生物
免疫系统
免疫疗法
癌症研究
结直肠癌
微卫星不稳定性
基因敲除
趋化因子
免疫检查点
多路复用
表型
免疫学
基因表达谱
癌症免疫疗法
T细胞
基因表达
基因
癌症
转移
免疫荧光
封锁
转录组
体内
抗体
作者
Xiaomao Yin,Jinran Wu,Midie Xu,Tongguan Tian,Lin Zhu,Jiexuan Wang,Xuan Dai,Xin Yang,Jingjing Qian,Wenqiang Wang,Liangchen Zhu,Zekun Zhao,Kai Xu,Yanping Xu,Xinxing Li,Zhiqian Hu
标识
DOI:10.1186/s12943-025-02502-8
摘要
The efficacy of immune checkpoint inhibitors (ICIs) in microsatellite stable colorectal cancer (MSS CRC) remains limited, highlighting an urgent need for predictive biomarkers. Through multi-omics analysis, we identified two novel MSS CRC subtypes, termed DUB-H and DUB-L. The DUB-L subtype exhibited an inflamed tumor immune microenvironment, a superior response to immune therapy, and better recurrence-free survival (RFS) compared to DUB-H. The classifier gene USP7 was selected as a gene of interest due to its specific expression profile, which is highly expressed in MSS CRC but not in microsatellite instability-high (MSI-H) tumors, and strongly correlated with suppressed immune infiltration. Large-scale clinical analyses confirmed associations between high USP7 expression, microsatellite stability, specific consensus molecular subtypes (CMS), and unfavorable prognosis. Single-cell analysis and multiplex immunofluorescence validated an immune-desert phenotype in USP7-high MSS tumors. Mechanistically, USP7 knockdown in MSS CRC cells enhances the secretion of T-cell-recruiting chemokines (CXCL9/10/11), promoting CD8⁺ T cell recruitment and cytotoxicity in vitro. In vivo experiments demonstrated that USP7 blockade enhanced the efficacy of anti-PD-1 treatment in MSS CRC models by remodeling the tumor immune microenvironment, increasing infiltration and function of CD8⁺ T and NK cells. Consistently, low USP7 expression is associated with a better response to anti-PD-1 therapy. Overall, we propose a novel DUB-based classification system for MSS CRC and demonstrate that targeting USP7 may overcome immunotherapy resistance by converting immunologically “cold” tumors into “hot” ones.
科研通智能强力驱动
Strongly Powered by AbleSci AI