Development of a method for predicting tumor immune activation by radiotherapy

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K08230
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 52040:Radiological sciences-related
• Research Institution: Sapporo Medical University
• Principal Investigator: HASEGAWA TOMOKAZU 札幌医科大学, 医学部, 助教 (80631168)
• Co-Investigator(Kenkyū-buntansha): 小塚 陽 札幌医科大学, 医学部, 訪問研究員 (50808160)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: ニューラルネットワーク / 腫瘍免疫

Outline of Final Research Achievements

The objective is to create a prediction model of tumor immune activation by radiotherapy using machine learning methods with high accuracy and easy clinical application. As a preliminary step, a radiotherapy effect prediction model was created by immunohistochemistry using biopsy specimens. Both prostate and hypopharyngeal cancers showed improved prediction accuracy in the analysis using ANN compared to the conventional method.
In addition, using samples of cervical cancer, we used QuPath software to classify and quantify the immunohistochemistry staining decisions, and verified that there was no significant difference compared to the results of manual counting. In the future, we will perform the same analysis on samples of mesopharyngeal carcinoma, aiming for objective evaluation and automation of the determination method of immunohistochemical staining.

Free Research Field

放射線治療

Academic Significance and Societal Importance of the Research Achievements

現在は、放射線治療が施行される場合、腫瘍の大きさや組織型が同じであれば画一的な線量が照射されているが、治療成績にばらつきがあり、癌組織の放射線感受性に応じた、個別化した放射線治療が求められている。放射線治療と免疫チェックポイント阻害剤の併用は行われ始めているが、最適な併用法や増感メカニズムなど未解明な点が多く、精度の高い臨床応用が容易な放射線治療による腫瘍免疫活性化の予測モデルの作成を目指す。