515 Background: Oral anticancer drug, UFT, has recently been reported to improve both overall survival (OS) and relapse-free survival (RFS) for patients with stage III colorectal cancer in Japan. We conducted a randomized clinical trial to compare S-1 with UFT for adjuvant therapy in patients with stage III colorectal cancer. Methods: Patients with stage III colorectal cancer (PS, 0 to 1; age, 20 to 80 years) were randomized to take either UFT (400mg/m2/day, 5days per week) or S-1 (80mg/m2/day, 28days per 6weeks) for 1 year started within 6 weeks following curative resection. The primary endpoint was to investigate relations of relapse-free survival (RFS) and tissue mRNA levels of 5-FU metabolism-related enzymes. The secondary endpoint was over all survival (OS) and safety. Results: Between July 2005 and February 2008, a total of 100 patients were registered from 21 centers. There were no clear intergroup differences in background factors. When patients who relapsed within 1 year postoperatively were excluded, the percentages of patients who took expected dosage of medicine were 73% (30/41) and 72% (33/46) in UFT and S-1 groups, respectively. Grade 3 liver dysfunction (4%) or diarrhea (4%) occurred in UFT group, whereas grade3 to 4 myelosuppression (4%), diarrhea (4%), stomatitis (2%) were observed in S-1 group, but all of them were reversible. With a median follow-up period of 1,250 days, OS in patients of UFT and S-1 groups were 86.6% and 95.9%, respectively (p=0.06). The number of patients who relapsed in rectal cancer was significantly higher in UFT group (11/25) than in S-1 group (4/21, p<0.01). In a multivariate analysis, dihydropyrimidine dehydrogenase (DPD) mRNA level was shown as a factor of the recurrences. And S-1 improved RFS in patients with rectum cancer with high orotate phosphoribosil transferase (OPRT) mRNA level. Conclusions: Oral anticancer drug, S-1, was no less effective than UFT as an adjuvant therapy for stage III colorectal cancer. Tissue levels of both DPD and OPRT mRNA were important parameters for postoperative prognosis in patients with stage III colorectal cancer who take oral 5-FU derivatives for adjuvant therapy. No significant financial relationships to disclose.