Predict overall survival of spinal conventional chordoma: Development and assessment of a new predictive nomogram.

Abstract Objective To predict the 5-year overall survival (OS) rate in patients with conventional chordoma of the spine Patients and methods The Surveillance, Epidemiology, and End Results (SEER) Registry was used to identify patients with conventional chordoma of the spine from 1994 to 2013. The entire cohort(n = 294) was randomly divided into training (n = 147) and validation (n = 147) cohorts to construct a nomogram. We used the univariate Log-rank test and multivariate Cox model to examine the independent prognostic factors associated with OS. These prognostic factors were integrated to construct a nomogram through R studio. The predictive and validating capacity of the nomogram was calculated by Harrell’s concordance index (C-index) and calibration curves. Results A total of 294 patients were identified with conventional chordoma of the spine. The patients’ age at diagnosis, tumor size, EOD (extent of disease), and treatment were independent prognostic factors and associated with OS. These prognostic factors were incorporated to construct a nomogram. The concordance index for the nomogram was 0.771 and 0.732 in the training cohort and validation cohort, respectively. Internal and external calibration curves for 5-year OS showed excellent matching between nomogram prediction and observed outcomes. Conclusions The findings of this study provide population-based estimates of patients with conventional chordoma of the spine. Using this nomogram, surgeons can classify patients into different risk groups and achieve individualized treatment.