Diagnostic accuracy of endoscopy in determining rectal tumor proximity to the peritoneal reflection.

Abstract

Treatment of invasive rectal adenocarcinoma is stratified into upfront surgery versus neoadjuvant chemoradiotherapy, in part, based on tumor distance from the anal verge (AV). This study examines the correlation between tumor distance measurements (endoscopic and MRI) and relationship to the anterior peritoneal reflection (aPR) on MRI. A single-center retrospective study was performed at a tertiary center accredited by the National Accreditation Program for Rectal Cancer (NAPRC). 162 patients with invasive rectal cancer were seen between October of 2018 and April of 2022. Sensitivity and specificity were determined for MRI and endoscopic measurements in their ability to predict tumor location relative to the aPR. One hundred nineteen patients had tumors endoscopically and radiographically measured from the AV. Pelvic MRI characterized tumors as above (intraperitoneal) or at/straddles/below the aPR (extraperitoneal). True positives were defined as extraperitoneal tumors [Formula: see text] 10cm. True negatives were defined as intraperitoneal tumors > 10cm. Endoscopy was 81.9% sensitive and 64.3% specific in predicting tumor location with respect to the aPR. MRI was 86.7% sensitive and 92.9% specific. Utilizing a 12cm cutoff, sensitivity of both modalities increased (94.3%, 91.4%) but specificity decreased (50%, 64.3%). For locally invasive rectal cancers, tumor position relative to the aPR is an important factor in determining the role of neoadjuvant therapy. These results suggest endoscopic tumor measurements do not accurately predict tumor location relative to the aPR, and may lead to incorrect treatment stratification recommendation. When the aPR is not identified, MRI-reported tumor distance may be a better predictor of this relationship.