Colorectal Liver Metastases : Onco-surgical Strategies & Prognostic Factors

Abstract: The management of patients with colorectal cancer with liver metastases has improved over the past decades. There have been significant developments in terms of surgery, local therapies, anesthesia and oncological treatment. Nevertheless, there is no consensus on what constitutes optimal treatment strategy. The criteria of resectability have widened, and an increasing number of patients with both more advanced disease and age, are offered a hepatic resection. Reliable predictive models help clinicians to stratify patients into different risk categories and improve patient selection for both surgery and oncological treatment. Furthermore, these models can serve as important tools to help predict accurate information regarding prognosis. The aims of this thesis were to identify prognostic factors, to construct a reliable predictive model, to assess surgical strategies, and the role of oncological treatment.In paper I, patients who had undergone a hepatic resection were grouped according to the Glasgow Prognostic Score (GPS). Patients received a score of GPS 0, GPS 1, or GPS 2. This score is based on each patient’s preoperative values of acute phase reactants C reactive protein, and albumin. GPS is an independent prognostic factor, and can be used to stratify patients into different risk groups. The overall survival difference between GPS 0 and GPS 2, was 43 months.In paper II, a predictive model was constructed for patients with resectable colorectal liver metastases, which stratified patients into low, medium, and high risk groups. The performance of the model was assessed with discrimination (Concordance statistic), and calibration. The multifold cross-validation confirmed its internal validity. The model identified a group of patients (high risk) with poor overall survival of 23 months, in which surgery may be questionable.In paper III, the optimal surgical strategy was examined. This study compared the outcome of staged resection (colorectal-first, or liver-first) with simultaneous resection in patients who had undergone resection for colorectal liver metastases. Patients who had a simultaneous resection more often had a primary cancer of midgut embryonic origin, less extensive metastatic disease, and smaller hepatic resection. No differences were found in outcome between the treatment strategies neither before nor after propensity score matching.In paper IV, a population-based study was performed which compared patients with a singular resectable colorectal liver metastasis treated with oncological treatment compared to patients who were treated with surgery alone. Both before and after propensity score matching, no differences were discerned with regards to overall survival, morbidity or mortality between the groups.