Abstract
OBJECTIVES: To develop and validate a prognostic nomogram incorporating clinical and nutritional factors for individualized survival prediction in elderly patients with inoperable esophageal squamous cell carcinoma (ESCC) treated with intensity-modulated radiotherapy (IMRT). METHODS: This retrospective study enrolled 300 patients aged over 65 with inoperable ESCC who underwent IMRT. Patients were randomly allocated to a training set (n = 211) and a validation set (n = 89) in a 7:3 ratio. Prognostic variables were initially screened using univariate Cox regression, followed by variable selection via the least absolute shrinkage and selection operator regression. Independent predictors were identified through multivariate Cox analysis and incorporated into a nomogram model. Model performance was evaluated using the concordance index, receiver operating characteristic curves, and calibration curves, with the model’s discriminative ability further validated by Kaplan-Meier survival analysis. RESULTS: Four independent prognostic factors were identified: age ≥ 70 years (HR = 1.50, P = 0.019), Karnofsky performance status score ≥ 80 (HR = 0.54, P = 0.003), radiotherapy dose ≥ 58 Gy (HR = 0.53, P < 0.001), and serum albumin ≥ 40 g/L (HR = 0.70, P = 0.041). The nomogram demonstrated predictive ability for 2-year and 5-year overall survival (OS), with better predictive performance for 5-year OS. The area under the curve values in the training cohort were 0.732 and 0.681, while those in the validation cohort were 0.570 and 0.740, respectively. Calibration and decision curve analyses confirmed the clinical utility and accuracy of the model. Compared to the AJCC 8th edition TNM staging system, the nomogram exhibited superior long-term prognostic performance in this population. CONCLUSION: This study proposes a clinical nomogram model to assist clinicians in predicting the survival of elderly patients with inoperable ESCC undergoing IMRT and formulating treatment plans.