Radioactive Iodine and Female Fertility, Less Innocent Than we Thought

Radioactive iodine (I131) is used after surgery in the treatment of Differentiated Thyroid Carcinoma (DTC). There is no solid evidence about the potential deleterious effect of I131 on women fertility. The objective of this study is to assess the impact that I131 may have on fertility in women. All women followed by DTC in our department have been analyzed and women younger than 45 years old at the time of diagnosis and initial treatment were included. There were 40 women exposed to I131 (study group) and 11 women who were only treated with thyroidectomy (control group). Of the women exposed to I131, 40% went through early menopause, while no cases were reported among their controls. Furthermore, 29.2% of women exposed to I131 had decreased Antimüllerian Hormone (AMH), compared to the only 11% of unexposed women (not signicant). Regarding the fertility impairment "perceived" by patients, in the group of women exposed to iodine, 17.9% described being unable to complete their genesic desire whereas, none was registered in the control group. We conclude that radioactive iodine can affect a woman's fertility and shorten her reproductive life, so this is an aspect that should be taken into consideration.


Introduction
Thyroid cancer is the most frequent endocrine malignancy. Differentiated tumours derived from the follicular epithelium, papillary (85%) and follicular carcinoma (15%), account for around 90% of thyroid tumours and fortunately, are the ones with the best prognosis. Papillary carcinoma has two age peaks of presentation: between the second and third decade of life, followed by a second later peak. Follicular carcinoma usually appears in older individuals. In this type of cancer, the female/male relation reaches a ratio of approximately 3.5/1 1 . Given its good prognosis, it is expected to have a good quality of life after the treatment and achieve a life expectancy similar to that of the rest of the population; that is why the possible effects derived from the treatment of the disease, including those related to future reproductive health, must be assessed.
Radioiodine treatment in low-risk patients is currently a matter of discussion. In these patients, the prognosis is excellent, mortality is low and the controversy resides in whether iodine really provides bene ts in terms of recurrence [2][3] . The treatment with radioactive iodine is not innocuous since the uptake is not only at thyroid level and during its elimination the radiation is discouraged towards other tissues. The most frequent complication is sialoadenitis due to iodine uptake by the salivary glands.
Transitory loss of avours perception, buccal dryness and dental caries, nasolacrimal duct stenosis and even pulmonary brosis and radiation neumonitis could also appear. Complications, despite being rare, usually appear in patients with a high treatment load. A statistically signi cant increase in second malignancies has been reported closely related to the administered dose. The secondary most related neoplasias are leukaemia, salivary gland, breast, digestive tract, and bone cancer. Despite this, the overall risk is low 4 . Another adverse effect is the impairment of fertility 5-6 .
In men, I 131 therapy is associated with a temporal increase in FSH and Inhibin B decrease, being related serum FSH with the received dose. The alteration of these values shows the damage to the germinal epithelium that results in azoospermia and oligospermia. On the other hand, damage to Leydig cells manifests itself as an elevation of LH levels and testosterone lowering. The effect on the male gonad responds to a dose-dependent effect. While low doses of iodine do not have negative consequences for reproduction, only transient alteration of hormone levels and recovery is the rule, cumulative doses greater than 500-800 mCi raise the risk of permanent increases in FSH, which could lead to infertility [7][8] .
In women, studies show short and long-term consequences for fertility due to the iodine treatment. I 131 is associated in the short term with menstrual disorders that occur in the rst year after iodine administration. Around 20-30% of women go through a period of amenorrhea or metrorrhagia that occurs with at least one menstrual cycle of latency and whose duration would not exceed a year [9][10] . Later studies nd similar results and emphasize that the temporary absence of menses would occur more frequently in women with older ages [11][12] . Another fact to support the damage of iodine to female fertility is the nding of the age of menopause advancement in women treated with I 131 13 , which would imply a decrease on their fertile period. In 2015, the study by James Wu et al. uses a retrospective cohort of 18,000 women to demonstrate that patients with differentiated thyroid carcinoma who are treated with I 131 have a decreased birth rate compared to patients who are not administered this therapy, and that this phenomenon occurs, especially, in women who are treated at later ages 14 . This study leads us to believe that the decrease in births could be due to the harm of I 131 to the ovarian reserve. To approximate the ovarian reserve, the most accurate and sensitive parameter due to its low inter and intracycle variability is the determination of the Antimüllerian Hormone. AMH is produced in the granulose cells of growing ovarian follicles, especially in preantral and small antral follicles, preventing the development of other smaller follicles. AMH levels show a progressive decrease throughout the reproductive life as the follicular reserve declines, nally becoming undetectable shortly before menopause [15][16] . Despite having established itself as a relevant marker, there is little data in the literature on normal AMH values in healthy population, compared to women who attend fertility clinics. One of them is the article by La Marca et al., which establishes a nomogram of AMH values with a large population of healthy women 17 . F. Acibucu et al. in 2016 analyzed the potential effect of iodine on fertility, using the antimüllerian hormone for the rst time as indicator of the ovarian reserve. The study found signi cantly lower mean AMH values in women who were exposed to iodine 18 . Giusti et al. in 2018 also used the AMH to assess the effect of iodine in a population of 34 women treated with I 131 versus 23 women without exposure with an average follow-up period of 7 years. Although lower AMH values were found in exposed women, the results do not obtain statistical signi cance, probably due to the sample size 19 . In 2018 I Yaish et al. demonstrated how a single dose of radioactive iodine has an effect, on ovarian reserve, estimated through AMH before and after exposure over a year. The results describe in patients treated with higher doses of I 131 a signi cant decrease in the concentration of AMH, with the nadir occurring at 3 months, when the values are almost 50% lower than the prior levels to treatment. Subsequently, an incomplete recovery is seen towards the end of the rst year that stagnates at around 9 months, at which time the AMH values are 32% lower than the initial ones 20 . This nding is con rmed in other studies such as the one published by B Evranos in the same year 21 , or the most recent of EFS van Velsen published in 2020 22 . However, all these recent works are limited to the study of ovarian reserve after receiving radioiodine treatment, but do not evaluate the long-term consequences on the impact on the duration of the reproductive life years and on the achievement of the objectives of desire for offspring.
The aim of this study is to assess the impact that I 131 can have on fertility in women in our patients' population with differentiated thyroid carcinoma. For this purpose, we study the age of menopause if it has already taken place at the time of the study or the Antimüllerian hormone if women still have their periods and assess the achievement of their reproductive desire.

Study population
From the entire population of women monitored for Differentiated Thyroid Carcinoma by the Health Department of Alcoy, just women with ages less or equal to 45 years at the time of initial diagnosis and treatment are included in the study. Patients are classi ed based on the administration of radioactive iodine as part of treatment in addition to surgery. Women treated with surgery and I 131 are the cases and women treated only with surgery without exposure to iodine are their controls.
Currently 87 women are being monitored by the Endocrinology Department for having suffered from DTC.
Thirty-four women were excluded from the study because they were diagnosed being older than 45 years of age and 2 more women because they were already menopausal at the time of diagnosis.
Fifty-one women with history of DTC aged less or equal to 45 years at the time of initial diagnosis and treatment were included in the study. None of the 51 women included had a history of previous pelvic radiotherapy or surgery.

Study protocol
Fertility is studied by assessing the age of menopause onset if it has already occurred, the Antimüllerian Hormone (AMH) in menstruating women at the time of the study, and analysing the obstetric history and the reproductive desire of patients.
Through the medical history and the follow-up, data such as characteristics of the menstrual cycles, age of menopause, pregnancies, miscarriages, use of assisted reproductive techniques, and the appearance of second malignancies are collected.
The usual analysis of the patients included the determination of AMH if they still had menses and FSH and estrogens in women without menstrual cycles.
A review of available reports about pathological anatomy of surgical pieces allowed us to describe the histological variant and its characteristics. Likewise, a genetic study of BRAF mutations was performed in papillary carcinomas and KRAS in follicular carcinomas for which we had a sample of su cient size.
Valuation criteria Spontaneous menopause. Women who have had a hysterectomy during follow-up have been excluded from this evaluation. It is considered early menopause when it appears at an age below 45 years old and it has a compatible hormonal analysis.
Decreased ovarian reserve. Determination of AMH in women who have not yet undergone spontaneous menopause, although they have undergone a hysterectomy without associated oophorectomy and who do not have a history of polycystic ovary syndrome. Patients have been considered to have a compromised ovarian reserve when their AMH value is below the 5th percentile of the AMH nomogram published by La Marca et al. 17 .
Reproductive desire ful lment. It is evaluated from data of the obstetric history and the own evaluation of each woman about the ful lment of her reproductive desire. The need for the use of Assisted Reproduction Techniques (ART) has also been assessed.

Statistical analysis
For continuous variables, mean, standard deviation, median and distribution by quartiles are used.
Normality tests are performed using Kolmogorov-Smirnov test on samples with more than 50 records and Shapiro-Wilk test on smaller samples. For means comparison on continuous variables, Student's T test was used in those samples with normal distribution and Mann-Whitney's U test was used in variables that did not meet normality criteria. For categorical variables, percentages are used. In the frequency comparison, the exact Fisher statistic is used since it is not possible with all parameters under study to collect a minimum of 5 expected frequencies in all cells. Odds Ratio (OR) calculation is performed to determine the risk supposed by exposure of radioactive iodine to produce a decreased ovarian reserve, early menopausal age and impaired fertility. To assess the relationship between variables, Spearman correlation coe cient is used, since the studied variables do not follow a normal distribution. Since menopause is a unique and irreversible event in a woman's life, the age of menopause is also studied using a Kaplan-Meier survival analysis.

Ethical considerations and Informed consent
All procedures performed were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. All subjects gave written consent to participate.

General characteristics of the study population
The study population is made up of 51 patients with a mean follow-up time of 12 As we see, women who have received iodine treatment are signi cantly younger at the time of initial diagnosis and treatment. As expected, they have larger tumours and are at higher risk according to the ATA (American Thyroid Association) classi cation.
Inside the group of women who were only treated with surgery, all had histological diagnosis of papillary thyroid carcinoma and all of them were classi ed as low-risk patients. There were microcarcinomas in 72.7% of the cases. In the group of I 131 exposed patients, 57.5% are considered of low risk, 37.5% intermediate risk and 5% high risk, one of the cases due to the presence of lung metastases and the other due to extensive local extension. There are 37 papillary carcinomas (92.5%) and 3 follicular carcinomas (7.5%). Regarding the size of the tumors, only 4 (11.1%) are microcarcinomas.
The study of genetic mutations reveals a positivity of 50-55% for the BRAF mutation in papillary carcinomas and 33.3% in follicular carcinomas.
When we study the dose used to treat DTC, almost 70% of women receive high doses of iodine (> 100 mCi). Only 15% of patients received less than 50 mCi.

Menopause age
Women undergoing hysterectomy are excluded from the analysis (four in the group exposed to radioactive iodine and one in the control group).
Eleven women (23.9%) showed menopause. The frequency of menopausal cases distributed according to age is shown in Table 2.  Table 2. Cases of women with menopause classi ed by age of onset before or after 45 years.
When comparing the frequency of spontaneous menopause between both groups, there are no signi cant differences between the women exposed to iodine and their controls. Similarly, given the scarcity of cases, we cannot speak of statistically signi cant differences in the appearance of early menopause in our series of patients. However, there seems to be a trend towards early menopause in women exposed to I 131 (Figure 1).

AMH alteration
When we compare mean AMH level in the radioiodine group and the control group, regardless of the age, there are no signi cant differences (Table 3).  Once deeply studied the AMH level behaviour, we see there is always a negative correlation between age and AMH level (Figure 2).
The relationship is statistically signi cant for both groups, exposed women (p = 0.001) and the control group (p-value = 0.002). What varies is the strength of the association, in the control group, Spearman's Rho statistic is -0.851 (strong association) and in the group of exposed women, Spearman's Rho is -0.582 (moderate association).
When comparing the individual AMH values with the proposed nomogram of La Marca et al. 17 it is obtained that there are 13 patients whose value is below the 5th percentile. Eleven of those women with low ovarian reserve belong to the iodine exposure group. Only 2 patients with decreased AMH values have not been treated with radioactive iodine ( Table 3).
The measure of the association between decreased AMH and I 131 exposure gives an Odds Ratio of 2.3158. However, it does not turn out to be statistically signi cant (CI: 0.4514 -12.9097).

Impairment of fertility
Among women in our study, 8 expressed to have failed to ful l their reproductive desire. This represents 16% of the sample. One of the women who considered their fertility to be affected belongs to the control group, while the remaining 7 women were exposed to radioactive iodine.
The woman who belongs to the control group, prior to the diagnosis of thyroid carcinoma, had already consulted for sterility, therefore we cannot implicate the carcinoma in the impairment of fertility in this case. This means that there are 7/40 women whose fertility was affected after the diagnosis, and all of them were exposed to I 131 , or what is the same, 17.5% of the women who were treated with I 131 consider to have not achieved their reproductive desire.
The difference in frequency of fertility impairment according to I 131 exposure was not statistically signi cant. The same as the association measure (OR) calculation whose con dence interval is 0.239 -19.987.

Discussion
The data presented, although not conclusive, should raise alarms among specialists who face differentiated thyroid carcinoma. This study includes a small patient sample, similar to the studies found in recent literature.
Regarding the advancement of menopausal age, our results are relevant. Firstly, it is observed that women who belong to the group of patients exposed to radioactive iodine are signi cantly younger at the time of initial diagnosis and treatment, on average 29.6 years old, while their controls are on average 35.2 years old. On the other hand, all registered cases of early menopause are in women who have received radioactive iodine treatment and, while in the general population the prevalence of early menopause reaches an approximate percentage of 10% 23 , among women who have been treated with I 131 in the study, this percentage rises to 40%. In the population of Alcoy, the age for searching the rst child in 2016 is estimated at 30 years 24 . The advancement of the age of menopause reduces even further the reproductive period of women who today plan to search for pregnancy at increasingly later ages. The study by Ceccarelli et al. 13 in 2001 shows data and graphs similar those of this work. When comparing the menopausal age in women with DTC treated with I 131 and patients with goiter, we nd a signi cantly earlier age of menopause onset in women with cancer. We agree that this phenomenon could be due to iodine-induced ovarian damage. This damage seems to contribute to the decline in ovarian function and hasten the process of follicular atresia in premenopausal women, especially those with a reduced pool of viable follicles.
Antimüllerian hormone is recognized as the most robust ovarian reserve marker available. AC de Kat et al. 25 clari es that the AMH estimates the reproductive period of women regardless of chronological age. In other words, that a woman with a low AMH level will have a lower ovarian reserve and therefore a shorter period of time until menopause than a woman of the same chronological age with higher AMH values. However, in all women, over the years the reserve decreases and with it, the antimüllerian hormone also diminishes. The speed of descent is not the same in all women, although we can establish a correlation between age and AMH value with a powerful association. In our work, we corroborate this fact by correlating the age and AMH values of our patients. The result in the case of women from control group is a negative and signi cant correlation that shows a strong association between the variables.
However, this correlation, in the case of women exposed to I 131 , continues to be signi cant but it is distorted and the association strength goes from being strong to moderate. This tells us that despite not nding differences in the mean AMH value between the exposed group and the control group, I 131 effectively leaves an indelible mark on the ovarian reserve.
The nal consequence of the menopause age advancement and the decline of the ovarian reserve is the di culty in obtaining spontaneous pregnancies and the non-ful lment of the reproductive desire. In this work, there are 17.5% women exposed to I 131 who report not having achieved their genesic desire. Among these women, there are 3 affected by early menopause, one of them started with the typical signs and symptoms of menopause immediately after treatment at the age of 28; Three women with a highly compromised ovarian reserve, who had repeated abortions before pregnancy and one of them was forced to resort to assisted reproduction techniques to achieve pregnancy; and a last woman who had a very intense metrorrhage after treatment with I 131 and underwent a hysterectomy with associated oophorectomy. It should be noted that two women with potential fertility impairment have not been included in the calculations; one who does not have or had reproductive desire and another young woman whose fertility has not been considered to be affected because she has not yet sought for pregnancy. Both women have greatly decreased AMH levels for their age, below the 5th percentile, and in case they had sought for pregnancy, they probably would have found di cult to ful l their reproductive desire and the percentage of women with a desire for fertility affected would amount up to 22.5%.
In view of the results, we think that the group of patients who should receive iodine must be very well selected, especially in low-risk tumours, and when deciding on treatment with I 131 it is important to inform about the possible consequences that I 131 might take on fertility and to perform reproductive counselling. A good option for patients who would like to have offspring in the future and have not yet ful lled their desire for reproduction could be to carry out an ovarian reserve study. Antimüllerian hormone determination in these patients is adequate to assess their follicular reserve [15][16] . In those patients with a compromised ovarian reserve, as recommended by the ASCO (American Society of Clinical Oncology), possible interventions aimed at preserving fertility, such as oocytes cryopreservation [26][27] , could be considered. Going a step further, those patients we have in follow-up after the diagnosis and treatment of a differentiated thyroid carcinoma, should be asked about their future reproductive desire, since even if they have already received the treatment with radioactive iodine, it may be convenient to advance the search for gestation or to preserve oocytes, before an irreversible deterioration can occur. Author contributions statement P.N. planned the study, carried out the study design and included the participants. P.N. and S.R. processed and analysed the data with the collaboration and supervision of PM and SO. Finally PN and SR interpreted and discussed the results and wrote the manuscript. All authors reviewed the manuscript. Kaplan-Meier graphic for menopausal women in I131 exposed group and their control.