Comparative validation of nodal classifications in resected non-small cell lung cancer based on the anatomical location and number of involved stations.

INTRODUCTION

Nodal staging of non-small cell lung cancer (NSCLC) is important for treatment decision-making. Comparative validation of various classifications based on anatomical location and station count, alone or in combination, remains lacking.

METHODS

We analyzed 1,601 patients with pathological stage I-III disease undergoing complete resection between 2002 and 2019. Nodal classifications were based on "TNM 8th," "the International Association for the Study of Lung Cancer (IASLC) exploratory proposal in TNM 8th," "TNM 9th," and "the number of involved nodal stations/zones." The ability of each classification to predict recurrence-free survival was examined using multivariate Cox regression analysis and time-dependent receiver operating characteristic (ROC) curves. Agreement between clinical and pathological nodal staging (cN and pN) was also assessed.

RESULTS

The IASLC exploratory proposal classification in TNM 8th revealed a better prognosis (P < 0.05) for single-station skip N2 within the N2 group, overlapping with N1. Other classifications stratified N descriptors into prognostically distinct subgroups, with time-dependent ROC curves showing no clinically significant differences between the classifications. However, N descriptors based on the number of involved stations/zones formed distinct prognostic subgroups when further stratified by anatomical location. cN and pN agreement was generally low for N1 (44.7 %) and very low for N1 based on the number of involved stations (in the 20 % range).

CONCLUSIONS

Future revisions of NSCLC nodal staging should integrate anatomical location and the number of involved stations, as in TNM 9th. Moreover, N1 subclassification should be avoided, given the difficulty in counting cN1 stations. From a prognostic perspective, single-station skip N2 should be integrated with N1.