1. Introduction

1. Introduction

Thyroid cancer (TC) is the most common malignancy of the endocrine system, and it is estimated that approximately 585,000 new cases were diagnosed globally in 2020, which is about 3% of all cancer cases [1]. Around 75% of patients that are diagnosed with TC are women, with an incidence rate of 10.1/100,000 females per year whereas the incidence rate for males is 3.1/100,000. Over the last three decades there has been a worldwide increase in the incidence of thyroid cancer which may be partly the result of overdiagnosis due to increased use of radiological investigations, including neck ultrasound and cross-sectional imaging; however there is also evidence of an increase in larger clinically significant tumors [2].

TC is divided according to the cell of origin, if it originates from follicular epithelial cells or parafollicular C cells. Parafollicular C cells give rise to medullary thyroid cancer, whereas follicular cells give rise to four histological types: papillary thyroid cancer (PTC), which accounts for 80–85% of TC, follicular thyroid cancer (FTC) accounts for 10–15%, poorly-differentiated thyroid cancer (PDTC) accounts for 1–2%, and anaplastic thyroid cancer (ATC), which accounts for less than <2%. Well-differentiated thyroid cancers (DTC) include PTC and FTC, as well as subtypes like follicular–oncocytic thyroid carcinomas (FTC-OV) and Hurthle cell carcinoma.

DTC is mainly diagnosed at an early stage and is treated with surgery with or without adjuvant radioactive iodine (RAI) treatment and TSH suppression. DTC is one of the most curable cancers, with excellent long-term prognosis, but up to 20% of patients develop local or regional recurrences, and up to 10% develop distant metastases [4]. With longer follow ups, up to 35% of patients with DTC can recur after 40 years, with 20% failing to concentrate iodine, resulting in overall close to 10% of patients dying of their disease [5,6,7]. A study looking at the SEER database showed that independent predictive factors for distant metastasis are tumor size, age at diagnosis, thyroidectomy, node positive disease, T3-T4 stage, and histopathology for female DTC patients [8].

It is worth pointing out that even patients with metastatic disease have a much better prognosis compared with other solid tumors, with the 10-year survival being about 45% due to the use of RAI for metastatic disease. Around 60%, will eventually develop disease refractory to RAI, because cancer cells lose their ability to concentrate iodine. Resistance to RAI is a poor prognostic factor and once patients develop RAIR TC, the five-year overall survival is quoted to decrease down to 10% [9].

2. Molecular Biology of Differentiated Thyroid Cancer

In the pathogenesis of DTC, somatic mutations are involved affecting the RAS (Rat Sarcoma Virus)/BRAF (B-Raf proto-oncogene, Serine/Threonine Kinase), MAPK (Mitogen-Activated Protein Kinase) pathway, and PI3K (phosphatidylinositol 3-kinase)/Akt (Protein Kinase B) pathways. In PTC, beyond mutations in BRAF and RAS, fusions of the RET (Rearranged during Transfection) and NTRK (Neurotrophic Tyrosine Receptor Kinase) transmembrane proteins are also seen. In FTC, the most common events are mutations to RAS and PAX8/PPARy (Paired Box 8/Peroxisome Proliferator-Activated Receptors) rearrangements. During cancer development, proliferation and dedifferentiation, additional mutations that affect the PI3K-AKT pathway, are also generated [10].

Data from the Cancer Genome Atlas (TCGA) project, published in 2014, identified a large number of tumor driver mutations [11]. In fact, in only 3.5% of PTC cases there were no putative pathologic mutations identified. Known driver mutations identified include RAS, RET/PTC, TP53, TRK, PTEN, β-catenin, PAX8/PPARy, BRAF, PIK3CA, BRAF/AKAP9, AKT1, ETV6/NTRK3, and STRN/ALK [12]. According to the TCGA, driver mutations are mostly point mutations accounting for 75% of all cases, while approximately 15% are gene fusions and the rest are gene amplifications.

The TCGA provides evidence that BRAF alterations are the most common causative molecular event in PTCs, and it is found in approximately 58.5% of PTC cases [11]. BRAF V600E is the most common mutation found in the BRAF gene. BRAF mutations are also associated with worse prognosis in terms of overall survival, need of re-operation, and higher risk of local recurrence in patients older than 65 years old [13,14,15]. Also, BRAF mutations are thought to lead to loss of the ability of follicular cells to concentrate iodine uptake, hence leading to dedifferentiated RAIR disease [15,16,17,18]. Finally, it should be borne in mind that BRAF mutations on nodules with inconclusive cytology results from FNA are rarely found in benign neoplasms, since 99% of nodules with BRAF mutations are positive for malignancy on final histopathology [19].

The second most common genetic alteration in DTC are point mutations in RAS family genes (KRAS, NRAS, and HRAS). NRAS mutations are the most common in TC and they are predominantly alterations in exons 12 and 61 [20]. Mutations in RAS genes are identified as driver mutations in approximately 12.7% of cases according to TCGA. They are found especially with follicular histology, and sometimes in patients with PTC too [11]. Most of the cases of RAS-mutated FTCs are follicular variant PTCs. RAS mutation is also found in follicular adenomas of the thyroid, which have a high risk of progressing to malignancy. This suggests that RAS mutations are an early event in the carcinogenesis pathways, and further mutational events are needed for cancer development [20].

RET/PTC rearrangements are accounted for as the causative genetic alteration in 6.3% of PTC, according to TCGA [10]. RET is a transmembrane protein with tyrosine kinase activity to its intracellular domain which regulates both RAS and PI3K-AKT pathways. During RET/PTC rearrangement the protein product remains in a constant activated form, leading to enhanced cell proliferation. They are the result of chromosomal rearrangements and the most common are paracentric chromosomal inversions leading to RET/PTC1 and RET/PTC3 [21]. The prevalence of these mutations is higher in younger patients [22]. RET/PTC is directly correlated with exposure to ionizing radiation, based on data gathered after the Chernobyl nuclear factory accident [23]. RET/PTC rearrangements are more common in patients younger than 45 years old [24].

Another genomic alteration in PTC is the amplification of PIK3CA, present in 53.1% of patients [25]. PIK3CA amplification leads to enhanced signaling of PI3K, which has a fundamental role in thyroid cancer carcinogenesis. PIK3CA is normally activated by tyrosine kinase proteins or by RAS signaling [26]. There are also mutations in PIK3CA in 2% of patients diagnosed with PTC. They are more frequent in FTC, poorly differentiated, anaplastic, and undifferentiated TCs, reaching 10–15% [25,27].

Telomerase reverse transcriptase (TERT) mutations and TP53 mutations were also identified in patients with DTC. TERT mutations lead to amplified telomerase activity and enhanced proliferative potential of cancer cells. TERT mutations are found more frequently in poorly differentiated TC and only account for 1% of DTC, however they are associated with poor prognosis and higher risk of relapse [12]. A synergistic effect as a poor prognostic factor was observed between TERT, RAS, and BRAF mutations, leading to tumors with more aggressive behaviors [28].

TRK fusions are found in 2% of adult patients diagnosed with PTC but are higher in pediatric and young adult PTCs, where they account for about 6–15% of all PTCs [29,30,31]. TRK is normally expressed in embryogenesis and in adult life it takes part in neurological functions such as pain, proprioception, appetite, and memory. TRK fusions lead to overexpression of the chimeric protein, resulting in constitutively active, ligand-independent downstream signaling which promotes carcinogenesis.

ALK fusions are found in about 0.8% of PTCs. Case reports of RAIR thyroid cancer treated with Crizotinib are described in the literature. The most typical histologic appearance of ALK fusion-positive thyroid cancer is PTC, however, it is found with increased relative incidence in up to 4–9% of patients with PDTC [32].

Finally, PAX8/PPARy rearrangements are the result of a fusion of PAX8, which is a transcription factor, with the PPARy gene. This mutation is mainly found in FTC and in a follicular variant of PTC. PAX8/PPARy rearrangements have an incident rate of 30–35% in patients with FTC [33].

3. Radioactive Iodine Treatment (RAI) and Radioactive Iodine Refractory (RAIR) Disease

The hallmark of treatment of DTC in patients with localized and locally advanced disease remains surgery followed by RAI treatment and TSH suppression. RAI is part of the adjuvant treatment for patients with high-risk features. In the adjuvant setting, RAI ablates both residual cancer cells to reduce the risk of cancer recurrence and also ablates remnant thyroid tissue to facilitate follow up with thyroglobulin monitoring [36]. Post-RAI treatment whole body iodine imaging should be assessed in all cases of RAI treatment to evaluate efficacy and evidence of residual thyroid cancer.

RAI is also the standard first-line treatment for recurrent or metastatic disease. The treatment outcome and response to RAI treatment should be assessed with an iodine whole body scan. However, in up to 15% of patients, RAI treatment is no longer effective as a result of loss of the expression of the sodium iodide symporter (NIS), which occurs following loss of thyroid differentiation [37]. This results in no RAI uptake in the post-RAI scan. In patients with iodine-negative post-treatment scans, but with strong clinical or radiological suspicion of recurrence of metastatic disease, there is the option of undertaking an FDG PET/CT scan, because FDG PET/CT scans can identify lesions from tissues that are non-iodine avid [38]. In these patients, the disease is classified as RAI Refractory (RAIR) disease.

There is no global agreement regarding the definition of RAIR. It has been proposed that RAIR can occur in any of the following four conditions: (a) the absence of uptake of RAI in all lesions on scintigraphy, (b) the absence of RAI uptake in some but not all lesions, (c) progression despite uptake of RAI, or (d) reaching the maximum recommended activity of RAI [39].

On the cellular level, iodine is used by follicular cells to synthesize thyroid hormones and enters the cell by passing through the basal membrane, making use of the sodium iodine symporter mechanism (NIS). Once iodine is inside the thyroid follicle, it gets oxidized by TPO (thyroid peroxidase) and then is further processed by TG (thyroglobulin) before becoming T3 and T4 hormones. The whole process is regulated by the thyroid stimulating hormone (TSH) which binds on the TSH receptor (TSH-R). TSH-R activates adenylyl cyclase, resulting in the accumulation of cyclic AMP (cAMP) within thyroid cells. cAMP induces NIS transcription by stimulating thyroid-specific transcriptional factors (TTFs) including paired box 8 (PAX8). Activation of both MAPK/ERK and PI3K/AKT pathways results in inhibition of TTFs and loss of NIS expression [37]. The BRAF V600E mutation has been correlated to the loss of NIS expression and RAIR [40]. BRAF activation represses PAX8 binding to the NIS promoter, and results in dysregulation of involved proteins in this process e.g., NIS, TPO, TG, TSH-R, and thus results in the emergence of RAIR TC.

4. Localized Treatment for RAIR DTC

Not all patients with a rising thyroglobulin and a negative diagnostic iodine scan have RAIR metastatic disease. These patients require further investigations, including conventional neck, brain, thorax, abdomen, and bone imaging, while, increasingly, PET scans are being used to detect recurrent or metastatic disease. Causes of a false negative iodine scan also need to be excluded; the TSH level at the time of the diagnostic scan must be elevated to or above 30 mU/L, and iodine contamination (e.g., history of recent iodine contrast radiography) or a high iodine diet also need to be excluded [41]. Finally, some centers, to be absolutely sure that this is iodine refractory disease, advocate the use of high-dose iodine therapy in patients with raised thyroglobulin and a negative diagnostic iodine scan [41].

Once RAIR has been confirmed, all patients with recurrent RAIR disease, and in the absence of specific contraindications, should have TSH suppression aimed at a serum level of <0.1 μIU/mL [42]. In a small minority of patients with RAIR and isolated recurrence, confirmed on PET to be in the neck or with single or very limited oligometastatic disease, an attempt to offer surgery to render the patient disease-free may be indicated and is normally reserved for patients with excellent performance status and lack of significant comorbidity [42,43].

For patients with more extensive disease or patients with oligometastatic disease who cannot have surgery, other local modality therapies, e.g., radiotherapy or ablative techniques, may be considered with the aim to obtain disease control or palliate disease-related symptoms. Radiofrequency ablation (RFA) can be considered for symptomatic, small (max 3cm) solitary lesions not eligible for surgery or requiring extensive resection [43]. Vertebroplasty can also be considered for patients with vertebral metastases. Bisphosphonates or denosumab should be considered in patients with TC and multiple bone metastases to reduce skeletal related events (SRE) [43]. These decisions should be taken jointly at the Multi-Disciplinary Team (MDT) meeting, after discussion with surgical, medical, and radiation oncologists, alongside nuclear medicine physicians and endocrinologists.

5. Timing of Initiation of Systemic Treatment for RAIR DTC

For patients with multiple foci of metastatic disease, e.g., multiple lung metastases or multiple bone metastases, the question then becomes as to the timing of the initiation of the systemic therapy. An increase in serum thyroglobulin (Tg) levels in the absence of radiologically evident disease progression should not be used to select patients requiring systemic therapy [43]. Instead, the rate of Tg rise or Tg doubling time should be used to guide frequency of imaging during follow up, in conjunction with tempo of disease as judged by previous imaging. For patients with Tg antibodies, these commonly result in false negative Tg results, but they can also produce false positive Tg results [41], thus, physicians should depend on radiological imaging and symptoms for monitoring disease activity/tempo of the disease and not make decisions regarding follow up of patients based on the Tg result.

Care should be exercised regarding the timing of the initiation of systemic therapy for patients with RAIR, as current treatments are non-curative, they are aimed at palliation of symptoms and prolongation of survival; these treatments are essentially life-long and are associated with significant toxicity [34,42,43]. Given that many patients may be asymptomatic and with a slow disease tempo, many patients may not need to start systemic therapy immediately, and instead a period of observation would be indicated. This period of observation in some patients may extend for years, as the natural history of DTC is quite variable, with rates of disease progression ranging from a few months to many years. However, for patients with heavy disease burden, more rapid disease tempo, and who are symptomatic from their disease, initiation of systemic therapy with tyrosine kinase inhibitors is appropriate [42,43].

6. 6. Systemic therapy

6.1 Chemotherapy

Historically, chemotherapy has been used for the treatment of RAIR DTC, however with minimal efficacy. [https://www.mdpi.com/2075-1729/14/1/22#B44-life-14-00022] Most commonly used chemotherapy agent was Doxorubicin. Combination Doxorubicin based therapy did not improve outcomes, so single agent Doxorubicin was the standard of care. [https://www.mdpi.com/2075-1729/14/1/22#B45-life-14-00022]. With the introduction of tyrosine kinase inhibitors (TKIs), there is no further use of chemotherapy for patients with RAIR.

6.2 Tyrosine Kinase Inhibitors

The development of multikinase tyrosine kinase inhibitors (TKIs) with effects on multiple tyrosine kinases and also acting on angiogenesis by inhibiting Vascular Endothelial Growth Factor Receptor (VEGFR) provided new effective drugs with the potential to improve outcomes in terms of symptom control, survival, and improved quality of life. The first multi TKI approved was Sorafenib based on the DECISION trial, a phase 3 Randomized Clinical Trial (RCT) which showed benefit in Progression Free Survival (PFS) benefit compared to placebo (10.8 months in Sorafenib arm versus 5.8 months in control arm). [https://www.mdpi.com/2075-1729/14/1/22#B48-life-14-00022] Lenvatinib, another multikinase TKI, also showed statistically significant benefit in terms of PFS compared to placebo in the SELECT RCT (18.8 months in Lenvatinib arm versus 3.6 in control arm). [https://www.mdpi.com/2075-1729/14/1/22#B55-life-14-00022] Using Lenvatinib at a lower starting dose of 18mg failed to show non-inferiority in a separate phase 3 RCT compared to the standard dose of 24mg. Hence, the starting dose of Lenvatinib 24 mg/day should be continued to be used and clinicians can subsequently adjust the dose as necessary [https://www.mdpi.com/2075-1729/14/1/22#B58-life-14-00022] In conclusion, Sorafenib and Lenvatinib are the preferred TKIs for the treatment of RAIR DTC as first line treatment. Lenvatinib presents optimal efficacy in terms of response, however, sorafenib is mostly better tolerated.

Cabozatinib showed benefit in PFS as second- or third-line treatment for RAIR DTC, in patients who have already received Lenvatinib and or Sorafenib (median PFS not reached in the Cabozatinib arm versus 1.9 months in the placebo arm, HR=0,22). This is based on the results of COSMIC-311 a phase 3 placebo controlled clinical trial. [https://www.mdpi.com/2075-1729/14/1/22#B51-life-14-00022]

Apatinib is a multi TKI that also showed clinical efficacy in RIALITY, a multi-centered placebo controlled, double blind Chinese phase 3 RCT. [https://www.mdpi.com/2075-1729/14/1/22#B52-life-14-00022] Donafenib also presented efficacy in terms of PFS based on the DIRECTION trial. [https://www.mdpi.com/2075-1729/14/1/22#B53-life-14-00022] Currently, Apatinib and Donafenib do not have FDA or EMA approval.

There is evidence for the efficacy of TKIs including Vandetanib, Sunitinib, Axitinib, and Pazopanib based on their prospective phase II studies. Of note that again these drugs do not have FDA or EMA approval. [https://www.mdpi.com/2075-1729/14/1/22#B64-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B60-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B68-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B67-life-14-00022]

6.3 Targeted treatment – RET inhibitors

In the presence of RET fusions patients can receive Selperactinib or Pralsetinib, RET inhibitors. The results were based on open label phase I/II trials which showed high objective response rates in the 79-89% (ORR) and median PFS greater than a year. In Europe, Selperactinib is the only approved RET inhibitor and only for second- or third-line treatment. In USA both Selperactinib and Pralsetinib are approved regardless of previous treatment. [https://www.mdpi.com/2075-1729/14/1/22#B77-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B78-life-14-00022].

6.4 Targeted Therapy - NTRK inhibitors

Currently, there are two options for the treatment of solid tumors with NTRK fusions, Larotrectinib and Entrectinib. Larotrectinib in a phase I/II basket trial showed high ORR (75%) in patients diagnosed with solid tumor with NTRK fusions. Also, Entrectinib showed a high ORR (57%) in a similar design trial. In a pooled analysis from three phase I/II Larotrectinib clinical trials, Larotrectinib shown 71% ORR in the 28 patients receiving this treatment. Both drugs can be used after multikinase TKIs based on tumor agnostic approvals. [https://www.mdpi.com/2075-1729/14/1/22#B80-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B81-life-14-00022]

6.5 Targeted Therapy – BRAF inhibitors

Vemurafenib and Dabrafenib are treatment options for patients diagnosed with RAIR DTC with BRAF V600E mutation. Phase II trials showed encouraging activity prolongation of PFS and high ORR. [https://www.mdpi.com/2075-1729/14/1/22#B82-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B82-life-14-00022] The combination of Dabrafenib plus Trametinib has tumor agnostic approval based on basket trial, for all patients with BRAF V600E solid tumors.

6.6 Redifferentiation Therapy

BRAF mutations dysregulate the capacity of follicular cells for iodine uptake and in fact BRAFV600E mutant thyroid cancers are often refractory to RAI. Re-differentiation of TC with BRAF and/or MEK inhibitors is an emerging treatment option tested in clinical trials to enable the use again of RAI therapy. Selumetnib, Dabrafenib, Vemurafenib and the combination of Dabrafenib with Trametinib have shown promising results in terms of increased RAI uptake and radiological response after the use of RAI in subset of patients where there is RAI uptake after targeted treatment. [https://www.mdpi.com/2075-1729/14/1/22#B85-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B86-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B87-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B88-life-14-00022]

6.7 Immunotherapy

DTC has been considered to be poorly immunogenic due to low Tumor Mutation Burden (TMB), however it is highly infiltrated by immune cells. [ https://www.mdpi.com/2075-1729/14/1/22#B96-life-14-00022] Based on this evidence immunotherapy with immune checkpoint inhibitors has been evaluated in prospective trials but with modest results so far. Pembrolizumab and the combination of Ipilimumab plus Nivolumab were evaluated in phase Ib and phase II studies respectively and showed ORR around 9%, however those were durable responses. [https://www.mdpi.com/2075-1729/14/1/22#B98-life-14-00022, https://www.mdpi.com/2075-1729/14/1/22#B99-life-14-00022] Further studies with the combination of immune checkpoint inhibitors and TKIs are underway.

7. Conclusions

Local DTC is treated with surgery and adjuvant RAI. However, in 20% of cases patients develop recurrence metastatic disease necessitating RAI treatment and eventually 10% of patients become refractory hence develop RAIR DTC.

For patients with progressive or symptomatic RAIR disease, multikinase Tyrosine Kinase Inhibitors (TKI) provide an effective and promising treatment option, with Lenvatinib and Sorafenib having become the standard first-line systemic therapies. These agents should be used by oncologists with experience in their use and management of their toxicity. Cabozantinib is an option to treat patients with RAIR following progression on Lenvatinib or Sorafenib. All patients with RAIR DTC should undergo molecular testing to look at potentially actionable genetic alterations with next-generation sequencing (NGS) in order to evaluate the opportunity of using targeted therapies including RET/NTRK/BRAF inhibitors. Studies of redifferentiation therapy have shown promising results and redifferentiation treatment is an exciting strategy that may help in the future patients with RAIR DTC. Ongoing studies are examining the combination of immune checkpoint inhibitors and TKIs.