Prediction Models for Mediastinal Metastasis and Its Detection by Endobronchial Ultrasound-Guided Transbronchial Needle Aspiration in Potentially Operable Non-Small Cell Lung Cancer

BackgroundPrediction 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 patients with non-small cell lung cancer (NSCLC).Research QuestionCan mediastinal metastasis and its detection by EBUS-TBNA be predicted with prediction models in NSCLC?Study Design and MethodsFor the prospective development cohort, 589 potentially operable patients with NSCLC were evaluated (July 2016-June 2019) from five Korean teaching hospitals. Mediastinal staging was performed using EBUS-TBNA (with or without the transesophageal approach). Surgery was performed for patients without clinical N (cN) 2-3 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 period (June 2019-August 2021).ResultsThe prevalence of mediastinal metastasis diagnosed by EBUS-TBNA or surgery and the sensitivity of EBUS-TBNA in the development cohort were 35.3% and 87.0%, respectively. In PLUS-M, younger age (< 60 years and 60-70 years compared with ≥ 70 years), nonsquamous histology (adenocarcinoma and others), central tumor location, tumor size (> 3-5 cm), cN1 or cN2-3 stage by CT, and cN1 or cN2-3 stage by PET-CT were significant risk factors for N2-3 disease. Areas under the receiver operating characteristic curve (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: Hosmer-Lemeshow P = .658, Brier score = 0.129; PLUS-E: Hosmer-Lemeshow P = .569, Brier score = 0.118). In the validation cohort, PLUS-M (AUC, 0.859 [95% CI, 0.817-0.902], Hosmer-Lemeshow P = .609, Brier score = 0.144) and PLUS-E (AUC, 0.900 [95% CI, 0.865-0.936], Hosmer-Lemeshow P = .361, Brier score = 0.112) showed good discrimination ability and calibration.InterpretationPLUS-M and PLUS-E can be used effectively for decision-making for invasive mediastinal staging in NSCLC.Trial RegistryClinicalTrials.gov; No.: NCT02991924; URL: www.clinicaltrials.gov 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 patients with non-small cell lung cancer (NSCLC). Can mediastinal metastasis and its detection by EBUS-TBNA be predicted with prediction models in NSCLC? For the prospective development cohort, 589 potentially operable patients with NSCLC were evaluated (July 2016-June 2019) from five Korean teaching hospitals. Mediastinal staging was performed using EBUS-TBNA (with or without the transesophageal approach). Surgery was performed for patients without clinical N (cN) 2-3 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 period (June 2019-August 2021). The prevalence of mediastinal metastasis diagnosed by EBUS-TBNA or surgery and the sensitivity of EBUS-TBNA in the development cohort were 35.3% and 87.0%, respectively. In PLUS-M, younger age (< 60 years and 60-70 years compared with ≥ 70 years), nonsquamous histology (adenocarcinoma and others), central tumor location, tumor size (> 3-5 cm), cN1 or cN2-3 stage by CT, and cN1 or cN2-3 stage by PET-CT were significant risk factors for N2-3 disease. Areas under the receiver operating characteristic curve (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: Hosmer-Lemeshow P = .658, Brier score = 0.129; PLUS-E: Hosmer-Lemeshow P = .569, Brier score = 0.118). In the validation cohort, PLUS-M (AUC, 0.859 [95% CI, 0.817-0.902], Hosmer-Lemeshow P = .609, Brier score = 0.144) and PLUS-E (AUC, 0.900 [95% CI, 0.865-0.936], Hosmer-Lemeshow P = .361, Brier score = 0.112) showed good discrimination ability and calibration. PLUS-M and PLUS-E can be used effectively for decision-making for invasive mediastinal staging in NSCLC. ClinicalTrials.gov; No.: NCT02991924; URL: www.clinicaltrials.gov Take-home PointsStudy Question: Can mediastinal metastasis and its detection by endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) be predicted with prediction models in non-small cell lung cancer (NSCLC)?Results: The prediction model for lung cancer staging-mediastinal metastasis (PLUS-M) and the prediction model for lung cancer staging-mediastinal metastasis detection by EBUS-TBNA (PLUS-E) were developed using age, histologic type, tumor location, tumor size, clinical N stage by CT and clinical N stage by PET-CT. PLUS-M and PLUS-E showed good discrimination ability and calibration.Interpretation: PLUS-M and PLUS-E can be used effectively for decision-making for invasive mediastinal staging in NSCLC. Study Question: Can mediastinal metastasis and its detection by endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) be predicted with prediction models in non-small cell lung cancer (NSCLC)? Results: The prediction model for lung cancer staging-mediastinal metastasis (PLUS-M) and the prediction model for lung cancer staging-mediastinal metastasis detection by EBUS-TBNA (PLUS-E) were developed using age, histologic type, tumor location, tumor size, clinical N stage by CT and clinical N stage by PET-CT. PLUS-M and PLUS-E showed good discrimination ability and calibration. Interpretation: PLUS-M and PLUS-E can be used effectively for decision-making for invasive mediastinal staging in NSCLC. Invasive mediastinal staging is an important step that guides treatment decision-making in non-small cell lung cancer (NSCLC).1Silvestri G.A. Gonzalez A.V. Jantz M.A. et al.Methods for staging non-small cell lung cancer: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines.Chest. 2013; 143: e211S-e250SAbstract Full Text Full Text PDF PubMed Scopus (1120) Google Scholar, 2De Leyn P. Dooms C. Kuzdzal J. et al.Revised ESTS guidelines for preoperative mediastinal lymph node staging for non-small-cell lung cancer.Eur J Cardiothorac Surg. 2014; 45: 787-798Crossref PubMed Scopus (556) Google Scholar, 3Remon J. Soria J.C. Peters S. ESMO Guidelines CommitteeEarly and locally advanced non-small-cell lung cancer: an update of the ESMO clinical practice guidelines focusing on diagnosis, staging, systemic and local therapy.Ann Oncol. 2021; 32: 1637-1642Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar Currently, endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA), which has high diagnostic values comparable with those of mediastinoscopy,4Yasufuku K. Pierre A. Darling G. et al.A prospective controlled trial of endobronchial ultrasound-guided transbronchial needle aspiration compared with mediastinoscopy for mediastinal lymph node staging of lung cancer.J Thorac Cardiovasc Surg. 2011; 142: 1393-1400.e1391Abstract Full Text Full Text PDF PubMed Scopus (437) Google Scholar,5Um S.W. Kim H.K. Jung S.H. et al.Endobronchial ultrasound versus mediastinoscopy for mediastinal nodal staging of non-small-cell lung cancer.J Thorac Oncol. 2015; 10: 331-337Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar is a primary method for invasive staging.1Silvestri G.A. Gonzalez A.V. Jantz M.A. et al.Methods for staging non-small cell lung cancer: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines.Chest. 2013; 143: e211S-e250SAbstract Full Text Full Text PDF PubMed Scopus (1120) Google Scholar,2De Leyn P. Dooms C. Kuzdzal J. et al.Revised ESTS guidelines for preoperative mediastinal lymph node staging for non-small-cell lung cancer.Eur J Cardiothorac Surg. 2014; 45: 787-798Crossref PubMed Scopus (556) Google Scholar Practice guidelines for mediastinal staging in NSCLC recommend invasive staging for specific risk groups for mediastinal metastasis.1Silvestri G.A. Gonzalez A.V. Jantz M.A. et al.Methods for staging non-small cell lung cancer: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines.Chest. 2013; 143: e211S-e250SAbstract Full Text Full Text PDF PubMed Scopus (1120) Google Scholar,2De Leyn P. Dooms C. Kuzdzal J. et al.Revised ESTS guidelines for preoperative mediastinal lymph node staging for non-small-cell lung cancer.Eur J Cardiothorac Surg. 2014; 45: 787-798Crossref PubMed Scopus (556) Google Scholar,6Vilmann P. Clementsen P.F. Colella S. et al.Combined endobronchial and esophageal endosonography for the diagnosis and staging of lung cancer: European Society of Gastrointestinal Endoscopy (ESGE) guideline, in cooperation with the European Respiratory Society (ERS) and the European Society of Thoracic Surgeons (ESTS).Endoscopy. 2015; 47: 549-559Google Scholar, 7Farjah F. Tanner N.T. Mediastinal staging for lung cancer.Chest. 2021; 160: 1552-1559Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar, 8National Institute for Health and Care Excellence (NICE)Lung cancer: diagnosis and management (NICE guideline [NG122]). March 28, 2019. National Institute for Health and Care Excellence website.www.nice.org.uk/guidance/ng122Date accessed: October 20, 2022Google Scholar, 9National Comprehensive Cancer Network (NCCN)Non-small cell lung cancer, version 5.2022, NCCN Clinical Practice Guidelines in Oncology. September 26, 2022. National Comprehensive Cancer Network website.www.nccn.org/professionals/physician_gls/pdf/nscl.pdfDate accessed: October 20, 2022Google Scholar According to European Society of Thoracic Surgeons (ESTS) guidelines, preoperative staging is recommended in patients with clinical N (cN) 1-3 disease by CT or PET, a central tumor or a tumor > 3 cm in size.2De Leyn P. Dooms C. Kuzdzal J. et al.Revised ESTS guidelines for preoperative mediastinal lymph node staging for non-small-cell lung cancer.Eur J Cardiothorac Surg. 2014; 45: 787-798Crossref PubMed Scopus (556) Google Scholar CHEST guidelines recommend invasive staging for cN1-3 disease by CT, cN2-3 disease by PET and central tumors with grade 1 evidence and do not recommend invasive staging for peripheral tumors of ≤ 3 cm with cN0 disease by CT and PET.1Silvestri G.A. Gonzalez A.V. Jantz M.A. et al.Methods for staging non-small cell lung cancer: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines.Chest. 2013; 143: e211S-e250SAbstract Full Text Full Text PDF PubMed Scopus (1120) Google Scholar National Institute for Health and Care Excellence guidelines have a narrow recommendation for invasive mediastinal staging: cN1-3 disease by CT or PET-CT.8National Institute for Health and Care Excellence (NICE)Lung cancer: diagnosis and management (NICE guideline [NG122]). March 28, 2019. National Institute for Health and Care Excellence website.www.nice.org.uk/guidance/ng122Date accessed: October 20, 2022Google Scholar Younger age10Koike T. Koike T. Yamato Y. Yoshiya K. Toyabe S. Predictive risk factors for mediastinal lymph node metastasis in clinical stage IA non-small-cell lung cancer patients.J Thorac Oncol. 2012; 7: 1246-1251Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar, 11Shafazand S. Gould M.K. A clinical prediction rule to estimate the probability of mediastinal metastasis in patients with non-small cell lung cancer.J Thorac Oncol. 2006; 1: 953-959Abstract Full Text Full Text PDF PubMed Google Scholar, 12Chen K. Yang F. Jiang G. Li J. Wang J. Development and validation of a clinical prediction model for N2 lymph node metastasis in non-small cell lung cancer.Ann Thorac Surg. 2013; 96: 1761-1768Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar, 13O’Connell O.J. Almeida F.A. Simoff M.J. et al.A prediction model to help with the assessment of adenopathy in lung cancer: HAL.Am J Respir Crit Care Med. 2017; 195: 1651-1660Crossref PubMed Scopus (27) Google Scholar, 14Zhang Y. Sun Y. Xiang J. Zhang Y. Hu H. Chen H. A prediction model for N2 disease in T1 non-small cell lung cancer.J Thorac Cardiovasc Surg. 2012; 144: 1360-1364Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar, 15Jiang L. Jiang S. Lin Y. et al.Nomogram to predict occult N2 lymph nodes metastases in patients with squamous nonsmall cell lung cancer.Medicine (Baltimore). 2015; 94e2054Crossref Scopus (14) Google Scholar and adenocarcinoma histology11Shafazand S. Gould M.K. A clinical prediction rule to estimate the probability of mediastinal metastasis in patients with non-small cell lung cancer.J Thorac Oncol. 2006; 1: 953-959Abstract Full Text Full Text PDF PubMed Google Scholar, 12Chen K. Yang F. Jiang G. Li J. Wang J. Development and validation of a clinical prediction model for N2 lymph node metastasis in non-small cell lung cancer.Ann Thorac Surg. 2013; 96: 1761-1768Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar, 13O’Connell O.J. Almeida F.A. Simoff M.J. et al.A prediction model to help with the assessment of adenopathy in lung cancer: HAL.Am J Respir Crit Care Med. 2017; 195: 1651-1660Crossref PubMed Scopus (27) Google Scholar,16Fibla J.J. Molins L. Simon C. Perez J. Vidal G. The yield of mediastinoscopy with respect to lymph node size, cell type, and the location of the primary tumor.J Thorac Oncol. 2006; 1: 430-433Abstract Full Text Full Text PDF PubMed Google Scholar, 17Suzuki K. Nagai K. Yoshida J. Nishimura M. Takahashi K. Nishiwaki Y. Clinical predictors of N2 disease in the setting of a negative computed tomographic scan in patients with lung cancer.J Thorac Cardiovasc Surg. 1999; 117: 593-598Abstract Full Text Full Text PDF PubMed Scopus (81) Google Scholar, 18Lee P.C. Port J.L. Korst R.J. Liss Y. Meherally D.N. Altorki N.K. Risk factors for occult mediastinal metastases in clinical stage I non-small cell lung cancer.Ann Thorac Surg. 2007; 84: 177-181Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar, 19Cho S. Song I.H. Yang H.C. Kim K. Jheon S. Predictive factors for node metastasis in patients with clinical stage I non-small cell lung cancer.Ann Thorac Surg. 2013; 96: 239-245Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar are reported to be risk factors for mediastinal metastasis, but are not reflected clearly in the staging guidelines.1Silvestri G.A. Gonzalez A.V. Jantz M.A. et al.Methods for staging non-small cell lung cancer: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines.Chest. 2013; 143: e211S-e250SAbstract Full Text Full Text PDF PubMed Scopus (1120) Google Scholar,2De Leyn P. Dooms C. Kuzdzal J. et al.Revised ESTS guidelines for preoperative mediastinal lymph node staging for non-small-cell lung cancer.Eur J Cardiothorac Surg. 2014; 45: 787-798Crossref PubMed Scopus (556) Google Scholar,6Vilmann P. Clementsen P.F. Colella S. et al.Combined endobronchial and esophageal endosonography for the diagnosis and staging of lung cancer: European Society of Gastrointestinal Endoscopy (ESGE) guideline, in cooperation with the European Respiratory Society (ERS) and the European Society of Thoracic Surgeons (ESTS).Endoscopy. 2015; 47: 549-559Google Scholar, 7Farjah F. Tanner N.T. Mediastinal staging for lung cancer.Chest. 2021; 160: 1552-1559Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar, 8National Institute for Health and Care Excellence (NICE)Lung cancer: diagnosis and management (NICE guideline [NG122]). March 28, 2019. National Institute for Health and Care Excellence website.www.nice.org.uk/guidance/ng122Date accessed: October 20, 2022Google Scholar, 9National Comprehensive Cancer Network (NCCN)Non-small cell lung cancer, version 5.2022, NCCN Clinical Practice Guidelines in Oncology. September 26, 2022. National Comprehensive Cancer Network website.www.nccn.org/professionals/physician_gls/pdf/nscl.pdfDate accessed: October 20, 2022Google Scholar Factors related with mediastinal metastasis, such as radiologic lymph node (LN) abnormalities by CT or PET scan imaging, central tumor location, tumor size, age, and histologic type can be related with each other. However, current guidelines do not present estimations for the probability of N2-3 disease based on combinations of risk factors. The decision to perform invasive staging is influenced by the staging method’s diagnostic capability as well as the probability of mediastinal metastasis. A prediction model for the detection of N2-3 disease by EBUS-TBNA that takes into account multiple risk factors has been developed using the American College of Chest Physicians Quality Improvement Registry, Evaluation, and Education database: the A Prediction Model to Help with the Assessment of Adenopathy in Lung Cancer (HAL).13O’Connell O.J. Almeida F.A. Simoff M.J. et al.A prediction model to help with the assessment of adenopathy in lung cancer: HAL.Am J Respir Crit Care Med. 2017; 195: 1651-1660Crossref PubMed Scopus (27) Google Scholar The model focused on EBUS-TBNA results and did not include the surgical pathologic staging results, which are the gold standard for mediastinal staging. The aim of this study was to develop prediction models for mediastinal metastasis diagnosed by EBUS-TBNA or surgery and by EBUS-TBNA alone using a prospective cohort of patients with potentially operable NSCLC, which will be useful for decision-making regarding invasive staging. This prospective cohort study was conducted at five teaching hospitals in Korea (National Cancer Center [NCC], Seoul National University Bundang Hospital, Asan Medical Center, Samsung Medical Center, and Seoul National University Hospital). Six hundred patients with potentially operable confirmed (or highly suspicious) NSCLC (age range, 18-80 years) were enrolled prospectively from July 2016 through June 2019 (development cohort) (enrollment criteria presented in e-Table 1). The sample size was calculated to be sufficient to generate a two-sided 95% CI with a sample area under the receiver operating characteristic curve (AUC) of 0.8 with a width of 0.08 and an N2-3 disease prevalence of 35% (based on our previous studies20Hwangbo B. Lee G.K. Lee H.S. et al.Transbronchial and transesophageal fine-needle aspiration using an ultrasound bronchoscope in mediastinal staging of potentially operable lung cancer.Chest. 2010; 138: 795-802Abstract Full Text Full Text PDF PubMed Scopus (212) Google Scholar,21Kang H.J. Hwangbo B. Lee G.K. et al.EBUS-centred versus EUS-centred mediastinal staging in lung cancer: a randomised controlled trial.Thorax. 2014; 69: 261-268Crossref PubMed Scopus (84) Google Scholar). We excluded tumors with subsolid nodules with solid parts of ≤ 1 cm in diameter and those with solid nodules that were cT1aN0 stage by CT and PET-CT scan imaging. Contrast chest CT scans (axial slice thickness, ≤ 3 mm) and PET-CT scans (without contrast) were requested within 30 days before EBUS-TBNA. For model validation, we used a retrospective cohort from the NCC from a different period (June 2019-August 2021) with the same enrollment criteria and imaging requirements. This study was approved by the institutional review board of NCC (Identifiers: NCC-2016-0156 and 2021-0307) and other hospitals (Seoul National University Bundang Hospital Identifier: B-1608-360-301, Asan Medical Center Identifier: 2016-0713, Samsung Medical Center Identifier: 2016-07-125-003, and Seoul National University Hospital Identifier: 1608-006-784). Informed consent was obtained from all participants in the development cohort. EBUS-TBNA in the development cohort was performed by six experienced bronchoscopists (≥ 500 EBUS-TBNA procedures) under conscious sedation. We recommended LNs of ≥ 1 cm or mediastinal nodes showing positive PET scan findings be sampled when targets were accessible and smaller or that nodes with negative PET scan findings be sampled based on echo features, potential pathways for lymphatic metastasis, and impact of metastasis at the nodal station on treatment decision. We recommended that N3 nodes be sampled first, with at least two to three aspirations per target.22Lee H.S. Lee G.K. Lee H.S. et al.Real-time endobronchial ultrasound-guided transbronchial needle aspiration in mediastinal staging of non-small cell lung cancer: how many aspirations per target lymph node station?.Chest. 2008; 134: 368-374Abstract Full Text Full Text PDF PubMed Scopus (291) Google Scholar N1 staging was not performed routinely. Final determination of procedural details was made by the attending bronchoscopists, taking into consideration patient tolerance. Aspirations using a transesophageal approach (endoscopic ultrasound with bronchoscope-guided fine needle aspiration [EUS-B-FNA]) were allowed when LNs were difficult to access by EBUS-TBNA, but were reachable by EUS-B-FNA.20Hwangbo B. Lee G.K. Lee H.S. et al.Transbronchial and transesophageal fine-needle aspiration using an ultrasound bronchoscope in mediastinal staging of potentially operable lung cancer.Chest. 2010; 138: 795-802Abstract Full Text Full Text PDF PubMed Scopus (212) Google Scholar cN stage determined by EBUS-TBNA included EUS-B-FNA results. We recommended surgery with systematic LN dissection for patients without cN2-3 disease by EBUS-TBNA within 30 days after EBUS-TBNA. Data on age, sex, histologic findings, tumor location (central or peripheral), clinical T (cT) stage by size on axial chest CT scan and cN stage by CT, PET-CT, and EBUS-TBNA, as well as data on the EBUS-TBNA procedure, were collected. Age was converted into a categorical variable (< 60 years, 60-70 years, and ≥ 70 years). We reviewed all nonsurgical and surgical lung cancer histologic results and classified them into three groups: squamous cell carcinoma, adenocarcinoma (including NSCLC with adenocarcinoma component), and others, based on final pathologic diagnosis including surgical results. A central tumor was defined as one located in the inner one-third of the hemithorax23Detterbeck F.C. Jantz M.A. Wallace M. Vansteenkiste J. Silvestri G.A. American College of Chest PhysiciansInvasive mediastinal staging of lung cancer: ACCP Evidence-Based Clinical Practice Guidelines (2nd edition).Chest. 2007; 132: 202S-220SAbstract Full Text Full Text PDF PubMed Scopus (653) Google Scholar based on the innermost part of the tumor on CT scan.24Shin S.H. Jeong D.Y. Lee K.S. et al.Which definition of a central tumour is more predictive of occult mediastinal metastasis in nonsmall cell lung cancer patients with radiological N0 disease?.Eur Respir J. 2019; 531801508Crossref PubMed Scopus (33) Google Scholar,25Guinde J. Bourdages-Pageau E. Collin-Castonguay M.M. et al.A prediction model to optimize invasive mediastinal staging procedures for non-small cell lung cancer in patients with a radiologically normal mediastinum: the Quebec Prediction Model.Chest. 2021; 160: 2283-2292Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar Lines dividing the hemithorax into thirds were drawn as concentric lines arising from the midline.24Shin S.H. Jeong D.Y. Lee K.S. et al.Which definition of a central tumour is more predictive of occult mediastinal metastasis in nonsmall cell lung cancer patients with radiological N0 disease?.Eur Respir J. 2019; 531801508Crossref PubMed Scopus (33) Google Scholar,25Guinde J. Bourdages-Pageau E. Collin-Castonguay M.M. et al.A prediction model to optimize invasive mediastinal staging procedures for non-small cell lung cancer in patients with a radiologically normal mediastinum: the Quebec Prediction Model.Chest. 2021; 160: 2283-2292Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar The eighth edition of the International Association for the Study of Lung Cancer staging criteria was used for staging.26Asamura H. Chansky K. Crowley J. et al.The International Association for the Study of Lung Cancer Lung Cancer Staging Project: proposals for the revision of the N descriptors in the forthcoming 8th edition of the TNM Classification for Lung Cancer.J Thorac Oncol. 2015; 10: 1675-1684Abstract Full Text Full Text PDF PubMed Scopus (508) Google Scholar, 27Rami-Porta R. Bolejack V. Crowley J. et al.The IASLC Lung Cancer Staging Project: proposals for the revisions of the T descriptors in the forthcoming eighth edition of the TNM Classification for Lung Cancer.J Thorac Oncol. 2015; 10: 990-1003Abstract Full Text Full Text PDF PubMed Scopus (558) Google Scholar, 28Travis W.D. Asamura H. Bankier A.A. et al.The IASLC Lung Cancer Staging Project: proposals for coding T categories for subsolid nodules and assessment of tumor size in part-solid tumors in the forthcoming eighth edition of the TNM Classification of Lung Cancer.J Thorac Oncol. 2016; 11: 1204-1223Abstract Full Text Full Text PDF PubMed Scopus (485) Google Scholar For surgery cases, pathological N stage was reviewed. For patients with pathological Nx stage, at least 12 months of follow-up was performed to identify benign mediastinal LNs with a lack of radiologic disease progression.29Yasufuku K. Chiyo M. Koh E. et al.Endobronchial ultrasound guided transbronchial needle aspiration for staging of lung cancer.Lung Cancer. 2005; 50: 347-354Abstract Full Text Full Text PDF PubMed Scopus (421) Google Scholar The characteristics of the development and validation cohorts were compared using the χ2 test, Fisher exact test, Wilcoxon rank-sum test, or Student t test, as appropriate. The Prediction model for Lung cancer Staging-Mediastinal metastasis (PLUS-M) and the Prediction model for Lung cancer Staging-mediastinal metastasis detection by EBUS-TBNA (PLUS-E) were developed. The primary outcome for the development of PLUS-M was the presence of N2 or N3 metastasis by EBUS-TBNA, surgery, or follow-up after surgery (for pathological Nx stage). We performed univariable logistic regression analysis for the following variables: age, sex, histologic findings, tumor location, cT stage by size, cN stage by CT, and cN stage by PET-CT. All variables with P values of < .2 were included in the multivariable logistic regression analysis, and variables included in the final model were determined by the backward selection method with an elimination criterion of P ≥ .05. The primary outcome for development of PLUS-E was the presence of N2 or N3 metastasis by EBUS-TBNA. Univariable and multivariable logistic regression analyses were performed on the variables described above. Significant risk factors in PLUS-M were included in PLUS-E. The regression analysis results are presented as ORs with 95% CIs and P values. Clinical nomograms were constructed for PLUS-M and PLUS-E based on multivariable logistic regression analyses. For discrimination, AUCs for PLUS-M and PLUS-E were calculated in the development cohort. Internal validation was performed to calculate optimism-adjusted AUCs using 1,000 bootstrap samples. Calibration was assessed for PLUS-M and PLUS-E using the Hosmer-Lemeshow test and the Brier score. Calibration plots (observed vs predicted) were created by dividing the predicted risk into deciles. Data from the validation cohort were applied to PLUS-M and PLUS-E and AUCs were calculated. Calibration was evaluated in the same way. To optimize the clinical application of the prediction models, probabilities of N2-3 disease predicted by PLUS-M and PLUS-E were calculated based on risk factors present. Using the ESTS, modified CHEST (cN1-3 disease by CT or PET-CT or central tumor), and National Institute for Health and Care Excellence guidelines as well as multiple probability thresholds for PLUS-M and PLUS-E (≥ 10%, 8%, or 5%) as criteria for selecting EBUS-TBNA candidates in the development and validation cohorts, we calculated the sensitivity and negative predictive value of guidelines and models for N2-3 disease, expected detection rate of N2-3 disease by EBUS-TBNA in confirmed N2-3 cases, unforeseen N2-3 disease rate after surgery, and EBUS-TBNA procedures prevented. All statistical analyses were performed using SAS version 9.4 software (SAS Institute, Inc.) and R version 4.1.1 project software (R Foundation for Statistical Computing). A P value of < .05 was considered statistically significant. Among 600 patients enrolled, 589 (NCC, n = 459; Seoul National University Bundang Hospital, n = 52; Asan Medical Center, n = 42; Samsung Medical Center, n = 27; and Seoul National University Hospital, n = 9) were included in the development cohort (e-Fig 1). Characteristics of the development cohort and the EBUS-TBNA procedure are presented in Table 1 and e-Table 2 (broken down by institution). In the development cohort, N2-3 disease prevalence was 35.3% (208 of 589). EBUS-TBNA sensitivity was 87.0% (181 of 208). The sensitivity in patients with cN0-1 disease by PET-CT was 68.5% (37 of 54). The mean number of nodal stations per patient and number of aspirations per target for EBUS-TBNA (including EUS-B-FNA) were 3.1 and 3.2, respectively. At least two nodal stations were sampled in 565 patients (95.9%). At least two aspirations per target were performed in 1,781 targets (98.2%). EUS-B-FNA was performed in 7.8% of targets. Baseline characteristics of the validation cohort were similar with the development cohort, except for the number of targets and aspirations per target during EBUS-TBNA and the use of EUS-B-FNA (Table 1). The prevalence of N2-3 disease and the sensitivity of EBUS-TBNA in t