Drug repositioning in thyroid cancer: from point mutations to gene fusions

The diagnosis of thyroid cancer (TC) has increased dramatically in recent years. Papillary TC is the most frequent type and has shown a good prognosis. Conventional treatments for TC are surgery, hormonal therapy, radioactive iodine, chemotherapy, and targeted therapy. However, resistance to treatments is well documented in almost 20% of all cases. Genomic sequencing has provided valuable information to help identify variants that hinder the success of chemotherapy as well as to determine which of those represent potentially druggable targets. There is a plethora of targeted therapies for cancer, most of them directed toward point mutations; however, chromosomal rearrangements that generate fusion genes are becoming relevant in cancer but have been less explored in TC. Therefore, it is relevant to identify new potential inhibitors for genes that are recurrent in the formation of gene fusions. In this review, we focus on describing potentially druggable variants and propose both point variants and fusion genes as targets for drug repositioning in TC.


Introduction
Thyroid cancer (TC) is the most common malignant tumor of the endocrine system, with 586,202 new cases worldwide in 2020 (1).The overall incidence of TC has increased dramatically in the last 30 years.This increase may be due to overdiagnosis, thanks to improvements made in diagnostic procedures (2).Morphologically and clinically, TC is classified into two main groups: differentiated cancer-comprising papillary and follicular thyroid cancer-and undifferentiated TC, designated anaplastic carcinoma of the thyroid (3).The most prevalent is papillary thyroid cancer (PTC), which accounts for up to 85% and has a good prognosis (5-year survival rate of more than 95%, mainly in patients with stage I or II disease), as does follicular thyroid cancer (FTC), which is less prevalent, accounting for 15% of all cases (4).Patients with poorly differentiated or anaplastic TC, advanced-stage disease, or distant metastases have higher mortality rates (5).Moreover, about 20% of PTC patients manifest disease recurrence because of drug resistance, suggesting a change in treatment approaches.This points out the need to personalize treatments, including drug repositioning (6).Target therapy can be repositioned and offers greater success since it can be customized according to the patient's genomic alterations.In this review, we highlight therapeutic opportunities for TC, focusing on druggable genes with potential repositioning for personalized therapy.

Classical point mutations in thyroid cancer: windows of opportunity for the use of drug repositioning
Radioactive iodine administration and/or surgery remain the first line of treatment for TC; however, for advanced disease, chemotherapy (CT) becomes the systemic option of treatment available (7).Nevertheless, CT constantly faces resistance and severe secondary effects (8).Therefore, it is necessary to overcome resistance by recognizing drug-susceptible mutations, which may lead to the identification of a broad spectrum of target therapies that could be repositioned in TC (Figure 1).
Next-Generation Sequencing (NGS) has made it possible to sequence the genomes of different types of cancer, which has revealed that around 90% of patients with TC have one or more genetic abnormalities (9).Dysregulation of phosphatidylinositol 3-kinase (PI3KCA) and mitogen-activated protein kinase (MAPK) signaling pathways is mainly affected by point mutations in target genes such as B-Raf proto-oncogene, serine/threonine kinase (BRAF), A-Raf proto-oncogene, serine/threonine kinase (RAS), and ret protooncogene (RET) (10).One of the best-documented and highest prevalence point mutations in PTC is BRAF exon 15 p. V600E (45% of all cases), which is associated with poor prognosis and high recurrence (11).The BRAF exon 15 p. V600E variant has constitutively active BRAF serine-threonine leading to the activation of effectors of the MAPK pathway and, consequently, surveillance and proliferation (11).Vemurafenib has shown antitumor activity in patients with BRAF exon 15 p. V600E-positive progressive PTC, representing a potential new therapeutic option (12,13).Ipilimumab, nivolumab, dabrafenib, and trametinib are also approved target therapy options for BRAF mutations in melanoma ( 14) that could be repositioned to TC.In addition, drugs blocking phosphatase and tensin homolog (PTEN) and PI3KCA homogenize the font of the letter with that of the rest of the text effects (Table 1).
PI3KCA is another gene with several missense mutations in three subtypes of TC: follicular, papillary, and anaplastic.Interestingly, PI3KCA mutations are associated with drug resistance in BRAF exon 15 p. V600E-positive cases.In this scenario, it is worth looking at how alpelisib can counteract the resistance mechanism by diminishing the EPH receptor B2 (EPHB2)-induced signaling (38).Consistent with the latest, PTEN, which has a negative regulatory role in the same pathway, has reported variants in TC (39).
KRAS proto-oncogene, GTPase (KRAS), is a G protein that plays an important role in the PI3KCA/MAPK signaling pathway.Point mutations in KRAS usually occur at codons 12, 13, and 61 and have been found in FTC and PTC at a frequency of 50% and 20%, Drugs with potential to be investigated in thyroid cancer clinical trials according to mutational profile.(42).Selpercatinib and pralsetinib are RET-specific inhibitors approved for the MTC variant and have been well tolerated (43)(44)(45).Other multitargeted kinase inhibitors used to inhibit the PI3KCA/AKT/mechanistic target of the rapamycin kinase (MTOR) pathway in MTC are vandenitib and cabozantinib.The first one inhibits RET but also inhibits other kinases such as vascular endothelial growth factor receptor 2 (VEGFR2), vascular endothelial growth factor receptor 3 (VEGFR3), and epidermal growth factor receptor (EGFR), while cabozantinib inhibits RET, vascular endothelial growth factor (VEGF), MET proto-oncogene, receptor tyrosine kinase (MET), and ROS proto-oncogene 1, receptor tyrosine kinase (ROS1) (46).Both inhibitors have shown efficacy and improved overall survival in patients harboring RET exon 16 p.M918T mutation (47,48).
MET is a receptor tyrosine kinase that has an oncogene role in promoting angiogenesis due to downstream activation of RAS, PI3KCA, and signal transducer and activator of transcription 3 (STAT3) signaling pathways (49).Drugs that inhibit MET are amivantamab, cabozantinib, capmatinib, crizotinib, osimertinib, and tepotinib (50).Particularly, MET has a reported variant in NSCLC that skips the exon 14 and makes the protein constitutively active (51).In TC, it constitutes an inclusion criterion for thyroid gland medullary carcinoma (52,53).Currently, therapy targeting MET, although only indicated to treat NSCL, represents a potential target in TC.Furthermore, a PTC expression signature has been identified in which three genes are overexpressed, promoting metastasis and being associated with poor prognosis: dipeptidyl peptidase 4 (DPP4), MET, and catenin beta 1 (CTNNB1).The signature is associated with immunosuppression and correlates with tumor infiltration of tumor-associated macrophages, which leads to T-cell exclusion.Interestingly, sitagliptin, an FDA-approved drug to treat diabetes type II, has affinity not only to DPP4 (diabetes target) but also to MET and CTNNB1 (54-56).Moreover, the affinity for MET and CTNNB1 is even higher than FDA-approved inhibitors specific for each of them, like crizotinib and PNU-74654, respectively.Therefore, sitagliptin represents a multidrug therapy window for TC (18).Paired box 8 (PAX8), a gene implicated in proliferation and migration, is usually overexpressed in TC.Likewise, in high-grade serous ovarian cancer, PAX8 is upregulated (57).Remarkably, losartan and captopril, which are FDA-approved drugs, have been found effective at inhibiting PAX8 expression and function.This evidence suggests potential therapeutic opportunities using losartan and captopril, not only for ovarian cancer but also for TC (57).
Besides the variants reported in the above-mentioned genes, there are also polymorphisms associated with TC (58).For instance, neuregulin 1 (NRG1) acts as an oncogene through its role as a glycoprotein that mediates cell-to-cell signaling (59).In breast cancer, lapatinib may be used to inhibit EGFR and erb-b2 receptor tyrosine kinase 2 (HER2) kinases, two receptors of also relevant function in TC.Nonetheless, resistance is acquired and correlates with an increased expression of NRG1.By trying to overcome it, adding pertuzumab has shown promising results in decreasing NRG1-acquired resistance and tumor progression (25).
Similarly, secreted phosphoprotein 1 (SPP1), an integrinbinding glycophosphoprotein overexpressed in TC that promotes tumorigenesis through the inhibition of differentiation factors of thyroid cells, represents an opportunity for drug repositioning (60,61).Although there are no current FDA drugs approved for inhibiting SPP1, a recent publication showed a promising inhibitory drug for cervical cancer: entrectinib (26).This represents a highlight, as entrectinib is an FDA-approved drug for NTRK fusions in solid tumors, including TC (62).
As with SPP1, another window of opportunity for targeted treatment is BCL2-associated X, apoptosis regulator (BAX).This gene participates in mitochondrial regulation of cell death; however, in cancer, it contributes to cell death dysregulation (63).Importantly, in TC, BAX has a reported polymorphism positively correlated with PTC, and more importantly, the FDA-approved drug eltrombopag acts as a BAX inhibitor, which drives apoptosis induction (64,65).SPP1 and BAX are not the only genes in which polymorphism is related to TC. VDR stands for vitamin D receptor and has been associated with cancer development (66).It is not well established if TC contributes to or attenuates tumor growth; however, two polymorphisms, FokI and TaqI, are associated with a more aggressive type, and the heterozygous FokI to metastasis (67).Remarkably, it has been shown that antagonists of vitamin D have therapeutic effects as they inhibit downstream cell cycle proliferation.There is already an insight into potential therapies using VDR as a druggable target.For instance, PS121912 has shown promising therapeutic effects by acting as a selective VDR inhibitor (28).
In the immunology context, several profiles have been described causing differential expression and immune cell proliferation among TC subtypes (68).Interleukin-10 (IL-10) is one of several dysregulated cytokines in TC associated with immunological and apoptosis evasion and aggressiveness (68).This effect is caused by expression induction of BCL2 like 1 (bcl-xL) and BCL2 apoptosis regulator (BcCL2) and resistance to CD95-mediated apoptosis (69,70).Due to its oncological role, IL-10 figures as a potential therapeutic target.Rituximab has promising inhibitory effects against IL-10 through downregulation of BCL2 and sensitization of B-cell non-Hodgkin's lymphoma to apoptosis (29).However, resistance constitutes a problem due to broad kinase inhibitor activity and toxicity, which may limit their use and encourage the use of more specific inhibitors (71).
Lastly, telomerase reverse transcriptase (TERT), an enzyme known to be implicated in cancer, has been described as one of the most frequently mutated genes in TC, particularly in its promoter, which causes its overactivation.TERT inhibitors are currently under study, and NU-1 not only sensitizes the cell to chemo-and radiotherapy but also can inhibit proliferation and increase immune activity (30).
From NGS of long DNA fragments, gene fusions have been identified.When two genes conform to a fusion, they either lose or gain function.In cancer, they can contribute to tumor growth due to constitutive activation of an oncogene, such as BCR-ABL (72).Remarkably, some gene fusions are considered drivers, while others contribute to generating more genomic instability and disease development.There are gene fusions that are found across various cancers (73).These features of gene fusions represent an unprecedented opportunity to develop target therapies aimed at providing personalized medicine to patients.

Spotlight of novel therapies: gene fusions
Over 50 gene fusions have been identified in TC, which are mainly conformed by the RET, neurotrophic receptor tyrosine kinase (NTRK), ALK, and BRAF genes (74).These genes are tyrosine kinase overactivated mainly due to kinase retention and overexpression by transcription factors of the parental genes, making them druggable targets (75).Currently, three drugs are being used in clinics to treat TC-targeting gene fusions: pralsetinib, selpercatinib, and larotrectinib (76).The first two are RET inhibitors and were first set as a treatment for both point mutations and gene fusions; however, selpercatinib shows efficacy in specific RET variant genotypes that present pralsetinib resistance.For instance, BaF4/KIF5B-RET shows tumor growth despite treatment with pralsetinib, while selpercatinib can effectively inhibit growth (44,77).However, as with other variants, these gene fusions are not expressed across all subtypes of cancer, while some therapies face drug resistance and lack of treatment for greater, yet untargeted variants (Figure 1).
For RET, 19 fusions have been described; however, only therapies consisting of RET inhibitors are currently available, leaving the partner genes pharmacologically unexplored (78).This is of great importance as it has been described that the inhibitory sensitivity of several gene fusions varies depending on the partner gene; hence, drug screening should be performed testing not only the most common gene.For instance, the coiled-coil domain containing 6 (CCDC6) is a recurrent gene-forming fusion with RET in lung cancer, where it has shown potential druggability of EGFR inhibitors in combination with RET inhibitors, decreasing resistance to RET inhibitors while also enhancing sensitivity to PARP inhibitors (79).Particularly, the fusion CCDC6-RET is more sensitive to vandetanib due to the off-target inhibitory effect and crosstalk with EGFR pathway activation (80).Furthermore, this fusion and ERC1-RET have not had a response to the RET drug, cabozantinib, supporting the idea of focusing on the second gene as well (81).
Larotrectinib targets the NTRK genes, which are neurotrophic tyrosine kinase receptors.If binding occurs, the protein phosphorylates itself and activates the MAPK pathway.Therefore, as part of a gene fusion, it causes its constitutive activation (82).Several fusions involving NTRK1, NTRK2, and NTRK3 have been described in the lung, colon, brain, head and neck, and TC (83).For this reason, it has been a promising targeted therapy, as the same fusions can occur in several tissues.In TC, larotrectinib is administered to patients diagnosed with the anaplastic subtype, and tumor growth continues despite other treatments (82).An example of this is the ETV6-NTRK3 fusion, which has been described as a driver variant in secretory breast cancer with high efficacy upon larotrectinib treatment (84).However, larotrectinib therapy targets only the NTRK gene, while their partner genes remain untargeted.For instance, sequestosome 1 (SQSTM1) is a gene that conforms to fusions with both NTRKs and plays a role in autophagy, specifically through the AKT/protein kinase AMP-activated catalytic subunit alpha 2 (AMPK)/MTOR signaling reported in PTC (84).
Although only three drugs are being used in TC to target gene fusions, there are several other recurrent genes forming gene fusions that are already targeted in other cancers.On one hand, there is ALK, which is widely known for its oncogenic role, especially as part of gene fusions (85).Currently, ALK fusions do not have a regimen of treatment for TC, but its potential has already been evaluated.For example, STRN-ALK and EML4-ALK are promising targets in TC using the FDA-approved drug crizotinib, among other drugs such as ceritinib, alectinib, brigatinib, and lorlatinib (32, 33).
On the other hand, there are BRAF fusions, and remarkably, despite BRAF having several target drugs, none of them are used to treat TC.Furthermore, among all the gene partners of BRAF, staphylococcal nuclease and Tudor domain containing 1 (SND1), an oncogene in several types of cancer acts in addition to posttranscriptional modifications (86).This is a highlight for novel therapy, as a small molecule called suramin has been identified to inhibit their protein by impairing its interaction with several microRNAs and sensitizing the response to standard chemotherapy (87).
Interestingly, up to five gene fusions are involved in THADA armadillo repeat containing (THADA), which stands for thyroid adenoma-associated gene (Table 1) (74).This gene participates in metabolism and energy storage through the calcium pathway.In cancer, not only fusions but also polymorphisms are associated with the disease development (88).Particularly, it has been described that THADA is necessary to retain CD274 in the Golgi for maturation.On the contrary, if suppressed, the immune response is triggered through the infiltration of CD8 + T cells and increased toxicity (89).In addition to this finding, the IGF2BP3-THADA fusion has been demonstrated to cause overexpression of the partner gene IGF2BP3, leading to sustained growth and invasion through the activation of PI3KCA and MAPK pathways (34,90).For its part, insulin-like growth factor 2 mRNA-binding protein 3 (IGF2BP3) is associated with a poor prognosis implicated in several mechanisms leading to aberrant metabolism in cancer (91).Currently, there are no inhibitors for THADA; however, the data strongly point out THADA as a potential therapeutic target in TC.
Another gene found in 30%-35% of FTC is PAX8-PPARG, characterized as an oncogene due to its binding to several genomic regions that code for genes related to cell proliferation, apoptosis evasion, and motility (92).Contrary to the case of repurposing losartan to PAX8 alterations, this fusion promotes tumor progression due to the likely loss of functions of peroxisome proliferator-activated receptor gamma (PPARG).When inhibited with pioglitazone, anti-inflammatory effects and growth modulation are observed; however, the function of the gene fusion is not yet fully understood (35).Opposed to this idea, the antitumoral effect of PPARG inhibitor GW9662 has also been described, indicating the existence of independent pathways of PPARG (93).Remarkably, PAX8-PPARG is not the only fusion in TC involving PPARG; there is also CREB3L2-PPARG (94).
NUT-BDR4 is an oncogenic driver fusion that causes a rare type of cancer named NUT midline carcinoma.Bromodomain-containing protein 4 (BDR4) binds to the chromatin, while NUT midline carcinoma family member 1 (NUT) recruits histone acetyltransferase (HAT), promoting the expression of several associated oncogenes (95).This rare fusion has also been described in some TC cases, and it is associated with high expression of CD274 (96).The prognosis is low, with an estimated overall survival of 10 months, while therapy consists of radiotherapy and standard chemotherapy for large tumors.With no targeted therapy available, it is an urgent matter to start studying potential inhibitors for the treatment of these patients (97).Currently, only one inhibitor has been proposed to target the NUT-BDR4 fusion.It consists of a dual inhibitor of bromodomain and extra-terminal motif (BET) proteins and the p300 bromodomain, named NEO2734, with proven inhibition of tumor growth and improvement of overall survival (36).
Lastly, MET not only has point mutations in TC but also a gene fusion.It has been identified that TGF-MET fusion is present in sarcoma, glioma, and TC (37,98).Interestingly, in sarcoma, tumors that have this variant do not fit into a specific subtype, which is a remarkable finding due to the existence of effective MET inhibitors (50,98).

Conclusions and perspectives
It is relevant to recognize that in the era of personalized medicine, drug repositioning has a major impact on oncology.This is possible due to the identification of new therapeutic targets, which can be shared in different diseases and even between cancers.This opens a whole window of opportunity for the use of a plethora of drugs, reducing the time and costs involved in the production of new drugs, which has a positive impact on patients.In this review, we found that several drugs used in different types of cancer can be repositioned to TC, either by the presence of point mutations or by gene fusions.We found an area of opportunity for 13 genes with missense mutations and 10 for gene fusions.Among all these drugs, 22 are FDA-approved drugs, while the remaining five are inhibitors with proven efficacy in in vitro studies, both of which represent a promising area of therapy opportunity.It is the aim of this work to highlight the relevance of the identification of new potential inhibitors for genes that are part of recurrent fusion formation in TC as well as other types of cancer due to the likelihood of their contribution to disease development.Hence, it is of interest to the clinic to elucidate these variants' potential as biomarkers or prognostic or therapeutic targets.

TABLE 1
Variants in potentially druggable genes in thyroid cancer.