HTC, first described in 1894 by the pathologist Hürthle, is a rare tumor and is considered an oncocytic variant of FTC, accounting for 3–7 % of all thyroid carcinomas [36]. Little is known about its biological characteristics or clinical behavior. When Hürthle cells comprise more than 75 % of an encapsulated nodule, the lesion is classified as a Hürthle cell neoplasm, either an adenoma or carcinoma. As in FTC, there is no reliable preoperative diagnostic test to differentiate HTC from Hürthle cell adenoma [37]. The low incidence and lack of long-term follow-up data have caused controversy regarding its survival characteristics [38]. This poses a diagnostic and therapeutic challenge. HTC has different biological behavior – usually not observed in FTC – like nodal metastasis, intrathyroidal spread, and poor ¹³¹I uptake [39, 40]. The survival is described to be around 74 and 49 % at 5 and 10 years, respectively [40]. Pulmonary metastases are most common, but metastases also occur in the bone, liver, and central nervous system. Literature shows that about 67 % of patients with metastatic HTC have lung metastases, 58 % soft tissue metastases, 33 % nodal metastases, 17 % bone metastases, and 17 % brain metastases [40]. Like patients with PTC and FTC, sorafenib is also active in patients with advanced HTC [35].

Medullary thyroid carcinoma (MTC) is an oncological entity on its own and accounts for 5–8 % of all thyroid cancers, originating from the calcitonin-secreting C cells of the thyroid [3, 44]. MTC is mainly sporadic in nature, but hereditary patterns (multiple endocrine neoplasia type 2 (MEN type 2) and familial medullary thyroid carcinoma (FMTC)) are present in 20–30 % of patients. These genetic diseases are inherited in an autosomal-dominant pattern and are caused by mutations of the rearranged during transfection (RET) proto-oncogene [41]. The primary treatment of both hereditary and sporadic forms of MTC is surgery by means of total thyroidectomy and extensive lymph node dissection. Locoregional recurrence frequently occurs in MTC, and tumor control may be improved by standard central, bilateral, and upper mediastinal neck dissection [42]. In MTC patients, extrathyroidal extension and stage at diagnosis are the only independent predictors of (recurrence-free) life expectancy. Patients diagnosed in an early stage of disease and patients without detectable recurrence have a favorable life expectancy independent of biochemical cure [43]. Distant metastases are the main cause of death in patients with MTC and in 50 % they present synchronous and at multiple sites [41]. Metastases often simultaneously affect various organs, such as liver, lungs, and bones and, less frequently, brain and skin. Survival is 51 % at 1 year, 26 % at 5 years, and 10 % at 10 years. However, the prognosis can be variable and long-term survival of more than 15–20 years has been described [41, 44]. In the follow-up of MTC, measurements of serum calcitonin and carcinoembryonic antigen (CEA) are of great importance as they reflect the presence of persistent disease. Surgery is the only potential curative treatments for distant metastases but usually multiple distant metastases require systemic therapy [44]. Recently, vandetanib was studied in a phase III trial for advanced MTC and demonstrated therapeutic efficacy with a median PFS of 19.3 months in the placebo group and 30.5 months in the treatment group [45]. When MTC impairs quality of life caused by hyper-calcitonemia-associated diarrhea or pain from local tumor growth, a need might arise for efficient tumor reduction [46]. In patients with predominant liver involvement, thermoablation, embolization, or transarterial chemoembolization (TACE) can reduce tumor mass and ameliorate symptoms [41, 47].

ATC is the most lethal type of thyroid cancer, with an overall median survival time of less than 6 months from diagnosis. Anecdotal cases of long-term survivors have been reported, mainly in patients with intra-thyroid tumors at the time of surgical treatment [48]. Almost all patients are classified as stage IV, with the primary lesion restricted to the thyroid gland in stage IVA; locoregional lymph nodes may exist in IVA/IVB, whereas stage IVC disease is defined by distant metastases [49]. Besic et al. studied the sites of metastases of ATC in an autopsy series of 45 patients and observed metastases in 41 cases (91 %). The most common sites of metastases were the lungs (78 %), intrathoracic lymph nodes (58 %), neck lymph nodes (51 %), pleura (29 %), adrenal glands (24 %), liver (20 %), brain (18 %), heart (18 %), and retroperitoneal lymph nodes (18 %) [50]. Causes of cancer-related mortality are distant metastases in 51.5 % (range 12–68 %), local complications in 23.7 % (5–37 %), or both in 26.2 % of patients (25–51 %) [48, 49, 51]. There is no place for surgery for distant metastases in ATC as it does not contribute to survival. The main aim of treatment focuses on prevention of complications by local tumor progression and can be obtained by palliative treatment, e.g., external beam radiotherapy and possibly surgery.

The liver is frequently involved by metastatic disease. Although nearly all primary tumors can metastasize to the liver, especially digestive tract tumors often give rise to liver metastases. Drainage of these organs via the portal system directly toward the liver is the major contributing anatomical background. In the early days of liver surgery for metastatic disease, predominantly metastases of colorectal origin were treated by partial liver resection. The paradigm of limited liver involvement, no extrahepatic disease, and “easily” resectable metastases has undergone a dramatic change. Nowadays, unresectable liver tumors are converted into resectable liver tumors by increasing the future liver remnant with portal vein embolization and decreasing size and number by neoadjuvant chemotherapy or by combining partial liver resection with local ablative therapy like radiofrequency ablation (RFA) [52, 53]. Ablation of liver tumors is considered to be a new treatment strategy in comparison to partial liver resection, and as a consequence many require evidence before it can be introduced as a useful strategy. In this respect it is important to realize that thermoablation of liver was published as early as 1961, only 4 years after the first published liver resection [54]. Currently, several interventional approaches are used to treat liver tumors [54]. Since the introduction of less invasive, local or regional, treatment modalities of liver tumors, other more “exotic” metastatic tumor types than colorectal liver metastases are increasingly considered for treatment [55]. Locoregional liver-directed therapies can be classified as depicted in Table 13.2. For many patients with non-colorectal liver metastasis, decision making on the best type of therapy – if any – cannot be based on evidence. For these patients, personalized medicine using clinical reasoning – based on knowledge of tumor biology – and expertise in liver-directed treatments should guide the best therapeutic approach. In this context, the following variables will be of influence: (1) tumor progression (“stable” versus progressive disease); (2) localization (liver-only versus liver-predominant); (3) presence of symptoms (asymptomatic versus symptomatic); (4) the impact and risks of the intended therapy; (5) the aim of therapy (intentionally curative, not curative but aiming at prolongation of life, palliative with the aim to ameliorate symptoms); and (6) the availability, expected response, and toxicity of systemic treatments. In patients with distant metastases of thyroid cancer, reduction of large metastatic tumor masses (debulking) could be useful to make the ¹³¹I therapy for other smaller remaining metastases more effective [56]. On the one end of the spectrum is the (theoretical) patient with severe debilitating symptoms from a liver-only, relatively stable, potentially curable liver metastasis which can be treated by a minimal invasive – e.g., percutaneous – thermoablation, for whom the alternative would be a highly toxic systemic treatment. In such an example, the decision for treatment will be relatively straight forward. On the other side of the spectrum is the patient with equally debilitating symptoms, with rapidly progressive, widespread metastatic disease for whom a major liver resection needs to be performed, knowing that this will offer only palliation. These cases will probably be withheld from invasive treatment and will be offered best supportive care. In between these extremes are patients in whom decision making is far more complex and prone to extensive discussions. In conclusion, the benefit-risk ratio has dramatically shifted toward a lower risk for patients with liver metastases when treated by minimal invasive locoregional liver-directed treatments. This paves the way for a more liberal use of these treatments probably also in patients with liver metastases from thyroid cancer.

Over the last 30 years, 57 patients were reported in 25 case reports for which various treatments were used (Table 13.3) [1, 46, 47, 56–77]. Seven patients were found with liver metastases from FTC, 40 from MTC, six from PTC, and three from HTC.