Colorectal liver metastases: factors affecting outcome after surgery

Abstract

Colorectal cancer is the second leading cause of cancer related death in Europe. The overall survival rate of patients with colorectal cancer is greatly affected by the presence of liver metastases, which occurs in about 50% of patients. Radical resection of colorectal liver metastases means a possibility of cure, resulting in a 5 year survival of 30-60%. In case of irresectable metastases, radiofrequency ablation (RFA), using heat to destroy tumour tissue, is an alternative surgical treatment option. Survival rates vary widely after surgery and it can still not be predicted accurately which patient will experience a recurrence and which not. The identification of prognostic markers are needed to select patient that possibly benefit from surgery and/or additional therapies. This thesis describes factors associated with outcome of patients undergoing surgery for colorectal liver metastases. First we discuss factors associated with recurrence after RFA. We demonstrate that the presence of viable tumour cells attached to the RFA needle applicators is associated with an increased chance of local tumour progression. This problem might be overcome by ablating the needle track. Furthermore we show that tumours close to vessels can be ablated safe and effectively using a multipolar RFA device, where previous literature demonstrated a problem effectively ablating these tumours using a monopolar device. Second, factors associated with outcome after resection are examined. Preclinical research demonstrated that the protein CD95 upon binding with its ligand activates pathways involved with a more agressive phenotype of colorectal liver metastases. Testing levels of CD95L in blood of patients demonstrated that higher levels of CD95L were associated with poor prognosis of patients with colorectal liver metastases. We used genomics and proteomics to identify more factors associated with poor prognosis of which the protein Maspin was a factor identified by both platforms. Next we used immunohistochemistry to determine Maspin expression in biopsies of primary colorectal cancer. We discovered a stage specific role for Maspin being prognostic in stage III but not in stage II primary colorectal cancer. Both Maspin and CD95L might be used in the future to select patients for surgery. Gene expression was further used to find a gene expression profile able to predict disease free survival. Unfortunately this attempt failed, possibly due to the phenomenon of ‘overfitting’. We did however discover that the profile was greatly influenced by administration of neoadjuvant chemotherapy. Neoadjuvant chemotherapy upregulates genes involved in the immune system. These genes seemed to have a tumour stimulating role, generating the question whether this genes might play a role in chemotherapy resistance. This thesis furthermore describes the Hepatica study, a randomized controlled study examining the addition of a angiogenesis inhibitor (bevacizumab) to a CAPOX regime. This study demonstrates the addition of bevacizumab is safe and leads to an improved non significant DFS. In conclusion, this thesis makes a start finding factors affecting outcome after surgery of colorectal liver metastases. Insight in the mechanisms involved with recurrence might select patients that will benefit from surgery and eventually lead to new therapeutic targets.