Introduction

Over recent decades, the treatment of thyroid carcinoma has evolved due to advancements in diagnostic and therapeutic modalities including the emergence of novel molecular markers and targeted therapies, all of which have changed the landscape of how physicians approach the diagnosis, treatment, and surveillance of thyroid carcinoma.

The frequency of palpable thyroid nodules increases throughout life, with a prevalence of approximately 5% in the United States population for those 50 years of age and older [1]. Thyroid nodules are estimated to be four times more common in females compared to males. The overall prevalence of nodules is estimated to be even greater when taking into account examination of the thyroid gland intraoperatively, during autopsy, and through ultrasonography [1, 2].

While thyroid nodules are a common finding during examination of the thyroid gland, thyroid carcinoma is a more uncommon phenomenon, with a lifetime risk of 1.2% in the United States [3]. Thyroid carcinomas, much like thyroid nodules, are found more often in females than in males. By the year 2030, it is estimated that thyroid carcinoma will become the second most common cancer diagnosis in females and the ninth leading cancer diagnosis in males [4]. In general, mortality rates for thyroid carcinoma are low, with differentiated thyroid carcinomas demonstrating a greater than 90–95% 10-year survival rate. The exception is anaplastic thyroid carcinoma, which has a very poor prognosis [5].

Although thyroid carcinoma is a rather uncommon diagnosis, the incidence of thyroid carcinoma has increased on both the national and global scales. In a study examining thyroid carcinoma incidence and mortality rates using the Surveillance, Epidemiology, and End Results (SEER) cancer registry data from 1974 to 2013, the overall incidence of thyroid carcinoma increased by 3% annually, due in part to the rising annual incidence of differentiated thyroid carcinoma, particularly papillary thyroid carcinoma [6].

The different types of thyroid carcinoma can be generally classified according to histology and cell of origin. The main histologic types include – (1) differentiated thyroid carcinoma (i.e. papillary thyroid carcinoma, follicular thyroid carcinoma, Hürthle cell carcinoma), (2) medullary thyroid carcinoma, and (3) anaplastic thyroid carcinoma [7]. Differentiated, anaplastic, and poorly differentiated thyroid carcinomas arise from endoderm-derived follicular cells, while medullary thyroid carcinoma originates from neural crest-derived C- cells [7]. Furthermore, while thyroid carcinoma can occur sporadically or present as sequelae from prior exposure to ionizing radiation, there are specific genetic or familial syndromes, such as multiple endocrine neoplasia (MEN) 2 A and 2B, Cowden syndrome, familial non-medullary thyroid cancer, and Carney complex, that place individuals at increased risk for thyroid carcinoma [8].

Diagnostic Techniques and Imaging in Thyroid Carcinoma

After decades of increase, rates of thyroid carcinoma incidence are more recently demonstrating a decline in both females and males of 2.5% annually from the years 2014 to 2018 [9]. This decline is attributed to an emphasis on changes to clinical practice with the goal to avoid over-detection.

In 2015, the American Thyroid Association (ATA) published management guidelines advocating for more conservative biopsy criteria [10]. These updated guidelines focus on the use of sonographic risk pattern to help providers determine the need for fine needle aspiration biopsy (FNAB), and is rooted in three main principles: 1) high-risk and low-risk features tend to associate, 2) a pattern of features has a higher sensitivity and specificity than consideration of any singular feature, and 3) intraobserver correlation is preferred for overall sonographic patterns rather than individual sonographic features [10]. These guidelines therefore give way to five malignancy risk patterns: 1) benign, 2) very low suspicion, 3) low suspicion, 4) intermediate suspicion, and (5) high suspicion [10]. Furthermore, the ATA has indicated that diagnostic FNAB is no longer recommended for nodules < 10 mm, including nodules with suspicious sonographic pattern [10].

Similarly, given that thyroid nodules are exceedingly common, which often leads to costly interventions for several lesions that ultimately prove benign, the American College of Radiology (ACR) developed a risk stratification system known as the Thyroid Imaging Reporting & Data System (TI-RADS) in order to provide management guidelines for nodules found incidentally on imaging, to create a universal lexicon to classify thyroid nodules on sonography, and to develop a standardized system based on the lexicon that would inform physicians of which nodules warrant biopsy [11]. The ACR identified five categories to be used to describe thyroid nodules – composition, echogenicity, shape, margin, and echogenic foci. These categories would then collectively produce a TI-RADS level score that stratifies nodules based on need for biopsy [11]. The TI-RADS score is comprised of five levels: TR1 (benign, no FNAB), TR2 (not suspicious, no FNAB), TR3 (mildly suspicious; FNAB if ≥ 2.5 cm, follow if ≥ 1.5 cm), TR4 (moderately suspicious; FNAB if ≥ 1.5 cm, follow if ≥ 1 cm), and TR5 (highly suspicious; FNAB if ≥ 1 cm, follow if ≥ 0.5 cm) [11].

While surgery is recommended for patients with cytology positive for primary thyroid malignancy, patients with very low risk tumors may now undergo active surveillance [10]. Moreover, as of 2017, the U.S. Preventive Services Task Force (USPSTF) recommends against thyroid cancer screening in asymptomatic adults given the relative rarity of thyroid carcinoma, presence of observational data demonstrating no change in mortality over time after implementation of a mass screening program, and evidence suggesting that over-diagnosis and over-treatment are likely to be significant with population-based screening [12].

Molecular Diagnostic Testing in Thyroid Carcinoma

Our understanding of the genetic mechanisms of thyroid carcinomas has significantly increased over the last two decades. Currently, many thyroid carcinomas have known genetic drivers [13]. There have been significant advances in molecular testing that have helped guide preoperative diagnosis and prognostication, operative management, and postoperative treatment of thyroid carcinoma while minimizing overtreatment of benign thyroid disease [13].

Fine needle aspiration biopsy (FNAB) and cytological assessment have been widely incorporated as a basic preoperative diagnostic assessment since the 1980s. While FNAB has decreased diagnostic surgery for benign thyroid nodules, there is a significant number of patients with nodules classified as indeterminate [13]. A meta-analysis review of 11 studies in the United States found a median of 72% (range 62–85%) of FNAB were benign, 5% (1–8%) were malignant, 17% (10–26%) were indeterminate, and 6% (1–11%) were non-diagnostic. In patients with indeterminate cytology, a median of 34% (range 14–48%) were ultimately found to have a malignancy [14].

In 2007, the United States National Cancer Institute (NCI) created the Bethesda classification that separated indeterminate FNAB results into three categories: atypia of undetermined significance (AUS) or follicular lesion of undetermined significance (FLUS), with malignancy in 5–10% of cases; follicular neoplasm (FN) or suspicious for follicular neoplasm (SFN), with malignancy in 20–30% of cases; and suspicious for malignancy, with malignancy in 50–75% of cases [15]. Current recommendations suggest nodules with AUS/FLUS cytology can be managed with observation, repeat FNA, molecular testing or surgical intervention; nodules with FN/SFN cytology should be managed with diagnostic lobectomy or molecular testing to guide further management [16]. Of note, large FNAB studies were conducted utilizing the Bethesda classification. For patients who underwent diagnostic surgery for AUS or FLUS, malignancy was found in 7–48% of cases. This suggested that watchful waiting may not be the appropriate management in many patients with atypia of undetermined significance or follicular lesion of undetermined significance [17,18,19,20,21,22].

Molecular diagnostic testing has allowed for more definitive risk stratification of indeterminate thyroid nodules. Immunocytochemistry (ICC) and immunohistochemistry (IHC) are well-standardized molecular tests. The advantages of ICC/IHC are use of existing cytological sample and minimal cost. The disadvantages are that protein expression is not easily quantitated and ICC staining may be variable. Genetic marker testing can also be employed with advantages including high specificity and positive predictive value (PPV). Gene expression marker testing (mRNA and microRNA) can also be used to predict which nodules are benign versus malignant with the disadvantage of RNA instability. Both genetic marker and gene expression testing may require additional FNAB and are expensive tests. However, analyses of cost effectiveness for both methods have shown a favorable profile [16]. Peripheral blood sampling may also be used to detect common gene mutations, such has BRAFV600E in differentiated thyroid cancer [16, 23].

Genetic marker testing can also be used to guide surgical management and prognostication of thyroid carcinoma. Most thyroid cancer oncogenes activate the mitogen activated protein kinase (MAPK) pathway and/or the phosphoinositide 3-OH kinase (PI3K) pathway [16]. Genomic mutations leading to protein expressions that selectively activate MAPK (BRAFV600E) cause papillary thyroid carcinoma (PTC) while proteins that activate PI3K (PTEN loss) lead to follicular thyroid adenomas and carcinomas [16]. Proteins that activate both MAPK and PI3K pathways can cause PTC or follicular variants of PTC (RET fusions) or follicular thyroid carcinomas (RAS mutations) [16]. For example, the BRAF mutation, typically V600E, is the most common genetic mutation in PTC as well as poorly differentiated and anaplastic carcinomas rising from PTC [13, 24]. It is associated with extrathyroidal invasion, advanced disease stages, lymph node metastasis, and tumor recurrence in PTC [24]. While the presence of BRAF mutation may not be an independent predictor of outcome, research shows tumors with multiple genetic markers present in conjunction (BRAF or RAS mutations with hTERT promoter mutations) have a poor outcome and aggressive course. Thus, while identifying presence of genetic biomarkers is not always sufficient to guide management, presence of combinations of genomic mutations via gene panel testing can help inform cancer risk stratification prior to surgery and guide extent of surgical intervention [13, 14, 16].

There is benefit in utilizing molecular testing in specific thyroid cancer patient populations. In patients with BRAFV600E mutated anaplastic thyroid cancer, research has shown response to therapy with BRAFV600E inhibitor dabrafenib and MEK inhibitor trametinib. Current recommendations are to obtain rapid BRAFV600E testing in all patients with anaplastic thyroid cancer to begin urgent therapy if applicable [16].

Gene expression marker testing, specifically miRNAs, can be used as adjuncts to genetic marker identification or as isolated information to further inform disease development, progression, and prognosis in thyroid cancer. Studies have shown specifically miR-146b can help determine poor prognosis; differentiate malignancy from benign lesions from FNA and plasma samples via quantitative polymerase chain reaction (PCR); be utilized as a biomarker for PTC recurrence; distinguish PTC from FTC and ATC; characterize classic PTC subtypes; and predict central neck lymph node metastasis preoperatively [24].

There are currently three molecular diagnostic tests available in the United States: Afirma Gene Sequencing Classifier (GSC), Thyroseq version 3, and ThyGenNEXT + ThyraMIR. Afirma GSC uses mRNA sequency and analyzes expression patterns of more than 10,000 genes; it can identify parathyroid lesions, medullary thyroid carcinoma, BRAF V600E mutations, at RET/PTC1 or RET/PTC3 fusions. ThyroSeq version 3 uses next general sequencing (NGS) of DNA and RNA to identify 112 genes, 12,135 insertions/deletions, 19 gene expression alterations, 10 copy alterations and more than 120 gene fusions. It can detect hTERT gene associated with poor prognosis as mentioned prior. ThyGenNEXT + ThyraMIR uses NGS to identify 41 mRNA fusion transcripts and 10 DNA mutations [16].

There is significant benefit in continuing to improve our understanding of molecular mechanisms of thyroid carcinoma. With the introduction and increased usage of targeted therapies, these aforementioned molecular pathways may serve as targets for drug treatments. As advancements in molecular testing improve, it offers more accurate and informed preoperative, operative and postoperative management of patients with thyroid carcinoma.

Differentiated Thyroid Carcinoma

Differentiated (i.e. papillary, follicular, Hürthle cell) thyroid carcinoma comprises a majority of thyroid carcinomas, with papillary thyroid carcinoma accounting for approximately 85% of all thyroid cancers, and follicular and Hürthle cell thyroid carcinomas accounting for an estimated 5% overall [7]. Differentiated thyroid carcinoma is typically asymptomatic and often presents as a solitary thyroid nodule.

Due to its often asymptomatic course, the surgical treatment for differentiated thyroid carcinoma must address the risk of under-treatment in high-risk cases and over-treatment in low-risk patients. The importance of this delicate balance is reflected in the updated American Joint Committee on Cancer (AJCC) 8th edition published in 2017, which outlines new staging criteria in an effort to address the physiology of the disease while taking into account patients’ quality of life as well as the effects on the healthcare system as a whole, specifically the delivery of cost-effective treatments due to the downstaging of an estimated 29–38% of patients [25, 26]. Recent changes to the staging criteria for differentiated thyroid carcinoma include: (1) increase in age limit from 45 to 55 years at time of diagnosis and stratification of patients with metastatic disease to lower versus higher risk for death according to age; (2) removal of lymph node metastases and minimal extra-thyroidal extension reported on histology for T3 disease; (3) T3 tumors to now include T3a (tumors > 4 cm but confined within the thyroid) and T3b lesions (gross extra-thyroidal extension into strap muscles); (4) patients > 55 years of age with evidence of N1 disease will be stage II; (5) level VII lymph nodes are now deemed central neck lymph nodes along with level VI lymph nodes; and (6) older patients with differentiated thyroid carcinoma and distant metastases are now stage IVB [27].

It is worth noting that in 2016, the encapsulated forms of follicular variants, formerly known as non-invasive encapsulated follicular variant of papillary thyroid carcinoma (EFVPTC), were re-classified as non-invasive follicular neoplasms with papillary-like nuclear features (NIFT-P) given their benign, indolent nature [11]. This not only addressed the overtreatment of an otherwise indolent lesion, thereby decreasing the number of patients diagnosed with thyroid carcinoma, but it also avoided the “carcinoma” terminology and its associated psychological impact on patients [28, 29].

Given the recent emphasis on personalized medicine and crafted treatment plans tailored to patients’ age, comorbidities, and life expectancy, active surveillance without surgical intervention has increasingly become a suitable alternative for very low risk patients with papillary thyroid carcinoma. Moreover, elderly patients with low-risk tumors may benefit from active surveillance without surgical intervention. Zambeli-Ljepović et al. conducted a study using the SEER-Medicare database to examine 3,341 patients ages 66 years and older who were treated between 1996 and 2011 for papillary thyroid carcinoma with lesions ≤2 cm, of which 67.6% underwent total thyroidectomy and 32.4% underwent lobectomy, and found that total thyroidectomy was associated with readmissions and complications, such as higher rates of hoarseness, hypocalcemia, and hypoparathyroidism, particularly for female and black patients [30].

A substantial proportion of papillary thyroid carcinomas present as a papillary microcarcinoma or follicular variant thyroid microcarcinoma, both of which often have an indolent course of disease. Therefore, active surveillance has been utilized by several institutions for thyroid nodules measuring up to 1.5 cm in diameter. Similarly, Sakai et al. conducted a prospective trial of active surveillance for T1b tumors compared to those with T1a tumors, and found that there were no significant differences in the development of lymph node metastasis or increase in tumor burden between groups, though younger age, non-calcification pattern, and rich vascularity were risk factors for the development of lymph node metastasis and increase in tumor burden, respectively [31]. These findings suggest that active surveillance could safely be utilized for selected patients with T1b tumors [31].

While active surveillance may be appropriate in certain circumstances given the indolent nature of papillary microcarcinoma, there still remains contraindications for active surveillance. Contraindications to active surveillance include high risk features, such as involvement of the trachea or recurrent laryngeal nerve, clinical lymph node or distant metastasis at the time of diagnosis, and high-grade malignancy seen on cytology [32]. Nonetheless, active surveillance could potentially be first line therapy for the treatment of papillary microcarcinoma as it not only protects against potential post-operative complications, but also is a cost-effective option for healthcare systems [32].

The current standard of care for differentiated thyroid carcinoma involves surgery. Among those who are undergoing active surveillance, surgery should be considered if the lesion increases in greatest dimension by at least 3 mm; tumor volume increases by 50%; or new suspicious metastatic lymph node disease develops with positive cytology [33, 34]. Surgical options include lobectomy/subtotal thyroidectomy for smaller lesions and near total/total thyroidectomy with or without lymph node dissection for more advanced lesions [35]. However, the decision between ipsilateral lobectomy versus total thyroidectomy remains controversial for lower-risk papillary carcinoma. The National Comprehensive Cancer Network (NCCN) generally recommends total thyroidectomy for biopsy-proven papillary carcinoma with significant N1 metastases given its associated improved disease-free survival as well as within the context of reports that patients undergoing only lobectomy demonstrate a 5–10% recurrence rate in the contralateral lobe and an overall long-term recurrence rate greater than 30%; this is in contrast to an overall long-term recurrence rate of 1% after total thyroidectomy [36,37,38].

Unilateral lobectomy is favored by some prominent thyroid specialists, including the NCCN guideline recommendations, for patients with papillary and follicular carcinoma due to the low mortality associated with the disease and the high post-operative complication rates with thyroidectomy [35, 39, 40]. According to the ATA 2015 guidelines, patients with thyroid cancer > 1 cm and < 4 cm, without evidence of extrathyroidal involvement or lymph node metastases (cN0), the initial surgical approach may be either an unilateral lobectomy or near-total or total thyroidectomy [35]. While lobectomy alone may be appropriate initial treatment for low risk papillary and follicular carcinomas, the treatment team may elect to perform a total thyroidectomy to allow for radioactive iodine (RAI) therapy or to increase follow-up based on disease features or patient preferences [35]. For patients with thyroid cancer > 4 cm, or with gross extrathyroidal involvement (clinical T4), or clinically positive metastatic nodal disease (clinical N1) or metastatic disease to distant sites (clinical M1), the initial surgical treatment should be a near-total or total thyroidectomy and gross removal of all primary tumor unless contraindications exist [35]. For those with thyroid cancer < 1 cm, without evidence of extrathyroidal involvement or cN0 disease, who elect to undergo surgery, should be recommended unilateral lobectomy unless there are indications to remove the contralateral lobe [35]. Small, unifocal, intrathyroidal carcinomas may be adequately treated with lobectomy alone given that the patient does not have a prior history of head and neck radiation, familial thyroid carcinoma, or clinically detectable cervical lymph node metastases [35].

Per the NCCN guidelines, completion thyroidectomy is recommended for lesions > 4 cm, positive resection margins, gross extrathyroidal extension, macroscopic multifocal disease or nodal metastases, contralateral disease, or vascular involvement [40]. However, the NCCN guidelines recommend total lobectomy without radioactive iodine (RAI), and not completion of thyroidectomy, for papillary carcinoma with incidental small-volume pathologic N1A metastases (< 5 nodes involved with no metastasis > 5 mm in largest dimension) [40]. Thyroid lobectomy would also be adequate for noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) as well as minimally invasive follicular thyroid carcinomas [40].

Patients often undergo completion thyroidectomy in anticipation of remnant ablation, as large thyroid remnants are difficult to ablate, or if serum thyroglobulin with or without whole body iodine-131 imaging will be used for long-term monitoring [40]. Additionally, Grigsby et al. sought to investigate the rate of contralateral papillary thyroid carcinoma in patients with low-risk papillary thyroid carcinoma who had undergone thyroid lobectomy, followed by completion thyroidectomy and RAI, and found that of the 150 patients, 41% had papillary thyroid carcinoma in the contralateral lobe; however, there were no recurrence or survival differences in those with versus without contralateral disease following resection and RAI [41]. Similarly, Pasieka et al. evaluated patients with presumed unilateral well-differentiated thyroid cancer who had undergone unilateral thyroid lobectomy, and found that 62% of the 60 patients in the study who had undergone completion thyroidectomy were found to have residual carcinoma in the contralateral lobe [42].

In certain circumstances, iodine-123 or low-dose iodine-131 whole body imaging is recommended post-surgery for determination of completeness of thyroidectomy and to aid in identification of any residual disease, such as nodal involvement or distant metastases [40]. However, a phenomenon known as “stunning” may occur when imaging doses of iodine-131 cause follicular cell damage, decreasing uptake in the thyroid remnant or sites of metastases, which would compromise the therapeutic efficacy of subsequent iodine-131 for remnant ablation or adjuvant therapy to target suspected micrometastases [43, 44]. Per the 2015 ATA guidelines, the potentially negative impact of such diagnostic imaging with iodine-131 on RAI therapeutic efficacy for remnant ablation can be avoided by using iodine-123 or low-dose iodine-131 [35]. RAI uptake with iodine-123 can also be used to help calculate the dose for iodine-131 therapy [45]. The 2015 ATA guidelines recommend RAI therapy for high-risk patients with RAI-avid disease and selected patients with intermediate risk disease [35]. RAI therapy is not routinely recommended for low-risk cases, per ATA guidelines [35]. Of note, empirically administered doses of RAI to treat metastatic differentiated thyroid carcinoma may lead to suboptimal dosing with little tumoricidal effects and increased adverse effects [35].

Medullary Thyroid Carcinoma

Medullary thyroid carcinoma comprises approximately 3–5% of all thyroid malignancies. An estimated 75% of cases are sporadic, while about 25% are attributable to genetic or familial syndromes, such as MEN 2 A, MEN 2B, and familial medullary thyroid carcinoma [46, 47]. Per the 2015 ATA guidelines, total thyroidectomy is recommended with cervical lymph node dissection pending intra-operative, serological, or imaging findings [47]. Specifically, for tumors ≥1 cm or bilateral disease, total thyroidectomy and bilateral central neck dissection (level VI) are indicated; tumors < 1 cm or unilateral disease is treated with total thyroidectomy with consideration for neck dissection [40].

Notably, if a patient with hereditary risk for medullary thyroid cancer is diagnosed early, the recommendation is a prophylactic total thyroidectomy by age 5 years, or for older patients, when the mutation is identified, particularly those with codon 609, 611, 618, 620, 630, or 634 RET mutations [47]. Additionally, the presence of certain RET mutations confer a higher risk for medullary thyroid carcinoma. Patients with MEN2B with codon 883, 918, or compound heterozygous RET mutations are recommended to undergo a total thyroidectomy within the first year of life or at time of diagnosis as these mutations are associated with the highest risk for medullary thyroid carcinoma [40]. However, patients with lower risk mutations (i.e. 768, 790, 791, 804, and 891 RET mutations) may be monitored with annual basal calcitonin levels and yearly ultrasound. If these annual tests are within normal limits and there is no family history of aggressive medullary thyroid carcinoma, then total thyroidectomy and central node dissection may be deferred [40].

Although surgery is the mainstay of treatment for medullary thyroid carcinoma, there are variations in strategy to be considered depending on the need for concurrent parathyroid resection for hyperparathyroidism or the level of risk for locoregional node metastases [47]. While bilateral central neck dissection (level VI) may be performed for MEN2B patients, MEN2A patients who are post-prophylactic thyroidectomy are recommended to undergo therapeutic ipsilateral or bilateral central neck dissection (level VI) if they exhibit elevated calcitonin or carcinoembryonic (CEA) levels, or if they have an abnormal thyroid ultrasound. In addition, more involved lymph node dissection (levels II-V) may be warranted for tumors ≥1 cm (> 0.5 cm for MEN2B patients) or central compartment lymph node metastases [40].

It is imperative to note that pre-operatively, patients should be examined for signs of hyperparathyroidism and pheochromocytoma, whether the patient appears to have sporadic or familial medullary thyroid carcinoma, as the possibility of MEN2 prompts testing for the germline RET proto-oncogene mutation. In the event that the patient has concurrent hyperparathyroidism secondary to multiglandular hyperplasia, excess parathyroid tissue should be removed and the equivalent mass of one normal parathyroid gland should remain or be auto-transplanted. For those in whom medullary thyroid carcinoma is diagnosed post-surgically, additional workup is needed to determine the need for future surgical management, including completion thyroidectomy with or without neck dissection. In the event that a pheochromocytoma is detected, it should be surgically removed prior to thyroid surgery to avoid hypertensive crisis [40].

Patients with extensive local disease, residual disease, or extra-thyroidal involvement may be recommended external beam radiotherapy (EBRT) or intensity-modulated radiation therapy (IMRT) to the neck [47]. Therapeutic EBRT or IMRT can be used for grossly incomplete resections when additional surgical resections are contraindicated or for palliation of bone metastases. Adjuvant EBRT or IMRT is rarely recommended [40].

For cases involving symptomatic recurrent disease or unresectable progression, the NCCN recommends the use of systemic therapy or enrollment in a clinical trial [40]. As RET is a type of tyrosine kinase receptor, tyrosine kinase inhibitors (TKIs), such as vandetanib and cabozantinib, may be appropriate for patients with locally advanced disease, recurrent disease that is unresectable, or symptomatic metastatic disease, and generally may not be appropriate for patients with stable or slowly progressive disease [40, 47]. Several clinical trials are underway to investigate the potential utility of TKIs. In a phase III randomized clinical trial conducted by Wells et al., patients with either locally advanced, metastatic, or inoperable medullary thyroid carcinoma who were treated with vandetanib were found to have increased progression-free survival compared to the placebo group [48]. This finding, along with others, led the Food and Drug Administration (FDA) to approve the use of vandetanib for patients with locally advanced or metastatic disease not otherwise amenable to surgery [49]. Similarly, Elisei et al. conducted a phase III randomized clinical trial whereby patients with locally advanced or metastatic medullary thyroid carcinoma were treated with either cabozantinib or placebo, and found that cabozantinib led to increased median progression-free survival compared to the placebo group; this led the FDA to approve cabozantinib for those with progressive, metastatic disease [50].

Of note, medullary thyroid carcinoma cells do not concentrate RAI and also do not demonstrate an effective response to cytotoxic chemotherapy, thus iodine-131 imaging and RAI therapy are not utilized in the treatment of medullary thyroid carcinoma. Post-operative levothyroxine is indicated in all patients and thyroid-stimulating hormone (TSH) should be within the normal range as suppression is unnecessary as C cells lack TSH receptors [47].

Anaplastic Thyroid Carcinoma

Anaplastic thyroid carcinomas are highly aggressive, undifferentiated tumors with a very poor prognosis, with a median survival of 5–12 months and a 1-year survival rate of 20–40% [7, 51]. This disease is relatively uncommon and accounts for approximately 1.7% of all thyroid malignancies in the U.S. and an estimated 3.6% of all global cases of thyroid carcinomas, with a mean age at diagnosis of about 71 years [51]. All patients with anaplastic thyroid carcinoma, regardless of T, N, or M, are considered to have stage IV disease [51]. Though the incidence of this disease is declining due to improved treatment of differentiated thyroid carcinoma and increased dietary iodine supplementation, anaplastic thyroid carcinoma has a propensity to be associated with pre-existing goiter or differentiated thyroid carcinoma [51]. Due to the undifferentiated nature of these lesions, iodine-131 imaging typically cannot be utilized.

As this disease process responds poorly to conventional therapy and RAI therapy is ineffective, it is crucial that palliative care and support is initiated early [52]. Death is typically secondary to either upper airway compromise, often despite presence of a tracheostomy, or complications related to disease spread or treatment. Surgical resection may be appropriate depending on extent of disease and experience of the surgeon. However, most cases are unresectable or have evidence of metastatic disease. If surgical management is appropriate, total thyroidectomy with complete gross tumor resection should be attempted, along with excision of all involved locoregional structures and nodes, followed by local/regional radiation with or without systemic chemotherapy [51, 52]. For disease confined within the thyroid, near-total or total thyroidectomy accompanied by therapeutic lymph node dissection is appropriate, while disease with extra-thyroidal involvement may call for an en bloc resection [51]. Disease not amenable to resection and with no extra-thyroidal extension may be initially treated with radiation with or without systemic chemotherapy with the hopes for an appropriate clinical response in tumor burden to facilitate resection [51].

EBRT or IMRT can be used to improve local control and palliation, and may even increase short-term survival in some patients [52, 53]. Saeed et al. conducted a retrospective analysis using the National Cancer Database and examined patients with non-metastatic anaplastic thyroid carcinoma who had undergone non-palliative resection, and found that adjuvant radiation therapy, in conjunction with concurrent chemotherapy, is associated with improved survival [54]. Of note, in cases of unresected disease or incomplete resection, radiation therapy should be implemented as soon as possible. Similarly, adjuvant radiation therapy should be started ideally 2 to 3 weeks post-operatively for R0 or R1 resection [40].

NCCN guidelines recommend molecular testing to aid in selecting the appropriate systemic therapy, particularly targeted therapies [40]. While anaplastic thyroid carcinomas involve mutations in multiple genes, BRAF and RAS are generally more common mutations [51]. The combination of dabrafenib and trametinib is an option for BRAF V600E mutation-positive tumors, and was approved by the FDA for anaplastic thyroid carcinoma with BRAF V600E mutation after a phase II, open-label trial demonstrated an overall response rate of 69%, as well as a direction of response and overall survival of 80% and 90%, respectively [55]. Furthermore, other targeted therapy options include larotrectinib or entrectinib for NTRK gene fusion-positive tumors; and selpercatinib or pralsetinib for RET fusion-positive disease [56, 57]. Monotherapy with paclitaxel or doxorubicin can also be used [52].

Conclusions

The management of thyroid carcinoma is constantly evolving with the advent of new diagnostic modalities and management options, including targeted therapy treatments, all of which help to enhance patient-centered care and emphasize the importance of patient-tailored surgical and medical therapies. While existing guidelines create a foundation upon which current treatment algorithms are rooted, several novel therapeutic strategies have emerged that have not only improved overall survival, but also pushed the boundary of what is known of the molecular landscape of thyroid carcinoma. These continuing improvements, in conjunction with surgical management, pave the way for creating treatment methods that will further transform care of thyroid carcinoma patients and improve quality of life for these patients.