biostudies-literatureUnknown13(1)Shao YAt present, no study has established a survival prediction model for non-metastatic primary malignant bone tumors of the spine (PMBS) patients. The clinical features and prognostic limitations of PMBS patients still require further exploration. Data on patients with non-metastatic PBMS from 2004 to 2015 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Multivariate regression analysis using Cox, Best-subset and Lasso regression methods was performed to identify the best combination of independent predictors. Then two nomograms were structured based on these factors for overall survival (OS) and cancer-specific survival (CSS). The accuracy and applicability of the nomograms were assessed by area under the curve (AUC) values, calibration curves and decision curve analysis (DCA). Results: The C-index indicated that the nomograms of OS (C-index 0.753) and CSS (C-index 0.812) had good discriminative power. The calibration curve displays a great match between the model's predictions and actual observations. DCA curves show our models for OS (range: 0.09-0.741) and CSS (range: 0.075-0.580) have clinical value within a specific threshold probability range compared with the two extreme cases. Two nomograms and web-based survival calculators based on established clinical characteristics was developed for OS and CSS. These can provide a reference for clinicians to formulate treatment plans for patients.Scientific reports3503https://www.ebi.ac.uk/biostudies/studies/S-EPMC9977926biostudies-literatureDevelopment and validation of nomograms predicting overall and cancer-specific survival for non-metastatic primary malignant bone tumor of spine patients.PMC9977926Shao YWang YWang ZShi XfalseDevelopment and validation of nomograms predicting overall and cancer-specific survival for non-metastatic primary malignant bone tumor of spine patients.At present, no study has established a survival prediction model for non-metastatic primary malignant bone tumors of the spine (PMBS) patients. The clinical features and prognostic limitations of PMBS patients still require further exploration. Data on patients with non-metastatic PBMS from 2004 to 2015 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Multivariate regression analysis using Cox, Best-subset and Lasso regression methods was performed to identify the best combination of independent predictors. Then two nomograms were structured based on these factors for overall survival (OS) and cancer-specific survival (CSS). The accuracy and applicability of the nomograms were assessed by area under the curve (AUC) values, calibration curves and decision curve analysis (DCA). Results: The C-index indicated that the nomograms of OS (C-index 0.753) and CSS (C-index 0.812) had good discriminative power. The calibration curve displays a great match between the model's predictions and actual observations. DCA curves show our models for OS (range: 0.09-0.741) and CSS (range: 0.075-0.580) have clinical value within a specific threshold probability range compared with the two extreme cases. Two nomograms and web-based survival calculators based on established clinical characteristics was developed for OS and CSS. These can provide a reference for clinicians to formulate treatment plans for patients.2023-01-01T00:00:00Z2023 Mar2026-07-14T15:44:19.321Z2025-04-19T09:20:46.399ZS-EPMC99779263685946510.1038/s41598-023-30509-y