SIRFLOX: Randomized trial comparing first-line mFOLFOX6 +/- bevacizumab versus mFOLFOX6 + selective internal radiation therapy (SIRT) +/- bevacizumab in patients with metastatic colorectal cancer (mCRC)-analysis by presence or absence of extra-hepatic metastases and bevacizumab treatment.

Abstract

Background: Liver metastases are the dominant site of disease in mCRC and the dominant cause of death. The SIRFLOX study was designed to assess the efficacy and safety of combining FOLFOX chemotherapy (+/- bevacizumab) with SIRT using yttrium-90 (Y-90) resin microspheres as first-line treatment of patients with liver metastases from mCRC. Method(s): SIRFLOX was an international, multi-centre, open-label, RCT in chemotherapy-naive patients with non-resectable, liver-only or liver-dominant (liver plus lung and/or lymph node metastases) mCRC. Arm A: mFOLFOX6 (+/- bevacizumab, at investigator discretion) was compared to arm B: mFOLFOX6 + SIRT (SIR-Spheres; Sirtex) administered once with cycle 1 (+/- bevacizumab, at investigator discretion) until disease progression. The primary endpoint was Progression-Free Survival (PFS) using RECIST v1.0 by independent central imaging review in the intention-to-treat (ITT) cohort. PFS in the liver, as defined by Cumulative Incidence of Liver Progression (Liver CIP), was assessed by Competing Risk analysis. Stratification variables included presence of extra-hepatic metastases (liver-only vs. liver-dominant), degree of liver involvement (<=25% vs. >25%), and ITT with bevacizumab (with vs. no bevacizumab). Result(s): From October 2006 to April 2013, 530 patients were randomised (arm A, n = 263; arm B, n = 267); 212 (40%) had extra-hepatic metastases; 292 (55%) were stratified to receive bevacizumab. Median follow-up was 36.1 months. The median overall PFS was 10.2 vs. 10.7 months in arms Avs. B, respectively (hazard ratio [HR]: 0.93; 95% CI 0.77-1.12; p = 0.429) by Kaplan-Meier analysis. The median Liver PFS was 12.6 vs. 20.5 months in arm Avs. B (HR: 0.69; 95% CI 0.55-0.90; p = 0.002). Median Liver PFS was 12.4 vs. 21.1 months in arm Avs. B (HR: 0.64; 95% CI 0.48-0.86; p = 0.003) for patients with liver-only metastases, and 12.6 vs. 16.7 months (HR: 0.77, 95% CI 0.54-1.09; p = 0.147) for those with liver and extra-hepatic metastases. Median Liver PFS was 10.6 vs. 18.9 months in arm Avs. B (HR: 0.69; 95% CI 0.50-0.96; p = 0.028) for patients with ITT for no bevacizumab, and 12.7 vs. 21.0 months (HR: 0.69, 95% CI 0.50-0.94; p = 0.018) for those with ITT to receive bevacizumab. Overall response rate ( partial response + complete response) was 68.1% vs. 76.4% in arm Avs. B, respectively (p = 0.113). Hepatic response rate was 68.8% vs. 78.7% in arm Avs. B (p = 0.042), including complete response rate 1.9% vs. 6.0% (p = 0.02). All-causality grade >=3 adverse events were noted in 73.3% vs. 85.4% (not significant [NS]) of patients in arm Avs. B. Grade 5 events occurred in 1.9% vs. 3.7% (NS) in arm Avs. B. The most common grade >=3 chemotherapy-associated toxicity was neutropenia (28.5% vs. 40.7%; p = 0.004); the most common grade >=3 SIRT-associated event was gastric or duodenal ulcers (0.0% vs. 3.7%; p = 0.001) in treatment arms A vs. B, respectively. Conclusion(s): In first-line treatment of patients with non-resectable CRC liver metastases, the addition of SIRT to a standard chemotherapy regimen failed to improve overall PFS. However, major gains in PFS in the Liver were achieved. The addition of SIRT was associated with acceptable toxicity.