MEN2 is an autosomal dominant disorder with an estimated prevalence of 1 to 10 per 100,000 in the general population. It can be subdivided into two independent syndromes: MEN 2A (Sipple syndrome) and MEN 2B. Manifestations of MEN 2A include MCT, pheochromocytoma, and hyperparathyroidism. MEN 2B includes MCT, pheochromocytoma, and a number of somatic manifestations (Table 22–2; Figure 22–4), but hyperparathyroidism is rare. Penetrance of MEN2 is greater than 80% in individuals harboring the defective gene.
Medullary carcinoma of the thyroid is the most common manifestation of MEN2 and often represents the first clinical presentation in individuals with multiorgan involvement. It also dominates the clinical course of patients affected with the disease. Eighty percent to 100% of individuals at risk develop MCT at some point during their lifetime. The classic thyroid lesion of MEN2 is hyperplasia of the calcitonin-producing parafollicular cells, which typically serves as the precursor of MCT. These tumors in MEN2 patients tend to be multicentric and concentrated in the upper third of the thyroid gland, reflecting the normal distribution of parafollicular cells.
As much as one-fourth of all MCT is genetic in origin. Roughly 30% of the heritable fraction is attributable to MEN 2A; 65% occurs as an isolated entity (isolated familial MCT), and 5% is found in MEN 2B kindreds. Isolated familial MCT is defined as MCT presenting as the only clinical manifestation in more than 10 carriers in a kindred and multiple carriers or affected members over the age of 50 with adequate history and laboratory support to rule out the presence of pheochromocytoma or hyperparathyroidism. The disease tends to behave more aggressively in MEN 2B than with either MEN 2A or familial MCT, with earlier presentation (often before age 5) and more rapid progression.
Biochemical diagnosis depends heavily on the calcitonin–producing properties of the hyperplastic parafollicular cells or MCT. These lesions respond to pentagastrin or calcium infusions with significant increments in plasma calcitonin levels (see Chapter 8). Occasionally, immunohistochemical staining of poorly differentiated thyroid tumors for calcitonin reveals the identity of the malignancy. The presence of extracellular amyloid is also one of the identifying features of these tumors. This material reacts with anticalcitonin antisera, suggesting that it includes aggregated hormone released from neighboring tumor cells. MCT spreads initially within the thyroid bed and to regional lymph nodes. Distant metastases to liver, lung, and bone occur late in the course of the disease.
Pheochromocytomas develop in approximately 50% of individuals harboring the Men2 gene. They are usually located in the adrenal bed, often are bilateral, and are rarely malignant. Diagnosis is based on standard clinical criteria (eg, hypertension, presence of headaches, palpitations, diaphoresis), elevations in plasma or urine catecholamines or catecholamine metabolites (eg, urinary or plasma metanephrine or normetanephrine), and demonstration of an adrenal mass on conventional abdominal imaging. As noted above for MCT, pheochromocytomas in MEN2 are preceded by a hyperplastic phase (adrenal medullary hyperplasia), although, unlike parafollicular cell hyperplasia, the adrenal precursor lesion can be difficult to detect with conventional biochemical testing.
As in MEN1, hyperparathyroidism in MEN2 is due to hyperplasia of the parathyroid glands. It is seen in about 25% of patients harboring the MEN2A gene and is reported, but rarely seen, as part of MEN 2B. The disease is usually less aggressive than its counterpart in MEN1 and approximates more closely the behavior of sporadic disease. It responds well to surgical management.
A number of other phenotypic features are associated with the MEN2 syndromes. Cutaneous lichen amyloidosis is a pruritic erythematous skin lesion that is seen coincident with or often preceding the development of MCT in MEN2. Amyloid in these lesions is composed of keratin rather than calcitonin, as seen in MCT. Cutaneous lichen amyloidosis is thought to result from an abnormality in cutaneous innervation that leads to hyperesthesia and pruritis. Repeated scratching of the pruritic area results in epidermal thickening and increased pigmentation. It has been noted more frequently in association with specific mutations of the MEN2 gene (specifically Cys634 to Tyr634). In a second variant, MEN 2A or familial MCT is associated with Hirschsprung disease (congenital megacolon). This is most frequently found with RET gene mutations involving Cys609, Cys615, and Cys620(see below). The intestinal ganglioneuromatosis, mucosal neuromas, marfanoid habitus, and medullated corneal nerves seen in MEN 2B appear to be related to the underlying genetic defect The intestinal lesions can disrupt gut motility, resulting in periods of severe constipation or diarrhea.
The pathogenesis of the MEN2 syndromes has been worked out in elegant detail. Traditional genetic linkage studies localized the defective genes in MEN 2A, MEN 2B, and familial MCT to the pericentromeric region of chromosome 10. Subsequent refinement in these analyses indicated that the defective gene was either closely linked to or identical with the ret proto-oncogene. RET is a single transmembrane domain, tyrosine kinase-linked protein that forms part of the receptor for the glial cell line—derived neurotrophic factor (GDNF) (see Figure 22–5). This receptor, GDNFR (glial cell line–derived neurotrophic factor receptor), is a glycosyl phosphatidyl inositol-linked cell-surface protein. Additional ligands, neurturin, persephin, and artemin have been shown to associate with and activate RET through their cognate receptors GDNFR1-4. As depicted schematically in Figure 22–6, its most striking structural feature is a series of cysteine residues clustered just outside the membrane-spanning segment. These cysteine residues are thought to exert a tonic inhibitory control on RET activity in the normal cell. RET also has a cadherin-like domain in that portion of the molecule projecting into the extracellular space and a tyrosine kinase-like domain in the intracellular portion of the molecule. RET is expressed endogenously in a variety of cells of neural crest origin, and it appears to play an important role in development. Knockout of the RET gene locus in mice results in the absence of myenteric ganglia in the submucosa of the small and large intestine and a variety of genitourinary anomalies, implying an important role in renal development.
Structure of GDNFR-α/RET receptor complex. GDNF identifies the ligand, glial cell line–derived neurotrophic factor, and GDNFR-α the GDNF receptor (TKD, tyrosine kinase domain).
Structural schematic of wild-type RET, including amino acids that have been shown to be mutated in different disease states. MEN 2A is associated with mutations of Cys609, Cys611, Cys618, Cys620, Cys630, Cys634, and, rarely, Leu790 and Tyr791. Familial MCT of the thyroid is frequently associated with mutations at the same Cys residues with the exception of Cys634, as well as mutations at Glu768, Leu790, Tyr791, Val804, and Ser891. MEN 2B mutations typically involve Met918 or, more rarely, Ala883, Ser922, or Val804/Ser904. Hirschsprung disease is associated with multiple mutations or deletions extending over the full length of the RET molecule.
Characterization of the RET gene in patients with MEN 2A demonstrated a number of mutations in affected kindred members that were not present in their normal counterparts. The mutations were clustered in the cysteines located in RET's extracellular, juxtamembrane domain (see Figure 22–6). Structurally, these cysteines are encoded by nucleotides in exons 10 and 11 of the RET gene. A number of these mutations were simple missense mutations, whereas others involved deletion or insertion of small segments of DNA, but in each case one of the aforementioned cysteines proved to be involved. Selective mutation of one of these six cysteines has now been shown to account for more than 97% of all RET mutations associated with MEN 2A. The most frequently mutated residue is Cys634. This amino acid is mutated to arginine (Arg), phenylalanine (Phe), serine (Ser), glycine (Gly), tyrosine (Tyr), or tryptophan (Trp) in approximately 84% of affected MEN 2A kindreds. It has been suggested that mutation of Cys634 to Arg634 is associated with the phenotypic expression of hyperparathyroidism, whereas mutation of Cys634 to any of the amino acids indicated above is linked to pheochromocytoma. It should be noted that the Cys-to-Arg mutation at position 634 is also the most common mutation at this position, accounting for about 64% of all codon changes at this location. Interestingly, a rare but unique mutation involving a four-amino-acid insertion between Cys634 and Arg635 has been described that results in MCT and a high incidence of hyperparathyroidism but not pheochromocytoma.
Ironically, patients with familial MCT display many of the same cysteine mutations identified in MEN 2A. This implies the existence of independent modulatory genomic factors that restrict the effects of the RET mutation to the parafollicular cells of the thyroid in familial MCT. The exception is the mutation of Cys634 to Arg634, which is almost uniformly associated with MEN 2A. Several additional mutations have been identified that may be more specific for familial MCT (Figure 22–6). The observation that mutations of Cys634 have higher transforming potential than mutations at these other residues has led to the suggestion that MEN 2A represents the more severe phenotype (vs familial MCT) along a spectrum of disease resulting from RET activation. In fact, the noncysteine mutations, which are more common in familial MCT (as high as 60% of familial MCT in one series), are associated with delayed onset of parafollicular C cell disease, but other clinical features (tumor size, bilaterality, the presence of nodal metastases) were not different in patients harboring cysteine versus noncysteine mutations.
Interestingly, patients with MEN 2B do not harbor mutations in the cysteine residues affected in MEN 2A and familial MCT. Instead, the majority possess a single-point mutation involving conversion of Met918 to Thr918. This mutation is found in more than 95% of cases of MEN 2B. A minority of patients have been shown to harbor an independent mutation of Ala883 to Phe883, Ser922 to Tyr922, or a mono-allelic combination of Val804 to Met804 plus Ser904 to Cys904. MEN 2B arises from spontaneous de novo mutations in as many as 50% of affected individuals. For unknown reasons, these mutations are found almost exclusively on the paternal allele.
A number of the germline mutations identified in MEN 2A and familial MCT and the Met918 mutation in MEN 2B have been demonstrated as somatic mutations in sporadic MCT (∼30%-40%) and pheochromocytoma (<10%). Sporadic parathyroid disease due to these mutations is not known to occur. The presence of the Met918 mutation, in particular, is associated with a less favorable clinical outcome in sporadic MCT.
Independent studies have shown a close linkage between the RET gene locus and Hirschsprung disease, a disorder characterized by failure of myenteric ganglia to develop normally in the hindgut of affected individuals. This leads to impaired gut motility and, in severe cases, megacolon (a phenotype similar to that reported for the RET knockout mice). Examination of RET coding sequence in Hirschsprung kindreds revealed a variety of mutations in both the intracellular and extracellular domains, some of which (eg, deletions) would be predicted to eliminate normal expression of the RET gene. This, together with findings from the RET, GDNF, and GDNFRα-1 knockout mice, suggests that Hirschsprung disease represents the null phenotype for the RET locus. Interestingly, several patients have been described who possess features of both MEN2 and Hirschsprung disease. The RET mutations in these cases have involved conversion of Cys609, Cys618, or Cys620 to Arg. These mutations, while promoting dimerization and increasing tyrosine kinase activity in the RET protein (see below), also appear to have difficulty trafficking to and accumulating in the plasma membrane. It is conceivable that predominance of one or the other of these mechanisms in different cell types could result in a phenotype characterized by both activation (eg, MEN2) and suppression (eg, Hirschsprung disease) of RET activity in the same individual.
By inference, the defect in RET function in MEN2 or familial MCT arises from increased or altered activity of the RET tyrosine kinase. In the case of RETMEN2A, the increase in activity appears to arise from interference with the tonic inhibition of RET tyrosine kinase activity by the clustered cysteine residues in the extracellular domain. This leads to increased dimer formation, autophosphorylation, and tyrosine kinase activity in the mutant RET molecules. In the case of RETMEN2B, there appears to be a change in substrate specificity of the tyrosine kinase that contributes to the phenotype. The activity of RETMEN2B—rather than being restricted to conventional RET substrates (RET substrates are similar to those recognized by the epidermal growth factor receptor)—is capable of phosphorylating substrates normally recognized by members of the Src and Abl families of cytoplasmic tyrosine kinases, signaling pathways which are closely identified with the regulation of cell growth. Thus, it appears that RETMEN2B has acquired the capacity for activation of a potent mitogenic pathway in the expressing endocrine cells merely by altering its selection of substrates for phosphorylation. RETMEN2B also potentiates phosphorylation of Tyr1062 more effectively than RETMEN2A. This tyrosine serves as a docking site for multiple effector proteins, including Shc and phosphatidyl inositol 3′-kinase, implying that RETMEN2B may be more effective in triggering downstream signaling pathways.
Treatment of heritable MCT should include total thyroidectomy with at least central lymph node dissection. Given the multicentric nature of the disease, subtotal thyroidectomy predictably results in recurrent disease. Basal or stimulated calcitonin levels are used in the postoperative setting to evaluate the presence of residual disease. The precise timing of surgery in patients with subclinical disease (ie, positive by genetic testing but without clinical or laboratory abnormalities) is controversial (see later), but most clinicians would agree that in kindreds with MEN 2B or clinically aggressive MCT, patients should undergo surgery as soon as the genetic defect is demonstrated. Typically, this is before age 6 months in MEN 2B (eg, mutation at Ala883 or Met918) and before age 5 years in MEN 2A (eg, mutation at Cys611, Cys 618, Cys620, or Cys 634). Foci of microscopic MCT are common, and metastatic disease has been described in the first year of life in patients with MEN 2B. The presence of a thyroid nodule or lymph node metastases at the time of diagnosis is strongly associated with persistent or recurrent disease. Mutations associated with familial MCT (eg, Cys609, Glu768, Leu790, Tyr791, Val804, or Ser891) tend to be associated with more indolent disease. Some have suggested that surgery in these patients can be deferred until plasma calcitonin levels (basal or stimulated) are elevated. Patients should always be screened for the presence of pheochromocytoma before undergoing neck exploration. Surgery for metastatic disease is palliative and targeted at reducing tumor burden rather than cure. Localization techniques (eg, MRI or selective venous sampling for calcitonin) can be helpful in identifying foci of malignant tissue. Recent clinical trials have used a tyrosine kinase inhibitor D6474 (Zactima) to inhibit the RET kinase with an objective remission rate of approximately 30%. Radiation and chemotherapy are of limited utility and are largely confined to later stages of the disease.
Treatment of pheochromocytomas in MEN2 is similar to that for sporadic pheochromocytomas (see Chapter 11). Alpha- and (occasionally) beta-adrenergic blockade is used to control blood pressure and associated hyperadrenergic symptoms and to restore normal intravascular volume in preparation for surgical resection of the tumor. Given the propensity for bilaterality of pheochromocytomas in this disorder, some have favored bilateral adrenalectomy at the time of initial surgery. However, because the incidence of bilaterality is well under 100% and because these tumors are rarely malignant, the most prudent strategy in the face of unilateral adrenal enlargement would appear to be unilateral adrenalectomy at the initial surgery with careful attention at follow-up, looking for the presence of disease in the contralateral adrenal gland. This vigilant approach has the advantage of minimizing morbidity from recurrent pheochromocytoma while sparing the patient the risks associated with lifelong adrenal insufficiency.
Genetic screening for MEN 2A, MEN 2B, or familial MCT is routinely carried out using polymerase chain reaction (PCR)-based tests designed to identify specific mutations in the RET coding sequence (Figure 22–7; see www.genetests.org for laboratories performing these tests). Known RET mutations account for more than 95% of all instances of MEN2, and selected mutations (eg, Cys634 to Arg634 in MEN 2A) account for a disproportionate number of affected individuals. Individuals lacking any of the known RET mutations can be tested using conventional haplotype analysis if informative genetic markers and affected family members are available. Biochemical testing using basal or stimulated plasma calcitonin levels has been largely supplanted by genetic screens. The biochemical tests remain useful, however, in identifying residual disease after thyroidectomy.
Screening for MEN 2A. Genetic screening has largely supplanted biochemical testing in identifying individuals at risk. Details of decisions regarding treatment are discussed in the text. Preoperative plasma calcitonin levels are potentially useful for follow-up but should not supplant genetic testing in assessing the need for surgery (MIBG, metaiodobenzylguanidine; RET, rearranged during transfection proto-oncogene). (See also Figure 7–48.)
In view of the fact that a high proportion of cases of MCT are familial to begin with and as much as 6% of patients with apparently sporadic MCT harbor germline RET mutations, genetic testing for RET germline mutations is probably indicated for all patients presenting with MCT. Controversy persists, however, in terms of what should be done for patients once the mutation has been identified. Some investigators citing incomplete clinical penetrance (according to published data, 40% of gene carriers do not present symptomatically prior to age 70) have argued that employing solely genetic criteria in making the decision for operative intervention subjects a small minority of patients to premature thyroidectomy. They argue that genetic testing should be used to identify those patients who require close clinical and biochemical surveillance to assist with the timing of surgery. Ideally, such biochemical testing (eg, pentagastrin stimulation) should be performed on an annual basis. Exceptions to this general approach might include patients with MEN 2B or a particularly aggressive form of familial MCT where the potential for significant morbidity and mortality would justify operation in any patient harboring the genetic defect regardless of the physical or biochemical manifestations of the disease.
The more widely shared view is that the true penetrance of MCT—combined clinical and preclinical disease—in MEN 2A is closer to 100%. This, when coupled with the high degree of sensitivity and specificity of the PCR-based genetic screens, difficulties encountered in obtaining adequate long-term patient follow-up and biochemical screening, and the potential for false-positive pentagastrin stimulation tests—even within MEN2 kindreds—has led to the recommendation that total thyroidectomy should be performed in all individuals harboring an MEN2–associated RET mutation. This argument has now been supported by several clinical studies in which parafollicular cell hyperplasia as well as early MCT have been identified in operative specimens taken from genetically affected individuals despite normal pentagastrin stimulation tests. False-positive biochemical tests are also a concern. There are several reports in the literature of patients in affected kindreds who have undergone total thyroidectomy following positive pentagastrin stimulation tests but did not, in fact, harbor the MEN2 gene mutation. Histologic examination of excised tissues revealed parafollicular cell hyperplasia, presumably unrelated to MEN2, but no MCT. Collectively, these findings point out the relative deficiencies of biochemical versus genetic testing and offer a compelling argument for early operation as a means of reliably eradicating the disease.