Recurrence Risk Stratification by Station- and Zone-Based Nodal Classification in Resected Non-Small Cell Lung Cancer

OBJECTIVES: Although a novel tumour-node-metastasis (TNM) classification (ninth edition) for N staging was recently proposed, there is still room for improvement. We assessed whether classifying N stage based on zones and the number of involved lymph node stations could better predict postoperative recurrence risk. METHODS: We included 671 patients with pathological N1 or N2, M0 non-small-cell lung cancer (NSCLC). Lymph node zones were categorized as peripheral and hilar for N1 nodes, and upper mediastinal, lower mediastinal, aortopulmonary, and subcarinal for N2 nodes. The N category was subdivided according to the location and number of involved lymph node stations: N1a (single-zone/single-station N1), N1b (multiple-zone/multiple-station N1), N2a1 (single-zone/single-station N2 without N1 involvement), N2a2 (single-zone/single-station N2 with N1 involvement), and N2b (multiple-zone/multiple-station N2, regardless of N1 involvement). RESULTS: Recurrence-free survival (RFS) and cumulative incidence of recurrence (CIR) were stratified according to the pathological N1, N2a, and N2b groups based on the TNM (ninth edition) N descriptors. RFS (P < .001 for both) and CIR (P < .001 for both) were effectively stratified using both station- and zone-based classifications. Station- and zone-based N2a1 exhibited RFS (hazard ratio [HR], 1.04; 95% CI, 0.66-1.63 and HR, 1.12; 95% CI, 0.66-1.89, respectively) comparable to those of N1b. Considering station-based N2b, zone-based N2a2 (HR, 1.98; 95% CI, 0.69-5.66) and N2b (HR, 1.99; 95% CI, 0.72-5.52) tended to have worse RFS than zone-based N2a1. CONCLUSIONS: Both station- and zone-based nodal classifications effectively stratify recurrence risk and are good candidates for further N category stratification in NSCLC.