International prognostic algorithm for allogeneic stem cell transplantation by machine learning of comprehensive HLA and clinical information

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K08391
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 54010:Hematology and medical oncology-related
• Research Institution: Kyoto University
• Principal Investigator: KANDA JUNYA 京都大学, 医学研究科, 講師 (30636311)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 同種造血幹細胞移植 / 臍帯血移植 / 機械学習

Outline of Final Research Achievements

We conducted an analysis using the databases of the Japanese Society for Transplantation and Cell Therapy and Eurocord. In this study, we demonstrated the impact of T-cell epitope matching on graft-versus-host disease (GVHD) prevention strategies after HLA-mismatched related donor transplantation. Furthermore, regarding the mid- to long-term complications after allogeneic transplantation, we showed that incorporating events of acute GVHD improves the predictive ability of machine learning models. Additionally, we examined umbilical cord blood transplantation for malignant lymphoma in Japan and Europe, highlighting the importance of total body irradiation in transplantation for lymphoma while also noting the differences in the significance of HLA mismatch between Europe and Japan, suggesting the influence of ethnic differences.

Free Research Field

血液学

Academic Significance and Societal Importance of the Research Achievements

本研究において、移植前因子のみならず移植後因子も含めることで、機械学習モデルの予測能が改善することが示されており、移植後因子も積極的に収集することの重要性が示されている。また、国際間でHLA不適合の影響は異なっており、集団の違いや危険因子を個別に分析することの重要性が示された。これらを考慮しながら同種移植の国際標準予後予測アルゴリズムの改善が必要である。。