%0 Journal Article %A Rong, Chao %A Liu, Yun %A Xiang, Fang %A Zhao, Xin %A Zhang, Jinjin %A Xiao, Zuorun %A Wang, Jinsha %A Chen, Lin %A Guo, Zhiqi %A Zhang, Ziyu %A An, Jingnan %A Shen, Jing %A Hess, Jochen %A Yuan, Xiaodong %A Zhang, Qiong %A Wang, Shouli %T Integrative bioinformatics analysis and experimental validation identify CHEK1 as an unfavorable prognostic biomarker related to immunosuppressive phenotypes in soft tissue sarcomas. %J npj precision oncology %V 9 %N 1 %@ 2397-768X %C [London] %I Springer Nature %M DKFZ-2025-01610 %P 268 %D 2025 %X Soft tissue sarcomas (STS), including rhabdomyosarcoma (RMS), exhibit significant heterogeneity and limited responsiveness to immune checkpoint blockade (ICB). Unsupervised tumor immune phenotype based on multi-omics expression profiling of STS has been less studied. To reveal the tumor immune phenotype of STS and identify promising therapeutic targets, multi-omics expression profiling across various subtypes of STS was investigated. Here, we established a novel molecular classifier based on immune cell subsets related to TGFβ1 and IFNγ to identify distinct immune phenotypes with higher or lower cytotoxic contents. Immune-high clusters demonstrated enriched immune cell infiltration, elevated IFNγ-related signatures, and favorable clinical outcomes. In contrast, immune-low clusters were enriched for immunosuppressive cell types and exhibited poor survival. CHEK1 emerged as a key node associated with immunosuppressive phenotypes and was significantly overexpressed in immune-low tumors. In situ analysis of independent validation cohorts revealed the significant correlation between CHEK1 and tumor-infiltrating immune cells. Collectively, our findings establish a novel risk assessment strategy for RMS and STS patients, and highlight the potential of CHEK1 as a promising therapeutic target in combination with immune checkpoint inhibitor therapy. %F PUB:(DE-HGF)16 %9 Journal Article %$ pmid:40751074 %2 pmc:PMC12317078 %R 10.1038/s41698-025-01064-8 %U https://inrepo02.dkfz.de/record/303360