Thyroid nodules and cancer in children

INTRODUCTION — Approximately 2 percent of children have palpable thyroid nodules. The majority are benign, including inflammatory lesions or follicular adenomas, but up to 25 percent are malignant.

The thyroid gland in children is particularly susceptible to irradiation and carcinogenesis, which may explain why children with thyroid cancer tend to present with advanced disease. Compared with adults, children with thyroid cancer display a greater frequency of lymph node metastases and distant metastases at the time of diagnosis and higher rates of recurrence. Despite these characteristics, children with thyroid cancer generally have a good prognosis.

The evaluation of a child presenting with a thyroid nodule and an overview of the treatment of thyroid cancer are discussed in this topic review. Congenital and acquired goiter and thyroid cysts are discussed separately. (See "Approach to congenital goiter in newborns and infants" and "Approach to acquired goiter in children and adolescents".)

EPIDEMIOLOGY

●Thyroid nodules – In a study of 5179 schoolchildren between the ages of 11 and 18 years conducted in the southwestern United States, thyroid nodules detectable by palpation were present in 1.8 percent [1]. In a study of 440 schoolchildren aged 5 to 18 years in Athens, Greece, thyroid nodules were present in 5.1 percent by ultrasonography [2]. By contrast, a survey of nearly 300,000 infants and children around Fukushima intended to document the baseline rate of nodules around the time of the 2011 nuclear accident reported that only 0.76 percent had nodules larger than 5 mm detected by ultrasound examination [3]. The differences in the prevalence of thyroid nodules among these three populations may be explained by different genetic makeup or environmental risk factors (such as iodine sufficiency) or variation in ultrasound technique or interpretation. In a group of children without suspected thyroid disease who underwent contrast-enhanced computed tomography (CT) of the chest, thyroid nodules were detected in 1.4 percent, which is much lower than the rate of unsuspected thyroid nodules detected on CT in adults [4]. Thyroid nodules increase in prevalence during adolescence: In the Fukushima study, nodules larger than 5 mm were present in 0.3 percent of children under age 15 years but in 1.7 percent of those age 15 to 19 years [3]. The prevalence of thyroid nodules is equal in females and males prior to 10 years of age, but nodules are approximately twice as common in females as in males after this age, perhaps because of pubertal effects that are not well understood.

●Malignant thyroid tumors – Most thyroid nodules in children are benign, but the percentage of nodules harboring cancer in children is higher than the rate of 5 to 15 percent in adults. Rates of malignancy among pediatric nodules evaluated clinically by ultrasound range from 19 to 34 percent [5-8]. In a study that used a standardized diagnostic process for thyroid nodules, the rate of cancer was 19 percent in children, compared with 12 percent in adults [9]. However, the malignancy rate appears to be lower (4 to 6 percent) in nodules discovered incidentally on ultrasound, CT, or magnetic resonance imaging performed for other indications [4,10].

The incidence of differentiated thyroid cancer (DTC) in children is increasing worldwide [11]. A large registry study reported a total of 1806 pediatric patients (<20 years of age) diagnosed with thyroid cancer between 1973 and 2013, with an annual incidence of 0.60 cases per 100,000 population [12]. The annual incidence was 0.48 cases per 100,000 person-years in 1973 and increased to 1.14 cases per 100,000 in 2013. The annual percent change rose gradually between 1973 and 2006 and then increased sharply to 9.56 percent between 2006 and 2013. Another study reported a similar 3 to 4 percent increase between 1984 and 2010 for children, adolescents, and young adults [13]. This increase appears to be a result of both enhanced detection and a true increase in the incidence of thyroid cancer in children.

Prepubertal children with thyroid cancer tend to have more advanced disease at presentation compared with adolescents and adults [14-16]. This was illustrated in a review of 540 children from nine large centers, in which 71 to 90 percent of the children had cancer that spread to regional lymph nodes, 20 to 60 percent had extracapsular extension (some with invasion of the trachea), and 30 percent had involvement of the recurrent laryngeal nerve [15]. Distant metastases are usually in the lungs or, rarely, in the bones or brain [17].

HISTOPATHOLOGY

Benign lesions — The majority of thyroid nodules in children are benign (see 'Epidemiology' above). Causes include benign thyroid adenomas (usually colloid or follicular adenomas), thyroid cysts, and inflammatory lesions (table 1). Follicular adenomas are the most common tumor.

Malignant thyroid tumors — Types of malignant thyroid tumors are (table 1) [11]:

●Differentiated thyroid cancers (DTCs)

•Papillary thyroid cancer (PTC) – Papillary carcinoma accounts for approximately 86 percent of pediatric thyroid cancers, with sharp increases in this type during the past two decades (see 'Epidemiology' above). One-quarter of these are "follicular variant," a subtype of PTC. A "noninvasive follicular thyroid neoplasm with papillary-like nuclear features" (NIFTP), previously referred to as a noninvasive follicular-variant PTC (FVPTC), was renamed NIFTP to highlight its indolent behavior and <1 percent recurrence rate. These neoplasms are considered benign, but they have malignant potential, and surgical excision is required to confirm that they are noninvasive.

•Follicular thyroid cancer (FTC) – Follicular carcinoma accounts for 8 to 9 percent of pediatric thyroid cancers.

●Medullary thyroid cancer (MTC) – Medullary carcinoma accounts for 4 percent of pediatric thyroid cancers but a higher proportion in young children [11]. Most cases are associated with multiple endocrine neoplasia type 2 (MEN2) (table 2) [18].

●Other – Other types of thyroid cancers, which are rare in children, are anaplastic thyroid carcinoma, primary thyroid lymphoma, and cancers that have metastasized to the thyroid. (See "Anaplastic thyroid cancer".)

PATHOGENESIS OF THYROID CANCER — Most pediatric thyroid cancer is sporadic and arises from a de novo somatic genetic variant [18]. However, a minority of patients have specific risk factors for thyroid cancer, including ionizing radiation or a genetic predisposition.

Radiation — Ionizing radiation is the most common nongenetic risk factor for thyroid cancer in children. Young children (eg, younger than 10 years of age) are particularly susceptible to the carcinogenic effects of irradiation.

Most cases are related to exposure to external irradiation for treatment of nonthyroidal malignancies, including Hodgkin lymphoma or other tumors of the head and neck, or conditioning for hematopoietic stem cell transplantation. For children with this exposure, the increased risk for developing thyroid nodules and/or cancer continues into adulthood [19-22]. Other cases are related to exposure due to nuclear accidents. As examples, follow-up of residents <18 years of age living in Belarus at the time of the Chernobyl nuclear accident (April 1986) showed a nearly fourfold increase in risk of having neoplastic thyroid nodules and the risk increased with the dose of radiation exposure and with younger age at exposure [23]. The pathogenesis of radiation-induced thyroid cancer is discussed in detail separately. (See "Radiation-induced thyroid disease".)

Genetic predisposition — Thyroid cancer is included in several familial cancer syndromes and a nonsyndromic form. Children with a thyroid nodule and a family history of one of these disorders are at increased risk for malignancy. For children with a family history of these disorders but no thyroid nodule, screening recommendations are discussed in the linked topic reviews.

●Syndromic forms – Several familial cancer predisposition syndromes are associated with thyroid cancer. All children with thyroid cancer should be evaluated for clinical features that would suggest these syndromes (table 3), and children with such a family history should be evaluated for thyroid cancer.

•APC (adenomatous polyposis coli)-associated polyposis syndromes, including Gardner syndrome, are characterized by familial adenomatous polyps in the gastrointestinal tract and a predisposition to papillary thyroid cancer (PTC) and several other tumors [24]. The risk of associated thyroid cancer approximates 10 percent [18], but thyroid cancer usually presents in adulthood. In a report of 129 patients with familial adenomatous polyposis, 8.5 percent had PTC, all of whom were female and between 18 to 47 years of age [25]. In a case series of 37 children with familial adenomatous polyposis who underwent ultrasound screening, thyroid cancer was diagnosed in two female patients (9 percent) between 17 to 19 years of age but not in any male patient [26]. APC-associated PTC frequently has a distinct histopathology (cribriform morular variant), and this variant of PTC should prompt evaluation for a germline APC variant. (See "Gardner syndrome".)

•PTEN (phosphatase and tensin homolog) hamartoma tumor syndromes (including Cowden and Bannayan-Riley-Ruvalcaba syndromes) are autosomal dominant inherited tumor syndromes that are characterized by hamartomas in the skin and other tissues and an increased predisposition to differentiated thyroid cancer (DTC) [27]. The prevalence of thyroid cancer in children with PTEN hamartoma tumor syndromes is 2 to 7 percent [28-30]. Follicular thyroid cancer (FTC) is overrepresented compared with the general population [31]. Both Cowden syndrome and Bannayan-Riley-Ruvalcaba syndrome are associated with germline mutations in the PTEN gene. (See "PTEN hamartoma tumor syndromes, including Cowden syndrome", section on 'Cowden syndrome' and "PTEN hamartoma tumor syndromes, including Cowden syndrome", section on 'Bannayan-Riley-Ruvalcaba syndrome'.)

Clinical features associated with PTEN gene mutations in pediatric patients are macrocephaly, autism or developmental delay, vascular malformations, and penile freckling or other benign skin lesions (table 3) [28]. Adult relatives with PTEN hamartoma tumor syndrome may have a history of breast or uterine cancer.

•DICER1 syndrome is an autosomal dominant disorder resulting from germline mutations in the DICER1 gene, which predispose to pleiotropic tumors including pleuropulmonary blastoma, cystic nephroma, ovarian Sertoli-Leydig cell tumors, pineal tumors, and nasal chondromesenchymal hamartomas. One study from the National Cancer Institute registry reported multinodular goiter in 13 percent of affected individuals [32], while another study from the same registry reported that thyroid cancer developed in 4 percent of nonproband carriers, a 20-fold increase in the standardized incidence ratio [33]. Most thyroid cancers in individuals with DICER1 syndrome are FTC or follicular variants of PTC [31]. (See "Congenital pulmonary airway malformation", section on 'Pleuropulmonary blastoma'.)

•Carney complex type 1 is characterized by a primary pigmented nodular adrenocortical disease, other endocrine tumors including PTC or FTC, and nonendocrine tumors such as myxomas and breast adenomas [34]. It is associated with a mutation in the PRKAR1A gene (protein kinase A regulatory 1-alpha subunit). (See "Cushing syndrome due to primary pigmented nodular adrenocortical disease", section on 'Carney complex (CNC)'.)

•Werner syndrome is characterized by connective tissue disease, causing symptoms of premature aging (progeria) and increased risk for osteosarcoma, soft tissue sarcomas, melanoma, and PTC or FTC [34,35]. (See "Calcinosis cutis: Etiology and patient evaluation", section on 'Genodermatoses'.)

•Multiple endocrine neoplasia type 2 (MEN2) is strongly associated with medullary thyroid cancer (MTC) and is transmitted in an autosomal dominant fashion. Three distinct subtypes of MEN2 are associated with MTC: MEN2A, MEN2B, and familial MTC (table 2). MEN2A may be associated with Hirschsprung disease, and both MEN2A and MEN2B may be associated with pheochromocytoma. These disorders are caused by mutations in the RET protooncogene. (See "Medullary thyroid cancer: Clinical manifestations, diagnosis, and staging" and "Classification and genetics of multiple endocrine neoplasia type 2".)

●Nonsyndromic forms – Familial patterns of inheritance for nonmedullary thyroid cancer, in the absence of the above syndromes, account for 3 to 9 percent of thyroid cancer in all age groups [34]. More than 85 percent of these are PTC, and the remainder are follicular or other types [36]. Linkage analysis has identified several genetic loci that may contribute to this predisposition, but no specific genes have been confirmed.

Other — Thyroid cancer has been found in children with thyroglossal duct cysts [37], and it also has been reported in children with congenital goiter due to dyshormonogenesis [38]. In these situations, it appears that unknown somatic cell mutations along with chronic thyroid-stimulating hormone (TSH) stimulation may play a pathogenic role. It remains unclear whether the risk of thyroid cancer may be increased in autoimmune thyroid disease, ie, Hashimoto thyroiditis [39,40] and Graves disease [41]. (See "Approach to congenital goiter in newborns and infants", section on 'Thyroglossal duct cysts'.)

CLINICAL PRESENTATION — Both differentiated thyroid cancer (DTC) and medullary thyroid cancer (MTC) in children usually present as asymptomatic solitary nodules. The vast majority of children with nodules have normal thyroid function. A minority of children with a thyroid nodule present with clinical and laboratory features of hyperthyroidism due to an autonomous nodule ("toxic adenoma"), which is relatively rare in children.

●DTC – DTC includes papillary thyroid cancer (PTC) and follicular thyroid cancer (FTC). Clinical findings that increase the likelihood of cancer in a child presenting with a thyroid nodule are male sex, history of external radiation to the head and neck or exposure to nuclear fallout, history of rapid growth of the nodule, a firm or fixed mass, hoarseness or dysphagia, and cervical adenopathy.

●MTC – Children with MTC typically present with a solitary nodule or they are discovered incidentally when a family member is diagnosed with MTC, typically as part of multiple endocrine neoplasia type 2A (MEN2A) or MEN2B (table 2). Children with MEN2B may have a Marfanoid body habitus, mucosal neuromas of the tongue and conjunctivae, and medullated (thickened) corneal nerves. (See "Clinical manifestations and diagnosis of multiple endocrine neoplasia type 2".)

EVALUATION OF THYROID NODULES — A diagnostic process using thyroid function tests, thyroid scintigraphy, ultrasound, and fine-needle aspiration (FNA) is used to identify patients with cancer. The diagnostic process and indications for FNA, although similar to those used in adults, have some elements specific to children (algorithm 1). Although the majority of nodules are benign, children with thyroid nodules are somewhat more likely to have cancer than adults. In a study that used this diagnostic process, the rate of cancer in the group undergoing FNA was 19 percent, as compared with 12 percent in adults evaluated by a similar process [9]. (See "Diagnostic approach to and treatment of thyroid nodules".)

History and physical examination — Most thyroid nodules in children are initially discovered either by the patient, parent, or clinician.

The history should include:

●Past medical history of head and neck radiation (eg, given as treatment for head and neck tumors)

●Questions about features of thyroid cancer syndromes, such as developmental delay or autism (table 3)

●Family history for autoimmune thyroid disease, thyroid cancer, and other features associated with thyroid cancer syndromes, such as endocrine tumors, gastrointestinal polyps, or other early-onset cancers (eg, breast, cervix, uterus)

The physical examination should include:

●Inspection and palpation of the thyroid gland, noting whether it is symmetric, whether it is cystic or firm, whether it is tender, and recording location and dimension(s) of nodule(s)

●Careful examination of the lateral neck should be undertaken for lymph nodes:

•Mobile, rubbery lymph nodes under the mandible (level II) are common and usually benign

•Firm, immobile lymph nodes in the lower neck (levels III and IV) are suspicious for malignancy

●In addition, a careful general examination should be carried out for findings associated with the genetic thyroid cancer syndromes (table 3)

Thyroid function tests — In children presenting with thyroid nodule(s), thyroid-stimulating hormone (TSH) should be assessed initially. Most patients will have normal TSH and should be evaluated further with a neck ultrasound. Patients with low serum TSH may have an autonomous nodule and should be evaluated further with thyroid scintigraphy as well as ultrasound.

Thyroid scintigraphy — In general, thyroid scintigraphy (radioisotope scan) is indicated only if the serum TSH is subnormal (suggesting an autonomous nodule). Scintigraphy should be performed using iodine-123 (123-I) as the tracer. If the nodule is autonomous (hyperfunctioning or "hot"), it is seen as an area of increased uptake, with absent or suppressed uptake in the remainder of the gland. A serum TSH <0.3 mIU/L at the time of scintigraphy generally ensures that uptake is suppressed in the surrounding thyroid tissue and permits optimal evaluation of nodule autonomy [42]. In our experience, children with borderline low TSH in whom an autonomous nodule is suspected can be treated with levothyroxine (50 to 75 mcg daily) for one to two weeks to suppress TSH further prior to scintigraphy.

The majority of autonomous nodules are benign, and FNA generally is not required. In a study using rigorous criteria to define autonomous thyroid nodules in children (both autonomous 123-I uptake into the nodule and suppression of uptake in the normal thyroid parenchyma observed when TSH is <0.3 mIU/L), the observed cancer rate in biopsied or resected nodules was 0 percent [42]. Rarely, a hyperfunctioning nodule may harbor papillary thyroid cancer (PTC) or follicular thyroid cancer (FTC) [43], so FNA should be considered in the presence of atypical features such as solid parenchyma or abnormal lymph nodes on ultrasound or extrathyroidal uptake on scintigraphy. Depending on symptoms and the severity of hyperthyroidism, treatment may include surgical resection, radioactive iodine ablation, or observation with or without methimazole to treat hyperthyroidism. (See 'Benign thyroid nodules' below.)

Neck ultrasound — Ultrasound examination of the neck will define the presence, size, and sonographic characteristics of thyroid nodules (table 4). Nodule characteristics that are associated with an increased risk of cancer include solid parenchyma, hypoechogenicity, irregular margins, and calcifications, as well as the presence of abnormal-appearing lymph nodes in the neck [6-8]. Simple cysts are almost always benign. In any child with a thyroid nodule, images should also be obtained to assess the morphology of the cervical lymph nodes. Benign nodes are elongated with a fatty hilum, whereas abnormal nodes may be rounded, lack a fatty hilum, or contain calcifications or cystic spaces. The ultrasound examination should be performed by a clinician experienced in interpreting thyroid ultrasounds in children.

Fine-needle aspiration

Indications — FNA is the most useful test to differentiate benign thyroid nodules from cancer and is preferred over core biopsy, which is more invasive and has a higher risk of complications. FNA has high diagnostic accuracy in children [44,45].

Standardized sonographic risk assessment systems have been validated in adults to assist with selection of nodules for FNA, including the 2015 American Thyroid Association system [46] and the more recent American College of Radiology Thyroid Imaging, Reporting and Data System (ACR-TIRADS) [47]. These systems are useful for assessing suspicious sonographic features but may not be directly applicable in the pediatric population, in part because their recommended size criteria for performing FNA in adults (≥1 to 1.5 cm) may not be appropriate in a growing child whose thyroid gland may normally be one-half the size of that of an adult. In particular, one series found that ACR-TIRADS missed or delayed the diagnosis of 22 percent of pediatric thyroid cancers, primarily those that were below the ACR-TIRADS size thresholds for FNA or sonographic follow-up or those that had few suspicious features [48]. Therefore, the American Thyroid Association recommends that "ultrasound characteristics and clinical context should be used rather than size alone to identify nodules that warrant FNA" [49].

Accordingly, we perform FNA biopsy in children on nodules with the following characteristics:

●Nodules ≥1 cm that are solid or predominantly solid or have other suspicious features.

●Nodules ≥1.5 cm that are mixed solid/cystic, with or without other suspicious features.

●Nodules <1 cm with ultrasound characteristics that are highly suspicious for cancer, such as those with calcifications or abnormal cervical lymph nodes (table 4) [50], particularly in individuals at increased risk of malignancy (such as history of irradiation). (See "Diagnostic approach to and treatment of thyroid nodules", section on 'Sonographic criteria for FNA'.)

●Nodules with documented enlargement on repeat ultrasound examinations.

Ultrasound guidance should be used for FNA of thyroid nodules in children because this improves diagnostic accuracy and reduces the risk of complications [49]. An experienced team, usually including a radiologist and a cytopathologist, is required to perform and interpret the FNA.

Interpretation and next steps — The Bethesda System for Reporting Thyroid Cytopathology (BSRTC), initially developed to classify FNA results in adults, should also be used for children [49,51]. FNA results in the BSRTC fall into six categories (algorithm 1). The risk of thyroid cancer in each category is presented in the table (table 5). Malignancy rates in intermediate cytologic categories (III, IV, and V) are higher in children than in adults [9].

●Nondiagnostic or unsatisfactory (BSRTC category I) – These lesions should undergo repeat FNA, particularly if solid or predominantly solid. If the repeat FNA is still nondiagnostic, surgical lobectomy should be considered.

●Benign (BSRTC category II) – These lesions are usually benign but should be followed clinically. (See 'Benign thyroid nodules' below.)

●Atypical cells of undetermined significance or follicular lesion of undetermined significance (BSRTC category III) – This category carries a 30 to 45 percent risk of thyroid cancer (table 5). For nodules with this cytology, we generally recommend repeating the FNA in three to six months because repeat cytology may be benign in 25 to 30 percent of cases [9] and benign cytology on repeat FNA is associated with a relatively low risk of malignancy (one of nine patients in one study) [52]. However, some clinicians and families prefer to proceed directly to surgical lobectomy for nodules with BSRTC category III cytology. There is some evidence that within this cytologic category, nuclear atypia (abnormal changes in cell nuclei) is associated with a higher risk of malignancy than architectural atypia (changes in cell arrangement) [51,53]. (See "Diagnostic approach to and treatment of thyroid nodules".)

●Follicular neoplasm (BSRTC category IV) – Specimens in this category carry a significant risk (30 to 70 percent) of thyroid cancer, often follicular carcinoma or follicular-variant papillary carcinoma (table 5). For these nodules, surgery (lobectomy) is generally recommended. (See 'Papillary thyroid cancer' below and 'Follicular thyroid cancer' below.)

For nodules in BSRTC category III or IV, another option is to undertake further evaluation of the FNA specimen for somatic mutations by genetic testing. Although genetic testing platforms developed for adults have uncertain applicability in children, studies show that tests that detect known oncogenic drivers are useful in detecting malignancy. If an oncogene panel is positive for a pathogenic molecular variant (BRAF V600E, RET/PTC, or NTRK fusion), there is a nearly 100 percent likelihood of PTC [18]. However, other genetic variants such as RAS, DICER1, and PTEN can be associated with either benign or malignant tumors. Moreover, the negative predictive value of molecular testing in children is unknown; therefore, a "benign" or "negative" result from a molecular test should not be relied upon to exclude malignancy in a child and such nodules should be managed based on cytology and clinical factors. For this reason, we do not recommend the use in children of gene expression panels that predict benign lesions (such as Afirma GSC). (See "Evaluation and management of thyroid nodules with indeterminate cytology in adults", section on 'Molecular markers'.)

●Suspicious for malignancy (BSRTC category V) – Children with this result should undergo surgery since most (70 to 85 percent) have thyroid cancer (table 5) [18,54,55].

●Malignant (BSRTC category VI) – This result is diagnostic of thyroid cancer and should be treated with thyroidectomy. (See 'Papillary thyroid cancer' below.)

Genetic testing for germline mutations — Genetic testing for germline mutations is indicated for some children, guided by clinical features, results of the FNA biopsy, and family history:

●Children with FNA biopsy results consistent with possible PTC or FTC should be carefully evaluated for clinical features of one of the associated genetic syndromes (table 3). Particular attention should be given to features of PTEN hamartoma syndrome (Cowden or Bannayan-Riley-Ruvalcaba syndrome) and DICER1 syndrome. PTEN hamartoma syndrome includes macrocephaly, autism or developmental delay, penile freckling, and other benign skin lesions. DICER1 syndrome includes pleuropulmonary blastoma, cystic nephroma, ovarian Sertoli-Leydig cell tumors, and pineal tumors. Children with any of these associated features, particularly macrocephaly, or those with a family history of similar findings should be tested for a PTEN gene mutation [28]. (See 'Genetic predisposition' above and "PTEN hamartoma tumor syndromes, including Cowden syndrome", section on 'Pilarski et al diagnostic criteria'.)

●Children with a history of gastrointestinal polyps (or a family history of gastrointestinal polyps or cancer) who present with a thyroid nodule should be considered for APC gene mutation testing (for familial adenomatous polyposis, including Gardner syndrome). Children with the cribriform-morular variant of PTC should also be tested for an APC gene mutation. (See "Clinical manifestations and diagnosis of familial adenomatous polyposis".)

●Children who are at risk for medullary thyroid cancer (MTC), usually based on discovery of a relative with multiple endocrine neoplasia type 2 (MEN2) or MTC, should undergo RET protooncogene analysis [56]. (See "Medullary thyroid cancer: Clinical manifestations, diagnosis, and staging", section on 'Inherited MTC'.)

MANAGEMENT — Thyroid nodules in children are managed primarily based on the results of fine-needle aspiration (FNA) (algorithm 1) [49]. (See "Diagnostic approach to and treatment of thyroid nodules".)

Benign thyroid nodules

●Euthyroid nodules – Nodules with benign (Bethesda system for reporting thyroid cytopathology [BSRTC] category II) cytology can be monitored by periodic neck palpation and ultrasound examinations. In the vast majority of these patients, thyroid function is normal. Patients with hypothyroidism should be treated with levothyroxine.

The optimal interval and duration of ultrasound monitoring for benign nodules in children are unknown. We perform initial repeat ultrasound in one year and subsequent surveillance every one to two years (or longer if significant changes are not observed). A significant increase in nodule size (increase in volume by ≥50 percent, or increase in diameter by ≥20 percent and by at least 2 mm in at least two dimensions) should prompt a repeat FNA because a small percentage of nodules initially read as benign harbor cancer [54]. A large study of cytologically benign nodules in children followed for a mean of 3.4 years found that 2.5 percent developed thyroid cancer [52]. Cancer was more common in nodules >4 cm at presentation or in nodules that grew during follow-up. Because of their increased risk of false-negative FNA (up to 15 percent), nodules >4 cm should be considered for resection [52]. Benign nodules may also be resected if they cause symptoms.

●Simple cysts and obvious inflammatory nodular lesions – Simple cysts and inflammatory lesions (eg, associated with some form of thyroiditis) generally can be monitored by periodic neck palpation. If the nodule increases in size, an ultrasound examination should be performed to record dimensions and look for new features that might warrant a repeat FNA (table 4).

●Noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) – NIFTP, previously referred to as a noninvasive follicular variant papillary thyroid cancer (FVPTC), was renamed NIFTP to highlight its indolent behavior and <1 percent recurrence rate. These neoplasms are considered benign, but they have malignant potential, and surgical excision (lobectomy) is required to confirm the diagnosis [57]. NIFTP appears to be rare in children, and data remain limited regarding its behavior in the pediatric population.

●Autonomous nodule – An autonomous nodule ("toxic adenoma") is diagnosed by the combination of low thyroid-stimulating hormone (TSH) and a radioisotope scan showing focal uptake of iodine-123 [123-I] in the nodule with suppressed uptake in the rest of the gland. For nodules with these characteristics, the risk of malignancy is very low. However, because these adenomas are unlikely to resolve spontaneously, surgical excision is often pursued, particularly if symptoms of hyperthyroidism (eg, palpitations, heat intolerance, anxiety) or local mass effect (eg, dysphagia, cosmetic effect) are present. Radioactive iodine ablation is an option in older adolescents (algorithm 1). For patients who have overt hyperthyroidism but do not desire surgery or ablation, a low dose of methimazole may be used. (See "Diagnostic approach to and treatment of thyroid nodules", section on 'Management'.)

If the hyperthyroidism is subclinical (asymptomatic or minimal symptoms), with suppressed TSH and free T4 and total T3 in the normal range, surgery may be deferred. However, FNA should be performed in those with ultrasound features suspicious for thyroid cancer, such as solid parenchyma or abnormal lymph nodes on ultrasound. This is because a hyperfunctioning nodule rarely may harbor papillary or follicular thyroid cancer (FTC) [43].

Papillary thyroid cancer — PTC accounts for more than 80 percent of thyroid cancer in children and is the most common form of differentiated thyroid cancer (DTC). (See 'Malignant thyroid tumors' above.)

Thyroidectomy — For most children with PTC that is diagnosed or strongly suspected based on preoperative evaluation, we suggest total or near-total thyroidectomy rather than lobectomy. Neck dissection (central and/or lateral) should be performed using a compartment-based approach in any compartment with clinically evident lymph node metastasis. Total thyroidectomy has long been the recommended initial surgical procedure because PTC in children is bilateral in 35 to 45 percent of cases [58,59] and this feature may not be detectable on preoperative imaging [59-61]. In addition, bilateral thyroid resection has been associated with lower rates of disease recurrence compared with unilateral resection in some studies. In an observational study of long-term outcomes for children with PTC, the incidence of disease recurrence was 25 percent among children who underwent bilateral lobar resection compared with 65 percent for those who underwent unilateral lobectomy [62]. Similarly, in another retrospective study of 235 pediatric patients with PTC who were treated at a single center between 1973 and 2002, patients who underwent total or near-total thyroidectomy (n = 172) had fewer local recurrences compared with those who underwent more limited surgical treatment (n = 63; 0.6 versus 17 percent, respectively) [63]. After adjusting for other variables, the extent of surgery remained a strong independent predictor of recurrence, with a nearly tenfold increased risk among patients who did not undergo total or near-total thyroidectomy (odds ratio 9.5, 95% CI 1.2-78.1).

In adults, lobectomy may be sufficient treatment for localized, noninvasive PTC, but the long-term outcome of this approach in children remains uncertain. Children with small PTCs (<1 cm) and no clinically evident lymph node metastasis have a low risk of bilateral disease and of recurrence [58,64]. In such patients, it may be reasonable to consider lobectomy as the initial surgery since total thyroidectomy is associated with a higher risk of operative complications (hypoparathyroidism and recurrent laryngeal nerve injury) and a need for lifelong levothyroxine treatment. One registry-based observational study compared 163 children with low-risk PTC up to 4 cm in size who underwent total thyroidectomy with 163 propensity-matched controls who underwent lobectomy and found no difference in 10-year survival [65]. However, this study was limited by relatively short median follow-up duration (five to eight years) and a low overall mortality rate (2.4 percent) that yielded low power to discern differences in outcome over this time frame.

In children with PTC who have undergone initial lobectomy, the presence of multifocal tumor in the resected lobe is associated with an increased risk of disease in the contralateral lobe (50 to 65 percent) [58,64]. Therefore, completion thyroidectomy should be strongly considered if multifocal PTC is present. Children under age 10 years may have an increased risk of bilateral disease [59], but whether young age alone necessitates total thyroidectomy remains uncertain.

The utility of prophylactic neck dissection (ie, in the absence of clinically evident lymph node metastasis) in children with PTC is uncertain. The number of metastatic lymph nodes is directly correlated with the risk of persistent/recurrent disease [66], implying that lymph node dissection during initial surgery may provide useful information. However, lymph node dissection increases complication rates and it remains unclear whether the prophylactic resection of non-clinically evident (microscopic) lymph node metastases improves outcomes in children with PTC.

We recommend that thyroid surgery in children be performed by an experienced, "high-volume" thyroid surgeon to minimize the risk of complications, which is higher in children than in adults [67,68]. One analysis of 1199 patients reported that recurrent laryngeal nerve injury occurred in 3.8 percent of children aged 0 to 6 years, 1.1 percent in children aged 7 to 12 years, and 0.6 percent in children aged 13 to 17 years [68]. Another multicenter study of pediatric tertiary care hospitals reported similar rates of laryngeal nerve injury as well as hypoparathyroidism in 7.2 percent, with permanent hypoparathyroidism in 3.3 percent of cases [69].

Stratification by risk for persistent postoperative disease — The American Thyroid Association Guidelines Task Force defined three categories based on the risk for persistent postoperative disease in children with PTC (table 6) [49,70]:

●Low risk – PTC confined to the thyroid, with no or a small number (≤5) of microscopic metastases (<0.2 cm) to central compartment neck lymph nodes. Serum thyroglobulin (Tg) while taking levothyroxine (nonstimulated Tg) is generally <1 ng/mL postoperatively.

●Intermediate risk – PTC with either extensive central compartment (1a) lymph node metastases (>5 in number or >0.2 cm) or minimal lateral neck (1b) lymph node metastases (≤10 lymph node metastases, ≤3 cm in size).

●High risk – Regionally extensive PTC (lateral compartment lymph node [1b] metastases >10 in number or >3 cm in size) or locally invasive PTC (gross tumor invasion into skeletal muscle, larynx, trachea, esophagus, blood vessels, or nerves), with or without distant metastases.

This risk stratification is intended to reduce the use of postoperative radioactive iodine therapy in low-risk patients, for whom radioactive iodine offers no benefit. The strategy reflects the unique features of PTC in children, which include an excellent prognosis but also increased risk for second primary malignancies decades later (see 'Prognosis' below). As an example, a long-term study reported a higher-than-predicted number of deaths from second primary malignancies after 30 and 50 years of follow-up [62]. In this analysis of all-cause mortality, two-thirds of the deaths resulted from nonthyroidal second malignancies and three-quarters of these patients had received some form of postoperative therapeutic irradiation. In two large studies of pediatric and young adult patients diagnosed with DTC up to age 25 to 29 years, treatment with radioactive iodine was associated with a 1.5- to 1.6-fold increased risk of developing a second primary malignancy [71,72]. One of these studies estimated that radioactive iodine treatment in young patients may result in one secondary malignancy for every 150 patients treated over 20 years of follow-up [72]. The second primary malignancies observed were mainly leukemias and cancers of the salivary gland and breast.

Patients with low-risk PTC — Thyroidectomy is the primary treatment for low-risk PTC; radioactive iodine therapy is not indicated. Key management steps for patients with low-risk PTC are (table 6) [49]:

●Levothyroxine – Initiate levothyroxine therapy and adjust dose to achieve a TSH concentration of 0.5 to 1 mIU/L.

●Postoperative staging

•Measure serum Tg on levothyroxine therapy (nonstimulated Tg) approximately 4 to 12 weeks after surgery; a nonstimulated Tg <1 ng/mL is consistent with low-risk disease

•Perform a neck ultrasound at six months postoperatively

●Radioactive iodine therapy is NOT necessary – We suggest not using radioactive iodine therapy in patients with low-risk PTC. The rationale is as follows:

•Excellent survival – Patients with low-risk PTC have excellent survival with or without radioactive iodine therapy (20-year overall survival >95 percent) [16].

•Uncertain impact on recurrence – The recurrence risk in this population is low (approximately 6 to 10 percent over 12 years) [70,73]. The extent to which radioactive iodine therapy reduces recurrence in low-risk children is uncertain as data are lacking. (See 'Stratification by risk for persistent postoperative disease' above.)

•Adverse effects of radioactive iodine therapy – Radioactive iodine therapy appears to increase the risk of nonthyroid secondary malignancies over long-term follow-up [71,72]. An observational study of 3850 patients found an excess risk of 4.4 secondary malignancies per 10,000 patient-years attributable to radioactive iodine (equivalent to 1 excess malignancy per 227 treated patients over 10 years) [74]. (See "Differentiated thyroid cancer: Radioiodine treatment", section on 'Secondary malignancy'.)

Patients with intermediate- or high-risk PTC — After thyroidectomy, key management steps for patients with intermediate- or high-risk PTC are (table 6) [49]:

●Levothyroxine therapy – Initiate levothyroxine therapy. The target TSH range is between 0.1 and 0.5 mIU/L for intermediate-risk patients and <0.1 mIU/L for high-risk patients.

●Postoperative staging – Postoperative staging is performed around 4 to 12 weeks after thyroid surgery.

•Measure TSH-stimulated Tg – In children, this is most easily accomplished by discontinuation of levothyroxine for two weeks, followed by a serum TSH measurement; in the majority of patients, serum TSH will rise to >30 mU/L. If the TSH is <30, repeat after another week off of levothyroxine [75]. Alternatively, recombinant human TSH may be used to obtain a stimulated Tg ("off-label" use in children under 16 years) [76].

•Perform a diagnostic whole-body scan with 123-I when serum TSH is >30 mU/L. This is recommended for most patients at intermediate risk and all patients at high risk.

●Other imaging – If the 123-I scan shows regional uptake outside of the thyroid bed, the next step is to undertake anatomic imaging to locate the abnormal lymph node(s). Imaging typically starts with an ultrasound examination, but if abnormal lymph nodes are not identified, single-photon emission CT (SPECT/CT), CT (with or without contrast), or magnetic resonance imaging may be considered. Administration of iodine-containing contrast for CT must be considered carefully since it will delay potential radioactive iodine therapy. If abnormal lymph nodes are identified in a location that is amenable to surgery, surgical excision may be a reasonable treatment option. Patients with uptake outside of the thyroid bed that is not amenable to surgical resection or those who have distant metastases are candidates for radioactive iodine treatment.

●Radioactive iodine therapy – After thyroidectomy, for children with intermediate- or high-risk PTC and evidence of significant residual thyroid tissue or residual disease, we suggest treatment with radioactive iodine (131-I). The decision to treat depends upon the TSH-stimulated Tg level and whole-body 123-I scan. Radioactive iodine therapy is almost always indicated if TSH-stimulated Tg is >10 ng/mL and sometimes indicated for those with levels between 2 and 10 ng/mL.

Radioactive iodine treatment has been associated in observational studies with a lower rate of disease recurrence [63,77]. It also improves the sensitivity of post-treatment surveillance with Tg. In a retrospective study of 235 pediatric patients with PTC who were treated at a single center between 1973 and 2002, patients treated with radioactive iodine (n = 174) had fewer recurrences compared with those who did not receive radioactive iodine (n = 61; 0.6 versus 18 percent, respectively) [63]. After adjusting for other variables, radioactive iodine therapy remained a strong independent predictor of recurrence, with more than tenfold increased risk among patients who were not treated with radioactive iodine (odds ratio 11, 95% CI 1.3-95). The efficacy of radioactive iodine is also supported by indirect evidence from studies in adult patients with intermediate- and high-risk PTC. These data are discussed separately. (See "Differentiated thyroid cancer: Radioiodine treatment", section on 'Intermediate risk' and "Differentiated thyroid cancer: Radioiodine treatment", section on 'High risk'.)

In children, the treatment dose of radioactive iodine approximates 2 millicuries 131-I/kg (mCi/kg) body weight; selection of a specific dose typically is based on estimates of disease burden and requires clinical judgment. In patients with extensive pulmonary metastases, dosimetry may help to calculate a treatment dose [78].

●Subsequent levothyroxine therapy – Continue levothyroxine therapy using the TSH targets above.

Monitoring for recurrence — Ultrasound examination and serum Tg determinations are useful in monitoring for recurrence [49,70,79]. The timing of follow-up testing depends upon the patient's risk category (table 6).

Patients in whom ultrasound shows no abnormal findings and Tg levels are low (nonstimulated Tg <0.2 ng/mL or TSH-stimulated Tg <1 ng/mL in a patient who has undergone thyroidectomy and radioiodine ablation) have an excellent response to therapy and a low risk of recurrence. In intermediate- or high-risk patients who maintain an excellent response for five years after initial treatment, the initial TSH target range may be relaxed to 0.5 to 2 mIU/L.

Residual or recurrent disease is suggested by elevations in Tg. In a patient who has undergone thyroidectomy and radioactive iodine therapy, a nonstimulated serum Tg concentration ≥1 ng/mL or a TSH-stimulated Tg concentration ≥10 ng/mL likely reflects persistent or recurrent disease. Nonstimulated Tg ≥0.2 ng/mL or TSH-stimulated Tg ≥1 ng/mL is indeterminate for persistent disease. Suspicious ultrasonographic findings may also reveal persistent or recurrent disease. Patients with these findings should undergo studies to identify the location and extent of the residual or recurrent disease. Helpful studies include ultrasound examinations, 123-I radioisotope scans, and computed tomography (CT) scans. Treatment options include surgical excision of disease in localized lymph nodes or radioactive iodine treatment. Clinical judgment is required since some children may have evidence of "low-burden" persistent or recurrent disease yet do well with monitoring alone.

The presence of Tg antibodies (TgAb) may complicate the interpretation of serum Tg levels. In one study, 41 percent of children with thyroid cancer had positive TgAb, which disappeared in nearly one-half of the cohort within one to two years [80]. There is some evidence that the TgAb level can be used as a surrogate estimate of serum Tg (eg, a falling TgAb level is reassuring, whereas a rising level may correlate with disease recurrence) [81]. Assays for Tg using mass spectrometry are available and may decrease antibody interference in patients with TgAb, but some mass spectrometry assays may have suboptimal sensitivity for Tg [82]. (See "Differentiated thyroid cancer: Role of serum thyroglobulin", section on 'Monitoring response to therapy'.)

Follicular thyroid cancer — FTC represents approximately 8 to 9 percent of DTC in children. In contrast with PTC, FTC tends to be unifocal; it typically occurs in nodules with FNA cytology in BSRTC III or IV. FTC is more common in patients with PTEN hamartoma syndrome or DICER1 syndrome than in the general population; conversely, if a diagnosis of FTC has been made, patients should be examined carefully for clinical features of these syndromes (table 3). (See "PTEN hamartoma tumor syndromes, including Cowden syndrome".)

For patients with FNA findings concerning for a follicular neoplasm for whom surgery is recommended, the first step generally is a diagnostic lobectomy. If FTC is confirmed, the next step depends on whether angioinvasion is minimal (less than four vessels) or more extensive, a distinction that requires full pathologic examination after resection. For minimally invasive FTC, lobectomy usually is curative. If a widely invasive FTC is diagnosed, patients then undergo completion thyroidectomy to facilitate radioactive iodine scintigraphy and treatment. If FTC metastasizes, it is more likely by a hematogenous than lymphatic route. Thus, postoperative staging typically includes a 123-I whole-body scan. Metastases occur most commonly to the lung. Radioactive iodine therapy should be used for patients with distant metastases or a high risk of disease recurrence.

Medullary thyroid cancer — Total thyroidectomy is the treatment of choice for children with medullary thyroid cancer (MTC) [13]. (See "Approach to therapy in multiple endocrine neoplasia type 2", section on 'Thyroidectomy'.)

For pediatric patients without MTC but with a high risk for developing MTC because of RET gene mutations, "prophylactic" total thyroidectomy is recommended during infancy or early childhood [56]. The recommended age of "prophylactic" thyroidectomy in asymptomatic RET gene-positive infants or children is based on their risk level, which correlates with the specific RET mutation (table 7). In patients with highest-risk mutations (multiple endocrine neoplasia type 2B [MEN2B]), thyroidectomy is recommended during the first year of life. (See "Approach to therapy in multiple endocrine neoplasia type 2", section on 'Preventive surgery'.)

While awaiting the recommended age for thyroidectomy, patients with high- or moderate-risk RET gene mutations can be monitored by serum calcitonin measurements and periodic thyroid ultrasound. If ultrasound discloses a thyroid nodule or worrisome lymph nodes, the diagnosis can be confirmed by FNA and surgery should be undertaken. In addition, if serum calcitonin levels show an upward trend or are elevated, thyroidectomy is undertaken. Prophylactic central neck dissection may be considered if the preoperative calcitonin level is significantly elevated (>40 pg/mL).

PROGNOSIS

●Survival – Despite having more widespread disease at discovery compared with adults, children with thyroid cancer have higher survival rates, even in those with distant metastasis or recurrent disease [83]. This is illustrated in the following studies:

•In the multicenter review cited above, recurrence rates ranged from 10 to 35 percent with involvement of the thyroid bed, regional nodes, or distant sites [15]. However, only 2.5 percent died of their disease over periods ranging from 12 to 33 years. The presence of regional lymph node spread does not affect survival.

•In a nationwide study from the Netherlands of 170 children with differentiated thyroid cancer (DTC) treated between 1970 and 2013, overall survival was 99.4 percent after a median follow-up of 13.5 years (range 0.3 to 44.7 years) [84]. With this excellent survival, the authors concluded that minimizing treatment morbidity was a major priority.

•In the Surveillance, Epidemiology, and End Results (SEER) registry report of 2504 children with thyroid cancer, the 5-, 15-, and 30-year survival rates for patients with papillary thyroid cancer (PTC) were 100, 100, and 99 percent, respectively [85]. Even among children with distantly metastatic PTC, 30-year survival was 97 percent. In another study, survival rates were slightly lower for patients with follicular thyroid cancer (FTC; 96, 95, and 92 percent, respectively) and significantly lower for patients with medullary thyroid cancer (MTC; 95, 86, and 86 percent, respectively) [86].

•Long-term outcomes in 190 children and adolescents with PTC treated from 1936 through 2015 reported a 30-year cause-specific mortality of 1.1 percent [16].

●Nonthyroid secondary malignancy – Children with thyroid cancer have increased rates of nonthyroid secondary malignancies in long-term follow-up. As an example, in a cohort of 215 children with median follow-up of 25 years, 15 subjects (7 percent) developed a nonthyroid secondary malignancy [62]. Although the risk factors for developing nonthyroid secondary malignancies are not well defined, radioactive iodine therapy likely contributes to this risk.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Thyroid nodules and cancer".)

SUMMARY AND RECOMMENDATIONS

●Implications of thyroid nodules in children – Approximately 2 percent of children and adolescents have palpable thyroid nodules, while somewhere between 0.5 and 5 percent have nodules demonstrated by ultrasonography. Most thyroid nodules are asymptomatic. The majority of children are euthyroid, but children with an autonomous nodule will have a low serum thyroid-stimulating hormone (TSH). Most thyroid nodules in children are benign, but approximately 20 percent represent differentiated thyroid cancer (DTC) (table 1). (See 'Epidemiology' above and 'Clinical presentation' above.)

●Pathogenesis of thyroid cancer – Thyroid cancer in children is usually sporadic and arises from a de novo somatic genetic variant. A minority of patients have specific risk factors for thyroid cancer, including ionizing radiation (eg, treatment for head and neck cancers) or a predisposing genetic syndrome (table 3), such as PTEN hamartoma syndrome, DICER1 syndrome, or multiple endocrine neoplasia type 2 (MEN2) (table 2). Rarely, DTC has been found in association with thyroglossal duct cysts and congenital goiters associated with dyshormonogenesis. (See 'Pathogenesis of thyroid cancer' above.)

●Evaluation of thyroid nodules

•Children with thyroid nodules should be evaluated with serum TSH and thyroid ultrasound. Patients with low TSH are evaluated with scintigraphy for a toxic adenoma. In patients with normal/high TSH and solid nodules ≥1 cm, mixed solid/cystic nodules ≥1.5 cm with or without suspicious features, or nodules <1 cm and suspicious clinical context or features on ultrasound (table 4), we perform a fine-needle aspiration (FNA) of the lesion and the results inform decisions about surgery or further monitoring (algorithm 1). (See 'Evaluation of thyroid nodules' above and 'Fine-needle aspiration' above.)

•Children with suspected or confirmed thyroid cancer should be carefully evaluated for clinical features of genetic syndromes, especially PTEN hamartoma syndrome (Cowden or Bannayan-Riley-Ruvalcaba syndrome) and DICER1 syndrome, as outlined in the table (table 3). (See 'Genetic predisposition' above and 'History and physical examination' above.)

●Management of thyroid cancer – The main categories of thyroid cancer in children are papillary thyroid cancer (PTC; approximately 86 percent), follicular thyroid cancer (FTC; approximately 8 to 9 percent), and medullary thyroid cancer (MTC; approximately 4 percent). PTC and FTC are both DTCs. (See 'Histopathology' above.)

•Papillary thyroid cancer

-For most pediatric patients with PTC, we suggest total or near-total thyroidectomy rather than lobectomy (Grade 2C). This may reduce the risk for recurrence because PTC in children is often bilateral, which may not be detectable on preoperative imaging. (See 'Thyroidectomy' above.)

-Using criteria in the American Thyroid Association guidelines, patients with PTC can be classified postoperatively as low, intermediate, or high risk, and postoperative management and surveillance should be individualized based on risk level (table 6). After total thyroidectomy, children require treatment with levothyroxine. The dose is adjusted to keep the serum TSH concentration in the target range based on risk level. (See 'Stratification by risk for persistent postoperative disease' above.)

-For children with low-risk PTC, we suggest against radioactive iodine therapy (iodine-131 [131-I]) after thyroidectomy (Grade 2C). These children have excellent long-term outcomes, and avoiding radioactive iodine treatment may reduce their risk for secondary malignancy. (See 'Patients with low-risk PTC' above.)

-By contrast, for children with intermediate- or high-risk PTC and evidence of residual thyroid tissue or residual disease on postoperative staging, we suggest treatment with radioactive iodine (131-I) (Grade 2C). Radioactive iodine treatment has been associated with a lower rate of disease recurrence, and this benefit likely outweighs the risk of secondary malignancy associated with radioactive iodine treatment. (See 'Patients with intermediate- or high-risk PTC' above.)

-Noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP), previously referred to as a noninvasive follicular variant PTC (FVPTC), was renamed NIFTP to highlight its indolent behavior and <1 percent recurrence rate. NIFTP is rare in children and has not been well characterized in this population. These neoplasms are considered benign, but they have malignant potential and surgical excision (lobectomy) is required to confirm the diagnosis. (See 'Benign thyroid nodules' above.)

•Follicular thyroid cancer – FTC usually presents as a single thyroid nodule, suspected with a finding of FNA cytology in the Bethesda system for reporting thyroid cytopathology (BSRTC) category III or IV. It is more common in PTEN hamartoma syndrome and DICER1 syndrome than in the general population. Patients typically undergo initial lobectomy for diagnosis. For minimally invasive FTC, lobectomy usually is curative. Children with invasive FTC require completion thyroidectomy followed by evaluation for metastases by a 123-I whole-body scan. (See 'Follicular thyroid cancer' above.)

•Medullary thyroid cancer – MTC in children almost always occurs in association with MEN2. Children at risk for MTC, usually based on discovery of a relative with MEN2 or MTC, should undergo RET protooncogene analysis. Those who have a confirmed RET gene mutation are usually treated with prophylactic total thyroidectomy. Recommendations for preventive surgery in patients with MEN2 are summarized in the table (table 7) and discussed in detail separately. (See 'Genetic testing for germline mutations' above and 'Medullary thyroid cancer' above and "Approach to therapy in multiple endocrine neoplasia type 2", section on 'Preventive surgery'.)

●Prognosis – Long-term survival rates for PTC or FTC approximate 98 percent. Five-year survival rates for MTC are also over 90 percent, but 30-year survival rates are poor. (See 'Prognosis' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Stephen LaFranchi, MD, who contributed to earlier versions of this topic review.