Study to Evaluate the Diagnostic Performance of of MAGENTIQ-COLO During Colonoscopy.

A Multicenter Study to Evaluate the Diagnostic Performance of Computer-aided Polyp Size Classification of the MAGENTIQ-COLO During Real-time Colonoscopy

This is an international, multicenter, study to evaluate the diagnostic performance of the CADx polyp sizing modality of the MAGENTIQ-COLO.

Colonoscopy is the gold standard for the detection and removal of premalignant colorectal polyps. Recommended post-polypectomy surveillance intervals are primarily based on pathological diagnosis and polyp size. However, accurate estimation of polyp size remains challenging, potentially influencing post-polypectomy surveillance intervals. Inaccuracies in size estimation may lead to either unnecessary or prematurely scheduled surveillance colonoscopies when overestimating size or lead to an increased risk of post-colonoscopy colorectal cancer or advanced neoplasia when underestimating size. Furthermore, diminutive (1-5mm) polyps pose challenges due to their high incidence and frequent pathological assessment. Proposed strategies to reduce this burden, such as the European Society of Gastrointestinal Endoscopy (ESGE) 'resect-and-discard' strategy, are infrequently used as non-expert endoscopists often do not meet diagnostic thresholds when using standard visual inspection. The MAGENTIQ-COLO computer-aided diagnosis (CADx) system by Magentiq Eye LTD, Haifa, Israel, addresses these challenges by providing real-time polyp size estimation and polyp characterization. Additionally, this study evaluates the diagnostic accuracy of the MAGENTIQ-COLO CADx system in an average-risk screening and surveillance colonoscopy population.

The primary objective of the study is to assess the diagnostic performance of the endoscopist performing a MAGENTIQ-COLO CADx-assisted colonoscopy to classify polyps as diminutive (≤5mm) or non-diminutive (>5mm) compared to the size classification using open biopsy forceps, or polypectomy snares, of known diameter.

This will be measured by comparing the sensitivity and specificity between the two size classifications. Secondary objective include:

• To assess the diagnostic performance of the endoscopist performing a MAGENTIQ-COLO CADx-assisted colonoscopy to diagnose diminutive (rectosigmoid) colorectal polyps as neoplastic (adenoma or sessile serrated lesion (SSL)) with high-confidence compared to the pathology diagnosis. This will be measured by the sensitivity and specificity between the two diagnoses;
• To assess the diagnostic performance of the CADx system in measuring the size in millimeters and the size classification of colorectal polyps compared to the size measurement using open biopsy forceps, or polypectomy snare, of known diameter; this correlation will also be conducted for classifying polyps into diminutive (≤5mm), small (6-9mm), and advanced (≥10mm) size categories.

The sensitivity of the CADx-assisted optical diagnosis in classifying colorectal polyps as diminutive (≤5mm) or non-diminutive (>5mm) compared to the reference gold standard, which is the size classification of the colorectal polyp using open biopsy forceps, or polypectomy snare, of known diameter.

This is an international, multicenter, study to evaluate the diagnostic performance of the CADx polyp sizing modality of the MAGENTIQ-COLO. Study subjects who are already referred for screening or surveillance colonoscopy, will undergo colonoscopy with the real-time use of the MAGENTIQ-COLO technology. Endoscopists will assess all colorectal polyps detected during colonoscopy with and without the MAGENTIQ-COLO. Diagnostic performance of polyp size classification and optical diagnosis with and without MAGENTIQ-COLO is evaluated with reference to open biopsy forceps, or polypectomy snare of known diameter size classification, and pathology-based diagnosis, respectively. There is no formal study subject follow-up. Study procedures will be performed intraprocedural during the colonoscopy. Final pathological diagnosis will be recorded from the electronic health record.

The unit of analysis is the colorectal polyp rather than a study subject. The study is planned to include 396 colorectal polyps. Based on an expected detection rate of approximately 1.20 polyps per colonoscopy in the study population, we assume that enrollment of 330 subjects will be sufficient to meet the study objectives.

Inclusion Criteria:

• Individuals aged ≥45 - ≤80 years old, who are scheduled for non-iFOBT screening or surveillance colonoscopy.

Exclusion Criteria:

1. In situ polyps with known histology detected in a previous colonoscopy.
2. No colorectal polyps detected during colonoscopy.
3. Known or suspected inflammatory bowel disease.
4. Polyposis syndrome (e.g., familial adenomatous polyposis, serrated polyposis).
5. Non-hereditary polyposis syndromes (e.g. Lynch syndrome).
6. History of chemotherapy or radiation therapy for colorectal lesions.
7. Pregnancy.
8. Has a referral for therapeutic procedure (i.e., endoscopic mucosal resection, intervention to stop a lower gastro-intestinal bleeding, etc.).
9. Inability to undergo polypectomy (e.g., incorrect continued use of anticoagulants, comorbidities) or patient refusal, as assessed by the endoscopist.
10. Inability to provide informed consent.