Colon cancer : management and outcome in a Swedish population

Abstract

Colon cancer is common in Sweden, with about 3500 new cases every year. Tumours of the colon and rectum are usually addressed as an entity. Great effort has been made to improve the outcome after rectal cancer treatment with subsequent improvement of survival. Only few studies have addressed the specific issue of colon cancer and how to improve the outcome for this large group of patients. As a consequence, the 5-year survival after colon cancer treatment in Sweden is now poorer than after rectal cancer treatment.

Since 1996, the Stockholm-Gotland region has a common management protocol for patients with colon cancer. As part of this protocol, data on all patients with newly diagnosed colon cancer in the region are prospectively collected in a database at the Oncologic Centre in Stockholm. The database includes information on age, sex, tumour location and stage, emergency or elective surgery, type of surgery performed, postoperative mortality, histopathology of the tumour and follow-up data on recurrence and survival. The database is continuously validated and updated through comparison to other registers with information on healthcare consumption, diagnoses according to the international classification of diseases (ICD) and causes of death.

This thesis is based on information from the Oncologic Centre database and includes all patients diagnosed with colon cancer in the Stockholm-Gotland region during 1996-2000, followed until January 2005. The aim of the thesis was to achieve knowledge on how patients with colon cancer have been managed in the region during these years and to assess the outcome in terms of postoperative mortality, loco-regional and distant recurrence and survival. Another aim was to identify risk factors for death and recurrence.

During the study period, 2855 patients were diagnosed with colon cancer. After the exclusion of 80 patients diagnosed at autopsy, 2775 were eligible for follow-up. The crude 5-year survival for all patients was 46 per cent. Nine hospitals managed these patients, and differences in overall survival and risk for local recurrence between the hospitals were present despite the common management protocol.

The cumulative risk for loco-regional recurrence was 11 per cent. Tumour location in the right flexure and sigmoid colon, more advanced T-stage and N-stage, bowel perforation, emergent surgery and poor tumour differentiation were identified as risk factors for loco-regional recurrence. After complete resection of loco-regional recurrences, the estimated 5-year survival was 43 per cent, while there were no 5-year survivors among patients where a complete resection of the recurrence could not be accomplished.

Liver metastases were detected in 24 per cent of the patients during follow-up. The hepatic resection rate was four per cent, which is remarkably low. A retrospective evaluation of radiological images of the liver showed that ten per cent of the patients might have been candidates for liver surgery.

An evaluation of tumour volume as a prognostic factor showed that an increased tumour volume was associated with poorer survival even after adjusting for other postoperatively known factors.

Some areas of possible improvement were identified. A multidisciplinary approach to improve preoperative staging, surgery, histopathologic staging and selection of patients for medical oncologic treatment could probably improve the outcome for patients with colon cancer.