE) as minimally invasive esophagectomy for thoracic esophageal cancer has become widely prevalent around the world over the past decade. However, results from a large-scale, multicenter randomized controlled trial that compared long-term survival as the primary endpoint between TE and conventional open transthoracic esophagectomy (OE) have not yet been demonstrated. We conducted a multicenter randomized phase III study (JCOG1409) to confirm the non-inferiority of TE to OE in terms of overall survival (OS) for thoracic esophageal cancer (UMIN000017628). Methods: Eligible patients with clinical stage I-III, excluding T4, thoracic esophageal squamous cell carcinoma were randomized to undergo either OE or TE. The primary endpoint was OS, with secondary endpoints including relapse-free survival (RFS), the proportion of patients achieving R0 resection, the proportion of patients needing conversion from TE to OE, adverse events, the proportion of patients requiring re-operation, changes in postoperative respiratory dysfunction, and postoperative quality-of-life score. Non-inferiority would be confirmed if the upper limit of the confidence interval (CI) for the hazard ratio (HR) does not exceed 1.44, the predefined non-inferiority margin. Results: We randomized 300 patients, assigning 150 to the TE group and 150 to the OE group between May 2015 and June 2022. The second planned interim analysis with 64 OS events was conducted in June 2023. Median follow-up (interquartile range) was 2.6 (1.4-4.9) years. Three-year OS was 82.0% (95% CI, 73.8%-87.8%) in the TE group, and 70.9% (61.6%-78.4%) in the OE group. The Data and Safety Monitoring Committee recommended terminating the trial and publishing the results because the non-inferiority of TE to OE in terms of OS was demonstrated after adjusting for multiplicity (HR 0.64; 98.8% CI, 0.34-1.21 [<1.44], one-sided p-value for non-inferiority = 0.000726 < 0.00616). Three-year RFS in the TE group was markedly better than that in the OE group (72.9% vs. 61.9%) with HR of 0.68 (95% CI, 0.46-1.01). R0 resection proportions were 95.3% in the TE group and 90.0% in the OE group. One patient in the TE group needed conversion from TE to OE intraoperatively. The postoperative overall morbidity was similar between the two groups, but the proportion of re-operation was 2.0% in the TE group and 4.1% in the OE group. Respiratory dysfunction proportion at 3 months after surgery were significantly lower in the TE group than in the OE group. Conclusions: TE was shown to be a standard treatment for patients with clinical stage I-III thoracic esophageal cancer. Clinical trial information: UMIN000017628 .