Development and validation of prognostic nomograms for medullary thyroid cancer

BACKGROUND: This aim of study was to develop and validate clinical nomograms to predict the survival of patients with medullary thyroid cancer. PATIENTS AND METHODS: Patient data were collected from the Surveillance, Epidemiology, and End Results database between 2004 and 2013. All included patients were randomly assigned into the training and validation sets. Multivariate analysis using Cox proportional hazards regression was performed, and nomograms were constructed. Model performance was evaluated by discrimination and calibration plots. RESULTS: A total of 1,657 patients were retrospectively analyzed. The multivariate Cox model identified age, tumor size, extrathyroidal extension, N stage, and M stage as independent covariates associated with overall survival (OS) and cancer-specific survival (CSS). Nomograms predicting OS and CSS were constructed based on these covariates. The nomograms predicting both OS and CSS exhibited superior discrimination power to that of TNM staging system in the training and validation sets. Calibration plots indicated that both the nomograms in OS and CSS exhibited high correlation to actual observed results. CONCLUSION: The nomograms established in this study provided an alternative tool for prognostic prediction, which may thereby improve individualized assessment of survival risks and lead to the creation of additional clinical therapies.