The Author declares no conflict of interest related to this work.

Correspondence to:
Mehtap Cakir, MD, Division of Endocrinology and Metabolism, Meram School of Medicine, Selcuk University
42080 Konya, TURKEY; Phone: +90 332 223 77 39; Fax: +90 332 323 72 10; email: cakirmehtap@yahoo.com

ABSTRACT
Thyroid cancers are the most common endocrine cancers and constitute around 1% of all human malignancies. Papillary thyroid carcinoma (PTC), follicular thyroid carcinoma (FTC) and anaplastic thyroic carcinoma (ATC) are derived from follicular epithelium. Among these PTC and FTC are coined under the term differentiated thyroid cancer (DTC) and comprises 90% of all thyroid cancers. The prognosis of DTC are in general favourable with an initial treatment of total thyroidectomy and radioactive iodine ablation, and TSH suppression in the follow-up period. This combined treatment regimen called the conventional treatment is highly successful in patients without metastases. Less than 10% of patients with DTC present with or ultimately develop distant metastases and around 40- 50% of these subjects are unresponsive to conventional treatment. In radioiodine refractory subjects 10-year survival is around 20%. Conventional chemotherapeutics have also been tried in these subjects with success rates less than 25%. On the other hand, for medullary thyroid cancer (MTC), which originates from the parafollicular C cells, the only curable treatment is surgery. However, 60- 80% of these subjects have metastases on admission and 5-year survival rates are less than 50%. ATC has the worst prognosis among thyroid cancers, with an expected survival less than 6 months after diagnosis. Thus for radioiodine treatment refractory DTCs, locally advanced and metastatic MTCs and ATCs new treatment modalities are needed. This review will focus on the use of tyrosine kinase inhibitors (TKIs) in thyroid cancers and a summary of the results of clinical trials to date will be presented.

INTRODUCTION
The management of patients with thyroid cancer depends on the histological subtype and stage of the tumour. Among thyroid cancers, a great majority of DTC patients can be cured with the conventional treatment regimen composed of surgery, radioactive iodine and thyroid hormone therapy. Less than 10% of DTC patients present with or ultimately develop distant metastases (1-3). Around 40-50% of these subjects are refractory to radioactive iodine treatment (1,4) and have a 10- year survival rate of 10-20%.`In radioactive iodine resistant subjects, conventional chemotherapy response rates are low and long term cure is rare (5). For patients with MTC, which comprises around 5% of thyroid cancers, the only curable treatment choice is surgery. For the treatment of ATC, which is luckily the least common histologic type of thyroid cancer (~1%), surgery, radiotherapy and chemotherapy have been used with disappointing results. Treatment choices for radioactive iodine refractory DTC, locally advanced and metastatic MTC and ATC are limited. In recent years, use of TKIs in clinical trials of thyroid cancer patients have shown promising results. Considering the role of Ras/Raf/ mitogen-activated ERK kinase (MEK)/ extracellular signal-regulated protein kinase (ERK) 1/2 signalling pathway in the pathogenesis of thyroid cancers, TKIs targeting the upstream of this pathway, seem a rational choice of treatment. This review will be a summary of the mechanism of action of TKIs, their expected role in the treatment of thyroid cancers and results of clinical trials.
RECEPTOR TYROSINE KINASES and DOWNSTREAM PATHWAYS
Receptor tyrosine kinases (RTKs) transmit signals that regulate cell proliferation and differentiation, promote cell migration and survival, and modulate cellular metabolism (expert). Epidermal growth factor receptor (EGFR), vascular endothelial growth factor receptor (VEGFR), fibroblast growth factor receptor (FGFR), glial cell line-derived neurotrophic factor receptor (RET), platelet-derived growth factor receptor (PDGFR), hepatocyte growth factor receptor (MET) are among the several RTKs. Growth factors activate these RTKs, which then activate two key signal-transduction components: the lipid kinase phosphoinositide 3-kinase (PI3K)/Akt and the Ras/Raf/MEK/ERK 1/2 signalling pathways (Figure 1, Figure 2).
Any of Ras/ERK 1/2 and PI3K/Akt elements, when converted to a constitutively active state, is sufficient to drive 'susceptible' but otherwise normal cells into a state that exhibits at least some properties of an oncogenically transformed cell (6). The Ras/ERK 1/2 pathway is a key signaling pathway that is involved in the regulation of normal cell proliferation, survival, growth and differentiation (6). Ras is the most frequently mutated oncogene in human tumours (7). The ERK 1/2 pathway is dysregulated in approximately one-third of all human cancers. PI3Ks are a family of proteins involved in the regulation of cell growth, metabolism, proliferation, glucose homeostasis and vesicle trafficking. Mutation in one or another PI3K pathway component accounts for up to 30% of all human cancers.

Figure 1- A simplified diagram of the Ras/ERK 1/2 and PI3K/Akt pathway. Yellow lines show activation, red lines show inhibition of the corresponding protein.



ABBREVIATIONS: BAD: BCL2-antagonist of death; CcnD1: Cyclin D1; CcnE: Cyclin E; ERK: Extracellular> signal-regulated protein kinases; FOXO: Forkhead box O; GSK3: Glycogen synthase kinase-3; HDM2: Human homolog of murine double minute ubiquitin ligase; IKK: Iκ B kinase; MEK: Mitogen-activated protein/extracellular signal-regulated kinase (ERK) kinase; mTOR: Mammalian target of Rapamycin; p21: Cyclin dependent kinase inhibitor p21Cip1/WAF1; p27: Cyclin dependent kinase inhibitor p27 Kip1; PDK1: Phosphoinositide-dependent kinase 1; RTK: Receptor tyrosine kinase; TSC: Tuberous sclerosis.

Regarding thyroid cancers, several mechanisms including overexpression or mutation of RTKs, rearrangements producing chimeric oncogenes, activating mutations of oncogenes, inactivation of tumour supressor genes due to mutations, deletions and epigenetic changes, have all been reported with changing rates (Figure 2). Figure 2 shows the demonstrated genetic changes that have been detected in PTC, FTC and MTC. To start with receptors, the role of RET is very well established in MTC (8). In addition to its well-known role in hereditary MTC, it is now known that approximately 50% of sporadic MTCs harbor activating RET mutations (9,10). Overexpression of EGFR (11), MET (12) and FGFR4 (13) were also reported in MTCs. On the other hand, RET/PTC represents a recombinant protein product from a chromosomal rearrangement with the combination of the 3' portion of the RET gene and the 5' portion of a partner gene (14). This recombination results in constitutive activation of the tyrosine kinase in RET. Among the more than 10 types of RET/PTC, which are mainly found in thyroid cancer, the most common and important types are RET/PTC1 and RET/PTC3 (14). RET/PTC occurs in about 15-20% of adult PTC patients (15). As RET/PTC was also reported to occur with high prevalences in benign thyroid tumours and Hashimoto's thyroiditis, it appears that, RET/PTC alone, under physiological expression, may not be sufficiently oncogenic and additional genetic alterations may be required for thyroid cell transformation (14). Ras mutations are common particularly in the follicular variant of PTC, FTC and poorly-differentiated thyroid cancer (PDTC) with a prevalence around 20-40% in most series (14). B-Raf which is an important component of RTK downstream pathway, is mutant in an avarage of 44% of PTC and its mutation was also frequently associated with the absence of radioiodine sensitivity (16). PI3KCA gene, which encodes for the catalytic subunit of PI3K, was found to be amplified in 24% of FTC (17) and 42% of ATC (18). An epigenetic inactivating mechanism through aberrant methylation of the PTEN gene was also reported in FTC and particularly in ATC (19,20). Activation of PI3K/Akt pathway in thyroid tumours was also noted with PPARγ/Pax8 rearrangement (14). As a result, in general PI3K/Akt pathway is commonly activated in FTC, while Raf/ERK 1/2 pathway is commonly activated in PTC (14).

Figure 2- A simple diagram showing RTKs, their ligands and downstream pathways. Some of the genetic changes that have been demonstrated to date in PTC, FTC and MTC are listed in rectangles.



ABBREVIATIONS: EGF(R): Epidermal growth factor (receptor); ERK: extracellular signal-regulated protein kinase; FGF(R): fibroblast growth factor (receptor); FTC: Follicular thyroid cancer; GDNF: Glial cell derived neurotrophic factor; HGF: hepatocyte growth factor; MET: hepatocyte growth factor receptor; MTC: Medullary thyroid cancer; PDGF(R): platelet-derived growth factor (receptor); PI3K: phosphoinositide 3-kinase; PTC: Papillary thyroid cancer; PTEN: Phosphatase and tensin homolog deleted on chromosome 10; RET: Glial cell line-derived neurotrophic factor receptor; TKI: Tyrosine kinase inhibitors; VEGF(R): vascular endothelial growth factor (receptor).

TYROSINE KINASE INHIBITORS
Generally, RTKs are activated through ligand-induced oligomerisation, typically dimerisation, which juxtaposes the cytoplasmic tyrosine kinase domains (21). For most RTKs, this juxtaposition facilitates autophosphorylation in trans of tyrosine residues in the kinase activation loop or juxtamembrane region, inducing conformational changes that serve to stabilise the active state of the kinase (21). These and other phosphotyrosine residues serve as recruitment sites for a host of downstream signalling proteins. TKIs are a group of small molecules that interfere with the interaction between the kinase domain and ATP, thereby inhibiting phosphorylation of the kinase and downstream substrates (8). TKIs used in the treatment of thyroid cancers and the corresponsing RTKs inhibited by are shown in Table 1.

TABLE 1- Tyrosine kinase inhibitors and the corresponding tyrosine kinases inhibited by the drug (8,34,38).



ABBREVIATIONS: C-KIT: Stem cell factor receptor; EGFR: Epidermal growth factor receptor; FGFR: fibroblast growth factor receptor; MET: hepatocyte growth factor receptor; PDGFR: platelet-derived growth factor receptor; RET: Glial cell line-derived neurotrophic factor receptor; TKI: Tyrosine kinase inhibitors; VEGFR: vascular endothelial growth factor receptor.

CLINICAL TRIALS
Table 2 shows a list of clinical trials performed with TKIs to date. Case reports and phase I trials have not been included in this list.

TABLE 2- Clinical trials of tyrosine kinase inhibitors performed in thyroid cancers.





ABBREVIATIONS: ATC: Anaplastic thyroid cancer; DTC: Differentiated thyroid cancer; MTC: Medullary thyroid
cancer; PDGFR: Platelet-derived growth factor receptor; PTC: Papillary thyroid cancer.

Importantly in all these studies patient accruel, follow-up and treatment response were performed according to Response Evaluation Criteria in Solid Tumours (RECIST) 1.0 guidelines (22). According to these criteria, only patients with measurable lesions should be included in the studies to evaluate objective drug response. All measurable lesions up to a maximum of five lesions per organ and 10 lesions in total, representative of all involved organs, should be identified as target lesions and recorded and measured at baseline. The evaluation of target lesions should be as follows; complete response—the disappearance of all target lesions; partial response—at least a 30% decrease in the sum of the longest diameter of target lesions, taking as reference the baseline sum longest diameter; progressive disease—at least a 20% increase in the sum of the longest diameter of target lesions, taking as reference the smallest sum longest diameter recorded since the treatment started or the appearance of one or more new lesions; stable disease—neither sufficient shrinkage to qualify for partial response nor sufficient increase to qualify for progressive disease, taking as reference the smallest sum longest diameter since the treatment started (22). The subjects included in these studies are patients with refractory DTC, locally advanced or metastatic MTC and ATC. Patients with radioiodine refractory disease may be defined as patients who have at least one lesion without radioiodine uptake or that has progressed within 1 year after radioiodine treatment (23). Imatinib was the first TKI to be used in thyroid cancers. It was used in MTC patients in 3 studies (24-26) without any objective response and considerable toxicity reported in two of these studies (24,25). Additionally, imatinib was used in ATC in a very recent study with 25% partial response at the end of 8 weeks and was well-tolerated (27). The authors stated that, due to the difficulty of accruing patients with this rare malignancy at a single institution, further investigation of imatinib in ATC should better be in a multi-institutional setting.
Motesanib has been used both in MTC and DTC patients (28,29). In the MTC study only a 2% partial response was noted (28). Among patients with tumor marker analysis, 69 (83%) of 83 and 63 (75%) of 84 had decreased serum calcitonin and carcinoembryonic antigen during treatment, respectively, compared with baseline. Although the objective response rate was low, a significant proportion of MTC patients (81%) achieved stable disease while receiving motesanib. In the study of DTC patients, in which 67% of patients had PTC, the objective response rate was 14% (29). Among the 75 patients in whom thyroglobulin analysis was performed, 81% had decreased serum thyroglobulin concentrations during treatment, as compared with baseline levels.
Sorafenib, the only TKI which also inhibits B-Raf, was studied in hereditary MTC and DTC and a small sumber of ATC patients (30-33). In the study of 15 hereditary MTC patients, 1 partial response and 1 fatal toxicity was reported (30). In a study performed in a mixed population of patients with mainly PTC and FTC and 1 MTC and 2 ATC/PDTC, an overall clinical benefit rate (partial response + stable disease) of 77%, median progress-free survival of 79 weeks, and an overall acceptable safety profile was observed (31). However a single patient died of liver failure that was likely treatment related (31). In a similar study, study arm A included 33 chemotherapy naïve PTC patients while arm B included 8 prior chemotherapy positive PTC patients in addition to FTC and ATC patients (32). A partial response of 15% was observed in the assessable 36 PTC patients (32). In 14 (78%) of 18 Tg-assessable PTC patients, Tg declined more than 25%. B-Raf mutation was detected in 17 (77%) of 22 PTCs analysed (32). Four of 10 paired tumor biopsies from PTC patients showed a reduction in levels of VEGFR phosphorylation, ERK phosphorylation, and in VEGF expression during sorafenib therapy (32). No objective response was noted for other tumours. In another sorafenib study in DTC patients, the primary endpoint was reinduction of radioactive iodine uptake at 26 weeks of treatment (33). A 25% partial response was noted however no reinduction of radioiodine uptake at metastatic sites was observed (33).
With regard to axitinib (34) and gefitinib (35), there is one published phase II trial for each, which included all histological types of thyroids cancers. In the axitinib study, there was a partial response of 30% in the whole study group and the most significant side effect was hypertension. In the study with gefitinib which included 25 thyroid cancers with different histologic types there was no objective response and only 12% stable disease (35).
Sunitinib has been used for the treatment of thyroid cancer in two phase II studies to date (36,37). In this study by Carr and colleagues, both MTC and DTC patients were included (36). One complete response and 28% partial response were observed. The other study included only MTC patients, with a partial response rate of 35% (37).
Among TKI, one of the most promising drug seems to be pazopanib which produced in DTC patients a 49% partial response (38). The two patients who died during treatment were reported to have pre-existing contributory disorders.
Probably the most commonly studied TKI in thyroid cancers is vandetanib (39-42). Wells and colleagues reported 20% partial response with 300 mg/day (39) while Robinson and colleagues reported 16% partial response with 100 mg. daily vandetanib (40). Moreover in the former study, in 24 patients, serum calcitonin levels showed a 50% or greater decrease from baseline that was maintained for at least 4 weeks; 16 patients showed a similar reduction in serum carcinoembryonic antigen levels (39). In International Thyroid Congress, September 2010, Leboulleux and colleagues presented a randomised placebo controlled study of vandetanib in DTC patients (41). Statistically significant progress-free survival prolongation was observed for vandetanib vs. placebo (41). However, the objective response rate (8.3% vs. 5.5%) did not reach statistical significance. The only double-blind phase III trial with a TKI was performed with vandetanib in 331 MTC patients (42). Also called ZETA, the results of this study showed significantly increased progress-free survival in the vandetanib group vs. placebo [hazard ratio 0.45, 95% confidence interval (CI) 0.30–0.69, p=0.0001] (42).
In subjects given TKI treatment, moderate to severe toxicity has been noted in 30-40% of the subjects. In their elegant review, Le and colleagues have given a list of side effects reported in more than 20% of patients treated with TKIs (8). According to this list, fatigue, weight loss, anorexia, headache, rash, hand-foot syndrome, diarrhoea, nausea/vomiting, abdominal pain, dyspnea, hypertension are the side effects seen in several of TKIs reported with different rates (8).

CONCLUSIONS
Cytotoxic systemic chemotherapies for advanced, metastatic thyroid carcinomas have limited effectiveness. With significant advances in our understanding regarding the molecular basis of thyroid cancers it seems logical to propose the use of TKIs in thyroid cancers. To summarise the published trials mentioned in this review, highest partial response rates in DTC patients have been observed with pazopanib (38), in MTC patients with vandetanib (42) and in ATC patients with imatinib, although this latter study was able to recruit a small number of patients (27). On the other hand, even in the same pathologic group of tumours, there may be considerable differences between the mutations and molecular changes which may alter the drug response. In a study by Bass and colleagues, biomarkers were studied as predictors of response to treatment of metastatic MTC and DTC with motesanib (43). A certain rate of increase in serum placental growth factor (PlGF) and decrease in soluble VEGF receptor 2 levels after initiation of therapy predicted response to motesanib in patients with advanced DTC or metastatic MTC. Lower baseline VEGF levels were associated with longer progress-free survival. In a recent update of the phase II study with sorafenib there was a correlation between longer progress-free survival and the presence of B-Raf mutations (44). Specifically, for patients with PTC/FTC, the progress-free survival for those with wild-type B-Raf was 54 weeks compared to 84+ weeks for patients with B-Raf^V600E (p = 0.028). Thus, maybe in the future the use of TKIs in thyroid cancers may come along with certain molecular analyses beforehand to ensure treatment response.
As a conclusion, TKIs may open a new era in the treatment of radioactive iodine refractory DTC, advanced MTC and ATC patients in near future. However, the published clinical trials are relatively sparse compared to other malignancies and there is only one published phase III trial yet in thyroid cancers. A possible reason for this is, the difficulty in accrual of enough number of patients to these clinical trials. It may be possible to overcome this difficulty by multi-institutional trials recruiting patients from several centers. On the other hand, given that there is no proof yet TKIs improve overall survival and they have quite significant undesirable effects, patients must be selected carefully before initiation of therapy. Randomised clinical trials for several agents are underway that may lead to eventual approval of TKIs for thyroid cancer.