Department of Radiation Oncology, Tianjin Medical University Cancer Institute & Hospital, Key Laboratory of Cancer Prevention and Therapy, National Clinical Research Center for Cancer, Tianjin's Clinical Research Center for Cancer, Tianjin, China.
School of Mechanical Engineering, Tianjin University, Tianjin, China.
Medicine (Baltimore). 2024 Oct 4;103(40):e39798. doi: 10.1097/MD.0000000000039798.
Colorectal cancer is a common malignant tumor with the second incidence rate and the third mortality rate worldwide. In this study, we identified and validated an immune-related gene signature, explored the clinical and molecular characteristics of the signature-defined risk groups, and assessed its ability in predicting prognosis, immune cell infiltration and immunotherapy responses. The Cancer Genome Atlas database was used as the training set while GSE39582 database as the validation set. Immune-related hub genes were selected by the Least Absolute Shrinkage and Selection Operator-penalized Cox regression model, and the signature was then constructed by the selected genes and their relevant coefficients. Prognostic performance of the signature and the signature-base nomogram models were assessed by time-dependent receiver operating characteristic curves and calibration plots in both training and validation cohorts. Clinical and mutation-related data were downloaded and analyzed to explore their associations with signature-defined risk groups. Proportions of infiltrated immune cells was estimated via CIBERSORT algorithm and immunotherapy response was evaluated by immunophenoscore and tumor immune dysfunction and exclusion scores. Seven among 790 immune-related differentially-expressed genes were selected and use to construct the signature. The signature and signature-base nomograms showed promising prognostic performance in both training and validation cohorts. Signature-defined high-risk group was associated with advanced disease, poor pathological prognostic factors and less active immune infiltration microenvironment. Besides, the response to immunotherapy of high-risk group was predicted to be poorer by immunophenoscore and tumor immune dysfunction and exclusion scores. Our signature proved its efficacy in predicting prognosis, tumor immune microenvironment and responses to immunotherapy in colorectal cancer.
结直肠癌是全球发病率第二、死亡率第三的常见恶性肿瘤。本研究旨在鉴定并验证一个免疫相关基因signature，探索该 signature 定义的风险组的临床和分子特征，并评估其预测预后、免疫细胞浸润和免疫治疗反应的能力。本研究使用癌症基因组图谱（The Cancer Genome Atlas，TCGA）数据库作为训练集，GSE39582 数据库作为验证集。通过最小绝对收缩和选择算子惩罚 Cox 回归模型筛选免疫相关枢纽基因，并根据筛选出的基因及其相关系数构建 signature。通过时间依赖性接收者操作特征曲线和校准图在训练集和验证集中评估 signature 和 signature 列线图模型的预后性能。下载并分析临床和突变相关数据，以探索其与 signature 定义的风险组的关联。通过 CIBERSORT 算法估计浸润免疫细胞的比例，并通过免疫表型评分和肿瘤免疫功能障碍和排除评分评估免疫治疗反应。从 790 个免疫相关差异表达基因中筛选出 7 个基因用于构建 signature。signature 和 signature 列线图在训练集和验证集中均表现出良好的预后预测性能。signature 定义的高风险组与晚期疾病、较差的病理预后因素和较少活跃的免疫浸润微环境相关。此外，免疫表型评分和肿瘤免疫功能障碍和排除评分预测高风险组对免疫治疗的反应较差。我们的 signature 证明了其在预测结直肠癌预后、肿瘤免疫微环境和免疫治疗反应方面的有效性。
