From the late 1980s to the early 1990s, the Gastrointestinal Oncology Study Group of the Japan Clinical Oncology Group (GIOSG/JCOG) conducted several phase II studies, some of which evaluated oral fluoropyrimidines and others of which introduced Western regimens to Japanese patients. Thereafter, in the phase III study JOCG9205 comparing 5-fluorouracil (5-FU), 5-FU plus cisplatin (CDDP) (FP), and uracil and tegafur (UFT) plus mitomycin (UFTM), neither FP nor UFTM showed a survival benefit over 5-FU alone. Whereas irinotecan (CPT-11) and S-1 (new oral fluoropyrimidine) were developed with promising action against gastric cancer in the late 1990s, these agents cannot be used for patients with impaired oral intake and bowel passage caused by severe peritoneal metastasis. Sequential methotrexate (MTX) and 5-FU (MF) therapy showed substantial action against peritoneal metastasis. Thus, GIOSG/JCOG followed different treatment strategies according to the presence or absence of severe peritoneal metastasis. The phase III study JCOG9912, comparing 5-FU, CPT-11 plus CDDP, and S-1, showed a highly significant noninferiority of S-1 to 5-FU in overall survival associated with acceptable toxicities and concluded that S-1 should be considered for the standard chemotherapy for gastric cancer without severe peritoneal metastasis. For patients with severe peritoneal metastasis, the phase III study JCOG0106 compares MF to 5-FU. In that study, patient enrollment has been completed and a final analysis is planned at the end of 2008. The randomized phase II study JCOG0407 compares the best available 5-FU with weekly paclitaxel after failure in first-line chemotherapy containing 5-FU.