STUDY DESIGN: Retrospective analysis. OBJECTIVE: To develop and validate nomograms to predict overall survival (OS) and cancer-specific survival (CSS) of spinal chondrosarcoma patients. SUMMARY OF BACKGROUND DATA: In this era of personalized medicine, data those are available to predict the survival of spinal chondrosarcoma patients are still limited due to the rarity of the disease. Nomogram, which has been widely used in clinical oncology, could conveniently and precisely predict survival outcome for individual patient. METHODS: We retrospectively collected 450 spinal chondrosarcoma patients from the Surveillance, Epidemiology, and End Results (SEER) database between 1984 and 2013. Univariate log-rank and multivariate Cox analyses were used to identify independent prognostic factors. These prognostic factors were included in the nomograms, which predict 3- and 5-year OS and CSS rate. The nomograms were bootstrap validated internally and externally. RESULTS: A total of 450 patients were collected and randomly assigned into the training (n = 225) and validation (n = 225) cohorts. Age, histologic subtype, grade, tumor size, stage, and surgery were identified as independent prognostic factors for OS and CSS (all P < 0.05) and were further incorporated to construct the nomograms. The concordance indices (C-indices) for internal validation of OS and CSS prediction were 0.807 and 0.821, while for external validation of OS and CSS prediction were 0.756 and 0.767. Internal and external calibration plots both revealed an excellent agreement between nomogram prediction and actual survival. CONCLUSION: Nomograms were developed to predict OS and CSS for spinal chondrosarcoma patients. The nomograms could assist clinicians in making more accurate survival evaluation and identifying patients with high risk of mortality. LEVEL OF EVIDENCE: 4.
STUDY DESIGN: Retrospective analysis. OBJECTIVE: To develop and validate nomograms to predict overall survival (OS) and cancer-specific survival (CSS) of spinal chondrosarcomapatients. SUMMARY OF BACKGROUND DATA: In this era of personalized medicine, data those are available to predict the survival of spinal chondrosarcomapatients are still limited due to the rarity of the disease. Nomogram, which has been widely used in clinical oncology, could conveniently and precisely predict survival outcome for individual patient. METHODS: We retrospectively collected 450 spinal chondrosarcomapatients from the Surveillance, Epidemiology, and End Results (SEER) database between 1984 and 2013. Univariate log-rank and multivariate Cox analyses were used to identify independent prognostic factors. These prognostic factors were included in the nomograms, which predict 3- and 5-year OS and CSS rate. The nomograms were bootstrap validated internally and externally. RESULTS: A total of 450 patients were collected and randomly assigned into the training (n = 225) and validation (n = 225) cohorts. Age, histologic subtype, grade, tumor size, stage, and surgery were identified as independent prognostic factors for OS and CSS (all P < 0.05) and were further incorporated to construct the nomograms. The concordance indices (C-indices) for internal validation of OS and CSS prediction were 0.807 and 0.821, while for external validation of OS and CSS prediction were 0.756 and 0.767. Internal and external calibration plots both revealed an excellent agreement between nomogram prediction and actual survival. CONCLUSION: Nomograms were developed to predict OS and CSS for spinal chondrosarcomapatients. The nomograms could assist clinicians in making more accurate survival evaluation and identifying patients with high risk of mortality. LEVEL OF EVIDENCE: 4.