AUTHOR=Chen Yan , Sun Zepang , Chen Wanlan , Liu Changyan , Chai Ruoyang , Ding Jingjing , Liu Wen , Feng Xianzhen , Zhou Jun , Shen Xiaoyi , Huang Shan , Xu Zhongqing TITLE=The Immune Subtypes and Landscape of Gastric Cancer and to Predict Based on the Whole-Slide Images Using Deep Learning JOURNAL=Frontiers in Immunology VOLUME=Volume 12 - 2021 YEAR=2021 URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2021.685992 DOI=10.3389/fimmu.2021.685992 ISSN=1664-3224 ABSTRACT=Abstract Background: Gastric cancer (GC) is a highly heterogeneous tumor with different response to immunotherapy. To identify immune subtypes and landscape of GC could improve immunotherapeutic strategies. Methods: Based on the abundance of tumor-infiltrating immune cells in GC patients from The Cancer Genome Atlas database, we used unsupervised consensus clustering algorithm to identify robust clusters of patients, and assessed their reproducibility in an independent cohort from Gene Expression Omnibus database. We further confirmed the feasibility of our immune subtypes in five independent pan-cancer cohorts. Finally, functional enrichment analyses were provided and a deep learning model studying the pathological images was constructed to identify the immune subtypes. Results: We identified and validated 3 reproducible immune subtypes presented with diverse components of tumor-infiltrating immune cells, molecular features and clinical characteristics. An immune-hot subtype 3, with better prognosis and the highest immune score, had the highest abundance of CD8+ T cells, CD4+ T activated cells, follicular helper T cells, M1 macrophages and NK cells among three subtypes. By contrast, immune-cold subtype 1 demonstrated the highest infiltration of CD4+ T resting cells, regulatory T cells, B cells and dendritic cells, while immune-cold subtype 2 demonstrated the highest infiltration of M2 macrophages and mast cells, and the lowest infiltration of M1 macrophages. Besides, higher proportion of EVB and MSI of TCGA molecular subtyping, over expression of CTLA4, PD1, PDL1 and TP53, and low expression of JAK1 were observed in immune subtype 3, which consisted with the results from Gene Set Enrichment Analysis. These subtypes may suggest different immunotherapy strategies. Finally, deep learning can predict the immune subtypes well. Conclusion: This study offers a conceptual frame to better understand the tumor immune microenvironment of GC. Future work is required to estimate its reference value for the design of immune-related studies and immunotherapy selection.