Surgery is the most important treatment modality for patients with CRLMs. Appropriate surgery in selected patients will result in long-term survival of up to 40%.1,2 This percentage has increased over the last two decades. There have been developments in surgical technique, including portal vein embolisation and safer liver division and resection, as well as improvements in post-operative management. As a result, resections are now more extensive and, due to improving expertise, are associated with reduced post-operative morbidity and mortality. A general principle is to resect all macroscopic disease, aiming for a potentially curative resection. In order to achieve this, it may be necessary to combine surgical excision with ablation, e.g. radiofrequency ablation, intra-operatively or percutaneously in the post-operative period. Often this avoids an unacceptably dangerous major resection; for example, an extended right hemi-hepatectomy can be combined with radiofrequency ablation of smaller lesions on the left side of the liver. However, it should be noted that initially only 15–20% of patients are suitable for surgical resection.3 As a result, an attempt to increase the resectability rate with additional treatments has been advocated.
There is increasing interest in the role of pre-operative or neoadjuvant chemotherapy to downsize liver metastases in an attempt to achieve resection of previously unresectable CRLMs. Accordingly, such treatment increases the proportion of patients able to achieve long-term survival. Recent studies have described resection rates of up to 20% in patients with initially unresectable liver metastases, with five-year survival rates in these patients approaching 50% (see Table 1).4–7
However, it should be emphasised that in order to achieve optimum results, careful selection is essential. Recent studies have demonstrated a strong correlation between response rate to chemotherapy and subsequent resection rate, which in selected patients may be 20–50% compared with 1–20% in non-selected patients.
As shown in Table 1, various chemotherapy regimes have been advocated and several have been subjected to prospective clinical trials. The overall conclusions are, in summary: folinic acid, fluorouracil and irinotecan (FOLFIRI) and 5-fluorouracil (5-FU), leucovorin and oxaliplatin (FOLFOX) are equally effective (response rates of 56 and 54%, respectively8); and 5-FU, leucovorin, oxaliplatin and irinotecan (FOLFOXIRI) is superior to FOLFIRI9 (response rates of 60 and 34%, respectively, and R0 resection rates of 36 and 12%, respectively).
Other studies have demonstrated response rates of between 8 and 41% following administration of oxaliplatin- or irinotecan-based regimes. Again, selection may be crucial in this regard. It would appear that three-drug combinations have a higher response rate and a higher resection rate, without any significant impact on either toxicity or surgical safety.
These studies are extremely important and demand our attention. Patients with apparently unresectable disease who are otherwise fit should be considered for neoadjuvant therapy in a multidisciplinary setting. The role of biological agents in this situation has also been extensively investigated. The two agents studied are bevacizumab (vascular endothelial growth factor [VEGF] monoclonal antibody) and cetuximab (epidermal growth factor receptor [EGFR] monoclonal antibody). Bevacizumab has been shown to improve the objective response rate and prolong survival: in one study, the response rate was 45–70% when combined with 5-FU, leucovorin and irinotecan.10 Cetuximab has been reported to result in resection rates of 19–30% in unselected patients.11 Patients refractory to conventional chemotherapy can be converted to cetuximab-based regimes, and hence increase the total proportion of initially unresectable patients to approximately 20%. It is suggested that resectability rate should be an end-point in randomised trials.
There is a concern that these agents will result in an increase in postoperative morbidity and possible mortality. Studies have demonstrated that post-operative morbidity is correlated with the number of cycles of chemotherapy and not necessarily the type of chemotherapy.
There are two major concerns in this regard: steatohepatitis (recognised particularly in patients with a high body mass index [BMI]) and vascular changes resulting in increased post-operative haemorrhage. However, there is controversy relating to these complications and several studies have failed to demonstrate any increased risk of morbidity or mortality.12
A few reports on the role of hepatic arterial infusion chemotherapy as neoadjuvant treatment have been described. Using these techniques, resection rates of between 6 and 47% are reported.13 Nevertheless, few centres utilise this technique in preference to systemic chemotherapy.
Peri- and Post-operative Adjuvant Chemotherapy
Controversies exist relating to the role of adjuvant chemotherapy following curative surgical resection. The European Organisation for Research and Treatment of Cancer (EORTC) trial 40983 randomised 364 patients into a two-arm study. Patients received either FOLFOX-4 followed by surgery and post-operative FOLFOX-4 over three months or surgery alone.
There was some improvement in disease-free survival with perioperative chemotherapy. In patients in whom the metastases were subsequently resected, progression-free survival in resected patients was 42.4% versus 33.2% in un-resected patients (hazard ratio [HR] 0.73; p=0.025), indicating a favourable response.13 In this study there were no major differences in post-operative complications. A new study is being devised in which cetuximab is added to the chemotherapy regime alone or with bevacizumab. These results are awaited with interest.
There is increasing interest in the role of radiofrequency ablation of liver metastases. Numerous reports have suggested improvements in survival in patients with unresectable disease. Five-year survival rates between 23 and 58% and even 10-year survival rates of up to 25% have been reported.14,15 This is a relatively minimally invasive technique associated with a low incidence of toxicity. Radiofrequency ablation can be used to extend the role of surgery by enabling a hemihepatectomy to be performed with radiofrequency ablation of lesions in the adjacent lobe carried out either percutaneously or intraoperatively. Other studies have even suggested that patients with resectable disease can be treated with radiofrequency ablation, thus avoiding the problems of major surgery.16–18 In the same way that surgery is combined with chemotherapy, so radiofrequency ablation should be combined with appropriate chemotherapy. There is no doubt that such therapy must be discussed in a multidisciplinary meeting, and careful selection is essential to achieve good results.
All otherwise fit patients with CRLMs should be referred to a multidisciplinary specialist team in order to ensure that the most effective therapy is initiated. Surgery has a major role and, whenever possible, all resectable disease should be removed. Patients whose disease is extensive and inoperable may be converted to operable disease by appropriate neoadjuvant therapy, and this treatment should be considered and discussed with the patient. Undoubtedly, increased resection rates can be achieved and therefore outcome improved.
Patients who have undergone appropriate resection with removal of all macroscopic disease may benefit from adjuvant chemotherapy with the addition of biological agents. Several studies are at present ongoing. It is important to remember the role of radiofrequency ablation in increasing resectability rates and dealing with recurrent disease following surgery. A combination of radiofrequency ablation with chemotherapy may be used to downstage inoperable disease towards a resectable state.