中国肺癌杂志2024，Vol.27Issue(4)：266-275,10.DOI:10.3779/j.issn.1009-3419.2024.102.14
囊腔型肺腺癌临床多特征分析及浸润性风险预测模型的构建
Clinical Multi-features Analysis of Cystic Lung Adenocarcinoma and Construction of Invasive Risk Prediction Model
摘要
Abstract
Background and objective Cystic lung cancer,a special type of lung cancer,has been paid more and more attention.The most common pathological type of cystic lung cancer is adenocarcinoma.The invasiveness of cystic lung adenocarcinoma is vital for the selection of clinical treatment and prognosis.The aim of this study is to analyze the multiple clinical features of cystic lung adenocarcinoma,explore the independent risk factors of its invasiveness,and establish a risk pre-diction model.Methods A total of 129 cases of cystic lung adenocarcinoma admitted to the Department of Thoracic Surgery of the First Affiliated Hospital of Nanjing Medical University from January 2021 to July 2022 were retrospectively analyzed and divided into pre-invasive group[atypical adenomatous hyperplasia(AAH),adenocarcinoma in situ(AIS)and minimally invasive adenocarcinoma(MIA)]and invasive group[invasive adenocarcinoma(IAC)]according to pathological findings.There were 47 cases in the pre-invasive group,including 19 males and 28 females,with an average age of(51.23±14.96)years.There were 82 cases in the invasive group,including 60 males and 22 females,with an average age of(61.27±11.74)years.Mul-tiple clinical features of the two groups were collected,including baseline data,imaging data and tumor markers.Univariate analysis,LASSO regression and multivariate Logistic regression analysis were used to screen out the independent risk factors of the invasiveness of cystic lung adenocarcinoma,and the risk prediction model was established.Results In univariate analysis,age,gender,smoking history,history of emphysema,neuron-specific enolase(NSE),number of cystic airspaces,lesion di-ameter,cystic cavity diameter,nodule diameter,solid components diameter,cyst wall nodule,smoothness of cyst wall,shape of cystic airspace,lobulation,short burr sign,pleural retraction,vascular penetration and bronchial penetration were statisti-cally different between the pre-invasive group and invasive groups(P<0.05).The above variables were processed by LASSO regression dimensionality reduction and screened as follows:age,gender,smoking history,NSE,number of cystic airspaces,lesion diameter,cystic cavity diameter,cyst wall nodule,smoothness of cyst wall and lobulation.Then the above variables were included in multivariate Logistic regression analysis.Cyst wall nodule(P=0.035)and lobulation(P=0.001)were found to be independent risk factors for the invasiveness of cystic lung adenocarcinoma(P<0.05).The prediction model was established as follows:P=e^x/(1+e^x),x=-7.927+1.476* cyst wall nodule+2.407* lobulation,and area under the curve(AUC)was 0.950.Conclusion Cyst wall nodule and lobulation are independent risk factors for the invasiveness of cystic lung adenocarcinoma,which have certain guiding significance for the prediction of the invasiveness of cystic lung adenocarcinoma.关键词
囊腔/肺肿瘤/浸润性/预测模型Key words
Cyst/Lung neoplasms/Invasiveness/Prediction model引用本文复制引用
王强,符程皓,王坤,任千睿,陈爱萍,徐心峰,陈亮,朱全..囊腔型肺腺癌临床多特征分析及浸润性风险预测模型的构建[J].中国肺癌杂志,2024,27(4):266-275,10.基金项目
本研究受江苏省卫生健康委医学科研项目(No.K2023077)资助 This study was supported by the grant from Medical Research Project of Jiangsu Provincial Health Commission(No.K2023077)(to Quan ZHU). （No.K2023077）
