Introduction Standards for Colorectal cancer (CRC) resection specimen histology reporting consider factors thought to have apparent significance for prognosis and further therapy. Whilst well validated for surgical resection, the increasing use of advanced endoscopic resection for polyps containing previously unknown early CRC presents challenges in interpretation of these factors. In addition to tumour budding, unfavourable tumour grade, and vascular invasion, Ueno et al[1] proposed parameters for width and depth of submucosal invasion as risk for adverse outcome. This study aims to analyse any association between pathological factors and outcome with endoscopic resection of early CRC.

Methods Retrospective review of all CRC removed endoscopically between March 2006 and March 2011. All endoscopic and surgical resection specimens were reviewed by two expert gastrointestinal histopathologists, with measurement of width and depth of submucosal invasion made. All follow up procedures, including radiology, were reviewed.

Results 35 cases were identified (24 males, 11 females, median age 69 years). All patients were alive after median follow-up period of 32 months; no residual/recurrent cancers were found in any patient managed with endoscopic therapy alone. Of the 12 patients who had further surgical intervention due to reported incomplete endoscopic resection on histology, none had residual carcinoma in the subsequent resection specimen. Three patients (8.6%) were found to have Dukes C1 cancers (all T1 N1 M0). These cancers were not associated with poor differentiation or lymphovascular invasion (p = 0.546) or tumour budding of low or high intensity (p = 1.000). The relationship between the width and depth of submucosal invasion and Dukes C1 did not reach statistical significance (p = 0.096), although these three cancers did fulfil Ueno criteria. Presence of lymph node metastases was associated with Haggitt level 4 (p = 0.03), but not with the presence of tumour at the excision margin (p = 1.000) in the subsequent surgical resection group.

Conclusion Our experience highlights the challenges in applying histopathological criteria to individual cases of early CRC resected via endoscopic therapy. Most patients underwent surgery for an unclear resection margin, however no residual cancer was present in the resection specimens and aside from a Haggit level 4, found no other predictors of risk lymph node metastases. Suggestions for future studies include piloting a more minimally invasive approach, such as regional lymph node dissection in selected cases as well as studying biomarkers for refining risk stratification.

Disclosure of Interest None Declared.

Reference

1. Ueno H et al. Risk factors for an adverse outcome in early invasive colorectal carcinoma. Gastroenterology 2004; 127: 385–394