Computer-Assisted Colonoscopy in High–Adenoma Detection Rate Settings in a High-Risk Population: A Randomized Clinical Trial
Mené à Taïwan auprès de 1 356 participants (âge moyen : 60 ans ; 50 % de femmes et 50 % d'hommes), cet essai multicentrique randomisé évalue la non infériorité, par rapport à une coloscopie standard et du point de vue du taux de détection d'adénomes colorectaux, d'une coloscopie avec détection assistée par ordinateur
Importance : Computer-aided detection (CAD) systems can enhance adenoma detection, but their effectiveness in high-performance settings and among patients with positive fecal immunochemical test (FIT) results remains uncertain.
Objective : To evaluate the impact of CAD on adenoma detection in routine practice, focusing on patients with positive FIT results.
Design, Setting, and Participants : This multicenter, open-label, randomized clinical trial was conducted at 4 tertiary hospitals in Taiwan from February 23, 2022, to November 27, 2024. Adults aged 40 to 79 years who were scheduled for a colonoscopy owing to FIT positivity, symptoms, screening, or surveillance were randomized 1:1 to CAD-assisted or standard colonoscopy. Data were analyzed from December 1, 2024, to February 28, 2025.
Exposures : Colonoscopy performed with a real-time CAD system or standard high-definition colonoscopy.
Main Outcomes and Measures : The primary outcome was adenoma detection rate (ADR), defined as the proportion of patients with at least 1 histologically confirmed adenoma. Secondary outcomes included adenomas per colonoscopy (APC), sessile serrated lesion detection rate (SSLDR), and postpolypectomy surveillance intervals according to the US Multi-Society Task Force (USMSTF) and European Society of Gastrointestinal Endoscopy criteria.
Results : Of 1356 randomized participants (mean [SD] age, 60.0 [9.4] years; 678 [50.0%] female and 678 [50.0%] male), CAD-assisted colonoscopy met noninferiority criteria for ADR compared with standard colonoscopy (395 of 675 [58.5%] vs 363 of 681 [53.3%]; absolute difference, 5.2 percentage points [95% CI, −0.1 to 10.5 percentage points]). Superiority was not statistically significant. CAD significantly increased mean (SD) APC (1.41 [1.95] vs 1.20 [1.88]; P = .01), driven mainly by detection of diminutive adenomas. In exploratory analyses of 864 patients with FIT-positive findings, CAD significantly increased ADR (288 of 441 [65.3%] vs 243 of 423 [57.4%]; P = .02; adjusted odds ratio [AOR], 1.39 [95% CI, 1.05-1.86]) and APC (mean [SD], 1.64 [2.08] vs 1.39 [2.09]; P = .01). SSLDR did not differ between groups. Consequently, CAD led to more intensive surveillance recommendations under USMSTF criteria, particularly in patients with FIT-positive findings (58 of 441 [13.2%] vs 31 of 423 [7.3%]; AOR, 1.94 [95% CI, 1.22-3.09]).
Conclusions and Relevance : In this randomized clinical trial, CAD-assisted colonoscopy met noninferiority criteria for adenoma detection. Superiority was not statistically significant overall, with significant improvements limited to the exploratory FIT-positive subgroup, driven largely by diminutive adenomas. CAD also increased intensive surveillance assignments. The incremental benefit of CAD in reducing interval cancer risk requires further investigation.
Trial Registration : ClinicalTrials.gov Identifier: NCT03842059
JAMA Network Open , article en libre accès, 2026