LAUSR

Background Parathyroid carcinoma (PC) is a rare but often lethal malignancy for which staging system, prognostic indicators, and treatment guidelines are still not established. We aimed to explore the prognostic parameters and construct a nomogram for cancer-specific survival (CSS) of PC. Methods A retrospective analysis of 604 PC patients in the SEER database from 2001 through 2018 was performed. All the cases were randomly assigned to the training cohort (n = 424) or the validation cohort (n = 180) at a ratio of 7:3. The Kaplan–Meier method and Cox regression model were applied to estimate the CSS and risk factors, and a nomogram was constructed. The predictive accuracy and discriminative ability of the nomogram in CSS were assessed by concordance index (C-index), the area under the curve (AUC) of receiver operating characteristics (ROC), and the calibration curve. Results Age at diagnosis > 70 years [hazard ratio (HR): 3.55, 95% CI: 1.07–11.78, p = 0.039] and tumor size > 35 mm (HR 4.22, 95% CI: 1.67–10.68, p = 0.002) were associated with worse CSS. Compared with distant metastasis, localized (HR 0.17, 95% CI: 0.06–0.47, p = 0.001) and regional lesions (HR 0.22, 95% CI: 0.07–0.66, p = 0.007) showed an improved CSS rate. Parathyroidectomy was the recommended treatment (p = 0.02). The C-index of the nomogram was 0.826, and the AUC for 5-, 10-, and 15-year CSS was 83.7%, 79.7%, and 80.7%, respectively. The calibration curve presented good agreement between prediction by nomogram and actual observation. Conclusion Age at diagnosis > 70 years, tumor size > 35 mm, and distant metastasis were independent risk factors for PC-specific mortality. Parathyroidectomy was currently the most recommended treatment for PC. This nomogram provided individualized assessment and reliable prognostic prediction for patients with PC.