Synergistic Effect of G-Eye Balloon for Behind the Folds Visualization With CADe (Discovery System) on Adenoma Detection Rate. (DiscoveryIII)

Synergistic Effect of G-Eye Balloon for Behind the Folds Visualization With CADe (Discovery System) on Adenoma Detection Rate.

Synergistic Effect of G-Eye Balloon for Behind the Folds Visualization With Artificial Intelligence Assisted Polyp Detection (Discovery System) on Adenoma Detection Rate. 'Discovery III Study'

Colonoscopy is the gold standard for colorectal screening. The diagnostic accuracy of colonoscopy highly depends on the quality of inspection of the colon during the procedure. To increase detection new polyp detection systems based on artificial intelligence (AI) have been developed. However, these systems still depend on the ability of the endoscopist to adequately visualize the complete colonic mucosa, especially to detect smaller and more subtle lesions, or lesions hidden behind folds in the colon. With this study we want to combine a device to flatten the folds in the colon combined with an artificial intelligence system to further improve the detection rate of lesions during colonoscopy.

Rationale: Colonoscopy is the gold standard for CRC screening. The adenoma detection rate (ADR) is the most important quality parameter for colonoscopy, because of its inverse association with the risk of interval CRC. Yet, the adenoma miss rate (AMR) in conventional colonoscopy is reported in meta-analyses to vary between 22-26%. The diagnostic accuracy of colonoscopy highly depends on the quality of inspection of the colonic mucosa during the procedure. To increase the adenoma detection rate (ADR), new polyp/adenoma detection systems based on artificial intelligence (AI) have been developed, i.e., computer assisted detection (CADe). However, these systems still depend on the ability of the endoscopist to adequately visualize the complete colonic mucosa, especially to detect smaller and more subtle lesions. Therefore, we hypothesize that ADR can further be improved by combining a CADe system, the Discovery system, with a behind the fold (BTF) visualization technique, the G-Eye. Study design: International multicenter prospective interventional cohort, compared with a cohort from the Discovery II study (NL73127.091.20, trial code NL9135). All subjects will be undergoing colonoscopy with a combined BFT and CADe assisted approach. Outcomes will be corrected for confounders using regression modeling. Study population: 194 Adult patients (>18 years) from 3 hospitals, scheduled for diagnostic, screening (non-iFOBT based), or surveillance colonoscopy. Exclusion criteria: inflammatory bowel disease (IBD), known polyp or tumor upon referral, therapeutic procedure (e.g. endoscopic mucosal resection), prior surgical resection of any portion of the colon, American Society of Anesthesiologists score of ≥3, Inadequately corrected anticoagulation disorders or anticoagulation medication use, inability to provide informed consent. Main study endpoints: The primary objective of the present study is to compare ADR between CADe assisted and a combined BTF/CADe assisted colonoscopy. Secondary objectives include advanced neoplasia rate (including advanced adenomas and/or CRC), polyp detection rate, size and histopathology, mean number of polyps per patient, procedure times, bowel cleaning levels, adverse events, inter-operator variability. Nature and extent of the burden and risks associated with participation, benefit and group relatedness: Eligible patients who are scheduled for surveillance colonoscopy will undergo one BFT and CADe assisted colonoscopy. There will be no burden for participants regarding the colonoscopy procedure. Colonoscopy is a commonly performed procedure, and the overall serious adverse event rate is low with an estimated risk 2.8 per 1000 colonoscopies. The risk of experiencing a (serious) adverse event with G-Eye and Discovery guided colonoscopy is believed to be equivalent to conventional colonoscopy. The benefit for patients is a higher likelihood of lesion detection during colonoscopy.

International multicenter prospective interventional cohort, compared with a cohort from the Discovery II study (NL73127.091.20, trial code NL9135).

This multicenter single-arm colonoscopy trial will compare colonoscopy assisted by both balloon-BFT visualizing and CADe to CADe-only assisted colonoscopy in diagnostic, screening (non-iFOBT based)), or surveillance colonoscopy. The latter group will be selected from the interventional arm from the Discovery II trial, using corresponding participating centers. All subjects in this trial will undergo the same treatment, i.e., combined balloon and CADe-assisted colonoscopy.

colonoscopy assisted by both balloon-Behind The Folds visualizing and Computer assisted detection (CADe)

All participants will be subjected to a colonoscopy procedure, assisted by both G-eye balloon and the Discovery CADe system. Standard of care regarding the colonoscopy procedure will be applied to all study subjects. Any lesions detected during the procedure will be removed directly

Calculated as the number of patients in whom at least one adenoma is detected during the colonoscopy procedure, divided by the total number of patients that underwent the colonoscopy procedure.

calculated as the number of patients in whom at least one advanced adenoma is detected during the colonoscopy procedure, divided by the total number of patients that underwent the colonoscopy procedure

calculated as the number of patients in whom at least one polyp is detected during the colonoscopy procedure, divided by the total number of patients that underwent the colonoscopy procedure

calculated as the number of patients in whom at least one SSL is detected during the colonoscopy procedure, divided by the total number of patients that underwent the colonoscopy procedure

Morphological characteristics of the lesion using the Paris classification (Ip, Is, IIa, IIb, IIc, III)

Morphological characteristics of the lesion using the Paris classification (Ip, Is, IIa, IIb, IIc, III)

ADR of the first 20% of patients that have undergone colonoscopy by each endoscopist will be compared with the final 20% of patients in each arm to identify any changes in ADR throughout the trial

ADR of the first 20% of patients that have undergone colonoscopy by each endoscopist will be compared with the final 20% of patients in each arm to identify any changes in ADR throughout the trial

Procedure times with both techniques (i.e., total procedure time, mean polypectomy time and withdrawal time)

Procedure times with both techniques (i.e., total procedure time, mean polypectomy time and withdrawal time)

SAEs will be subcategorized into colonoscopy related, cardiac (cardiac ischemia, heart failure, arrhythmia, other) pulmonary (exacerbation COPD, infectious, other), neurological (stroke, cerebrovascular accident, bleeding, other), and other (surgical interventions etc.)

Inclusion Criteria: Adult patients (>18 years) Referred and scheduled for diagnostic, screening (non-iFOBT based), or surveillance colonoscopy. Exclusion Criteria: Inflammatory bowel disease (IBD) Known polyp or tumor upon referral Therapeutic procedure (e.g., endoscopic mucosal resection) Prior surgical resection of any portion of the colon American Society of Anesthesiologists score of ≥3 Inadequately corrected anticoagulation disorder or anticoagulation medication use inability to provide informed consent.

The dataset used during this study is available from the corresponding author upon reasonable request.