Mucin lakes or perforation?

Abstract

INTRODUCTION

Mucinous adenocarcinomas account for about 10-15% of all adenocarcinomas. They are comprised of at least 60% mucus (1). The findings suggest that it represents a genetically distinct variant of colorectal adenocarcinoma and has implications for the development of targeted therapies and their clinico-pathologic features (3). It is thought that the presence of mucin allows cancer cells to spread faster. As a result, it is considered more aggressive than regular adenocarcinomas (1). Conflicting results are found in the published literature regarding the relationship between mucinous colorectal cancer and survival (4). Some studies report a 5-year survival of 11% for mucinous carcinomas.

The appearance of high signal T2W sequences of mucinous rectal tumours on MRI help to stage disease pre and post radiotherapy period. It is reported that MRI is not useful for gauging disease activity of persistent abnormalities in mucinous tumours that often represented inactive mucin lakes(4).

CASE PRESENTATION 1

A 67 year old patient presented to the colorectal clinic with a history of diarrhoea and bleeding. He underwent colonoscopy, CT and MRI scanning. Colonoscopy revealed a tumour 6 cm from the anal verge and biopsy confirmed intestinal poorly differentiated adenocarcinoma with mucinous component.

CT scan showed no evidence of distant metastases. MRI has shown a T3 N1 mucinous type rectal carcinoma 9cm from the anal verge with negative circumferential resection margin (CRM).

Patient had been offered neo-adjuvant chemo/radiation after MDT discussion.

Restaging MRI revealed reduction in tumour size with areas within the mesorectum consistent with perforation/infection that lies close to anterior circumferential resection margin.

A Laparoscopic low anterior resection and loop ileostomy formation was done and histology showed recto-sigmoid mucinous adenocarcinoma pT3pN1pMXR0, Duke’s C1 completely excised at all margins.. No extramural venous invasion, no perforation and no growth on the mucosal site on macroscopic examination. There were mucin deposits 14mm beyond muscularis propria and 3mm from the non peritonealised circumferential margin containing very occasional viable tumour cells. Several lymph nodes contained acellular mucin and a single node contained small viable tumor deposits.

The patient recovered successfully after surgery and had been offered adjuvant chemotherapy.

CASE PRESENTATION 2

A 75 year lady was referred by her community doctor with a history of rectal bleeding and palpable tumour on rectal digital examination. She was referred for urgent colonoscopy, CT and MRI. Biopsies showed an adenocarcinoma arising in a tubulovillous adenoma. The patient received a colostomy due to obstructive symptoms. CT showed no evidence of distant metastases. MRI showed bulky polypoid tumour and a lot of high signal change within the rectal lumen suggesting a mucinous nature of the polypoid mass. Predicted MRI staging was T2N1 with threatened resection margin due its low position and unclear separation from levator ani muscles. The patient started neoadjuvant chemo-radiotherapy.

Final MRI scans were preformed 4 weeks post chemo-radiation. Though the intraluminal mass had reduced a few small high signals out-pouchings through the anterior rectal wall into thin mesorectal space. This iprodcued a predicted final MRI staging of no more than T2N0. These changes of uncertain nature could be due to perforation or mucinous lakes protruding through invaded rectal wall. Subsequently the patient was offered and received a laparoscopic abdominoperineal resection.

Histology revealed two polypoid lesions with residual tubulovillous adenoma with low and high grade dysplasia, with no residual invasive carcinoma identified. Mucin pools were present in the submucosa, muscularis propria and subserosa. All this was interpreted as a single large adenomatous lesion giving rise to an invasive adenocarcinoma, which had regressed with therapy. Differentiation was not assessable; there was no extramural venous invasion. Mucin pools extended to <1 cm from the non-peritonealised circumferential margin. Resultant staging was pTxpN0pMXR0, with no possiblity of calculating Duke’s staging.

The patient recovered after surgery well.

DISCUSSION

It is the first case series of two patients with mucinous adenocarcinoma with unusual changes in post radiotherapy MRI. The post chemoradiation MRI findings suggested features of perforation masking the “mucin lakes”. We were unable to fully explain pathogenesis or evident reason for this type of finding after looking at the literature. Although patients post chemoradiotherapy MRI reported reduction in tumour bulk, there are a few questions raised in both cases: are these “Mucin lakes” mimicking perforation, post radiation changes or persistence of the disease? How do mucin lakes extend beyond the muscularis propria?

There are case reports describing mucin pools extending to subserosa and submucosa, but no description of mucin lakes extending 14mm beyond the muscularis propria. Absence of the literature evidence about it causes confusion on the pre-chemoradiotherapy MRI (reported as perforation).

The theory of “mucin lakes” formation is that the low-grade neoplasm produces mucin that distends the bowel lumen. This increase in intramural pressure, where the perforating vessels traverse the muscularis propria allows neoplastic epithelium to herniate through these weak areas in the wall (4) and thereby mimic perforation.

In our experience, patient 1 had occasional cancer cells in mucin pools extending 14 mm beyond the muscularis mucosa, but circumferential resection margin was not compromised (2 mm). In the second patient mucin pools were acellular, histopathology showed minimal margin of <1mm, and again the tumour was completely excised. This cellularity of the mucin lakes poses questions on staging.

One more unusual observation from comparison of these two case reports: In first case study pre-chemoradiation MRI did not show obvious mucin lakes, but they became evident on post-chemoradiation MRI extending beyond the muscularis propria despite tumour mass having completely regressed. Does this imply that wall tissues become less resistant to cancer pool intrusion after chemoradiation?

Second patient had defunctioning colostomy after first MRI and mucin pools were only at submucosa and subserosa. Does the high intramural pressure have an impact on the protruding of the cancer pools beyond the muscularis propria?

Pathogenesis and prediction of the mucin lakes behaviour prior and post chemoradiotherapy is unclear and requires future study, it might change surgical management. MRI diagnostic features are not always straight forward in the recognition of the tumour extent. The presence of acellular mesothelial reaction and inflammation extending to the serosal surface causes considerable confusion and may result in upstaging or overstaging of mucinous adenocarcinoma.

REFERENCES