New Horizons: Emerging Therapies and Targets in Thyroid Cancer

Abstract

The treatment of patients with progressive metastatic follicular cell-derived and medullary thyroid cancers that do not respond to standard therapeutic modalities presents a therapeutic challenge. As a deeper understanding of the molecular drivers for these tumors has occurred and more potent and specific compounds are developed, the number of Food and Drug Administration (FDA)-approved treatments for thyroid cancer has expanded. In addition, with the advent of disease-agnostic target-directed FDA approvals an ever-broadening number of therapeutic options are available for clinicians and patients. However, to date, complete remissions are rare, the average durations of response are relatively modest, and toxicities are common. These factors accentuate the need for further understanding of the mechanisms of resistance that result in treatment failures, the development of biomarkers that can improve patient selection for treatment earlier in the disease process, and the continued need for new therapeutic strategies. In this article, recent approvals relevant to thyroid cancer will be discussed along with selected new potential avenues that might be exploited for future therapies.

Metastatic I-131 nonresponsive progressive thyroid cancer and progressive metastatic medullary thyroid cancer (MTC) are relatively rare entities that present a therapeutic challenge. The clinical course can be highly variable, ranging from long-term stability that can last for decades to early rapidly progressive disease. In addition, some slowly progressive and/or stable tumors can evolve into a more progressive state over time. The early clinical courses often can be predicted by the histological subtypes and rates of growth in the primary tumor in the thyroid, as evidenced by patients with anaplastic thyroid cancer (ATC). Traditionally, only patients in a progressive phase of disease or those with symptoms are considered for systemic treatments beyond standard therapies due to the nondurable responses and toxicities associated with available compounds (1-3). With the expansion of more effective and better tolerated options, the development of new combinatorial strategies, and the advent of more robust biomarkers that may enable use of disease agnostic but target-based therapies, there are new opportunities to consider the types of tumors appropriate for these treatments and the timing of their use. In this review, there will be a focus on recent advances in thyroid cancer therapy and selected opportunities for new approaches.

ATP-competitive and Allosteric Kinase Inhibitors

Multikinase inhibitors and RET inhibitors

Most of the Food and Drug Administration (FDA)-approved medications for differentiated thyroid cancer (DTC), ATC, and MTC competitively block 5′-adenosine triphosphate (ATP) binding and thereby inhibit the activities of protein kinases with varying degrees of specificity (Fig. 1) (4). In DTC, the multikinase inhibitors (MKIs) sorafenib and lenvatinib achieved FDA approval based on improved progression-free survival rates that last on average 12 to 24 months with similar results reported in postapproval clinical experience (5-8). In both cases, the clinical antitumor activities do not correlate with the presence of specific mutations, thus molecular testing does not predict clinical responses. In MTC, initial FDA approvals were for the MKIs vandetanib and cabozantinib, which include RET among their kinase targets. These compounds are active against RET-mutated and RET wildtype MTCs in clinical trials, although 1 post hoc analysis suggests enhanced activity of cabozantinib in MTCs with a RET^M918T mutation (9, 10). In addition, drug-resistant mutations in RET have been reported to be induced with therapy, suggesting that anti-RET activity may play a role in the antitumor activity of this MKI (11, 12). More recently, 2 potent and much more selective second generation inhibitors of RET, selpercatinib (LOXO-292) and pralsetinib (BLU-667), have been studied in RET-mutated MTC, and in DTCs (papillary thyroid cancers [PTC]) and lung cancers with RET rearrangements. Selpercatinib was FDA approved in 2020 for the aforementioned tumors and both compounds have high overall response rates in RET-mutated and RET-rearranged thyroid cancer with relatively fewer toxicities (12, 13). While both compounds are active also versus the classical “gatekeeper” mutants, MTCs that develop resistance to selpercatinib develop mutations in the solvent front of RET in trans (eg, G810S) that can block access to the ATP binding site (14, 15). Non–RET-mutated MTCs most commonly harbor mutually exclusive activating mutations in the RAS oncogenes, potential strategies for targeting these tumors more specifically are discussed below.

Figure 1.

Therapeutic targets in thyroid cancer. Thyroid cancer are characterized by mutations (*) or rearrangements (**) in a number of genes that result in enhanced activity of signaling pathways leading to cancer development and progression. Mutational activation occurs at the level of receptor tyrosine kinases (eg, RET) and in signaling molecules such as BRAF, Rat Sarcoma (RAS), PI3K, and AKT. Chromosomal rearrangements that lead to fusion proteins linking the promoter and exons of 1 gene to the kinase domain of a tyrosine kinase (eg, RET, NTRK, anaplastic lymphoma kinase) or serine threonine kinase (eg, BRAF) result in overexpression and activation of the wildtype kinase in the cell cytoplasm. Additional therapeutic targets include proteins that regulate the immune checkpoint and those that control tumor cell interactions with the microenvironment. Red notates the proteins with FDA-approved compounds with specific targeting. Bold and italicized names indicate receptor tyrosine kinases that are inhibited by FDA-approved multikinase inhibitors.

BRAF and MEK inhibitors

BRAF ^V600E is the most common driver oncogene in PTC. Mutually exclusive rearrangements involving RET and activating mutations in RAS (mostly NRAS) are found in radiation induced PTC and follicular variant of PTC, respectively in adults. Mitogen Activated Protein kinase (MAPK) activation in one of the critical common signaling pathways for these tumors (Fig. 1) (16). In vitro and murine models provide strong evidence that BRAFV600E-mediated MAPK activation drives thyroid cancer growth and dedifferentiation (17, 18). Interest in sorafenib in thyroid cancer was originally motivated by its RAF kinase inhibiting properties; however, the clinical activity of this MKI also does not correlate with the mutation and its inhibition of RAF is modest (4, 19). Subsequent, second-generation competitive RAF kinase inhibitors that have BRAF^V600E specificity (eg, vemurafenib and dabrafenib) have been studied in PTC and ATC (20-22). Vemurafenib and dabrafenib have been studied and FDA approved in melanoma in combination with allosteric inhibitors of MEK1/2 (trametinib) to suppress secondary reactivation of MAPK signaling (23). In 2018, the FDA approved dabrafenib/trametinib combination therapy for patients with BRAF^V600E-mutated ATC based on nondurable but sometimes dramatic responses (22). In thyroid cancer systems, resistance to single-agent BRAF inhibitors also has been shown to involve HER3 overexpression and activation of p21 activated kinase (PAK), both of which have been exploited to improve efficacy of activity (24, 25). PAK also has been targeted in RAS and Rac family small GTPase (RAC)-driven melanoma but has not yet been studied in those contexts in thyroid cancer (Fig. 1) (26, 27).

Because increased MAPK signaling driven by BRAF^V600E also drives dedifferentiation, BRAFV600E and/or MEK inhibitors have been used as a short-term pretreatment to redifferentiate and sensitize thyroid cancers to I-131 by improving Na, I symporter expression and function (20, 22, 28, 29). An additional potential benefit of this approach are data demonstrating that MAPK inhibition sensitizes BRAF^V600E and RAS-mutated thyroid cancer cells to radiation-induced DNA damage (30). Improved biomarkers to better select patients likely to benefit from redifferentiation and radiation sensitization to augment I-131 therapy and external beam radiation therapy are under investigation.

Additional targets and compounds

Advanced thyroid cancers uncommonly harbor rearrangements or mutations in other oncogenes for which there are FDA-approved compounds, either in a tumor-type agnostic model in which all tumors harboring a mutation are approved, or in a disease-specific manner. Examples include inhibitors of Anaplastic lymphoma kinase (ALK) mutations and rearrangements, tropomyosin receptor kinase rearrangements, and mammalian target of rapamycin (mTOR) mutations (Fig. 1) (4, 31). For these targets, tumor responses in patients with thyroid cancer are reported in clinical trials and/or case studies prompting recommendations for broader genomic testing in patients with progressive disease from multiple societies (1-3).

There a multitude of studies evaluating FDA-approved or preclinical targets in thyroid cancer. Space limitations preclude a comprehensive review of these in this context but a number of reviews have been published with recent updates on this topic (4, 31, 32). These include a number of additional receptor tyrosine kinase targets, intracellular kinases such as PI3K/Akt and others, cyclin-dependent kinases, and others both alone and in combinatorial strategies.

New approaches to target signaling molecules

The majority of compounds studied in thyroid cancer are compounds that inhibit kinase activity either as competitive ATP or allosteric inhibitors. However, there are a number of other emerging approaches with future potential for thyroid cancer.

Superenhancer transcriptional inhibitors.

In all eukaryotic cells, including cancer cells, transcription of coding mRNAs begins with the binding of RNA polymerase to upstream promoters and is subsequently regulated by activating or inhibiting transcription factors and coactivators. Selectivity of expression is further enhanced (“superenhanced”) by upstream DNA sequences that bind additional transcription factors/coactivators (Fig. 2). Recognition of these superenhancer sites occurs by modifications in histones (methylation and acetylation) that are highly regulated. More specifically, superenhancers occur when clusters of enhancers located very far from the transcriptional start sites are “looped” together and localized near the promoter site by a group of proteins including mediator, cohesin, and CCCTC-Binding factor, that recognize acetylation of lysine 27 on Histone 3 (H3K27ac) and other epigenetic marks (Fig. 2) (33-35). Superenhanced genes are important normally in defining lineage specificity. A role for aberrant superenhancing in cancer has been identified over the past decade when it was determined that some critical cancer genes, such as C-MYC and MYB, are selectively “superenhanced” and overexpressed. Cancer-specific superenhancing may occur through a variety of mechanisms, including mutations in the regulatory regions of the superenhancer or through chromosomal rearrangements where a superenhancer for proximal component of the fusion gene leads to overexpression of the oncogenic partner in the fusion (34, 35). The development of superenhancers has been shown to be functionally involved in the development of therapeutic resistance (33, 35, 36). Thus, regulators of superenhancer function have become therapeutic targets in the context of resistant cancer. Specific examples of current clinical trial targets include inhibitors of bromodomain subfamily members such as BRD4 (BET inhibitors) that regulate mediator, and inhibitors of CDK 7 and CDK9 that regulate RNA polymerase II function (Fig. 2) (35). In thyroid cancer, BET inhibitors, and CDK7 and CDK9 inhibitor have been reported to block expression of superenhanced oncogenes including as MYC in ATC and RET in MTC (37-40). This differs from use of more general epigenetic inhibitors, such as Histone Deacetylase inhibitors that were not effective in progressive DTC (41). There remain concerns about off-tissue effects when used long term; however, a number of compounds are in phase 1 and 2 clinical trials and data regarding efficacy and toxicity are forthcoming.

Figure 2.

Superenhancer schematic. (A) Gene transcription of a normally enhanced gene in which an upstream enhancer (red) interacts with a transcription (trans) factor supported by BRD4 linking to the transcriptional machinery by mediator in a complex folded together by cohesion. RNA polymerase 11 (RNA Pol II) requires phosphorylation by cyclin dependent kinases (CDK) 7 and 9 to initiate and engage transcription leading to production of mRNA. (B) Superenhanced genes are regulated by a group of far upstream (often >10kb) enhancers (red) marked by specific epigenetic changes (blue ovals) that engage groups of transcription factors, BRD4 molecules and mediator molecules folded together by cohesin leading to greater activation of transcription at the promoter site leading to increased gene transcription and subsequent mRNA production. Green stars denote the therapeutic targets with compounds in clinical trial or studied in thyroid cancer.

A number of other nucleic acid–based technologies with a goal to reduce expression of target genes have been utilized with therapeutic intent in cancer. These are not discussed in detail here but include delivery of siRNAs and microRNAs using various nanotechnologies to cancer cells, and CISPR-CAS9 methods (42). Off-target effects, and mechanisms of resistance for gene knockouts are areas of active research, but with recent FDA approvals there is opportunity using these approaches in the future, as reviewed in detail elsewhere (43).

Covalent inhibitors

RAS and other G proteins with function regulated by GTP binding and subsequent hydrolysis to GDP into an inactive state have proven to be challenging targets for drug development. Mutations in NRAS are particularly common in follicular thyroid cancer and follicular variant of PTC; mutations in HRAS and KRAS occur in MTC (Fig. 1) (16, 44). Activating mutations in KRAS (such as G12C) occur commonly in pancreatic cancer and other solid tumors, and have been induced by treatment with BRAF^V600E-targeted treatments in thyroid cancer (45). Compounds are now under development and clinical testing that can maintain KRAS G12C into a GDP-bound inactive state covalently, thereby blocking further activation of the mutant protein. This approach is currently being extended other small G protein mutants and other signaling molecules, and represents a potential approach for RAS-mutated by targeting RAS or its downstream signaling molecules (46, 47).

Protein “degraders.”

Another approach to removing an oncogenic protein is the use of compounds that reduce their half-lives through enhanced degradation. Such an approach has been utilized in thyroid cancer in a nonspecific manner. For example, heat shock protein 90 is a molecule that stabilizes a large number of client proteins, including key signaling molecules in thyroid cancer such as Akt. Geldanamycin and modified versions of this compound, such as 17AAG, have been utilized in thyroid cancer preclinical and clinical trials (48). Efficacy of the nonspecific approach has been modest and toxicities are high. However, a more specific approach has been developed using proteolysis targeted chimeras in which the compound links a target-specific binding domain to a protein that recruits E3 ligases that engage with the protein degrading machinery thereby enabling specific protein degradation (49). Although not yet reported in thyroid cancer, specific degraders of BRAF^V600E and AKT have been recently reported that are of relevance to thyroid cancer (50, 51).

Immune-targeted therapies

The relationship between thyroid cancer and the immune system has long been studied in the field owing to the common co-occurrence of PTC and Hashimoto’s thyroiditis (52). Activation of antigen presentation and an activated tumor inhibitory immune response in PTC in this context is thought to confer an improved prognosis for many patients when present (53). As DTCs dedifferentiate changes occur in the immune population that allow cancers to evade this immune response, including expression of programmed death-1 (PD1) and PD ligands 1 and 2 that allow for immune evasion and expansion of tumor-associated macrophage populations that inhibit immune activation and enable progression (M2 macrophages) (54, 55). Preclinical studies demonstrate that disruption of tumor-associated macrophage recruitment can reduce progression of BRAF^V600E-induced thyroid cancer in vivo (56). Finally, utilizing immune-based therapies such vaccines, monoclonal antibodies or genetically engineered antitumor Chimeric Antigen Receptor (CAR)-T and NK-cell approaches are areas of current exploration in thyroid cancer (57). Several of these are reviewed below.

Immune checkpoint inhibitors

With the development and subsequent FDA approval of PD1 and PDL-1 targeted therapies for cancer, and the identification of biomarkers to predict activity, there has been emphasis on pursuing these approaches for poorly differentiated, medullary, and anaplastic thyroid cancers that often have appropriate immune environments (Fig. 1) (54, 58, 59). A recent phase II single arm study of the PD1 inhibitor spartalizumab showed efficacy in 29% of patients with PDL1-positive tumors versus 0% of patients with PDL1-negative tumors confirming the importance of the biomarker in ATC (60). Similar results were reported in a recent phase 1b study of the PD1 inhibitor pembrolizumab in PDL-1 progressive DTC (61). Current trials are enrolling using combinations of kinase inhibitors and immune checkpoint treatments and combinations of PD1 and PDL-1 inhibitors. In all cases, it appears that selection of proper tumors using biomarkers, or demonstration that kinase inhibitors are able to enhance PDL1 expression in tumor samples may be critical for patient selection.

Vaccines and Chimeric Antigen Receptor-expressing cells

The evolution of vaccine technologies and the development of CAR-T and CAR-NK therapies for liquid and solid tumors over the past decade has led to studies in thyroid cancer model systems. A challenge for most vaccines is the identification of tumor antigens to enable specificity of the therapy to cancer cells. There has been particular interest in dendritic cell and anti-Carinoembryonic Antigen (CEA) vaccines in MTC and other neuroendocrine tumors (62, 63). Indeed, a recent study recently reported that MTCs are characterized by an active immune environment with potential antigen expression suggesting this tumor type might be amenable to CAR-T therapy (58). Large studies have not yet been reported but evidence of induction of an immune response has been reported (64). CAR-T therapy targeting ICAM-1 has shown efficacy in ATC mouse models but it has not been reported yet in patients (57).

Peptide Receptor Radionuclide Therapy

Radioiodine therapy has been one of the mainstays of treatment of thyroid cancer and uptake on I-131 scanning has served as a biological predictor of therapeutic efficacy. Somatostatin analogues that bind somatostatin receptors have been used for similar purposes in DTC, MTC, and neuroendocrine tumors (65). Although not studied in randomized clinical trials in thyroid cancer, there is evidence that the use of somatostatin analogues to deliver therapeutic radiation may have benefit in patients with DTC and MTC (44, 65-68). The recent FDA approval of 177Lu-DOTATATE for somatostatin receptor–expressing midgut neuroendocrine tumors, and the availability of DOTATE for positron emission tomography scanning opens opportunities for further clinical trials of this approach in thyroid cancer.

Summary

Treatment options for patients with progressive metastatic DTC, MTC, and ATC have expanded dramatically over the past decade owing to extensive work on the biology of these cancers, human tumor correlative studies, and expansion of clinical trials. Clear evidence of improved length of survival and quality of life underscore the need for improved therapies. There are a number of avenues to improve inhibition of known targets, define mechanisms of secondary resistance and harness the host–tumor response to improve the care of patients with thyroid cancer.

Glossary

Abbreviations

ATC

anaplastic thyroid cancer

ATP

5′-adenosine triphosphate

CAR

Chimeric Antigen Receptor

DTC

differentiated thyroid cancer

FDA

Food and Drug Administration

MKI

multikinase inhibitor

MTC

medullary thyroid cancer

PAK

p21 activated kinase

PTC

papillary thyroid cancer

Additional Information

Disclosure Summary : None.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.