Characterization of the patient profile and efficacy of empirical radioiodine therapy in differentiated thyroid cancer with biochemical evidence without identified iodine-avid structural metastatic disease

Abstract

Objective

To evaluate the profile of patients and tumor features of differentiated thyroid carcinoma (DTC) with biochemical evidence without structurally demonstrable metastatic disease in the postoperative follow-up, regarding the response to empirical radioiodine therapy (RIT).

Methods

A retrospective study of 196 patients with DTC followed up at a tertiary service from 1990 to 2018 was performed. All of them presented negative diagnostic whole-body scan (WBS) and high TG levels. Based on criteria indicating greater severity, as higher TG levels, 72 of them received empirical RIT. All remained in follow-up with their outcomes assessed.

Results

Carcinoma histopathological characteristics were similar between the groups of patients. In all the assessed moments, TG levels were higher in patients who received RIT than in patients who did not receive it. Regarding WBS after RIT, 57.74% of cases showed uptake, with one-third showing evidence of distant metastasis and 23.94% with cervical uptake in remnant/recurrence of thyroid or lymph node tissue. Patients who received RIT showed a partial reduction in suppressed serum TG, while patients who did not receive RIT exhibited stability of TG during follow-up.

Conclusion

After radioiodine therapy administered with biochemical evidence of metastases but without identification of iodine-avid structural disease, metastatic foci avid for iodine were identified in 57% of cases, with a subsequent reduction in serum TG, indicating a potential therapeutic benefit. Despite similar tumor histopathological characteristics, the outcomes differed between patients who received RIT and those who did not, highlighting the importance of individualizing this indication.

Introduction

Thyroid cancer is the most common malignant neoplasm within the endocrine system (1). Differentiated thyroid carcinoma (DTC) is the most common type, accounting for approximately 95% of all cases (2). Treatment with I-131, along with suppressive therapy with levothyroxine to reduce thyroid-stimulating hormone (TSH) levels, constitutes the main adjuvant therapy to surgical treatment of DTC, especially in patients with a high risk of recurrence or who present evidence of biochemical or structural disease (3, 4). The iodine activity varies from 100 to 200 mCi for lymph node, pulmonary, or bone metastases (3, 5). There is no consensus on the administered activity, given the challenges in conducting randomized prospective studies due to the natural history of the disease and the long time required to detect changes during its progression (5).

Although most thyroid cancers have a favorable prognosis, the increasing incidence has contributed to a growing number of patients with recurrent disease, rapid disease progression, and poor or no response to conventional treatments (1). The challenge is to distinguish patients who warrant more aggressive treatment while sparing others from unnecessary treatments and procedures. Therefore, for therapeutic planning and determining the best way to monitor patients with DTC, it is essential to assess the risk of recurrence and disease progression.

Surveillance for recurrences and metastases of DTC includes serum thyroglobulin (TG) levels and anti-thyroglobulin antibodies (TgAbs), as well as cervical ultrasonography and chest computed tomography to localize the focus of recurrence or metastasis for appropriate treatment planning. In general, undetectable levels of TG and TgAbs suggest a disease-free state, while detectable or elevated concentrations are suggestive of local or distant metastases, which usually uptake I-131.

Previous studies have shown that negative diagnostic WBS and elevated serum TG occur in 10–27% of patients with DTC after initial therapy (6, 7) and are associated with a lower overall survival (OS) (7, 8, 9). Elevated serum TG indicates persistent or recurrent disease with high sensitivity and specificity (10). If WBS shows only slight uptake in the thyroid bed, the presence of metastases should also be investigated, considering that this is likely not the source of the elevated TG (11).

The American Thyroid Association (ATA) guidelines (3) recommend considering the use of empirical radioiodine activity in patients with significantly higher or rapidly rising serum TG levels in whom cross-sectional and/or 18F-FDG PET/CT imaging did not reveal tumor amenable to targeted therapy.

Overall, the goal of treatment with I-131 is to reduce recurrence and mortality from carcinoma by eradicating microscopic or macroscopic tumor foci, as well as to enhance the accuracy of TG measurements and WBS in screening for metastasis. Numerous studies report improved survival after I-131 treatment in patients with residual, metastatic, or high-risk disease based on clinicopathological criteria (12, 13).

However, patients showing indeterminate response could be followed with thyroid hormone therapy alone in the absence of structurally demonstrable disease, stimulated serum TG < 10 ng/mL after thyroid hormone withdrawal or <5 ng/mL with rhTSH. In addition, there is no documented cutoff value for a serum TG level above which a patient should be empirically treated with radioiodine (3). This study aims to evaluate demographic and tumor characteristics of patients with DTC regarding the indication of empirical radioiodine activity and influence on the clinical evolution of the disease.

Methods

Study design and patients

This is a retrospective study conducted through a review of medical records of patients with DTC under follow-up at the Thyroid Cancer Service, Division of Endocrinology, from 1990 to 2018. The study included patients with detectable serum thyroglobulin levels concurrent with negative diagnostic whole-body scans at any time during follow-up after total thyroidectomy and ablative radioiodine therapy. The study was approved by the local Human Research Ethics Committee (CAAE 55558322.3.0000.5404).

Inclusion criteria were patients over 18 years old at diagnosis, of both sexes; adequate clinical and pathological data to allow for precise classification of initial risk stratification; and a minimum follow-up of 24 months after the administration of the empirical dose. Patients with elevated TgAbs, with incomplete data in the medical records or who have lost to follow-up in the service were excluded.

Clinical data comprised the following: sex; age at the date of medical record evaluation and at the DTC diagnosis; previous presence of anti-thyroid antibodies; histological type and subtype of thyroid carcinoma; TNM staging (14); initial recurrence risk stratification (3); ablative therapy with iodine; mean cumulative dose of radioactive iodine (in mCi); outpatient follow-up time until serum thyroglobulin levels rise, absolute TG values; presence of iodine uptake on scintigraphy prior to the negative scan; presence of uptake on 18F-FDG PET, when performed; empirical radioiodine activity administered; presence or absence of uptake on post-empirical dose WBS; classification based on response to empirical radioiodine therapy; and refractoriness to treatment.

All patients underwent total thyroidectomy followed by a diagnostic WBS, ablative RIT, and a WBS five days after radioiodine therapy. In all patients with a negative WBS and biochemical markers present, cervical ultrasound and chest computed tomography were performed to locate residual/metastatic disease. Only after these examinations showed no evident lesion was an empirical dose of radioiodine therapy considered. All patients were submitted to treatment for hypothyroidism after levothyroxine withdrawal (recombinant TSH was not administered in any case). All WBSs were performed with SPECT/CT images.

Refractoriness of DTC to RIT was defined as follows: i) malignant/metastatic tissue does not concentrate radioactive iodine on the diagnostic WBS; ii) malignant tissue does not concentrate radioactive iodine on the post-activity WBS; iii) malignant tissue loses the ability to concentrate radioactive iodine after prior evidence of iodine-avid disease; and iv) metastatic disease progresses despite a significant concentration of radioactive iodine (3, 15, 16, 17). The present study adopted all the above criteria to classify DTC as refractory.

The suppressed thyroglobulin levels of all patients during follow-up were analyzed and classified in the periods of 1 year after the first ablative RIT (which all patients underwent) and at the patient’s last consultation. All thyroglobulin levels were measured against a TSH target defined in the literature according to the response and risk of tumor recurrence. In this way, the tendency of TG variation was established and classified into four different categories: negative, partial reduction, stable, or elevation. The classification was as follows: suppressed thyroglobulin levels below 0.2 ng/mL were considered negative, a reduction in values of more than 50% was considered a partial reduction, a reduction below 50% was considered stable TG, and any increase was considered as rising thyroglobulin.

Statistical analysis

The results of quantitative variables were described using means, standard deviations, medians, and first and third quartile values. Descriptive analysis was performed with the presentation of frequency tables for categorical variables and measures of central tendency and dispersion for continuous variables. The chi-square test or Fisher’s exact test, when necessary, was used for comparing proportions. The Mann–Whitney test was applied to compare continuous measures between two groups. The significance level adopted for statistical tests was 5%.

Results

A total of 1,312 patients with differentiated thyroid cancer were evaluated, and 196 patients (14.93%) presented negative diagnostic whole-body scan (dxWBS) and elevated TG levels. These patients were divided into two groups according to whether they had received (n = 72; 5.48%) or not (n = 124) empirical RIT.

The age at diagnosis was 43.97 ± 14.27 years in the empirical RIT group, similar to 41.27 ± 14.72 years in the non-treated group (P = 0.1767). There was no difference between the groups regarding sex, family history of thyroid neoplasia, and reason for surgery (data not shown). There was a higher prevalence of women (n = 54, 75% in empirical RIT patients, and n = 101, 81.45% in non-treated patients) in both groups.

There were no differences between the groups when analyzing tumor characteristics: diameter of the largest nodule (cm), histological type of the tumor, capsular/vascular or lymphatic invasion, presence of extrathyroidal tumor invasion, degree of tumor differentiation, lymph node metastasis, and TNM staging (Table 1).

Table 1.

Tumor characteristics in patients with differentiated thyroid carcinoma receiving or not empirical radioiodine therapy.

Variables	Empirical radioiodine therapy		P-value
	Yes	No
Largest nodule diameter (cm)	2.67 ± 1.84	2.88 ± 1.82	0.3657
Histological type
Papillary	66 (91.67%)	114 (91.94%)	0.7966
Follicular	5 (6.94%)	10 (8.06%)
Capsule invasion	17 (47.22%)	32 (51.61%)	0.6752
Vascular invasion	23 (46.94%)	31 (33.70%)	0.1235
Lymphatic invasion	9 (20.93%)	15 (18.75%)	0.7711
Extrathyroidal tumor invasion	38 (61.29%)	50 (48.54%)	0.1119
Well-differentiated tumor	68 (94.44%)	118 (95.16%)	>0.999
Lymph node metastasis	43 (59.72%)	61 (49.19%)	0.1545
TNM
T1	20 (28.17%)	31 (25.41%)	0.3093
T2	11 (15.49%)	32 (26.23%)
T3	12 (16.90%)	22 (18.03%)
T4	28 (39.44%)	37 (30.33%)
N0	28 (38.89%)	59 (47.58%)	0.2377
N1	44 (61.11%)	65 (52.42%)
M0	61 (87.14%)	114 (91.94%)	0.2808
M1	9 (12.86%)	10 (8.06%)

Categorical variables are represented by absolute number and frequency; numerical variables are represented by mean ± standard deviation. TNM, tumor–node–metastasis; T, primary tumor; N, lymph node metastasis; M, distant metastasis.

There were no differences between the groups in the percentage of cervical uptake on the dxWBS, the ablative activity of the administered radioiodine, the number of positive WBS results before the first negative WBS during follow-up, and the time between the ablative RIT and the first negative WBS. The total radioiodine activity in the empirical RIT group (ablative + empirical) was 508.59 mCi, while the activity of radioiodine administered empirically was 222.08 ± 64.85 mCi. There was a difference in the total radioiodine dose in the empirical RIT group compared to the group that did not receive it, which means 185.56 mCi. Refractoriness to radioiodine was associated with the empirical RIT group (Table 2).

Table 2.

Diagnostic and therapeutic strategy of radioiodine therapy when comparing patients who did or did not receive empirical RIT.

Variables	Empirical RIT		P-value
	Yes	No
Cervical uptake in dxWBS (%)	2.67 ± 3.88	2.42 ± 3.03	0.6727
Ablative radioiodine therapy (mCi)	135.42 ± 45.07	134.76 ± 43.92	0.4370
Number of positive WBSs prior to the negative WBS	1.42 ± 0.76	1.38 ± 0.83	0.3311
Time between ablative RIT and first negative WBS (months)	33.85 ± 37.47	26.92 ± 25.21	0.6431
Empirical radioiodine activity (mCi)	222.08 ± 64.85	-	-
Total radioiodine activity (mCi)	508.59 ± 240.29	185.56 ± 134.90	<0.001
Radioiodine refractoriness	40 (56.34%)	14 (11.29%)	<0.001

Categorical variables are represented by absolute number and frequency, and numerical variables are represented by mean ± standard deviation. dxWBS, diagnostic whole-body scan; WBS, whole-body scan; RIT, radioactive iodine therapy.

Regarding WBS after the empirical dose, 42.25% remained negative, while in 57.74% of cases, there was post-dose uptake. Among the positive ones, in 33.80% of patients, the uptake showed distant metastasis, 16.90% presented cervical uptake, and 7.04% showed cervical lymph node uptake.

As shown in Tables 3 and 4, the stimulated thyroglobulin (TG) levels in the first WBS, suppressed TG after the first year of follow-up after the first RIT (ablative), stimulated TG at the time of negative PCI and the last TG of the clinical follow-up were higher in the group that received empirical RIT than in the group that did not. There was no difference between the groups regarding the performance of neck dissection at diagnosis and late neck dissection (Table 3).

Table 3.

Patient outcomes with differentiated thyroid carcinoma when comparing the empirical RIT group and the non-RIT group.

Variables	Empirical RIT		P-value
	Yes	No
Stimulated TG at ablative RIT (ng/mL)	299.72 ± 970.45	38.51 ± 108.81	<0.001
Suppressed TG one year after ablative RIT (ng/mL)	131.54 ± 743.17	4.73 ± 20.46	<0.001
Stimulated TG at first negative WBS (ng/mL)	271.10 ± 1,051.73	32.58 ± 240.55	<0.001
Suppressed TG at last evaluation (ng/mL)	516.46 ± 1,851.45	111.51 ± 927.26	<0.001
Cervical emptying at diagnosis	35 (48.61%)	43 (34.68%)	0.0547
Uptake in 18F-FDG PET	34/43 (79.07%)	18/21 (85.71%)	0.7357
Late cervical emptying	16 (22.54%)	16 (12.90%)	0.0806
Late metastases (months)	34.12 ± 29.40	49.93 ± 35.25	0.0342
Disease-free survival (months)	4.24 ± 16.07	22.42 ± 41.13	<0.001
Follow-up time (years)	12.73 ± 6.83	10.04 ± 5.41	0.0067
Tumor-related deaths	4 (5.56%)	1 (0.81%)	0.0618

Categorical variables are represented by absolute number and frequency, and numerical variables are represented by mean ± standard deviation. RIT, radioiodine therapy; TG, thyroglobulin; WBS, whole-body scan; 18F-FDG PET, positron emission tomography with fluorodeoxyglucose.

Table 4.

Evolution of thyroglobulin levels during follow-up of patients with DTC when comparing the groups that received or did not receive empirical RIT.

Variables	Empirical RIT		P-value
	Yes	No
Stimulated TG at ablative RIT (ng/mL)			<0.001
Mean ± SD	299.72 ± 970.45	38.51 ± 108.81
Median (1st – 3rd quartile)	33.50 (16.02–127.75)	6.23 (2.21–14.60)
Suppressed TG one year after ablative RIT (ng/mL)			<0.001
Mean ± SD	131.54 ± 743.17	4.73 ± 20.46
Median (1st – 3rd quartile)	2.82 (0.31–10.97)	0.25 (0.20–0.81)
Stimulated TG at 1st negative WBS (ng/mL)			<0.001
Mean ± SD	271.10 ± 1,051.73	32.58 ± 240.55
Median (1st – 3rd quartile)	33.98 (11.15–90.10)	2.30 (0.87–6.58)
Suppressed TG at last evaluation (ng/mL)			<0.001
Mean ± SD	516.46 ± 1,851.45	111.51 ± 927.26
Median (1st – 3rd quartile)	0.57 (0.20–7.26)	0.20 (0.20–0.43)

TG, thyroglobulin; RIT, radioiodine therapy; WBS, whole-body scan.

Patients who underwent 18F-FDG PET, all with iodine-non-avid disease in the WBS (performed with SPECT/CT images), had their results analyzed in relation to the presence or absence of uptake, regardless of the uptake site. In group 1, of the 43 patients who underwent the exam, 34 had an uptake, while in group 2, of the 21 exams performed, 18 had an uptake, with no statistical difference between the groups, as shown in Table 3.

The appearance of metastases was later in the group that did not receive the treatment when compared to the empirical RIT group, as well as a greater disease-free survival. The mean follow-up of patients was 12.73 years in the empirical RIT group versus 10.04 years in the group that did not receive the treatment. There was no difference when evaluating the number of tumor-related deaths.

The group that did not receive empirical RIT was associated with a low risk of persistence/recurrence at diagnosis, while the empirical RIT group was associated with a high risk. In the latter group, 80% of patients had disease persistence/recurrence at the last evaluation, a prevalence higher than that in the group that did not receive the empirical dose (50% of patients) (Table 5).

Table 5.

Risk of persistence or recurrence of differentiated thyroid carcinoma, treatment response after 1 year of ablative RIT, and at the last evaluation in patients who did or did not receive empirical RIT.

Variables	Empirical RIT		P-value
	Yes	No
Risk of persistence/recurrence at diagnosis			<0.001
Low	7 (9.72%)	44 (35.48%)
Intermediate	11 (15.28%)	29 (23.39%)
High	54 (75%)	51 (41.13%)
Response 1 year after ablative RIT			<0.001
Indeterminate	8 (11.76%)	45 (37.50%)
Biochemical incomplete	26 (38.24%)	12 (10%)
Structural incomplete	23 (33.82%)	21 (17.50%)
Excellent	11 (16.18%)	42 (35%)
Treatment response at last evaluation			<0.001
Indeterminate	17 (23.94%)	37 (29.84%)
Biochemical incomplete	10 (14.08%)	11 (8.87%)
Structural incomplete	30 (42.25%)	15 (12.10%)
Excellent	14 (19.72%)	61 (49.19%)
Disease status at last evaluation			<0.001
Disease-free	14 (19.72%)	61 (49.19%)
Persistence/recurrence	57 (80.28%)	63 (50.81%)

Categorical variables are represented by absolute number and frequency. RIT, radioiodine therapy.

Table 6 presents the treatment responses at different follow-up times. One year after the first RIT (performed by all patients), the group that received empirical RIT showed more frequently incomplete biochemical and structural responses during follow-up, and in the last evaluation, an incomplete structural response. A generalized estimating equations (GEE) analysis was conducted, which showed a change in response between the analyzed time points, independent of the group (P = 0.048), with no difference in response between the groups (P = 0.053) or in the time–group interaction (P = 0.096).

Table 6.

Treatment response in patients who received or not empirical radioiodine therapy: comparison 1 year after ablative radioiodine therapy, 1 year after empirical RIT, and at the last evaluation.

ATA response	Empirical RIT
	Yes			No
	1 year after ablative RIT	1 year after empirical RIT	Last evaluation	1 year after ablative RIT	Last evaluation
Indeterminate	8 (11.76%)	8 (12.12%)	17 (23.94%)	45 (37.5%)	37 (29.84%)
Incomplete biochemical	26 (38.24%)	14 (21.21%)	10 (14.08%)	12 (10%)	11 (8.87%)
Incomplete structural	23 (33.82%)	29 (43.94%)	30 (42.25%)	21 (17.5%)	15 (12.10%)
Excellent	11 (16.18%)	15 (22.73%)	14 (19.72%)	42 (35%)	61 (49.19%)

Categorical variables are represented by absolute number and frequency. Generalized estimating equations (GEE): change in response between the times 1 year after ablative RIT and last evaluation, regardless of the group (P 0.048), with no difference in response between the groups (P 0.053) or in the time–group interaction (P 0.096). ATA, American Thyroid Association; RIT, radioiodine therapy.

The suppressed TG (in ng/mL) of all patients was analyzed during follow-up, establishing the trend of TG variation in both groups (Fig. 1). The group that received empirical RIT was associated with a partial reduction in suppressed serum TG, while the group that did not receive RIT was associated with TG stability (P < 0.001).

Figure 1.

Trend of suppressed thyroglobulin course during evolution after treatment. Suppressed thyroglobulin levels below 0.2 ng/mL were considered negative, a reduction in values of more than 50% was considered a partial reduction, a reduction below 50% was considered stable TG, and any increase was considered as rising thyroglobulin. The group that received empirical RIT was associated with a partial reduction in suppressed serum TG, while the group that did not receive RIT was associated with TG stability. Categorical variables are represented by absolute numbers and frequencies.

Discussion

In the present study, we sought to determine the profile of patients with DTC and biochemical evidence without identified iodine-avid structural metastatic disease who received empirical RIT as well as its efficacy in a cohort from a specialized thyroid cancer service analyzed over a long follow-up period. We found that patients treated empirically with radioiodine showed greater tumor aggressiveness, with higher pre-treatment TG, a greater prevalence of refractoriness to radioiodine, associated with a high risk of disease persistence or recurrence, but without differences in the histopathological characteristics of DTC in relation to untreated patients. WBS uptake post-empirical RIT was noted in 57.74% of cases, followed by a partial reduction in suppressed serum TG in almost half of these patients. Despite the worse prognosis of patients who received RIT, contrary to expectations, there was no difference in the change in treatment response patterns over time when comparing the two groups, pointing to a possible improvement in the outcome associated with empirical RIT.

Negative diagnostic WBS with high serum TG occurs in 10–27% of patients with DTC after initial ablative RIT (6, 7), similarly to the prevalence of 14.93% found in this study. This situation indicates persistent or recurrent disease after ablative and/or adjuvant RIT and often raises the condition of radioiodine-refractory carcinoma, which is known to have a worse prognosis (7, 18, 19). However, it is not uncommon in these cases to empirically indicate radioiodine therapy, albeit without certainty regarding treatment efficacy. Many patients benefit from empirical treatment, while others do not (3, 9, 20, 21). Tramontin et al. (20) did not confirm differences during long-term follow-up between patients treated and untreated with empirical RIT. In contrast, Piscopo et al. (21) showed that the empirical administration of radioiodine had a beneficial effect on patient outcomes with DTC, with a lower risk of structural recurrence compared to patients who did not receive radioiodine.

Among DTCs, the papillary histological type accounts for 50–80% of cases, followed by follicular, 15–20% of cases (2). The prevalence in the evaluated patients was 91.84% papillary and 7.65% follicular. The same proportion was observed when analyzing the patients who received the empirical RIT, 91.67% papillary and 6.94% follicular. Regarding sex, there was a prevalence of females, similarly to the literature, indicating that the overall prevalence of thyroid cancer affects women three times more than men (22).

When analyzing patients with DTC presenting biochemical evidence of disease but no scintigraphy identification of structural disease on the I-131 WBS, despite similar histopathological tumor characteristics between the groups, we observed distinct outcomes between patients who received empirical RIT and those who did not. Compared to the untreated patients, patients who received empirical RIT showed higher stimulated TG during the post-thyroidectomy WBS and at the time of the negative WBS, as well as higher suppressed TG during the first year of follow-up after the ablative RIT, and the last TG of the follow-up, indicating greater tumor aggressiveness. These findings supported the recommendation for empirical therapy, as the decision was primarily based on the magnitude and behavior of the disease, since there is no TG cutoff value from which treatment is indicated. This decision could be interpreted as a selection bias; however, the same team of specialists evaluated the indication in the case of all patients, always according to the biological characteristics and evolution of the tumor. Smallridge et al. showed that depending on the TG concentration, a considerable percentage of physicians would consider indicating empirical RIT for patients with high TG/negative WBS (23).

The success of RIT requires adequate uptake and retention of iodine in the target tissue, which is referred to as iodine avidity. In the treatment of known metastases or those at high risk of developing them, avidity is often unknown. Thus, currently, the size of the primary tumor and the presence of any initial lymph node metastases guide the choice of radioiodine activity to be administered (24).

Refractory DTC does not respond to conventional radioiodine therapy (RIT). This treatment exploits the iodide uptake function of thyroid follicular cells, as well as the expression of specific glandular genes, such as the sodium/iodine symporter, thyroglobulin, TSH receptor, and thyroperoxidase, along with thyroid-specific transcription factors (25). The molecular basis underlying the loss of radioiodine avidity in RIT-refractory thyroid cancer is the silencing of thyroid-specific gene expression and transcription factors (25). Among these factors, decreased expression of NIS and/or reduced translocation of NIS to the membrane, primarily caused by genetic/epigenetic alterations and dysregulated signaling pathways, are the main mechanisms underlying radioiodine refractoriness (1, 25).

The arguments for indicating RIT without structural evidence of iodine-avid disease are based on the potential location of the disease, as determined by uptake after radioiodine administration, and consequently, a possible therapeutic benefit. The 2025 guidelines from the American Thyroid Association (ATA) (3) recommend considering the use of empirical radioiodine in patients with more significantly elevated or rapidly rising serum TG levels in the presence of a negative diagnostic WBS and without evidence of structural disease amenable to direct therapy. Activity of 100–200 mCi could be administered in these cases. In the present study, the mean empirical dose administered was 222 mCi, nearly half of the total activity (ablative + empirical = 508.59 mCi), and higher than that used in the untreated group, 185.56 mCi. Furthermore, 56.34% of patients who received empirical RIT met criteria for radioiodine refractoriness (17) during follow-up, in contrast to 11.29% of patients in the untreated group.

The percentage of cervical uptake in the first diagnostic WBS, the number of positive WBSs prior to the first negative WBS during follow-up, and the time between the first RIT and the first negative WBS were similar between patients who received the empirical dose and those whose dose was administered solely based on evidence of disease on the WBS.

Regarding the uptake in the WBS post-empirical RIT, 42.25% remained negative, while 57.74% showed an uptake. We also found that in 33.80% of patients, uptake indicated distant metastases, with cervical uptake in 16.90% and lymph node uptake in 7.04%. These data showed that after the empirical therapy, it was possible to identify metastatic foci, in addition to the potential therapeutic benefit due to the presence of uptake after activity administration.

When classifying treatment response after one year and at the last evaluation, high risk of persistence or recurrence was associated with patients who received empirical RIT, while low risk was associated with the untreated group, confirming lower aggressiveness in these patients.

The group of patients who received empirical RIT was associated with incomplete biochemical and structural responses after one year of ablative RIT and with incomplete structural response after the administration of empirical RIT. An increase in the prevalence of patients with excellent response after empirical RIT was also observed, rising from 16.18 to 19.72%, along with an increase in the frequency of indeterminate response from 11.76 to 23.94% of cases. Overall, there was an increase in the percentage of excellent responses and a decrease in incomplete biochemical response. A change in the pattern of treatment responses was noted concerning time factors, specifically between the pre-empirical RIT moments and the last patient evaluation. However, no difference was found between groups or in the time–group interaction, indicating that even with poorer prognostic characteristics, the behavior of the empirical RIT group resembled that of the untreated group, suggesting that the treatment may have influenced this outcome.

In this regard, when observing the trend of suppressed TG over follow-up, a partial reduction after empirical RIT was demonstrated. It is possible that the administration of an empirical dose may have contributed to long-term response and slowed disease progression, given that almost half of the patients, despite 75% being classified as high risk of recurrence or persistence, showed a partial reduction in TG after empirical RIT.

On the other hand, the untreated group was associated with stability of suppressed TG, which can be associated with the initial low risk of persistence or recurrence, and the absence of additional treatment during follow-up (3). Despite not receiving additional treatment, almost half of the patients showed stability in TG, and an increase in 21.77%.

In both groups, there was a percentage of patients in whom TG became negative. Serum TG may remain detectable after initial treatment and later disappear without any further interventions (26), as demonstrated in 14.52% of our patients who did not receive empirical RIT or any other complementary treatments.

In this regard, we found that patients in the untreated group took longer to develop late metastases compared to those who received empirical RIT, as well as having a greater disease-free survival. However, mortality due to the tumor did not differ between the two groups. These findings may be related to the administration of the empirical dose in patients who characteristically had higher TG levels and a greater prevalence of treatment refractoriness, factors associated with worse outcomes and treatment responses.

High or repeated doses of radioiodine increase the risk of unwanted effects on normal tissues (salivary and lacrimal glands, bone marrow, and lungs). Reproductive effects and the need for family planning in women of childbearing age are recognized concerns. Therefore, individualizing the indication and dose used is essential to reduce risks, ensuring that the benefits always outweigh the risks and avoiding excessive treatments.

The main limitation of our study was the retrospective nature, as causality cannot be determined. The long follow-up period, the single specialized service for thyroid cancer treatment allowed for higher homogeneity in the indication of the empirical RIT (reducing selection bias), and the presence of a group of untreated patients are strengths of the study.

In conclusion, despite similar histopathological tumor characteristics, the outcomes differed between patients who received or not empirical RIT, highlighting the importance of individualizing this indication. The group empirically treated with radioiodine was associated with a high risk of recurrence or persistence of DTC at diagnosis, higher serum TG levels, and greater prevalence of treatment refractoriness compared to the untreated group, justifying the choice of performing empirical RIT in this specific group. Despite having features of greater tumor aggressiveness, nearly half of the patients who received empirical RIT showed a partial reduction in TG after the treatment. The empirical RIT was able to identify metastatic foci, indicating a potential therapeutic benefit.

Therefore, we cannot exclude the possibility that the empirical RIT may have contributed to long-term response and slowed disease progression. This is supported by the absence of a difference in response to treatment between the groups, with the effect of follow-up time being independent of the group.

Further studies are needed to validate the results presented.

Declaration of interest

The authors declare that they have no conflict of interest.

Funding

DEZ-W had a National Council of Technological and Scientific Development Scholarship (CNPq) (303068/2021-3). This work did not receive any specific grant from any funding agency in the public, commercial, or non-profit sector.

Author contribution statement

FFG, LVMA, ECSCE, and DEZ-W conceived the study. FFG collected the data. DEZ-W supervised the study. FFG and DEZ-W analyzed the data and wrote the paper. FFG, ECSCE, and DEZ-W revised the paper.