Background: FOLFIRI and FOLFOX showed equivalent efficacy for metastatic colorectal cancer (mCRC), however their combination with bevacizumab (Bev) have not been directly compared. Methods: The WJOG4407G study was a randomized, open-label, phase III trial. Patients (pts) with previously untreated mCRC were randomized to receive either mFOLFOX6+Bev (group A; oxaliplatin 85mg/m², l-leucovorin 200mg/m², bolus 5-FU 400mg/m², infusional 5-FU 2,400mg/m², and Bev 5mg/kg; q2w) or FOLFIRI+Bev (group B; irinotecan 150mg/m²instead of oxaliplatin), stratified by institution, adjuvant chemotherapy, and liver limited disease. The primary objective was to investigate non-inferiority of B to A in progression free survival (PFS) assessed in the full analysis population (all randomized pts without major violation of eligibility criteria); non-inferiority margin of hazard ratio (HR) 1.25 based on the assumption of median PFS 10/11 months in A/B (power 0.85, 1-sided alpha 0.025). The secondary endpoints were response rate, overall survival (OS), safety, and quality of life (QOL). Results: From September 2008 to January 2012, 395 (A/B 198/197) of 402 enrolled pts were eligible for efficacy analysis. Data cut-off date was July 2013. Median PFS in A/B were 10.7/12.0 months (HR 0.905, 95% CI 0.723-1.133; p=0.003 for non-inferiority and p=0.427 for superiority; events in 78% pts). There was no interaction effect between the treatment and KRAS(codon12, 13) mutation status. Subsequent treatments were performed in 168/157 pts after A/B. Median OS in A/B were 28.9/31.8 months (HR 0.901, 95% CI 0.683-1.189; events in 50% pts), and response rates were 62.2/63.8%. The common grade 3 or 4 adverse events in A/B were leukopenia 4.5/11.3%, neutropenia 35.4/45.6%, diarrhea 5.1/8.7%, febrile neutropenia 1.5/5.1%, peripheral neuropathy 21.7/0%, and venous thromboembolism 1.5/6.2%. The QOL assessed by FACT-C (TOI-PFC) and FACT/GOG-Ntx were favorable for B throughout 18 months. Conclusions: FOLFIRI+Bev was non-inferior to mFOLFOX6+Bev with respect to PFS as 1st-line treatment for mCRC, and showed better trend in QOL. Clinical trial information: UMIN000001396.