There is currently no widely accepted prognostic model specifically for elderly patients with esophageal squamous cell carcinoma (ESCC) undergoing radiotherapy. This study aimed to develop a nomogram incorporating metabolic imaging parameters from 18F-fluorodeoxyglucose positron emission tomography-computed tomography (18F-FDG PET/CT) to predict overall survival (OS) in this patient population. The clinical need for such a prediction model is significant given the challenges of treatment planning in elderly patients with ESCC undergoing radiotherapy. A retrospective analysis was conducted on 118 elderly patients with ESCC treated with radiotherapy. The patients were evaluated using 18F-FDG PET/CT imaging prior to treatment, and the spleen:liver ratio (SLR) and length of visual tumor (Lv) were identified as potential prognostic indicators. These variables, along with clinical tumor, node, metastasis (cTNM) staging, were used to develop a nomogram model. Key baseline clinical factors, PET variables, inclusion criteria, and follow-up procedures were documented. The model's predictive accuracy was assessed using time-dependent receiver operating characteristic (ROC) curves, the concordance index (C-index), and decision curve analysis (DCA). The patient cohort was stratified into three risk groups based on the total scores derived from the nomogram. SLR and Lv were found to be independent predictors of OS in elderly patients with ESCC. The nomogram developed by incorporating these factors, along with cTNM staging, showed superior predictive power compared to the traditional TNM staging system. ROC curve analysis demonstrated greater accuracy in predicting 1-, 2-, and 3-year OS rates, with area under the curve (AUC) values of 0.771, 0.763, and 0.815, respectively. DCA confirmed that the nomogram provided a greater clinical benefit. Patients were stratified into low-risk, intermediate-risk, and high-risk groups, with corresponding 3-year OS rates of 60.3%, 25.0%, and 3.6%, respectively. The developed nomogram incorporating SLR, Lv, and cTNM staging offers a reliable tool for the risk stratification of elderly patients with ESCC undergoing radiotherapy. This model may serve as a reference for personalized treatment planning, potentially improving clinical outcomes in this patient population.
Read full abstract