Diagnostic value of tumor volume for differentiating between parathyroid carcinoma and adenoma.

BACKGROUND: The distinction between parathyroid carcinoma (PC) and parathyroid adenoma (PA) remains a considerable clinical challenge. This study evaluated the discriminatory performance of pathology-derived tumor volume for differentiating PC from PA and compared it with tumor length and routine biochemical parameters. METHODS: This retrospective matched case-control study included 15 PC cases and 45 age- and sex-matched PA controls treated between 2008 and 2024. Tumor volume was calculated from three-dimensional pathological specimen measurements using the ellipsoid formula. Discriminatory performance was assessed by receiver operating characteristic (ROC) analysis, with areas under the curve (AUCs) compared using DeLong's test. Parsimonious multivariable logistic regression models assessed the independent association between tumor volume and PC. RESULTS: Tumor volume showed the highest discriminatory performance (AUC = 0.985, 95% CI: 0.914-1.000), with an exploratory pathology-derived cut-off of > 4.71 cm³ yielding 100.0% sensitivity and 97.8% specificity. Its AUC was significantly higher than those of parathyroid hormone, alkaline phosphatase, calcium, and albumin (all P < 0.05), whereas the difference from tumor length was not statistically significant (AUC = 0.937; ΔAUC = 0.048; P = 0.055). Tumor volume remained independently associated with PC after separate adjustment for PTH, ALP, calcium, and albumin (adjusted OR range, 2.017-2.446; all P ≤ 0.003). CONCLUSION: Pathology-derived tumor volume demonstrated excellent discriminatory performance for differentiating PC from PA and outperformed the evaluated biochemical markers, although its advantage over tumor length was not statistically significant. As the identified cut-off is pathology specimen-derived, prospective validation using standardized preoperative radiological volumetry is required before its clinical utility can be established.