Abstract | BACKGROUND: METHOD: The immunologic role of CD93 is analyzed across TCGA pan- cancers. The correlation between CD93 and BLCA clinical and tumor microenvironment features, predicted immunotherapy pathways, molecular subtypes, therapeutic signatures and mutation status was evaluated in TCGA-BLCA and other two BLCA cohorts. The impact of CD93 on immunotherapy response was validated by five real-world cohorts, and chemotherapy response was assessed with IC50. CD93-based risk model was constructed with LASSO regression and validated by seven independent cohorts. RESULT: CD93 is positively correlated with immunomodulators, tumor-infiltrating lymphocytes (TILs) and immune checkpoints across pan- cancers. In BLCA, CD93 leads to higher T cell inflamed score and expression of immune checkpoints. However, CD93 is indicative of more aggressive clinical features, worse survival, more tumor-associated macrophages and regulatory T cells recruitment, less recognition and killing of cancer cells by T cells, lower predicted chemotherapy and immunotherapy response, which is further validated by immunotherapy cohorts (IMvigor210: 16.11% vs 29.53%; GSE176307: 15.56% vs 20.93%). Notably, CD93 correlates with enriched neuroendocrine subtype and epithelial-mesenchymal transition differentiation, while CD93-low group has enriched luminal subtype. Pathways including hypoxia and Wnt-β- catenin are enriched along with CD93 expression, and more frequent FGFR3 mutation is also observed. Lastly, the CD93-based risk model, validated by seven independent cohorts, is powerful in distinguishing the survival probability of BLCA (3-year AUC 0.808). CONCLUSION: CD93 plays a critical role in tumor immune regulation. CD93 expression indicates more aggressive clinicopathological status and molecular subtypes of BLCA and worse therapy response, which implies that combing anti-CD93 therapy with immunotherapy (or chemotherapy) may be potentially beneficial for BLCA in clinical practice.
|
Authors | Xiaonan Zheng, Hang Xu, Tianhai Lin, Ping Tan, Qiao Xiong, Xianyanling Yi, Shi Qiu, Lu Yang, Bairong Shen, Jianzhong Ai, Qiang Wei |
Journal | Computers in biology and medicine
(Comput Biol Med)
Vol. 147
Pg. 105727
(08 2022)
ISSN: 1879-0534 [Electronic] United States |
PMID | 35785664
(Publication Type: Journal Article, Research Support, Non-U.S. Gov't)
|
Copyright | Copyright © 2022 Elsevier Ltd. All rights reserved. |
Topics |
- Humans
- Immunotherapy
- Mutation
- Tumor Microenvironment
- Urinary Bladder Neoplasms
(genetics, therapy)
|