MEN2 (OMIM 171400) has an estimated annual incidence of 5 per 10,000,000 persons and a prevalence of 1 in 30,000. MEN2 follows an autosomal dominant mode of inheritance. MEN2 patients can develop MTC (medullary thyroid carcinoma), PCC, and/or PHPT (primary hyperparathyroidism), the latter resulting from hyperplasia or from parathyroid adenomas. The syndrome is classified into three subtypes, MEN2A, MEN2B, and familial MTC (FMTC), each defined according to the combination of pathologies developed by the individuals affected (Table 2.1). MEN2A patients may develop all three pathologies. They are also more likely to develop “cutaneous lichen amyloidosis,” a pruritic skin lesion in the upper area of the back caused by the uncontrolled deposition of amyloid protein between the dermis and epidermis. In addition, these patients may occasionally develop Hirschsprung disease (HSCR). Patients are classified as MEN2B if they develop, in the absence of parathyroid disease, MTC; PCC; multiple mucocutaneous neuromas involving the lips, tongue, and eyelids; corneal nerve myelination; intestinal ganglioneuromas (hyperganglionic megacolon); or marfanoid habitus, including skeletal deformities and hypermobility of joints. Finally, families in which affected members have developed only MTC or C-cell hyperplasia (CCH) are considered to have the third subtype, FMTC, but only if more than ten members have MTC. A less conservative definition has been proposed by [16], based on the presence in at least four family members of MTC without other manifestations of MEN2A. Defining and distinguishing FMTC from MEN2A is challenging, and an exhaustive clinical follow-up of these families is required to rule out the presence of other tumors characteristic of MEN2, especially in older family members.

In the early 1990s, activating mutations in the rearranged during transfection (RET) proto-oncogene were identified as the genetic basis for MEN2A, MEN2B, and FMTC [17–20]. Since then, germline RET mutations have been identified in 98 % of patients with MEN2A, 95 % of patients with MEN2B, and 88 % of patients with FMTC [21].

RET is located on chromosomal band 10q11.2 and encodes a tyrosine kinase receptor that is mainly expressed in cells derived from the neural crest (C cells, parafollicular thyroid cells, and adrenal medulla cells, among others) and in urogenital system precursor cells [22]. The genetic testing of RET is relatively simple, since the mutations associated with the development of MEN2 are mainly located on exons 10, 11 and 13–16. Additional, less frequent mutations on exon 5 and 8 have been also reported in MEN2 patients. Sequencing of the entire coding region of RET is recommended only for those who meet clinical criteria for MEN2 but in whom initial sequencing of selected exons gives negative results [23] or if there is a discrepancy between the MEN2 phenotype and the expected genotype [24]. If a RET variant is detected at a noncanonical position, it is important to consult different databases for evidences that the variant may be a polymorphism with no clinical significance [25]. A sequence change in RET is considered to be a causative MEN2 mutation if it segregates with the clinical expression of disease within a family including at least two affected individuals having the MEN2 phenotype.

The established genotype–phenotype relationships for MEN2 syndrome are based on the classification of individual mutations according to their transforming ability, and therefore the expected aggressiveness with which the disease they cause will develop [24]. The impact of RET mutation testing on the management of MEN2 patients is without doubt one of the most robust examples of the utility of genetic diagnosis in personalizing clinical follow-up [24].

Approximately 50 % of MEN2 patients develop PCC in their lifetime, and the mean age at diagnosis is 35 years. A PCC is the first manifestation of MEN2 in only 12–15 % of cases, and so RET explains relatively few cases of non-syndromic disease (around 5 %), compared to other syndromes [26, 27]. RET mutations are very rarely found in patients diagnosed with PCC before age 20 [5–7]. Thus, RET is not a priority in the genetic testing of pediatric patients, although it should still be included in genetic diagnosis algorithms [5]. PCCs developed by RET mutation carriers are bilateral in 50–80 % of patients, they tend to be epinephrine-secreting tumors, and very few (no more than 5 %) are malignant.

It is recommended that screening for PCC begins between ages 5 and 16 for carriers of highest-, high-, and moderate-risk mutations [24, 27].

VHL (OMIM 193300), with an incidence of 1 in 36,000 live births, is a dominantly inherited familial cancer syndrome caused by germline mutations in the VHL tumor suppressor gene [28, 29]. This gene encodes three gene products: a protein comprising 213 amino acids and two shorter isoforms, one produced by alternative splicing (excluding exon 2) and the other by alternative initiation. While the protein (pVHL) is involved in multiple processes, its best characterized role is the regulation of the proteasomal degradation of hypoxia-inducible factors (HIFs) [30]. Under normal oxygen tension, the α subunits of HIF (HIF-1α, HIF-2α, and HIF-3α) are hydroxylated by the dioxygenases PHD (PHD1, PHD2, and PHD3). The hydroxylated HIF-α are then targeted by pVHL for proteasomal degradation [31, 32]. Under hypoxia conditions, HIF-α is stabilized and binds to the HIF-β subunit to form an active transcription factor that regulates expression of a large repertory of genes involved in angiogenesis, cell survival, erythropoiesis, and tumor progression [30]. This explains the highly vascularized nature of the tumors associated with VHL syndrome [33].

The penetrance of causal mutations is age dependent, and the disease demonstrates marked phenotypic variability. Patients with VHL are at a higher risk of developing hemangioblastomas (HBs) of the retina and central nervous system (CNS), PCC and/or PGL, clear cell renal cell carcinoma (ccRCC), renal and pancreatic cysts (serous cystadenoma), neuroendocrine pancreas tumors, endolymphatic sac tumors, pancreatic serous cystadenomas, and papillary cystadenomas of the epididymis in men and of the broad ligament in women (Table 2.1) [27, 33, 34]. After identifying VHL in 1993, the phenotype associated with VHL gene mutations was expanded to include VHL disease, dominantly inherited familial PCC, and, in cases with particular mutations, autosomal recessive congenital polycythemia (also known as familial erythrocytosis-2; MIM# 263400) [35, 36]. It has been suggested that VHL accounts for approximately a third of patients with a CNS HBs, >50 % of patients with a retinal angioma, 1 % of patients with RCC, 50 % of patients with apparently isolated familial PCC, and 11 % of patients with an apparently sporadic PCC [37].

The disease is classified into four subtypes (1, 2A–2C) based on the clinical phenotype. VHL type 1 families have a relatively low risk of developing PCC, but may present with any of the other tumors associated with the disease. VHL type 2 is subdivided into three categories corresponding to a low (2A) or high risk (2B) of developing ccRCC or to a higher risk of PCC and PGLs as the only clinical sign of the disease (2C) (Table 2.2).

Systematic characterization of germline VHL mutations has led to the identification of genotype–phenotype correlations such that germline mutations causing amino acid changes on the surface of pVHL are associated with a higher risk of PCC [34]. VHL germline deletions, mutations predicted to cause a truncated protein, and missense mutations that disrupt the structural integrity of the VHL protein are associated with a lower risk of PCC (and therefore associated with type 1 VHL syndrome). However, phenotypes related to VHL gene deletions appear to be influenced by the retention of some genes surrounding VHL. In particular, VHL deletions associated with the loss of the actin regulator HSPC300 gene (also known as BRICK1) are associated with protection against ccRCC [38–40]. These clinical findings led McNeill et al. (2009) to suggest that the subclassification of VHL syndrome should take into account not only clinical phenotype but also VHL mutation data [40] (Table 2.2).

Neurofibromatosis type 1 (NF1), formerly known as von Recklinghausen disease, is a common hereditary disease with an incidence of 1 per 2,500–3,300 newborns that primarily involves the skin and nervous system. The condition is usually diagnosed in early childhood, when cutaneous manifestations are apparent. It is characterized by the appearance of multiple neurofibromas, cafe au lait spots, freckling in the armpits and groin, iris hamartomas (Lisch nodules), bone lesions such as scoliosis, sphenoid dysplasia or pseudoarthrosis, macrocephaly, learning disorders, cognitive deficits, predisposition to optic and CNS glioma, and leukemia [27, 54]. Other malignancies occur less frequently in patients with NF1, including PCC, rhabdomyosarcoma, leukemia, and brain tumors other than optic gliomas [55]. Some features of neurofibromatosis 1 are present at birth, and others are age-related manifestations, which means that periodic monitoring is required to address ongoing health and developmental problems and to minimize the risk of serious medical complications [56].

The gene responsible for NF1, NF1 (17q11.2), encodes the protein neurofibromin, which is expressed primarily in the nervous system and has the role of suppressing cell proliferation by inactivating RAS proteins. Loss-of-function mutations in NF1 lead to the activation of RAS and the PI3K/AKT/mTOR pathway, which depends on RAS [54].

The detection of mutations in NF1 by DNA analysis has proven to be challenging because of the gene’s large size (it has 58 exons), the lack of mutation hot spots, and the existence of pseudogenes. NF1 mutations are predominantly truncating and often accompanied by loss of the wild-type allele in the tumor. Although molecular technology is now available to detect most mutations in NF1 [57], it is typically not required because in 95 % of cases a diagnosis of NF1 can be made by age 11 years on the basis of clinical findings alone. NF1 has one of the highest rates of spontaneous mutation of any gene in the human genome. This in part explains why between 30 and 50 % of patients have de novo mutations, which if they occur post-zygotically, can give rise to mosaic phenotypes [58]. Events of germline mosaicism are very rare in this condition [59].