# Histological prediction and choice of the best resection strategy in front of a colorectal lesion > 2 cm: prospective comparison of endoscopic characterization, non-targeted and targeted biopsies

## TL;DR

This study compares endoscopic characterization and biopsies for predicting the best treatment for large colorectal lesions, finding that characterization is more effective.

## Contribution

The study provides a prospective comparison of endoscopic characterization versus biopsy methods for treatment strategy selection in large colorectal lesions.

## Key findings

- Endoscopic characterization proposed adequate treatment in 52.3 to 70.5% of cases, while biopsies had higher under treatment rates.
- Biopsies failed to accurately assess cancer invasion depth and risked under treatment in nearly a third of cases.
- Endoscopic characterization using CONECCT classification reduced under treatment and unnecessary surgeries for non-malignant lesions.

## Abstract

Accurate endoscopic characterization of colorectal lesions is essential to predict histology and select the best treatment strategy but remains very difficult. Instead of the recommended endoscopic characterization, many gastroenterologists routinely perform biopsies of the lesion to propose endoscopic resection with or without R0 intent. The aim of this study was to determine which of endoscopic characterization or biopsies, either targeted (TB) or non-targeted (NTB), is the most effective to determine the best treatment strategy for colorectal neoplasia > 2 cm.

We prospectively assessed the best strategy between endoscopic characterization and targeted or non-targeted biopsies, so that the proposed resection technique offered a level of quality of tumor resection adapted to the definitive histology of the lesion on R0-resected specimen.

84 patients with 88 lesions were included. “Adequate treatment” was proposed by endoscopic characterization in 52.3 to 70.5% of cases, “under treatment” in 2.3 to 9.1% and “over treatment” in 20.5 to 45.5%. Two severe events were recorded. “Adequate treatment” was proposed by TB and NTB in respectively 72.7 and 69.3% of cases, “under treatment” in respectively 27.3 and 30.7% and no case of “over treatment” was reported. TB and NTB were ineffective to evaluate the depth of invasion in the submucosa and to differentiate superficial invasive from deep invasive adenocarcinomas.

Biopsies-based strategies are unable to predict depth of cancer invasion and could be associated with a risk of under treatment of large colorectal lesions in near a third of the cases compared to only around 5% with endoscopic characterization. Endoscopic characterization could lead to over treatment, but mainly by endoscopic submucosal dissection with low morbidity. Characterization with the CONECCT classification could decrease the risk of under treatment and avoid surgeries for non-malignant colorectal lesions. Other endoscopic criteria should be determined to better characterize colorectal lesions and to improve the best adapted treatment for each lesion.