A major component in the lung cancer TNM staging system is the status of lymph node (LN) involvement, and accurate LN assessment is crucial in patient management. The N component is currently stratified into categories of N0 to N3 solely on the basis of the anatomic stations with metastatic LN involvement. Options for revising the N classification include counting LNs, LN stations, or LN zones, as well as detecting extranodal extension (ENE) [1].

For metastatic LNs, ENE is defined as extension of metastatic cells beyond the LN capsule into perinodal tissue. ENE has been incorporated into N category determination for head and neck cancers and for vulvar cancer. In a meta-analysis of pathologic ENE in patients with non–small cell lung cancer (NSCLC), ENE was associated with poor prognosis in terms of disease recurrence and all-cause mortality [2].

This retrospective study of 382 patients with NSCLC investigated prognostic implications of CT-based ENE and the diagnostic performance of CT-based ENE in predicting pathologic ENE. Indistinct LN margin, coalescent LNs, direct invasion of adjacent structures, and central necrosis were found to be possible CT findings of ENE. These findings were adopted from previously explored CT findings of ENE in head and neck cancer [3]. Two chest radiologists also classified CT examinations for overall confidence in presence of ENE on a patient level as no ENE, possible or probable ENE, and unambiguous ENE. For both readers, 5-year overall survival rates were significantly different among these three tiers. Unambiguous ENE independently predicted worse overall survival and was highly specific in predicting pathologic ENE.

This is the first study, to my knowledge, to evaluate the impact of CT-based ENE in predicting prognosis and pathologic ENE. Although identifying ENE on CT is challenging, this study opens a new window into CT-based evaluation for ENE of metastatic LNs in NSCLC.