Early postoperative calcitonin-to-preoperative calcitonin ratio as a predictive marker for structural recurrence in sporadic medullary thyroid cancer: A retrospective study

BACKGROUND: Calcitonin (Ctn) is widely used as a marker in the diagnosis, prognosis, and postoperative follow-up of patients with medullary thyroid carcinoma (MTC). The prognostic value of postoperative calcitonin-to-preoperative calcitonin ratio (CR), reflecting the change in Ctn level of response to initial treatment, remains uncertain in long-term disease outcomes. This study aims to determine the cut-off value of CR for predicting structural recurrence and assess the prognostic role of CR in patients with MTC. METHODS: We retrospectively reviewed patients with MTC in Sun Yat-sen University Cancer Center (SYSUCC) between 2000 and 2022. CR is defined as the ratio of postoperative Ctn level on the day of discharge divided by preoperative Ctn level. In order to determine the optimal cut-off value of CR, the receiver operating characteristic (ROC) analysis was performed. We evaluate the effect of CR on recurrence-free survival (RFS) by using the Kaplan-Meier method and Cox regression analysis. Then, a nomogram based on CR was constructed. RESULTS: In total, 112 sporadic MTC patients were included in this study. The optimal cut-off value of CR that predicted disease recurrence was 0.125. Patients with CR≥0.125 showed significantly worse RFS than patients with CR <0.125, respectively (3-years RFS rate of 63.1 vs. 94.7%, 5-years RFS rate of 50.7 vs. 90.3%, P < 0.001). In the multivariate analysis, CR was the strongest independent predictor of structural recurrence (HR: 5.050, 95% CI: 2.247–11.349, P <0.001). Tumor size (HR: 1.321, 95% CI: 1.010–1.726, P =0.042), multifocality (HR: 2.258, 95% CI: 1.008–5.058, P =0.048) and metastasized lymph nodes (HR: 3.793, 95% CI: 1.617–8.897, P <0.001) were also independent predictors of structural recurrence. The uncorrected concordance index (c-index) of the nomogram was 0.827 (95% CI, 0.729-0.925) for RFS, and bias-corrected c-index were similar. As compared to TNM stage, the nomogram based on CR provided better discrimination accuracy. CONCLUSIONS: We demonstrate that CR is a strong prognostic marker to predict structural recurrence in patients with sporadic MTC. The nomogram incorporating CR provided useful prediction of RFS for patients with sporadic MTC to provide personalized treatment.