Background

Thyroid cancer: overview

Thyroid cancer is a rare cancer, representing only 1% of all malignancies in England and Wales.1 It is caused by the growth of abnormal cells in the thyroid gland. This is a small gland at the base of the neck that secretes three hormones: triiodothyronine (T₃), thyroxine (T₄) and calcitonin. T₃ and T₄ control the rate of metabolism in the body, and calcitonin works with the parathyroid hormone to control the amount of calcium in the blood.2 Thyroid cancer is usually asymptomatic and is often discovered incidentally via imaging studies [e.g. sonograms, computerised tomography (CT) scans and magnetic resonance imaging (MRI)] that are carried out for another reason, or when patients present with a large palpable nodule in the neck.3 The actual diagnosis of thyroid cancer is usually made using ultrasonography and biopsy (typically, a fine-needle aspiration).4

The incidence of thyroid cancer is increasing worldwide.4–10 In the UK, between the period 2003–5 and the period 2012–14, thyroid cancer incidence rates increased by 74% (Figure 1).1 In 2014, there were 3404 patients diagnosed with thyroid cancer in the UK, 2941 in England and 123 in Wales.1 The reasons for the increase in incidence are unknown, but are thought to be, at least in part, attributable to improved diagnostic and detection techniques.11

FIGURE 1 Average number of new cases per year per 100,000 people in the UK. Based on a graphic created by Cancer Research UK.1 (Figure not reproduced here; see the original article.)

The incidence of thyroid cancer is 2.5 times greater in women than in men.1 The reasons for this disparity are unclear.12 Thyroid cancer incidence is strongly related to age, with the highest incidence rates being in older males, and the highest incidence rates among females being in younger and middle-aged women (Figure 2).

FIGURE 2 Age-specific incidence rates per 100,000 people in the UK. Based on a graphic created by Cancer Research UK.1 (Figure not reproduced here; see the original article.)

In the UK, thyroid cancer accounts for < 1% of male cancer deaths and < 1% of female cancer deaths.13 The mortality rate in the UK is reported to be < 1 death per 100,000 people.13 In 2014, there were 376 thyroid cancer deaths in the UK, 154 (41%) in males and 222 (59%) in females, giving a male-to-female ratio of around 7 : 10. In England and Wales, there were 331 thyroid cancer deaths: 137 in males and 194 in females.13

Although the incidence of thyroid cancer in the UK increased between the period 2003–5 and the period 2012–14, overall mortality rates remained stable during this time (Figure 3);13 however, between 1970 and 2014, thyroid cancer mortality rates decreased by 46% in the UK, the decrease being more marked in females (54%) than in males (24%).13 Mortality rates for thyroid cancer are projected to rise in the future: in the UK, it is expected that between 2014 and 2035 mortality will increase by 7%; however, the overall rate is expected to remain relatively low at 1 death per 100,000 people.13

FIGURE 3 European age-standardised thyroid cancer mortality rates in the UK: 1971–2014. Based on a graphic created by Cancer Research UK.13 (Figure not reproduced here; see the original article.)

Differentiated thyroid cancer

The most common form of thyroid cancer is differentiated thyroid cancer (DTC); DTC is reported to account for approximately 94% of thyroid carcinomas.14,15 Less common types of thyroid cancer include medullary carcinoma and anaplastic carcinoma; these have been reported to account for approximately 4% and approximately 2% of all thyroid carcinomas, respectively.15

Differentiated thyroid cancer is a specific type of thyroid cancer made up of different subtypes including papillary carcinoma (PTC), follicular carcinoma (FTC) and Hürthle cell carcinoma. PTC is the most common type of DTC, accounting for approximately 83%15 to 86%16 of all cases; FTC accounts for approximately 10%16 to 13%,15 and Hürthle cell carcinoma accounts for approximately 3%15 to 4%.16 Hürthle cell carcinomas are usually grouped with FTCs because they present and behave similarly.17

The median age for all patients with DTC is reported to be 45 years;18 however, estimates of the median age at onset for the subtypes of DTC have been reported to vary:

• Papillary carcinoma often affects people aged < 40 years,17 but it is also reported that the median age of patients with PTC is 45 years.19
• The peak age for the onset of FTC has been stated to be between 40 and 60 years,20 but, again, the median age has been reported to be approximately 45 years.21
• The median age of patients with Hürthle cell carcinoma has been reported to be 55 years.21

In general, the prognosis for patients with DTC is relatively good. The overall 10-year survival rate for middle-aged adults is reported to be 80–90%.4 It has also been reported that > 85% of patients with DTC have a ‘normal’ life expectancy;22 however, the prognosis generally gets worse with increasing age at the time of diagnosis, particularly for patients aged ≥ 45 years.4 In addition, young children (aged < 10 years) are at higher risk of recurrence than older children.4 Prognosis may also be affected by DTC subtype (histology). An analysis of US National Cancer Database data on 41,375 patients with DTC who were treated between 1985 and 199515 has shown that the 10-year relative survival for patients with PTC is 93%, whereas for patients with FTC it is 85%, and for patients with Hürthle cell carcinoma it is 76%.

The size and spread of the tumour affect prognosis. Studies cited by the British Thyroid Association (BTA)4 are reported to show that the risk of recurrence and mortality correlates with the size of the primary tumour. Extrathyroidal invasion, lymph node metastases and distant metastases are also reported to be important prognostic factors.4

First-line treatment options for patients with differentiated thyroid cancer

There are currently no National Institute for Health and Care Excellence (NICE) guidelines and no NICE guidance for treating patients with DTC or any other type of thyroid cancer. Other clinical guidelines do present some recommendations. In chronological order from date of publication, relevant clinical guidelines include the European Society for Medical Oncology (ESMO) guidelines (2012),23 BTA guidelines (2014),4 American Thyroid Association guidelines (2015)24 and National Comprehensive Cancer Network (NCCN) guidelines (2017).25

Owing to the indolent course of the disease, many patients with DTC, even if they have metastatic disease, do not require therapy for several years after diagnosis.26 Treatments for DTC depend on factors including age, extent of disease, and histology, but usually involve surgery to remove all or part of the thyroid gland (thyroidectomy) followed by lifelong thyroxine for thyroid-stimulating hormone (TSH) suppression from the low–normal to fully suppressed range dependent on risk factors.4,23–25

Treatment options for patients with differentiated thyroid cancer that has progressed following surgery

Following initial surgery, it is estimated that between 5% and 20% of patients with DTC develop local or regional recurrences (approximately two-thirds of these involve cervical lymph nodes27) and between 10% and 15% of patients with DTC develop distant metastases.4,24 The most common sites for metastases are reported to be the lungs (50%), bones (25%), lungs and bones (20%) or other (5%).24 It has been noted that the presence of bone metastases has been associated with a worse prognosis than metastases in other sites.23

The sites that DTC is most likely to spread to vary by histology. For patients aged > 40 years, it has been reported that 10% of patients with PTC, 25% of patients with FTC and 35% of patients with Hürthle cell carcinoma develop distant metastases.28,29 PTC tends to spread to lymph nodes in the neck, whereas FTC usually spreads to the bones or lungs.17 Hürthle cell carcinoma is more likely than FTC to spread to lymph nodes in the neck.30

A radioactive iodine uptake test is commonly used to determine whether or not DTC has spread. The test involves a patient being given a liquid or capsule containing radioactive iodine (iodine-123) to swallow. Two separate uptake measurements are then commonly obtained at different time points within a 24-hour period. The patient is then scanned to see how much of this radioactive iodine has been absorbed by the thyroid (radioactive uptake). Positive results (evidence of iodine-123 uptake) denote the presence of disease, whereas negative results (no radioactive uptake) denote the absence of disease.

It is recommended in clinical guidelines4,23–25 that patients with DTC and evidence of radioactive iodine uptake should undergo treatment with radioactive iodine (also known as radioactive iodine ablation) to treat residual, recurrent or metastatic disease. Patients are typically tested 1–2 months after surgery. Radioactive iodine treatment has been used for > 60 years. It is administered in hospital (during an inpatient stay) and can be given to patients on more than one occasion, as necessary.4

Like the radioactive iodine uptake test used to diagnose DTC, radioactive iodine treatment involves swallowing radioactive iodine in either liquid or capsule form; however, the radioactive iodine is a different form (iodine-131) to that used for scans (iodine-123): the purpose of radioactive iodine treatment is to destroy cancerous cells. Thus, patients with iodine-131 uptake are responsive to treatment, which can be confirmed by imaging studies.

Approximately 33% of patients with advanced disease can be cured and many others achieve long-term disease stabilisation.31 From published French registry data,32 the 10-year survival rate for patients with distant metastases who successfully responded to treatment with radioactive iodine is 92%.32

Radioactive iodine-refractory differentiated thyroid cancer

Although for many patients radioactive iodine is an effective treatment, some patients become resistant to the treatment (decreased or no radioactive iodine uptake) or are unable to safely tolerate additional doses. These patients are considered to have radioactive iodine-refractory DTC (RR-DTC) and are the focus of this multiple technology appraisal (MTA).

Although clinical criteria and algorithms have been developed and reported in clinical guidelines,4,23–25 there is no agreed precise definition of RR-DTC;33 however, a review of the literature published in February 201731 highlights key features that can be considered in defining RR-DTC:

• Metastatic disease that does not take up radioactive iodine at the time of the first radioactive iodine treatment.
• Ability to take up radioactive iodine has been lost after previous evidence of uptake of radioactive iodine.
• Radioactive iodine uptake is retained in some lesions but not in others.
• Metastatic disease that progresses despite substantial uptake of radioactive iodine.
• Absence of complete response to treatment after > 600 mCi of cumulative activity of radioactive iodine.
• Evidence of high uptake of fludeoxyglucose (FDG) ¹⁸F on positron emission tomography or CT scan; however, importantly, the authors of this review31 state that this reason alone should not be used to abandon radioactive iodine treatment.

Before deciding whether or not a patient’s disease can be described as being RR-DTC, it is important to determine that decreased radioactive iodine uptake is not due to iodine contamination or insufficient TSH.34

Radioactive iodine-refractory DTC is a life-threatening form of thyroid cancer with a tendency to progress and metastasise.14 From published French registry data,32 the 10-year survival rate and median overall survival (OS) for patients with distant metastases who failed to respond to treatment (no iodine-131 uptake) was 10% and 3 years, respectively. For those who appear to respond to radioactive iodine treatment (iodine-131 uptake) but who did not then attain negative imaging studies, the 10-year survival and median OS was 29% and 6 years, respectively. A separate analysis of patients with lung and/or bone metastases35 found that 10-year survival and median OS for those who did not have a complete response to treatment with radioactive iodine was 14% and 5 years, respectively. Data from Canada5 have suggested that the median OS for patients with RR-DTC may be between 2.5 and 3.5 years.

The proportion of patients whose disease becomes refractory to treatment with radioactive iodine is relatively small, and so RR-DTC is described as an ultra-orphan condition.7,8 Estimates of the proportion of patients who become refractory vary but commonly lie within the range of 5–15%.7,8,14,16,32,35–37

As with early-stage DTC, many patients with RR-DTC are initially asymptomatic. As highlighted in a literature review published by Schmidt et al.,31 even patients with distant metastases may have a disease that does not progress for many years; however, as noted by Thyroid Cancer Canada,5 the cancer continues to progress with no obvious symptoms.

For patients with rapidly progressing disease, which is characterised by symptomatic disease, the symptoms of RR-DTC can be severe, profoundly debilitating and result in patients becoming increasingly dependent on carers.8 Clinical advice to the Assessment Group (AG) is that the percentage of patients with RR-DTC with rapidly progressing disease is likely to be approximately 25% to 30%. As a result of their symptoms, patients with clinically significant progressive RR-DTC may suffer a poor quality of life and the psychological impact of the disease can also be substantial, resulting in low mood and fatigue.38 It has also been stated that patients with RR-DTC often experience multiple complications.39

Treatment options for patients with radioactive iodine-refractory differentiated thyroid cancer

Radioactive iodine-refractory DTC is typically asymptomatic, but symptoms start to occur as the disease progresses. Symptoms associated with lymph nodes of the neck include difficulty swallowing and/or breathing, pain or sensitivity in the front of the neck or throat, hoarseness or other voice changes, and swelling of the lymph nodes in the neck.4 Symptoms associated with lung metastases also include swallowing and breathing difficulties.26 Pain often presents as the principal symptom of metastatic bone involvement.29,40 Fractures and spinal cord compression are also associated with bone metastases.

Because many treatments, particularly systemic treatments, can have severe side effects and impact significantly on health-related quality of life (HRQoL), clinical advice to the AG is that best supportive care (BSC) tends to be the preferred treatment option, at least until symptoms occur. BSC typically entails TSH suppression therapy and imaging every 3 to 12 months. Palliative radiotherapy and symptom relief are also offered when necessary.

Patients experiencing RR-DTC symptoms and/or patients with rapidly progressing disease are those in need of systemic treatment,31 as reflected in clinical guidelines.4,23–25 The aim of systemic treatment for patients with rapidly progressing and/or symptomatic RR-DTC is to gain local disease control in the neck and manage systemic disease.41 Another important objective of treatment is to prolong survival;27 however, treatment options for patients with RR-DTC are limited. Within the ESMO guidelines published in 2012,23 it is stated that chemotherapy should not be given to patients with RR-DTC as it is associated with significant toxicity with no proven evidence of effectiveness. The authors of these guidelines stated that surgical resection and external beam radiotherapy represented the only therapeutic options and they strongly encouraged enrolment of patients in experimental trials with targeted therapy. Similarly, the authors of the guidelines published by the BTA in 20144 only recommended chemotherapy for patients with rapidly progressive, symptomatic RR-DTC who have good performance status (PS), and only when access to targeted therapies in clinical trials is unavailable or when targeted therapies have proved unsuccessful. The authors of the more recent US guidelines published by the American Thyroid Association24 and NCCN25 recommend that patients with RR-DTC should usually avoid treatment with chemotherapy. Clinical advice received by the AG is that chemotherapy is rarely used to treat RR-DTC in UK NHS practice.

Targeted therapies were not widely available and were only the subject of clinical trials between 2012 and 2014, when the ESMO guidelines23 and the BTA guidelines4 were published. The authors of the BTA guidelines4 considered the most promising targeted therapies at that time to be lenvatinib and sorafenib.4 By 2017, the authors of the NCCN guidelines25 recommended lenvatinib or sorafenib as the treatment of choice for patients with progressive and/or symptomatic disease; lenvatinib is stated to be the ‘preferred’ option based on a response rate of 65% for lenvatinib, compared with 12% for sorafenib, although these agents have not been directly compared. However, the authors state that the decision should be based on the individual patient, taking into account the likelihood of response and comorbidities.25 In cases in which lenvatinib or sorafenib are not available or not appropriate, drugs not regulated by the US Food and Drug Administration (FDA) but used in the context of clinical trials are also recommended by the authors of the NCCN guidelines.25

Description of technology under assessment

The two interventions under consideration in this MTA are lenvatinib (Lenvima), manufactured by Eisai Ltd, and sorafenib (Nexavar), manufactured by Bayer HealthCare. Both are a type of tyrosine kinase inhibitor (TKI) known as a multikinase inhibitor (MKI).

A brief comparison of the key features of the two interventions is given in Table 1. The AG notes that lenvatinib and sorafenib appear to have slightly different mechanisms of action.42 Both drugs have been approved for treating RR-DTC in the USA43,44 and Europe,49,50 with sorafenib being the first of the two agents to be approved in both jurisdictions. In the USA and Europe, the marketing indications for both lenvatinib and sorafenib are for identical patient populations. Approval in the USA and Europe was based largely on evidence from two Phase III randomised controlled trials (RCTs): SELECT,51 in which lenvatinib was compared with placebo, and DECISION (StuDy of sorafEnib in loCally advanced or metastatIc patientS with radioactive Iodine-refractory thyrOid caNcer),52 in which sorafenib was compared with placebo.

Approval for use in NHS Scotland was granted to sorafenib in June 201548 and to lenvatinib in September 2016.38 Both approvals are for the treatment of patients with progressive, locally advanced or metastatic RR-DTC. In NHS Scotland, the use of both lenvatinib and sorafenib is contingent on the continuing availability of Patient Access Scheme (PAS) prices that have been assessed by the PAS Assessment Group.

Sorafenib has been available in England, since July 2016, via the Cancer Drugs Fund (CDF). It is currently funded for all patients with RR-DTC for whom the treating specialist has established that treatment with sorafenib may be beneficial. According to Bayer HealthCare, based on its analysis of notification data from July 2013 to June 2016, sorafenib has become the standard of care for patients for whom systemic treatment is appropriate.7 Lenvatinib is not currently available to patients treated in the English or Welsh NHS.

Eisai Ltd8 has estimated the incidence of patients in England and Wales with RR-DTC who are potentially eligible for treatment with lenvatinib or sorafenib to be approximately 280 each year. Bayer HealthCare7 has estimated the incidence to be approximately 225 patients per year. The AG notes that the estimates given by the companies differ in how they are calculated. The estimates provided by the companies are reflective of the population defined by the agreed final scope of this appraisal; however, neither estimate appears to account for the fact that lenvatinib and sorafenib are likely to only be preferred for patients with symptomatic and/or rapidly progressing disease. Clinical advice to the AG is that there are no generally agreed definitions of ‘symptomatic’ or ‘rapidly progressive disease’ and that, in clinical practice, definition of a patient’s disease status depends on individual patient characteristics. Therefore, it is difficult to further segment the population.