Zheng Chang, Sun Liang, Zhou Baosen, Wang Aiping
Department of Clinical Epidemiology and Evidence-Based Medicine, First Affiliated Hospital of China Medical University, Shenyang, China.
Department of Emergency, First Affiliated Hospital of China Medical University, Shenyang, China.
Ann Transl Med. 2021 Aug;9(16):1312. doi: 10.21037/atm-21-3234.
The immunosuppressive tumor microenvironment produced by cancer cells is a key mechanisms of cancer immune escape. In this study, we investigated the relationship between the metabolic patterns and tumor immune environment in the TME of lung adenocarcinoma (LUAD) with the p53 mutation.
The clinical data of 495 LUAD patients was obtained from The Cancer Genome Atlas as transcriptomic and somatic mutation data. Using differential analysis, survival analysis, and a LASSO regression model based on metabolic unigenes from KEGG pathways, a tumor metabolic model was constructed to predict the prognosis of LUAD patients. Subsequently, nomogram, receiver operating characteristic, and decision curve analyses were conducted to assess the predictive ability of the model. In addition, the ESTIMATE and CIBERSORT algorithms were used to detect tumor purity and estimate the fractions of 22 immune cell types in each patient, respectively. We found a correlation between the composition of immune cells and the tumor metabolic model. The results were validated using an independent GSE72094 dataset with 442 patients, as well as an immunohistochemistry assay, RT-qPCR, and western blot.
The tumor metabolic model reassigned the risk score of every patient, and a tumor metabolic risk score (TMRS) was generated to show the predictive ability for patient prognoses (hazard ratio =0.39; 95% confidence interval: 0.18-0.85). Using a combination of TMRS and clinical features, a nomogram was produced with a predictive accuracy of 0.72. Further analysis showed that CD4 memory resting T cells and M1 macrophages may by correlated with the TMRS, which corresponded to immunoediting in p53 mutant patients. Additionally, the similar expression of ALDH3A1 and MGAT5B were also verified by wetlab experiments.
Based on the identified tumor metabolism-immune landscape, we were able to predict a metabolism risk score for patient prognosis and identify a correlation with two types of infiltrating lymphocytes in the TME of p53-mutated LUAD. This landscape provides insights that will help identify the molecular mechanisms of immune-editing tumor metabolism.
癌细胞产生的免疫抑制性肿瘤微环境是癌症免疫逃逸的关键机制。在本研究中，我们调查了p53突变的肺腺癌（LUAD）肿瘤微环境（TME）中代谢模式与肿瘤免疫环境之间的关系。
从癌症基因组图谱获取495例LUAD患者的临床数据作为转录组和体细胞突变数据。使用差异分析、生存分析以及基于KEGG通路代谢单基因的LASSO回归模型，构建肿瘤代谢模型以预测LUAD患者的预后。随后，进行列线图、受试者工作特征和决策曲线分析以评估模型的预测能力。此外，分别使用ESTIMATE和CIBERSORT算法检测肿瘤纯度并估计每位患者22种免疫细胞类型的比例。我们发现免疫细胞组成与肿瘤代谢模型之间存在相关性。使用包含442例患者的独立GSE72094数据集以及免疫组织化学检测、RT-qPCR和蛋白质印迹对结果进行验证。
肿瘤代谢模型重新分配了每位患者的风险评分，并生成了肿瘤代谢风险评分（TMRS）以显示对患者预后的预测能力（风险比=0.39；95%置信区间：0.18 - 0.85）。结合TMRS和临床特征，制作了预测准确率为0.72的列线图。进一步分析表明，CD4记忆静止T细胞和M1巨噬细胞可能与TMRS相关，这与p53突变患者的免疫编辑相对应。此外，ALDH3A1和MGAT5B的相似表达也通过湿实验室实验得到验证。
基于所确定的肿瘤代谢 - 免疫格局，我们能够预测患者预后的代谢风险评分，并确定其与p53突变LUAD的TME中两种浸润淋巴细胞类型的相关性。这一格局提供了有助于识别免疫编辑肿瘤代谢分子机制的见解。
