| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL REVIEW 158 |
Washington Hospital Center, MedStar Research Institute (M.B.-B., M.D.R.), Washington, D.C. 20010; and SEMPR, Serviço de Endocrinologia e Metabologia do Hospital das Clínicas da Universidade Federal do Paraná (M.B.-B.), Curitiba 80.060-240, Brazil
Address all correspondence and requests for reprints to: Matthew D. Ringel, M.D., 110 Irving Street NW, Room 2A-46B, Washington, D.C. 20010. E-mail: matthew.ringel{at}medstar.net.
Abstract
One of the greatest challenges in the management of patients with follicular cell-derived thyroid cancer is the treatment of tumors that progress despite surgery, radioiodine, and T4 suppression of TSH. As knowledge of thyroid cancer biology improves, the potential exists to develop compounds targeted to treat thyroid cancers that do not respond to traditional therapy. Recently, the development of therapies targeted against specific molecular pathways involved in cancer progression has resulted in dramatic responses in patients with chronic myelogenous leukemia, gastrointestinal stromal tumors, and other cancers. A number of compounds are currently being evaluated in clinical trials that alter pathways involved thyroid cancer, and several of these agents have been tested in thyroid cancer in vitro and in vivo. In this review we will discuss the mechanisms of action and preclinical/clinical data for several of these compounds that have the potential to play an important role in the management of thyroid cancer in the future.
THE ABILITY TO retain features of normal thyroid cells, such as iodine uptake, TSH receptor expression, and thyroglobulin production, is used to classify epithelial-derived malignant thyroid tumors into differentiated (papillary and follicular) and undifferentiated (anaplastic) tumors. These differentiated features facilitate monitoring for thyroid cancer recurrence or progression and allow for thyroid-specific radiotherapy and suppression of growth. However, over time some thyroid cancers display a reduced ability to capture iodine, produce thyroglobulin, and/or express the TSH receptor, thereby becoming more difficult to monitor and less responsive to traditional therapeutic modalities. These types of tumors probably account for about 2–5% of all thyroid cancers and cause the vast majority of deaths attributable to thyroid cancer; therefore, they represent a critical area of research for thyroid cancer management.
Alternative therapies have been used in some cases of progressive thyroid cancer. Conventional chemotherapy has proven to be ineffective for most patients with progressive metastases from thyroid cancer. This has been further confirmed in a recent Phase II trial with etoposide, where no response was observed (1). The lack of an effective therapy for tumors that are resistant to radioiodine and TSH-suppressive therapy underscores the importance of developing new options for patients with thyroid cancer. Because endocrinologists are the primary caregivers for patients with thyroid cancer, it is imperative that they become facile with the new directions in oncology research and take an active role in defining the future directions of clinical research in this area.
Over the past 2 decades, the pathways involved in thyroid cancer development and progression have begun to be elucidated. Overexpression and/or uncontrolled activation of receptor tyrosine kinases, downstream signaling molecules, and inhibition of programmed cell death (apoptosis) have all been demonstrated to occur in thyroid cancer. In some cases animal models have clearly demonstrated a direct oncogenic effect on thyroid glands. More recently, the factors responsible for thyroid tumor progression, angiogenesis, and distant spread are also being intensely evaluated.
Several new agents are currently being tested in vitro, in xenograft models, and in clinical studies that are logical choices to consider for thyroid cancer based on thyroid cancer cell biology. In this review we will focus on agents currently in clinical trials in the United States for differentiated thyroid cancer or solid tumors as well as preclinical agents whose mechanisms of action suggest that they might be useful in thyroid cancer. Furthermore, certain current trials for patients with solid tumors other than thyroid cancer will also be discussed. The discussion will be organized both by the biological plausibility for the agent to be active against thyroid cancer and by the availability for enrollment of thyroid cancer patients in clinical trials. By its nature, this review is speculative, as there are no published studies of these agents as therapy for thyroid cancer patients.
Materials and Methods
To include as many available and completed clinical trials in this review, a clinical trial search was performed at the NCI website (www.nci.nih.gov) where details regarding disease status and patient recruitment are displayed. Preliminary unpublished results, when available, were retrieved from the Proceedings of the American Society of Clinical Oncology Meetings (www.asco.org) and in Proceedings of the American Association for Cancer Research Meetings (www.aacr.org). A Medline search was performed to analyze results from concluded trials and preclinical (in vitro and in vivo) studies. Searches were performed using the 1966 to the present databases using the chemical and compound names of all agents identified in the initial search, first cross-referenced with "thyroid cancer" and "thyroid," then using the MESH terms "cancer" and "tumor." Manuscripts were compiled and reviewed. The results of this review are presented below.
Suggested Clinician’s Approach
This review is organized by mechanism of action and the frequency that a particular pathway is abnormal in thyroid cancer, as many of the individual agents will ultimately not be pursued for U. S. Food and Drug Administration approval due to lack of efficacy or unacceptable side effects. Because it is now recommended by the National Cancer Center Network to consider patients with iodine nonresponsive progressive thyroid cancer as candidates for clinical trials, an understanding of both the pathways and the types of agents being evaluated is invaluable to individual clinicians.
Choosing a particular clinical trial for a patient depends on study location and availability, the data supporting activity against thyroid cancer, the medical condition of the patient, the side effects of the study medication, and the mechanism of action. Using these principles, the following hierarchy for choosing a particular clinical trial is suggested: 1) agents proven to have efficacy against thyroid cancer in clinical studies with acceptable side effects; 2) agents in Phase II studies with activity against other solid tumors with preclinical activity against human thyroid cancer; 3) agents in Phase II studies with activity against other solid tumors that are logical choices for thyroid cancer based on mechanism of action, but no preclinical thyroid cancer data available; 4) agents in Phase I studies with preclinical data demonstrating activity against thyroid cancer; and 5) agents in Phase I studies with a mechanism of action that suggests clinical activity against thyroid cancer would be possible.
Results and Discussion
Drugs that target signaling molecules
Like many cancers, thyroid cancer is characterized by genetic alterations that result in dysregulated cell growth and death. Several compounds currently being tested in preclinical and clinical studies target intracellular molecules involved in these processes. Complex interactions between pathways make many of these compounds active in more than one pathway. These agents are not tumor specific, as these pathways are active in normal and malignant cells, but are thought to be tumor selective because the cancers demonstrate higher levels of pathway activation, making them more sensitive than normal cells at lower concentrations. This factor may lead to significant toxicity that may ultimately reduce clinical utility. This section will focus on compounds that act on pathways involved in cell growth, apoptosis, and angiogenesis, with a focus on those tested for activity in thyroid cancer.
Ras pathway
Ras activation is central to the pathogenesis of some thyroid cancers, and it can occur through mutations in the genes encoding Ras or through activation of upstream regulators. Because of this independent role for Ras separate from receptor-mediated activation, and because of its important role in thyroid cancer, the Ras pathway will be considered as a separate entity from other tyrosine kinase-regulated pathways.
Ras is a small GTP-binding protein (G protein) regularly expressed in normal thyroid cells (2), and its protein product is involved in several important functions, including proliferation, differentiation, and cell survival (Fig. 1
). In normal cells Ras is largely activated by receptor tyrosine kinases; these receptors autophosphorylate at specific residues after the appropriate ligand binds its extracellular domain (in some cases receptor dimerization is required), leading to activation of several signaling pathways. For Ras activation, specific docking sites for an adaptor protein, GRB2, must be available in the activated receptor tyrosine kinase. The interaction of GRB2 with an activated receptor tyrosine kinase leads to binding of another protein known as son of sevenless (SOS), which subsequently activates Ras by altering its conformation, allowing a GTP to be exchanged for the stably bound GDP. GTP-bound Ras then phosphorylates its targets, resulting in a cascade of events leading to cellular proliferation. Ras is then inactivated by hydrolyzing the GTP, allowing for a new GDP to bind. In this manner, Ras is then poised for another cycle of activation in the GDP-bound state.
|
In addition to activating mutations, Ras overactivation can occur secondary to receptor overactivation. Enhanced signaling of receptor tyrosine kinases is a common event in thyroid cancer, particularly papillary thyroid cancer. Thus, when considering Ras as a therapeutic target for thyroid cancers, it appears to be relative thyroid tumor specific, occurs in follicular cancer, and also occurs in papillary cancer, albeit by distinct mechanisms. For these reasons, Ras is a reasonable molecular target to consider for novel forms of thyroid cancer therapy.
Ras-directed therapy
Ras antisense compounds.
Antisense compounds are small synthetic DNA sequences of up to 25 oligonucleotides comprised of a complementary sequence to a particular targeted mRNA. When bound to mRNA, these drugs serve as substrates for ribonuclease H that cleaves the mRNA strand, but not the antisense compound. These substances also interfere with ribosomal assembly, blocking gene expression and inhibiting protein synthesis (Fig. 2). Antisense compounds designed to inhibit the expression of cell signaling molecules have been synthesized and are being tested as potential drugs for the treatment of cancer. Two antisense drugs, ISIS 2503 and ISIS 5132, target elements from the Ras pathway. Both are phosphorothiotate oligonucleotides in which the presence of oxygen in the phosphodiester bonds is substituted by sulfur, thus conveying nuclease resistance and enhancing bioavailability.
|
Phenylacetate. Phenylacetate is a naturally occurring aromatic fatty acid product of phenylalanine metabolism that has been evaluated as an antineoplastic drug in several preclinical studies. Although the mechanisms of action of this compound in malignant cells are not yet clear, it may inhibit cell growth by affecting posttranslational processing of Ras. In laboratory studies phenylacetate decreases the cell growth rate and induces dedifferentiation in several human cancer cell lines in vitro and in vivo (14, 15, 16, 17). In a Phase I study of iv infusion twice a day for 2 wk, a partial response in two patients (refractory malignant glioma and prostate cancer) (18) was noted. Side effects were limited; however, a low response was observed in a Phase II trial involving patients with recurrent malignant glioma (19).
In human thyroid cancer cell lines (follicular, papillary, and Hürthle cell carcinoma), phenylacetate decreases TSH- and non-TSH-induced growth and induces increased radioiodine uptake and thyroglobulin secretion (20). Moreover, vascular endothelial growth factor (VEGF) secretion was inhibited in all cell lines, suggesting additional interference with angiogenesis. In follicular cancer cell lines, phenylacetate was also demonstrated to act in cooperation with all-trans-retinoic acid to reduce the growth rate (20). This agent may, therefore, warrant further study in patients with thyroid cancer, either alone or in combination with other agents.
Farnesyl transferase inhibition. Translocation of activated Ras to the cytoplasmic membrane is a key step its activation. For this step, several posttranslational modifications must occur to allow membrane localization. The initial modification step involves farnesylation (i.e. addition of a farnesyl moiety to a cysteine residue) of Ras by the enzyme farnesyl transferase. Inhibition of farnesyl transferase has been shown to inhibit membrane accumulation of Ras in vitro and therefore reduce Ras signal transduction (21). Currently, four farnesyl transferase inhibitors have been developed and are in clinical studies (R115777, L-778,123, SCH66336, and BMS-214662). Another inhibitor, manumycin, has been used in laboratory studies, but it has not been evaluated in clinical trials due to toxicity in animal models. Finally, it is noteworthy that protein other than Ras that are similarly involved in cell signaling also undergo farnesylation and are, therefore, inhibited in the presence of farnesyl transferase inhibitors.
Manumycin, a natural product of streptomyces with antibiotic properties, has been previously shown to possess antineoplastic properties in vivo through farnesyl transferase inhibition (22, 23). In vitro, manumycin was demonstrated to reduce the number of living cells from six anaplastic thyroid cancer cell lines with variable sensitivity, either alone or in combination with paclitaxel, doxorubicin, or cisplatin (24). The combination of manumycin plus paclitaxel also inhibited angiogenesis in a mouse xenograft model of anaplastic thyroid carcinoma, suggesting that it may be useful clinically in thyroid cancer (25).
R115777 (tipifarnib, Zarnestra, Johnson & Johnson, Raritan, NJ) has been extensively evaluated, both alone and in combination with other agents, in Phase I and II clinical studies of patients with hematological malignancies and solid tumors and is currently being investigated in Phase III trials. A Phase I trial of R115777 in combination with topotecan, a topoisomerase I inhibitor, is open and recruiting patients with advanced solid tumors, including thyroid cancer.
The pharmacokinetics properties and dose toxicity of L-778,123 (Merck \|[amp ]\| Co., Inc., West Point, PA) have been evaluated in a Phase I trial that involved 25 patients with advanced solid carcinomas (1 patient with thyroid carcinoma) (26). Effective farnesylation inhibition was confirmed by inhibition of HDJ2 prenylation in peripheral blood, but significant side effects were observed with the highest dose evaluated. In another Phase I trial, L-778,123 was combined with radiotherapy to treat patients with advanced lung and head and neck cancer. In this study two of three patients with head and neck cancer showed a complete response (27); however, further studies evaluating this agent have been discontinued due to toxicity.
A Phase I study evaluating SHC66336 (lonafarnib, Sarasar, Schering-Plough Corp., Kenilworth, NJ) in patients with localized or metastatic cancer demonstrated more mild side effects. No studies have been conducted in vitro or in vivo in patients with thyroid cancer. Current open Phase I and II trials are recruiting pediatric patients with brain tumors and adult patients with urinary tract carcinoma to be treated with SHC 66336 in combination with gemcitabine.
Two Phase I clinical trials are ongoing for patients with solid tumors, including thyroid cancer, to evaluate dose-related toxicity of BMS-214662. In another Phase I trial, patients with advanced tumors that express Her2/neu receptor, which is commonly overexpressed in papillary thyroid tumors, are being recruited to be treated with BMS-214662 in combination with herceptin, a human humanized monoclonal antibody that binds to the Her2/neu receptor.
Raf inhibition.
Many cellular responses that follow Ras activation occur via its downstream effector, Raf (Fig. 1
). Three isoforms of Raf have been identified, A-Raf, B-Raf, and C-Raf (Raf-1). Activated (GTP-bound) Ras localizes Raf to the membrane, and once activated, Raf phosphorylates MAPK kinase (MEK), initiating a cascade of events resulting in cell growth and reduced cell death. The antisense compound designed to inhibit c-Raf, ISIS 5132 (code CGP69846A, Isis Pharmaceuticals, Inc.) has been demonstrated to inhibit the growth of malignant cells derived from several different tumors in vitro and in vivo (28), inducing apoptosis and cell death (29). Furthermore, several chemotherapeutic agents in xenograft models enhanced the antiproliferative effects of ISIS 5132 (30). Initial Phase I studies demonstrated minimal side effects (31, 32), with a more than 50% reduction of c-Raf activity (33); however, further clinical Phase II evaluation failed to demonstrate disease improvement in patients with colorectal, lung, and prostate cancer (34, 35, 36). Although this compound may be active against thyroid cancer cells, this has not yet been tested.
BAY 43-9006 is a small molecule that specifically inhibits Raf kinase and was demonstrated to inhibit cell proliferation and bear cytostatic activity in preclinical models. In a Phase I study involving patients with multiple malignancies, BAY 43-9006 was able to induce a partial response in a few patients with mild to moderate toxicity. Currently, this drug is under investigation in combination with other chemotherapeutic agents.
MEK inhibition.
As noted above, MEK is a key protein involved in signaling for several oncogenic pathways, including the Ras-activated pathway (Fig. 1). Two isoforms, MEK1 and MEK2, have been identified, and phosphorylation of both can be inhibited by the compound CI-1040 (also known as PD-184352). This drug is administered orally and has been shown to inhibit the tumor growth of colon carcinoma in mice (37). Initial clinical studies, presented in abstract form only, have shown association with mild side effects at concentrations capable of inhibition of MEK activity in peripheral blood cells (38) and in tissue samples (39). The agent is currently in Phase II trials for patients with lung, colon, breast, and pancreatic cancer.
Drugs that target receptor tyrosine kinases (Fig. 3)
Receptor tyrosine kinases are activated by a wide variety of ligands, are frequently mutated in a manner that results in constitutive activation, and are the most commonly overexpressed receptors in thyroid cancers. In follicular cell-derived thyroid cancers, genetic rearrangements resulting in the expression of chimeric proteins involving the tyrosine kinase domain of RET are common in radiation-related and sporadic papillary thyroid cancer. In addition, overexpression of other receptor tyrosine kinases, including fibroblast growth factor (FGF), epidermal growth factor (EGF), hepatocyte growth factor (c-Met), VEGF, insulin, and IGF-I receptors are commonly identified. Several of these are common to many cancers, are expressed at very low levels in nonneoplastic tissues, and may be associated with angiogenesis or progression, making them excellent therapeutic targets.
|
The formation of new blood vessels is a crucial step in determining tumor expansion and is greatly dependent on proangiogenic factors that are produced in a paracrine fashion by tumor cells undergoing hypoxia or mechanical compression. Several growth factors are involved in the process of angiogenesis in malignant tumors; among them, VEGF appears to be the most prominent. Besides the functional activity of stimulating vascular proliferation and permeability and inducing metastasis, VEGF may function as an apoptotic protector for the newly formed vessels via the phosphatidylinositol 3-kinase/Akt signaling pathway (40). Significantly increased levels of VEGF have recently been demonstrated in the serum of patients with well differentiated metastatic thyroid tumors compared with lower levels found in patients considered to be in a complete remission (41). Similar results in thyroid tumors have been previously demonstrated by immunohistochemistry (42), where high levels of VEGF have been associated with the occurrence of metastasis (43) and possibly also with a worse prognosis (44). In patients with other malignancies, a worse prognosis was observed in those who expressed higher levels of VEGF in their tumors, probably due to increased vessel formation and development of metastasis (45). Although two VEGF tyrosine kinase receptors (VEGF-1 and VEGF-2) have been identified, VEGF-2 is considered to be the dominant receptor for signal transduction pathway stimulation.
EGF receptor (EGFR)
Four structurally related receptors (Her1, ErbB1, or EGFR; Her2/neu or ErbB2; Her3 or ErbB3; and Her4 or ErbB4) are part of the EGFR superfamily. When bound to ligands, these receptors either homodimerize or heterodimerize, leading to subsequent signal transduction. Her2/neu is the preferred heterodimerization partner for EGFR, and both EGFR and Her2/neu have been implicated in thyroid cancer progression.
The EGFR is commonly expressed in differentiated thyroid tumors (46), and its overexpression has been associated with a worse prognosis (47). Several high affinity ligands of the EGFR have been identified, but EGF and TGF
are apparently the most prominent. Once bound to a ligand, the EGFR triggers pathways that lead to cell cycle progression and apoptosis. EGFR blocking leads to cell cycle arrest in G1, apoptosis, antiangiogenesis, and down-regulation of matrix metalloproteinase, resulting in a decreased incidence of metastases.
Gene amplification of Her2/neu has also been detected in various solid tumors and has been correlated with a poor prognosis. In mouse fibroblasts, overexpressed Her2/neu has been shown to act as a potent oncogene (48). Although several ligands have been identified, the kinase activity of Her2/neu can be activated without any ligand when overexpressed, homodimerized, or heterodimerized (49). In differentiated thyroid tumors, Her2/neu has been demonstrated to be up-regulated, particularly in papillary thyroid cancer (50).
Considering the important role of both EGF and VEGF receptors in the development and progression of malignant tumors, significant attention has been dedicated to new drugs that potentially block pathways related to these receptors. Currently, it is possible to pharmacologically interrupt the activation of these receptors at different levels through neutralization of the ligand with antibodies, blocking the receptor with small molecules, or inhibiting mRNA with antisense compounds.
Antibodies
Anti-VEGF antibodies (recombinant human monoclonal antibody against VEGF). Monoclonal antibodies directed against VEGF have been demonstrated to neutralize VEGF (51) and reduce angiogenesis in several human xenograft models (52). Initial results from a Phase I study to evaluate bevacizumab (Avastin, Genentech, Inc., South San Francisco, CA) demonstrated only mild to moderate side effects in patients with advanced cancer (53). Currently, several Phase I and II trials are open and recruiting patients with different malignancies for treatment with bevacizumab alone or in combination with other chemotherapeutic agents or radiotherapy. Preliminary abstract results show minimal side effects with stabilization or, less frequently, regression of disease in patients with metastatic renal cancer (54) and advanced solid tumors (55). A Phase II trial of bevacizumab in association with fluorouracil, hydroxyurea and radiotherapy is currently open for patients with advanced head and neck cancer.
Anti-EGFR antibodies (monoclonal antibodies 528 and 4253). Monoclonal antibodies have been designed to bind exclusively to the extracellular domain of the EGFR, causing internalization of the receptor and thereby reducing its availability. The combination of monoclonal antibodies and chemotherapy or radiotherapy has been shown to act against several human tumor cell lines in xenograft models. In two studies that evaluated the use of anti-EGF monoclonal antibody in thyroid cancer cell lines, a reduction in tumor growth was noted (56, 57). Clinical trials with these particular anti-EGFR monoclonal antibodies are currently being designed.
Anti-Her2/neu antibodies. Herceptin (code 4D5, Trastuzumab, Genentech, Inc.) is a humanized anti-Her2/neu monoclonal antibody that binds with high affinity to Her2/neu and inhibits the growth of Her2/neu-expressing cancer cells and xenograft models (58). Several Phase I/II trials are currently recruiting patients with solid malignancies to be treated with herceptin in association with other chemotherapeutic agents. Thyroid cancer has not been tested in particular.
Receptor antagonists
VEGF receptor inhibition (VEGFR-2 antagonists). SU5416 (Semaxanib, Sugen, Inc., South San Francisco, CA) is a selective VEGF receptor 2 antagonist that has been shown to inhibit tumor growth (59) and metastases (60) in xenograft models for several solid malignancies and is being tested in active trials for patients with several malignancies. Two studies of SU5416 in combination with paclitaxel are recruiting patients with locally advanced metastatic cancer of the head and neck and patients with advanced malignancies in general.
Small molecules
ZD1839 (Iressa, AstraZeneca Pharmaceuticals, Newark, NJ). The kinase activity of the EGFR can be blocked by small molecule tyrosine kinase inhibitors that interfere with ATP binding to the receptor. ZD1839 (Iressa) has been demonstrated to inhibit growth in a wide range of tumor cell lines and human tumor xenografts. It has been widely evaluated in preclinical and Phase I and II studies and showed promising results when associated with conventional chemotherapy. Although no studies have been conducted to evaluate the effects of ZD1839 in thyroid cancer cell lines or in clinical trials, a Phase I study is currently recruiting children with refractory solid tumors for therapy with ZD1839 alone. A similar agent has been demonstrated to also inhibit signaling of activated RET found in medullary thyroid cancer in vitro. The concept that the small molecule inhibitors of receptor tyrosine kinases may not be completely specific requires further investigation in thyroid cancers with RET rearrangements.
ZD6474. ZD6474 was designed to be an orally available alternative small molecule inhibitor of the VEGF receptor family tyrosine kinase domain. Wedge et al. (61) reported that this agent is orally absorbed and exhibited dose-dependent inhibition of tumor growth, VEGF signaling, and angiogenesis in animal xenograft models. This agent, however, is not entirely specific for the VEGF receptor family. It also inhibits the EGFR, TIE-2, and was recently demonstrated to inhibit the family of RET oncoproteins, including RET/MEN2A, RET/MEN2B, and RET/PTC3, in vitro and in vivo (62). These exciting in vitro and animal data suggest potential activity against medullary and papillary thyroid cancer.
Drugs that target angiogenesis through other pathways
Thalidomide. Thalidomide, originally designed as a sedative drug, has long been known to demonstrate antineoplastic properties in animal tumor models, but it was only recently that attention was given to thalidomide in cancer therapy, after it was found to have antiangiogenic properties (63). Although the mechanisms by which thalidomide exerts growth inhibition on new vessels are not yet fully understood, it is currently being tested in several clinical trials for the treatment of different malignancies. A Phase II trial recruiting patients with metastatic follicular, papillary, or medullary thyroid carcinoma is ongoing to evaluate drug toxicity and response in radioiodine-resistant tumors. Preliminary data indicate that thalidomide has promise in reducing disease progression (Celgene Corp., Warren, NJ; press release 11/02); confirmation of this early report is pending.
Combretastatins. Combretastatins are a family of tubulin-binding proteins derived from the African willow with unique vascular-targeting properties. The agents appear to be most active against endothelial cells, although direct cytotoxic effects have been demonstrated against cancer cells, including thyroid cancer cell lines (64). Although there are several combretastatins under evaluation, combretastatin A4 phosphate is the compound that has been most carefully tested to date.
A Phase I study of combretastatin A4 phosphate was recently completed that included 3 patients with anaplastic thyroid cancer limited to the neck and 2 patients with metastatic medullary thyroid cancer (65). One patient with anaplastic thyroid cancer demonstrated a complete response, whereas disease progression was halted in 1 individual with metastatic medullary thyroid cancer. Two other patients with anaplastic thyroid cancer were treated in the study; no clinical responses were noted, but 1 had acute pain at the site of bulky disease. The major toxicity of the medication was cardiac, with prolongation of the QT interval or arrhythmia in 6 of 25 patients, and myocardial infarction in 1 patient. These data demonstrate that this class of agents may be useful in the management of patients with poorly differentiated thyroid cancer, but careful toxicity evaluation of combretastatin A4 or other members of this group of compounds will be required.
Drugs that target Akt/mammalian target of rapamycin (mTOR)
Several critical cellular functions are controlled by the mTOR (also known as FRAP, RAFT1, and RAPT1), a serine/threonine member of the phosphatidylinositol 3-kinase-activating cascade. Upon activation, mTOR signals to increase cell cycle progression and cell growth. Several targets of mTOR have been found to be dysregulated in thyroid cancer, such as the cell cycle stimulators c-Myc and cyclin-D1 (66) and the cell cycle inhibitor p27kip1 (p27) (67), suggesting a potential role for mTOR in thyroid cancer progression (Fig. 4). In particular, c-Myc and cyclin D1 are overexpressed in malignant thyroid tissues (68), and the level of expression correlates with tumor aggressiveness (69). mTOR is under control by another important kinase, Akt, which is elevated in several different types of cancer, including anaplastic, papillary, and follicular thyroid cancer (70), and may be an important regulator of mTOR in these cancers. These data suggest that mTOR is an appropriate target for thyroid cancer therapy.
|
Drugs that target apoptotic pathways (programmed cell death)
Thyroid cancer cells, like many other cancer cells, demonstrate reduced sensitivity to cell death. This ability to resist cell death is thought to lead to the ability of cancer cells to sustain genetic alterations, but continue to grow. Apoptosis is an orderly process leading to cell death through specific signaling pathways. This process can be initiated via intracellular and extracellular stimuli. Several new agents designed to initiate apoptosis in cancer cells are currently being investigated, some of which may have particular utility for thyroid cancer.
TNF-related apoptosis-inducing ligand (TRAIL; recombinant soluble Apo2 ligand).
TRAIL is a member of the TNF family that is expressed in most benign and malignant tissues and induces apoptosis through activation of the caspase pathway via activation of specific receptors (Fig. 5). Compared with Fas ligand, another apoptosis-inducing protein, TRAIL appears to induce apoptosis primarily in cancer cells, suggesting that it might be a potential therapeutic agent for cancer-specific therapy.
|
TRM-1 is a human monoclonal antibody that binds with high affinity to the TRAIL-R1 receptor, one of the two TRAIL receptors, to induce apoptosis. In vitro, TRM-1 has been shown to induce cell death in several cancer cell lines, and its effect may be increased by association with chemotherapeutic agents (83). However, the efficacy of TRM-1 has not yet been reported in animal models or clinical trials.
Bcl-2 inhibition. Phosphorylation of Bcl-2 is associated with resistance to apoptosis, and overexpression of Bcl-2 leads to cell proliferation in the absence of growth factors, including in thyroid cancer cells. Overexpression and overactivation of Bcl-2 have been shown to be increased in several cancer cell lines in vitro; treatment with a Bcl-2 antisense compound, G3139 (oblimersen sodium, Genasense, Genta, Inc., Berkeley Heights, NJ) (84, 85), resulted in cell death and reduced Bcl-2 expression and activity. In a Phase I trial that involved patients with solid malignancies, G3139 administration was associated with mild side effects, particularly fatigue (86). In other studies chemotherapy sensitivity was increased by Bcl-2 mRNA antisense treatment in patients with melanoma (87) and prostate carcinoma (88). Several Phase I/II/III trials are currently open, recruiting patients with other malignancies to be treated with G3139 in association with conventional chemotherapy. This agent, although logical for thyroid cancer, has not been tested in vitro.
Other potential agents for thyroid cancer
Cyclooxygenase-2 (COX-2) inhibitors. Overexpression and overactivation of COX-2, a key enzyme in the synthesis of prostaglandins, occur frequently in many types of cancers, including differentiated thyroid carcinomas (89). When activated, COX-2 inhibits apoptosis and enhances angiogenesis (90). Therefore, COX-2 inhibitors have been considered potential therapeutic agents for the management of different malignant conditions. In vitro, COX-2 inhibitors induce growth arrest and inhibit tumor formation in xenograft models (91). Several trials are currently open to recruit patients with malignancies, including thyroid cancer, for treatment with celecoxib, a COX-2 inhibitor, alone or in combination with chemotherapeutic agents.
The 90-kDa heat shock protein (Hsp90) inhibitors. Hsp90 protein is a chaperone molecule involved in activation and stabilization of critical proteins in signal transduction pathways. Specifically, the serine/threonine kinases Raf1 and Akt are proteins that are dependent on Hsp90 for stabilization and localization. Blockade of Hsp90 results in enhanced degradation of these signaling molecules, causing decreased activation. Because the activation of Akt and Raf is implicated in thyroid cell growth and thyroid cancer, inhibition of Hsp90 is a logical consideration for thyroid cancer therapy.
Geldanamycin and its related compound, 17-N-allylamino-17-demethoxy-geldanamycin (17-AAG), inhibit Hsp90by binding to its ATP-binding domain. In vitro, both have been shown to induce growth arrest in several cancer cell lines. 17-AAG inhibits cellular growth and induces apoptosis in thyroid cancer cell lines cultures through depletion of Akt and Raf and by inducing apoptosis (our unpublished data). Phase I and II studies are currently recruiting patients with solid malignancies for treatment with 17-AAG.
Demethylating agents. Hypermethylation of DNA occurs frequently in the promoter regions of genes, resulting in altered binding of cofactors leading to altered (usually reduced) gene expression. The enzyme methyltransferase transfers a methyl group to cytosine rings in CpG islands (sequences rich in cytosine and located in the 5' regulatory region of many genes), thereby reducing binding of transcription factors to the promoters of regulated genes. Blockers of methyltransferase have been used to induce reexpression of tumor suppressor genes and other genes important for facilitating therapy or reducing cell growth.
5-Azacytidine and 5-aza-2'-deoxycytidine are compounds that are incorporated into DNA and are able to inhibit methylation. In thyroid cancer cell lines, hypermethylation of the sodium-iodide symporter (NIS) gene promoter has been demonstrated, and treatment with 5-azacytidine restores NIS expression as well as the expression of several important thyroid cell transcription factors, resulting in enhanced iodide uptake in some cell lines (92). Currently, a Phase I study is open to patients with advanced solid tumors for treatment with 5-azacytidine and phenylbutyrate, although toxicities and inhibition of activity have been reported. The concept of inhibiting methyltransferase activity is attractive for thyroid cancer, particularly for patients with tumors that no longer effectively concentrate iodine.
Histone deacetylase inhibitors. Histones are small positively charged proteins that compose most of the protein structure of the chromosomes, binding tightly to the negatively charged DNA, forming a condensed protein-DNA complex. Relaxation of this dense structure via modification of histones by acetyl transferase and histone deacetylase (HDAC) is required to allow for transcription of DNA into mRNA. Disruption of this process using HDAC inhibitors has been demonstrated to induce cell cycle arrest and differentiation; however, the mechanisms mediating this activity remain unclear. In vitro, incubation of follicular and anaplastic thyroid cancer cells with the HDAC inhibitor depsipeptide (code FR901228) was shown to increase the expression of thyroglobulin and NIS, with consequent intracellular iodine accumulation (93). Also, HDAC inhibitors are able to induce apoptosis in several cancer cell lines, probably through derepression of specific cell death genes. Phase I studies of similar agents have been performed with modest response rates and toxicities (despite the nonspecific nature of the mechanism).
Proteasome inhibitors. PS-341 is a selective inhibitor of the 26S proteasome, an important protease in the ubiquitin-proteasome pathway involved in the process of elimination of damaged proteins and in other functions related to apoptosis and cell cycle progression. Inhibition of the 26S proteasome in cancer cell cultures is related to growth arrest, inhibition of angiogenesis, and enhanced radiosensitivity and chemosensitivity (93). A Phase I trial to evaluate the combination of PS-341 and paclitaxel or doxorubicin in patients with advanced solid tumors is currently open. A planned Phase I trial of PS-341 in combination with fludarabine and Phase II monotherapy will recruit patients with thyroid cancer.
Although proteasome activation is likely to be important in thyroid cancer cell biology, this particular agent has not been tested in vitro in thyroid cancer cells. Side effect profiles of such a generally active agent will require careful evaluation in Phase I studies as proteosome abnormalities may be involved in several diseases, including Parkinson’s disease.
Gene therapy
Gene therapy has been used in the laboratory for decades to induce the expression of genes not normally expressed in particular cells, to induce reexpression of silenced genes, or to inhibit the expression of abnormal genes. This process has been used in animal models in cancers to induce cell death (suicide gene therapy) or to enhance the therapeutic effect of other agents. The possibility of reexpressing specific genes that are mutated or inactivated in malignant cells in patients represents a promising field in oncology research. Initial clinical studies have confirmed adequate intratumoral gene delivery with the use of direct injection of cDNAs or viral vectors containing cDNA encoding the genes of interest. Systemic therapy of metastatic disease has been more problematic due to host immune response to the vector and the first pass effect in liver. Several genes that have been evaluated in thyroid cancer or that might have particular utility in thyroid cancer will be discussed below.
p53. The p53 protein is a critical regulator of cell cycle progression; its activation allows for repair of DNA mismatches that can occur normally over time or in response to some external event, such as exposure to radiation. When enough damage occurs, p53 activates a cascade of events that results in apoptosis. Mutant p53 proteins can be caused by missense mutations in the coding regions of the p53 gene in either one (heterozygous mutation) or both (homozygous mutation) alleles. Heterozygous mutations can result in reduced function of the normal p53 by binding to the normal protein or by inhibiting activity directly (dominant negative effect). Homozygous mutations result in the production of p53 proteins with reduced or absent activity. In both cases, inhibition of normal p53 activity results in more rapid cell cycle progression, thus not allowing for appropriate DNA repair or apoptosis in response to cellular damage or aging.
Inactivating mutations of p53 are among the most common gene mutations found in human malignancies. Malignant cells bearing wild-type p53 are typically more susceptible to a broad range of chemotherapeutic agents compared with cells expressing mutant forms of p53. In addition, p53 mutations are generally more common in poorly differentiated cancers in most forms of solid tumors. Similar to those in other cancers, p53 mutations in thyroid cancer are more frequently found in anaplastic compared with well differentiated cancers, although all tumor subtypes can harbor a p53 mutation. A direct role for p53 mutation in thyroid cancer is supported by data demonstrating that reexpression of wild-type p53 in a thyroid cancer cell line harboring a p53 mutation leads to growth arrest (94). Moreover, p53-dependent chemotherapy sensitization has been demonstrated in another thyroid cancer cell line (95). Although no clinical studies are currently being conducted in patients with thyroid cancer, patients with other malignancies are being evaluated with intratumoral, ip, or iv vector injections (96).
NIS. Another potential clinical application for gene therapy in thyroid carcinomas is restoration of iodine uptake in thyroid cancers by reexpression of the NIS. A functional NIS protein is fundamental for iodide incorporation in benign and differentiated malignant thyroid cells. Defective iodine uptake may be consequent to hypermethylation of the NIS gene promoter (97), altered subcellular localization of the protein (98), or reduced NIS gene expression by other mechanisms (95). In preclinical studies, successful induction NIS reexpression has been demonstrated using a gene therapy approach in several malignant cell lines (99), including the follicular thyroid cancer cell line, FTC133 (100). In vivo, increased iodide uptake was demonstrated in a xenograft model of human NIS-transfected FTC133 (101). This approach has also been used to induce iodine uptake into nonthyroidal tumors, such as prostate and breast carcinomas (102).
Suicide gene therapy. The use of gene therapy, leading to the expression of proteins in cancer cells that are either directly toxic or that induce the sensitivity of cancer cells to particular medications, is the goal of suicide gene therapy. One well studied method is to induce the expression of the viral enzyme thymidine kinase (TK) in target cells by creating a construct in which the gene encoding TK is controlled by a cell type-specific promoter (such as the thyroglobulin promoter), resulting in the expression of TK only in cancer cells. Once expressed, TK is able to activate the antiviral drug, ganciclovir, resulting in DNA strand breaks and subsequent cell death. Because mammalian cells normally do not express TK, only cells that express the enzyme will respond to the drug. Therefore, treatment of patients with ganciclovir after administration of gene therapy could result in cancer-specific cell death.
Suicide gene therapy has been successfully demonstrated in vitro and in xenograft models of several follicular cell-derived and medullary thyroid cancer cell lines. In follicular cell-derived thyroid cancer cell lines (WRO, follicular; FRO, anaplastic), the expression of TK leads to increased sensitivity to ganciclovir and enhances the efficacy of 
-radiation (103). Also, direct injection of tumor xenografts with a retrovirus carrying TK-thyroglobulin promoter sequence has been shown to induce ganciclovir sensitivity to these cells (104). This approach has not yet been used in patients, but is a promising avenue for future therapy.
Conclusions
Current therapeutic measures for patients with progressive thyroid cancer are relatively ineffective, and there is a clear need to develop new alternatives for their management. In this review we have attempted to summarize a few of the most exciting areas in cancer research that are likely to result in advances in the management of these patients over the next few years based on preclinical and clinical data or on the mechanism of action. Treatment areas, such as immunotherapy or tumor vaccines, were not discussed in detail due to a paucity of data for thyroid cancer, but may also be important alternatives for patients in the future. Enrollment of patients with thyroid cancer into Phase I and II studies will be needed to define the best therapeutic options. Endocrinologists, as the primary care providers for these individuals, need to take an active role in this process to advance the care of thyroid cancer patients in the future.
|
|
|
This work was supported by grants from the American Cancer Society (RSG CNE-103291) and the NIH (CAN8339479), to M.D.R.
Abbreviations: 17-AAG, 17-N-allylamino-17-demethoxy-geldanamycin; COX-2, cyclooxygenase-2; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; HDAC, histone deacetylase; Hsp90, 90-kDa heat shock protein; MEK, MAPK kinase; mTOR, mammalian target of rapamycin; NIS, sodium-iodide symporter; TK, thymidine kinase; TRAIL, TNF-related apoptosis-inducing ligand; VEGF, vascular endothelial growth factor.
Received November 26, 2002.
Accepted February 11, 2003.
References
