Recent advances and emerging therapies in anaplastic thyroid carcinoma

Introduction

Anaplastic thyroid carcinoma (ATC) is a rare form of thyroid cancer and one of the most aggressive cancers in humans. In part because of ATC’s rarity and short overall survival, very little progress has been made in the treatment of this disease until recently. An understanding of the molecular aberrations in this disease has advanced the field, leading to one of the most important treatment discoveries—BRAF targeting in BRAF V600ELN876569-mutant ATC. While this is an exciting discovery and represents substantial progress, only 25–45% of ATCs harbor a BRAF mutation^1–3, and this treatment is not curative. However, a renewed interest in this disease will likely lead to a deeper comprehension of the mechanisms by which these tumors survive, that one day may result in better therapies and eventually a cure for this devastating disease.

Background

The usual clinical presentation of ATC is of a patient with a rapidly expanding neck mass that may lead to compressive symptoms such as dysphagia and dyspnea. Patients may also have hoarseness due to vocal cord paralysis secondary to laryngeal nerve paralysis that can progress to stridor. However, some patients may have few symptoms until the disease is quite advanced and, therefore, most patients with ATC are not surgical candidates. Furthermore, approximately 46% of newly diagnosed patients present with widely metastatic disease at diagnosis⁴. Patients may give a past history of long-standing goiter, thyroid nodules, or prior thyroid cancer. According to one series, in more than 80% of cases, there is a co-existing differentiated thyroid cancer (DTC) giving rise to a less differentiated or undifferentiated tumor⁵.

The diagnosis of ATC can be made on fine needle aspiration, core biopsy, or surgical resection material. Fine needle aspiration is often the first attempt at biopsy, since this is the standard of care for thyroid nodules and suspected thyroid cancers. However, if the diagnosis of ATC is suspected, a core needle biopsy is preferable in order to perform genetic mutational analysis of the tumor. Many patients are not suspected to have ATC and are diagnosed only after undergoing thyroidectomy when the surgical pathology is reviewed.

The differential diagnosis of ATC includes lymphoma, poorly differentiated thyroid carcinoma (PDTC), poorly differentiated medullary thyroid carcinoma (MTC), squamous cell carcinoma from an adjacent site, and metastasis from other solid tumors. The diagnosis can be challenging because of the frequent loss of thyroid (TTF-1 and PAX8) and epithelial cell markers (cytokeratins). Morphologically, ATC is highly variable, from sarcomatoid, squamoid, and epithelioid to pleomorphic/giant cell cytologic features. ATC often grows as discohesive sheets of overtly malignant cells with high-grade features including marked necrosis and mitoses. An associated mixed inflammatory component is also present in the background.

According to the American Joint Committee on Cancer (AJCC) thyroid cancer staging system⁶, all ATC patients are considered to have stage IV disease owing to the extremely high mortality rate. Stage IV disease is further broken down into stage IVA (intrathyroidal and surgically resectable without distant metastatic disease), stage IVB (extrathyroidal, with or without lymph node metastases but without distant metastatic disease), and stage IVC (distant metastatic disease at presentation).

Overall prognosis for ATC remains poor despite recent advances. The median overall survival remains around 3–5 months with a 1-year survival of approximately 20%⁷. Patient factors associated with poorer prognosis include advanced age (>60–70 years), male gender, presence of leukocytosis (>10,000), and symptoms (such as rapidly growing tumor, pain in the neck, dyspnea, dysphagia, and hoarseness). Histologically, larger tumor size (>5 cm), extrathyroidal invasion, giant cell and pleomorphic pattern, and/or presence of distant metastases are tumor factors that can portend a poor survival and predict treatment failure. Tumors with co-existing well-differentiated papillary thyroid carcinoma appear to fare better^2,7–9. In terms of treatment factors, patients with tumors that are surgically resectable and who are able to undergo external beam radiation have improved survival^4,8–16. The dose of external beam radiation is also prognostic, with patients who are able to undergo higher doses (>60 Gy) surviving longer, even those who have not undergone complete resection^12,17. Somatic tumor mutations may also affect outcomes. ATC tumors with TERT promotor mutations had a significantly shorter overall survival, particularly when these co-existed with either BRAF or RAS mutations¹. In another study, ATC patients with TP53 mutations were shown to have a trend towards shorter time to failure after primary treatment².

Standard treatment for patients with stage IVA disease and some IVB patients is curative intent surgery, as this has been associated with longer survival^{10,14,18–20}. Surgery should be performed by a surgeon with experience in advanced thyroid cancer. Post-operative radiation with radiosensitizing chemotherapy should be started within 4 weeks of surgery. Chemotherapy regimens are detailed in the American Thyroid Association guidelines for the management of patients with ATC²¹. Unfortunately, surgery is not possible in the majority of patients, as most present with locally advanced and invasive disease. In general, if the patient has unresectable IVB disease, chemoradiation can be considered with the goal of local control. If the radiation field is too large or the chemoradiation treatment deemed too morbid for the patient, systemic therapy can be considered in these patients, particularly the novel targeted treatments discussed in the next section. Patients with IVC disease are the most challenging to manage, as balancing the need to control both local and distant metastatic disease can be difficult. Treatment guidelines recommend palliative radiation to control locoregional disease; however, our experience has been that palliative radiation provides little control. Thus, our practice is to treat these patients on clinical trials (preferable) or outside a trial with novel targeted therapies. A better understanding of the genetics of thyroid cancers has led to the use of novel therapies for ATC, which is the theme of this manuscript.

Genetics of anaplastic thyroid cancer

The MAP kinase and PI3 kinase pathways are commonly activated in thyroid cancer. The common driver mutations that are seen in ATC are the same as those in DTC, as these tumors are usually derived from this class of tumor, particularly papillary, follicular, and Hurthle cell thyroid cancers. Thus, BRAF, NRAS, KRAS, and HRAS are common driver mutations seen in ATC. Mutations in tumor suppressor genes include p53, NF1, and PTEN. Secondary mutations such as PIK3CA, TERT promotor, CTNNB1, EIF1AX, MTOR, CDKN2A, and AKT1 can co-exist with driver mutations and tumor suppressor gene mutations. Figure 1 shows the frequency of point mutations in ATC. The genetic rearrangements observed in papillary thyroid cancer, RET/PTC and NTRK, have also been described in ATC. Most recently, ALK and ROS1 genetic rearrangements have also been described^22,23.

Figure 1. Distribution of oncogenic mutations in anaplastic thyroid carcinoma.

Data compiled from three sources: COSMIC (cancer.sanger.ac.uk)³, Kunstman et al.²⁴, and Landa et al.¹. Genetic rearrangements are not represented on the graph.

Of all of the genetic mutations that occur in ATC, BRAF V600E is the most common (25–48%^1–3,24) actionable mutation, and therefore the most is known regarding novel therapies that target this mutation in ATC. However, other rare mutations and genetic aberrations may also prove to be actionable, such as ALK translocations. Unfortunately, at this time, the majority of mutations in ATC are not actionable, but targeting of the tumor microenvironment or common pathways is an alternative approach. Figure 2 shows drugs and drug targets of interest in thyroid cancer.

Figure 2. Schematic overview of the two most common pathways with genetic alterations in thyroid cancer—mitogen-activated protein kinase (MAPK) and phosphatidylinositide 3-kinase (PI3K) pathways.

The pathways can be activated by receptor tyrosine kinases (RTKs) being activated by growth factors or by overexpression of the receptor. Constitutive activation of the signaling pathways can be caused by mutations along the pathway, such as BRAF, RAS, PI3K, AKT, or mammalian target of rapamycin (mTOR), mutations in tumor suppressor genes, such as neurofibromin 1 (NF1) or phosphatase and tensin homolog (PTEN) mutations, or fusion RTKs, such as those in RET, neurotropic tropomyosin receptor kinase (NTRK), or anaplastic lymphoma kinase (ALK). Drugs that can inhibit at critical points in the pathway are listed in grey boxes. FGFR, fibroblast growth factor receptor; TSC1, tuberous sclerosis complex 1; TSC2, tuberous sclerosis complex 2; VEGFR, vascular endothelial growth factor receptor.

Novel and promising therapies for ATC

Summary of the evolving management and future directions in ATC

Owing to the staggeringly high failure rate in ATC and recent renewed interest in this disease, new paradigm shifts are rapidly evolving and being adopted by many leaders in the field. While routine molecular testing of ATC tumors is not currently in treatment guidelines and is controversial, our group feels strongly that all of these patients’ tumors should be tested upon diagnosis, at minimum, to determine BRAF status. Patients with BRAF mutations can be offered treatment with combination selective BRAF plus MEK inhibitor therapy. Some patients with initially unresectable disease may become resectable and could be considered for surgery if they respond to BRAF/MEK combination therapy. Patients who are well enough and want to consider clinical trials should have a broader panel of mutations and fusions interrogated in order to identify those who would be appropriate for selective inhibitor trials such as the selective NTRK or RET inhibitor trials. Other driver mutations that are identified may help identify a clinical trial that is appropriate for the patient, particularly as more targeted therapies become available. Figure 3 shows a proposed algorithm for selecting treatment in ATC patients and how the molecular information of the tumor can be used to help guide therapy, particularly in patients with a BRAF V600E mutation, since currently the most promising treatments are in these patients.

Figure 3. Proposed algorithm to determine best treatment in anaplastic thyroid carcinoma (ATC).

The most critical steps that lead to important decision points are in colored boxes and should be performed in tandem. Brown box: the airway should also be assessed by laryngoscopy. An unstable airway will require immediate management, which might include steroids, intubation, tracheostomy, or hospice if the patient has a poor performance status. Green boxes: full staging with cross-sectional, contrast-enhanced imaging of the brain, neck, chest, and abdomen as well as bone scan or PET/CT (preferable) should be performed in order to determine if there are distant metastases. If brain metastases are present, these may require radiation or surgery prior to any therapy. Based on the staging images and the patient’s performance status, the patient should be evaluated to determine if the tumor is resectable and a good surgical candidate. Blue box: a rapid test to determine BRAF status should be performed as soon as possible. Patients with a stable airway can be triaged to BRAF-directed therapy if a BRAF V600E mutation is present. If there is a treatment response that leads to resectability, surgery by an experienced surgeon may be considered for patients with low burden or no distant metastatic disease. In the absence of a BRAF V600E mutation, external beam radiation to the neck or systemic therapy on clinical trial is recommended. The decision between systemic therapy and external beam radiation in the BRAF wild-type subgroup will depend on local versus distant disease burden.

However, in order for patients to begin therapy before they deteriorate clinically, it is necessary to be able to rapidly test patients for pertinent mutations. The gold standard of molecular testing with next-generation sequencing can take over 1 month before results are known (far too long for patients with a rapidly lethal disease). Thus, the use of liquid biopsy, the testing of blood for circulating cell-free tumor DNA, which has been shown to be highly sensitive for identifying BRAF mutations in circulating DNA, should be employed routinely in patients who do not have available tissue specimens⁵². In those with available tissue, immunohistochemistry for BRAF V600E on surgical specimens or core needle biopsy has proven to be a useful, rapid tool to identify BRAF-mutated ATC patients⁵³.

Resistance to kinase inhibitors remains a major obstacle in the treatment of ATC. Thus, moving away from single-agent therapy for ATC is one step in the right direction. Combinations of targeted therapy with immunotherapy may help overcome resistance and are currently being studied in clinical trials. Radiation plus systemic therapy with either immunotherapy or targeted therapies are also promising strategies for ATC patients. Also, due to the very high recurrence rate after definitive treatment with surgery and radiation, a systemic adjuvant approach should be studied in this setting.

The future is much more promising for ATC patients; however, many physicians continue to recommend hospice as the only option for these patients. Our hope in the current environment with potential promising targeted therapies is that more patients are aware of and will participate in clinical trials, particularly those whose tumors are molecularly driven. Increasing medical and public awareness of expanding clinical trial options for ATC, coupled with enhanced availability of tumor sequencing, may prime a cycle for rapid enrollment and outcome data in this rare disease.