Thyroid Cancer: Molecular Aspects and New Therapeutic Strategies

Despite that thyroid cancer accounts for over 90% of tumors that arise from the endocrine system, these tumors barely represent 2% of solid tumors in adults. Many entities are grouped under the general term of thyroid cancer, and they differ in histological features as well as molecular and clinical behavior. Thus, the prognosis for patients with thyroid cancer ranges from a survival rate of >97% at 5 years, in the case of differentiated thyroid tumors sensitive to radioactive iodine, to a 4-month median survival for anaplastic tumors. The high vascularity in these tumors and the important role that oncogenic mutations may have in the RAS/RAF/MEK pathway and oncogenicity (as suggested by activating mutations and rearrangements of the RET gene) have led to the development of multitarget inhibitors in different histological subgroups of patients. The correct molecular characterization of patients with thyroid cancer is thought to be a key aspect for the future clinical management of these patients.

Thyroid cancer is considered a nonfrequent entity because it represents only 1% of all solid tumors in adults. Women are three-times more likely to suffer this disease, and the incidence of DTC has increased 2.4-fold over the past 30 years [3]. Most DTC patients have a very good prognosis if diagnosed at early stages, and 91% of patients are alive at 20 years when the classical treatment with surgery followed by radioiodine ablation and suppression of thyroid stimulating hormone (TSH) is employed. However, patients with DTC resistant to radioactive iodine, or those with MTC or ATC, have limited options for treatment. Up to now, the chemotherapy drugs most often used worldwide to treat thyroid cancer are doxorubicin and cisplatin. However, the most referenced guidelines in the field do not always recommend the routine use of these agents in the clinic. Doxorubicin and cisplatin showed some activity in a study conducted in the 1970s in a total of 92 patients with all types of thyroid cancer; a progression-free survival of 3 months; an overall survival of 7 months being achieved following treatment [4]. Currently, the use of chemotherapy in patients with thyroid cancer is limited to patients with ATC, those with poorly differentiated histology with a high rate of proliferation, or in highly symptomatic patients who are not candidates for other local or systemic therapies.
The last decade has seen advances in the molecular biology that may underlie the development and progression of these tumors. For example, DTC is mainly associated with mutations in the RAS/RAF/MAPK intracellular signaling  [5].
Classically, thyroid tumors are associated with high vascularization and high levels of vascular endothelial growth factor, fibroblast growth factor, and platelet-derived growth factor (VEGF, FGF, and PDGF, resp.) [6]. The importance of angiogenesis in these neoplasms became clear when several studies showed that tumors with high levels of VEGF in the stroma correlated with a greater propensity towards metastases at distance, as well as with other markers of tumor aggressiveness. Of considerable note is that thyroid cell cultures have a reduced proliferation when the VEGF pathway is blocked, thus demonstrating a direct antitumor activity of antiangiogenic drugs on thyroid tumor cells [7]. Additionally, VEGFR3 is involved in the lymphangiogenesis process, that is, of special interest in papillary thyroid carcinomas that commonly metastasizes in regional lymph nodes [8].
Tyrosine kinases are enzymes that transfer phosphate groups from adenosine triphosphate (ATP) to tyrosine residues of another protein. Many tyrosine kinases are documented in humans as being involved in key processes of cellular control such as survival, proliferation, differentiation, function, and cell motility. In recent years, a large group of low molecular weight agents capable of inhibiting the function of tyrosine kinases have changed the natural history and management of various solid tumors such as kidney cancer, liver cancer, gastrointestinal stromal tumor (GIST), and, more recently, pancreatic neuroendocrine tumor. These drugs are called multitarget tyrosine kinase inhibitors and, due to their biochemical structure being similar to that of ATP, they are able to block the intracellular activation of several membrane receptors or proteins in the intracellular signaling cascade with tyrosine kinase activity. The degree of affinity and selectivity of this inhibitory activity is very variable (Table 1) [6].
Advances in understanding the molecular pathogenesis of different subtypes of thyroid cancer have, currently, made this field of research one of the busiest in the world of endocrine oncology. The application of discoveries in the laboratory to the patient (the "bench-to-bed" transfer) is one of the biggest challenges in translational research area, since these tumors can become a paradigm of therapy based on molecular features that govern the development and growth of tumors in each individual patient. This paper aims to summarize the key molecular determinants of each histological subtype of thyroid cancer and the data derived in recent years from clinical studies conducted with multitarget agents.

Clinical Management of DTC Patients Sensitive to
Radioactive Iodine. DTC represents the vast majority of thyroid tumors. Treatment is based on thyroidectomy with or without lymphadenectomy regarding papillary or follicular histology, followed by radioiodine ablation in those patients at high risk of lymph node metastases and levothyroxine in doses that suppress TSH levels. The semiannual or annual monitoring using cervical ultrasound, measurement of thyroglobulin levels as a tumor marker, and whole-body iodine scan is appropriate for these patients.
It is estimated that between 10 and 15% of patients undergoing DTC will have disease relapse. In 3 out of 4 cases, recurrence occurs at the cervical level in the thyroid bed or lymph node level. In these patients salvage surgery, radioiodine therapy and external radiotherapy is recommended in highly selected patients.
Distant metastases occur in <10% of DTC patients, and only half of them are apparent at the time of diagnosis. The lungs are the main receptor organs of these tumor metastases (50%), followed by bone (25%). Initial treatment of patients with disseminated disease includes TSH-suppressive doses of levothyroxine plus radioiodine administered in conditions of high serum TSH levels in those patients with positive total body scan uptake, albeit complete remission is achieved in only 30% of subjects [9].

Clinical Management of Patients with Radioactive Iodine-Refractory DTC.
A patient is considered resistant to the use of radioactive iodine when at least one tumor lesion is observed that does not show uptake of radioiodine, or when the lesion radiologically progresses in the first 12 months post-radioiodine administration, or when the patient has persistent disease following the administration of an accumulated dose of radioactive iodine of >600 mCi Journal of Thyroid Research 3 [10]. The median survival of patients with radioiodineresistant DTC and distant metastases ranges between 3 and 6 years. It is in these patients that the application of new multitarget inhibitors has its main testing ground [8]. The use of classical chemotherapy, both as monotherapy and in combination with other treatments, in patients with radioiodine-refractory DTC has not shown any significant benefit in these patients and, conversely, has shown high toxicity. The most recent study conducted with doxorubicin was published in 2008 by Matuszczyk et al. [11]. Among the 22 recruited patients with radioiodine-refractory DTC, only one (5%) achieved a partial positive response that lasted for 6 months. Adverse effects included alopecia (42%), nausea (23%), pneumonia (20%), and neutropenia (10%). Of the 22 patients, 12 showed disease progression during the first measurement of treatment response, and only 42% of the patients recruited for the trial had stable disease for a median of 7 months. Other authors have also tested several combination regimes with specific chemotherapy agents such as taxanes, platinum, or gemcitabine, with little added clinical benefit; response rates ranging between 0 and 22% and without an amelioration of progression-free survival (PFS) of the disease [12]. (Figure 1). In recent years, several different genetic events have been identified as being related to the genesis of the DTC. Further, a correlation has been reported between the manifestation of histological alterations of the disease and the presence of changes in the regulation of intracellular molecular pathways that confer a distinct clinical behavior. Two major molecular determinants that involve the same molecular pathway are responsible for the appearance of a PTC: the alteration of gene regulation of the RET gene expression and the abnormal activation of signaling pathways, RAS/RAF/MAPK. There is an increasing body of evidence showing that the role of PI3K/Akt/mTOR pathway may be crucial for tumor development and may become an attractive target in the future [13].

Molecular Biology of DTC
The abnormal chromosome rearrangement of protooncogene RET (rearranged during transfection) with the tyrosine kinase domain of other genes located in the same or other chromosomes results in the fusion oncogene RET/PTC that plays a key role in the pathogenesis of up to 30% of patients with PTC [14]. Their involvement is even greater in childhood PTC and in patients whose tumor results from exposure to ionizing radiation (approximately 50-80% of cases). The RET proto-oncogene is activated by fusion of the RET TK domain with the 5 -terminal sequence of one of different heterologous genes via rearrangements that generate a series of chimeric-transforming oncogenes collectively described as RET/PTCs. The results of RET/PTC reorganization are abnormal proteins constitutively activated from RET kinase, independently of the binding to extracellular ligands, that activate multiple intracellular signaling pathways leading to clonal expansion and neoplastic transformation of the thyroid follicle cells. It is widely known that RET oncogene rearrangements plays a crucial role in radiation-associated papillary thyroid carcinogenesis.  Other proto-oncogene rearrangements have been identified affecting to the NTRK1 gene in PTCs. The frequency of these rearrangements could be seen in up to 25% of patients with PTC. NTRK1 proto-oncogene rearrangements result in the formation of chimeric genes composed of the tyrosine kinase domain of NTRK1 fused to 5 sequences of different genes (TPM3, MET, and TRK-T2). It is estimated that NTRK1 proto-oncogene is activated by rearrangement with a similar frequency in "spontaneous" and radiation-associated thyroid tumours. Moreover, NTRK1 proto-oncogene activating rearrangements play a role in the development of a minority of radiation-associated PTC but not in adenomas [15].
Molecular aberrations observed in FTC are significantly different from those described in PTC, including RAS mutations in up to 45% of patients and rearrangements involving the transcription factor gene of the paired-box 8 (PAX8) and peroxisome proliferator-activated receptor-γ (PPARγ) resulting in a PAX8/PPARγ rearrangement observed in up to 35% of FTC [12].
Genetic alterations in the signaling pathway of RAS/RAF/ MAPK constitute the most frequent genetic/molecular anomaly in patients with DTC. BRAF is a serine-threonine kinase that is key in the intracellular regulation of this cellular pathway. Activating mutations have been observed in the V600E codon of the BRAF gene in up to 70% of patients with PTC. Similarly, patients with mutations in BRAF have more chances of developing resistance to radioactive iodine resulting from alterations in the normal patterns of mobilization and metabolism of iodine in the thyroid follicle cell. These features confer poorer clinical prognosis to patients carrying the mutation. The V600E BRAF activating mutation is associated with a loss of expression of the unidirectional cotransporter (symport) sodium iodine and, consequently, a lower uptake of radioiodine by the metastatic lesions of tumors that harbor the mutation. This decreased expression of symport is mediated via the TGFβ autocrine pathway. Point mutations in the gene encoding the RAS oncoprotein occur to a lesser extent (15%) in PTC, but their frequencies are increased in patients with FTC and less differentiated histologies. Mutations in any gene encoding for RAS (HRAS, KRAS and NRAS) involve abnormal activation of mitogenactivated protein kinase (MAPK) and the PI3K/Akt/mTOR alternative parallel route [12,33].
Mutations in BRAF, RAS and RET rearrangements are involved in >70% of PTC. However, these molecular alterations are mutually exclusive in most cases and are rarely expressed simultaneously within the same tumor. As such, they are independently able to promote PTC.
Activation and dysregulation of the intracellular signaling cascade PI3K/Akt/mTOR through loss of expression of tumor suppressor phosphatase PTEN has been implicated in the development of DTC and undifferentiated tumors of the thyroid. Likewise, mutations in the gene PI3CA, encoding for the phosphatidyl-inositol-tri-phosphate kinase (PI3K), play a role as initial triggers that activate the intracellular signaling pathway ( Figure 2).
Another molecular alteration frequently found in DTC is amplification of the gene encoding for the hepatocyte growth factor (c-MET) receptor. The MET membrane receptor cooperates with VEGF receptor to promote angiogenesis, cell mobility, and survival of thyroid cells. Amplification occurs in 80% of papillary thyroid tumors, while its silencing in animal models induces tumor regression, reduction in tumor size, and the appearance of new distant metastases [12].
Angiogenesis is another potential area for the design of new molecular targets. Known receptors for VEGF (VEGFR-1, -2, and -3) and the FGF (FGFR) and PDGF (PDGFR-α and -β) have been shown to be associated with an increased risk of distant metastases, an increased risk of disease recurrence, less protracted PFS, and increased presence of mutations in BRAF. In in vitro experimental models, blockade of the VEGF pathway has succeeded in delaying DTC growth [10].

Multitarget Agents in Current Clinical Development for
the Treatment of DTC (Table 2) 2.4.1. Sorafenib. Sorafenib is a multitarget agent for several major molecular pathways involved in the development and progression of PTC, including the serine-threonine kinase BRAF and the tyrosine kinases from the membrane receptors VEGFR-2 and PDGFR-β (Table 1). From

2.4.2.
Vandetanib. Vandetanib (ZD6474) is a low molecular weight inhibitor of tyrosine kinases, which acts mainly against VEGFR-2, VEGFR-3, and RET and, in higher concentrations, on the epidermal growth factor receptor (EGFR) ( Table 1). Vandetanib was one of the first drugs to demonstrate activity in thyroid cancer cell lines, mainly through its action on the rearrangements RET/PTC and on RET mutations. Vandetanib has been the subject of report of the largest randomized clinical trial conducted to date in patients with DTC refractory to radioiodine. A total of 145 patients received vandetanib at doses of 300 mg/day or placebo. PFS, which was the primary endpoint, was significantly higher from the clinical as well as statistical point of view, in patients receiving the study drug (11 months) as compared to those receiving placebo (5.8 months) with a relative risk (hazard ratio) of 0.63 (95% CI: 0.43-0.92). However, the objective response rate was less than 5% in the vandetanib arm, and is difficult to explain. The most frequent adverse events found in the vandetanib arm were diarrhea (74%), asthenia (26%), fatigue (23%), nausea (25%), hypertension (34%), hyporexia (26%), and skin rash (25%

Sunitinib.
Sunitinib is a multitarget inhibitor of 3 known receptors of VEGF, as well as of RET and PDGFRα and -β, and of the protein derived from the RET/PTC realignment (Table 1). In a phase II study [23], sunitinib, administered at a dose of 37.5 mg/day in continuous schedule, induced metabolic complete response in 28 DTC patients, with uptake of distant disease on PET/CT scan a partial response in 7 cases (25%), and disease stabilization in 14 (50%). The median time to progression (TTP) was 12.8 months, and the decline in the uptake of fluorodeoxyglucose (FDG) at 7 days of treatment with sunitinib was superior in those patients who subsequently achieved positive radiological response (by RECIST criteria). In another phase II study conducted in 31 evaluable patients with DTC with metastatic disease progression, sunitinib administered at a dose of 50 mg/day for 4 weeks followed by 2 weeks off treatment, achieved a partial response rate of 13%, with 63% stabilizations [24]. In a retrospective analysis that we conducted in 17 patients with thyroid cancer of different histologies treated with sunitinib for advanced disease, we found an objective reduction of the size of metastases in 4 (33%) of the evaluable patient, and a median PFS of 13.3 months in the overall patient groups, 8.6 months being the PFS in those patients who had failed to one prior systemic treatment with sorafenib. The main grade 3-4 toxicities found in the study were asthenia (7.1%), thrombocytopenia (7.1%), and mucositis (5.9%) [34].

Lenvatinib (E7080). Lenvatinib (E7080) is a potent
inhibitor in the nanomolar range of VEGFR-1 and -2 as well as PDGFR-β and the fibroblast growth factor receptor-1 (FGFR-1) ( Table 1). In an international, open, phase II study which treated a total of 99 patients who had advanced thyroid cancer, lenvatinib showed an objective reduction in the size of metastases (according to RECIST criteria 1.1) in 29 (50%) of the 58 patients with histology of DTC. Also, 21 (36%) patients achieved disease stabilization, and the median PFS in all DTC patients was 12.6 months [25]. Among the 41 (70%) patients who had not received prior antiangiogenic therapy for advanced disease, the radiological objective response rate was 54% while, in the 17 patients who had previously received anti-VEGF, the observed partial response rate was 41%. However, dose reduction was required in 35% of patients, and 23% of them discontinued treatment due to toxicity. The most frequent grade 3 or 4 toxicities that led to dose reductions were hypertension (10%), proteinuria (10%), decreased weight (7%), diarrhea (10%), and fatigue (7%). Based on these promising data, we began a phase III registration study in which 360 patients with DTC refractory to radioactive iodine, which had been previously treated with other anti-VEGF agents, were randomized to receive lenvatinib or placebo. PFS was the primary aim of the study (NCT01321554).

2.4.6.
Motesanib. Motesanib (AMG706) is an oral inhibitor of multiple kinases, including VEGFR-1, -2, and -3 as well as the wild and mutant forms of the membrane receptor RET. In one of the initial phase I studies, focusing on recommended doses and major toxicities, motesanib was shown to achieve objective radiographic partial response in 3 of the 5 patients with thyroid cancer. This activity was not expected in these patients and led on to a prospective phase II study in 93 patients with DTC being carried out. One in every 3 patients remained on treatment for 48 weeks after commencement. The most common adverse events found at any grade were diarrhea (59%), hypertension (56%), fatigue (46%), and weight loss (40%). The radiological response rate assessed by independent reviewers was 14%, and up to 35% of patients benefited from a stabilization of their disease beyond 6 months. The median PFS was 9.3 months and, although the drug does not inhibit BRAF, the investigators observed that patients carrying a mutation in this pathway were less likely to progress if they received motesanib than patients without the mutation This finding suggested a greater involvement of angiogenesis in patients with the BRAF mutation [26].

2.4.7.
Pazopanib. Pazopanib shares the main targets with the other agents discussed above. However, it appears to have a marginal activity on RET/PTC and BRAF that is, its activity in these tumors is due mainly to its antiangiogenic effect rather than any intrinsic antitumor effect (Table 1). In a phase II study, pazopanib administered at a dose of 800 mg/day, induced a radiographic response rate of 49% in 37 patients with DTC who had disease progression over the previous 12 months. This response rate is among the highest achieved by any multitarget agent reported in this context. Up to 66% of patients who achieved a positive objective radiological response rate were able to maintain the response one year from the commencement of treatment. The median PFS in the total patient sample was 11.8 months.

Clinical Management of Patients with MTC.
Currently, there were no officially approved therapy for the treatment of metastatic MTC until the recent approval of vandetanib by the American Food and Drugs Administration (FDA) and the European Medicine Agency (EMA). Following the failure of initial treatment by thyroidectomy and cervical lymphadenectomy, patients have received only symptomatic care measures. Therefore, this tumor has represented, and currently represents, a niche for the development of new active drugs, especially considering that between 35 and 50% of patients diagnosed with metastatic disease are likely to be diagnosed as having lympho-ganglionar metastases, and up to 15% of patients will develop distant metastases after diagnosis. Half of the patients diagnosed with MTC will ultimately present distant metastases during the clinical evolution of the disease. MTC is considered a neuroendocrine tumor caused by the uncontrolled proliferation of para-follicular C cells producing calcitonin derived from the neural crest. It accounts for between 4 and 10% of thyroid cancers. In up to 30% of cases, the CMT is associated with an inherited syndrome and multiple endocrine neoplasia type 2 (MEN 2A or 2B) or with familial MTC (FMTC). Most hereditary MTC occur within MEN 2A that is often associated with pheochromocytoma or hyperparathyroidism [35].
Sporadic MTC is diagnosed in the 5th or 6th decade of life, while the hereditary versions of the disease can be diagnosed in 5-year-old children. Conventional therapy with chemotherapy has not proven to be particularly useful in patients with metastatic MTC. The most commonly used chemotherapy regimen has traditionally been that of doxorubicin monotherapy. This treatment achieves a clinical benefit, including the partial responses plus stable disease of 6 months, in about 21% of patients, with 79% of patients progressing within 5 months of treatment [27]. Regimens for neuroendocrine tumors have also been used for MTC, including a combination of streptozocin, dacarbazine, and 5-fluorouracil. Up to 30% of patients achieved objective response rates in small nonrandomized studies [36]. (Figure 1). Approximately 50% of patients with sporadic MTC have a somatic mutation at codon 918 of the RET gene. These mutations are considered "gain of function" type. The presence of this mutation is also associated with a high probability of lympho-ganglionary metastases, recurrent/persistent disease, and decreased survival. By contrast, germinal mutations at different loci of the RET gene are present in virtually all hereditary syndromes associated with MTC. There is a strong genotype-phenotype correlation between the location of the mutation in RET and the associated MEN2 type. This enables the patients to be classified into different risk groups according to the mutation they carry and as such, to be referred (or not) for prophylactic thyroidectomy [35,37].

Molecular Biology of MTC
The abnormal activation of RET by any of the several mutations affecting the coding part of the gene can trigger activation of multiple intracellular signaling pathways such as RAS/RAF/MAPK, PI3K/Akt/mTOR, or Rac/JNK. (Table 3) 3

Cabozantinib (XL184).
Cabozantinib is an oral inhibitor that, in addition to acting against VEGFR-1 and -2, and c-KIT as are most of the drugs discussed above, is able to inhibit c-MET which, as also discussed above, is amplified in a large proportion of patients with DTC and MTC. It also acts against RET and against RET/PTC rearrangements (  [32]. In patients with RET mutation and treated with sunitinib, the probability of remaining progression-free at one year was 88% (95% CI: 43%-98%).

Anaplastic Thyroid Cancer (ATC)
4.1. Clinical Management of Patients with ATC. ATC contributes less than 2% of all thyroid-derived tumors but is responsible for 14 to 39% of deaths from thyroid cancer. With a frequency 5-fold higher in men than in women, the median survival range depends on status between 3 and 9 months after diagnosis. Approximately 50% of patients diagnosed with ATC already have metastases at the time of diagnosis, and another 25% will eventually develop metastases. Treatment options are surgery, palliative radiotherapy, and chemotherapy. ATC tumors are considered chemo resistant. Doxorubicin in monotherapy achieves positive radiological response rates in <20% of patients, and this response is rarely protracted. Other platinum-based combinations such as taxanes, gemcitabine or vinorelbine can achieve up to 53% radiological response but does not have much impact on the overall survival of patients [38]. (Figure 1).

Combretastatin A4 Phosphate (CA4P).
CA4P is a vascular-disrupting agent, also called fosbretabulin tromethamine or combretastatin A4 phosphate. Unlike other antiangiogenesis agents that block tumor blood vessel formation, CA4P prevents blood flow in existing vessels. This drug has shown some activity as a single agent in patients with metastatic ATC who have not responded to at least one prior systemic therapy but in whom, on rare occasions, positive responses were observed and objective radiological PFS of 3 months was achieved. The final results in overall survival of the largest randomized trial conducted in patients with metastatic ATC were presented at American Society of Clinical Oncology (ASCO) annual meeting in 2011. A total of 80 of the planned 180 patients with ATC were recruited and randomized to receive carboplatin + paclitaxel with or without associated CA4P in a 2 : 1 randomization. The median overall survival was higher in the CA4P arm but did not reach statistical significance (5.2 versus 4.0 months; HR 0.65; 95% CI: 0.38-1.10); 27% of patients were alive at 1 year in the arm that received the experimental therapy compared to 9% in the standard treatment arm. The planned total patient sample could not be reached due to low recruitment [40].

Conclusion
Thyroid cancers are one of the greatest challenges for those treating advanced tumors. Thyroid cancer offers extensive opportunities to work in interdisciplinary teams that include not only oncologists and endocrinologists but also specialists in pathology, nuclear medicine, surgery, otolaryngology, radiology, clinical laboratory, and radiotherapy. Coordinated administration (whether concomitant or sequential) of the different therapeutic options for these patients make collaborative and efficient functioning of surgical and medical professionals more important than ever. The opportunity that presents itself is to apply the various treatments becoming available based on the genotypic characteristics of patients. Moreover, we have to consider the possibility of administering tyrosine kinase inhibitors for a long time with caution. We assume that these drugs are quite specific cytostatic and not cytotoxic agents, therefore, there is a need to give these agents until progression of the disease or toxicity. The concept of long-term treatment management of these new agents to patients in a childbearing age, with known cardiovascular toxicity, and perhaps many different side effects that could arise in the long-time followup, makes the monitoring of patients receiving inhibitors tyrosine kinase a must to be done very closely.
Maximizing clinical outcomes of tyrosine kinase therapy requires clear, effective communication, anticipation of side effects, and early intervention to avoid treatment delays and dose-limiting toxicities. There is a direct correlation between the plasmatic levels of the tyrosine kinase inhibitors with the activity of the drug, therefore, we would need to maintain as far as we can the highest dose of the targeted agent we consider is the most appropriate to our patients. In this sense, the experience in the management of tyrosine kinase inhibitors is crucial and probably the development of therapy management guidelines would be necessary in the next years.
To summarize, there is a need for a greater understanding of the molecular basis of the disease by the various health professionals charged with the care of these patient that could lead to a better optimization of treatment options in the advanced setting.