■癌症开始,programming,免疫逃避依赖于肿瘤微环境(TME)。因此,了解TME免疫结构对于了解肿瘤转移和治疗反应至关重要.这项研究旨在使用通过生态类型分析丰富的大量RNA-seq数据来创建免疫细胞状态(CS)图谱,以解决TME中的复杂免疫结构。
■我们雇用了EcoTyper,机器学习(ML)框架,为了研究免疫CSs和多细胞生态系统的真实世界预后意义,利用来自ORIENAvatar队列中接受免疫检查点抑制剂(ICI)治疗的1,610例多种恶性肿瘤患者的分子数据,一个注释良好的现实世界数据集。
■我们的分析揭示了在我们的泛癌症数据集中一致的ICI特异性预后TME癌生态型(CEs)(包括CE1,CE9,CE10),其中CE1更缺乏淋巴细胞和CE10更多的促炎。此外,对不同癌症的特异性免疫CSs的分析显示了一致的CD8+和CD4+T细胞CS分布模式.此外,ORIENICI队列的生存分析表明,生态型CE9与最有利的生存结局相关,而CE2与最不利的结果有关。值得注意的是,黑色素瘤特异性预后EcoTyper模型证实,较低的预测风险评分与提高生存率和更好的免疫治疗反应相关.最后,在ORIENICI数据集中重新发现生态型,确定生态型E3与较差的生存结局显著相关。
■我们的发现为在现实世界实践中完善免疫治疗的患者选择过程提供了重要的见解,并指导创建针对TME中特定生态型的新型治疗策略。试用注册:NCT04526730。
UNASSIGNED: Cancer initiation, progression, and immune evasion depend on the tumor microenvironment (TME). Thus, understanding the TME immune architecture is essential for understanding tumor metastasis and therapy response. This study aimed to create an immune cell states (CSs) atlas using bulk RNA-seq data enriched by eco-type analyses to resolve the complex immune architectures in the TME.
UNASSIGNED: We employed EcoTyper, a machine-learning (ML) framework, to study the real-world prognostic significance of immune CSs and multicellular ecosystems, utilizing molecular data from 1,610 patients with multiple malignancies who underwent immune checkpoint inhibitor (ICI) therapy within the ORIEN Avatar cohort, a well-annotated real-world dataset.
UNASSIGNED: Our analysis revealed consistent ICI-specific prognostic TME carcinoma ecotypes (CEs) (including CE1, CE9, CE10) across our pan-cancer dataset, where CE1 being more lymphocyte-deficient and CE10 being more proinflammatory. Also, the analysis of specific immune CSs across different cancers showed consistent CD8+ and CD4+ T cell CS distribution patterns. Furthermore, survival analysis of the ORIEN ICI cohort demonstrated that
ecotype CE9 is associated with the most favorable survival outcomes, while CE2 is linked to the least favorable outcomes. Notably, the melanoma-specific prognostic EcoTyper model confirmed that lower predicted risk scores are associated with improved survival and better response to immunotherapy. Finally, de novo discovery of ecotypes in the ORIEN ICI dataset identified
Ecotype E3 as significantly associated with poorer survival outcomes.
UNASSIGNED: Our findings offer important insights into refining the patient selection process for immunotherapy in real-world practice and guiding the creation of novel therapeutic strategies to target specific ecotypes within the TME.