Thyroid cancer (TC) is currently the most common malignancy of the endocrine system and its incidence has steadily increased worldwide over the past 20 years (1,2). TC exhibits considerable heterogeneity and shares typical histopathological features with other tumors (3). The two varieties of epithelial cells that constitute the thyroid gland are follicular and parafollicular cells. Follicular cells convert iodine into T4 and T3, and the hormones T3 and T4 (tyrosine) regulate metabolism (4). By contrast, parafollicular cells are epithelial cells that produce calcitonin. Primary TC predominantly originates from thyroid follicular cells and is classified as an epithelial tumor, which can be categorized into three primary pathological types. The most prevalent type is papillary TC (PTC), followed by follicular TC (FTC), with the most aggressive form being anaplastic TC (ATC). Furthermore, medullary TC (MTC) arises from parafollicular cells (5).

Tumorigenesis is mainly driven by somatic mutations that occur during the initial stages of transformation. According to various genetic studies on TC, several oncogenes and tumor suppressor genes have been identified that can be used as diagnostic markers and therapeutic targets (6–8). For example, pathogenic loss-of-function mutations in genes responsible for tumor suppression constitute a common genetic event involved in TC. Crucial examples include phosphatase and tensin homolog and tumor protein 53, which are frequently inactivated via such mutations. These alterations are highly prevalent and well-characterized events driving the initiation and progression of TC. The use of targeted therapy in affected patients has become possible due to the development of tyrosine kinase inhibitors (TKIs) (9–11). The major driver mutations involved in TC mainly include B-Raf proto-oncogene (BRAF), RET (mutations/fusions) and RAS mutations. In addition, gene mutations related to the VEGFR and PI3K/AKT/mTOR signaling pathways, as well as mutations in drug resistance-associated genes such as mesenchymal epithelial transition and neurofibromin 2, are also closely connected with the regulation of TKI efficacy and the mechanisms of drug resistance. TKIs function by competing with adenosine triphosphate (ATP) for binding sites, thus inhibiting phosphorylation by kinases, and ultimately preventing the signaling and proliferation of tumor cells (12). Novel and state-of-the-art genetic testing based on advanced next-generation sequencing has promoted the development of tumor-targeted therapies, in which the molecular drivers of tumorigenesis act as the therapeutic targets. Previous studies (13,14) have compared targeted therapies with non-targeted agents and immunotherapeutic agents lacking well-defined molecular targets, including chemotherapeutic drugs for advanced anaplastic thyroid carcinoma (ATC) and radioiodine-refractory differentiated thyroid carcinoma (RAIR-DTC), such as taxanes (paclitaxel, docetaxel), anthracyclines (doxorubicin), platinum-based agents (cisplatin, carboplatin, etc.), and PD-1/PD-L1 inhibitors (pembrolizumab, nivolumab). The results indicate that targeted therapies achieve superior efficacy and a lower incidence of off-target adverse events. The aim of the present study was to examine and summarize the role and organization of the rearranged during transfection (RET) proto-oncogene, along with its disease-causing mechanisms, in different variants of TC. In addition, the present revied aimed to investigate small-molecule inhibitors of tyrosine kinases that target RET mutations, and present a summary of advances in research on the management and treatment of TC linked to alterations in the RET gene.

The RET gene, which is located on chromosome 10q11.2 and encodes a 170 kDa transmembrane receptor tyrosine kinase, was first discovered by researchers in 1985 (15). Structurally, the RET protein comprises an extracellular region featuring four cadherin-like domains (CLD1-4), a calcium-binding site and a cysteine-rich domain (CRD), a transmembrane region, and an intracellular region containing the juxtamembrane domain and a bilobed tyrosine kinase domain (TKD). Under normal physiological conditions, RET activation is ligand-dependent. The binding of the glial cell line-derived neurotrophic factor (GDNF) family ligands, along with their co-receptor GDNF family receptor α, induces RET dimerization. This dimerization triggers the trans-autophosphorylation of specific tyrosine residues within the TKD (16–18). Subsequently, these phosphorylated residues serve as docking sites for downstream adaptor proteins, initiating crucial signaling cascades such as the RAS/MAPK and PI3K/AKT pathways, which govern essential cellular processes, including proliferation, differentiation and survival (Fig. 1) (19).

The constitutive, ligand-independent activation of RET, primarily driven by specific mutations, is central to the pathogenesis of MTC (20–22). These activating mutations occur either in the germline (hereditary, 95–98%) or somatically (sporadic, 25–50%) (23–25). The 95–98% refers to germline mutations accounting for the majority of mutations in hereditary TC cases, while the 25–50% refers to somatic mutations in all TC cases (including sporadic ones); these two proportions describe different mutation subsets and are not mutually exclusive. The global age-adjusted incidence rate of germline RET mutations is estimated to be 0.06 cases per 100,000 persons/year, and the corresponding prevalence is 1.3 cases per 100,000 individuals (26,27).

RET-activated tumors may arise from chromosomal rearrangements, which involve the fusion of the RET kinase domain with various protein partners that possess dimerization domains (48–50). This phenomenon occurs along with germline and somatic mutations that activate RET. Previous studies have proposed that the erroneous repair of DNA double-strand breaks serves a critical role in the molecular mechanisms that lead to RET fusion. Specifically, these chromosomal breaks facilitate the fusion of the 3′ sequence that encodes the RET mRNA kinase domain with the 5′ sequence that encodes domains responsible for both the dimerization and localization of an upstream partner gene. This fusion ultimately results in the production of active RET fusion proteins, which contribute to tumorigenesis (51–53). The RET fusion protein can cause cancer through two pathways: First, by sending a message that promotes the continued proliferation of cancerous cells due to the ability to send such messages without a ligand; and secondly, due to the unique endocytosis process, which prevents RET from being inhibited by blocking the attachment of ubiquitin to the protein receptor (54).

In 1987, Grieco et al (55) reported the initial human chromosomal arrangement of RET in PTC, demonstrating fusion between the RET TKD and the 5′ terminal segment of coiled-coil domain-containing 6 (CCDC6). Breakpoints within RET intron 11 are the most common sites for gene fusion, resulting in the binding of the RET cytoplasmic kinase domain. Additional cases have been shown to occur in introns 7 and 10, resulting in merging of the transmembrane domain of RET (56,57). According to the literature, among the several RET fusions that are associated with PTC, the most common fusions are rearrangements involving the CCDC6 gene (known as RET/PTC1) and the coiled-coil domain of the nuclear receptor coactivator gene (known as RET/PTC3) (58–60). The kinesin family member 5B/RET fusion is the most common fusion found in lung adenocarcinoma (61,62). The nuclear receptor coactivator 4-RET gene and CCDC6-RET fusions are caused by paracentric chromosomal inversion involving chromosome 10, which does not include the centromere, whereas the pericentric inversion of chromosome 10, which includes the centromere, causes the kinesin family member 5B-RET fusion (63,64). The RET gene is likely to be fused with the following additional genes: Protein kinase cAMP-dependent regulatory subunit type I-α, tripartite motif-containing 24, Golgin A5, Kinectin 1, tripartite motif-containing 33 and RET finger protein gene 9 (historical/legacy nomenclature) (65–67). Notably, the term ‘RET finger protein gene 9’ primarily refers to the tripartite motif-containing protein 9 gene (TRIM9), which belongs to the TRIM family of proteins characterized by typical RING finger, B-box and coiled-coil domains. The term ‘RET finger protein’ is a historical descriptive term for proteins harboring RING finger domains, similar to the RET-associated RFP/TRIM27, and does not indicate a direct association with the RET gene itself. The legitimate repair of double-strand breaks in DNA leads to recombination events (68). The occurrence of RET fusions in PTC is fundamentally associated with exposure to specific damaging agents such as ionizing radiation and reactive oxygen species. These two factors serve critical roles in the cellular mechanisms that lead to cancer development (69–71). Both ionizing radiation and reactive oxygen species can induce notable genomic instability by causing double-strand breaks in DNA. Such disruptions in DNA structure are pivotal in the progression of PTC as they contribute to alterations in genetic material that drive oncogenesis. RET fusions are often seen in children and young adults with TC (72–74).

The rate at which RET fusion occurs is even higher in radiation-induced TC. A previous study was performed to analyze molecular genetic aberrations and associated phenotypes in the pathological tissues of 191 patients with PTC who were exposed to radioactive iodine from the Chernobyl reactor as young children. The results revealed that the frequency of RET gene rearrangements increased to 62.3% in the first decade following exposure to radiation. In addition, ELE1/RET (PTC3) rearrangements were markedly more common than H4/RET (PTC1) rearrangements (75). Note that NCOA4 is the official symbol for the nuclear receptor coactivator gene involved in this rearrangement, which is also widely known by the aliases ELE1 and ARA70. RET/PTC1 is defined as the CCDC6 (formerly H4)-RET gene fusion, whereas RET/PTC3 corresponds to the NCOA4 (ELE1/ARA70)-RET gene fusion. An additional investigation of the survivors of the atomic bombings that took place in Hiroshima and Nagasaki (Japan), revealed that among 50 patients with PTC who were exposed to nuclear radiation, 11 (22%) exhibited RET rearrangements, whereas only 5% of the 21 unexposed patients exhibited the same RET rearrangements. To the best of our knowledge, this study is the first to clearly demonstrate a statistically significant positive association between radiation exposure dose and the incidence rate of RET/PTC rearrangement in a large-scale human sample (analyzing 249 PTC tumor samples from atomic bomb survivors). Specifically, the higher the radiation dose received by the survivors, the higher the proportion of RET/PTC rearrangement in the PTC tumors they later developed (76).

The specific phosphotyrosine residues on the activated RET intracellular tail recruit distinct adaptor and effector proteins, thereby activating multiple downstream signaling pathways critical for both normal development and oncogenesis (77–79). The RAS/RAF/MEK/ERK pathway (80–84) is primarily initiated through phosphorylated-Y1062 and -Y1096 (these phosphorylated tyrosine residues are located on the RET protein. Specifically, phosphorylated-Y1062 and phosphorylated-Y1096 are tyrosine phosphorylation sites on the intracellular domain of the activated RET receptor tyrosine kinase), and this pathway is a key driver of cellular proliferation and differentiation. The PI3K/AKT pathway (85–87) is a pro-survival pathway that is activated through several mechanisms, including the recruitment of SHP2 to phosphorylated-Y687 (phosphorylated-Y687 is also a tyrosine phosphorylation site on the intracellular domain of the activated RET receptor tyrosine kinase). The JAK2/STAT3 pathway (88,89) is activated following the binding of STAT3 to phosphorylated-Y752 and -Y928 (phosphorylated-Y752 and -Y928 are tyrosine phosphorylation sites of the RET protein), leading to nuclear translocation and the regulation of target gene expression. The phospholipase (PLC)γ/protein kinase C pathway (90) is triggered by PLCγ binding to phosphorylated-Y1015 (phosphorylated-Y1015 is a tyrosine phosphorylation site on the intracellular domain of the activated RET receptor tyrosine kinase); this pathway contributes to various cellular responses.

In cancer, genetic alterations subvert this tightly regulated system. In MTC, extracellular cysteine mutations (C609, C611, C618, C620, C630 and C634) cause constitutive dimerization (91,92), whereas intracellular kinase domain mutations (for example, M918T and V804M) stabilize the active conformation, enhance ATP binding or impair autoinhibitory mechanisms. In PTC, chromosomal rearrangements create RET fusion genes (93–96). These chimeric proteins, often lacking the transmembrane domain and fused to dimerization partners, exhibit ligand-independent, constitutive kinase activity localized in the cytoplasm. Collectively, these diverse genetic events lead to the persistent and unregulated activation of RET-driven oncogenic signaling, fueling tumor initiation, growth and progression.

Numerous TKI medications have advanced to the stages of preclinical and clinical development (97–99). For example, the 2025 National Comprehensive Cancer Network Guidelines recommendations on TKI inhibitors for TC are as follows (100): For iodine-refractory PTC and FTC, both lenvatinib and sorafenib are Category 1 recommendations, with lenvatinib as the preferred first-line option for locally recurrent/metastatic, progressive radioactive iodine-refractory PTC and FTC. For MTC, first-line agents include vandetanib and cabozantinib (Category 1 recommendations). For MTC with RET mutations, selpercatinib and pralsetinib (highly selective RET inhibitors) are preferred. For ATC, treatment is mainly chemotherapy combined with immunotherapy, whereas TKIs may be used in the setting of positive specific targets. For example, dabrafenib plus trametinib can be used for tumors with the BRAF V600E mutation.

The recommendations on TKI inhibitors for TC in the 2025 American Thyroid Association Guidelines (101) include: For TC with RET fusions, highly selective TKIs, such as selpercatinib (for RET) and larotrectinib (for neurotrophic tyrosine receptor kinase), are preferred first-line treatments. These agents demonstrate superior efficacy and safety compared with conventional multikinase inhibitors, such as lenvatinib and sorafenib. In the absence of specific targets, lenvatinib is the first-line choice, and sorafenib is an alternative. For TC with the BRAF V600E mutation, dabrafenib plus trametinib may be used as first-line in patients intolerant to multi-target TKIs; in those who are tolerant, this combination is reserved for later-line therapy. These kinase inhibitors are organic compounds with small molecular structures that interact with the nucleotide-binding site within the kinase domain, either entirely or in part, and block kinase function (102,103). Depending on the orientation of the activation loop, kinases can adopt either an active or an inactive conformation. A kinase achieves its active state when the aspartate-phenylalanine-glycine (DFG) motif is located at the N-terminus of the activation loop; this configuration is referred to as DFG-in. By contrast, when the DFG motif is found at the C-terminus of the activation loop, the kinase is considered to be in an inactive conformation, referred to as DFG-out (104–106). TKIs are divided into two primary categories based on their preference for DFG-in (Type I) or DFG-out (Type II) conformations. Type I TKIs function by competing with ATP at the active site. By contrast, Type II inhibitors promote the inactive state of kinases by binding both to the ATP-binding pocket and the nearby allosteric site, which can only be accessed in the DFG-out conformation (107,108).

Several multikinase inhibitors (MKIs) exhibit anti-RET activity and are approved for the treatment of TC, including vandetanib (MTC), cabozantinib (MTC), regorafenib and sorafenib (differentiated TC). A variety of highly selective RET inhibitors have been developed. These agents target specific RET mutations and are associated with lower toxicity, lower required doses, and lower discontinuation rates compared with MKIs, such as vandetanib, cabozantinib, regorafenib and sorafenib (109–112).

The LIBRETTO-001 (NCT03157128) trial was an international, multicenter, open-label, phase I/II trial assessing the safety and efficacy of selpercatinib (49). A total of 531 patients aged 12 years were enrolled in the study. All of the patients had locally advanced or metastatic solid tumors of various types. This occurs with the activation of RET alterations. Of the 531 patients, 162 had RET-altered TC. In the TC subgroup, 55 patients had RET-mutated medullary TC (MTC) previously treated with cabozantinib and vandetanib, 88 patients had RET-mutated MTC without prior cabozantinib or vandetanib treatment, and 19 patients had RET fusion-positive non-MTC. Notably, these 19 patients had been previously treated for fusion-positive non-MTC, not for RET-altered MTC. Based on the findings from the clinical trial, RET-mutant patients with MTC who were not previously treated with cabozantinib and vandetanib exhibited an overall response rate (ORR) of 69% and progression-free survival (PFS) rate of 82% after 1 year. By contrast, patients with RET-mutant MTC who had not previously received these treatments had a 73% ORR and a 1-year PFS rate of 92%. For patients with RET fusion-positive non-MTC, the ORR reached 79%, with treatment effectiveness noted across various histological subtypes and a corresponding 1-year progression-free survival (PFS) rate. Furthermore, selpercatinib demonstrated overall treatment effectiveness in all patients with MTC. All patients enrolled in the LIBRETTO-001 trial, including the 162 patients with RET-altered TC described in the text, were treated with selpercatinib, regardless of their previous exposure to MKIs, radioactive iodine, or the specific type of RET mutation or fusion.

Pralsetinib specifically inhibits RET and is an ATP-competitive inhibitor (113). According to the imaging results of the ARROW trial, a single-arm phase I/II clinical study, pralsetinib led to a 90% reduction in tumors among patients with PTC and MTC (114,115). In patients with RET mutations in MTC, the ORR was 63% and the disease control rate was 94%. Furthermore, nine individuals with MTC who experienced brain metastases had an intracranial response rate of 56% (116). Pralsetinib has broad applicability, independent of the RET fusion partner (117). Consequently, owing to these results, the Food and Drug Administration approved pralsetinib for use in treating RET-mutated MTC in 2020, and in March 2021, the National Medical Products Administration approved this drug for use in patients with MTC in China (Fig. 3).

Based on the current research status and clinical practice, it is proposed that future studies on TKIs targeting RET alterations should focus on the following aspects: i) The development of more efficient selective RET inhibitors to overcome existing resistance mechanisms; ii) the exploration of combinations of RET inhibitors with other targeted therapies or immunotherapies to enhance treatment efficacy; and iii) the utilization of technologies such as liquid biopsy to monitor RET mutations and resistance in real time, thereby providing a foundation for personalized treatment.

In conclusion, TC is an increasingly common endocrine malignancy with genetic drivers, particularly RET mutations and rearrangements, which serve key roles in MTC and PTC subtypes, respectively. While targeted RET inhibitors represent a promising therapeutic advance, future research should prioritize deeper molecular studies of RET signaling, the interplay between genetic and environmental risk factors such as radiation, and the development of improved strategies for early detection and precision treatment. Advancing these areas will be critical to enhancing patient outcomes and understanding TC pathogenesis.