Tumors Invaded in the Central Airway in Predicting Severe Immune Checkpoint Inhibitor-Related Pneumonitis Based on Propensity Score Matching
10.3969/j.issn.1005-5185.2025.06.014
- VernacularTitle:基于倾向性评分评价肿瘤中央气道侵犯预测重度免疫抑制剂相关性肺炎的价值
- Author:
Bofeng ZHAO
1
;
Yaming ZHANG
;
Ping CHEN
;
Wei FENG
;
Kejun NAN
;
Jinpeng LIU
;
Baoying CHEN
Author Information
1. 西安国际医学中心医院影像诊疗中心,陕西 西安 710700
- Publication Type:Journal Article
- Keywords:
Immune checkpoint inhibitor-related pneumonitis;
Drug-related side effects and adverse reactions;
Tomography,X-ray computed;
Tumor invasion in the central airway;
Propensity score matching
- From:
Chinese Journal of Medical Imaging
2025;33(6):645-650
- CountryChina
- Language:Chinese
-
Abstract:
Purpose To evaluate the value of tumors invasion in the central airway(TICA)in predicting the severe immune checkpoint inhibitor-related pneumonitis(S-CIP)in lung cancer patients using propensity score matching(PSM).Materials and Methods The intact data of 162 consecutive lung cancer patients who received treatment with immune checkpoint inhibitors in Xi'an International Medical Center Hospital from September 2019 to March 2022 were retrospectively collected.Patients were divided into S-CIP group(23 cases)and non-S-CIP group(139 cases)according to the presence of S-CIP.The demographic information of the patients,including gender,age,history of smoking,thoracic radiotherapy histology,baseline lung diseases,classification,TNM stage,tumor location as well as TICA were collected.A binary Logistic regression was used to analyze the confounding factors and independent risk factors of S-CIP and to predict the development of S-CIP.A 1:1 matching was performed by the nearest neighbor method for PSM.The PSM was used to pair the two groups,and the value of TICA in predicting S-CIP before and after PSM was compared.The receiver operating characteristic curve and the area under the curve were used for model performance based on TICA.Results Before PSM,the proportion of baseline lung diseases(78.3%vs.32.4%,OR=6.802,P=0.001),thoracic radiotherapy history(69.6%vs.30.2%,OR=5.300,P=0.002)and TICA(65.2%vs.27.3%,OR=5.882,P=0.001)in the S-CIP group was higher than those in the non-S-CIP group,and were independent risk factor for predicting S-CIP.After PSM,20 patients were included in each group.The presence of TICA was higher in S-CIP group than that in the non-S-CIP group(60.0%vs.20.0%,OR=6.000,P=0.013).The area under the curves of Logistic regression model based on TICA was 0.700(95%CI 0.534-0.866).Conclusion TICA is an independent risk factor for development of S-CIP,which has moderate degree of accuracy in predicting S-CIP,can be used for risk prediction and early intervention to reduce the poor prognosis of S-CIP patients.
