Hepatic resection following selective internal radiotherapy in the FOXFIRE clinical trial: survival, safety, and histopathology


Session type:


Pradeep S. Virdee1,Helen Winter2,Joe Rassam3,Rob Goldin3,Harpreet S. Wasan4,Ricky A. Sharma5
1Centre for Statistics in Medicine, University of Oxford,2Green Templeton College, University of Oxford,3Centre for Pathology, Imperial College,4Imperial College Healthcare NHS Trust and Imperial College, Hammersmith Hospital,5University College London Hospitals Biomedical Research Centre, UCL Cancer Institute



Colorectal cancer (CRC) commonly metastasises to the liver and is a leading cause of cancer-related death. The FOXFIRE trial compared the safety and efficacy of radiosensitising chemotherapy (OxMdG: oxaliplatin, 5-fluorouracil and folic acid) with selective internal radiotherapy (SIRT) using yttrium-90 resin microspheres (SIR-Spheres®; Sirtex Medical Limited) to OxMdG alone as first-line management for liver-dominant metastatic CRC. In patients downsized to potentially-curative hepatic resection, we explored survival, safety, and histopathological findings.


FOXFIRE (ISRCTN83867919) was an open-label, multicentre, randomised (1:1) trial of 12 cycles of OxMdG with or without SIRT (Wasan HS et al. Lancet Oncol 2017). Eligible patients provided written informed consent and were considered untreatable by surgical resection or local ablation. Suitability for surgery was reassessed after 3 months of treatment. Surgical complications were graded using Dindo D et al. Annals of Surgery 2004. Cox models and a 5% significance level were used.


FOXFIRE randomised 364 patients: 182 per treatment group. Seventy-one (20%) underwent hepatic resections following first-line treatment: 38 (21%) in the OxMdG+SIRT and 33 (18%) in the OxMdG group. Among those resected, 22 in the OxMdG+SIRT and 19 in the OxMdG group had primary tumour in situ. Common surgeries performed were right hepatectomy (n=22) and segmentectomy (n=12). Among those resected, overall survival from the time of resection did not differ significantly between groups (OxMdG median=25.2 months; OxMdG+SIRT median=21.9 months; HR=1.55, 95% CI=0.83-2.89). Twenty patients had grade I/II complications: 11 in the OxMdG and 9 in the OxMdG+SIRT group. Less viable tumour was histologically observed in patients receiving SIRT. Zonal analysis demonstrated that median microsphere density was highest at the tumour periphery and lowest in non-neoplastic tissue.


This study reports that hepatic resection following SIRT with OxMdG chemotherapy has an acceptable safety profile and demonstrates the histopathological distribution of microspheres in liver metastases.