The defined 5 types of thyroid abnormalities: (a) hypothyroidism: abnormal laboratory results (thyroid stimulating hormone [TSH] > 5.5 μIU/mL; thyroxine [T4] < 0.63) and clinical manifestations; (b) hyperthyroidism: abnormal laboratory results (TSH<0.36 μIU/mL, free thyroxine [fT4] > 1.4ng/dL) and clinical manifestations; (c) subclinical hypothyroidism: abnormal laboratory results (TSH>5.5 μIU/mL, fT4 and free triiodothyronine [fT3] within the normal range) in absence of compatible symptoms; (d) subclinical hyperthyroidism (TSH<0.36 μIU/mL or undetectable and fT4 and fT3 within the normal range) in absence of symptoms and (e) thyroid nodules: detected by ultrasound.

We generated contingency tables grouping the 5 abnormalities into 3 categories: hypothyroidism overall (clinical hypothyroidism+subclinical hypothyroidism), hyperthyroidism overall (clinical hyperthyroidism+subclinical hyperthyroidism) and thyroid nodules. We assessed the association of these categories with the RTX, alkylating agents and combined RTX+alkylating agents treatment variables.

The study was designed and carried out in adherence to the International Code of Medical Ethics and the principles of the Declaration of Helsinki. Confidentiality was strictly safeguarded in the handling of personal data, conforming to the current legislation on personal data protection: Organic Law 3/2018, of 5 December, on the Protection of Personal Data and the Guarantee of Digital Rights, and Regulation (EU) 2016/679 of the European Parliament and of the Council. The data used in the study were retrieved from the health records of the patients and were identified using a numerical code and collected in a research file in the centre. The study did not require direct contact with the patients and all the data were collected retrospectively for a study period spanning 15 years, so we requested the exemption from informed consent as established in current law. The project was approved by the Ethics Committee of Research with Medicines of our hospital (file R-0089/22; 31 January 2023).

In our study, the overall frequency of thyroid abnormalities (laboratory and sonographic) was 16.7%. Of these abnormalities, 6.7% corresponded to cases of hypothyroidism (clinical or subclinical). Hypothyroidism is the thyroid complication described most frequently in survivors of childhood HL.9 Some studies have found frequencies of up to 60%, depending on the population under study and the criteria used to define hypothyroidism.5 The proportion of cases in our study, which is considerably lower, was probably due to the small sample size compared to most other studies and the shorter duration of follow-up.

We found a statistically significant association between the use of RTX and the development of hypothyroidism. Neck irradiation is the main risk factor for the development of hypothyroidism in oncological patients, and in most studies the disorder tends to appear in approximately 30% of cases within 10 years of completing RTX.5 However, the risk of radiation-induced hypothyroidism continues to be high even 25 years after exposure to radiation.10 Furthermore, there is evidence that the risk of developing hypothyroidism after RTX is dose-dependent.9 Although we did not find statistically significant differences in the RTX dose between the group that did not develop thyroid abnormalities and the group that did, the impact of the use of RTX in the development of thyroid complications was strong and sizeable. Taking into account the small sample size in our study, the fact that these associations were statistically significant corroborates the role of RTX as a cause of thyroid damage. The median radiation dose received by patients with thyroid abnormalities was 29.8Gy. A study conducted by Sklar et al. found a prevalence of hypothyroidism of 30% associated to doses of 30–45Gy and of up to 50% for doses greater than 45Gy. In addition, in the latter group, the risk of developing hypothyroidism within 20 years of diagnosis was 50%.4 Duarte et al. found that radiation doses greater than 35Gy had a significant impact (P<.01) on the development of hypothyroidism. The findings of this study also suggested that the risk was greater in patients treated with neck and craniospinal irradiation compared to total body irradiation,11 in opposition to the study of Madanat et al., who did not find differences based on the irradiation field.12 There is also evidence of the benefits of hyperfractionated RTX compared to conventional RTX and the use of protons instead of photons to minimise the exit dose and thus the damage beyond the target tissues.8 Other risk factors associated with the development of hypothyroidism secondary to RTX are the age of exposure, dose of RTX, duration of exposure and time elapsed since the completion of treatment.8,10

The association between the use of alkylating agents and the development of hypothyroidism was not statistically significant in our study (P=.528). Few studies support this association. One of them, a report of the Childhood Cancer Survivor Study, attributed 9.4% of the total cases of hypothyroidism to the use of bleomycin and alkylating agents (cyclophosphamide and lomustine).10 However, there is controversy in regard to the potential causative role of CTX and, for the time being, there are little data that support it.5,13

Of the 5 nodules detected in our case series, 4 were diagnosed as papillary thyroid cancer after the fine-needle aspiration and biopsy. As occurred in relation to hypothyroidism, RTX was a risk factor for the development of thyroid neoplasms, benign or malignant. In paediatric patient it is even more damaging, as children aged 5–10 years are more susceptible to the noxious effects of radiation.14–16 Thyroid cancer, especially the papillary form, is one of the most frequently diagnosed second neoplasms in survivors of childhood cancer.17 Broadly speaking, the time elapsed from the diagnosis of the primary cancer and the development of secondary neoplasms varies widely in the literature,18 as they can develop within 4 years of treatment completion6 or even 20 years after.17 And while the increase in risk declines gradually over time, the risk never becomes as low as the risk in the general population.19

In our case series we found a statistically significant association between the use of RTX and the development of thyroid nodules (P= .01). It has been described that any increment in the dose of radiation increases the risk of second thyroid cancer,8 although there seems to be a downturn in the dose-response relationship from approximately 20Gy. This suggests that the greater the radiation dose the thyroid parenchyma is exposed to, the greater the direct cell-killing effect and therefore the fewer viable thyrocytes that remain to undergo malignant transformation.8,20,21 However, in our study, the median RTX dose in the group of patients who developed thyroid cancer was 29.8Gy.

When it came to age and sex in relation to the development of thyroid neoplasms, we did not find any significant differences. However, a study conducted in HL survivors with second thyroid cancer found that patients irradiated at younger ages and female patients were at increased risk.22 The multivariate analysis of a more recent study found that age less than 20 years (P= .032) and female sex (P= .016) were associated with a greater probability of second cancer.11 The disagreement of our results with the overall trends reported in the literature is very likely due to the limitations imposed by the small sample size and the short duration of follow-up.

The role of alkylating agents in the development of neoplasms, as occurred with hypothyroidism, is not well established. A report from the Childhood Cancer Survivor Study shows that 119 cases of thyroid cancer were identified in a cohort of 12 547 patients with a median duration of follow-up of 5 years. There was a statistically significant association between the detection of thyroid cancer and combined treatment with RTX (at a dose < 20Gy) and alkylating agents (relative risk [RR], 4.7; 95% CI, 1.5–15; P=.04). 23 Of the 5 patients with thyroid nodules in our case series, all received RTX and 3 did not receive chemotherapy with alkylating agents. Given the small size of the groups in the sample (5 patients with nodules and 8 with thyroid abnormalities of any kind), we were not able to assess causality in the association between the development of thyroid cancer and the use of alkylating agents.

Due to the small number of patients with papillary thyroid cancer in the case series, we could not assess the association between this disease and laboratory abnormalities.

Only one case of hyperthyroidism (subclinical) was detected, corresponding to a frequency of 1.7% (clinical hyperthyroidism+subclinical hyperthyroidism) in our patients. Of all the possible thyroid complications, hyperthyroidism is the least frequent5 and is usually subclinical. In addition, it may be a transient complication due to a certain level of thyrotoxicosis resulting from the direct destruction of parenchyma by RTX that eventually resolves (euthyroidism).8 For this reason, assessment of thyroid function is recommended, at least a few months after RTX has been completed to avoid biased results due to the thyroiditis susceptible period that follows irradiation.11

When it came to the development of thyroid abnormalities overall, we found a strong association between the use of RTX and these abnormalities (P= .003). The multivariate analysis also found a significant association (P= .028), and, while the 95% CI was broad (1.279–82.056) due to the dispersion of the few data that we had available, the identified association suggested consistency. In addition to demonstrating this association based on the presence or absence of a history of RTX, we also estimated that for each additional Gy of radiation, the risk of developing thyroid disease increased by 8% (P= .017; HR, 1.081; 95% CI, 1.014–1.152). These findings were consistent with those of other authors such as Steven et al., who calculated an increase of 1.06 in the relative risk of hypothyroidism with each additional Gy administered (P= .000001).

Given all that has been discussed this far, we ought to highlight the importance of close monitoring and follow-up in these patients. The number of HL survivors continues to grow, and the broad variety of sequelae and involvement at different levels require strategies with a multidisciplinary approach.24 The findings of our study and the existing literature suggest that the risk of thyroid abnormalities in survivors of childhood HL increases consistently with time. The cumulative incidence of thyroid cancer continues to grow even 30 years after the diagnosis of the primary cancer,25 and, while these radiation-induced second cancers do not appear to be more aggressive than sporadic cancers, it is important to provide adequate screening and follow-up.26 Endocrinology societies propose routine ultrasound screening in patients with a history of RTX 5, while others argue that palpation and the physical examination are sufficient and reduce the associated costs and the potential psychological distress that may be associated with a diagnostic procedure.8 Close long-term follow-up of HL survivors is necessary to detect life-threatening thyroid changes in a timely manner.

The main limitations of the study are its small sample size and its retrospective design. The duration of follow-up, short in comparison with other studies, is another limitation.