Ret Oncogene and Thyroid Carcinoma

Thyroid cancer is a malignant neoplasm that originates from follicular or parafollicular thyroid cells and is categorized as papillary (PTC), follicular (FTC), anaplastic (ATC) or medullary thyroid carcinoma (MTC). The alteration of the Rearranged during trasfection (RET) (proto-oncogene, a gene coding for a tyrosine-kinase receptor involved in the control of cell differentiation and proliferation, has been found to cause PTC and MTC. In particular, RET/PTC rearrangements and RET point mutations are related to PTC and MTC, respectively. Although RET/PTC rearrangements have been identified in both spontaneous and radiation-induced PTC, they occur more frequently in radiation-associated tumors. RET/PTC rearrangements have also been reported in follicular adenomas. Although controversial, correlations between RET/PTC rearrangements, especially RET/PTC3, and a more aggressive phenotype and a more advanced stage have been identified. Germline point mutations in the RET proto-oncogene are associated with nearly all cases of hereditary MTC, and a strict correlation between genotype and phenotype has been demonstrated. A large spectrum of RET point mutations has been reported to date. Somatic RET mutations, almost all of which are point mutations, have been described in approximately 40-50% of sporadic MTC cases. Although RET somatic mutations have been found in different codons, the M918T mutation in exon 16 is the most frequent. Detecting germline RET point mutations is a useful tool for identifying subjects who are clinically unaffected but who will develop the disease soon or late in their life, thus allowing presymptomatic treatment. From a clinical point of view, detecting RET somatic mutations in tumoral tissue is able to predict a poor outcome of the disease. In recent years, several drugs capable of inhibiting RET activity have been demonstrated to play an important role in the treatment of radiorefractory advanced thyroid carcinomas. Journal of Genetic Syndromes & Gene Therapy J o u r n a l o f G en eti c Sy ndromes &ene T h e r a p y ISSN: 2157-7412 Citation: Elisei R, Molinaro E, Agate L, Bottici V, Viola D, et al. (2014) Ret Oncogene and Thyroid Carcinoma. J Genet Syndr Gene Ther 5: 214. doi:10.4172/2157-7412.1000214


Introduction
Thyroid cancer is a malignant neoplasm that originates from follicular or parafollicular thyroid cells. Approximately 95% of malignant lesions are derived from thyroid follicular cells and are classified as papillary (PTC), follicular (FTC), or anaplastic (ATC) thyroid carcinoma [1]. A relatively small percentage of thyroid carcinomas (7.5-10%) [2] are derived from parafollicular C cells and are called medullary thyroid carcinomas (MTC). MTC can be sporadic (75% of cases) or familial (25% of cases), in which the disease is inherited in an autosomal dominant manner. In these cases, other organs (i.e., parathyroid and adrenal glands) can be involved, and three different syndromes, MEN 2A, MEN 2B and FMTC, are distinguished [3][4][5].
In the last 25 years, many studies have been conducted to identify the genetic alterations related to the pathogenesis of thyroid tumors. Most of these alterations affect genes coding for either some tyrosine kinases receptors, such as RET and NTRK1 or other proteins such as RAS and BRAF. All mutated genes are involved in the activation of the mitogen-activated protein kinase (MAPK) pathway suggesting that activation of this signaling pathway is essential for tumor initiation and that the alteration of a single effector of the pathway is sufficient for cell transformation [6]. In the majority of cases these mutations are mutually exclusive and are present in about 70% of PTC and 65% of MTC [6][7][8] .
It is of interest that two different thyroid tumors such as PTC and MTC are characterized by the activation of the same RET gene with different type of alterations: RET/PTC rearrangements in PTC, and RET point mutations in MTC. Aim of this review is to describe and discuss the pathogenic and clinical roles of RET gene alterations in these thyroid tumors.

Ret Proto-Oncogene
RET gene is subjected to alternative splicing and is translated into three functional isoforms that contain the first 19 exons of the RET gene, RET51, RET9, and RET43; RET splice variants differ in the number of amino acids at their C terminal end [11]. These isoforms are highly conserved between hu mans and mice [12].
The RET gene is expressed in tissues derived from the neural crest, including thyroid C cells and adrenal medulla; while, it is not, or only very moderately, expressed in normal thyroid follicular cells [14,15].
RET proto-oncogene genetic alterations cause both medullary thyroid carcinoma (MTC), which is derived from parafollicular C cells, and papillary thyroid carcinoma (PTC), which is derived from follicular cells. In MTC, point mutations are commonly found, whereas in PTC, the most frequent genetic alterations are gene rearrangements that carry the tyrosine kinase domain of ret under the control of the promoter of ubiquitously expressed genes. In addition to activating point mutations, several single-nucleotide polymorphisms have been described, of which several seem to play a role in the predisposition to MTC [16].

Ret/PTC Rearrangements in PTC
RET/PTC rearrangement was first identified in 1987 [17] by DNA transfection analysis using NIH3T3 cells and tumoral tissue derived from an irradiated PTC. This oncogene was a chimeric gene generated by the fusion of the RET tyrosine kinase domain with the 5' terminal region of the COILED-Coil Domain-Containing Protein 6 (CCD6) formerly named H4 [18]. Subsequently, several activating gene rearrangements with RET to form RET/PTC by juxtaposing the genomic region coding for the tyrosine kinase domain with the 5' terminal regions of several unrelated genes have been described [19] ( Figure 2). The several genes reported to be rearranged with RET are ubiquitously present in all cells and harbour a coiled-coil domain which is responsible for the dimerization leading to the tyrosine hyperphosphorylation thus inducing the transforming activity of the RET/PTC oncogenes.
The presence of the rearrangement determines the constitutive activation of the intracellular domain of the receptor, which is followed by uncontrolled proliferation of the follicular cells and the development of malignancy. RET/PTC1 and RET/PTC3 are the most frequent chimeras identified in PTC [20]. The transforming ability of RET/PTC rearrangements has been unequivocally demonstrated with in vitro and in vivo studies: the RET/PTC oncogene is in fact capable of transforming normal rat thyroid PCCl3 cells into malignant cells [21] and transgenic mice that express RET/PTC oncogenes under the control of the thyroglobulin promoter, which is specific for follicular cells, develop classical PTC [22].
The reported RET/PTC prevalence in thyroid tumors varies greatly in different series [23][24][25][26][27][28][29][30][31][32]. This difference can be attributed to ethnic and geographic variations, as well as to the different sensitivities of the detection methods [33]. Although RET/PTC rearrangements have been identified in both spontaneous and radiation-induced PTC, they seem to occur more frequently in radiation-associated tumors: on this regard, it is worth to note that the first RET/PTC rearrangement was discovered in an irradiated PTC [17]. As a matter of fact, ionizing radiation is a well-known risk factor for PTC [34,35]; in particular, the risk of thyroid cancer has a linear dose response for doses 0.1-2 Gy, with flattening of the dose-response curve at higher doses [1,35,36]. RET/PTC1 is more common after γ-radiation exposure, whereas RET/PTC3 was the predominant type found in patients exposed to I-131 who developed thyroid cancer 10 years after the Chernobyl accident (50-80%) [37][38][39][40][41][42][43][44]. RET/PTC rearrangements have also been found to be more prevalent in children than in adults for both irradiated and non-irradiated PTC [38,45]. According to our previous report, the RET/PTC prevalence is approximately 55% in post-Chernobyl tumors and 37% in spontaneous tumors ( Figure 3). The relationship between the radiation exposure and the development of RET/PTC 1 and 3 rearrangements has been clearly demonstrated by in vitro studies, which showed that irradiated thyroid cells develop RET/PTC rearrangements, and in particular, radiation exposure, under fragile site-inducing conditions, can generate breaks at fragile sites in the RET and CCDC6 genes, which could result in the formation of cancer-causing RET/PTC1 rearrangements [46][47][48]. The evidence that hydrogen peroxidase (H 2 O 2 ) induces DNA single-and double-strand breaks in thyroid cells arose the question of whether H 2 O 2 could be a potential mutagen for thyroid [49]. The link between radiation exposure, production of H 2 O 2 and development of RET/PTC1 rearrangement was recently proved by an elegant in vitro study [50].
As the latency period has increased, the prevalence of RET/PTC rearrangements in post-Chernobyl thyroid tumors has declined, and among the RET/PTC-positive tumors, the percentage of tumors with RET/PTC1 has increased, whereas the proportion of tumors with RET/PTC3 has decreased in parallel [40]. These findings confirm the relationship between the development of PTC and exposure to radiation, which has tended to decline over the years. A significant temporal decrease in the prevalence of RET/PTC rearrangements was also observed in spontaneous PTC [51,52], in parallel with an increase in the BRAF V600E point mutation, which represents the other most important genetic alteration in PTC [53]. These results suggest that the modification of the genetic profile of PTC can be related to the exposure to different environmental and etiological agents over time.
Despite an initial belief that RET/PTC rearrangements were exclusively associated with PTC, they were subsequently found in 10-45% of thyroid adenomas and other non-neoplastic thyroid lesions such as Hashimoto's thyroiditis [24]. RET/PTC rearrangements have also been reported in follicular adenomas in subjects previously exposed to therapeutic external radiation [43,45]. The clinical meaning of these genetic alterations in benign thyroid diseases is still undefined; however, the presence of a one or a few malignant cells within the follicular adenoma or in the context of Hashimoto's thyroiditis cannot be excluded [45,[54][55][56]. Additionally, other RET oncogene rearrangements, different from those described to date in PTC, have recently been reported in a few cases of lung carcinoma and leukemia [57,58].

Clinical Relevance Of RET/PTC Rearrangements in Ptc Prognosis
The biological and clinical significance of RET activation with regard to the outcome of PTC is controversial. Some studies have reported higher rates of local extension and lymph node involvement in RET/PTC-positive cases than in RET/PTC-negative cases [59], whereas other studies were unable to find any relationship between RET/PTC rearrangements and classical prognostic factors [60][61][62][63].
Correlations with a more aggressive phenotype (i.e. rapidly, progressive disease) and a more advanced stage (i.e. T4/N1/M1) have been clearly identified for RET/PTC3 rearrangements [38,64] in cases of post-Chernobyl childhood thyroid cancer because RET/PTC3 positivity was more frequently associated with the more aggressive solid variant. This evidence was confirmed in a subsequent study on sporadic PTC, in which we demonstrated positive correlations between the presence of RET/PTC3 rearrangements and both a larger tumor size and a more advanced stage at the time of diagnosis [63]. Other studies have reported that PTC with RET/PTC3 rearrangements may be more prone to dedifferentiate, showing a more aggressive behavior [65,66], and others claimed that RET/PTC3 plays a role in metastatic spread [66,67]. In contrast, RET/PTC1 rearrangements seem to be correlated with a more favorable behavior [68].
In our experience, RET/PTC rearrangements do not seem to be correlated with any clinical or pathological feature of aggressiveness according to studies of both the expression of the RET/PTC protein by immunohistochemistry [62] or the mRNA expression of RET/PTC rearrangements by real-time RT-PCR [63].

Diagnosis
Fine needle aspiration cytology (FNAC) is the first-choice method to distinguish benign and malignant thyroid nodules [69][70][71]. Nevertheless, FNAC cannot make a definitive diagnosis for 25% of thyroid nodules [72,73], and the final diagnosis is made only by the histological examination of the surgically removed thyroid nodule. The progress recently made in the knowledge of the genetic alterations responsible for thyroid carcinoma has made possible the development of molecular methods for improving the presurgical diagnosis of PTC. The American Thyroid Association guidelines for the management of thyroid nodules recommend the use of molecular testing in the presurgical diagnosis of thyroid nodules [69]. Recent, large prospective studies have confirmed the ability of genetic markers to improve preoperative diagnostic accuracy for patients with indeterminate thyroid nodules. Many of these markers are available for commercial use in reference laboratories but have not yet been widely applied in clinical practice. However, the biggest diagnostic impact can be achieved only by testing FNA samples for a panel of mutations (i.e., BRAF, RET/PTC, RAS, TRK) rather than for a single mutation [74,75]. In particular, the detection of RET/PTC rearrangements does not provide a high positive predictive value because a low, but not negligible, percentage of nodules that were positive for RET/PTC rearrangements on FNA turned out to be benign according to histology [74,75].

Ret Point Mutations in Hereditary MTC (Men 2)
Multiple Endocrine Neoplasia type 2 (MEN 2) is an autosomal dominant inherited disease characterized by the presence of MTC plus other endocrine tumors (pheochromocytoma and/or parathyroid adenomas in MEN 2A, pheochromocytoma and mucosal neurinomas in MEN 2B). MTC alone, lacking an association with other endocrine tumors, may also be inherited (FMTC). In these syndromes, MTC develops in approximately 100% of the affected individuals, whereas pheochromocytoma and parathyroid adenoma develop in 50% and 20%, respectively [2,76]. In 1993, germline point mutations in the RET proto-oncogene were recognized to be associated with MEN 2 [77,78]. In the subsequent years, an International RET Consortium (IRC) [79] was established, and clinical and RET mutational data for 477 kindred affected by MEN 2A, MEN 2B and FMTC were collected. The causative role of the germline RET mutation in MEN 2 was clearly defined, and a strict correlation between genotype and phenotype was demonstrated. It was immediately observed that the classical MEN 2A phenotype was primarily associated with mutations in the RET cysteine codons 609, 611, 618, and 620 in exon 10 and, mostly, with the Cys634Arg mutation in exon 11, whereas MEN 2B was almost exclusively associated with the Met918Thr mutation in exon 16. Presently, the only alternative RET mutation to the classical Met918Thr is the Ala883Phe mutation in exon 15 that has been reported in a few cases of MEN 2B [80]. In contrast, a more widespread distribution of RET mutations was observed in FMTC families [81]. This correlation was confirmed in more recent studies in which several other RET mutations, not known at the time of the IRC, have been studied [2,76] (Figure 4). In fact, in the last 20 years, many other RET mutations have been discovered and reported. These mutations affect not only the classical cysteine codons but also noncysteine codons, such as codon 804 in exon 14, codon 883 in exon 15 and others. These non-cysteine codon mutations are usually associated with FMTC [82][83][84]. Nearly all of the mutations reported to date are listed in public databases (www.hgmd.cf.ac.uk; www.arup.utah.edu/ database/MEN2; www.ensembl.org).
Despite the well-established genotype-phenotype correlation, there is clinical evidence that the same RET mutation can manifest differently in different members of the same family, with some individuals affected by the complete spectrum of the MEN 2A syndrome and others by only one or two of the concurrent endocrine neoplasia. The most validated hypothesis to explain this phenomenon is that, the genetic background of each person can modulate the expression of the genetic alteration and this was clearly demonstrated in transgenic mice [85].
By analyzing the genotype-phenotype correlation it appears that not all mutations confer the same aggressiveness to MTC. It is well known that many cases of FMTC are rather indolent, and they are more frequently associated with non-cysteine RET mutated codons. In particular, mutations at codon 790 and 791 in exon 13 are associated with a non-aggressive form of MEN 2 [86,87] that is concentrated in a particular geographic area (i.e., Germany). Recently, it has also been postulated that mutations at codon 791 may more likely to be singlenucleotide polymorphisms rather than a real transforming mutations [88]. In this regard, it is noteworthy that several of these newly identified RET mutations are very rare, being present only in a few families and a few family members or even in a single affected member (i.e., private mutations), thus raising doubts as to whether they represent the driving force of the tumoral disease. In fact, many of these rare RET mutations are found as the result of very careful genetic screening in "apparently sporadic" MTC, with neither a familial history of MTC nor an association with other endocrine neoplasia [89,90]. These peculiar RET mutations have been defined as variants of unknown significance (VUS), and a comprehensive description is provided in Table 1.
Although RET germline mutations are generally unique and heterozygous, in few cases, tandem RET mutations of codons 805, 806, and 904 in a cis configuration with the Val804Met mutation have been reported in individuals with MEN 2B [91,92]. The likelihood of having a homozygous or compound heterozygous RET mutation is rare, but such mutations have nevertheless been described in the literature and can explain the acquired transforming role of RET mutations with very low or null transforming activity [93,94].
RET mutations have been found in more than 98% of individuals with MEN 2, and only few families affected by hereditary MTC are "orphans" of germline mutations. However, the prevalence and type distribution of RET germline mutations in MEN 2 families are clearly different in different countries [79,95,96], likely due to ethnic variations or to a founder effect. In particular, in a recent Italian series [96], the most frequent RET amino acid substitution was Val804Met. In other European series, although RET mutations at codon 634 were the most prevalent, a high frequency was found for the Leu790Phe and Tyr791Phe mutations [79,95].

Ret Point Mutations in Sporadic MTC
Approximately 7-10% of all apparently sporadic MTC patients harbor a germline RET mutation that is discovered at the time of genetic screening [82,97]. These cases are then reclassified as hereditary forms and the genetic screening of the first degree relatives is strongly suggested [98].
It is of great interest that the same RET mutations can be found at the somatic level in the tumoral tissue of true sporadic MTC cases. According to data reported in a public database (COSMIC), RET somatic mutations have been found in 518/1212 (43%) MTC tumoral tissues. As shown in Table 2, in the majority of cases, these mutations are point mutations (n=475), but deletions, insertions and complex mutations are also reported (n = 43). Although RET somatic mutations have been found at different codons, the Met918Thr mutation in exon 16 is the most frequent (361/475, 76%). As mentioned above for germline mutations, several uncommon somatic mutations have also been reported in isolated cases, but their transforming ability has not always been proven. In vitro and/or in silico analysis is mandatory for both newly discovered somatic and germline RET mutations to establish the transforming ability of these new genetic alteration [90,99] [100] and induce MTC in transgenic mice [101].
Recently, we analyzed the presence of RET somatic mutations in micro MTC (i.e., <2 cm), and we observed that the prevalence of the mutation was significantly lower than in larger tumors [102]. This finding can be interpreted as indicating that RET somatic mutation is a late event that, when it occurs, modifies the biological behavior of the tumor, thereby making it more aggressive and increasing the growth rate.
As an alternative explanation, we can hypothesize that two different types of MTC, RET-negative small indolent tumors and RET-positive aggressive tumors, do exist from the beginning [102].

Clinical Relevance of Ret Point Mutations in MTC Diagnosis
The primary clinical implication of RET mutations in MEN 2 is the ability to screen family members to find those who harbor the same germline mutation previously detected in an MTC index case. This screening allows for the identification of the "gene carriers" when they are clinically unaffected or at an early stage of the disease and the exclusion of "non-gene carriers" from further testing for the rest of their life. Several series have confirmed the effectiveness of this approach. Gene carriers have been detected with a frequency ranging from 15.5% to 69.0% [82,[103][104][105][106][107] in nearly every decade of life, but more frequently at a very young age. A classification of the level of risk  for developing a more or less aggressive disease has been reported in the American Thyroid Association guidelines for the management of MTC patients. According to this classification, RET mutations are classified into 4 risk levels (i.e., A-B-C and D), and different indications for surgical treatment are proposed on the basis of the risk level [108].
RET genetic testing is also indicated in apparently sporadic MTC cases. Screening sporadic MTCs for germline RET mutations may help differentiate patients who exhibit truly sporadic disease from those with unrecognized hereditary disease. If an unexpected germline RET mutation is found, the physician will be alerted to the possible coexistence or future development of adrenal or parathyroid disease. This screening may be extended to the patient's first-degree relatives, thus detecting additional gene carriers, usually in the preclinical phase of the disease. Patients negative for germline RET mutations can be reassured on the sporadic nature of the disease, thus avoiding the need to look for other endocrine neoplasia and to screen his or her relatives. Very detailed guidelines for RET genetic screening and the management of subjects with RET mutation have been recently published by the European Thyroid Association [98].

Prognosis
Not all RET mutations confer the same aggressiveness to the tumor. It is well recognized that Met918Thr has a very high transforming activity [109] and, as expected, is associated with a very aggressive behavior of the MTC that develops in MEN 2B patients, who, especially in the past years, rarely survived to the pubertal age [110]. Although to a lesser extent, the Cys634Arg has also a very high transforming ability, and the MTC associated with MEN 2A with this mutation is rather aggressive [109]. All RET mutations are classified according to their level of risk [108]. The correlation between the presence of RET somatic mutations and the clinical outcome of MTC patients has been controversial for some time. Several years ago, a correlation between the presence of a somatic RET mutation and a more aggressive phenotype of the sporadic MTC was reported by several groups of researchers [111][112][113]. We recently provided evidence that MTC patients with a somatic RET mutation not only have a greater likelihood of unsuccessful treatment but also have a higher likelihood of dying from the disease, as demonstrated by their significantly worse 30-year survival rate with respect to that of patients without somatic RET mutations [114]. This evidence was confirmed in subsequent studies [115,116].

Ret Oncogene Mutations and Target Therapy
Recently, several new drugs that are capable of blocking activated tyrosine kinase receptors have been developed [117,118]. The majority of these drugs are inhibitors of several receptors, and nearly all are active against the vascular endothelial growth factors (VEGFR1, 2 and 3). There are evidences that RET enzymatic activity is refractory to several of these drugs as for example several tyrphostins [119]. At variance, there are drugs, such as motesanib (AMG706), vandetanib (ZD6474), lenvatinib (E7080), cabozantinib (XL184), sorafenib (BAY-439006) and sunitinib, that showed also a specific anti-RET activity when tested either in vitro or in vivo [120][121][122][123][124] (Table 3). These drugs can exert their antioncogenic activity not only as antiangiogenic drugs by blocking the VEGF receptors but also as antineoplastic drugs by inhibiting the signaling and tranforming ability of ret oncoproteins.
Dedifferentiated thyroid cancer that is not responsive to radioiodine and advanced MTC represent real clinical challenges because of the absence of any effective therapy among the conventional strategies, such as chemotherapy and radiotherapy [85]. The high prevalence of RET genetic alterations, especially in advanced MTC, represents the rationale for treatment using the above mentioned drugs.
At the present, both vandetanib and cabozantinib have shown a statistically significant prolongation of progression-free survival (PFS) in MTC patients who are treated with either one or the other [125,126]. In particular, the ZETA trial showed a longer progression-free survival (PFS) for patients receiving vandetanib than for patients receiving the placebo (30.5 months versus 19.3 months; hazard ratio [HR], 0.46). Similarly, the EXAM trial demonstrated a significantly longer PFS for patients who received cabozantinib rather than the placebo (11.2 months versus 4.0 months; HR, 0.28). Presently, vandetanib (@ Caprelsa, AstraZeneca) is available in the USA and Europe following the approval by both the FDA and EMA, whereas cabozantinib (@ Cometriq, Exilixis) has been approved by the FDA and is under the EMA evaluation. Based on very promising results obtained in a phase 2 multicentric study, a phase 3 trial with vandetanib for the treatment of advanced radiorefractory differentiated thyroid cancer is now enrolling patients. Very interesting results in terms of PFS prolongation have been obtained in the DECISION study, which is a phase 3 study comparing the results for advanced radiorefractory differentiated thyroid cancer treated with sorafenib (@Nexavar, Bayer) or a placebo (10.8 months vs. 5.8 months; HR, 0.58) [data presented at the ASCO meeting, 2013] In all of the studies, the correlation between the mutational status and the drug response has been analyzed. In the ZETA trial, MTC patients with RET somatic mutations showed better responses to the drug, but a good response was also obtained for RET negative and RET unknown cases [125,126]. Similar results have been obtained in the EXAM study. These results seem to confirm that the drug activity against the RET mutation is very important for the clinical benefit of the patients but also that other mechanisms are involved in the control of the disease, as likely happens in RET-negative cases.

Conclusions
The RET oncogene is of great relevance in the pathogenesis of both PTC and MTC, although different molecular mechanisms are responsible for its constitutive activation in these two thyroid tumors. Well-established diagnostic and prognostic roles of RET oncogene mutations have been globally recognized. Presently, the relevance of these molecular alterations in the development and use of new target therapies for the treatment of radiorefractory advanced thyroid cancer is under evaluation, with promising preliminary results.