Abstract
BACKGROUND: Signet-ring cell carcinoma (SRCC) of the bile duct is a rare malignancy with poorly characterized clinical features and prognosis. Given its rarity and the critical need to differentiate its behavior from common cholangiocarcinoma (CCA), this population-based study aimed to characterize SRCC and compare survival outcomes with CCA, while developing prognostic nomograms for individualized prediction. METHODS: We analyzed data from the Surveillance, Epidemiology, and End Results (SEER) database [2000-2021], identifying 98 SRCC and 18,979 CCA cases. Propensity score matching (PSM) analysis (1:5 ratio) was conducted to balance baseline characteristics between groups (98 SRCC vs. 490 CCA). Independent prognostic factors for overall survival (OS) and cancer-specific survival (CSS) were identified using multivariable Cox regression. These factors were incorporated into nomograms for OS and CSS prediction. Model performance was evaluated using the concordance index (C-index), calibration curves, receiver operating characteristic (ROC) curve analysis, and decision curve analysis (DCA). RESULTS: SRCC predominantly involved the extrahepatic bile duct (86.7%) and exhibited poor survival [median OS (mOS) 7.0 months, 95% confidence interval (CI): 4.0-10.0; median CSS (mCSS) 8.0 months, 95% CI: 5.0-12.0]. After PSM, no significant survival difference was observed between SRCC and CCA (P>0.05). Older age and distant metastasis were independent factors associated with poor survival, whereas surgery and chemotherapy were independently associated with better survival. The nomograms demonstrated moderate predictive accuracy (C-index: OS 0.78, 95% CI: 0.71-0.84; CSS 0.78, 95% CI: 0.70-0.85). Calibration curves showed excellent agreement between predicted and observed survival, ROC analysis confirmed discriminative ability, and DCA indicated strong clinical utility. CONCLUSIONS: SRCC of the bile duct is a rare, aggressive malignancy with a prognosis similar to CCA. The developed nomograms, integrating readily available clinical factors, provide clinically applicable tools for individualized prognosis prediction.