Measurement of plasma CTN is part of the routine evaluation of patients with thyroid nodules. Up to 3 % of patients with thyroid nodules have pathological CTN concentrations, and about 0.6 % have an MTC (Costante et al. 2007). The prevalence of MTC was nearly 100 % when basal CTN levels were >100 pg/ml and pentagastrin-stimulated CTN levels were >1000 pg/ml as measured with specific and sensitive ICMAs (Scheuba et al. 2009).

When using basal CTN as a screening tool, one must take into account that CTN can also be slightly elevated in patients with various clinical conditions who present with no clinical evidence of MTC. These conditions include normal elevation during early childhood and pregnancy and pathological elevation in patients with chronic renal failure or autoimmune thyroiditis, in those who take proton pump inhibitors or who smoke, and in patients with small-cell and large-cell lung cancers, prostate cancer, mastocytosis, and various enteric and pulmonary neuroendocrine tumors (Whang et al. 1998; Schuetz et al. 2006; Machens et al. 2000; Toledo et al. 2009). Notably, the serum CTN levels in patients with various non-MTC malignancies do not increase in response to calcium or pentagastrin stimulation, and compared to MTC, the tumors usually produce less CTN per gram of tissue. The presence of heterophilic antibodies can also interfere with ICMA testing, causing falsely elevated CTN values (Karanikas et al. 2004). With the improved sensitivity of new specific two-site assays, basal and stimulated CTN tests have similar accuracy, the number of false-positive elevated CTN results has decreased, and the relevance of stimulated CTN is reduced (Mian et al. 2014). Notably, many increases in CTN levels are unrelated to MTC. In particular in the 10–30 pg/ml range for basal CTN, such increases are commonly caused by C-cell hyperplasia and are not related to MTC.

The determination of the levels of CTN mRNA or CTN gene-related peptide mRNA as extracted from peripheral blood represents a diagnostic tool that is an alternative to basal and stimulated CTN measurement. These determinations have a higher positive predictive value than do basal or stimulated CTN levels (Camacho et al. 2013). Careful evaluation of CTN in nodular thyroid disease allows early diagnosis of MTC, with a reduction in primary tumor diameter at first diagnosis (Machens and Dralle 2010), early curative surgery, and a reduction in the significant mortality associated with this malignant tumor (Costante et al. 2007; Niccoli et al. 1997; Ozgen et al. 1999; Hahm et al. 2001; Elisei et al. 2004). Accordingly, measurement of plasma CTN in patients with thyroid nodules has been advocated as a routine procedure by some European consensus groups (Pacini et al. 2006). Compared with fine-needle aspiration, the sensitivity of the CTN measurement for preoperative diagnosis of MTC is higher (approximately 100 % sensitivity and 95 % specificity) (Bugalho et al. 2005; Hasselgren et al. 2010; Papi et al. 2006). There are a number of other proteins, including carcinoembryonic antigen (CEA), PDN-21 (katacalcin) (Blind et al. 1992), chromogranin A (Blind et al. 1992), neuron-specific enolase (Grauer et al. 1987), somatostatin (Grauer et al. 1995), and ACTH, that are sometimes produced and secreted by MTC and that could serve in some cases as tumor markers for diagnosis and follow-up.

Neoplastic C cells also produce CEA, which can be used as a prognostic marker during the follow-up of individuals with MTC (Machens et al. 2007; Busnardo et al. 1984; Wells et al. 1978). CEA is not a specific biomarker for MTC and is not useful in the early diagnosis of MTC. However, serum CEA levels might be used for the risk stratification of individuals with known MTC. CEA levels >30 ng/ml are suggestive of lymph node metastases in the ipsilateral central and lateral neck compartments, while levels >100 ng/ml correlate with contralateral lymph node metastases and distant metastases. CEA values >30 ng/ml correlate with low cure rates (Machens et al. 2007). Some patients with progressive disease demonstrate increasing serum CEA levels that are associated with stable or declining serum CTN levels. This is considered an indication of a poorly differentiated MTC and is supported by CEA and CTN immunohistochemistry findings (Mendelsohn et al. 1984).

MTC without CTN and CEA secretion has been reported, but it is rare. The prevalence of nonsecretory MTC in a series of 839 patients was low (0.83 %) (Frank-Raue et al. 2013). Nonsecretory MTC cannot be detected by serum CTN screening as an integral part of the diagnostic evaluation of thyroid nodules. However, it is detected more often postoperatively at advanced tumor stages because of its characteristic histology and can be confirmed by positive immunohistochemistry for CTN, CEA, and chromogranin A. In some cases, CTN or CEA levels may rise during follow-up but to an extent that is entirely disproportionate relative to the tumor mass (Bockhorn et al. 2004; Dora et al. 2008). Nonsecretory MTC is a rare disease with a poor prognosis, and sometimes it shows advanced dedifferentiation. The aggressive biological behavior of the tumor is characterized by poorly differentiated histology, a high Ki-67 proliferation index, and a high proportion of RET M918T mutated cells. Nonsecretory MTC is markedly heterogeneous in its histological and immunohistological appearance and in its clinical course and prognosis (Frank-Raue et al. 2013).

The gene responsible for MEN2 was discovered in 1985 (Takahashi et al. 1985) and localized to centromeric chromosome 10 (10q11.2) by genetic linkage analysis in 1987 (Mathew et al. 1987). Activating germline point mutations in the RET proto-oncogene were identified in (Donis-Keller et al. 1993). Analysis of the RET gene in families with MEN2 revealed that only affected family members had germline missense mutations in 8 closely located exons.

The RET gene has 21 exons and encodes a single-pass transmembrane receptor tyrosine kinase that appears to transduce growth and differentiation signals in several developing tissues, including those derived from the neural crest, the branchial arches, and the urogenital system (Pachnis et al. 1993). RET is expressed in cells such as C cells, which are the precursors of MTC, as well as in pheochromocytomas. The RET gene codes for a receptor with three domains: a large extracellular cysteine-rich domain, which is thought to be involved in ligand binding; a short transmembrane domain; and a cytoplasmic tyrosine kinase (TK1 and TK2) domain that is activated upon ligand-induced dimerization. Hereditary MTC is caused by autosomal dominant gain-of-function mutations in the RET proto-oncogene that affect exons 5, 8, 10, 11, and 13–16. Mutation of the extracellular cysteine at exon 11, codon 634, causes ligand-independent dimerization of receptor molecules, enhanced phosphorylation of intracellular substrates, and cell transformation. Mutation of the intracellular tyrosine kinase (codon 918) has no effect on receptor dimerization but causes constitutive activation of intracellular signaling pathways and also results in cellular transformation (de Groot et al. 2006). There is a significant age-related progression from C-cell hyperplasia to MTC that correlates with the transforming capacity of the different RET mutations.

Mutation analysis has identified over 100 different missense mutations, duplications, insertions, and deletions involving RET that are associated with the development of MEN2 (Raue et al. 2012; Margraf et al. 2009). Although there is some overlap between RET mutations and the resulting clinical subtype of MEN2, 85 % of patients with MEN2A have a mutation in codon 634 (exon 11). Mutations in codons 609, 611, 618, and 620 account for an additional 10–15 % of cases. Pheochromocytoma is associated with codon 634 and codon 918 mutations in approximately 50 % of patients, and in 20 % of patients, pheochromocytoma is associated with mutations in exon 10 (codons 609, 611, 618, 620) or exon 15 (codons 791, 804) (Quayle et al. 2007). Hyperparathyroidism in MEN2A is most commonly associated with codon 634 mutations and in particular with the C634R mutation. In FMTC, germline mutations are distributed throughout the RET gene, with many in exon 13 (codons 768, 790, 791), exon 14 (codons 804, 844), and, rarely, exon 10 (codons 618, 620). More than 95 % of MEN2B patients have mutations in codon 918 (exon 16), but mutations are rarely identified in codon 883 in exon 15. Approximately 75 % of patients with MEN2B have mutations that occur sporadically as de novo mutations, almost always from the paternal allele (Gimm et al. 1997; Eng et al. 1994; Schuffenecker et al. 1997). The association between disease phenotype and RET mutation genotype has important implications for the clinical management of MEN2 patients and their families. In particular, there is a correlation between specific germline RET mutations and the age of onset with the aggressiveness of MTC development and the presence of nodal metastases (Task et al. 2009).

Approximately 23–60 % of sporadic MTCs have an acquired RET codon 918 somatic mutation in tumor tissue that is termed somatic mutation T918M. This mutation is identical to the germline mutation found in MEN2B. Patients with sporadic MTC with T918M have more aggressive tumor growth and a poor prognosis (Schilling et al. 2001; Romei et al. 1996; Marsh et al. 1996; Moura et al. 2009; Elisei et al. 2008; Eng et al. 1996). The prevalence of somatic M918T RET mutations varies depending on tumor size: small tumors (<1 cm) rarely have the mutation (11.3 %), while T918M is found in 58.8 % of patients with tumors >3 cm (Romei et al. 2012). The discovery that mutation-positive and mutation-negative regions can coexist in the same sporadic MTC tumor suggests that such genetically heterogeneous MTCs may not be clonally derived from a single initiating tumor cell with a RET mutation. RET does not seem to be the early initiator of tumor growth in sporadic MTC; rather, RET is activated later in oncogenesis as a driver of tumor growth, and other genes must play a significant role in MTC onset.

Mutations in codons 618, 630, 634, 768, 804, and 883, as well as partial deletion of the RET gene, have been identified in a few tumors (Moura et al. 2011). It was recently discovered that 18–80 % of sporadic MTCs lacking somatic RET mutations have somatic mutations in KRAS, HRAS, or, rarely, in NRAS (Moura et al. 2011; Boichard et al. 2012; Ciampi et al. 2013). No other common genetic mutation has been detected in subsequent exome sequencing studies of MTCs (Agrawal et al. 2013).

The most common clinical presentation of sporadic MTC is an indolent and usually solitary single nodule or thyroid mass with or without associated lymphadenopathy in the neck. Typically, the nodule or mass is found incidentally during routine examination or is an incidental finding during an imaging examination of the neck. Diagnosis of the sporadic form of MTC is usually established late in life (approximately during the fifth or sixth decade), but age of onset shows a wide range. The presentation of sporadic MTC does not differ from that observed for papillary or follicular thyroid carcinoma. The patient may present with no symptoms or with local symptoms related solely to the neck mass, and presentation may include findings such as dysphagia, dysphasia, and dyspnea. Metastases to cervical and mediastinal lymph nodes are found in two-thirds of the patients at the time of initial clinical presentation (Moley et al. 1999; Scollo et al. 2003; Machens et al. 2002; Scheuba et al. 2007). In contrast, when MTC was diagnosed by CTN screening, the diameters of the tumors were <10 mm, and only 11 % had cervical nodal involvement (Scheuba et al. 2007). Distant metastases to the lung, liver, and bone occur late in the course of the disease. At presentation, about 5 % of patients have distant metastatic disease (Pacini et al. 2010). Patients sometimes present with painful bone metastases. Diarrhea is the most prominent of the hormone-mediated clinical features of MTC, while flushing can be present but is rare; notably, diarrhea is often seen in patients with advanced disease. In addition, occasionally tumors secrete ACTH ectopically, causing Cushing’s syndrome (Barbosa et al. 2005). An MEN2 index patient might present with Hirschsprung’s disease or with a cutaneous lichen in the back region. In a young patient, facial features that include a centrofacial ganglioneuroma of the lip or tongue may suggest MEN2B.

A patient with a palpable anterior neck mass and an associated endocrine neoplasm (pheochromocytoma, hyperparathyroidism) and/or a suggestive family history of thyroid tumors, early sudden death (suggestive of pheochromocytoma), or nephrolithiasis (primary hyperparathyroidism) in first-degree relatives might be an index patient with MEN2.

In general, a thyroid nodule that is identified by physical examination is subsequently evaluated by ultrasonography and radioisotopic scanning (Fig. 1). MTC shows hypoechogenic regions, sometimes with calcifications; however, there are no specific ultrasound features that are pathognomonic for thyroid cancer. Some nodules that have a higher risk of malignancy show ultrasound characteristics that include hypoechogenicity, microcalcifications, irregular margins, predominantly central vascularization, and the presence of enlarged neck lymph nodes. Nevertheless, there are no differences in the echogenicity or in the presence or type of calcifications between MTC and papillary thyroid cancer (Lee et al. 2010; Kim et al. 2009; Saller et al. 2002). One important limitation of ultrasound is that it is operator-dependent; therefore, the results vary according to the operator’s experience.