Cediranib with mFOLFOX6 versus bevacizumab with mFOLFOX6 as first-line treatment for patients with advanced colorectal cancer: a double-blind, randomized phase III study (HORIZON III).

PURPOSE To compare the efficacy of cediranib (a vascular endothelial growth factor receptor tyrosine kinase inhibitor [VEGFR TKI]) with that of bevacizumab (anti-VEGF-A monoclonal antibody) in combination with chemotherapy as first-line treatment for advanced metastatic colorectal cancer (mCRC). PATIENTS AND METHODS HORIZON III [Cediranib Plus FOLFOX6 Versus Bevacizumab Plus FOLFOX6 in Patients With Untreated Metastatic Colorectal Cancer] had an adaptive phase II/III design. Patients randomly assigned 1:1:1 received mFOLFOX6 [oxaliplatin 85 mg/m(2) and leucovorin 400 mg/m(2) intravenously followed by fluorouracil 400 mg/m(2) intravenously on day 1 and then continuous infusion of 2,400 mg/m(2) over the next 46 hours every 2 weeks] with cediranib (20 or 30 mg per day) or bevacizumab (5 mg/kg every 14 days). An independent end-of-phase II analysis concluded that mFOLFOX6/cediranib 20 mg met predefined criteria for continuation; subsequent patients received mFOLFOX6/cediranib 20 mg or mFOLFOX6/bevacizumab (randomly assigned 1:1). The primary objective was to compare progression-free survival (PFS). RESULTS In all, 1,422 patients received mFOLFOX6/cediranib 20 mg (n = 709) or mFOLFOX6/bevacizumab (n = 713). Primary analysis revealed no significant difference between arms for PFS (hazard ratio [HR], 1.10; 95% CI, 0.97 to 1.25; P = .119), overall survival (OS; HR, 0.95; 95% CI, 0.82 to 1.10; P = .541), or overall response rate (46.3% v 47.3%). Median PFS and OS were 9.9 and 22.8 months for mFOLFOX6/cediranib and 10.3 and 21.3 months for mFOLFOX6/bevacizumab. The PFS upper 95% CI was outside the predefined noninferiority limit (HR < 1.2). Common adverse events with more than 5% incidence in the cediranib arm included diarrhea, neutropenia, and hypertension. Cediranib-treated patients completed fewer chemotherapy cycles than bevacizumab-treated patients (median 10 v 12 cycles). Patient-reported outcomes (PROs) were significantly less favorable in cediranib-treated versus bevacizumab-treated patients (P < .001). CONCLUSION Cediranib activity, in terms of PFS and OS, was comparable to that of bevacizumab when added to mFOLFOX6; however, the predefined boundary for PFS noninferiority was not met. The cediranib safety profile was consistent with previous studies but led to less favorable PROs compared with bevacizumab. Investigation of oral TKIs in CRC continues.