INTRODUCTION: Some unfavorable prognostic factors for stage I non-small-cell lung cancers have been reported; however, they are not reflected in the current Tumor-Node-Metastasis classification. METHODS: We retrospectively reviewed 629 patients who underwent complete resection of pathological stage I adenocarcinomas (ADs) or squamous cell carcinomas (SQs) at two institutes between 1996 and 2011. The correlation between clinicopathological characteristics and survival rates was analyzed to identify prognostic factors. RESULTS: Multivariate analysis indicated that among ADs, high serum carcinoembryonic antigen levels (p = 0.04 for overall survival [OS]; p < 0.01 for recurrence-free survival [RFS]; p = 0.02 for disease-specific survival [DSS]), lymphatic permeation (p < 0.01 for RFS and DSS), and vascular invasion (p < 0.01 for OS and RFS; p = 0.03 for DSS) were independent prognostic factors. Among SQs, high squamous cell carcinoma antigen (SCC) (p < 0.05 for OS), and vascular invasion (p < 0.05 for RFS and DSS) were independently prognostic. We suggest that among completely resected tumors less than or equal to 5 cm without lymph node metastasis, the current stages IA and IB AD with high serum carcinoembryonic antigen levels, lymphatic permeation, or vascular invasion should be upgraded to stage IB and IIA, respectively. The current stage IA SQ with high SCC antigen levels or vascular invasion should be upgraded to stage IB. These reclassifications accurately reflect survival status (p < 0.04 in all comparisons). CONCLUSIONS: Some important differences in prognostic factors were observed between AD and SQ. High preoperative serum tumor marker levels and lymphovascular invasion should be included as additional criteria in the forthcoming Tumor-Node-Metastasis staging.
INTRODUCTION: Some unfavorable prognostic factors for stage I non-small-cell lung cancers have been reported; however, they are not reflected in the current Tumor-Node-Metastasis classification. METHODS: We retrospectively reviewed 629 patients who underwent complete resection of pathological stage I adenocarcinomas (ADs) or squamous cell carcinomas (SQs) at two institutes between 1996 and 2011. The correlation between clinicopathological characteristics and survival rates was analyzed to identify prognostic factors. RESULTS: Multivariate analysis indicated that among ADs, high serum carcinoembryonic antigen levels (p = 0.04 for overall survival [OS]; p < 0.01 for recurrence-free survival [RFS]; p = 0.02 for disease-specific survival [DSS]), lymphatic permeation (p < 0.01 for RFS and DSS), and vascular invasion (p < 0.01 for OS and RFS; p = 0.03 for DSS) were independent prognostic factors. Among SQs, high squamous cell carcinoma antigen (SCC) (p < 0.05 for OS), and vascular invasion (p < 0.05 for RFS and DSS) were independently prognostic. We suggest that among completely resected tumors less than or equal to 5 cm without lymph node metastasis, the current stages IA and IB AD with high serum carcinoembryonic antigen levels, lymphatic permeation, or vascular invasion should be upgraded to stage IB and IIA, respectively. The current stage IA SQ with high SCC antigen levels or vascular invasion should be upgraded to stage IB. These reclassifications accurately reflect survival status (p < 0.04 in all comparisons). CONCLUSIONS: Some important differences in prognostic factors were observed between AD and SQ. High preoperative serum tumor marker levels and lymphovascular invasion should be included as additional criteria in the forthcoming Tumor-Node-Metastasis staging.