LAUSR

BACKGROUND Parathyroid carcinoma is a rare endocrine malignancy that lacks an established system for risk categorization. This study evaluated a prognostic scoring system for recurrence-free survival (RFS) of patients with parathyroid carcinoma. STUDY DESIGN Patients diagnosed and confirmed to have parathyroid carcinoma and who were treated between 1980 and 2016 at The University of Texas MD Anderson Cancer Center were studied retrospectively. Univariate and multivariate Cox proportional hazards regression analyses of RFS were conducted. A prognostic scoring system was created based on multivariate analysis. RESULTS Sixty-eight patients were evaluated. After a median follow-up of 4.6 years, 26 patients experienced a recurrence. The Kaplan-Meier RFS rates were 85% at 1 year (95% CI 77% to 95%), 67% at 2 years (95% CI 55% to 81%), and 51% at 10 years (95% CI 36% to 72%) after initial operation. Multivariate analysis demonstrated that age older than 65 years, serum calcium level >15 mg/dL, and vascular invasion were negatively correlated with RFS rate. Combining these adverse variables into a prognostic scoring system, we stratified patients into 3 risk groups: low (0 variable; 2-year RFS rate, 93%), moderate (1 variable; 2-year RFS rate, 72%), and high (2 variables; 2-year RFS rate, 27%) (p = 0.001 [log-rank test]; concordance index, 0.70; 95% CI 0.47 to 0.92). CONCLUSIONS A prognostic scoring system using vascular invasion, age, and serum calcium level at initial parathyroidectomy can be used to predict RFS. This categorization might be helpful for clinical decisions relative to the timing and use of adjuvant therapy. Comprehensive validation using multiple cohorts will be needed to confirm applicability.