Chen Jun, Zhou Chao, Liu Ying
Department of Oncology, The First Affiliated Hospital of Nanchang University, Nanchang, China.
Department of Neurology, Jiangxi Provincial People's Hospital Affiliated to Nanchang University, Nanchang, China.
Front Oncol. 2022 Feb 23;11:771988. doi: 10.3389/fonc.2021.771988. eCollection 2021.
Tumor-associated macrophages are important components of the tumor microenvironment, and the macrophage phenotypic switch has been shown to correlate with tumor development. However, the use of a macrophage phenotypic switch-related gene (MRG)-based prognosis signature for lung adenocarcinoma (LADC) has not yet been investigated.
In total, 1,114 LADC cases from two different databases were collected. The samples from TCGA were used as the training set (N = 490), whereas two independent datasets (GSE31210 and GSE72094) from the GEO database were used as the validation sets (N = 624). A robust MRG signature that predicted clinical outcomes of LADC patients was identified through multivariate COX and Lasso regression analysis. Gene set enrichment analysis was applied to analyze molecular pathways associated with the MRG signature. Moreover, the fractions of 22 immune cells were estimated using CIBERSORT algorithm.
An eight MRG-based signature comprising CTSL, ECT2, HCFC2, HNRNPK, LRIG1, OSBPL5, P4HA1, and TUBA4A was used to estimate the LADC patients' overall survival. The MRG model was capable of distinguishing high-risk patients from low-risk patients and accurately predict survival in both the training and validation cohorts. Subsequently, the eight MRG-based signature and other features were used to construct a nomogram to better predict the survival of LADC patients. Calibration plots and decision curve analysis exhibited good consistency between the nomogram predictions and actual observation. ROC curves displayed that the signature had good robustness to predict LADC patients' prognostic outcome.
We identified a phenotypic switch-related signature for predicting the survival of patients with LADC.
肿瘤相关巨噬细胞是肿瘤微环境的重要组成部分，巨噬细胞表型转换已被证明与肿瘤发展相关。然而，基于巨噬细胞表型转换相关基因（MRG）的预后特征在肺腺癌（LADC）中的应用尚未得到研究。
总共收集了来自两个不同数据库的1114例LADC病例。来自TCGA的样本用作训练集（N = 490），而来自GEO数据库的两个独立数据集（GSE31210和GSE72094）用作验证集（N = 624）。通过多变量COX和Lasso回归分析确定了一个预测LADC患者临床结局的强大MRG特征。应用基因集富集分析来分析与MRG特征相关的分子途径。此外，使用CIBERSORT算法估计22种免疫细胞的比例。
一个基于8个MRG的特征，包括CTSL、ECT2、HCFC2、HNRNPK、LRIG1、OSBPL5、P4HA1和TUBA4A，用于估计LADC患者的总生存期。MRG模型能够区分高风险患者和低风险患者，并准确预测训练和验证队列中的生存期。随后，基于8个MRG的特征和其他特征被用于构建列线图，以更好地预测LADC患者的生存期。校准图和决策曲线分析显示列线图预测与实际观察之间具有良好的一致性。ROC曲线显示该特征在预测LADC患者的预后结局方面具有良好的稳健性。
我们确定了一个与表型转换相关的特征，用于预测LADC患者的生存期。
