Metabolic pathway-based molecular subtyping of colon cancer reveals clinical immunotherapy potential and prognosis
Metabolic pathway-based molecular subtyping of colon cancer reveals clinical immunotherapy potential and prognosis
复制标题
DOI：
10.1007/s00432-022-04070-6
复制
发表时间：
2022-06-22
影响因子：
3.6
通讯作者：
Song, Jinglue
中科院分区：
文献类型：
--
作者：
Dai, Zhujiang;Peng, Xiang;Song, Jinglue
目的结肠癌的高度异质性给临床诊断和治疗带来了挑战。为了更有效和方便地诊断和治疗结肠癌，我们致力于通过开创基于代谢途径的分类系统来表征结肠癌的分子特征。方法在前一阶段收集的113条代谢途径和基因的基础上，利用ssGSEA对训练集中每个样本的代谢途径进行评分和筛选，得到16条与结肠癌复发相关的代谢途径。在具有复发相关代谢途径评分的训练集样本的一致聚类中，我们确定了结肠癌的两种稳健分子亚型（MC 1和MC 2）。并对各亚型的生存差异、代谢特征、临床特征、功能富集、免疫浸润、与其他亚型的差异、干性指标、TIDE预测、药物敏感性等进行多角度分析，最终构建结肠癌预后模型。结果MC 1亚型具有较高的免疫活性和免疫检查点基因表达，预后较差。MC 2亚型与高代谢活性和免疫检查点基因的低表达以及更好的预后相关。MC 2亚型比MC 1亚型对PD-L1免疫治疗更敏感。然而，我们没有观察到两者之间肿瘤突变负荷的显著差异。结论基于代谢途径的两种分子亚型结肠癌具有不同的免疫特征。构建基于亚型差异基因的预后模型为针对肿瘤代谢特征的个体化治疗提供了有价值的参考。
Purpose Colon cancer presents challenges to clinical diagnosis and management due to its high heterogeneity. For more efficient and convenient diagnosis and treatment of colon cancer, we are committed to characterizing the molecular features of colon cancer by pioneering a classification system based on metabolic pathways. Methods Based on the 113 metabolic pathways and genes collected in the previous stage, we scored and filtered the metabolic pathways of each sample in the training set by ssGSEA, and obtained 16 metabolic pathways related to colon cancer recurrence. In consistent clustering of training set samples with recurrence-related metabolic pathway scores, we identified two robust molecular subtypes of colon cancer (MC1 and MC2). Furthermore, we performed multi-angle analysis on the survival differences of subtypes, metabolic characteristics, clinical characteristics, functional enrichment, immune infiltration, differences with other subtypes, stemness indices, TIDE prediction, and drug sensitivity, and finally constructed colon cancer prognostic model. Results The results showed that the MC1 subtype had a poor prognosis based on higher immune activity and immune checkpoint gene expression. The MC2 subtype is associated with high metabolic activity and low expression of immune checkpoint genes and a better prognosis. The MC2 subtype was more responsive to PD-L1 immunotherapy than the MC1 subclass. However, we did not observe significant differences in tumor mutational burden between the two. Conclusion Two molecular subtypes of colon cancer based on metabolic pathways have distinct immune signatures. Constructing prognostic models based on subtype differential genes provides valuable reference for personalized therapy targeting unique tumor metabolic signatures.