MOE Key Lab of Bioinformatics and Bioinformatics Division, BNRIST; Department of Automation; Tsinghua-Peking Joint Center for Life Sciences; Center for Synthetic and Systems Biology, Tsinghua University, Beijing, 100084, China.
MOE Key Laboratory of Carcinogenesis and Translational Research; Department of Human Anatomy, Histology and Embryology; State Key Laboratory of Natural and Biomimetic Drugs, Peking University Health Science Center, Beijing, 100191, China.
Sci China Life Sci. 2021 Aug;64(8):1311-1325. doi: 10.1007/s11427-020-1811-5. Epub 2020 Oct 30.
Colorectal cancer (CRC) progression is associated with cancer cell dedifferentiation and sternness acquisition. Several methods have been developed to identify sternness signatures in CRCs. However, studies that directly measured the degree of dedifferentiation in CRC tissues are limited. It is unclear how the differentiation states change during CRC progression. To address this, we develop a method to analyze the tissue differentiation spectrum in colorectal cancer using normal gastrointestinal single-cell transcriptome data. Applying this method on 281 tumor samples from The Cancer Genome Atlas Colon Adenocarcinoma dataset, we identified three major CRC subtypes with distinct tissue differentiation pattern. We observed that differentiation states are closely correlated with anti-tumor immune response and patient outcomes in CRC. Highly dedifferentiated CRC samples escaped the immune surveillance and exhibited poor outcomes; mildly dedifferentiated CRC samples showed resistance to anti-tumor immune responses and had a worse survival rate; well-differentiated CRC samples showed sustained anti-tumor immune responses and had a good prognosis. Overall, the spectrum of tissue differentiation observed in CRCs can be used for future clinical risk stratification and subtype-based therapy selection.
结直肠癌（CRC）的进展与癌细胞去分化和获得干性有关。已经开发了几种方法来鉴定 CRC 中的干性特征。然而，直接测量 CRC 组织中去分化程度的研究是有限的。目前尚不清楚在 CRC 进展过程中分化状态如何变化。为了解决这个问题，我们开发了一种使用正常胃肠道单细胞转录组数据分析结直肠癌组织分化谱的方法。将该方法应用于 TCGA 结肠腺癌数据集的 281 个肿瘤样本，我们确定了三种具有不同组织分化模式的主要 CRC 亚型。我们观察到，分化状态与 CRC 中的抗肿瘤免疫反应和患者预后密切相关。高度去分化的 CRC 样本逃避了免疫监视，预后较差；轻度去分化的 CRC 样本对抗肿瘤免疫反应具有抗性，生存率较差；分化良好的 CRC 样本表现出持续的抗肿瘤免疫反应，预后良好。总体而言，在 CRC 中观察到的组织分化谱可用于未来的临床风险分层和基于亚型的治疗选择。
