Clinical features of thyroid cancer in paediatric age. Experience of a tertiary centre in the 2000–2020 period

To describe the clinical features of a paediatric cohort affected by differentiated thyroid cancer (DTC) followed in a tertiary Department of Paediatric Endocrinology. Clinical data of 41 patients affected by DTC in the 2000–2020 period were reviewed. The main risk factor was autoimmune thyroiditis (39%). Cytological categories were TIR3b in 39%, TIR4 in 9.8%, TIR5 in 51.2%. After total thyroidectomy, radioiodine treatment was performed in 38 subjects (92.7%). ATA low-risk category was assigned in 11 (30.5%), intermediate-risk category in 15 (41.7%), and high-risk category in 10 patients (27.8%). Age at diagnosis was 15.1 ± 0.92 years in low-risk category, 14.7 ± 0.59 in intermediate-risk category, 11.7 ± 0.89 years in high-risk category (p = 0.01). TIR3b was manly observed in low-risk class (63.6%), while TIR5 was mainly reported in intermediate and high-risk class (60 and 80% respectively) (p = 0.04). Post-surgery stimulated thyroglobulin was increased in high-risk class (407.8 ± 307.1 ng/ml) [p = 0.04]. Tumour size was larger in high-risk category (42.6 ± 2.6 mm), than in low and intermediate-risk categories (19.4 ± 3.5 mm and 28.5 ± 3.9 mm, respectively) (p = 0.008). Patients in intermediate and high-risk categories displayed more tumour multifocality (60 and 90% respectively) (p < 0.005). Disease relapse was mainly observed in high risk category (40%, p = 0.04). DTC in childhood is more aggressive than in adults, but the overall survival rate is excellent. The therapeutic approach is still heterogeneous, especially in low-risk category. Further studies are needed to standardise management and reduce disease persistence in childhood.


Introduction
Thyroid nodules in paediatric age are less frequent than in adults (0.2-0.5 vs 19-35%), but the rate of malignancy is higher, reaching up to 20-25% during childhood [1,2]. Diagnostic evaluation include clinical, biochemical and ultrasound assessment. Fine-needle agobiopsy (FNAB) for cytological category determination is required when suspicious ultrasound features of malignancy or clinical risk factors are present [3][4][5][6][7][8][9][10][11][12][13]. Previous radiation exposure or cancer treatment, underlying thyroid disease, family history for thyroid cancer and predisposing genetic mutations represents known risk factors for thyroid cancer [1]. Based on the cytological assessment, the decision on further management is made, from clinical-ultrasound follow-up, repeat FNAB, lobectomy or total thyroidectomy. The diagnosis of thyroid cancer is defined in the case of a cytological category of certain malignancy or a definitive histological finding for thyroid cancer after surgery.
Differentiated thyroid cancer (DTC) represents the most common solid tumour in paediatric age, with a reported incidence of 0.2-1 per million child-years, which is higher in adolescence and in females. Papillary thyroid cancer (PTC) is the most frequent histological finding and accounts for 90% of DTC; follicular thyroid cancer (FTC) is less common in paediatric age (5%), as well as the mixed histologic type (3%), while medullary thyroid cancer (MTC) represents 2%.
Although children more frequently display multifocality, local extension, lymph node involvement, distant metastases, persistent disease and risk of recurrence, the overall survival rate is higher than in adults and the prognosis is excellent [27][28][29]. Younger age and advanced T and N stages are risk factors for multifocality, which is predictive of bilateral disease [30,31]. The lungs are the most frequent site of distant metastases, occurring in 6-33% of children with DTC [32]. Follicular histologic type, basal pre-ablation thyroglobulin >14 ng/ml, or stimulated thyroglobulin >154 ng/ml are the main known factors associated with a higher risk of distant metastases [33][34][35].
Persistent disease occurs more frequently with higher T stage and when lymph node or distant metastases are present or when high stimulated thyroglobulin is observed after surgery [36][37][38].
The risk of recurrence has been estimated to be 20-40% at 10 years and surveillance protocols are mandatory [39]. Younger age, partial thyroidectomy, involvement of more than 5 lymph nodes or incomplete primary lymph node excision, distant metastases and absence of RAI ablation are the main risk factors for recurrence [38,40].
Total thyroidectomy with local lymph node dissection is in most cases the first-line treatment [11,12,41]. Some authors suggest considering lobectomy in case of tumour size <10 mm and absence of extra-thyroidal invasion, although this approach is still under discussion [42][43][44]. American Thyroid Association (ATA) guidelines established risk stratification (low, intermediate, and high) according to the American Joint Committee on Cancer (AJCC) Tumour, Nodes, and Metastases (TNM) classification system [12]. Some authors have suggested dynamic risk stratification (DRS) to increase the accuracy of risk class assignment [42]. Radioiodine (RAI) therapy is necessary in cases of tumour larger than 40 mm, lymph node involvement and distant metastases [11,12,41]. The use of RAI in patients in the low-risk category patients is still under discussion. Post-treatment staging with 131 Iscintiscan is useful for determining residual disease and for evaluating subsequent follow-up strategies.
The aim of this study is to describe the clinical features of a paediatric cohort affected by differentiated thyroid cancer followed in a tertiary Department of Paediatric Endocrinology for a period of 20 years.

Materials and methods
The data of all paediatric (<18 years) patients affected by differentiated thyroid cancer referred to the Department of Paediatric Endocrinology at Regina Margherita Children's Hospital of Turin in the period 2000-2020 were retrospectively reviewed. Demographic, clinical and biochemical (thyroid hormone profile, thyroglobulin and calcitonin levels and anti-thyroglobulin antibodies) data were analysed for each patient. Patients with suspicious ultrasound of malignancy or with known risk factors for developing DTC were referred for FNAB. All specimens were evaluated in the same hospital Pathological Anatomy Department. The cytological category was assigned according to the Italian Society of Anatomical Pathology and Diagnostic Cytology (SIAPeC). The American Joint Committee on Cancer (AJCC) Tumour, Nodes, metastases (TNM) classification system was used to assign the post-surgical class of risk, according to ATA guidelines.
All patients underwent total thyroidectomy and lymph node dissection. Surgery was performed in a high-volume surgery centre for thyroid disease (>150 thyroidectomies/ year). RAI therapy was performed in all patients included in the intermediate and high risk categories, while case-bycase evaluation was performed in the low-risk category. In all patients, pre-ablation TSH increase was achieved by discontinuing levothyroxine for 30-40 days before RAI.
Persistent disease was considered in case of subtotal thyroidectomy, elevated post-ablation stimulated thyroglobulin and residual tissue caption at 131 Iscintiscan investigation.
Disease recurrence was considered in case of significant increase of thyroglobulin level during follow-up after surgery and/or RAI treatment.
Follow-up was performed by clinical, radiological (ultrasound or 131 Iscintiscan) and biochemical (thyroid hormone profile, basal and stimulated thyroglobulin levels) assessment after treatment.
Statistical analyses were performed through Graphpad 7 software (GraphPad Software, La Jolla, CA, USA), using ANOVA test to compare means and chi-square test to compare differences between groups. Significant difference between means was considered with p < 0.05.
The study was conducted according to the guidelines of the Declaration of Helsinki and received the approval of the Ethics Committee of the City of Health and Science University Hospital of Turin (number of approval 13942/A1). Informed parental consent was obtained in all cases.
Of the 36 subjects affected by PTC, according to ATA criteria, 11 patients were assigned to the low-risk (30.5%), 15 to the intermediate-risk (41.7%) and 10 to the high-risk category (27.8%) [ Table 2]. The three patient who did not receive RAI treatment were assigned to the low-risk category and presented tumour dimension lower than 1 cm.
No differences were observed between risk categories with respect to gender, lymph node involvement and distant metastases, positive titration of anti-thyroglobulin antibodies, RAI treatment and 131 I dose.
Age at diagnosis was 15.1 ± 0.92 years in the low-risk category, 14.7 ± 0.59 in the intermediate-risk category and 11.7 ± 0.89 years in the high-risk category (p = 0.01).
Patients assigned to intermediate and high-risk categories showed tumour multifocality more frequently, found in 9/15 (60%) and 9/10 (90%) patients, respectively (p < 0.005). Persistence of disease was observed in 2 (20%) patients assigned to the high-risk class. Radical thyroidectomy was not possible in both patients.

Discussion
The management of thyroid cancer in paediatric age is still a matter of debate. Compared to adults, young patients have more advanced disease at presentation, but the clinical course is not complicated, and overall survival is excellent [27][28][29].
Among thyroid disease, autoimmune thyroiditis is the most frequent risk factor associated with thyroid nodule, but it is not clear whether the rate of malignancy is also higher [45]. Grave's disease and congenital hypothyroidism are also considered predisposing factors for the development of thyroid cancer [1]. In such cases, a periodic ultrasound evaluation is therefore suggested, even if to date there is no unanimous opinion. To avoid redundant procedures and patient and familial anxiety, evaluation may be indicated every 2 years in AIT, every year in Grave's disease and every 3-4 years in congenital hypothyroidism with eutopic gland or ectopic thyroidal tissue.
Previous radiation exposure, particularly of the head and neck region, is a known risk factor for thyroid malignancy [46][47][48][49][50]. An irradiation dose <30 Gy is related to an increased risk of second cancer, while higher doses lead to tissue necrosis. Many chemotherapy agents are also related to thyroid malignancy in childhood cancer survivors. Again, there is no solid evidence of ultrasound monitoring in these subjects, but evaluation every 2 years beginning 5 years after cancer off-therapy, may be considered reasonable.
The most approved surgical approach in DTC is total thyroidectomy, although some authors suggest lobectomy in case of low-risk thyroid neoplasia (dimension <10 mm and non-extrathyroidal involvement) [11,12,42,43]. Total thyroidectomy is related to a lower risk of recurrence, more effective radioiodine ablation, and more reliable thyroglobulin measurement during the postoperative follow-up. Conversely, lobectomy is related to a lower rate of surgical complications such as hypoparathyroidism, recurrent laryngeal nerve injury, and bleeding. In our cohort, almost all patients underwent total thyroidectomy, with no reported cases of permanent recurrent laryngeal nerve injury and only one case of permanent hypoparathyroidism. Near-total thyroidectomy was performed in 2 patients due to tumour adherences to important anatomic structures such as the oesophagus and recurrent laryngeal nerve. Both patients showed persistent but stationary disease after three sessions of RAI, confirming the need for a radical surgical approach, when possible. To reduce total thyroidectomy surgery complications, persistent disease rates, and tumour relapse, it is critical that total thyroidectomy is performed by an experienced surgeon with highvolume surgery for thyroid diseases [51]. After surgery, RAI treatment is required in case of tumour size >40 mm, lymph node involvement and distant metastases, although recent European guidelines recommend RAI ablation in all patients with DTC. In the present cohort, we decided to perform RAI treatment in all patients but three subjects included in the low-risk ATA class. Considering the young age of children or adolescents with DTC and the limited evidence of RAI safety data, a case-bycase approach may be reasonable in subjects assigned to the low-risk category, rather than absolute recommendation for all patients [11,12,41,52,53].
The majority of patients in our cohort were assigned to the intermediate-risk class followed by low and high-risk categories. We observed no differences in gender, lymph node involvement and distant metastases, positive titration of anti-thyroglobulin antibodies, RAI treatment and 131 I dose among the three risk categories.
According to literature data, the patients assigned to the high-risk class in our study were younger [12]. The indeterminate cytological category (TIR3b) was mainly observed in the low-risk category. To our knowledge this data has not been reported before in the literature.
Tumour size greater than 40 mm and lobar multifocality were mainly observed in intermediate and high-risk classes, confirming their negative prognostic role. We observed more severe induced hypothyroidism before RAI ablation in high-risk patients, probably due to longer withdrawal period to increase tissue avidity for 131 I. The postoperative level of stimulated thyroglobulin was increased in the high-risk class compared to the intermediate and low-risk category, indicating a probable association with a negative prognostic role. Subjects included in the high-risk class showed a higher relapse rate than other risk categories, in accordance with literature data. Only 1 subject with relapsed disease was included in the low-risk ATA category. In this case, post-surgical basal thyroglobulin was persistently elevated, even though there were no other associated negative prognostic factors. These data confirm the need to increase the accuracy of the ATA criteria for risk class assignment by adding the well-known prognostic factors to the initial class assignment or by following a dynamic risk stratification approach [42].
In conclusion, we confirm the more aggressive behaviour of DTC in paediatric age compared to adults. Despite this, the overall survival rate is excellent. The therapeutic approach is still heterogeneous, especially in the low-risk category. Further studies are needed to standardise management across tertiary centres and reduce disease persistence in childhood.
Author contributions All authors contributed to the study design and conception. Material preparation, data collection and analysis were performed by A.M., J.M. and G.T. The clinical management of the enroled subjects was managed by A.C., P.M., G.T., J.M., F.Q. and L.D. Sanctis. The first draft of the manuscript was written by G.T. and J.M. and all authors have commented on previous versions of the manuscript. All authors have read and approved the final manuscript.

Compliance with ethical standards
Conflict of interest The authors declare no competing interests.
Consent to publish Consent to publish has been received from all partecipants.
Consent to partecipate Written parental consent was obtained.
Ethics approval This study was conducted in line with the principles of the Declaration of Helsinki. The approval was granted by the Ethics Committee of the City of Health and Science University Hospital of Turin (01/10/2021, 13942/A1).