预测非小细胞肺癌（NSCLC）可手术患者纵向队列中纵隔转移及内镜支气管超声引导经支气管针吸（EBUS-TBNA）检测的预测模型尚未建立。在纵向开发队列中，从韩国五所教学医院评估了589位潜在可手术的NSCLC患者（2016年7月至2019年6月）。同时进行使用EBUS-TBNA（±经食管途径）的纵隔分期，对于无cN2-3疾病的患者进行内窥镜分期手术。多变量逻辑回归分析开发了肺癌分期-纵隔转移预测模型（PLUS-M）和EBUS-TBNA检测纵隔转移预测模型（PLUS-E）。使用不同时间段的回顾性队列（n=309）进行验证（2019年6月至2021年8月）。在开发队列中，EBUS-TBNA配合手术诊断的纵隔转移的患病率和EBUS-TBNA的灵敏度分别为35.3%和87.0%。在PLUS-M中，年龄较小（<60或60-70岁与≥70岁相比）、腺癌、其他非鳞状细胞癌、中央肿瘤位置、肿瘤大小（>3-5厘米）和CT或PET-CT分期是N2-3疾病的重要风险因素。PLUS-M和PLUS-E的受试者操作特征曲线下面积（AUC）分别为0.876（95% CI，0.845-0.906）和0.889（95% CI，0.859-0.918）。模型适合度良好（PLUS-M：Homer-Lemeshow P=0.658，Brier评分=0.129；PLUS-E：Homer-Lemeshow P=0.569，Brier评分=0.118）。在验证队列中，PLUS-M（AUC，0.859 [95% CI，0.817-0.902]，Homer-Lemeshow P=0.609，Brier评分=0.144）和PLUS-E（AUC，0.900 [95% CI，0.865-0.936]，Homer-Lemeshow P=0.361，Brier评分=0.112）均表现出良好的判别能力和校准性。在NSCLC中，PLUS-M和PLUS-E可有效用于纵隔浸润分期的决策。 版权 © 2023年Elsevier Inc. 发布。

Prediction models for mediastinal metastasis and its detection by endobronchial ultrasound guided transbronchial needle aspiration (EBUS-TBNA) have not been developed using a prospective cohort of potentially operable non-small cell lung cancer (NSCLC) patients.Can mediastinal metastasis and its detection by EBUS-TBNA be predicted with prediction models in NSCLC?For the prospective development cohort, 589 potentially operable NSCLC patients were evaluated (July 2016-June 2019) from 5 Korean teaching hospitals. Mediastinal staging was performed using EBUS-TBNA (± trans-esophageal approach). Surgery was performed for patients without cN2-3 disease by endoscopic staging. The Prediction model for Lung cancer Staging-Mediastinal metastasis (PLUS-M) and a model for mediastinal metastasis detection by EBUS-TBNA (PLUS-E) were developed using multivariable logistic regression analyses. Validation was performed using a retrospective cohort (n=309) from a different time period (June 2019-August 2021).The prevalence of mediastinal metastasis diagnosed by EBUS-TBNA plus surgery and the sensitivity of EBUS-TBNA in the development cohort were 35.3% and 87.0%, respectively. In PLUS-M, younger age (<60 and 60-70 compared to ≥70), adenocarcinoma, other non-squamous cell carcinoma, central tumor location, tumor size (> 3-5 cm) and cN1 or cN2-3 stage by CT or PET-CT were significant risk factors for N2-3 disease. Areas under the receiver operating characteristic curves (AUCs) for PLUS-M and PLUS-E were 0.876 (95% CI, 0.845-0.906) and 0.889 (95% CI, 0.859-0.918), respectively. Model fit was good (PLUS-M: Homer-Lemeshow P=0.658, Brier score=0.129; PLUS-E: Homer-Lemeshow P=0.569, Brier score=0.118). In the validation cohort, PLUS-M (AUC, 0.859 [95% CI, 0.817-0.902], Homer-Lemeshow P=0.609, Brier score=0.144) and PLUS-E (AUC, 0.900 [95% CI, 0.865-0.936], Homer-Lemeshow P=0.361, Brier score=0.112) showed good discrimination ability and calibration.PLUS-M and PLUS-E can be effectively used for decision making for invasive mediastinal staging in NSCLC.Copyright © 2023. Published by Elsevier Inc.