国际医学放射学杂志2024，Vol.47Issue(3)：267-273,7.DOI:10.19300/j.2024.L21308
CT影像形态学特征对肺腺癌免疫表型术前预测的价值
The value of CT imaging morphological features for preoperative prediction of immune phenotypes in lung adenocarcinoma
摘要
Abstract
Objective To investigate the value of CT morphological features in predicting the immune phenotypes of lung adenocarcinoma.Methods A retrospective analysis was conducted on preoperative CT images and clinical data from 302 patients with pathologically confirmed invasive lung adenocarcinoma.The patients were randomly divided into a training set(211 cases)and a validation set(91 cases)in a 7∶3 ratio.Tumor immune phenotypes were defined based on lymphocyte density and spatial distribution characteristics in lung adenocarcinoma,classifying patients into hot tumor groups(training set:80 cases;validation set:32 cases)and cold tumor groups(training set:131 cases;validation set:59 cases).Survival analysis was performed using the Kaplan-Meier method and Log-rank test.Univariate and multivariate logistic regression analyses were used to construct a preoperative prediction model for lung adenocarcinoma immune phenotypes based on CT imaging morphological features,with the predictive performance of the model evaluated by the area under the receiver operating characteristic curve(AUC).Results The Kaplan-Meier curves indicated that the disease-free survival period was shorter in the cold tumor group compared to the hot tumor group in both the training set(P=0.001)and validation set(P=0.005).A CT imaging model was constructed using the lesion location(peripheral/central),maximum diameter of the solid component,spicule sign(presence/absence),and border(clear/unclear).The AUCs of the prediction model in the training and validation sets were 0.70(95%CI:0.60～0.83)and 0.72(95%CI:0.63～0.77),respectively.Conclusion The immune phenotype of lung adenocarcinoma can assist in prognostic stratification beyond the pathological TNM staging.The model based on CT imaging morphological features can effectively predict the immune phenotype of lung adenocarcinoma preoperatively.关键词
肺腺癌/免疫表型/体层摄影术,X线计算机/预后Key words
Lung adenocarcinoma/Immune phenotype/Tomography,X-ray computed/Prognosis分类
医药卫生引用本文复制引用
莫梓阳,林欢,梁演婷,潘细朋,冯拯云,刘再毅..CT影像形态学特征对肺腺癌免疫表型术前预测的价值[J].国际医学放射学杂志,2024,47(3):267-273,7.基金项目
国家重点研发计划(2021YFF1201003) （2021YFF1201003）
广东省科技计划项目(2022B1212010011) （2022B1212010011）
国家青年科学基金(62002082) （62002082）
高水平医院建设科研专项(DFJHBF202105) （DFJHBF202105）
