Imaging of Pediatric Thyroid Tumors: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper

Abstract

Pediatric thyroid cancer is rare in children, however incidence is increasing. Papillary thyroid cancer and follicular thyroid cancer are the most common subtypes, comprising about 90% and 10% of cases respectively. This manuscript provides consensus imaging recommendations for evaluation of pediatric patients with thyroid cancer at diagnosis and during follow-up.

Introduction

Thyroid cancer is relatively rare in children and adolescents, accounting for only 1.8% of the total thyroid cancers diagnosed in the United States annually, but the incidence is increasing.^1–4 Female predominance of thyroid cancer increases with age, approximating a 6:1 female to male ratio by age 15 to 19 years, making thyroid cancer the second most common malignancy in this age/gender group.^1,5 In children, thyroid cancer may be asymptomatic and discovered incidentally at imaging for another purpose, or may present as a painless palpable thyroid mass, neck mass related to cervical lymphadenopathy, difficulty breathing, and/or hoarseness⁶

Thyroid cancer in children is classified into differentiated, poorly differentiated, and medullary subtypes. Subtypes of differentiated thyroid cancer include papillary thyroid cancer (PTC) and follicular thyroid cancer (FTC) subtypes. PTC is most common, accounting for up to 90%, and is further subdivided into classical, follicular, solid, and diffuse sclerosing variants.^1,2 FTC accounts for up to 10% of thyroid cancer.²

Risk factors for developing PTC and FTC include autoimmune thyroid disease (i.e. Hashimoto’s thyroiditis and Graves’ disease), iodine deficiency, and prior therapeutic radiation exposure.^6,7 Genetic predisposition syndromes including APC-associated polyposis, DICER1 syndrome, Carney Complex, PTEN hamartoma syndrome, and Werner syndrome may account for up to 5% of differentiated thyroid cancers.^6,7 Additionally, Li-Fraumeni, Peutz-Jeghers, familial paraganglioma, McCune-Albright, and Beckwith-Wiedemann syndromes are also associated with an increased risk of differentiated thyroid cancer.⁷ Genetic mutations of the BRAF, RET/PTC, ALK, RAS, and PAX8/PPARγ genes play a role in thyroid cancer pathogenesis.^4,8 The RET/PTC genetic rearrangement is relatively more common in children compared to adult thyroid cancers, while BRAF mutations are less commonly encountered in children than adults with PTC.^6,9 Finally, thyroid cancer is one of the most common secondary malignancies in childhood cancer survivors.^7,10

Poorly differentiated and medullary thyroid cancers are rare in the pediatric population.⁷ Most medullary thyroid carcinomas are hereditary, related to germline RET mutations causing multiple endocrine neoplasia (MEN) type 2A, MEN 2B, or familial medullary thyroid carcinoma (FMTC) and incidence is highest in the 0 to 4 year age group.^1,2,7

In children, thyroid nodules occur in 0.5-5% of the population, however, there is a 19-25% rate of malignancy, highlighting the importance of nodule identification.^11–13 Compared to adults, children with thyroid cancer are more likely to present with larger tumor size and extrathyroidal extension.^9,14 Children also have increased rates of regional lymph node and pulmonary metastasis compared to adults, at 60-80% and 10-25% respectively.^9,14 However, despite more extensive disease at presentation, the mortality rate from thyroid cancer is lower in children than adults, with overall survival of 95% at 20 years in children.^9,14

Imaging in Tumor Staging

The American Joint Committee on Cancer (AJCC) and Union International Contre le Cancer (UICC) Tumor-Node-Metastasis (TNM) staging system is recommended for pediatric thyroid carcinoma (GRADE: A; SOR: 1.21, very strong recommendation). Use of this post-operative staging system is recommended by the American Thyroid Association (ATA).⁹ The AJCC/UICC-TNM staging system incorporates size of tumor and extrathyroidal extension, regional lymph node status, and distant metastasis, and its main purpose is to define an anatomic-based classification of the extent of disease, to guide further management (Table 1).

TABLE 1.

American Joint Committee on Cancer (AJCC) Tumor-Node-Metastasis (TNM) staging system (8^th edition) for differentiated thyroid carcinoma (adapted from reference 9).

Tumor (T)
T Category		Criteria
TX		Primary tumor cannot be assessed
T0		No evidence of primary tumor
T1		Tumor ≤ 2cm limited to the thyroid
	T1a	Tumor ≤ 1cm
	T1b	Tumor > 1cm but ≤ 2cm
T2		Tumor > 2cm but ≤ 4cm limited to the thyroid
T3		Tumor > 4 cm limited to the thyroid or gross extrathyroidal extension invading only the strap muscles (sternohyoid, sternothyroid, thyrohyoid, or omohyoid muscles)
	T3a	Tumor > 4 cm
	T3b	Gross extrathyroidal extension invading only the strap muscles from tumor of any size
T4		Gross extrathyroidal extension into major neck structures
	T4a	Gross extrathyroidal extension invading subcutaneous soft tissues, larynx, trachea, esophagus, or recurrent laryngeal nerve from a tumor of any size
	T4b	Gross extrathyroidal extension invading prevertebral fascia or encasing carotid artery or mediastinal vessels from a tumor of any size
Regional Lymph Node (N)
N Category		Criteria
NX		Regional lymph nodes
N0		No evidence of metastatic regional lymph nodes
	N0a	No evidence of lymph node metastasis confirmed cytological or histologically
	N0b	No radiologic or clinical evidence of locoregional lymph node metastasis
N1		Metastasis to regional lymph nodes
	N1a	Metastasis to lymph node levels VI or VII (pretracheal, paratracheal, or prelaryngeal/Delphian or upper mediastinal). Either uni- or bilateral disease
	N1b	Metastasis to unilateral, bilateral, or contralateral lateral neck lymph nodes (levels I, II, III, IV, or V) or retropharyngeal lymph nodes
Distant Metastasis (M)
M Category		Criteria
M0		No distant metastasis
M1		Distant metastasis

Based on the AJCC/UICC-TNM classification system, ATA guidelines classify children with thyroid carcinoma into risk groups (Table 2). Stratification of patients into ATA risk groups is recommended as this allows estimation of disease-free survival (Table 2), identifies patients at risk for residual/persistent cervical lymphadenopathy, and guides further management by identifying patients that may require further imaging to define distant metastasis and that may require ¹³¹I therapy. (GRADE: B; SOR 1.36, very strong recommendation)

TABLE 2.

Pediatric thyroid cancer patient risk groups defined by ATA guidelines⁹

Low-Risk	Tumor confined to the thyroid gland, N0/NX, or incidental minimal (microscopic) N1a to few central neck (Level VI) lymph nodes. Possible risk for residual cervical disease, mainly if no surgical central nodal dissection. Low risk for distant metastatic disease.
Intermediate-Risk	Extensive (Level VI) N1a nodes or minimal metastases to unilateral, bilateral or contralateral Levels I, II, III, IV or V cervical or superior mediastinal (N1b) nodes. High risk for incomplete surgical lymph node resection. High risk for persistent cervical disease. Low risk for distant metastasis.
High-Risk	Extensive regional (N1b) disease, or extrathyroidal invasive disease (T4), +/− distant metastasis. High risk for incomplete surgical resection. High risk for persistent disease. High risk for distant metastasis.

Imaging Modalities

Ultrasound is the cornerstone of imaging thyroid cancer, and plays a role in diagnosis, staging, operative planning, and disease monitoring in pediatric patients.¹⁵ The advantages of ultrasound are that it is portable, can be performed quickly, is cost-effective, and requires no patient preparation and no sedation or anesthesia.¹⁶ The high spatial resolution and ability to detect microcalcifications as a finding of tumor are additional advantages, and ultrasound is superior to other imaging modalities at characterizing thyroid nodules.¹⁷ The primary disadvantage of ultrasound is that it is operator-dependent, and screening for cervical nodal metastatic disease requires experience and meticulous technique.

Computed tomography (CT) may play a role in imaging locoregional metastatic disease as well as pulmonary metastases. However, use of iodinated IV contrast material is discouraged because it interferes with radioactive iodine (RAI) uptake, and if RAI therapy is deemed necessary, treatment must be delayed for weeks to 3 months. CT scanning parameters should be optimized for patient size to minimize radiation exposure. Additionally, sedation or anesthesia may be required to achieve breath-holds necessary to detect miliary pulmonary metastases in patients unable to comply with breath holding instructions.

Magnetic resonance imaging (MRI) in patients with thyroid cancer is primarily used in the evaluation of bulky cervical metastatic disease and extrathyroidal invasion.¹⁸ Although the spatial resolution of MRI is lower than both ultrasound and CT, the tissue contrast resolution is higher. However, punctate calcifications relevant to identifying disease are not apparent by MRI. MRI examinations are lengthy, which may require use of sedation or anesthesia to ensure patient immobility.

RAI scintigraphy capitalizes on the physiologic uptake of iodine by most differentiated thyroid cancers and may be used in post-surgical disease staging. ¹⁸F-FDG PET/CT is not typically used but may play a role in the evaluation of undifferentiated tumors that are not iodine avid. ⁶⁸Ga-DOTATATE PET/CT may be useful for medullary thyroid cancer disease evaluation.¹⁹

Imaging at Diagnosis

High resolution ultrasound of the thyroid and neck with a high frequency (12-18 MHz) linear transducer is recommended as the primary imaging modality for tumor diagnosis. (GRADE: A; SOR 1.07, very strong recommendation) This enables evaluation of the morphologic features of the thyroid nodule as well as location within the thyroid, evaluation for extrathyroidal extension, and involvement of important adjacent anatomic structures, which may impact management.

At ultrasound, there are both pattern-based and point-based risk stratification guidelines that have been evaluated in both adults and children to differentiate benign and malignant thyroid nodules.^{9,12,16,20–25} Generally, features such as solid composition, taller than wide orientation, irregular margins, microcalcifications or punctate echogenic foci, and extrathyroidal extension are considered suspicious for malignancy. It is important to recognize intrathyroidal ectopic thymic tissue at ultrasound, which appears as a hypoechoic nodule with linear and punctate echogenic foci, is unique to children, and should not be mistaken for PTC.^26,27 Although color Doppler may be useful for distinguishing solid components from debris in nodules and may be useful to predict bleeding risk during fine-needle aspiration (FNA), Doppler pattern appears to be less helpful in determining malignancy than grayscale appearance.^28,29 An important distinction between adult and pediatric guidelines is the size threshold to guide FNA decisions. Although adult guidelines specify nodule size cut-offs to proceed to FNA, in children, size thresholds are not recommended in the decision-making process, but rather the ultrasound appearance is prioritized and a lower threshold to proceed with further diagnostic work-up recommended, particularly in children with risk factors.^{5,9,12,16,23,30} (GRADE: B; SOR 1.64, strong recommendation) This is because in adults, the goal of imaging is not to diagnose every thyroid malignancy, but to balance the benefit of identifying clinically significant cancers against the cost of subjecting patients with benign nodules or indolent cancers to unnecessary treatment.^20,31

Ultrasound lymph node mapping of the neck with a meticulous evaluation of lymph node levels in the central neck (level 6), lateral neck (levels 1-5), and mediastinum (level 7) is required because PTC metastasizes to regional lymph nodes in most children.⁹ (GRADE: B; SOR 1.57, strong recommendation) Ultrasound is highly sensitive and specific for predicting cervical lymph node metastasis preoperatively.³² Preoperative suspicion of locoregional metastatic disease is important to plan an appropriate, compartment-oriented lymph node dissection at the time of initial surgery.⁶ Preoperative ultrasound has been shown to improve surgical outcome, decrease rate of recurrence or need for more surgeries, and to guide further medical therapy.^33–35

Ultrasound guidance is recommended for FNA of the thyroid nodule, targeted to the solid or most suspicious component of the nodule to provide the highest diagnostic yield specimen. (GRADE: A; SOR 1.28, very strong recommendation) Ultrasound guidance is also recommended to guide FNA of suspicious lymph nodes, if needed preoperatively to plan lymph node dissection approach.⁹ (GRADE: B; SOR 1.93, strong recommendation) Ultrasound-guided FNA is both sensitive and specific to diagnose pediatric thyroid cancer. Without ultrasound guidance, rates of non-diagnostic and false negative thyroid nodule cytologic results are higher.^9,16

CT or MRI of the neck is not routinely recommended, but is reserved for select cases where bulky lymphadenopathy or large tumor burden can hinder ultrasound visualization of the deep compartments of the neck (levels 6 and 7, retropharyngeal, and supraclavicular regions) or if local invasion is suspected.^18,36 Neck CT requires iodinated IV contrast material injection to adequately visualize anatomy, and therefore is not recommended. Neck CT without IV contrast is not recommended. Therefore, neck MRI is preferred over CT in the evaluation of the extent of bulky cervical metastatic disease prior to surgery. (GRADE C; SOR 2.0, moderate recommendation)

Although current ATA guidelines recommend either chest radiographs or CT in intermediate and high-risk patients to evaluate for pulmonary metastatic disease, CT is the most sensitive imaging modality for this purpose.³⁷ Therefore, CT of the chest without IV contrast should be performed in initial staging to detect pulmonary metastases in patients in the ATA Intermediate and High-Risk categories. (GRADE: C; SOR 1.92, strong recommendation) Chest CT is not routinely recommended in patients categorized as Low-Risk. (GRADE: C; SOR 1.86, strong recommendation) While intravenous contrast material can improve detection of mediastinal and hilar lymphadenopathy in the chest, pulmonary metastases can be detected without IV contrast. Axial imaging with 3 mm or smaller slice thickness complemented by coronal and sagittal reconstructions is recommended, with maximal intensity projections (MIPs). Use of MIPs has been shown to improve the detection of small pulmonary nodules.³⁸

Imaging Post-Thyroidectomy

Whole-body RAI scintigraphy with ¹²³I is recommended within 12 weeks following surgery for postoperative staging in ATA Intermediate and High-Risk patients to detect residual locoregional disease and distant metastasis ³⁷ in order to identify patients who may benefit from additional surgery or RAI for remnant ablation or therapy.³⁹ (GRADE A; SOR 1.07, very strong recommendation) ¹²³I is preferred due to superior imaging resolution, the ability to utilize single photon emission computed tomography with integrated conventional CT (SPECT/CT), and slightly lower dose to the patient, however higher cost may be prohibitive. In such cases, ¹³¹I remains an acceptable alternative. While planar imaging is generally used for whole-body RAI scintigraphy, the addition of targeted SPECT/CT when focal abnormal uptake is identified offers improved disease localization and characterization.⁹ Therefore, when focal RAI uptake is identified on planar imaging, SPECT/CT is recommended. (GRADE C; SOR 2.36, moderate recommendation).

Imaging Off Therapy/Surveillance

Six months postoperatively, all patients without evidence of active disease should undergo surveillance imaging with neck ultrasound, including the surgical bed, central and lateral neck, and upper mediastinum.⁹ (GRADE C; SOR 2.0, moderate recommendation) For patients in the ATA pediatric low risk level, ultrasound surveillance should be performed annually for 5 years. In both the ATA pediatric Intermediate and High-Risk groups, ultrasound surveillance should be performed every 6-12 months for 5 years. Neck ultrasound can continue less frequently after 5 years for the intermediate and high-risk levels based on individual recurrence risk.⁹ Thyroglobulin levels can be useful to guide imaging assessment as a marker of residual or recurrent disease and are performed at similar time points.

¹²³I-diagnostic whole-body scan is recommended in patients treated with RAI at 1-2 years following therapy. (GRADE C; SOR 1.36, very strong recommendation) ¹²³I-diagnostic whole-body scan is also recommended in patients with increasing thyroglobulin levels or with concern for recurrence at other imaging.⁹ (GRADE A; SOR 1.07, very strong recommendation) Table 3 details timing of imaging surveillance.

TABLE 3.

Surveillance imaging timeline based on ATA risk group

	ATA Low-Risk	ATA Intermediate-Risk	ATA High-Risk
Neck ultrasound	Annually	Every 6-12 months	Every 6-12 months
Whole-body 123I scan	As needed if rising thyroglobulin	1-2 years following 131I therapy or if rising thyroglobulin	1-2 years following 131I therapy or if rising thyroglobulin

Are there late effects that change the goal of surveillance imaging?

Differentiated thyroid cancer recurrence has been reported several decades after initial treatment, therefore long-term surveillance is necessary.⁹ In addition, relative risk of a secondary primary malignancy, most commonly leukemia, is increased in those treated with RAI.^39–43

Advancements in imaging

Novel ultrasound techniques may prove useful in the future for both the diagnosis and follow-up of children with thyroid cancer but are not currently standard of care. Ultrasound elastography is a non-invasive method to evaluate the stiffness of tissue. Its use has been studied in adults and shown to increase the sensitivity and negative predictive value in the assessment of thyroid nodules for malignancy.⁴⁴ Ultrasound elastography may also be useful in the future to help distinguish benign from malignant cervical lymph nodes.⁴⁵ However, further study is needed in children.²⁵

Contrast-enhanced ultrasound (CEUS) is another newer ultrasound technique that uses the IV administration of an ultrasound contrast agent to assess enhancement features. CEUS has been shown to increase the accuracy of thyroid nodule diagnosis in adults, particularly when combined with grayscale ultrasound features.^46,47 Confirmatory studies of its utility in children with thyroid nodules are needed.

Abbreviations key:

PTC

Papillary thyroid cancer

FTC

Follicular thyroid cancer

FMTC

Familial medullary thyroid carcinoma

AJCC

American Joint Committee on Cancer

UICC

Union International Contre le Cancer

TNM

Tumor-Node-Metastasis

ATA

American Thyroid Association

CT

Computed tomography

RAI

Radioactive iodine

MRI

Magnetic resonance imaging

FNA

Fine needle aspiration

SPECT

Single photon emission computed tomography

CEUS

Contrast-enhanced ultrasound