The screening of optimal primary tumor resection candidates in patients with small cell lung cancer: a population-based predictive model.

Abstract

Background: Although a strong survival benefit has been observed among small cell lung cancer (SCLC) patients undergoing surgery, not all SCLC patients benefit from surgery. To help clinicians make choices and decisions regarding surgical intervention, we have developed an effective model to screen beneficial candidates based on population and tumor characteristics.

Methods: Patients with SCLC were acquired from the Surveillance, Epidemiology, and End Results database. Propensity score matching (PSM) was performed to balance covariates between the surgery and non-surgery groups. We assumed that patients undergoing surgery between 2014 and 2018 would benefit from the procedure if their median cancer-specific survival (CSS) time was longer than that of non-surgical patients. Univariate and multivariable logistic analyses were used to identify independent factors of surgical benefit in the surgery group. According to these preoperative factors, a nomogram was built and then internal and external validation were performed.

Results: In total, 35,214 patients with complete data were included for subsequent analysis, 1,364 of whom underwent surgery. Before and after PSM, surgery was an independent factor of long-term survival, with a median CSS time of 37.00 months for the surgery group compared to 16.00 months for the non-surgery group. A multivariable logistic model identified T stage, N stage, M stage, tumor site, and age as independent factors, which were used to establish a stable predictive model.

Conclusions: We have built a preoperative predictive model for SCLC patients to screen for optimal surgery candidates. This model has the potential to help clinicians determine whether it is beneficial to operate on patients with SCLC.