Anterior resection specimens for rectal cancer contain a variable amount of mesorectum depending on the position of the tumour and the quality of the surgical dissection. There is also marked variation in the overall volume of the mesorectum and the height of the peritoneal reflections between individuals. During the operation, the surgeon should attempt to sharply dissect immediately outside the mesorectal fascia in the so-called holy plane of TME surgery described by Heald [5]. This ensures that the mesorectum is removed as an intact package that contains the entire primary tumour along with all possible routes of metastasis. The outer non-peritonealised surface of the specimen forms what is termed the CRM, a surgically created margin that should be extensively covered with mesorectal fascia if the dissection has been performed in the correct tissue plane.

Tumour involvement of the CRM, defined as tumour at or within 1 mm of the margin, is a poor prognostic feature and is strongly linked to local disease recurrence [6–10]. The method of involvement may be by direct spread or tumour deposit or within a vessel or nerve bundle with all showing a similarly poor prognosis [11]. A tumour deposit contained wholly within a lymph node may have a lesser risk, but this has yet to be proven. Others have suggested that tumour within 2 mm of the CRM is associated with a worse outcome, but sufficient data does not currently exist to support this position [12]. Prior to the description of TME surgery, the rates of CRM involvement were as high as 36 % [6] and varied markedly between individual surgeons [11]. The introduction of TME over recent years has been shown to substantially reduce the rate of local disease recurrence and improve survival, initially in single-centre series [13, 14], followed by large-scale population series and clinical trials [15–17]. A recently published UK trial showed CRM involvement rates to be as low as 8 % in the latter patient cohort with an increase in the median distance between the tumour and the CRM from 5 to 8 mm over the duration of the study [18]. It is believed that a major factor in the success of TME surgery is the marked reduction in CRM involvement and tumour perforation associated with the technique [11]. However, the more advanced tumours may still extend to within 1 mm of the CRM or even breach the mesorectal fascia, and therefore, TME surgery on its own will not result in complete resection of the tumour. Surgeons may therefore attempt to perform a more extensive operation removing additional extra-fascial tissues or organs en bloc or more frequently use preoperative therapy to attempt to shrink the tumour into a surgically resectable state. Radiologists are now very good at predicting tumour involvement of the CRM using MRI [19, 20] and are therefore critical in selecting patients for neoadjuvant radiotherapy.

TME/anterior resection specimens for rectal cancer should be carefully examined and dissected by pathologists to ensure that all the important features are identified and described prior to irreversible slicing of the specimen. A meticulous and consistent approach is required to ensure that nothing is forgotten. We recommend the dissection protocol developed in Leeds [21] that has been adopted into the United Kingdom Royal College of Pathologists dataset for colorectal cancer [1]. It is important that the specimen is received intact and unopened, preferably in the fresh state, as any disruptions or incisions may impair the assessment of the CRM. On a fresh intact specimen, the CRM should be bound by mesorectal fascia, which appears as a shiny smooth layer that often becomes dulled and opaque after formalin fixation (Fig. 22.3). If the dissection extends as low as the prostate gland in males, an additional layer of fascia, the fascia of Denonvilliers, may also be noted [22].

An assessment by the pathologist of the integrity of the surface of the specimen forms an important component of the assessment of the quality of surgery that must be fed back to the surgical team. A three-point grading system was initially developed for the MRC CLASICC trial [23] and has subsequently been shown to predict CRM involvement, local recurrence and patient survival in both small series [24–26] and larger multicentre studies [18, 27]. The recommended mesorectal grading system is described in Table 22.1.

It is helpful if the specimen is photographed at this stage to form a permanent record of the plane of surgery (Fig. 22.4). Digital images should be taken of the front and back of the specimen alongside a metric scale, with additional close ups of any significant mesorectal defects or perforations. These images should be stored in a departmental archive and can be used to facilitate feedback to surgeons in multidisciplinary team meetings and utilised for subsequent research/audit. Feeding back the plane of surgery in such a way to the operating team has been shown to improve the planes of dissection over time leading to a reduction in CRM involvement, which should ultimately improve patient outcomes [18].

Perforations, defined as a communication between the surface of the specimen and the lumen of the bowel, should similarly be documented in the report. Tumour perforations above the peritoneal reflections risk intraperitoneal recurrence and are associated with a poor prognosis [28, 29]. Under TNM staging rules, these tumours should be classified as pT4 [30]. However, infraperitoneal perforations are much more frequent and can occur at the time of surgery, often at the anterior aspect where the mesorectal thickness is at a minimum or in the area of the puborectalis. These surgical failures are not strictly pT4 tumours under TNM staging, but due to the high risk of local recurrence and reduced survival [31], we feel that they should be classified as pT4 on the basis of a poor anticipated prognosis.

Following a detailed macroscopic description, the specimen should be inked over the entire area of the CRM to allow for accurate histological measurement between the tumour and this margin. There should be no confusion between the CRM and the peritoneal surface as tumour involvement of both of these structures has different potential sequelae. Peritoneal disease, as described above, risks intraperitoneal recurrence in the abdominal cavity, whereas CRM involvement risks local pelvic recurrence. After inking, the tumour segment should undergo serial cross-sectional slicing at 3–5 mm intervals. The slices should be laid out in order and photographed to record the relationship between the tumour and the CRM and also to provide additional evidence for the plane of surgery. These images can be very useful for comparing to the radiological appearance of the tumour if any discrepancies arise. Ideally a minimum of five tumour blocks should be taken and at least one of these must demonstrate the tumour closest to the inked CRM. The pathological report must include a comment as to whether or not the CRM is involved by tumour so that the patient may be offered further adjuvant therapy, if appropriate, to reduce the chances of local disease recurrence. The distance between the tumour and the closest CRM to the nearest millimetre should be fed back to radiologists to allow them to audit the accuracy of their preoperative prediction of CRM status. If the MRI scan predicted a clear margin but the resection specimen shows evidence of tumour involvement, then it is helpful if the pathologist can indicate why this is so. Possible reasons include a tiny tumour deposit at the margin which would not be visible on MRI or alternatively that the area of CRM involvement relates to a region where the TME plane was not followed by the surgeon. In post-neoadjuvant therapy cases, it is also helpful for the pathologist to indicate whether any fibrosis extends beyond the tumour towards the CRM for correlation with the pre-radiotherapy MRI scan.