Background: Recently, the International Association for the Study of Lung Cancer (IASLC) has proposed significant modifications to the existing TNM and stage grouping classifications affecting the T4 and M descriptors. We set out to validate this staging system for bronchioloalveolar carcinoma (BAC) cases using data from the California Cancer Registry (CCR).

Methods: We identified 1909 patients from the CCR between 1999 and 2003 with histologically confirmed BAC and complete TNM staging and reclassified them according to the IASLC proposed staging revisions. There were 657 patients with stage IIIB and IV disease who formed the primary analysis of the changes to T4 and M descriptors. Surveillance Epidemiology and End Results (SEER) extent of disease codes (EOD) were used to identify various T4 and M descriptors. The primary outcome measured was overall survival (OS) for stage-specific comparisons of the existing to the proposed staging systems, using the Kaplan-Meier method. Multivariate survival analyses were performed using Cox proportional hazards ratios.

Results: Using the proposed criteria, 162 (25%) of the 657 patients with advanced BAC were reclassified: 73 patients with multiple lesions in the same lobe as T3 (stage II T3N0M0 [n = 53], stage IIIA T3N1-2M0 [n = 18], stage IIIB T3N3M0 [n = 1] or T3NXM0 [n = 1]); 89 patients with ipsilateral intrapulmonary metastasis were reclassified as T4 (stage IIIA T4N0-N1M0 [n = 54], stage IIIB T4N2-3M0 [n = 23] or T4NXM0 [n = 12]). Univariate and multivariate survival analysis of this validation set revealed an improved fit for the proposed IASLC staging system compared with the existing staging system.

Conclusions: The proposed IASLC staging system modifications accurately reflect survival characteristics for BAC and represent an improvement compared with the existing staging system.