BACKGROUND AND OBJECTIVES: It is unclear whether the prognostic significance of the 8th American Joint Committee on Cancer (AJCC) tumor, node, metastasis (TNM) staging system for non-small-cell lung cancer (NSCLC) is applicable to lung cancer as a second primary malignancy (LCSPM). This study used a population-based database to evaluate this relationship. METHODS: Patients diagnosed with second primary lung cancer after a nonpulmonary malignancy were identified from the Surveillance, Epidemiology and End Results (SEER) registry from 2004 to 2015. Cumulative incidence function (CIF) and multivariable CIF regression analyses were performed to estimate the difference in disease-specific mortality (DSM) among different TNM stages. RESULTS: Our cohort included 2687 patients from the SEER database. After CIF analysis, although rates of 1-year, 3-year, and 5-year DSM trended higher with increasing TNM stages, the DSM curves overlapped for many subcategories. In a multivariable regression analysis, hazards ratios (HRs) for subcategories of stage Ι demonstrated no significant difference compared with the reference stage ΙA1 ([ΙA2 HR = 1.120; 95% confidence interval [CI], 0.477-2.626]; [ΙA3 HR = 1.762; 95% CI, 0.752-4.126]; [ΙB HR = 2.003; 95% CI, 0.804-4.911]). The following HRs trended higher for increasing TNM stages but with overlapping CIs among adjacent stage groupings. CONCLUSION: The 8th edition AJCC TNM staging system fails to provide accurate prognostic value for LCSPM.
BACKGROUND AND OBJECTIVES: It is unclear whether the prognostic significance of the 8th American Joint Committee on Cancer (AJCC) tumor, node, metastasis (TNM) staging system for non-small-cell lung cancer (NSCLC) is applicable to lung cancer as a second primary malignancy (LCSPM). This study used a population-based database to evaluate this relationship. METHODS:Patients diagnosed with second primary lung cancer after a nonpulmonary malignancy were identified from the Surveillance, Epidemiology and End Results (SEER) registry from 2004 to 2015. Cumulative incidence function (CIF) and multivariable CIF regression analyses were performed to estimate the difference in disease-specific mortality (DSM) among different TNM stages. RESULTS: Our cohort included 2687 patients from the SEER database. After CIF analysis, although rates of 1-year, 3-year, and 5-year DSM trended higher with increasing TNM stages, the DSM curves overlapped for many subcategories. In a multivariable regression analysis, hazards ratios (HRs) for subcategories of stage Ι demonstrated no significant difference compared with the reference stage ΙA1 ([ΙA2 HR = 1.120; 95% confidence interval [CI], 0.477-2.626]; [ΙA3 HR = 1.762; 95% CI, 0.752-4.126]; [ΙB HR = 2.003; 95% CI, 0.804-4.911]). The following HRs trended higher for increasing TNM stages but with overlapping CIs among adjacent stage groupings. CONCLUSION: The 8th edition AJCC TNM staging system fails to provide accurate prognostic value for LCSPM.