Cancer of the upper rectum.

Rectal cancer constitutes one-third of all colorectal cancers, and the incidence in Denmark increasing. In 2012, 1.400 cases were registered, and of these 38% were located in the upper rectum. There have been several key advances in the optimal management of rectal cancer during the past decades, primarily by standardisation and improvement of the surgical procedure. There is now general agreement that the optimal surgical treatment involves the concept of total mesorectal excision and that a resection with tumour-free margins is crucial. Controversy exists as to whether total mesorectal excision (TME) is necessary for upper rectal cancers or if a partial mesorectal excision (PME) with mesorectal transection 5 cm below the tumour is adequate. Furthermore, there is no agreement as to whether surgery alone is sufficient or whether neoadjuvant radio- and/or chemotherapy should be administered for tumours of the upper rectum. This thesis aims to discuss aspects of the treatment of rectal cancer with regard to the adequacy of mesorectal excision and oncological outcome with a particular focus on cancer of the upper rectum. In study I, the extent and completeness of mesorectal excision was estimated by postoperative magnetic resonance imaging of the pelvis in patients with primary surgery for rectal cancer. In the 136 patients with post-operative MRI, inadvertent residual mesorectal tissue was evident in 40%, especially following PME, suggesting suboptimal surgery performed. Additionally in patients who had PME, the distal margin was found to be less than 3 cm in more than 50% of patients, suggesting a discrepancy between guidelines and the actual surgery performed. In study II, we estimated the risk of local recurrence in the previously audited cohort of patients, with a particular focus on patients with upper rectal cancer treated by PME and without neo-adjuvant therapy as standard. Using Kaplan-Meier analysis, the total three-year local recurrence rate was 7% with tumour stage and an involved circumferential margin as the most important predictors of local recurrence. The local recurrence rate after PME was significantly higher than for TME (14% vs. 3%; p=0.032), and were diagnosed earlier (p=0.001). In all cases with local recurrence following PME there was evidence of either inadvertent residual mesorectum and/or an insufficient distal resection margin. In study III, we investigated the length of the distal resection margin and degree of tissue shrinkage after surgical removal and fixation by using MRI of the fresh and fixed specimen. We found that the length of the specimen and the distal margin was reduced by 30% after surgical removal and fixation. If a 5-cm distal margin below the luminal level of the primary tumour on the fresh specimen is the objective for advanced cancer of the upper rectum treated with PME surgery, a margin of at least 3.5 cm of mesorectum on the fixed specimen should be attained for the pathologist to accurately establish distal radicality.