Development and validation of a novel nomogram for predicting long-term survival in gallbladder cancer: a large population-based study using the SEER database.

Background: Gallbladder cancer (GBC) remains an aggressive malignancy with a generally poor long-term outlook. In this study, we sought to develop and internally validate a prognostic nomogram for predicting overall survival (OS), aiming to improve individualized risk assessment for these patients. Methods: We identified 3,892 patients diagnosed with GBC between 2000 and 2021 from the Surveillance, Epidemiology, and End Results (SEER) database. Eligible cases were randomly divided into a training cohort (70%, n=2,724) and an internal validation cohort (30%, n=1,168). Candidate predictors covered demographic characteristics, tumor‑related factors and treatment information. Univariate and multivariate Cox proportional hazards regression were performed sequentially on the training set to identify independent prognostic factors, which were then used to construct a nomogram predicting 3-, 5- and 8-year OS. Model performance was evaluated comprehensively using the concordance index (C‑index), time‑dependent receiver operating characteristic (ROC) curves, area under the curve (AUC), calibration plots, net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA). All statistical tests were two‑sided, with significance set at P<0.05. Results: Independent prognostic factors included age, tumor grade, T stage, N stage, M stage, surgery, and radiotherapy. The nomogram yielded a C‑index of 0.708 [95% confidence interval (CI): 0.670-0.745] in the training set and 0.623 (95% CI: 0.583-0.622) in the internal validation set. Time‑dependent AUCs for 3-, 5- and 8‑year OS were 0.752, 0.757, and 0.763 in the training set, and 0.758, 0.767, and 0.771 in the validation set, respectively. Compared with the American Joint Committee on Cancer (AJCC) staging system, our nomogram showed significantly improved risk reclassification at 3, 5 and 8 years (NRI =0.460, 0.467 and 0.494; all P<0.001) as well as better IDI (IDI =0.058, 0.044 and 0.041; all P<0.001) in the internal validation cohort. Calibration curves indicated excellent agreement between predicted and observed survival probabilities. DCA confirmed that the nomogram offered a higher net clinical benefit across a wide range of threshold probabilities, suggesting its potential usefulness in clinical decision‑making. Conclusions: Compared to the AJCC system, this novel nomogram provides better risk stratification for 3-, 5- and 8‑year survival in GBC, with moderate discrimination, good calibration, and clinical utility. It may serve as a practical tool for individualized prognosis and treatment planning.