Dynamic Nomogram

Dynamic Nomogram predicts overall survival (OS) probabilities for elderly patients with chondrosarcoma (CHS) using prognostic clinical variables from 595 cases in the Surveillance, Epidemiology, and End Results (SEER) database (2004–2018).

Key Features:

• Data source: Retrospective clinical data for 595 elderly chondrosarcoma (CHS) patients were obtained from the Surveillance, Epidemiology, and End Results (SEER) database for 2004–2018.
• Cohort split: The dataset was partitioned into training (419 cases) and validation (176 cases) cohorts at a 7:3 ratio.
• Statistical analyses: Univariate and multivariate Cox regression analyses were performed to identify independent prognostic factors.
• Prognostic variables: Independent factors included age, sex, grade, histology, M stage, surgery, and tumor size.
• Nomogram output: A nomogram was constructed to predict 12-, 24-, and 36-month overall survival (OS) probabilities.
• Validation: Model robustness was assessed using k-fold cross-validation (k=10).
• Performance metrics: Harrell’s concordance index (C-index), receiver operating characteristic (ROC) curves, area under the curve (AUC), calibration curves, decision curve analysis (DCA), integrated discrimination improvement (IDI), and net reclassification index (NRI) were used for evaluation.
• Reported performance: C-index was 0.800 (training) and 0.789 (validation); AUCs at 12-, 24-, and 36-months were 0.866, 0.855, 0.860 (training) and 0.839, 0.856, 0.840 (validation), respectively.
• Comparative performance: The nomogram outperformed the American Joint Committee on Cancer (AJCC) staging system across ROC, IDI, NRI, and DCA metrics.

Scientific Applications:

• Personalized prognosis: Provide individualized 12-, 24-, and 36-month OS probabilities for elderly CHS patients.
• Treatment decision support: Inform clinical management and therapeutic decision-making by integrating multiple prognostic factors.
• Research utility: Enable comparative evaluation of prognostic models and quantitative assessment of predictive performance in CHS studies.

Methodology:

Retrospective SEER data (2004–2018) for 595 elderly CHS patients were split 7:3 (419 training, 176 validation); univariate and multivariate Cox regression identified independent prognostic factors, a nomogram was constructed to predict 12-, 24-, and 36-month OS, validation used k-fold cross-validation (k=10), and performance was evaluated with C-index, ROC/AUC, calibration curves, DCA, IDI, and NRI.