Objectives: Platinum-based combination chemotherapy is the standard postoperative adjuvant treatment for pathological stage II/III non-small cell lung cancer (NSCLC). Oral S-1 therapy has good efficacy and relatively low toxicity for the treatment of advanced NSCLC. We investigated whether long-term S-1 monotherapy is also useful as an adjuvant therapy after surgery in patients with NSCLC. Patients and methods: We conducted a phase II randomized open-label multi-institutional study in patients with pathological stage II/IIIA NSCLC (7 th TNM classification) who underwent complete resection from 2009 to 2013. The primary endpoint, the 2-year disease-free survival (DFS) rate, was evaluated using the Bayesian method. Eligible patients were randomly assigned to two arms: oral S-1 monotherapy (S-1 arm) and S-1 plus cisplatin combination therapy followed by S-1 (S-1 plus cisplatin arm) both for a total of 1 year. Results: A total of 70 and 71 patients were enrolled in S-1 arm and S-1 plus cisplatin arm, respectively. The 2-year DFS rates were 52% (95% confidence interval [CI], 0.40–0.63) and 61% (95% CI, 0.48–0.70) for S-1 arm and S-1 plus cisplatin arm, respectively. Both arms met the primary endpoint. Neither DFS nor OS was significantly different between the arms (log-rank test: P = 0.1695 and P = 0.8684, respectively). The main G3/4 adverse events were loss of appetite and anemia (S-1 vs. S-1 plus cisplatin: 4.3% vs. 11.6% and 0% vs. 5.8%, respectively). The treatment completion rate did not differ between the two arms (S-1 vs. S-1 plus cisplatin: 45.7%, 95% CI, 41.9–66.3% vs. 43.5% 95% CI, 44.0–68.4%). Conclusions: Long-term adjuvant chemotherapy with S-1 was a feasible and promising treatment for patients with completely resected NSCLC, regardless of cisplatin addition. S-1 monotherapy should be investigated further, based on its low toxicity and practical convenience.
!!!All Science Journal Classification (ASJC) codes