Abstract
Purpose
Parathyroid carcinoma is a rare endocrine malignancy in which much of the long-term morbidity relates to chronic hypercalcemia. Conventional survival analyses treat deaths from causes other than the cancer as censoring events, which can distort estimates of cause-specific risk. We applied competing-risk methodology to characterize cause-specific mortality and to determine whether distinct clinical factors differentially influence cancer-specific and cardiovascular/renal mortality.Methods
Patients with malignant parathyroid carcinoma (primary site C75.0) were identified from the SEER Research Data, 17 Registries, November 2023 submission (2000-2021). Cause-specific, cardiovascular/renal, and other-cause deaths were defined a priori as competing events. Cumulative incidence functions were estimated using the Aalen-Johansen estimator, compared with Gray's test, and modeled with Fine-Gray subdistribution hazard regression, adjusting for age, sex, race, and stage.Results
Among 656 patients (median follow-up, 86 months), 213 died: 61 (9.3%) from cancer-specific causes, 68 (10.4%) from cardiovascular/renal causes, and 84 (12.8%) from other causes. Cancer-specific mortality plateaued after roughly a decade (15-year cumulative incidence, 11.0%), whereas cardiovascular/renal mortality continued to rise (12.3%) and ultimately exceeded it. In multivariable Fine-Gray models, advanced (regional/distant) stage was independently associated with cancer-specific death (subdistribution hazard ratio [sHR], 3.37; 95% CI, 1.72-6.60) but not cardiovascular/renal death, whereas age ≥ 60 years was independently associated with cardiovascular/renal death (sHR, 2.43; 95% CI, 1.49-3.96) but not cancer-specific death.Conclusions
Cancer-specific and cardiovascular/renal mortality follow distinct competing-risk trajectories in parathyroid carcinoma, with stage primarily determining cancer-specific mortality and age primarily determining cardiovascular/renal mortality. These findings support consideration of competing-risk profiles when planning long-term surveillance, including cardiovascular and renal follow-up.