Serum anti-Müllerian hormone is lower in patients with multiple radioiodine dose for treatment of pediatric thyroid cancer

Introduction

The increasing prevalence and excellent long-term survival rates of differentiated thyroid cancer (DTC) have generated worldwide interest in the reproductive health and risk of infertility among adolescents and young women exposed to treatment with iodine-131 (radioiodine (RAI)). However, controversy exists regarding the gonadotoxic effects of RAI in DTC survivors, which may include menstrual irregularity, miscarriage, premature ovarian failure, impaired fertility, and genetic damage causing congenital malformations (1, 2, 3, 4). Although transient oligospermia and decreased ovarian function have been reported in adults after RAI, infertility appears to be rare in these patients, except after high RAI doses (5). Temporary amenorrhea or oligomenorrhea lasting 4–10 months occurs in 20–27% of menstruating women after RAI therapy for thyroid cancer. Although the studies on this topic have included small sample sizes, the long-term rates of infertility, miscarriage, and fetal malformation do not appear to be increased in women after RAI therapy (6). Only a few studies have assessed the fertility and ovarian reserve of DTC survivors treated during childhood or adolescence (7, 8, 9, 10, 11), and little is known about the impact of RAI considering the different pubertal stages and ovary maturity.

The assessment of the ovarian follicular reserve is a crucial aspect of fertility evaluation in women. A reduced ovarian reserve may decrease the reproductive life span and, in some women, progress to premature ovarian insufficiency. Reduced ovarian reserve may present as a decreased reproductive capacity or reduced oocyte quality and quantity (12).

The anti-Müllerian hormone (AMH) measured in serum is a marker of ovarian reserve in the female sex (13, 14) and correlates directly with the number of growing follicles that decrease with aging (15). Previous studies have shown that AMH may decrease after high RAI activity in premenopausal women (16, 17). However, the risks of damage to the reproductive system and depletion of ovarian reserve caused by RAI in DTC survivors treated during childhood have not been well characterized and require further studies.

Based on these considerations, the aim of this study was to analyze the association between RAI treatment strategies and the extent of reproductive impairment and depletion of ovarian follicular reserve in women treated with RAI for DTC during childhood.

Materials and methods

Results

In total, 114 women were eligible for the study. Of these, 23 were excluded due to age >19 years at RAI treatment and 7 due to a history of other neoplasms. Of the 84 eligible patients, 47 had serum AMH levels measured between May 2022 and March 2023. In total, 26 patients could not be contacted due to unavailability of contact information (incomplete phone number and address) in their medical records, and 11 refused to participate due to the distance from their homes to the center where blood was collected.

Table 1 shows the general characteristics of the 47 participants with AMH measurements. Among these participants, the absolute AMH values were similar in those treated for the first time before vs after menarche, regardless of the age at treatment. Similarly, no significant differences between these two groups were observed regarding age at menarche (P = 0.67), use of contraceptive methods (P = 0.5), or cumulative RAI activity above 100 mCi. All 17 pregnancies were spontaneous and the percentages of patients with PCOS were not statistically different among the two groups( multiple or single treatment and pregnant or nonpregnant) However, when we restricted the analysis to patients who received an RAI cumulative activity >200 mCi, we found a significant difference in serum AMH levels between those who received a cumulative activity above vs below 200 mCi (P = 0.02) and between those who received single vs multiple doses (P = 0.012) (Figs. 1 and 2). Of note, these seemed to be independent factors for altered AMH levels in our cohort.

View Full Size

Serum levels of anti-Müllerian hormone related to contraceptive use, menarche before or after radioiodine, age at menarche.

Citation: European Thyroid Journal 13, 2; 10.1530/ETJ-23-0252

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Serum levels of anti-Müllerian hormone associated with multiple doses of radioiodine and cumulative activity of radioiodine above or below 200 mCi.

Citation: European Thyroid Journal 13, 2; 10.1530/ETJ-23-0252

• Download Figure
• Download figure as PowerPoint slide

The cohort was subsequently divided into two groups according to the number of RAI treatments (single vs multiple), regardless of the cumulative RAI activity (Table 2). The group that underwent multiple treatments compared with the one that underwent a single treatment had a longer interval between the last RAI dose and AMH measurement, AMH levels more frequently out of the reference range adjusted for age, lower median absolute AMH levels, and a higher median cumulative RAI activity, but the number of patients who became pregnant at least once was not different between these groups. Of note, all patients who underwent multiple RAI doses received a cumulative RAI activity >200 mCi.

In an analysis restricted to the participants with serum AMH levels out of the reference range for age, we observed a significant decrease in ovarian reserve after multiple vs single RAI treatment (P = 0.002) and at cumulative activity > 200 mCi vs below this level, but no difference in median cumulative activity between these groups (Table 3).

In our cohort there was one case of miscarriage, 1 case of early menopause but no reports of birth defects or premature delivery.

Discussion

The results of our study show a decreased level of the ovarian reserve marker AMH among women with a history of DTC who received RAI treatment during childhood or adolescence in different treatment regiments and also cumulative activity. Patients who received multiple RAI treatments (vs those who received a single RAI dose) and, in particular, those who received a cumulative RAI activity above 200 mCi (vs those who received less than that) were more likely to have lower serum AMH levels, independent of the age in which RAI was administered or pubertal status at treatment. Within the groups, AMH was more frequently out of the reference age when adjusted for age. Interestingly, the median time between the last RAI treatment and AMH measurement was more than 10 years, and AMH level appears to have remained lower even after this long period of time. The cumulative RAI activity was not different between the groups with AMH levels inside or outside the reference range.

Other authors have also shown an impact of RAI on AMH levels, particularly in young adults. Yaish et al. have shown that RAI could exert a direct effect on the ovarian reserve, assessed by AMH levels, and reported that patients treated with higher RAI have a significant decrease in AMH concentration. The same has been shown by Evranos et al., who also demonstrated that AMH level decreases shortly after RAI treatment and plateaus at a level lower than the one measured before RAI (17). This would explain our finding of lower AMH levels and higher proportions of AMH levels out of the reference range when more than 10 years had passed since RAI treatment. As shown by van Velsen et al. (22), a single RAI treatment significantly reduced the AMH concentration during the first year and remained stable thereafter, although further decline was seen in patients receiving multiple RAI treatments. A meta-analysis published recently by Piek et al. (23) showed a significant decrease in AMH levels after RAI therapy. Our study confirms and expands these previous findings of low serum AMH levels in DTC survivors treated with multiple RAI doses during childhood and adolescence. Our evaluation of AMH levels using absolute values in addition to values adjusted by age in a pediatric group was crucial to confirm this effect of RAI, as it is known that AMH typically decreases with age in young women (15).

Studies in patients treated with RAI before the age of 20 years (7, 8, 9) have shown no impact of RAI on fertility, miscarriage, prematurity, or major congenital anomalies. However, these studies included small cohorts and did not evaluate the ovarian reserve. In a Dutch study similar to ours, Nies et al. found that female DTC survivors who received RAI treatment during childhood did not appear to have major abnormalities in reproductive characteristics or predictors of ovarian failure (10). The authors observed no differences in AMH levels between women who were DTC survivors and a control group of 420 women not treated for cancer and retrieved from a database. In our study, most AMH levels were not abnormal, except in participants who underwent multiple RAI treatments, which was not analyzed separately in the study by Nies et al. In contrast to the Dutch study, the morbidity level of the patients in the present study seems to have been more advanced, as 45% of all patients had distant metastases, as opposed to only 10.7% in that study. This could explain the apparent higher cumulative RAI dose administered to the patients in the present study (i.e. a mean of 250 mCi, increasing up to 1150 mCi in patients who received repeated treatments). In the Dutch study, the median RAI dose was 200 mCi (the 75th percentile was 350 mCi). Indeed, this could explain the more concerning results found in the present study. Therefore, the finding that AMH level is lower in women who received multiple treatments and a cumulative dose above 200 mCi, not found in the less aggressive cases studied by Nies et al. (10), could be due to the more aggressive disease in our cohort and a larger number of children undergoing higher cumulative activities and number of treatments.

Aligned with previous studies (10, 16, 17), we found no relationship between the cumulative RAI dose and AMH level, despite the decrease in serum AMH levels observed after treatment. However, we found that the decrease in AMH levels was more evident above the dose of 200 mCi. Unlike the study by Nies et al., we found that multiple RAI doses influenced AMH measurements.

Issues concerning the effect of treatment regimens (single vs multiple doses) and cumulative activity on safety and efficacy remain controversial. Some cases including very young children have shown that for patients treated with ‘nontoxic’ cumulative RAI activity below 2 Gy, the administration of a single RAI dose was safe, but multiple RAI doses were associated with sequential increases in the overall frequency of cells carrying chromosome aberrations; this occurred mainly due to the induction of unstable chromosome aberrations, while the proportion of cells with stable chromosome aberrations remained mostly unchanged (24).

Aligned with most literature on this topic, we found no direct influence of RAI on pregnancy. Like the others, our study had a retrospective design and collected data on the occurrence or not of pregnancy. Also, a previous study comparing DTC adult patients treated with surgery and surgery followed by RAI showed similar AMH levels in both groups (25).However, information regarding the time elapsed between the decision to become pregnant and the pregnancy itself, along with other factors that could show increased difficulty in conceiving, could have been missed in all studies (15, 22, 23), including ours. Furthermore, most of our patients were of childbearing age at data collection and may become pregnant in the future with more or less difficulty. Another limitation of the present study is the absence of serum AMH levels measured before RAI treatment; without this information we are unable to know the pretreatment ovarian reserve of our patients. An additional limitation was the fact that the reference range for AMH was provided by the manufacturer, when, ideally, a reference range from a control population would be preferable. However, the reference range for AMH in the present study has been used in previous studies in the Brazilian population and has been shown to correlate well with ovarian reserve (26). Furthermore, besides the relatively small sample (only 47 patients), DTC in children is a rare disease, and our follow-up of a median of more than 11 years strengthens our results. They were all followed up at the same center, had a particularly aggressive disease (as our incidence of metastatic disease is higher compared with other studies), and were followed up by the same medical team, and we were very careful to select patients treated with RAI under 19 years of age, which makes our study different from others that included children and young adults mixed together. It is also worth mentioning that potential confounders of AMH such as PCOS, use of birth control pills, and BMI were not different among the groups. However, PCOS, in particular, is associated with higher levels of AMH and the number of PCOS patients was similar in both groups.

Conclusion

The results of the present study indicate that serum AMH levels are lower in women with a history of DTC who received multiple RAI doses during childhood compared with those who received a single RAI dose, particularly in patients who received cumulative doses above 200 mCi compared with those who received a lower dose than that. This finding should be considered during decisions about RAI activity and regimen and may favor individual dosimetry in patients with metastatic disease in order to accomplish treatment in a single dose. Prospective designed studies could be useful to access more specific data regarding difficulties and time until pregnancy along with pregnancy rates and also how slightly low levels of AMH can impact on pregnancy rates.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.

Funding

This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.