Early Diagnosis of Medullary Thyroid Cancer: Are Calcitonin Stimulation Tests Still Indicated in the Era of Highly Sensitive Calcitonin Immunoassays?

ABSTRACT

Background: Measurements of both basal (b) calcitonin (CT) and calcium (Ca)-stimulated CT (Ca-sCT) levels are performed to identify medullary thyroid cancer (MTC) at an early stage when used as part of the diagnostic workup of thyroid nodules (CT screening). Novel immunochemiluminometric assays, which are highly sensitive and specific for monomeric CT and avoid cross-reactivity, have been introduced over the past decade. No prospectively generated data have so far become available to answer the frequently raised question as to whether Ca-sCT in contrast to bCT alone is helpful and, therefore, still indicated for the early detection of MTC. Methods: Ca-stimulation tests were performed in 149 consecutive patients with thyroid nodules and elevated bCT. Regardless of Ca-sCT levels, all patients had an operation applying a uniform surgical protocol, including thyroidectomy and systematic lymph node dissection. Recently published sex-specific cutoff levels for the differentiation of MTC and other C-cell pathologies (C-cell hyperplasia [CCH]) were used to compare the diagnostic performance of bCT or Ca-sCT alone and in combination using receiver-operating characteristic (ROC) analysis. In addition, CT cutoff levels to predict lateral lymph node metastasis were evaluated for bCT compared with Ca-sCT. Follow-up for all patients was documented and correlated with initial CT levels. Results: MTC was identified in 76 (50.1%) patients, in 21/76 (27.6%) with lymph node and in 4 (5.3%) with distant metastasis. Using predefined cutoff levels, patients could effectively be subdivided into a group above the cutoff level with definitive diagnosis of MTC (100%) and below (gray zone) with a significant overlap of CCH and MTC (all classified as pT1a; males 19/58 [37.5%], females 7/41 [17.1%]). The areas under the ROC curve (AUC) were excellent for the diagnosis of MTC in all tests. Determination of bCT proved to be superior for both diagnosing MTC in males (AUC for bCT 0.894; AUC for Ca-sCT 0.849) and females (bCT 0.935; Ca-sCT 0.868) and also for diagnosing lymph node metastasis in the lateral compartment (males bCT 0.925; Ca-sCT 0.810; females bCT 0.797; Ca-sCT 0.674). Combining both tests did not improve diagnostic accuracy. Using a cutoff level of >85 pg/mL for females and >100 pg/mL for males, the sensitivity for diagnosing lateral neck lymph node metastasis was 100%. Below these cutoff levels, no patient showed persistent or recurrent disease (median follow-up 46 [ ± 27] months). Conclusions: Predefined sex-specific bCT cutoff levels are helpful for the early detection of MTC and for predicting lateral neck lymph node metastasis. Ca-sCT did not improve preoperative diagnostics. bCT levels >43 and >100 pg/mL for males and of >23 and >85 pg/mL for females are relevant for advising patients and planning the extent of surgery.