Efficacy of CADe System in Detecting Gastric Neoplasia

NCT07395570 | Active Not Recruiting

• 1 other identifier: CRE-2025.465
• Study Type: interventional
• Target: 260
• Locations: 1 country
• Sites: 1

Brief Summary

Gastric cancer remains the 5th most common cancers worldwide. It also ranked 5th in the cancer related mortality, causing more than 650'000 deaths per year. Survival of gastric cancer is directly related to the stage of the presentation, with early stage cancers having a significantly better survival. Patients with stage I gastric cancer generally have a 5-year survival of more than 90%. In particular, T1a cancer confined to the mucosa are amenable for endoscopic resection, and patients who underwent such treatment have an excellent survival of 97.2% at 5 years. These patients are not only able to survive longer but also with good quality of life through organ preservation. However, diagnosis of gastric cancer at an early stage has always been difficult. A meta-analysis of 22 studies from both East and Western population showed a gastric cancer miss rate of 9.4%. Early gastric cancer usually presents with subtle mucosal changes that are hard to detect endoscopically, especially for endoscopists with limited experience in early cancer diagnosis. Background chronic inflammation and high frequency of non-neoplastic lesions often pose significant diagnostic challenges for endoscopists to detect real neoplastic changes. In high incidence countries such as Japan and Korea, the combination of national screening programme as well as good endoscopy training program facilitated high proportion of early gastric cancer detection. Previous studies have showed that significant survival outcome difference between countries with high versus low early cancer detection rate. Artificial intelligence has emerged as one of the promising technologies that helps enhance endoscopic performance. Numerous high quality randomized studies have demonstrated that computer assisted detection (CADe) system significantly improved colonic adenoma detection rate during screening colonoscopy. Development of gastric cancer CADe system has been much slower than colonic polyp detection. Despite the publication of numerous retrospective studies utilizing endoscopic images in differentiating benign versus malignant gastric lesions, there were only very few completed systems available for clinical real time application. A single centre randomized controlled trial from China demonstrated an improvement in the gastric neoplasm miss rate from 27.3% to 6.1 % when utilizing a novel CADe system. A novel CADe prototype system (OIP-Ge1, Olympus Medical Corporations, Tokyo, Japan) has recently been developed. The system was developed through collaboration of multiple experts in diagnosing early gastric cancer, collecting more than 100'000 endoscopic images from dozens of high volume centres in Japan. There is currently no prospective clinical data on the actual performance of the prototype CADe system, especially when applied in regions with low proportion of early gastric cancer detection. The purpose of this study is to investigate the clinical utility of the new CADe system in detection of gastric neoplasia among high risk patients. If the current study confirms the significant difference in miss rate of gastric neoplasia with the CADe system, a multicentred international randomized controlled trial is planned to compare the efficacy of gastric neoplasia detection with or without the system.

Conditions

Gastric Neoplasia

Keywords

CADe; OIP-Ge1; GC-AI; gastric neoplasia

Outcome Measures

Primary Outcomes (1)

• Proportion of patients with missed gastric neoplasia during first endoscopy without CADe system

  Number of patients in whom at least one histologically confirmed gastric neoplasia (Vienna III-V) is not identified during the first endoscopy without CADe, but is identified during the second endoscopy with CADe, divided by the total number of patients undergoing both examinations (unit: %) Miss rate (%) = \[Number of patients with ≥1 neoplasia detected only on 2nd endoscopy with CADe\] ÷ \[Total number of patients undergoing both 1st and 2nd endoscopies\] × 100.

  1 day

Secondary Outcomes (7)

• Gastric neoplasia detection rate during second endoscopy with CADe System

  1 day

• Time to detect each gastric neoplasia with CADe System

  1 day

• Detection rate of non-neoplastic gastric lesions with CADe System

  1 day

• Miss rate of gastric neoplasia detection with CADe System

  1 day

• Total procedure time for screening endoscopy

  1 day

Study Arms (1)

CADe System Group

EXPERIMENTAL

During the procedure, the stomach would be examined with the assistance of CADe system by endoscopist.

Device: CADe System

Interventions

CADe System DEVICE

The endoscopy would be performed using standardized video processing system (EVIS-X1 CV-1500, Olympus Medical Corporations, Tokyo, Japan) and gastroscope (GIF-EZ1500, GIF-XZ1200). A soft black hood (MAJ-1989) would be attached to the distal end of the endoscope. The video processing system would be connected to the OIP-Ge1 (Olympus Medical Corporations, Tokyo, Japan), the novel CADe system, allowing simultaneous artificial intelligence assisted lesion detection, when turned on using conventional white light imaging (WLI).

CADe System Group

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Age \>=18
• Deemed at high risk of gastric cancer, defined as below:
• Family history of stomach cancer (1st degree relative) for screening, or
• Known gastric atrophy/ intestinal metaplasia requiring surveillance, or
• Suspicious lesion for repeat diagnostic OGD, or
• History of gastric dysplasia / early gastric cancer with endoscopic resection for surveillance
• Newly diagnosed early gastric cancer, workup for synchronous cancers

You may not qualify if:

• History of gastrectomy (For any reason, including benign and malignant disease)
• Patient who refused to participate
• Patients deemed not fit for consent, including minor patients
• Pregnancy
• Other cases deemed by the examining physician as unsuitable for safe treatment

Sponsors & Collaborators

• Chinese University of Hong Kong: lead

Study Sites (1)

Department of Surgery, Faculty of Medicine, the Chinese University of Hong Kong

Hong Kong, Hong Kong

Location

Related Publications (13)

• Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024 May-Jun;74(3):229-263. doi: 10.3322/caac.21834. Epub 2024 Apr 4.

  PMID: 38572751 BACKGROUND

• Hasuike N, Ono H, Boku N, Mizusawa J, Takizawa K, Fukuda H, Oda I, Doyama H, Kaneko K, Hori S, Iishi H, Kurokawa Y, Muto M; Gastrointestinal Endoscopy Group of Japan Clinical Oncology Group (JCOG-GIESG). A non-randomized confirmatory trial of an expanded indication for endoscopic submucosal dissection for intestinal-type gastric cancer (cT1a): the Japan Clinical Oncology Group study (JCOG0607). Gastric Cancer. 2018 Jan;21(1):114-123. doi: 10.1007/s10120-017-0704-y. Epub 2017 Feb 21.

  PMID: 28224238 BACKGROUND

• Pimenta-Melo AR, Monteiro-Soares M, Libanio D, Dinis-Ribeiro M. Missing rate for gastric cancer during upper gastrointestinal endoscopy: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol. 2016 Sep;28(9):1041-9. doi: 10.1097/MEG.0000000000000657.

  PMID: 27148773 BACKGROUND

• Allemani C, Matsuda T, Di Carlo V, Harewood R, Matz M, Niksic M, Bonaventure A, Valkov M, Johnson CJ, Esteve J, Ogunbiyi OJ, Azevedo E Silva G, Chen WQ, Eser S, Engholm G, Stiller CA, Monnereau A, Woods RR, Visser O, Lim GH, Aitken J, Weir HK, Coleman MP; CONCORD Working Group. Global surveillance of trends in cancer survival 2000-14 (CONCORD-3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries. Lancet. 2018 Mar 17;391(10125):1023-1075. doi: 10.1016/S0140-6736(17)33326-3. Epub 2018 Jan 31.

  PMID: 29395269 BACKGROUND

• Huang RJ, Koh H, Hwang JH; Summit Leaders. A Summary of the 2020 Gastric Cancer Summit at Stanford University. Gastroenterology. 2020 Oct;159(4):1221-1226. doi: 10.1053/j.gastro.2020.05.100. Epub 2020 Jul 21.

  PMID: 32707045 BACKGROUND

• Desai M, Ausk K, Brannan D, Chhabra R, Chan W, Chiorean M, Gross SA, Girotra M, Haber G, Hogan RB, Jacob B, Jonnalagadda S, Iles-Shih L, Kumar N, Law J, Lee L, Lin O, Mizrahi M, Pacheco P, Parasa S, Phan J, Reeves V, Sethi A, Snell D, Underwood J, Venu N, Visrodia K, Wong A, Winn J, Wright CH, Sharma P. Use of a Novel Artificial Intelligence System Leads to the Detection of Significantly Higher Number of Adenomas During Screening and Surveillance Colonoscopy: Results From a Large, Prospective, US Multicenter, Randomized Clinical Trial. Am J Gastroenterol. 2024 Jul 1;119(7):1383-1391. doi: 10.14309/ajg.0000000000002664. Epub 2024 Jan 18.

  PMID: 38235741 BACKGROUND

• Gong D, Wu L, Zhang J, Mu G, Shen L, Liu J, Wang Z, Zhou W, An P, Huang X, Jiang X, Li Y, Wan X, Hu S, Chen Y, Hu X, Xu Y, Zhu X, Li S, Yao L, He X, Chen D, Huang L, Wei X, Wang X, Yu H. Detection of colorectal adenomas with a real-time computer-aided system (ENDOANGEL): a randomised controlled study. Lancet Gastroenterol Hepatol. 2020 Apr;5(4):352-361. doi: 10.1016/S2468-1253(19)30413-3. Epub 2020 Jan 22.

  PMID: 31981518 BACKGROUND

• Lau LHS, Ho JCL, Lai JCT, Ho AHY, Wu CWK, Lo VWH, Lai CMS, Scheppach MW, Sia F, Ho KHK, Xiao X, Yip TCF, Lam TYT, Kwok HYH, Chan HCH, Lui RN, Chan TT, Wong MTL, Ho MF, Ko RCW, Hon SF, Chu S, Futaba K, Ng SSM, Yip HC, Tang RSY, Wong VWS, Chan FKL, Chiu PWY; ENDOAID-TRAIN study group. Effect of Real-Time Computer-Aided Polyp Detection System (ENDO-AID) on Adenoma Detection in Endoscopists-in-Training: A Randomized Trial. Clin Gastroenterol Hepatol. 2024 Mar;22(3):630-641.e4. doi: 10.1016/j.cgh.2023.10.019. Epub 2023 Nov 2.

  PMID: 37918685 BACKGROUND

• Seager A, Sharp L, Neilson LJ, Brand A, Hampton JS, Lee TJW, Evans R, Vale L, Whelpton J, Bestwick N, Rees CJ; COLO-DETECT trial team. Polyp detection with colonoscopy assisted by the GI Genius artificial intelligence endoscopy module compared with standard colonoscopy in routine colonoscopy practice (COLO-DETECT): a multicentre, open-label, parallel-arm, pragmatic randomised controlled trial. Lancet Gastroenterol Hepatol. 2024 Oct;9(10):911-923. doi: 10.1016/S2468-1253(24)00161-4. Epub 2024 Aug 14.

  PMID: 39153491 BACKGROUND

• Ochiai K, Ozawa T, Shibata J, Ishihara S, Tada T. Current Status of Artificial Intelligence-Based Computer-Assisted Diagnosis Systems for Gastric Cancer in Endoscopy. Diagnostics (Basel). 2022 Dec 13;12(12):3153. doi: 10.3390/diagnostics12123153.

  PMID: 36553160 BACKGROUND

• Wu L, Shang R, Sharma P, Zhou W, Liu J, Yao L, Dong Z, Yuan J, Zeng Z, Yu Y, He C, Xiong Q, Li Y, Deng Y, Cao Z, Huang C, Zhou R, Li H, Hu G, Chen Y, Wang Y, He X, Zhu Y, Yu H. Effect of a deep learning-based system on the miss rate of gastric neoplasms during upper gastrointestinal endoscopy: a single-centre, tandem, randomised controlled trial. Lancet Gastroenterol Hepatol. 2021 Sep;6(9):700-708. doi: 10.1016/S2468-1253(21)00216-8. Epub 2021 Jul 21.

  PMID: 34297944 BACKGROUND

• Yao K. The endoscopic diagnosis of early gastric cancer. Ann Gastroenterol. 2013;26(1):11-22.

  PMID: 24714327 BACKGROUND

• Muto M, Yao K, Kaise M, Kato M, Uedo N, Yagi K, Tajiri H. Magnifying endoscopy simple diagnostic algorithm for early gastric cancer (MESDA-G). Dig Endosc. 2016 May;28(4):379-393. doi: 10.1111/den.12638. Epub 2016 Apr 22.

  PMID: 26896760 BACKGROUND

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: NA
• Masking: NONE
• Purpose: TREATMENT
• Intervention Model: SINGLE GROUP
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Assistant Professor

Study Record Dates

• First Submitted: January 30, 2026
• First Posted: February 9, 2026
• Study Start: December 1, 2025
• Primary Completion (Estimated): January 31, 2028
• Study Completion (Estimated): July 31, 2028
• Last Updated: February 9, 2026

Record last verified: 2026-02

Data Sharing

• IPD Sharing: Will not share

Locations

HK (1)