early gastric cancer diagnostic panel using human Helicobacter pylori molecular markers + endoscopic AI diagnosis

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K07901
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 53010:Gastroenterology-related
• Research Institution: St. Marianna University School of Medicine
• Principal Investigator: Watanabe Yoshiyuki 聖マリアンナ医科大学, 医学部, 講師 (90329243)
• Co-Investigator(Kenkyū-buntansha): 山本 博幸 聖マリアンナ医科大学, 医学研究科, 教授 (40332910)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 早期胃がん / 遺伝子メチル化異常 / ピロリ菌遺伝子異常 / 胃洗浄廃液 / ESD / 内視鏡AI診断 / 診断マーカー / 白色光

Outline of Final Research Achievements

This research aims to diagnose gastric cancer at an extremely early stage by adding AI diagnosis to human plus H. pylori genomic abnormality molecular markers using gastric wash fluid. As a result, compared to the AUC (MINT25:0.465, SOX17:0.554, miR34:0.490, BARHL2:0.659) for gastric cancer molecular markers, the AUC (Area Under the ROC Curve) for AI diagnosis was very high at 0.999, and interestingly, NBI , AI diagnosis based on white light observation images tended to be better than magnified observation or indigo carmine staining (p=0.662, p<0.0001, P<0.0001).It was suggested that gastric cancer molecular markers may be effective for predictive diagnosis, and AI diagnosis for identification diagnosis. A paper will be submit to Journal of Clinical Laboratory Analysis soon.

Free Research Field

消化器

Academic Significance and Societal Importance of the Research Achievements

従来からの、胃内視鏡検査による診断能の限界を、胃洗浄廃液を用いた分子診断とAI診断を組み合わせることで、医師が内視鏡下、肉眼診断が困難な病変も診断可能となる可能性が示唆された。医師偏在、首都圏以外の専門医不足問題に対する解説策の一つとなる可能性から、社会的な意義は高いと考えている。