Development and clinical application of universal prognostic score for gastrointestinal tumors

2020 Fiscal Year Final Research Report

• Project/Area Number: 19K20403
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 62010:Life, health and medical informatics-related
• Research Institution: Kyushu University
• Principal Investigator: Hideyuki Shimizu 九州大学, 生体防御医学研究所, 学術研究員 (70826263)
• Project Period (FY): 2019-04-01 – 2021-03-31
• Keywords: 個別化医療 / 層別化 / 人工知能 / 消化器がん

Outline of Final Research Achievements

Cancer is the leading cause of death among Japanese people, and among them, there are many patients with colorectal and gastric cancer. In the treatment methods that have been used so far, patients have been divided into several groups, and treatment within each group has been generally the same. However, from the perspective of personalized medicine, more precise stratification is desirable. Therefore, we used statistics and machine learning to unravel the relationship between prognosis and gene expression in thousands of patients, and created a score called mPS_colon. The usefulness of this score was validated in a number of patient cohorts, and it outperformed existing stratification methods. We hope that it will be applied to clinical practice in the future.

Free Research Field

人工知能, がん, 分子生物学, システム生物学

Academic Significance and Societal Importance of the Research Achievements

これまで大腸がん・胃がんにおいてさまざまな予後規定遺伝子が報告されてきたが、いずれもごく限られた患者データや実験データから得られた報告であった。今回我々は数千人という大規模の患者データを統合し、具体的な予後層別化スコアを提唱したところに大きな学術的な意義がある。また社会的にも、このシステムを使えば本来は化学療法をしなくてもいい患者さんを同定できるため、過剰な治療を避けることにつながる。