The most common thyroid cancer is differentiated thyroid cancer (DTC) which accounts for over 95% of cases ^[1][ 1 ]. It had become the most common cancer in women by the year 2020 after breast cancer, colorectal cancer, lung and cervical cancer [GLOBOCAN2020]. Although it represents only 1% of all human malignancies, it has been found in 5% of thyroid nodules and, in turn, has a prevalence of 20-30% in the general population [ ^[2]2 ]. The histological derivation is the follicular epithelium from which it maintains histological characteristics such as the expression of sodium iodide symporter (NIS) ^[3][ 3]. Thyroid follicular epithelial cells represent the embryological derivation of DTC. Classically, it can be divided into three histological types: papillary thyroid cancer (also known as PTC, the most common), follicular thyroid cancer (FTC, the second most common) and Hürthle cell thyroid cancer. The latter was considered a particular type of FTC, but due to the significant histopathological and molecular differences it was classified as an independent tumor type in 2017 by the World Health Organization. Hürthle cell carcinoma is thus defined as a derivation of follicular cell carcinoma of the thyroid and not a variant ^[4][ 4 ].

Although the causes of thyroid cancer are unknown, many studies have evaluated risk factors. High-risk factors in the development of DTC are radiation exposure to the head and neck region, chromosomal alterations such as mutations in the RAS and BRAF genes and expression of the PAX8 / PPARγ fusion protein, and hereditary conditions such as bone marrow cancer. thyroid, syndromic and non-syndromic familial non-medullary thyroid cancer. Low risk factors are thyroid imaging with iodine 131 ( ¹³¹I), iodine deficiency, elevated levels of thyroid stimulating hormone (TSH), autoimmunity, presence of thyroid nodules and environmental pollutants, but also lifestyle, diet and elevated BMI. It is unclear whether estrogen plays a role in increasing risk: some studies have reported an increasing risk associated with exogenous estrogen, while ovarian estrogen loss lowers it. Estradiol is considered a stimulator of benign and malignant neoplasms [ 5 , 6 ^[5][6]]. Last but not least, histological type, in particular high cell cancer and columnar cell cancer, but also vascular, lymphatic and distant metastasis are also considered risk factors contributing to a poor prognosis ^[7][ 7 ].

Regarding the prognosis, the risk of cancer death from DTC was estimated by the TMN classification. It has been established in classes V: Group I includes patients under the age of 55 with any T and N, Mo, and the cancer-related risk of death is <2%. Class II represents patients with any T and N except M1 or patients aged 55 years or older and TMN T1 / T2 N1 M0 or T3a / T3b classification, any N and M0. In this group the risk is 5%. Group III included patients aged 55 or older with T4a, any N and M0 with a 5-20% risk. Class IVa identifies subjects with at least 55 years of T4b, any N, M0 and a risk> 50%. Class IVb has a risk of death> 80%, with any T, any N and M1 [ 16 ^[8][9] , 17 ] (seeTable 1 ).

European Association of Nuclear Medicine (EANM) guidelines recommended ¹³¹I therapy only for the ablation of thyroid remnants as a post-surgical adjuvant procedure, for microscopic or incompletely resectable DTC, other non-resectable lesions, or both purposes (See Table 3).

More than 80% of patients are identified as low risk ^[12][19]. The risk influences therapy management: for ATA iodine therapy is not recommended for the residual ablation in low-risk DTC, while adjuvant therapy is recommended in intermediate-risk with activities between 30 mCi (1110 MBq) and 150 mCi (5550 MBq). In 70-year-old or older patients, activities up to 200 mCi (7400 MBq) of ¹³¹I are desirable to avoid possible adverse events.

Hürthle cell, invasive FTC, tall cell, diffuse sclerosis variant, and columnar cell carcinomas are characterized by the worst prognosis due to their high vascularity and therefore a higher risk of hematogenic spread. It is common in patients with these histological types to have metastases at diagnosis ^[10][18]. Risk classification is a milestone for iodine treatment. The therapy of DTC is customized on the basis of histological classification and Tumor Nodes Metastasis (TNM) staging and risk of recurrence. In particular, the risk of structural recurrence has been valued by the American Thyroid Association (ATA) on the basis of histological characteristics such as capsular, lymph nodes, and vascular invasion, the presence of aggressive histologic type, the number of lymph nodes metastasis, and thyroglobulin serum levels. The patients are, therefore, classified at a low (recurrence risk <5%, it is the case of PTC and well-differentiated and minimally invasive FTC), intermediate (5–20%), or high (>20%) risk of recurrence ^[11][15] (See Table 2).

The administration of RAIT also for ATA and EANM ought to be limited to remnant ablation, adjuvant treatment, or treatment of known diseases such as locoregional metastasis and distant metastasis ^[13][20]. In particular, RAIT is recommended in microscopic DTC. It finds explication in better results of RAIT in microscopic or small macroscopic tumors than in larger lesions ^[14][12]. The intention of cure or palliation using RAIT should be considered case by case, which represents the best choice as adjuvant treatment post-surgery of persistent or recurrent DTC until the lesion is iodine avid. Lymph nodes, lung, and most tissue metastasis can be successfully treated by RAIT with or without the auxilium of surgery, while bone and brain metastasis does not have a high rate of cure ^[15][21]. Moreover, less differentiated RAIT tumor histotypes, such as papillary tall-cell, columnar cell or diffuse sclerosing or follicular widely invasive, poorly differentiated, or Hürthle cells, have a greater risk of relapse and a lower overall survival but, despite diminished NIS expression, such tumors may be responders to RAIT ^[16][22]. RAIT should be also considered in patients >45 years since the most aggressive forms are diffused in older age and have a reduced overall survival ^[11][15], but also in patients unable to tolerate surgery or other eventual other therapies such as chemotherapy. According to recent European guidelines, in addition to pregnancy and breastfeeding, exclusions for RAIT are patients with unifocal papillary thyroid cancer ≤1 cm sized without metastasis, thyroid capsule invasion, or a history of radiation exposure. Additionally, tall-cell, columnar cell, or diffuse sclerosing subtypes are excluded [3]. Relative exceptions are bone marrow depression, pulmonary function restriction, especially if lung metastasis is present, and salivary gland function restriction. It is also necessary to evaluate the presence of neurological symptoms or damage since inflammation and local edema due to RAIT effect on cerebral metastasis could result in severe compression effects. Furthermore, 18F-fluorodeoxyglucose (18F-FDG) uptake in metastases is an evident indicator of the presence of radioiodine non-avid disease and a considerable independent unfavorable prognostic indicator. Specifically, the numbers of FDG-avid lesions and the higher maximum standard uptake values (SUVmax) correlate with overall mortality ^[17][23]. Radioiodine therapy is commonly well tolerated even if it has been suggested to be a potential second primary malignancy superior to 30% in a study conducted on 13 types of cancers ^[18][24]. However, EAMN evaluated that risk to be inferior to 1% in a latency period of ≥5 years. More cases have been observed when cumulative radioiodine activities exceed 20–30 GBq ^[19][20][25,26]. To maximize therapeutic effects and minimize collateral events, the maximum activity allowed has been proposed. A single administration of 1–5 GBq is recommended for the ablation of post-surgical thyroid remnants, even if there are different clinical practices among centers. It remains controversial which activities between 1.11GBq, 1.85 GBq, or 3.7 GBq are the best option. A systematic review by Hackshaw et al. established that ablation success rates are similar using 1.11 GBq and 3.78 GBq of ¹³¹I ^[21][27]. The accepted dosimetry approach is the blood-based method that takes into account blood and the whole body as compartments under the assumption that a similar iodine concentration has been found in blood and red marrow ^[22][28]. Still, the presence of extended metastatic bone involvement represents a limitation for the method due to the underestimation of the absorbed dose in red marrow. Likewise, in the presence of diffuse pulmonary micro-metastases, this method should be applied with caution because the critical organ, in this case, is the lung and not red marrow so dosimetry could be applied to the lung ^[23][24][29,30].

For ATA, adjuvant treatment after total thyroidectomy should be deemed in intermediate-risk and high-risk DTC using activities between 30 mCi (1110 MBq) and 150 mCi (5550 MBq) ^[25][31]. However, a common practice was determining the activity through empiric fixed doses chosen by physicians on the basis of experiences, convention, and patient parameters ^[26][27][32,33]. The method mostly used was Beierwaltes’s proposal ^[28][34]. It considers that the administered activity should not exceed 5.55–6.2 GBq for regional nodes that could not be removed by surgery, 6.2–7.4 GBq for pulmonary metastasis, and 7.4 GBq for skeletal metastasis. Other intermediate activities were used such as Schlumberger et al. ^[29][35], which used an initial activity of 3.7 GBq of ¹³¹I for pulmonary and bone metastasis every 3–6 months, and Menzel et al., which used a fixed activity of 11.1 GBq at 3-month intervals ^[30][36]. In this regard, a riveting study by Iizuka and collaborators makes a comparison between clinical outcomes of intermediate-high risk patients treated with high-dose and low-dose iodine therapy ^[31][37]. For low-dose therapy is intended the administration of 1100 MBq of ¹³¹I, while the high-dose is defined as 2960–3700 MBq of ¹³¹I administration. They collocated initial success with a thyroglobulin (Tg) level inferior to 2.0 ng/mL without thyroid-stimulating hormone administration and no ¹³¹I uptake in the thyroid region at ¹³¹I scintigraphy performed 6–12 months after RAIT. The conducted analysis established that the success rate of iodine therapy tended to be superior in high-risk patients treated with high-dose therapy even if no statistical differences have been found between the two groups in terms of success rate. These results reveal that in high-risk patients, a high-dose treatment is preferable. On the other side, Qu et al. evaluated the therapeutic response between low- and intermediate-risk patients treated with high-activity (3.7 MBq) and low-activity (1.1 MBq) ^[32][38]. The ablation and therapeutic results were similar between the two groups. Lymph node involvement and serum Tg seemed to influence ablation and therapeutic response, respectively. The success rate of ablation was lower for patients in the N1b stage than for patients in the N0 stage, while increased with lower serum Tg levels. In particular, they noted that a pre-treatment Tg level was associated with a higher better response: a level of 0.47 μg/L was a cut-off for predicting ablation results and a level of 3.09 μg/L for predicting therapeutic response. Thirty-two years of follow-up in DTC patients after RAIT was conducted by Martins-Filho and colleagues, to evaluate tumor behavior in correlation with cumulative iodine doses and survival. Patients with higher Tg serum levels had progressive disease and needed more frequent and higher doses of iodine treatment ^[33][39]. They observed that a patient with an age below 45 years had a 70% chance of complete remission with a cumulative activity of 100 mCi. If the activity ranges in 1 Ci, there is a 27% chance of stabilizing the disease. A high chance of progression was evident for cumulative activities of 600 mCi in a patient with an age above 45 years, and also for cumulative activities of 800 mCi or higher in a patient under 45 years. The data suggest a careful evaluation of further RAIT. There also is no common line of decision making for children since some centers set the activity by body weight or surface area or by age ^[34][40], while for German guidelines it should be adjusted according to the 24 h thyroid bed uptake and body weight: if uptake is <5%, the activity should be 50 MBq/kg, 5–10% uptake would assure an activity of 25 MBq/kg and 10–20% uptake an activity of 15 MBq/kg ^[35][41].

A field that needs standardized applications again is RAIT in metastatic DTC. There had been multiple differentiated approaches in the past and present among centers. First of all, the empiric proposals: Beierwalters et al. provided 6.48 GBq (175 mCi) for lung and 7.4 GBq (200 mCi) for bone ^[28][34]; Schlunmberger et al. used 3.7 GBq (100 mCi) every 3 months to 6 months until whole-body scan (WBS) was negative ^[25][31]; Petrich et al. used a therapeutic activity of 3.7 GBq (100 mCi) and if metastasis were still evident, they retreated the patient with 7.4 GBq (200 mCi) ^[36][42]; Brown et al. used 5.5 GBq (150mCi) every 3–4 months until the scan was negative of there was progression; Menzel et al. administered 11.1 GBq (300 mCi) every 3 months ^[30][36]; Durante et al. opted for 3.7 GBq (100 mCi) every 3–9 months during the first 2 years, once a year after until no metastasis was evident ^[14][12]; Hindle et al. used 3.7 GBq (100 mCi) every 6 months if there was lung uptake and chest radiography was negative, but if a cumulative activity of 18.5 GBq (500 mCi) was reached and lung uptake was present, reduced the therapy once a year and then every 2 years ^[37][43]. Again, Hindle et al. recommended the use of activity between 3.7 and 5.5 GBq (100–150 mCi) every 6 months when lung uptake was evident but also chest radiography was ^[38][44]. Ghachem et al. compared the ablation rate between patients treated with 1.1 to 1.85 GBq and patients treated with 3.7 GBq of ¹³¹I ^[39][45]. They had a similar ablation rate but the likelihood to have remission was 1.83 times greater for higher activity determining that mini dose protocol is not more effective in ablation than the higher dose protocol. 2013/59 EURATOM recommended for metastatic DTC that the optimal activity is the one that permits the best response rate and low toxicity even if it is still in debate and the same international guidelines do not suggest if it is preferable to use empiric or dosimetry-based therapies in these patients ^[40][46].

The side effects of RAIT, especially on bone marrow, have been widely described even if the great majority of these studies do not evaluate them through dosimetry studies. In an intriguing review by Andresen et al. ^[41][47], adverse effects have been related to dose levels summarizing what was reported in the literature.

To exclude the possibility to have bone marrow damage, the absorbed dose to the blood should not exceed 2 Gy, and the whole-body retention after 48 h from the administration should not be superior to 4.4 or 3 GBq in the absence or presence of iodine-avid diffuse lung metastases, respectively ^[42][43][48,49].

The considered side effects such as nausea, neck pain, lacrimal and salivary dysfunction, and altered smell and taste seem to be acute effects that are more frequent in therapies with 100 mCi, and less frequent for dosages between 30 and 50 mCi (respectively, 13% vs. 4%, 17% vs. 7%, 10–24% vs. 8–20%, 5–16% vs. 6–13%, 6% vs. 0 and 2% vs. 0). Sialadenitis has a frequency between 2 and 67% and nasolacrimal duct obstruction 3.4% ^[44][50].

RAIT should be carefully evaluated in patients with restriction of salivary to minimize side effects (e.g., xerostomia) since salivary glands are the physiological site of iodine uptake. It happens because of the presence of NIS, a plasma membrane glycoprotein, that mediates active I(-) transport in various organs such as the thyroid, the previously written salivary glands, stomach, lactating breast, and small intestine ^[45][51]. Acute salivary gland swelling and pain decrease a few days after ¹³¹I-therapy in patients with DTC but, in some patients, the onset of these symptoms is slower. In other patients, the symptoms could lead to chronic radiation sialadenitis. These complications are clinically significant in 10–30% of subjects even if their frequency is not well known ^{[46][47][48][49][50][51][52]}[52,53,54,55,56,57,58]. In particular, in the prospective cohort study by Hyer et al., patients were treated with an ablative activity of 3GBq and a further of 5.5 GBq in case of residual disease. Of the 79 patients examined, 26% developed salivary gland toxicity after iodine therapy, with a median onset of 48 h after ¹³¹I administration and a median duration of symptoms of 12 months. The reported symptoms were xerostomia in all cases, 2 patients were referred with an episode of swelling and pain, 6 patients had two episodes, and 8 subjects had three or more episodes. The submandibular gland was interested in 50% of subjects, parotid in 39%, and 11% of patients had both glands affected. Among these patients, 55% had lymph node metastasis while only 32% of patients without salivary gland involvement had lymphatic metastasis. The mean cumulative activities were 8.57 GBq in patients with salivary complications, 9.04 GBq for the others. These results demonstrated that repeated activities were more associated with the development of salivary gland toxicity. Conversely, cumulative doses seemed not to be related suggesting how it is more damaging repeated doses than a single dose ^[53][59]. Since sialadenitis is not always clinically evident, many other studies using 99mTC-pertechnetate scintigraphy were conducted to evaluate salivary gland function. Abnormalities at salivary gland scintigraphy were noted with activities of more than 18.5 GBq ¹³¹I ^[54][60], but also with activities between 3.7 GBq and 38.7 GBq of ¹³¹I ^[55][61]. As regards the timing of symptoms that arise, it was reported from months to years after RAIT ^[52][56][58,62], and in a study by Allweiss et al. ^[51][57], they continued up to 2.5 years later. It was also reported an increase of risk in developing salivary gland neoplasm such as pleomorphic adenoma, non-Hodgkin’s lymphoma, and mucoepidermoid carcinoma and it is proportional to the ¹³¹I dose ^[54][60]. Moreover, it seems that the administration of lemon candy post-therapy increases the risk of acute and chronic salivary gland toxicity. The rationale for using lemon candy after iodine administration is based on the concept that ascorbic acid increases the salivary flow and then iodine clearance. On the other flip, it enhances blood flow to rising ¹³¹I uptake. In the context of diminished glomerular filtration rate determined by hypothyroidism, the serum level of ¹³¹I is high. Additionally, the remnant or metastatic tissues are not so avid as the normal thyroid. Consequently, the continuous consumption of lemon candy could only improve this condition leading to a greater amount of dose to the salivary gland ^[57][63].

Male and female infertility have subacute latency, differently from the upper cited sialadenitis but also from nasolacrimal duct obstruction, and the onset of secondary malignancy that are late effects that seem to be evident only for ¹³¹I activity major to 100 mCi. Notably, in males, a transient decrease of follicle-stimulating hormone (FSH) levels and reduced sperm motility were evident while in women it has been observed lower birth rate in the age group between 35 and 39 years.

A potential impact has been pointed out by Bourcigaux et al. on exocrine and endocrine testicular function after 3.7 GBq iodine administration. They observed Sertoli cell function and induced sperm chromosomal abnormalities 3 months and 13 months after therapy ^[58][64].

According to literature from 30 to 40 years old, recurrences were thought to occur 10–20 years after RAIT. Today, thyroid cancer is presented at an earlier stage and has a lower recurrence rate of 3–10% ^{[45][46][47][48]}[51,52,53,54]. Moreover, Carrillo et al. demonstrated that patients with DTC treated with ¹³¹I, which have a biochemical recurrence and no diagnostic WBS (DWBS) or extensive studies, have lower frequencies of second recurrence (SR) compared to the group that, prior RAIT, underwent a DWBS with 5 mCi of ¹³¹I and Magnetic Resonance Imaging (MRI) and/or 18F-FDG PET if DWBS was negative ^[49][55]. In a study by Rubino et al., second primary malignancies (SPM) and leukemia risk were evaluated. They selected a cohort of 6841 Swedish, Italian, and French patients with FTC or PTC diagnosed during the period 1934 and 1995 ^[19][25]. In total, 576 patients had SPM and among them, 13 presented a third malignant neoplasm less than 2 years after. The risk of cancer was globally increased by 27% compared to the general population of the same countries. The mean time between thyroid cancer diagnosis and SPM presentation was 15 years. No gender differences were observed. The cancer types with a significantly increased risk were cancers in the digestive tract, bone and soft tissue, skin melanoma, kidney, central nervous system, endocrine glands (excluded thyroid), leukemias, female breast cancer, and genital male malignancies. No association between SPM and external radiotherapy was observed, so no interactions between them could be described. In total, 13 cases of oral cancer were counted, and of them, 7 were caused by ¹³¹I administration. They also demonstrated a significant association between cumulative activities of ¹³¹I and the risk of solid cancer, with an increase of 4% per GBq.

Another noteworthy event to take into consideration for RAIT success is represented by the “stunning” phenomenon. It consists of the diminishing of iodine uptake due to the suboptimal therapeutic effects, biological effects, or both, caused by a previous diagnostic radioiodine administration. To avoid this phenomenon, it is suggested to shun unnecessary pre-therapeutic ¹³¹I administration, when the indications to RAIT are clear, or if necessary, to use lower radioiodine activities before RAIT. Hence, an activity of 3–10 MBq is recommended for uptake quantification and 10–185 MBq for WBS. An alternative could be opting for other tracers such as 123iodine (¹²³I), 99mTechnetium (99mTc), or ¹²⁴I in PET/CT. Dosimetry studies demonstrated that activities of 10–20 MBq represent a considerable radiation burden for thyroid cells. The stunning effect seems to be caused by direct radiation damage to thyrocytes ^{[59][60][61][62][63][64][65][66][67]}[65,66,67,68,69,70,71,72,73]. This results in a diminishing in iodine transportation by 50% also with an absorbed dose of 3 Gy ^[68][74] and down-regulation in NIS expression ^[69][75]. These phenomena lead to a diminishing in iodine uptake ^[63][69]. In this regard, Verburg and collaborators compared two groups of patients, one undergone a pre-ablative 24 h uptake measurement with 40 MBq ¹³¹I and the other without pre-ablative diagnostic scintigraphy ^[59][65]. They demonstrated how the success rate of ablation was 2 times greater in the group without pre-ablative scintigraphy than in the group that underwent a pre-therapeutic uptake study. Similar results were cited in the papers: Jeevanram et al., who reported a decrease of 20–25% uptake when the activity of 111–185 MBq was used for pre-ablation scan ^[70][76]; Lassman et al., after 74 MBq ^[60][66]; Muratet et al., for 111 MBq ^[71][77]; Hu et al. and Lees et al., after using 185 MBq ^[61][72][67,78]; and Park et al., for 370 MBq ^[73][79]. No stunning events were observed in groups undergoing 37 MBq scans ^[71][77], with ¹²³I or without any pre-therapeutic scans ^[72][78].

Notably, patients with reduced kidney function or hemodialysis reached higher bone marrow doses per unit of administered activity ^[27][33]. Recently, in low-risk cancer, the long-term cancer control rates were equivalent if therapeutic activities between 30–50 mCi or major or equal to 100 mCi were used ^[74][80]. In contrast, it was suggested that higher doses could better control cancer in intermediate- and high-risk patients ^[41][75][47,81]. In a recent phase III trials, Schlumberger et al. proved equivalent rates of thyroid remnant ablation with 30 versus 100 mCi ^[76][82]. Mallick et al., instead, asserted that low-dose using recombinant human thyrotropin alfa was as effective as high-dose radioiodine, but it has a lower rate of side effects ^[77][83]. In a paper by Khang et al., the mortality risk for SR is higher for patients with a cumulative ¹³¹I activity > 37 GBq. On the other hand, the risk of DTC-specific mortality is lower compared to the group which received a cumulative activity < 37 GBq and has similar all-cause mortality to low-activity RAIT patients ^[78][84]. In a study by Klubo-Gwiezdzinska et al., it was well demonstrated that the dosimetry approach has more efficacy compared to the empiric approach ^[79][85]. These considerations highlight how the dosimetry-based approach is more effective and represents a key role for a more personalized therapy, especially in patients with metastatic disease to minimize side effects.