High Definition White-Light Colonoscopy vs. Chromoendoscopy for Surveillance of Lynch Syndrome.

Overview

Adenomas in Lynch syndrome have an accelerated progression to colorectal cancer (CRC) which might occur despite a regular follow-up. Despite low evidence, high-definition technology (HD) and indigo-carmine chromoendoscopy (CE) are recommended for surveillance in Lynch syndrome.The investigators will conduct a prospective multicenter randomized non-inferiority study. The principal aim is to compare the adenoma detection rate with WLE vs CE. Our hypothesis is that HD-white-light endoscopy (WLE) is not inferior to CE. Therefore - under expert hands - HD-CE does not add any significant advantage over HD-WLE on adenoma detection rate in patients with Lynch syndrome.

The investigators will conduct a prospective multicenter randomized non-inferiority study. Eligible patients will be those with Lynch syndrome (known germline mutation in mismatch repair genes) who undergo surveillance colonoscopies. Patients will be sequentially assigned in a 1:1 ratio to HD-WLE or HD-CE. The method of stratified randomization based on partial colectomy history will be used to avoid proportion imbalance between groups. Participant centers must have an organized high-risk of CRC clinic and endoscopic unit provided with HD technology. Endoscopists must have a documented high adenoma detection rate and experience in performing CE in patients with high-risk conditions of CRC. The principal aim is to compare the adenoma detection rate with WLE vs CE. Principal outcome measures will be: 1) adenoma detection rate, defined as the proportion of patients with at least one adenoma in each arm; 2) number of adenomas per patient, defined as the total number of detected adenomas in each arm (HD-WLE or HD-CE) divided by the number of colonoscopies in each arm. The sample size calculation was determined for a non-inferiority study. Assuming an ADR of 28% with conventional chromoendoscopy in patients with Lynch syndrome, a 15% non-inferiority margin, a one-sided significance level of 0.05 powered at 80% and a 10% of drop-off. Based on these assumptions, it was determined that 122 patients were required for each arm (a total of 244).