Development and validation of a diagnostic and prognostic model for bone metastasis of intrahepatic cholangiocarcinoma: a population-based analysis

Background: Bone metastasis (BM) is a common site of metastasis in patients with intrahepatic cholangiocarcinoma (ICC), significantly impacting the quality of life and prognosis of affected individuals. This investigation aimed to assess the risk of BM development in ICC patients and to prognosticate for patients with ICC-associated BM (ICCBM) through the construction of two nomograms. Methods: We conducted a retrospective analysis of data from 2,651 ICC patients, including 148 cases of BM, documented in the Surveillance, Epidemiology, and End Results (SEER) database spanning 2010 to 2017. Independent predictors for the occurrence of BM in ICC patients were identified via univariate and multivariate logistic regression analyses; simultaneously, independent prognostic indicators for ICCBM patients were ascertained through univariate and multivariate Cox regression analyses. The utility of the nomograms was evaluated through calibration curves, receiver operating characteristic (ROC) curves, decision curve analysis (DCA), and Kaplan-Meier (KM) analysis. Results: Independent risk factors for BM in ICC included sex, tumor size, lung metastasis, brain metastasis, and intrahepatic metastasis. For ICCBM patients, independent prognostic factors comprised age, chemotherapy, and radiotherapy. The prognostic nomogram exhibited C-indexes of 0.737 [95% confidential interval (CI): 0.682-0.792] for the training cohort and 0.696 (95% CI: 0.623-0.769) for the validation cohort. Calibration curves demonstrated strong concordance between predicted outcomes and observed events. The areas under the curve (AUC) for 3-, 6-, and 12-month cancer-specific survival (CSS) were 0.853, 0.781, and 0.739, respectively, in the training cohort, and 0.794, 0.822, and 0.780 in the validation cohort. DCA illustrated significant net benefits across a broad spectrum of threshold probabilities. KM analysis revealed 1-, 2-, and 3-year CSS rates of 23.91%, 7.55%, and 2.35%, respectively, with a median CSS of 6 months, underscoring the nomograms' capacity to distinctly stratify patients according to survival risk. Conclusions: The development of these nomograms offers substantial clinical utility in forecasting BM risk among ICC patients and prognosticating for those with ICCBM, thereby facilitating the formulation of more efficacious treatment modalities.