背景:肝脏免疫耐受可能有助于免疫治疗的治疗抗性的发展。然而,解决这个问题是具有挑战性的,因为在肝转移(LM)的背景下,免疫疗法的疗效仍然很少研究。这里,我们旨在建立LM共同免疫特征(LMCIF)评分,以量化不同癌症类型中LM免疫耐受的特征,以协助临床疾病管理.
方法:收集大规模临床数据以确定LM的预后。使用来自分子特征数据库(MSigDB)的具有LM特殊免疫相关途径(LMSIP)的转移性癌症的多组学数据集来获得LMCIF簇。基于差异表达基因(DEGs),构建了LMCIF聚类的新LMCIF评分。此外,多组学,来自公众和内部队列的免疫组织化学(IHC)数据用于探索LM的特征,和LMCIF得分。
结果:当结合临床和RNA测序数据分析时,LM患者的预后较差,免疫细胞浸润明显低于其他器官转移患者。在利用29个LMSIP从373个样本中提取LMCIF簇并在微阵列队列中验证它们之后,LMCIF评分的建立是为了确认免疫抑制环境在不同癌症类型的LM预后不良中的作用.此外,该LMCIF评分可用于预测接受免疫治疗的癌症患者的免疫反应.最后,我们发现31个LMCIF基因中的大多数与LM患者的TME细胞呈负相关,其中一个,KRT19与LMCIF评分正相关最强,通过IHC分析证实具有免疫抑制作用。
结论:我们的结果表明,不同癌症类型的LM具有相似的免疫特征,可以诱导免疫耐受并逃避免疫监测。新的LMCIF评分代表了一个常见的肝转移免疫簇,用于预测免疫治疗反应,其结果可能有利于临床疾病管理。
BACKGROUND: Hepatic immune tolerance might contribute to the development of therapeutic resistance to immunotherapy. However, addressing this issue is challenging since the efficacy of immunotherapy in the context of liver metastasis (LM) remains poorly studied. Here, we aimed to establish an LM common immune feature (LMCIF) score to quantify the characteristics of LM immunotolerance across cancer types for assisting clinical disease management.
METHODS: Large-scale clinical data were collected to identify the prognosis of LM. Multi-omics datasets of metastatic cancers with LM special immune-related pathways (LMSIPs) from the Molecular Signatures Database (MSigDB)were used to obtain an LMCIF cluster. Based on differential expression genes (DEGs), a novel LMCIF score for the LMCIF cluster was constructed. In addition, multi-omics, and immunohistochemistry (IHC) data from the public and in-house cohorts were used to explore the features of LM, and LMCIF score.
RESULTS: Patients with LM had a worse prognosis and significantly lower infiltration of immune cells than patients with metastasis to other organs when analyzed with combined clinical and RNA sequencing data. After extracting the LMCIF cluster from 373 samples by utilizing 29 LMSIPs and validating them in a microarray cohort, an LMCIF score was established to confirm the role of the immunosuppressive environment as a contributor to the poor prognosis of LM across cancer types. Moreover, this LMCIF score could be used to predict the immune response of cancer patients undergoing immunotherapy. Finally, we identified that the majority of the 31 LMCIF genes exhibited a negative correlation with TME cells in LM patients, one of them, KRT19, which possessed the strongest positive correlation with LMCIF score, was confirmed to have an immunosuppressive effect through IHC analysis.
CONCLUSIONS: Our results suggest that LM across cancer types share similar immunological profiles that induce immunotolerance and escape from immune monitoring. The novel LMCIF score represents a common liver metastasis immune cluster for predicting immunotherapy response, the results of which might benefit clinical disease management.