Prediction of distant metastases at diagnosis in medullary thyroid cancer: combining calcitonin with lymph node imaging

This is the authors' abstract. We don't add key points for this paper.

Context Current guidelines recommend performing preoperative additional imaging in patients with medullary thyroid cancer (MTC) when calcitonin levels exceed 500 pg/mL to detect distant metastases (M1). However, this recommendation is based on limited evidence. Whether this is the optimal cutoff, including other diagnostic characteristics, has only been partially evaluated in research. Objective The aims of this study were to evaluate the current calcitonin-driven recommendation and to investigate whether other diagnostic characteristics can be added to improve prediction of M1 disease at diagnosis. Methods Adult MTC patients treated in a tertiary care hospital between 1984 and 2023 with a preoperative calcitonin measurement were retrospectively collected. M1 disease was detected by preoperative imaging or biopsy. Logistic regression was used to identify new predictors for M1 at diagnosis. Results In total, 81 patients with MTC were included for analysis. M1 disease at presentation was found in 27%. Sensitivity and specificity for the current calcitonin cutoff were 90.9% (95% CI, 72.2-97.5) and 47.5% (95% CI, 35.3-60.0), respectively. In multivariable analysis, presence of suspicious lymph nodes on preoperative ultrasound was the strongest predictor (odds ratio [OR] 6.7; 95% CI, 1.3-34.2; P = .022) followed by calcitonin (OR 1.9; 95% CI, 1.2-2.8; P = .005) for M1 disease. Conclusion To our knowledge, this is the first study investigating the optimal combination of predictors for M1 disease in MTC at diagnosis. Suspicious lymph nodes on ultrasound is the strongest predictor for M1 disease, well exceeding calcitonin levels. Incorporating suspicious lymph nodes and calcitonin into a novel clinical decision tool may optimize M1 detection while reducing unnecessary imaging.