Prognostic factors of survival in patients with non-small cell lung cancer: a competing risk model using the SEER database

Abstract

Background: To explore the prognostic factors of survival in non-small-cell lung cancer (NSCLC) patients using the competing risk analysis.Methods: This was a retrospective cohort study. NSCLC patients with complete data were selected from the Surveillance, Epidemiology, and End Results (SEER) database between 2010 and 2015. Outcomes were censored, cancer-specific mortality in NSCLC, and other-cause mortality. Gray's test was used in univariable analysis, and a multivariable Fine-Gray competing risk model with backward elimination was used to explore the prognostic factors of survival. The screened variables were incorporated into a random survival forest (RSF) model for the prediction of 1-, 3-, and 5-year survival in patients with NSCLC. Receiver operator characteristic (ROC) curves, calibration curves, the value of area under the curve (AUC), and decision curve analysis (DCA) were used to evaluate the performance.Results: Totally 1,251 eligible patients were included, 678 (54.20%) patients were cancer-specific mortality, and 128 (10.23%) patients were other-cause mortality. The median follow-up time was 26 months. Age, primary site, N stage, M stage, surgery type, tumor size, and lymph nodes (LNs) count were included in the multivariable Fine-Gray model for further analysis (P<0.05). The six most important features (surgery type, tumor size, M stage, LNs count, N stage, and primary site) were included in the competing risk analysis using the RSF model. The value of AUC for predicting 1-, 3-, and 5-year survival in the testing set were 0.796, 0.804, and 0.792, respectively. Calibration curves were well-fitted. DCA curves showed that the RSF model had similar or greater clinical net benefits in survival compared with the 8th edition the American Joint Committee on Cancer Staging (AJCC) staging. The good performance of the RSF model under different surgery types, T, N, and M stages. Conclusions:We conducted a competing risk analysis using the RSF model for predicting the 1-, 3-, and 5-year survival of NSCLC. We generated a web calculator (https://github.com/YingChen19/Prognosticfactors-of-long-term-survival-of-non-small-cell-lung-cancer), which could provide a convenient assessment and could help improve the prognosis and survival of NSCLC.

Assessing the prognostic value of KRAS mutation combined with tumor size in stage I-II non-small cell lung cancer: a retrospective analysis

“…One strategy is to stratify patients according to KRAS mutational status together with other key prognostic factors, such as tumor size. Primary tumor size is an established prognostic factor in NSCLC, with larger tumors being associated with poorer survival (24,(31)(32)(33)(34). The reason for this association is not yet fully understood but larger tumors may be more resistant to therapy due to having poorer blood supply, differential metabolism, and potentially a higher likelihood of micrometastatic disease compared to smaller tumors (35).…”

Combined assessment ofKRASmutational status and tumor size has no impact on prognosis in early-stage non-small cell lung cancer

“…One strategy is to stratify patients according to KRAS mutational status together with other key prognostic factors, such as tumor size. Primary tumor size is an established prognostic factor in NSCLC, with larger tumors being associated with poorer survival [21,[28][29][30][31]. The reason for this association is not yet fully understood but larger tumors may be more resistant to therapy due to having poorer blood supply, differential metabolism, and potentially a higher likelihood of micrometastatic disease compared to smaller tumors [32][33][34][35].…”

“…Primary tumor size is an established factor in prognosis, with larger tumors associated with poorer survival [6][7][8][9][10]. The reason for this association is not yet fully understood but it is believed that larger tumors are more resistant to therapy due to having poorer blood supply, a different metabolism and potentially a higher likelihood of micrometastatic disease compared to smaller tumors.…”