Prevalence and Risk Factors for Pathologic N2 Disease in Resected Lung Cancers Assessed as N0 or N1 Disease on Preoperative Imaging

Background: In certain patients with lung cancer, guidelines recommend endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) as the primary method to evaluate for metastatic mediastinal lymph nodes, defining pN2 disease. EBUS-TBNA has associated costs and complications and possibly limited availability. Objectives: To investigate the prevalence and risk factors for pN2 disease in patients undergoing resection of lung cancer assessed as having radiologic N0 or N1 disease. Methods: This retrospective study included 3581 patients (mean age, 63.8±9.4 years; 1917 male, 1664 female) with lung cancer who underwent chest CT and FDG PET/CT showing radiologic N0 or N1 disease before resection between January 2015 and December 2021. Tumor characteristics were assessed on chest CT. Patients were assessed for presence of guideline-based indications for EBUS-TBNA as evaluation for imaging-occult N2 disease. Pathologic N categories were determined from surgical specimens. Preoperative risk factors for pN2 disease were identified using logistic regression analyses. Results: A total of 1936 patients had radiologic rN0 disease without EBUS-TBNA indication, 1348 had radiologic N0 disease with EBUS-TBNA indication, and 297 had radiologic N1 disease. These groups had prevalence of pN2a disease of 4.1%, 6.5%, and 18.5%, respectively, and of pN2b disease of 1.2%, 2.4%, and 14.8%, respectively. In multivariable analyses, independent risk factors for pN2 disease in patients with radiologic N0 disease without EBUS-TBNA indication were female sex (OR=1.66), larger size of the tumor's solid portion (OR=1.05 [95% CI, 1.01-1.10]), pure-solid nodule (OR=5.53 [95% CI, 3.15-9.72]), and spiculation (OR=2.66 [95% CI, 1.72-4.11]); with radiologic N0 disease with EBUS-TBNA indication were younger age (OR=0.97 [95% CI, 0.96-0.99] per year), pure-solid nodule (OR=1.75 [95% CI, 1.10-2.80]), and lobulation (OR=1.96 [95% CI, 1.23-3.11]); and with radiologic N1 disease were younger age (OR=0.973 [95% CI, 0.948-0.999] per year), female sex (OR=2.91 [95% CI, 1.66-5.11]) and spiculation (OR=2.81 [95% CI, 1.66-4.76]). Conclusion: pN2b disease was uncommon in patients with radiologic N0 disease, regardless of indications for EBUS-TBNA, and increased in prevalence in patients with radiologic N1 disease. Clinical Impact: The identified risk factors can inform patient selection for EBUS-TBNA, to aid detection of occult pN2 disease.