Development of a prognostic nomogram and risk stratification system for elderly patients with esophageal squamous cell carcinoma undergoing definitive radiotherapy: a multicenter retrospective analysis (3JECROG R-03 A)

Background: Our goal is to develop a nomogram model to predict overall survival (OS) for elderly esophageal squamous cell carcinoma (ESCC) patients receiving definitive radiotherapy (RT) or concurrent chemoradiotherapy (CRT), aiding clinicians in personalized treatment planning with a risk stratification system.

Methods: A retrospective study was conducted on 718 elderly ESCC patients treated with RT or CRT at 10 medical centers (3JECROG) from January 2004 to November 2016. We identified independent prognostic factors using univariate and multifactorial Cox regression to construct a nomogram model. Its effectiveness was evaluated using concordance statistics (C-index), area under the curve (AUC), net reclassification index (NRI), and integrated discrimination improvement (IDI), and compared against the AJCC staging. Additionally, decision curve analysis (DCA) assessed the model's clinical benefit. Patients were stratified into low, intermediate, and high-risk groups using the nomogram, and their prognoses in various disease stages were analyzed.

Results: Significant prognostic factors identified included diabetes, tumor volume (GTVp), tumor length, location, and clinical stages (T, N, M), and RT response. Multivariate analysis confirmed these as independent factors for OS. The nomogram outperformed AJCC staging in prediction accuracy and discrimination, evidenced by a higher C-index, better AUC, and significant NRI and IDI values. Patients categorized by the nomogram demonstrated distinct 5-year OS rates, with a higher C-index than AJCC staging (0.597 vs. 0.562) .

Conclusions: The study identified key prognostic factors for elderly ESCC patients receiving RT or CRT. The nomogram model, based on these factors, showed enhanced prediction performance, discrimination, and clinical utility compared to AJCC staging. This risk stratification provided more accurate survival predictions and aided in personalized risk management.