2 The procedure
2.1.1 Colorectal cancer is the second most common cancer in women and the third most common cancer in men in the UK. The liver is the most common site for metastases.
2.1.2 Curative treatment for patients with liver metastases from colorectal cancer may be provided by liver resection. However, fitness for surgery and the number, location and size of the metastases may dictate the use of alternative treatment options. These include systemic chemotherapy, radiotherapy, thermal ablation, chemoembolisation and selective internal radiation therapy. Radiofrequency ablation may be indicated as the primary treatment for liver metastases if the patient is unfit for surgery or in the treatment of postresection recurrence. It may also be used as an adjunct to hepatic resection to ablate small-volume disease in the future remnant liver.
2.2.1 A percutaneous or intraoperative approach may be used, with the patient under local or general anaesthesia. Needle electrodes are inserted into the target tumour area (or areas) using imaging guidance. A high-frequency alternating current is applied, resulting in heat generation, which causes localised coagulative necrosis and tissue destruction around the electrodes.
2.2.2 Several different types of RF electrodes are available.
Sections 2.3 and 2.4 describe efficacy and safety outcomes from the published literature that the Committee considered as part of the evidence about this procedure. For more detailed information on the evidence, see the overview.
2.3.1 The selection criteria in the following studies means that patients who had RF ablation or RF ablation plus surgery could not have been treated by surgery alone.
2.3.2 A non-randomised controlled trial of 46 patients treated by RF ablation or surgical resection described in a systematic review reported that median survival after diagnosis of liver metastases was 44 months and 54 months, respectively (significance not stated). In the same systematic review, 6 case series reported survival ranging from 17% (1/6) at 11-month follow-up to 88% (7/8) at 2- to 6-month follow-up. A non-randomised controlled trial of 418 patients reported that patients treated by surgical resection had overall survival rates of 73%, 65% and 58% at 3-, 4- and 5-year follow-up, respectively. These rates were significantly higher than those from patients treated by RF ablation alone, or by RF ablation plus resection (absolute figures not stated) (p < 0.0001). In the same study, the difference in survival between patients treated by RF ablation alone and RF ablation plus surgical resection was not significant (p = 0.36). A non-randomised controlled trial of 258 patients reported that 3-year disease-free survival was significantly greater in patients treated by surgical resection alone (40%) than in those treated by RF ablation plus resection (34%) (absolute figures not stated) (p = 0.01).
2.3.3 A case series of 243 patients with unresectable metastases reported survival rates of 20% and 18% at 3- and 5-year follow-up, respectively (absolute figures not stated).
2.3.4 The non-randomised controlled trial of 418 patients reported that recurrence at any site occurred more often in the RF ablation group (84%) than in the surgical resection group (52%) at a median 21-month follow-up (absolute figures not stated) (p < 0.001). The non-randomised controlled trial of 258 patients reported that recurrence at any site at 1-year follow-up was significantly lower in the surgical resection group (24%) than in the RF ablation plus resection group (61%) or the RF ablation alone group (66%) (absolute figures not stated) (p < 0.001).
2.3.5 The Specialist Advisers listed key efficacy outcomes as overall survival and local recurrence rate.
2.4.1 Haemorrhage requiring transfusion (required hospital stay of more than 72 hours) was reported in 1% (7/617) of treatment sessions in a case series of 309 patients.
2.4.2 Three cases series described in the systematic review reported postoperative complication rates between 0% and 33%. The complications included bowel perforation, peritoneal seeding (metastasis), bile duct stricture, wound infection and postoperative bleeding (absolute figures not stated).
2.4.3 A case report of a patient previously treated by both surgical resection and cryoablation described cutaneous fistula formation between the site of electrode insertion and the gastric antrum at 3-week follow-up. This had resolved at 6-month follow-up.
2.4.4 Visceral thermal injury (required hospital stay of more than 72 hours) was reported in less than 1% (4/617) of treatment sessions in the case series of 309 patients.
2.4.5 A case series of 122 patients reported infected biloma development (requiring percutaneous drainage) in 1% (1/122) of patients, and biliary dilation and cholangitis leading to biliobronchial fistula in 1% (1/122) of patients. Minor complications (not requiring intervention) included development of a small haemoperitoneum in 2% (3/122), biliary dilation in 3% (4/122) and persistent pain (location not reported) in 2% (3/122) of patients.
2.4.6 Another case report described a patient with pain and fever (40°C) at 5-day follow-up. Abdominal ultrasound revealed an intrahepatic abscess which required drainage. The patient recovered uneventfully.
2.4.7 The Specialist Advisers listed anecdotal adverse events as damage to the biliary tree and bradycardia. They considered theoretical adverse events to include injury to the bowel or diaphragm, and damage to the pleura or lungs.