Anaplastic thyroid cancer (ATC) is a highly aggressive malignancy that accounts for more than 39% of all thyroid cancer deaths even though it represents less than 2% of all thyroid cancers diagnosed.1,2 Although the incidence of well-differentiated thyroid cancer (WDTC) has increased in recent years, newly diagnosed ATC is decreasing in the United States. This may be because of improvements in histologic techniques, the elimination of endemic goiter, or advances in the treatment of patients with WDTC.3 It is possible that a significant proportion of thyroid lymphomas, medullary thyroid carcinomas (MDCs), and other non-ATC were misdiagnosed as ATC in the past; histologic accuracy has benefited from developments in immunohistochemistry (Figure 7-1).4 Because ATC is associated with WDTC in more than 80% of cases, earlier removal of WDTC could, in theory, reduce anaplastic transformation and ATC incidence.5,6 Conversely, incomplete or delayed therapy for patients with WDTC increases the risk of anaplastic transformation.7

Although no known familial syndromes are associated with ATC, we have observed a probable increase in the frequency of ATC in patients with familial nonmedullary thyroid cancer. Some studies have found associations between ATC incidence and environmental factors. ATC is more common in areas with endemic goiter and low socioeconomic status.8 Furthermore, advanced age is a risk factor because ATC is primarily a disease of elderly people, and it typically occurs after the sixth decade of life.9

Although some of the genetic alterations found in patients with ATC have been described, the essential changes that drive anaplastic transformation remain incompletely understood.10,11 Anaplastic transformation likely entails both the loss of tumor suppressor function and oncogene activation.

Regarding tumor suppressor status, the majority of ATC studies have reported a loss of p53 function.12 Mutations of p53 are common in patients with ATC but rare in those with WDTC, even when they are determined within the same surgical specimen. This suggests that p53 loss is an essential event in anaplastic transformation.5

Regarding oncogene activation, several studies have compared global expression in WDTC with ATC, and others have investigated the expression of known oncogenes from other cancer types in those with ATC. Confirmed alterations in ATC include β-catenin, OEATC-1, Aurora B, c-myc, and NM23.13–15

Most patients with ATC present with a rapidly enlarging neck mass. Patients with long-standing goiters or indolent WDTC may abruptly develop new symptoms and an increase in size of the thyroid gland. These patients most likely had indolent, well-differentiated tumors that underwent anaplastic transformation. Some of the gross and histologic features of ATC are shown in Figure 7-2.

Furthermore, some patients also present with hyperthyroidism from pseudothyroiditis secondary to the rapid destruction of normal thyroid follicles by ATC invasion.16 Because ATC has a tendency to be locally invasive, it is often fixed to surrounding structures; may cause regional pain; and may invade the recurrent laryngeal nerves (RLNs), resulting in vocal cord dysfunction. ATCs are generally large and cause local symptoms from their mass effect, including a sensation of pressure, dysphagia, and dysphonia. They also frequently cause respiratory problems because of their large size, including tracheal obstruction or recurrent or vagal nerve dysfunction.

In our experience with ATC, the average tumor size is greater than 7 cm in diameter, and nearly every patient presents with symptoms that may be attributed to the mass effect from the primary tumor. Clinicians should carry a high index of suspicion for ATC in all elderly patients with a symptomatic or rapidly enlarging neck mass.

Because the prognosis in ATC is generally poor, all ATC is considered stage IV disease. This is also true for occult ATCs that are incidentally discovered during thyroidectomy for WDTC or benign thyroid disease. The 2002 American Joint Committee on Cancer tumor, node, metastasis (TNM) staging system for thyroid cancer subdivides ATC into three different subcategories of stage IV disease. In stage IVa, the primary tumor has not extended beyond the thyroid capsule. Given the strong propensity for local invasion in ATC, stage IVa disease is found only when a small ATC lesion is incidentally discovered in a thyroidectomy specimen or during a radiologic workup for another medical problem. Stage IVb disease represents tumors with local invasion, which often precludes curative resection. Stage IVc disease includes all ATCs with distant metastases. In general, long-term survival is exceedingly rare in patients with stage IVb and IVc ATC.

Distant metastases are common at presentation and may be found in up to 43% of patients.17 The common sites of distant ATC metastases are the lungs, bones, and brain. Although systemic metastases are common, the most frequent cause of death in ATC is caused by the primary tumor site.

Regarding prognosis, results are generally poor, and survival is typically measured on the order of months, rather than years. In a recent comprehensive review of the ATC literature from 1975 to 2002, the median survival from the time of diagnosis ranged from 4 to 12 months.18 In our recent experience at the University of California, San Francisco, from 1981 to 2007, the median survival of patients with ATC was 6 months from the time of diagnosis. We have two patients who have lived without evidence of disease for longer than a decade after thyroidectomy. Both of these patients underwent aggressive multimodality therapy, including surgical extirpation, external-beam radiation therapy (EBRT), and systemic chemotherapy. The tumor specimens from these survivors have been independently reviewed by endocrine pathologists to confirm the diagnosis of ATC. In our experience, both distant and regional metastases preclude curative therapy for ATC patients despite multimodality therapy.

Several studies have identified clinical and pathologic features that are predictive of survival in patients with ATC. Because ATC is uncommon, most reports are small and retrospective. As such, reports have varied with respect to predictive clinical and pathologic determinants of survival for patients with ATC. Although some reports have found white blood cell count, serum albumin, and thyroid function to be independent predictors, multiple studies have implicated patient age, extent of disease, and tumor size as prognostic clinicopathologic markers for survival.19–21

One recent study included 516 patients from the Surveillance, Epidemiology and End Results (SEER) database and found that patients younger than 60 years with tumors without extracapsular extension have significantly longer survival.22 In a recent Korean study of 121 cases of ATC, age younger than 60 years, tumor size less than 7 cm, and lesser extent of disease were independently predictive of lower mortality from ATC.23 In one group’s experience of 67 patients with ATC, young age, absence of symptoms, small tumor size, microvascular invasion, and surgical resection were independent predictors of improved survival.24 In contrast, our group found that the ability to attempt curative surgery, rather than patient age or tumor size, was the only predictive factor for survival in patients with ATC.25

As with any thyroid lesion, investigation should include a detailed history and physical examination. The interview should focus on the patient’s history of endocrinopathy and the potential symptoms that could be attributed to ATC. The physical examination should evaluate for local invasion and metastastic disease. Thyroid function tests are typically unremarkable unless the burden of ATC has replaced most of the normal thyroid gland. ATC can be readily diagnosed by fine-needle aspiration (FNA) biopsy with cytologic review.26

All patients with the diagnosis of ATC by FNA should undergo a detailed workup to determine the extent of disease and the potential for curative resection. Because ATCs tend to be large, ultrasonography is generally not needed for accurate FNA sampling. However, formal neck ultrasonography can determine the local extent of the ATC and the presence of abnormal cervical lymph nodes. Computed tomography (CT) and magnetic resonance imaging of the neck can characterize the extent of locoregional disease and the presence of invasion into the esophagus, trachea, and soft tissues (Figure 7-3A). The evaluation should include direct laryngoscopy to determine if tracheal invasion exists and to evaluate for vocal cord dysfunction from possible invasion of the RLN. Esophagoscopy with or without endoscopic ultrasonography can evaluate for esophageal invasion. Although positron emission tomography (PET) may demonstrate increased [18F]2-fluoro-2-deoxyglucose (FDG) avidity from the primary site and distant metastases (Figure 7-3B), not all ATCs are FDG avid.