Prognostic factors and prediction of long-term survival in ES-SCLC.

Abstract

e20114 Background: This study aims to investigate the prognostic factors associated with survival outcomes in patients with extensive-stage small cell lung cancer (ES-SCLC) and to establish a predictive model for identifying long-term survivors, providing valuable insights to guide future clinical practice. Methods: This single-center retrospective study analyzed data from 574 patients with ES-SCLC who received treatment at the Fourth Hospital of Hebei Medical University. Using Cox regression to explore independent risk factors affecting the prognosis of patients with extensive-stage small cell lung cancer (ES-SCLC), we will predict long-term survival (>2 years) by ROC curve analysis for the independent risk factors. Subsequently, the independent risk factors will be incorporated into a risk scoring model using the formula: Risk Score = β1X1 + β2X2 + ⋯ + βnXn. Finally, an ROC curve will be used to create a predictive model for long-term survival in patients with ES-SCLC. The predictive performance of individual indicators was compared to that of the combined multivariate indicators, highlighting the enhanced prognostic accuracy achieved through their integration. To prevent excessive complexity in the risk model from incorporating numerous variables, only the top three indicators with the highest predictive power were selected for inclusion in the final model. Differences were considered statistically significant at P<0.05. Results: The ROC curve identified a progression-free survival (PFS) cutoff value of 7.5 months, with an area under the curve (AUC) of 0.89, a sensitivity of 0.81, and a specificity of 0.86, demonstrating robust predictive accuracy. chemotherapy, smoking history, bone metastases, liver metastases, elevated LDH (LDH >250 U/L), suboptimal treatment response, and shorter PFS (≤7.5 months) were identified as independent prognostic factors for poorer overall survival (OS). The probabilities of long-term survival were evaluated based on various factors, including treatment regimen (0.51), smoking status (0.54), presence of bone metastases (0.56), presence of liver metastases (0.62), LDH levels (0.57), treatment response (0.81), and PFS (0.81). Among these, PFS demonstrated the highest predictive accuracy, with a sensitivity of 0.86 and a specificity of 0.77. The combined indicators, including PFS, liver metastases, and LDH levels, yielded an area under the ROC curve (AUC) of 0.76. The optimal predictive probability was achieved at a comprehensive risk cutoff value of 0.8, with a sensitivity of 0.89 and specificity of 0.58. Notably, the predictive performance of PFS alone surpassed that of the combined indicators, highlighting its superior prognostic value. Conclusions: The prognosis of small cell lung cancer patients is influenced by metastatic organs and serological indicators. Notably, a PFS of >7.5 months was strongly correlated with an increased probability of OS exceeding 2 years.