In Japan, colorectal cancer remains the second most common cancer in both genders and the most common cause of cancer mortality in women . The rate of recurrence of stage III colorectal cancer is reportedly 30.8% , and adjuvant chemotherapy is important in the treatment of stage III cancer after curative resection. In Western countries, oxaliplatin (Ox)-based regimens in addition to fluorouracil (FU) and leucovorin (LV) (FOLFOX (Ox + FU + LV), FLOX (Ox + FU(, or XELOX (Ox + capecitabine)) have been established as the gold standard adjuvant chemotherapy for stage III colon cancer based on the results of three large randomized controlled phase III studies conducted after 2000 [3–5]. However, favorable results using an FU-based regimen without Ox were reported in a Japanese randomized controlled study (JCOG 0205) . This study compared uracil and tegafur (UFT)/LV with 5-FU/LV as an adjuvant chemotherapy for patients with stage III colon cancer and demonstrated that the three-year disease-free survival (DFS) was 77.8% in the UFT/LV arm and 79.3% in the 5-FU/LV arm, respectively. This result was equal to the experimental arm of the MOSAIC trial using Ox (72.2% in stage III patients only) , the NSABP C-07 trial (71.8% including stage II patients)  and the XELOXA trial (70.9%) . Therefore, the FU-based regimen has been widely accepted as an adjuvant chemotherapy regimen for stage III colon cancer in Japan, and the Japanese guidelines recommend 5-FU/LV, UFT/LV and capecitabine as equal treatments to FOLFOX .
Oral FU is an effective, better tolerated, and convenient chemotherapy regimen compared to Ox-based regimens, as it does not require the use of a central infusion port system and there is less need to visit the outpatient clinic. Therefore, oral FU is a valuable treatment regimen for both patients and medical staffs. The equality of UFT/LV or capecitabine to 5-FU/LV has already been confirmed in large randomized trials [7, 8]. S-1 (TS-1, Taiho Pharmaceutical, Tokyo, Japan) is an oral preparation evolved from UFT, which combines tegafur (a prodrug that is converted by cells to fluorouracil), gimeracil (an inhibitor of dihydropyrimidine dehydrogenase, which degrades fluorouracil), and oteracil (which inhibits the phosphorylation of fluorouracil in the gastrointestinal tract, thereby reducing the gastrointestinal toxic effects of fluorouracil) at a molar ratio of 1:0.4:1. Two Japanese phase II trials of S-1 monotherapy for chemotherapy-naïve patients with metastatic colorectal cancer showed that the response rates were 35.5% and 39.5%, similar to the results from the joint study of UFT/LV in the United States and Japan (36.4% in Japan and 34.1% in the United States). In addition, the efficacy of S-1 in an adjuvant setting has been demonstrated in a Japanese phase III trial in patients with stage II or III gastric cancer (ACTS-GC trial) . Currently, two randomized clinical trials are ongoing in Japan to compare the effectiveness of S-1 as an adjuvant chemotherapy with that of UFT/LV or capecitabine in patients with stage III colon cancer [10, 11].
The standard schedule of S-1 monotherapy is an oral six-week cycle (four-weeks administration followed by two-weeks rest). However, the completion rate of scheduled treatment in the ACTS-GC trial was reported to be 65.8% , which is relatively low, and, moreover, some patients received a modified schedule, such as a three-week cycle (two-weeks administration followed by a one-week rest). Some studies indicated that patients who discontinued treatment early in the adjuvant setting for colon cancer showed poor survival [12, 13]. Therefore, it is an urgent necessity to establish the optimal treatment schedule in order to reduce the withdrawal rate of patients receiving S-1 therapy. Only one randomized phase II trial, in which a three-week cycle was compared with a standard six-week cycle has been published. An increase in compliance and a decrease in toxicity was observed in the patients treated with a three-week cycle, and this suggested its use for patients with stage III/IV squamous cell carcinoma of the head and neck after definitive treatment . In this trial, the completion rate along the planned schedule and dose was 40% and 20% in patients treated with a three-week cycle and a six-week cycle and severe adverse grade 3 events were observed in 14% and 27%, respectively.
Based on these findings, we designed the present randomized phase II trial to compare a six-week cycle with a three-week cycle, and to establish the optimal schedule of S-1 adjuvant therapy for patients with stage III colon cancer after curative resection.