EVALUATION OF ARTIFICIAL INTELLIGENCE-ASSISTED COLONOSCOPY FOR ADENOMA DETECTION IN LYNCH SYNDROME: A MULTICENTRE RANDOMIZED CONTROLLED TRIAL (TIMELY STUDY)

Introduction
Artificial intelligence-based polyp detection systems (CADe) have demonstrated to increase lesion detection during colonoscopy in average risk population although this seems to be mainly due to small and non-advance lesions. In Lynch syndrome (LS) adenomas which are the main precursor lesions are usually small with flat morphology and more often display advance histology which makes the detection of these lesions challenging.

Aim: To compare mean number of adenomas per colonoscopy (APC) between CADe versus white light endoscopy (WLE).

Methods
An international, prospective, paralell, randomized, multicenter (16 centers and 30 endoscopists) controlled study was conducted to compare CADe Gi genius Medtronic (intervention) with WLE (control) in individuals harboring pathogenic/likely pathogenic MLH1, MSH2, MSH6, or EpCam variants associated with LS. The randomization 1:1 stratified by center. The procedures, management and resection of lesions was made according to clinical practice and high confidence hyperplastic <5 mm rectosigmoid polyps were left in situ. Hystopathology was the gold standard. Based on previous data the sample size calculation on two negative binomial rates we estimated that 168 to 203 individuals per group will be needed [1-2]

Results
430 patients of 456 elegible individuals were randomized. Sixteen participants were excluded after randomization which led to finally 414 patients (204 CADe arm and 210 WLE arm).Baseline characteristics of patients and procedures were well distributed between both groups Mean age was 48.9 (Standard deviation [SD] 14.3). There were no differences on median of withdrawal time between both groups (13.1 min for CADe vs 12.6 min for WLE p=0.32).

There were no statistically significant differences for the main outcome the overall APC between groups CADe 0.64 (SD 1.59) versus WLE 0.64 (SD 1.17) adjusted rate ratio (aRR)=1.02 (95% CI, 0.71-1.44);p=0.93. Subgroup analysis of adenomas showed no differences on means between both arms for size, morphology (flat versus polypoid) or location. Table 1.

There were no differences on number of serrated lesions per colonoscopy CADe 0.58 (SD 0.94) versus WLE 0.46 (SD 0.95) aRR=1.34 (95% CI,0.94-1.93); p=0.11 but we did find better performance for CADe on 5-9mm subgroup of size 0.15 (SD 0.45) for CADe versus 0.06 (SD 0.24) for WLE aRR=2.99 (95% CI,1.42-6.33); p=0.04. No differences were found on polyp, adenoma or serrated detection rate. Table 2

Based on endoscopists detection rates on high (ADR ≥35%) and low (ADR<35%) among low detectors, there was a trend of higher APC in the CADe arm. For false positives based on histopathology there was a higher detection on CADe arm 0.23(SD 0.7) versus WLE 0.08(SD 0.31) aRR=2.79 (95% CI, 1.35-5.00); p=0.04.

Conclusions
CADe did not improve the detection of adenomas in Lynch syndrome when compared to WLE standard colonoscopy.