臨床研究等提出・公開システム

An exploratory randomized controlled trial evaluating the efficacy and safety of third-generation narrow band imaging, texture and color enhancement imaging, and white light imaging in detecting the early gastric cancer and/or gastric adenoma (3G-detection trial)

3G-detection trial (3G-detection trial)

901

Group A were male 228 (75.8%) and median age 73 years (46-85 years). Upper gastrointestinal endoscopy was performed in 174 cases for follow-up (stomach), 50 (esophageal), and 76 for detailed examination, respectively. Medical history was 170 cases of early gastric cancer, 10 cases of gastric adenoma, 56 cases of esophageal cancer (endoscopic treatment), 2 cases of esophageal cancer (radiation or chemotherapy), and 74 cases of none, respectively. In addition, 44 had no history of Helicobacter pylori infection, 198 had a history, and 59 had an unknown. The breakdown of those with a past history was 21 who had not been eradicated, 168 who had successfully eradicated, 5 who had failed to eradicate, and 4 who had not been tested after eradication. There were 258 patients who had not taken antiplatelet drugs or anticoagulants. Group B were male 228 (76.0%) and median age 73 years (42-85 years). Upper gastrointestinal endoscopy was performed in 179 cases for follow-up (stomach), 51 (esophageal), and 70 for detailed examination, respectively. Medical history was 180 cases of early gastric cancer, 8 cases of gastric adenoma, 58 cases of esophageal cancer (endoscopic treatment), 2 cases of esophageal cancer (radiation or chemotherapy), and 62 cases of none, respectively. In addition, 38 had no history of Helicobacter pylori infection, 208 had a history, and 54 had an unknown. The breakdown of those with a past history was 29 who had not been eradicated, 170 who had successfully eradicated, 6 who had failed to eradicate, and 3 who had not been tested after eradication. There were 239 patients who had not taken antiplatelet drugs or anticoagulants. Group C were male 227 (75.7%) and median age 73 years (35-85 years). Upper gastrointestinal endoscopy was performed in 177 cases for follow-up (stomach), 48 (esophageal), and 75 for detailed examination, respectively. Medical history was 192 cases of early gastric cancer, 7 cases of gastric adenoma, 55 cases of esophageal cancer (endoscopic treatment), 9 cases of esophageal cancer (radiation or chemotherapy), and 60 cases of none, respectively. In addition, 37 had no history of Helicobacter pylori infection, 216 had a history, and 47 had an unknown. The breakdown of those with a past history was 23 who had not been eradicated, 179 who had successfully eradicated, 6 who had failed to eradicate, and 8 who had not been tested after eradication. There were 240 patients who had not taken antiplatelet drugs or anticoagulants.

The initial observation was performed in 899 patients (group A 299, group B 300, and group C 300, respectively), excluding 2 patients who discontinued the study before the start of endoscopy. The target lesions observed at the initial observation were 29 lesions suspected of gastric adenoma (group A 8, group B 16, group C 5, respectively), 166 lesions suspected of gastric cancer (group A 49, group B 47, group C 70, respectively), for a total of 195 lesions. Treatment for target lesions included 40 lesions by endoscopic resection (group A 16, group B 14, group C 10, respectively), 5 lesions by surgical resection ( group A 2, group B 2, group C 1, respectively) and 150 lesions (group A 39, group B 47, group C 64, respectively) without treatment. In addition, the target lesions observed in the second observation were 4 lesions suspected of gastric adenoma ( group B 2, group C 2), 29 lesions suspected of gastric cancer (group A 9, group B 9, group C 11, respectively), totaling 33 lesions. Treatment for target lesions included 8 lesions by endoscopic resection (group A 3, group B 3, group C 2, respectively), 1 lesion by surgical resection in group C, and 24 lesions (group A 6, group B 8, group C 10, respectively) without treatment.

Adverse events based on physician judgment / CTCAE ver5.0-JCOG No adverse events occurred in this study. Failure rate Target: All treated cases (n=896) Group A (1st:WLI, 2nd:WLI):0/297=0% (95% CI 0-1.2%) Group B (1st:NBI, 2nd:WLI):1/299=0.3% (95% CI 0.0-1.8%) Group C (1st: TXI, 2nd:WLI):0/300=0% (95% CI 0-1.2%)

Primary endpoint Detection ratio of gastric epithelial tumor lesions in non-magnified observation at initial observation (3-group comparison) Target: All registered cases (n=901) Group A (WLI):17/301=5.6% (95% CI 3.3-8.9%) Group B (NBI):22/300=7.3% (95% CI 4.7-10.9%), p-value (Group A vs Group B)=0.4135 Group C (TXI):15/300=5.0% (95% CI 2.8-8.1%), p-value (group A vs group C)=0.8562 The point estimate of the detection rate of gastric epithelial tumor lesions in groups B (NBI) and C (TXI) in the unmagnified observation of the initial observation is 7.3% in group B (NBI) >5.0% in group C (TXI). Therefore, group B (NBI) was judged to be a more promising observation method. Observation method point estimate of >-1.0% for the selected study relative to the control group A (WLI) Because it exceeded, it was judged that the selected group B (NBI) was appropriate as the test observation group for the next confirmatory study. Secondary endpoint (1)Detection of overlooked gastric epithelial tumor lesions (lesions detected in non-magnified observation at the second observation) Proportion (comparison of 3 groups) Target: All registered cases (n=901) Group A (1st:WLI, 2nd:WLI):3/301=1.0% (95% CI 0.2-2.9%) Group B (1st:NBI, 2nd:WLI):3/300=1.0% (95% CI 0.2-2.9%), p-value (group A vs group B) =1.0000 Group C (1st:TXI, 2nd:WLI):2/300=0.7% (95% CI 0.1-2.4%), p-value (group A vs. group C) =1.0000 (2) Proportion of detection of gastric cancer lesions in non-magnified observation at initial observation (comparison of 3 groups) Target: All registered cases (n=901) Group A (WLI):17/301=5.6% (95% CI 3.3-8.9%) Group B (NBI):17/300=5.7% (95% CI 3.3-8.9%), p-value (Group A vs Group B)=1.0000 Group C (TXI):12/300=4.0% (95% CI 2.1-6.9%), p-value (Group A vs Group C)=0.4470 (3) Positive predictive value (PPV) for gastric epithelial tumor lesion diagnosis in non-magnified observation at initial observation (comparison of 3 groups) Subjects: 195 lesions detected by non-magnifying observation in the initial observation Group A (WLI):21/57=36.8% (95% CI 24.4-50.7%) Group B (NBI):23/63=36.5% (95% CI 24.7-49.6%), p-value (Group A vs Group B)=1.0000 Group C (TXI):16/75=21.3% (95% CI 12.7-32.3%), p-value (Group A vs Group C)=0.0538 (4) Sensitivity and specificity of gastric cancer diagnosis by NBI magnifying observation (including near-focus observation) Subjects: A total of 228 lesions with target lesions obtained in the first and second observations [Sensitivity] 43/61=70.5% (95% CI 57.4-81.5%) [Specificity] 148/167=88.6% (95% CI 82.8-93.0%)

Efficacy analysis (primary endpoint) detection rate of gastric epithelial tumor lesions in non-magnification observation at initial observation (comparison of 3 groups) was Group B (NBI) 7.3% > Group C (TXI) 5.0%, and B Group (NBI) exceeded the control group A (WLI) by 5.6% by more than 1.0%. Based on this, group B (NBI) is a more promising observation method and was judged to be appropriate as the observation group for the next confirmatory study.

May. 29, 2023

No

No

https://jrct.niph.go.jp/latest-detail/jRCT1032210213

Yano Tomonori

National Cancer Center Hospital East

6-5-1, Kashiwanoha, Kashiwa, Chiba 277-8577, Japan

toyano@east.ncc.go.jp

Kadota Tomohiro

National Cancer Center Hospital East

6-5-1, Kashiwanoha, Kashiwa, Chiba 277-8577, Japan

+81-4-7133-1111

tkadota@east.ncc.go.jp

Complete

Aug. 01, 2021

Interventional

randomized controlled trial

open(masking not used)

active control

parallel assignment

diagnostic purpose

1) Aged 20 to 85 years old at the time of obtaining consent.
2) Fulfill any of the following [1]-[3].
[1] Have a history of endoscopic treatment for early gastric cancer and/or gastric adenoma, and endoscopic examination is scheduled on the purpose of follow-up.
[2] Have a history of endoscopic treatment, chemotherapy, or radiotherapy for esophageal cancer, and endoscopic examination is scheduled on the purpose of follow-up.
[3] Has a lesion diagnosed as gastric cancer, gastric adenoma, or esophageal cancer, and endoscopic examination is scheduled on the purpose of detailed examination.
3) Written informed consent has been obtained for participation in the study.

1) Have a history of gastrectomy.
2) Have a history of gastric tube reconstruction after esophagectomy.
3) Judged as inappropriate for carrying out this study, including the underlying disease.

Both

gastric cancer,esophageal cancer

arm A: First observation: WLI + Second observation: WLI
arm B: First observation: NBI + Second observation: WLI
arm C: First observation: TXI + Second observation: WLI

Proportion of patients with gastric cancer and/or adenoma detected by non-magnifying endoscopy of the first observation

1) Proportion of patients with overlooked gastric cancer and/or adenoma (detected by non-magnifying endoscopy of the second observation)
2) Proportion of patients with gastric cancer detected by non-magnifying endoscopy of the first observation
3) Positive predictive value (PPV) for diagnosis of gastric cancer and/or adenoma in non-magnifying endoscopy of the first observation
4) Sensitivity and specificity of gastric cancer diagnosis by NBI with magnifying endoscoy (including near-focus mode)
5) Defect rate
6) Proportion of adverse events
7) Observation time

+81-4-7133-1111

June. 15, 2021

none