Background The purpose of this study was twofold: (1) to compare S-1 with infusional 5-fluorouracil (FU) to determine which would be a better partner of paclitaxel (PTX), and (2) to compare a concurrent strategy with a sequential one, the latter strategy being the one that is widely used in Japanese general practice. Methods The 161 eligible patients were randomized into four arms to receive the following regimens: A (sequential), intravenous 5-FU at 800 mg/m 2 for 5 days every 4 weeks followed by weekly PTX at 80 mg/m 2 ; B (sequential), S-1 at 80 mg/m 2 for 4 weeks and 2-week rest followed by PTX; C (concurrent), intravenous 5-FU at 600 mg/m 2 for 5 days and weekly PTX at 80 mg/m 2 every 4 weeks; and D (concurrent), S-1 for 14 days and PTX at 50 mg/m 2 on days 1 and 8 every 3 weeks. The primary endpoint was the overall survival (OS) rate at 10 months. Results The ten-month OS rates in arms A, B, C, and D were 63, 65, 61, and 73%, respectively. The OS was best in the concurrent S-1/PTX arm, with a mean survival time of
123Gastric Cancer (2012) 15:363-369 DOI 10.1007 15.4 months, but no significant difference was observed between the four arms. Response rates were higher in the concurrent arms than in the sequential arms. Conclusion Our study did not show sufficient prolongation of survival with the concurrent strategy to proceed to a phase-III trial; however, the sequential arms showed survival comparable to that in the concurrent arms, with less toxicity. In patients who are ineligible for cisplatin (CDDP), sequential treatment starting with S-1 and proceeding to PTX would be a good alternative strategy, considering quality of life (QOL) and the cost-benefits of an oral agent as first-line treatment.