Impact of Computer-aided Optical Diagnosis (CAD) in Predicting Histology of Diminutive Rectosigmoid Polyps: a Multicenter Prospective Trial (ABC Study). (ABC)

Impact of Computer-aided Optical Diagnosis (CAD) in Predicting Histology of Diminutive Rectosigmoid Polyps: a Multicenter Prospective Trial (Artificial Intelligence BLI Characterization - ABC Study).

Recently, a CNN-based artificial intelligence (AI) system for polyp characterization has been developed by Fujifilm Co., Tokyo, Japan. It works in conjunction with BLI system. In the present study we prospectively evaluate whether the evaluation of the endoscopist combined with the CAD system output achieve > 90% accuracy in characterization (i.e. as adenomas or non-adenomas) of diminutive rectosigmoid polyps having histopathology as reference standard. Consecutive adult outpatients undergoing elective colonoscopy, in which at least one diminutive (<5 mm) rectosigmoid polyp is detected are included. During endoscopic procedures all polyps identified by the endoscopist are documented for size, location and morphology. All diminutive polyps are characterized by a three sequential steps process: I) endoscopist prediction: the endoscopist evaluates the polyp by using BLI through the BASIC classification; the confidence level (high vs. low) in histology prediction is recorded; II) AI prediction: the AI system is switched on and the output of the automatic evaluation is recorded; this outcome is rated as stable or unstable, depending of the consistency over time of the outcome; III) combined prediction: a final classification is provided by endoscopist in light of the results of the first and of the second step; the confidence level is recorded. All polyps are resected and retrieved in separate jars and sent for pathology assessment. Only polyps characterized with high confidence will be included in the per-polyp analysis; the high-confidence characterization rate will be also calculated; the rate of polyps characterized with a CAD stable outcome will be calculated. Operative characteristics (sensitivity, specificity, positive and negative predictive value and accuracy) in distinguishing adenomatous from non-adenomatous polyps, evaluated with high confidence, will be calculated for each diminutive polyp and for each diminutive rectosigmoid polyp, having histopathology report as reference standard. The post-polypectomy surveillance intervals will be calculated on the basis of polyp histology (reference standard) in all patients according to both USMSTF and ESGE guidelines.

Study Overview

• Status: Completed
• Conditions: Colonic Adenomatous Polyp
• Intervention / Treatment: Diagnostic test: Polyp carachterization by combing endoscopist evaluation and Ai output

• Study Type: Observational
• Enrollment (Actual): 1134

Contacts and Locations

Study Locations

• Italy
  • Como, Italy, 22100
    • Gastroenterology Unit, Valduce Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 85 years (Adult, Older Adult)
• Accepts Healthy Volunteers: N/A
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

Consecutive adult outpatients undergoing elective colonoscopy, in which at least one diminutive (<5 mm) rectosigmoid polyp is detected

Description

Inclusion Criteria:

• Consecutive adult outpatients undergoing elective colonoscopy, in which at least one diminutive (<5 mm) rectosigmoid polyp is detected.

Exclusion Criteria:

• patients with CRC history or hereditary polyposis syndromes or hereditary non-polyposis colorectal cancer
• patients with inadequate bowel preparation
• patients scheduled for partial examinations
• polyps could not be resected due to ongoing anticoagulation preventing resection and pathologic assessment
• patients undergoing urgent colonoscopy

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Patients with at least one diminutive rectosigmoid polyp; Consecutive adult (>18 years) outpatients undergoing elective colonoscopy, in which at least one diminutive (<5 mm) rectosigmoid polyp is detected. Exclusion criteria: patients with CRC history or hereditary polyposis syndromes or hereditary non-polyposis colorectal cancer; patients with inadequate bowel preparation; patients in which caecal intubation was not achieved or scheduled for partial examinations; polyps could not be resected due to ongoing anticoagulation preventing resection and pathologic assessment; patients undergoing urgent colonoscopy.

Diagnostic test: Polyp carachterization by combing endoscopist evaluation and Ai output; A polyp characterization (adenoma vs. non adenoma) is provided by endoscopist in light of the results of this own evaluation and of the Ai system output. The confidence level (high vs. low) in polyp characterization is recorded. The combined evaluation is compared with histopathology results.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Agreement of combined prediction with PIVI I statement	To prospectively evaluate whether the evaluation of the endoscopist combined with the CAD system output achieve > 90% accuracy in characterization (i.e. as adenomas or non-adenomas) of diminutive rectosigmoid polyps (i.e. PIVI I threshold) having histopathology as reference standard.	6 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Endoscopist prediction	to calculate the performance measures (sensitivity, specificity, positive and negative predictive value) of the endoscopist alone in characterizing diminutive rectosigmoid polyps	6 months
Ai prediction	- to calculate the performance measures (sensitivity, specificity, positive and negative predictive value) of the AI system alone in characterizing diminutive rectosigmoid polyps	6 months
Agreement of combined prediction with PIVI II statement	- to evaluate if the evaluation of the endoscopist combined with the CAD system output achieve > 90% accuracy in the assignment of post-polypectomy surveillance intervals, according to US and EU guidelines, when combined with the histopathology assessment of polyps >5 mm in size	6 months

Collaborators and Investigators

• Sponsor: Valduce Hospital

Publications and helpful links

General Publications

• Rondonotti E, Hassan C, Tamanini G, Antonelli G, Andrisani G, Leonetti G, Paggi S, Amato A, Scardino G, Di Paolo D, Mandelli G, Lenoci N, Terreni N, Andrealli A, Maselli R, Spadaccini M, Galtieri PA, Correale L, Repici A, Di Matteo FM, Ambrosiani L, Filippi E, Sharma P, Radaelli F. Artificial intelligence-assisted optical diagnosis for the resect-and-discard strategy in clinical practice: the Artificial intelligence BLI Characterization (ABC) study. Endoscopy. 2023 Jan;55(1):14-22. doi: 10.1055/a-1852-0330. Epub 2022 May 13.

Study record dates

Study Major Dates

• Study Start (Actual): October 22, 2020
• Primary Completion (Actual): February 27, 2021
• Study Completion (Actual): March 30, 2021

Study Registration Dates

• First Submitted: October 22, 2020
• First Submitted That Met QC Criteria: October 22, 2020
• First Posted (Actual): October 28, 2020

Study Record Updates

• Last Update Posted (Actual): June 10, 2021
• Last Update Submitted That Met QC Criteria: June 9, 2021
• Last Verified: June 1, 2021

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Neoplasms by Histologic Type; Neoplasms; Neoplasms, Glandular and Epithelial; Pathological Conditions, Anatomical; Adenoma; Polyps; Adenomatous Polyps
• Other Study ID Numbers: 599/2020

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No