Liver metastases from colorectal cancer

Abstract

Introduction: Colorectal cancer (CRC) is the third most common cancer worldwide. At diagnosis of CRC 20-25% of patients have metastatic disease. The liver is the most common metastatic site and liver metastases are detected in 25-30% of all patients. A quarter of these patients are amenable for liver resection that results in a five-year survival exceeding 50%. The indications for liver resection continue to broaden and are no longer limited by number and size of liver metastases nor the presence of extrahepatic metastases. Currently liver resection is indicated when macroscopic tumour clearance can be achieved with preservation of a sufficient future liver remnant. Different strategies to improve resectability exist such as portal vein occlusion, two-stage resections, associating liver partition and portal vein ligation for staged hepatectomy and thermal ablation, mainly radiofrequency ablation or microwave ablation (MWA). Decisions on management of patients with metastatic CRC should ideally be made in a multidisciplinary team (MDT) setting. Failing to do so may result in suboptimal management and patients that could be resected are not necessarily offered curative-intended treatment. As a result of this there are known regional differences in the treatment of patients with liver metastases that may affect survival. For patients not suitable for resection, either due to the metastatic burden or comorbidity omitting extensive surgery, local ablation is an option.

Aims: The aim of Study I was to provide detailed population-based data of liver metastatic patterns, treatment and survival in patients with metastatic CRC. In Study II, the potentially improved resection rates were evaluated in a scenario where all patients with liver metastatic disease, irrespective of extrahepatic metastases, were assessed by a liver MDT. Study III aimed to describe the feasibility and safety of a multiple MWA strategy in patients with initially unresectable liver metastases. The primary aim of Study IV was to evaluate the accuracy and safety of antenna placement in stereotactic computed tomography-guided MWA of primary and secondary liver tumours. The secondary aims of Study IV were to evaluate the feasibility of the navigation system, to measure the procedure-related radiation dose and to assess the safety of high-frequency jet ventilation for target motion control.

Patients and Methods: In Studies I and II, a population-based cohort consisting of all patients diagnosed with CRC in the Stockholm and Gotland region during 2008, identified from the Swedish Colorectal Cancer Registry, was used. Details of metastatic spread, referral to a MDT conference and oncologic and surgical treatment were retrieved from electronic patient charts and recorded during a five-year follow-up period or until death. Predictors of survival in Studies I and III were estimated using a Cox proportional hazards model. Survival curves were illustrated using Kaplan-Meier estimates and survival functions were compared using the log-rank test (Studies I-III). For Study II, additional information on American Society of Anesthesiologists grade, comorbidity and patients’ own preferences towards treatment, were retrieved for the 272 patients with liver metastases. Each patient was presented at a fictive liver MDT conference, irrespective of previous management, and categorized as resectable, potentially resectable or unresectable. Treatment decisions were compared with the original management and factors associated with referral to the liver MDT were assessed using logistic regression. In Study III, a multiple MWA strategy was applied to 20 patients with initially unresectable liver metastases between October 2009 and September 2012. The feasibility and safety of the procedure as well as local recurrence rate was recorded. Overall and disease-free survival in the ablated group was compared with results from two historic cohorts from Study I, one treated palliatively and the other resected. In Study IV 20 patients with primary or secondary liver malignancy, where surgical resection was contraindicated or the lesions were not visible on ultrasound, were included for treatment with percutaneous MWA using a stereotactic navigation system (Cascination AG, Bern, Switzerland) that shows the actual position of the tracked antenna in real time with respect to pre-operative CT images. Descriptive statistics were used to evaluate the accuracy of antenna placement, the number of antenna readjustments, safety and radiation dose.

Results: In Study I 1026 patients with CRC were identified and liver metastases were detected in 272 (26.5%). Liver and lung metastases were more often diagnosed in hindgut (splenic flexure to rectum) compared with midgut cancer (caecum to splenic flexure) (28.4% versus 22.1%, p=0.029 and 19.7% versus 13.2%, p=0.010, respectively) but the extent of liver metastases was less for hindgut compared with midgut cancer (p=0.001). Five-year OS was significantly worse in liver metastatic midgut cancer compared with hindgut cancer (6.5% vs. 21.6%, p<0.001). In liver metastatic disease the presence of lung metastases did not significantly influence OS as assessed by multivariable analysis (HR 1.11, CI 0.80-1.53). At the fictive liver MDT in Study II, a further 22 patients (12.9%) of the 170 patients not previously referred to a liver MDT were considered as resectable or potentially resectable. Factors influencing referral to a liver MDT were age (OR 3.12, CI 1.72-5.65), ASA score (ASA 2 versus ASA 3, OR 0.34, CI 0.18-0.63) and number of liver metastases (OR 0.10, CI 0.04-0.22, 1-5 versus >10 liver metastases), while male gender (OR 1.39, CI 0.84-2.30) and treatment at a teaching hospital (OR 1.06, CI 0.62-1.81) were not. In Study III, the ablated group showed a four-year overall survival of 41% compared with 70% for the historic cohort of resected patients and 4% for palliatively treated patients. Eighteen patients had recurrence in the liver, 11 had extrahepatic recurrence and 10 out of 20 treated patients were alive at a median follow-up of 25 months. In Study IV, the antenna was placed with a mean target error of 5.8±3.2 mm in relation to the intended target at a mean total radiation dose of 958±557 mGy x cm.

Conclusions: Study I: Detailed population-based data on the metastatic pattern of CRC and survival could assist in more structured and individualized guidelines for follow-up of patients with CRC as well as personalized treatment, based on factors other than resectability as currently defined. Study II: A meaningful number of patients with liver metastases were not managed according to best available evidence and the potential for higher resection rates is considerable. Study III: The highly selected patients treated with a multiple MWA strategy had a survival benefit compared with patients treated with palliative chemotherapy but the recurrence rate was high. Study IV: Sufficient accuracy was achieved using percutaneous MWA with stereotactic navigation.