CD8+ T 细胞轨迹亚型解析肿瘤异质性并为肝细胞癌提供治疗建议。

INTRODUCTION: Mounting evidence has revealed that the interactions and dynamic alterations among immune cells are critical in shaping the tumor microenvironment and ultimately map onto heterogeneous clinical outcomes. Currently, the underlying clinical significance of immune cell evolutions remains largely unexplored in hepatocellular carcinoma (HCC). METHODS: A total of 3,817 immune cells and 1,750 HCC patients of 15 independent public datasets were retrieved. The Seurat and Monocle algorithms were used to depict T cell evolution, and nonnegative matrix factorization (NMF) was further applied to identify the molecular classification. Subsequently, the prognosis, biological characteristics, genomic variations, and immune landscape among distinct clusters were decoded. The clinical efficacy of multiple treatment approaches was further investigated. RESULTS: According to trajectory gene expression, three heterogeneous clusters with different clinical outcomes were identified. C2, with a more advanced pathological stage, presented the most dismal prognosis relative to C1 and C3. Eight independent external cohorts validated the robustness and reproducibility of the three clusters. Further explorations elucidated C1 to be characterized as lipid metabolic HCC, and C2 was referred to as cell-proliferative HCC, whereas C3 was defined as immune inflammatory HCC. Moreover, C2 also displayed the most conspicuous genomic instability, and C3 was deemed as "immune-hot", having abundant immune cells and an elevated expression of immune checkpoints. The assessments of therapeutic intervention suggested that patients in C1 were suitable for transcatheter arterial chemoembolization treatment, and patients in C2 were sensitive to tyrosine kinase inhibitors, while patients in C3 were more responsive to immunotherapy. We also identified numerous underlying therapeutic agents, which might be conducive to clinical transformation in the future. CONCLUSIONS: Our study developed three clusters with distinct characteristics based on immune cell evolutions. For specifically stratified patients, we proposed individualized treatment strategies to improve the clinical outcomes and facilitate the clinical management.

简介：越来越多的证据表明，免疫细胞之间的相互作用和动态变化对于塑造肿瘤微环境至关重要，并最终映射到异质的临床结果上。目前，肝癌（HCC）中免疫细胞演变的潜在临床意义在很大程度上仍未得到探索。

方法：共从 15 个独立的公共数据集检索到 3817 个免疫细胞和 1750 名 HCC 患者。使用 Seurat 和 Monocle 算法描绘 T 细胞进化，进一步应用非负矩阵分解（NMF）识别分子分类。随后，解码不同簇之间的预后、生物学特征、基因组变异和免疫景观。进一步研究了多种治疗方法的临床疗效。

结果：根据轨迹基因表达，鉴定出具有不同临床结局的三个异质性簇。与 C1 和 C3 相比，具有更先进病理阶段的 C2 表现出最糟糕的预后。八个独立的外部队列验证了这三个簇的稳健性和可重复性。进一步的探索表明，C1 被定义为脂质代谢 HCC，C2 被定义为细胞增殖性 HCC，而 C3 被定义为免疫炎症性 HCC。此外，C2 还显示出最明显的基因组不稳定性，而 C3 被认为是“免疫热”，具有丰富的免疫细胞和免疫检查点的高表达。治疗干预的评估表明，C1 中的患者适合经导管动脉化疗栓塞治疗，C2 中的患者对酪氨酸激酶抑制剂敏感，而 C3 中的患者对免疫治疗更敏感。我们还确定了许多潜在的治疗药物，这可能有助于未来的临床转化。

结论：我们根据免疫细胞进化，开发了具有不同特征的三个簇。对于特定分层的患者，我们提出了个性化的治疗策略，以提高临床结果并促进临床管理。