Thyroid function after diagnostic 123I-metaiodobenzylguanidine in children with neuroblastic tumors

Background Metaiodobenzylguanidine (MIBG) labeled with radioisotopes can be used for diagnostics 123I−) and treatment (131I−) in patients with neuroblastic tumors. Thyroid dysfunction has been reported in 52% of neuroblastoma (NBL) survivors after 131I-MIBG, despite thyroid protection. Diagnostic 123I-MIBG is not considered to be hazardous for thyroid function; however, this has never been investigated. Therefore, the aim of this study was to evaluate the prevalence of thyroid dysfunction in survivors of a neuroblastic tumor who received diagnostic 123I-MIBG only. Methods Thyroid function and uptake of 123I− in the thyroid gland after 123I-MIBG administrations were evaluated in 48 neuroblastic tumor survivors who had not been treated with 131I-MIBG. All patients had received thyroid prophylaxis consisting of potassium iodide or a combination of potassium iodide, thiamazole and thyroxine during exposure to 123I-MIBG. Results After a median follow-up of 6.6 years, thyroid function was normal in 46 of 48 survivors (95.8%). Two survivors [prevalence 4.2% (95% CI 1.2–14.0)] had mild thyroid dysfunction. In 29.2% of the patients and 11.1% of images 123I− uptake was visible in the thyroid. In 1 patient with thyroid dysfunction, weak uptake of 123I− was seen on 1 of 10 images. Conclusions The prevalence of thyroid dysfunction does not seem to be increased in patients with neuroblastic tumors who received 123I-MIBG combined with thyroid protection. Randomized controlled trials are required to investigate whether administration of 123I-MIBG without thyroid protection is harmful to the thyroid gland.


Introduction
Metaiodobenzylguanidine (MIBG) is a guanidine derivate and norepinephrine analogue, which is actively taken up in neuroendocrine cells via the norepinephrine transporter. Once labeled with radioisotopes it can be used for diagnostics (imaging, ( 123 I − ) or treatment ( 131 I − ) purposes in patients with neuroblastic tumors [1,2].
Thyroid damage has been previously reported in neuroblastoma (NBL) patients who have received therapeutic 131 I-MIBG. [3][4][5][6][7] This thyroid damage may present as mild thyroid dysfunction (TSH elevation) but also thyroid nodules and 3 cases with thyroid cancer have been described. This thyroidal damage most probably reflects follicular cell damage due to thyroid uptake of free circulating 131 I − . Thyroid protection during exposure to 131 I-MIBG has been proposed by administering potassium iodide (KI) (dilution and lowering uptake of the circulating 131 I − ). Due to the fact that KI does not protect the thyroid gland sufficiently [3], thiamazole (blocking the binding of iodide to thyroglobulin in the thyroid follicular cell resulting in a shorter exposure time to 131 I) and thyroxine (T4) (lower uptake of 131 I − due to lowering of the serum TSH) or perchlorate (CIO4) (lowering the uptake of 131 I − by blocking the sodium-iodide transporter) can be added to KI [5][6][7]. Despite these preventive measures, thyroid dysfunction occurs frequently [3,4,7,8], which raises the question whether the administered thyroid protection measures are sufficient or if NBL survivors are more susceptible to thyroid dysfunction regardless of given treatment. The 131 I/ 123 I-MIBG scintigraphy procedure guidelines for tumor imaging from the European Association of Nuclear Medicine (EANM) recommend the use of thyroid prophylaxis during both 131 I and 123 I-MIBG [9]. It may be questioned however whether thyroid protection during exposure to 123 I-MIBG is necessary. Due to the mainly gamma irradiation, the short half-life of 123 I − and the relative low dose that is given, the expected probability of developing thyroidal radiation damage is very low after exposure to 123 I − . To illustrate, 123 I − unbound to MIBG, is also used as diagnostic tool in children with suspected congenital thyroid disease to locate the thyroid gland and to objectify its function using the perchlorate discharge test [10]. In these children, the administration of 123 I − is not considered to be hazardous for thyroid function. On the other hand, although only a very small proportion of 123 I-MIBG will circulate as free 123 I − , children with NBL may have multiple courses of 123 I-MIBG administered (up to 20 times) in the course of treatment of NBL. As it is unknown whether these repeated and higher doses might affect the function of the thyroid gland, thyroid protection is advised during exposure to 123 I-MIBG. The aim of this retrospective cohort study was to evaluate the prevalence of thyroid dysfunction in survivors of a neuroblastic tumor who received 123 I-MIBG for diagnostic purposes only during childhood (i.e., no treatment with 131 I-MIBG) and to compare this with the prevalence of thyroid dysfunction in the general childhood population and with NBL patients who had been treated with 131 I-MIBG.

Study population
The study included all patients who were (1) diagnosed with a neuroblastic tumor (i.e., (ganglio) NBL or ganglioneuroma) in Emma Children's Hospital, Amsterdam, The Netherlands in the period 1989-2012; (2) given 123 I-MIBG for diagnostic purposes; (3) more than 1 year in follow-up after the last administration of 123 I-MIBG; (4) having stable disease or being in remission after completion of therapy at the time of follow-up; and (5) did not receive 131 I-MIBG treatment or external radiotherapy exposing the thyroid gland. Patients receiving 131 I-MIBG for diagnostic purposes as well as patients diagnosed with preexistent thyroid dysfunction requiring levothyroxine (LT4) treatment before exposure to 123 I-MIBG were excluded from this study. Follow-up time was defined as the period between the first administration of 123 I-MIBG and the last measured thyroid function. Written informed consent of patients and parents was obtained prior to screening for thyroid dysfunction for the purpose of this study. The present study was approved by the local Medical Ethical Committee of the Academic Medical Center, Amsterdam, The Netherlands. Ethical approval for extra blood sampling (to test thyroid function) was denied for children < 12 years who did not need a venipuncture on clinical grounds; five children were, therefore, excluded from the study.

Data collection
Demographic, tumor and treatment-related characteristics, as well as the frequency of 123 I-MIBG scanning and cumulative dose of 123 I-MIBG, and the prescribed type of thyroid prophylaxis were extracted from medical records. Disease stage was classified according to the International Neuroblastoma Staging System (INSS). Pre-and post-diagnostic TSH and free T4 (FT4) values (when available) were collected from medical records. Any documented symptoms of hypothyroidism and/or use of LT4 replacement therapy were noted. In case thyroid function was not determined more than 1 year after the last 123 I-MIBG scintigraphy, patients were actively screened for thyroid dysfunction by measurement of plasma TSH and FT4 at the time of their regular outpatient visit. Plasma levels of FT4 and TSH were measured using standard commercial immunoassays. TSH elevation (TE) was defined as a plasma TSH concentration above the institutional age-related reference ranges independent of the serum FT4 level, or if patients used LT4 replacement therapy at the last moment of follow-up.

Thyroid prophylaxis
To protect the thyroid gland from radiation during exposure to 123 I-MIBG, thyroid prophylaxis was administered according to previously described protocols [5,6]. In short, patients received 100 mg KI for 3 days during use of diagnostic 123 I-MIBG in the time period 1989-1999. From 1999 until present, thyroid protection consisted daily of T4 100-125 μg/m 2 in one dose and thiamazole 0.5 mg/kg body mass given in 2 doses for a period of 3 days, and KI 90 mg (100 mg per mL) was started on the morning of 123 I-MIBG administration for a period of 2 days.

Scan review
Radionuclide imaging in all patients was performed 24 h after injection of 123 I-MIBG. Thyroid uptake of 123 I − was evaluated by two observers (pediatric endocrinologist and nuclear physician), who were blinded for clinical data, using a four-grade scoring system ("0" = thyroid gland not visible, "1" = faint visibility of the thyroid gland, "2" = clear visibility of the thyroid gland, "3" = not possible to assess the thyroid region appropriately or missing scan).

Statistical analysis
Data are presented as median (range) for continues data, or n (proportion in %) for categorical variables. The prevalence of TE among survivors with (ganglio-)NBL (n = 39) was compared to the results of thyroid function measurements from two historical cohorts of NBL patients (n = 40) who had received diagnostic 123 I-MIBG as well as 131 I-MIBG treatment [3,4]. Inter-group differences were evaluated using a Chi-square test, Fisher's exact test or Mann-Whitney U test, depending on the type of variable. Using multivariable logistic regression analysis, we investigated the following risk factors for the occurrence of TE in the total group of NBL survivors: number of 131 I-MIBG therapy (y/n), number of diagnostic 123 I-MIBG administrations and chemotherapy (y/n). A p-value of < 0.05 was chosen to indicate statistical significance. Data analyses were performed with SPSS statistical software version 28.0 (SPSS, Chicago, IL, USA).

Study population
In the period 1989-2012, 60 survivors with neuroblastic tumors were identified who received 123 I-MIBG for diagnostic purposes and met all inclusion criteria. Of these, seven were lost to follow-up/followed at another institution and for 5 survivors no ethical approval was obtained for blood sampling. Forty-eight survivors were included for analysis of thyroid function. Demographic and clinical data of all included survivors are shown in Table 1.

Visibility of the thyroid gland on the 123 I-MIBG scans
In total 144 123 I-MIBG scans were evaluated for scintigraphic visibility of the thyroid gland. No thyroidal uptake of 123 I − ("0") was seen on 109 (76%) scans, weak uptake ("1") on 12 (8.3%) scans, and clear uptake ("2") on 4 (2.8%) scans. On 19 (13.2%) scans, it was not possible to classify thyroidal uptake of 123 I − because of interference due to metastases or missing images. In total, on 11.1% of images in 29.2% of the survivors uptake of 123 I − was visible in the thyroid gland, despite the prescribed thyroid prophylaxis. There was no difference in the frequency of thyroidal uptake with regards to the type of prescribed thyroid prophylaxis. Of the two patients with TE, in one of the 11 images weak uptake of 123 I − was seen.  Table 2 shows detailed information on the difference in patient demographics and clinical characteristics between a previous cohort of NBL patients treated with 131 I-MIBG therapy (n = 40) and the (ganglio-)NBL survivors of the current cohort with only 123 I-MIBG as diagnostics (n = 39). The prevalence of TE was significantly different between the patient groups (42.5% vs. 2.6% respectively, (p = < 0.01)). The patients who received 131 I-MIBG therapy were in general more often diagnosed with high-risk NBL (i.e., higher stages NBL and/or presence of MYCN gene amplification/chromosome 1-p-deletion), required more intensive therapy (i.e., (high-dose) chemotherapy) and had received 123 I-MIBG more frequently. Multivariable logistic regression analysis identified a significant association between 131 I-MIBG therapy [odds ratio 33.4 (95% confidence interval: 3.8-294.5)] and the occurrence of TE ( Table 3). The number of 123 I-MIBG and chemotherapy administrations was not predictive for the development of TE.

Discussion
To the best of our knowledge, we are the first to evaluate the prevalence of thyroid dysfunction in neuroblastic tumor patients who had only been exposed to 123 I-MIBG. Our results demonstrate that in this population given thyroid protection during exposure to 123 I − , the prevalence of TE is very low. It may be questioned whether thyroid protection is necessary during 123 I − exposure.
Patients surviving NBL are frequently reported to have an increased risk to develop thyroid damage [13,14]. This is most probably due to radiation exposure (i.e., radiotherapy exposing the thyroid gland and/or the administration of 131 I-MIBG therapy) at a young age. Due to the fact that in our previous studies on 131 I-MIBG, no correlation was found between the occurrence of TE and uptake of 131 I − in the thyroid gland, to the number of 131 I-MIBG treatments, the received total dose, or to the age at time of diagnosis, other causes for the high percentage of TE in these survivors were considered. For this reason, it was questioned whether children with NBL have more thyroid problems irrespective of given treatment [13][14][15]. The data here presented do not support the idea that NBL survivors are at increased risk to develop thyroid dysfunction independent of 131 I-MIBG treatment. We found biochemical evidence of mild TE in our study in only two patients (4.2%). When compared to the prevalence of TE or "subclinical hypothyroidism" in the general childhood population (1.7-9.5%), this seems not to be increased [11,12]. Whether the mild thyroid dysfunction in these two survivors should be attributed to the administration of 123 I-MIBG may be questioned. The number of patients diagnosed with TE in our study, was too small to allow for a detailed analysis regarding the independent effects of number of 123 I-MIBG administrations, total received 123 I − dose, visible thyroidal 123 I − uptake or chemotherapy.
When our cohort was compared to a historical cohort of NBL patients treated with 131 I-MIBG [3, 4] a significant increased prevalence of thyroid dysfunction was found for survivors after 131 I-MIBG therapy (42.5%), indicating that 131 I-MIBG therapy should be considered as the most likely causative factor. However, some confounders need to be considered.
First, more children in the 131 I-MIBG therapy group received (high-dose) chemotherapy compared with the non-131 I-MIBG therapy group. To date, the role of chemotherapy in causing TE remains unclear. Published data are conflicting; some claim chemotherapy to induce thyroid damage [16], others deny this role [17,18]. Chemotherapy may be   an additional hazard in patients who are also treated with combination radiation therapy [19]. In explorative multivariable logistic regression analysis, we did not find evidence an increased risk for treatment with chemotherapy on TE; however, the number of cases was too small to allow for strong conclusions. Second, the follow-up time for survivors treated with 131 I-MIBG was significantly longer when compared to the non-131 I-MIBG group (11.5 vs. 7.5 years respectively), which may explain the relative lower incidence of TE in the 123 I-MIBG group. However, the fact that thyroid dysfunction usually develops within the first 5 years after 131 I-MIBG administration disproves this argument [3,4].
Third, the fact that patients selected for 131 I-MIBG therapy were more frequently diagnosed with high-risk NBL compared to patients who did not receive 131 I-MIBG therapy may suggest that these patients have a different genetic profile, resulting in more advanced-stage NBL and an increased susceptibility for the development of thyroid damage, irrespective of given treatment.
The results of this study, in combination with the notion that 123 I-MIBG has a lower irradiation risk compared to 131 I-MIBG, strengthens the question whether administration of 123 I-MIBG without thyroid protection is damaging to the thyroid gland. In accordance with the 2010 EANM Guideline, most centers administer thyroid prophylaxis for the use of diagnostic 123 I-MIBG [9]. In 2014, Friedman et al. reported that thyroid uptake and scans for patients who received 123 I-MIBG for cardiac imaging did not differ between patients who received thyroid prophylaxis and those who did not [20]. The authors hypothesize that the accumulation of MIBG tracer in the thyroid sympathetic nerves, rather than in the thyroid itself, resulted in the total uptake seen in the patients of whom the thyroid gland was blocked. These results were confirmed by Giubbini et al. [21] In this study, heart-failure and Parkinson patients underwent cardiac 123 I-MIBG with or without thyroid blockade pre-treatment. Interestingly, there was no difference in thyroid parameters (thyroid/mediastinum ratio and washout) between patients who did and did not receive thyroid blockade. The authors stated that thyroid prophylaxis may not be justified in patients undergoing 123 I-MIBG imaging as the thyroid uptake is likely a reflection of sympathetic neuronal activity. The risk of adverse effects due to pre-treatment with KI (i.e., leukopenia and the risk for iodine allergy) could be higher than the risk of exposing the thyroid gland to unnecessary radiation. Collectively, we call for long-term randomized trials that detail the efficacy and safety of thyroid prophylaxis during 123 I-MIBG administration on thyroid function. These studies should be particularly performed in pediatric NBL patients, to answer the question whether repeated and high doses of 123 I-MIBG during childhood are damaging to the thyroid gland.
In summary, our study confirms our hypothesis that thyroid dysfunction is not prevalent in NBL survivors after exposure to 123 I-MIBG only. It may be questioned whether thyroid protection is necessary at all. Future studies are required to investigate whether protection of the thyroid gland is necessary during administration of diagnostic 123 I-MIBG.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.