Abstract
Objectives
Medullary thyroid carcinoma (MTC) is caused by a malignant transformation in the parafollicular C-cells of the thyroid, where calcitonin (CT) is released. Nowadays, CT is the main tumor marker used in the diagnosis and follow-up of MTC patients. Nonetheless, procalcitonin (PCT) has recently been proposed as a useful complementary/alternative biomarker in MTC. Our aims were to investigate the diagnostic performance of CT and PCT and their combination in the differential diagnosis between active and inactive MTC and between MTC and non-MTC thyroid diseases, respectively.
Methods
Serum samples were collected from 16 patients with active (i.e. primary tumour before surgery or post-surgical recurrent disease) and 23 with inactive (i.e. complete remission) MTC, 125 patients with non-MTC benign thyroid disease and 62 patients with non-MTC thyroid cancers, respectively. Elecsys® CT and PCT measurements were simultaneously performed on the Cobas e601 platform (Roche Diagnostics, Rotkreutz, Switzerland).
Results
Both CT and PCT median values in active MTC (94 pmol/L and 1.17 ng/mL, respectively) were significantly higher compared with inactive MTC (0.28 and 0.06) and either benign (0.37 and 0.06) or malignant (0.28 and 0.06) non-MTC. Undetectable PCT was found in five non-MTC patients with false positive CT results.
Conclusions
Elecsys® PCT assay is a highly sensitive and specific alternative MTC marker. At the very least it appears useful in patients with positive CT results as negative PCT values securely exclude active MTC. The availability of both markers on the same automated platform facilitates reflex or reflective strategies to refine the laboratory diagnosis.
Introduction
Medullary thyroid carcinoma (MTC) is a malignant tumor of the thyroid cells C cells, accounting for about 2–5% of thyroid cancers [1]. Calcitonin (CT), the main C cell secretory product, is considered a sensitive and specific marker for the diagnosis and follow-up of MTC [2]. However, CT assays suffer pre-analytic and analytic drawbacks: i. CT is prone to relatively rapid in vitro degradation by serum proteases which makes rapid processing of samples mandatory [3]; ii. the presence of various different immunoreactive isoforms and fragments, which can lead to inaccurate results (usually false low) as well as poor comparability of results obtained by different assays [4], [5], [6], [7]. Carcinoembryonic antigen (CEA), a nonspecific marker, is also used for monitoring MTC patients, being a reliable predictor of disease progression [8]; however, it is an insensitive marker for early postoperative detection of MTC persistence/relapse [9]. Administrating intravenous calcium or pentagastrin was largely used to improve the accuracy of CT measurement; however specificity is suboptimal and establishing clear reference guidelines for abnormal stimulated serum CT levels is a much more challenging task than establishing reference guidelines for abnormal basal CT levels. Accordingly, also considering additional costs and potential unpleasant side effects, most clinicians feel that provocative testing is no longer necessary [10]. Procalcitonin (PCT), the precursor of CT, circulates at low concentrations in healthy individuals; moreover, it has a concentration-independent and predictable in vivo half-life of 20–24 h and displays excellent in vitro stability in serum or plasma [11], [12]. A strong correlation between PCT and CT levels was proven in patients with MCT and multiple studies confirmed equivalent high sensitivity of CT and PCT in detecting and monitoring MTC [13], [14], [15]. On balance, the above-mentioned studies suggested that PCT, being less affected by analytical, physiological, pharmacological and pathological factors that can influence results of serum CT values, could be a new standard of care in the management of MTC [16], [17], [18], [19]. Accordingly, Lim and colleagues demonstrated a 98–99% negative predictive value for pathologically increased CT levels [20]. Recently, a new CT immunoassay become available on automated Cobas® platforms (Roche Diagnostic, Rotkreuz, Switzerland) [21], [22] making feasible both simultaneous and sequential (reflex) measurement of CT and/or PCT on the same platform. Therefore, the present study was undertaken to compare the clinical performance of the automated Elecsys® CT and PCT assays and their combination or sequential use for the diagnosis of MTC.
Materials and methods
This study was approved by the Ente Ospedaliero Cantonale Institutional Review Board and the Canton Tessin Ethical Committee, Bellinzona (Switzerland) [references CE 3466 and BASEC 2019-00662, respectively]. The study was performed in accordance with the World Medical Association Declaration of Helsinki regarding ethical conduct of research involving human subjects. A total of 226 subjects were included:
Patients with confirmed active MTC (n=16, 5 males, aged 31–78 yr; 11 females, aged 22–75 yr), based on clinical, imaging, or histological evidence of initial, persistent or recurrent disease. Patients with active MTC included six cases of newly diagnosed disease and 10 cases of metastatic disease.
Patients with confirmed MTC, deemed disease free by clinical assessment, imaging, and serum CT determinations (inactive MTC; n=23; five males, aged 22–73 yr; 18 females, aged 19–82 yr).
Patients with untreated benign thyroid diseases, including benign thyroid nodules, hyperthyroidism, hypothyroidism and goiter (n=125; 27 males, aged 26–75 yr; 98 females, aged 18–79 yr).
Patients with untreated (i.e. preoperative sampling) non-medullary differentiated thyroid cancer (DTC) (n=62; 19 males, ages 18–84 yr; 43 females, aged 20–76 yr).
The specimens were obtained from residual samples from patients of our Centre who had undergone CT and other thyroid testing at the EOLAB Central Laboratory (Bellinzona, Switzerland) between June 1st, 2016 and May 30th, 2018. Sera were stored at −80 °C immediately after clinical testing, until testing for Elecsys® CT and PCT was performed for the present study. The final diagnosis and assessment of disease status was based on the longitudinal review of the clinical, imaging and biochemical data and any cytology/histology data available. Patients with renal failure were excluded because this may result in serum CT and PCT elevations.
Analyte measurements
For the purpose of the present study CT and PCT were measured on a fully automated analyzer Cobas e601 (Roche Diagnostics, Rotkreuz, Switzerland) that uses electrochemiluminescence (ECL) technology for immunoassay analysis (https://diagnostics.roche.com/global/en/products/instruments/cobas-e-601.html as last accessed on 20.09.2020). The Elecsys® Calcitonin assay is standardized against IRP WHO 89/620 international standard. Limits of blank (LoB), detection (LoD) and quantification (LoQ, with a total allowable error of ≤30 %) values are 0.09 pmol/L, 0.14 pmol/L/L, 0.28 pmol/L, respectively (https://diagnostics.roche.com/global/en/products/params/elecsys-calcitonin.html as last accessed on 20.09.2020). As the intellectual property for commercial PCT assays is licensed out by BRAHMS® GmbH (Henningdorf, Germany) the Elecsys® BRAHMS procalcitonin assay is standardized against BRAHMS® PCT luminescent immunoassay (LIA). Analytical (AS) and functional (FS) sensitivity values are 0.02 ng/mL and 0.06 ng/mL, respectively (https://diagnostics.roche.com/fi/en/products/params/elecsys-brahms-procalcitonin-pct.html as last accessed on 20.09.2020). All assays were performed in strict adherence to the manufacturers’ instructions.
Data analysis and statistics
For the purpose of statistical analysis, measurement results below the FS were replaced with 0.28 pmol/L and 0.06 ng/mL for CT Elecsys® and PCT Elecsys®, respectively. For all variables, differences in the distribution of the values among the four different subgroups of patients were estimated with Kruskal-Wallis test and in case of positive result a post-hoc pairwise comparison of subgroups was performed according to Conover [23]. All continuous variables were dichotomized, using receiver-operating characteristic (ROC) analysis and Youden’s coefficient (index J) to identify the optimal cut-off point to discriminate between active MTC patients from subjects with complete-remission MTC. Data analysis and ROC curves plotting were performed using MedCalc Statistical Software version 19.4.1 (MedCalc Software Ltd, Ostend, Belgium; 2020). For all tests a p-value <0.05 was considered statistically significant.
Results
The distribution of serum CT and PCT concentrations in patients with active MTC, inactive MTC, benign and malignant non-MTC thyroid diseases is showed in Figure 1.
![Figure 1:
Distribution of serum CT and PCT in patients with active MTC, inactive MTC, non-MTC benign and non-MTC malignant thyroid diseases, respectively [squares, median values; bars, interquartile ranges (IQR)].](/document/doi/10.1515/cclm-2020-1424/asset/graphic/j_cclm-2020-1424_fig_001.jpg)
Distribution of serum CT and PCT in patients with active MTC, inactive MTC, non-MTC benign and non-MTC malignant thyroid diseases, respectively [squares, median values; bars, interquartile ranges (IQR)].
Median serum concentrations of both markers were significantly higher in patients with active MTC compared with other groups. In addition, higher CT levels were found in patients with non-MTC benign disease compared to both inactive MTC and non-malignant MTC, respectively (Table 1).
Comparison of calcitonin and procalcitonin levels in different patients’ groups (data are expressed as median and interquartile range, IQR).
| MTC | Non-MTC | Kruskal-Wallis p-Value | Conover test p<0.05 | |||
|---|---|---|---|---|---|---|
| Active (1) | Inactive (2) | Benign (3) | Malignant (4) | |||
| CT, pmol/L | 94.10 (13.1–283) | 0.28 (0.28–0.74) | 0.37 (0.28–1.13) | 0.28 (0.28–0.48) | <0,0001 | 1 vs. 2,3,4 2 vs. 1 3 vs. 1.4 4 vs. 1.3 |
| PCT, ng/mL | 1.17 (0.45–3.46) | 0.06 (0.06–0.06) | 0.06 (0.06–0.06) | 0.06 (0.06–0.06) | <0,0001 | 1 vs. 2,3,4 2 vs. 1 3 vs. 1 4 vs. 1 |
-
ROC curves for CT were plotted using MTC patients with active (sensitivity) and inactive (specificity) disease and showed a 100% sensitivity and specificity [area under the curve (AUC) of 1.00 (95%C.I. 0.91–1.00), p<0.001]. The best cutoff value was derived from ROC curve analysis (Youden index J) and settled at 3.1 pmol/L (Figure 2).
No overlaps of PCT values were observed between patients with active MCT compared to other groups. Accordingly, the highest value in non-MTC patients (i.e. 0.20 ng/mL) was adopted as cutoff point to maximize the specificity without detriment in sensitivity. Applying the above described cutoff levels the corresponding sensitivity, specificity, positive (LHR+) and negative (LHR-) likelihood ratio values were 100%, 100%, 210 and 0 respectively. Discrepant results were found in five patients affected by non-MTC benign disease showing true-negative PCT but false-positive CT results, respectively (Table 2).
Patients with discordant calcitonin (cutoff 3.1 pmol/L) and procalcitonin (cutoff: 0.20 ng/mL) results.
| Patient | Gender | Age | Diagnosis | CT (pmol/L) | PCT (ng/mL) |
|---|---|---|---|---|---|
| 1 | M | 49 | GD | 3.70 | 0.06 |
| 2 | M | 53 | MNG | 4.30 | 0.06 |
| 3 | F | 53 | MNG | 3.40 | 0.06 |
| 4 | F | 54 | MNG | 4.10 | 0.06 |
| 5 | F | 74 | MNG | 4.30 | 0.08 |
-
GD, Graves’ disease; MNG, multinodular goitre; CT, calcitonin; PCT, procalcitonin.
Since gender and thyroid volume is a relevant determinant for CT levels in non-thyroidectomized patients the performance of CT in discriminating patients with active MTC from those carrying benign and malignant non-MTC diseases was also evaluated by using i. the 97.5th percentile of CT distribution in males (3.7 pmol/L) and females (3.3 pmol/L) with non-MTC thyroid diseases (benign and malignant) and ii. the manufacturer-provided cutoff (males 2.8 pmol/L, females 1.9 pmol/L). Results are summarized in Table 3.
Calcitonin: discrimination of patients with active MTC from those with non-MTC thyroid diseases using different cutoff points.
| Cutoff (criteria) | Cutoff (pmol/L) | Sensitivity | Specificity | +LHR | −LHR |
|---|---|---|---|---|---|
| MTC ROC | 3.10 | 100% | 96.7% | 30 | 0 |
| Non-MTC 97.5tha | 3.70 | 100% | 96.7% | 30 | 0 |
| Non-MTC 97.5 thb | 3.30 | 100% | 97.6% | 42 | 0 |
| MCOa | 2.80 | 100% | 90.0% | 10 | 0 |
| MCOb | 1.90 | 100% | 95.0% | 21 | 0 |
-
+LHR, positive likelihood ratio; −LHR, negative likelihood ratio; MTC, medullary thyroid cancer; ROC, receiver operating characteristic curve; MTC ROC, ROC-derived cutoff obtained in MTC patients; non-MTC 97.5th, 97.5th percentile value obtained in non-MTC patients; MCO, manufacturer-provided cutoff; a malesb females.
Discussion
This retrospective study evaluated the use of Elecsys® CT and PCT as tumor markers in patients with MTC. First of all, no statistically significant difference was found between the two analytes in discriminating active MTC from MTC in complete remission. However, when the ROC curve-optimized cutoff used for the above purpose was adopted to discriminate active MTC from patients with non-MTC thyroid disease false-positive results were observed in five patients, affected by Graves’ disease (n=1) and multinodular goiter (n=4). This is not unexpected as CT levels increase in patients with enlarged thyroid glands; conversely these levels are low in patients with complete remission MTC due to previous surgery and absence of active disease. In addition gender-specific cutoff are suggested to improve accuracy. Accordingly, gender-specific cutoff levels obtained in our patients with non-malignant thyroid disease and those provided by the manufacturer and obtained in healthy subjects were also applied. No differences in sensitivity (100%) were observed using different cutoff levels (ranging from 2.8 to 3.7 pmol/L in males and 1.9–3.3 pmol/L in females, respectively). While manufacturer-provided cutoffs reduced specificity in both males and females those obtained in our patients with non-MTC thyroid diseases performed equally to ROC-derived one in males and slightly increased specificity and +LHR were observed in males and females as well. In summary, as the main results of our study i. both Elecsys® CT and PCT assays were highly sensitive and specific MTC markers; ii. in the few cases with false-positive CT results a negative PCT measurement safely excluded active MTC. Putting our results in current clinical setting and considering that MTC management is centered on CT measurement since decades it seems reasonable, at the very least, to adopt PCT as complementary if not even as a replacement marker in patients with thyroid nodules and positive CT screening as well as in MTC patients with unclear serial CT measurement patterns after thyroidectomy. Furthermore, the present results, again confirming the value of PCT in MTC care, should contribute to long-term follow-up studies in MTC using PCT in order to further facilitate adoption of this novel, more reliable marker in MTC care. Having both assays performed on the same automated platform greatly facilitates the adoption of reflexing or reflective testing strategies avoiding inappropriate, even invasive, procedures to our patients with attached costs.
Receiver operating characteristic (ROC) curve for calcitonin (sensitivity: active MTC, specificity, inactive MTC).
-
Research funding: Elecsys® CT and PCT kits were provided for free by Roche Diagnostics (Rotkreutz, Switzerland) for the purpose of the present study. L.G. is a member of Roche Diagnostics Advisory Board and has received research grants from Roche Diagnostics and speaker honoraria from Roche Diagnostics, BRAHMSGmbH and Sanofi-Genzyme. F.A.V. has received consultancy honoraria from EISAI, JubilantDraximage and Sanofi, speaker honoraria from Sanofi and research support from EISAI. Other authors disclose no conflicts of interest. This study was approved by the Ente Ospedaliero Cantonale Institutional Review Board and the Canton Tessin Ethical Committee, Bellinzona (Switzerland) [references CE 3466 and BASEC 2019-00662, respectively]. The study was performed in accordance with the World Medical Association Declaration of Helsinki regarding ethical conduct of research involving human subjects.
-
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
-
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
References
1. Trimboli, P, Giovanella, L, Crescenzi, A, Romanelli, F, Valabrega, S, Spriano, G, et al. Medullary thyroid cancer diagnosis: an appraisal. Head Neck 2014;36:1216–23. https://doi.org/10.1002/hed.23449.Search in Google Scholar PubMed
2. Bae, YJ, Schaab, M, Kratzsch, J. Calcitonin as biomarker for the medullary thyroid carcinoma. Rec Res Canc Res 2015;2014:117–37. https://doi.org/10.1007/978-3-319-22542-5_5.Search in Google Scholar PubMed
3. Frank-Raue, K, Schott, M, Raue, F. Recommendation for calcitonin screening in nodular goiter. Dtsch Med Wochenschr 2018;143:1065–69. https://doi.org/10.1055/a-0585-8097.Search in Google Scholar PubMed
4. Trimboli, P, Giovanella, L. Serum calcitonin negative medullary thyroid carcinoma: a systematic review of the literature. Clin Chem Lab Med 2015;53:1507–14. https://doi.org/10.1515/cclm-2015-0058.Search in Google Scholar PubMed
5. Austin, LA, Heath, H. Calcitonin: physiology and pathophysiology. N Engl J Med 1981;304:269–78. https://doi.org/10.1056/nejm198101293040505.Search in Google Scholar
6. Cavalier, E, Carlisi, A, Chapelle, JP, Delanaye, P. Analytical quality of calcitonin determination and its effect on the adequacy of screening for medullary carcinoma of the thyroid. Clin Chem 2008;54:929–30. https://doi.org/10.1373/clinchem.2007.100636.Search in Google Scholar PubMed
7. Giovanella, L, Imperiali, M, Ferrari, A, Palumbo, A, Lippa, L, Peretti, A, et al. Thyroid volume influences serum calcitonin levels in a thyroid-healthy population: results of a 3-assay, 519 subjects study. Clin Chem Lab Med 2012;50:895–900. https://doi.org/10.1515/cclm-2011-0920.Search in Google Scholar PubMed
8. Machens, A, Ukkat, J, Hauptmann, S, Dralle, H. Abnormal carcinoembryonic antigen levels and medullary thyroid cancer progression. A multivariate analysis. Arch Surg 2007;142:289–93. https://doi.org/10.1001/archsurg.142.3.289.Search in Google Scholar PubMed
9. van Veelen, W, de Groot, JWB, Acton, DS, Hofstra, RMW, Hoeppener, JWM, Links, TP, et al. Medullary thyroid carcinoma and biomarkers: past, present and future. J Intern Med 2009;266:126–40. https://doi.org/10.1111/j.1365-2796.2009.02106.x.Search in Google Scholar PubMed
10. Wells, SAJr, Asa, SL, Dralle, H, Elisei, R, Evans, DB, Gagel, RF, et al. Revised American Thyroid Association guidelines for the management of medullary thyroid carcinoma. Thyroid 2015;25:567–610. https://doi.org/10.1089/thy.2014.0335.Search in Google Scholar PubMed PubMed Central
11. Bihan, H, Becker, KL, Snider, RH, Nylen, E, Vittaz, L, Lauret, C, et al. Calcitonin precursor levels in human medullary thyroid carcinoma. Thyroid 2003;13:819–22. https://doi.org/10.1089/105072503768499716.Search in Google Scholar PubMed
12. Algeciras-Schimnich, A, Preissner, CM, Theobald, JP, Finseth, MS, Grebe, SK. Procalcitonin: a marker for the diagnosis and follow-up of patients with medullary thyroid carcinoma. J Clin Endocrinol Metab 2009;94:861–8. https://doi.org/10.1210/jc.2008-1862.Search in Google Scholar PubMed PubMed Central
13. Giovanella, L, Verburg, FA, Imperiali, M, Valabrega, S, Trimboli, P, Ceriani, L. Comparison of serum calcitonin and procalcitonin in detecting medullary thyroid carcinoma among patients with thyroid nodules. Clin Chem Lab Med 2013;51:1477–81. https://doi.org/10.1515/cclm-2012-0610.Search in Google Scholar PubMed
14. Giovanella, L, Imperiali, M, Piccardo, A, Taborelli, M, Verburg, FA, D’Aurizio, F, et al. Procalcitonin measurement to screen medullary thyroid carcinoma: a prospective evaluation in a series of 2705 patients with thyroid nodules. Eur J Clin Invest 2018;48: e12934. https://doi.org/10.1111/eci.12934.Search in Google Scholar PubMed PubMed Central
15. Machens, A, Lorenz, K, Dralle, H. Utility of serum procalcitonin for screening and risk stratification of medullary thyroid cancer. J Clin Endocrinol Metab 2014;99:2986–94. https://doi.org/10.1210/jc.2014-1278.Search in Google Scholar PubMed
16. Trimboli, P, Giovanella, L. Procalcitonin as marker of recurrent medullary thyroid carcinoma: a systematic review and meta-analysis. Endocrinol Metab 2018;33:204–10. https://doi.org/10.3803/enm.2018.33.2.204.Search in Google Scholar PubMed PubMed Central
17. Giovanella, L. Serum procalcitonin and calcitonin normal values before and after calcium gluconate infusion. Exp Clin Endocrinol Diabetes 2012;120:169–70. https://doi.org/10.1055/s-0031-1301290.Search in Google Scholar PubMed
18. Kratzsch, J, Petzold, A, Raue, F, Reinhardt, W, Bröcker-Preuss, M, Görges, R, et al. Basal and stimulated calcitonin and procalcitonin by various assays in patients with and without medullary thyroid cancer. Clin Chem 2011;57:467–74. https://doi.org/10.1373/clinchem.2010.151688.Search in Google Scholar PubMed
19. Giovanella, L, Giordani, I, Imperiali, M, Orlandi, F, Trimboli, P. Measuring procalcitonin to overcome heterophilic-antibody-induced spurious hypercalcitoninemia. Clin Chem Lab Med 2018;56:191–93. https://doi.org/10.1515/cclm-2017-0993.Search in Google Scholar PubMed
20. Lim, SK, Guéchot, J, Vaubourdolle, M. Negative predictive value of procalcitonin in medullary thyroid carcinoma. Ann Biol Clin 2016;74:213–8. https://doi.org/10.1684/abc.2015.1115.Search in Google Scholar PubMed
21. Schiettecatte, J, Strasser, O, Anckaert, H, Smitz, J. Performance evaluation of an automated electrochemiluminescent calcitonin (CT) immunoassay in diagnosis of medullary thyroid carcinoma. Clin Biochem 2016;49:929–31. https://doi.org/10.1016/j.clinbiochem.2016.05.006.Search in Google Scholar PubMed
22. Kahaly, GJ, Algeciras-Schimnich, A, Davis, TE, Diana, T, Feldkamp, J, Karger, S, et al. United States and European Multicenter prospective study for the analytical performance and clinical validation of a novel sensitive fully automated immunoassay for calcitonin. Clin Chem 2017;63:1489–96. https://doi.org/10.1373/clinchem.2016.270009.Search in Google Scholar PubMed
23. Conover, WJ. Practical nonparametric statistics, Hoboken, USA: 3rd ed. John Wiley & Sons; 1999.Search in Google Scholar
© 2020 Luca Giovanella et al., published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.
