Artificial intelligence-aided colonoscopy did not improve outcomes in Lynch syndrome

Computer-aided detection (CADe) colonoscopy was not associated with higher adenoma detection rates (ADR) compared with white light endoscopy (WLE) in participants with Lynch syndrome. Additionally, the false positive rate in the CADe group was higher than in the WLE group, according to the results of a randomised-controlled trial.

“Evidence indicates that artificial intelligence improves ADR, particularly in small non-advanced lesions,” said Mr Oswaldo Ortiz (Hospital Clinic Barcelona, Spain) [1]. “In patients with Lynch syndrome, it is extra important to detect every adenoma, since these patients tend to progress faster towards adenocarcinoma” [2]. Mr Ortiz and colleagues designed a randomised-controlled trial to compare the performance of CADe colonoscopy with WLE in 430 participants with Lynch syndrome [3]. Adenoma per colonoscopy was the primary endpoint of the study.

The mean adenoma per colonoscopy was 0.64 in both study arms, showing no difference between the 2 study procedures (RR 1.00; 95% CI 0.73–1.27; P=0.42). Stratifying the outcomes by location, size, or morphology did not reveal significant differences between the 2 study arms. Also, polyp detection rates and ADR were similar for the 2 interventions. However, the rate of serrated lesions per colonoscopy was higher in the CADe arm than in the WLE arm (0.58 vs 0.46; RR 1.26; P=0.01). Finally, Mr Ortiz noted that the false positive rate was higher in the CADe arm than in the WLE arm (0.21 vs 0.08; RR 2.8; P=0.04).

1. 
   
   1. Hassan C, et al. Ann Intern Med. 2023;176:1209–1220.
   2. Ahadova A, et al. Int J. Cancer. 2021;148:800–811.
   3. Ortiz O, et al. Evaluation of artificial intelligence-assisted colonoscopy for adenoma detection in Lynch syndrome: a multicentre randomized controlled trial (timely study). LB15, UEG Week 2023, 14–17 October, Copenhagen, Denmark.

Copyright ©2023 Medicom Medical Publishers



Posted on 7 December 20237 December 2023