Benefits and Risks of Neoadjuvant Therapy for Liver Metastases

Abstract

Colorectal cancer ranks second in Europe and third in the United States among the most common causes of death caused by cancer. Approximately 50% of patients with colorectal cancer will develop liver metastases during the course of their disease. Surgical resection is the only treatment and can currently ensure long-term survival in 25% to 40% of patients. However, using the current indications for surgery, only 15% to 20% of patients with colorectal cancer liver metastases are suitable for surgical resection. In the majority of patients, metastases are considered to be unresectable, and chemotherapy is the treatment of choice, but 5-year survivors with chemotherapy alone are rare. Recent progress, including more efficient chemotherapy regimens and targeted agents, has opened new perspectives in the treatment of both resectable and nonresectable colorectal liver metastases. In patients with unresectable liver metastases, emerging data have shown that neoadjuvant chemotherapy can render some metastases resectable, hence opening up the possibility of prolonged survival. In patients with initially resectable liver metastases, neoadjuvant chemotherapy is often administered, although the benefits in terms of long-term disease control have not yet been formally proven. Thus, many patients receive neoadjuvant chemotherapy before liver resection but without direct proof of benefit. Recently, the enthusiasm over the potential positive effects of neoadjuvant chemotherapy has been cooled down somewhat by the awareness of toxicity to the remnant liver caused by preoperative treatment, which could increase the risks of surgery and even preclude liver resection. There are now accumulating data showing that preoperative chemotherapy is associated with pathologic changes of liver parenchyma. The following two main types of chemotherapyassociated liver injuries have been reported: vascular changes and chemotherapy-associated steatohepatitis (CASH). The vascular changes include sinusoidal dilation with erythrocytic congestion, occasionally accompanied by perisinusoidal fibrosis and fibrotic venous occlusion, which could result in sinusoidal obstruction syndrome as observed in veno-occlusive disease. CASH is defined as the association of severe steatosis, lobular inflammation, and ballooning of hepatocytes. Few studies have evaluated the correlation between the type of liver injury and the neoadjuvant chemotherapy regimen. In this issue, Aloia et al assess the relationship between systemic chemotherapy and hepatoxicity by comparing 92 patients who received preoperative chemotherapy with 17 patients who received surgery without chemotherapy. Preoperative chemotherapy included fluorouracil (FU) plus leucovorin, either alone or combined with oxaliplatin. This study confirms, as suggested in other reports, that the oxaliplatin-based combination regimen is associated with an increased risk of vascular lesions of the liver. It is uncertain from this report whether this is specific for oxaliplatin or caused by the effects of more intensive combination chemotherapy. In other reports, irinotecan-containing regimens have also been associated with increased risks of steatosis and steatohepatitis. Steatohepatitis is observed more frequently after chemotherapy in patients with a higher body mass index. This could explain why CASH is more frequently reported in US studies, whereas vascular lesions are more often reported in studies from Europe. Therefore, it is possible that specific chemotherapy regimens do not automatically predispose to steatosis but can aggravate it when it already exists. That chemotherapy can be associated with liver injury is now well recognized. The main question is whether collective damages to the liver induced by preoperative chemotherapy have any clinical significance. Do they increase the risks of complications after surgery and lead to a change in treatment strategy? The article by Aloia et al in this issue of the Journal of Clinical Oncology addresses this question. It shows that perioperative mortality and morbidity rates were not significantly increased in patients who had received preoperative chemotherapy. However, patients who had received more than 12 cycles of preoperative chemotherapy had a higher risk of reoperation and a longer hospital stay. The safety data of European Organisation for Research and Treatment of Cancer (EORTC) study 40983, which compared perioperative chemotherapy with FU, leucovorin, and oxaliplatin (six cycles before surgery and six cycles after) with surgery alone in 364 patients, were presented at the 41st Annual Meeting of the American Society of Clinical Oncology in 2005. The results showed that the mortality rate was low (close to 1%) in the two treatment arms and that the rate of reversible complications was acceptable. Thus, administration of six cycles of FU, leucovorin, and oxaliplatin before surgery seems feasible. Another report by Karoui et al brought another insight into the relationship between duration of preoperative chemotherapy and perioperative morbidity and showed that administration of more than six cycles of neoadjuvant systemic chemotherapy increased morbidity after major liver resection but did not increase mortality. Some studies have looked at the relationship JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 24 NUMBER 31 NOVEMBER 1 2006