Objective: The effects of surgical and non-surgical treatment on the overall survival(OS)and the lung cancer specific survival(LCSS)of patients with stage IIB-IIIA small cell lung cancer(SCLC)was analyzed based on Surveillance,Epidemiology,and End Results(SEER)database.In addition,the independent prognostic factors affecting of stage IIBIIIA SCLC were screened out to further build a clinical prognostic prediction model.Methods: The retrospective analysis was performed on 1211 cases who were diagnosed with pathological stage IIB-IIIA SCLC from 2010 to 2018,and they were divided into surgery group and non-surgery group according to treatment methods.Kaplan-Meier survival curve was used to evaluate the effect of surgical treatment and non-surgical treatment on OS/LCSS in patients with stage IIB-IIIA SCLC.Meanwhile,propensity score matching(PSM)was applied to balance baseline characteristics,both groups were compared using survival analysis again.Cox regression models were employed to identify and screen stage IIB-IIIA SCLC risk factors,and a Nomogram prognostic prediction model was developed,whose predictive accuracy was verified by calibration plots and ROC curves.Results:1 Survival analysis1.1 Before PSM: the median OS in the surgery group was superior to the non-surgery group(2.08 years(1.22 to 2.94)vs.1.25 years(1.13 to 1.37),respectively)(P<0.001;hazard ratio(HR): 0.63,95% CI: 0.51 to 0.77).At the end point was LCSS,compared with the non-surgical group,the surgery group prolonged the median LCSS of patients by 2.58years(1.08 to 4.01)vs.1.42 years(1.30 to 1.54),respectively.(P<0.001;HR: 0.60,95%CI:0.48 to 0.75)1.2 After PSM: the median OS values was higher in patients who received surgery versus those who did not(2.00 years(1.33 to 2.67)vs.1.58 years(1.29 to 1.87)).There was still significant difference in OS between the two groups(P=0.041;HR:0.73,95%CI:0.53 to0.99).However,the median LCSS of the surgical group versus the non-surgical group was2.17 years(0.68 to 3.66)vs.1.67 years(1.26 to 2.08),with no statistically significant difference(P=0.058;HR: 0.72,95% CI: 0.51 to 1.02).2 Nomogram prognostic prediction model: multivariate Cox regression models showed that age,TNM stage,surgery and chemotherapy were independent factors affecting IIBIIIA SCLC prognosis.A Nomogram is constructed based on Cox regression proportional risk model.C-index,ROC curves and calibration plots have shown that some predictive stability and good inter-rater reliability of the model.Conclusions: Compared with non-surgical treatment,surgical treatment can improve the OS of patients with stage IIB-IIIA SCLC.The independent prognostic factors of stage IIBIIIA SCLC patients were age,TNM stage,surgery and chemotherapy.In addition,this Nomogram prognostic model for stage IIB-IIIA SCLC patients established in this study has some predictive power and accuracy,and is expected to provide a better prognostic means for SCLC patients. |