Low-Dose Radioiodine Ablation in Patients with Low-Risk Differentiated Thyroid Cancer

Abstract

Aim

Based on the response criteria of the 2015 American Thyroid Associations guidelines, our objectives were to ­determine the response rate when using a low dose of ­131-I GBq in patients with low-risk differentiated thyroid cancer (LRDTC) and the influence of clinical and analytical variables on the prediction of complete response.

Methods

We performed a multicentre and longitudinal study, including patients who were operated for LRDTC and who underwent radioiodine remnant ablation with a low-dose of 131-I. All patients were assessed at 6–12 months, and their status was classified as complete (excellent response) or incomplete response (structural incomplete, biochemical incomplete or indeterminate response). Various factors including age, gender, histology, tumour focality and size, stage, time from surgery to treatment, type of thyroid-stimulating hormone (TSH) stimulation, preablation serum thyroglobulin (pTg), antiTg antibodies (pAntiTgAb) and TSH (pTSH) levels were also analysed in order to predict the complete response rate.

Results

Of 108 patients, 79.6$ achieved complete response and the remaining showed incomplete response (2.9, 5.5 and 12$ due to biochemical incomplete, structural incomplete and indeterminate response respectively). Six patients received a new dose of 131-I. Tumour size and pAntiTgAb were the only factors related to therapeutic response (p = 0.03 and p < 0.01, respectively). However, pAntiTgAb was the only independent factor related to complete ­response. Patients with complete response showed lower pTg than those with incomplete response (5.1 ± 12.9 vs. 11.2 ± 25 ng/mL) although without statistical significance (p = 0.14). There was no significant difference in the response rate depending on the thyrotropin stimulation methods.

Conclusions

A low dose of 131-I was sufficient for reaching a complete response at 6–12 months of follow-up in the majority of patients with LRDTC. Tumour size and pAntiTgAb variables were related to therapeutic response.

Introduction

The incidence of thyroid cancer has increased in many countries during the past decade and this is due mainly to an improvement achieved in detection methods [1, 2]. Most cases are of the differentiated thyroid carcinoma and many patients undergo radioiodine remnant ablation (RRA) to eliminate the residual normal thyroid tissue after total thyroidectomy (TT) or near TT (NTT).

Previous works [3, 4, 5] have shown that a low dose of 131-I is as effective as a high dose for ablation of thyroid remnants in low and intermediate risk patients. Additionally, the low dose has clinical advantages such as the reduction of side effects and risk of a second primary cancer, but it also influences positively other extra-clinical factors such as the hospital stay and consequently, its financial costs. However, the rate of ablation success has an important variability, especially due to inconsistent definitions of successful ablation and patient selection criteria. Moreover, it has been observed that some elements might be potential prognostic factors for ablation success.

On the other hand, “the ablation success” has been the main end point for response to RRA. Nevertheless, the disease relapse and disease-specific mortality should be the real endpoints. Thus, different systems have been used to predict the prognosis of patients with thyroid cancer. Recently, the 2015 American Thyroid Associations (ATA) guidelines have added a dynamic risk assessment system, which classifies the patient based on the response to initial therapy (excellent, indeterminate, biochemical incomplete and structural incomplete response) using Preablation thyroglobulin (pTg), preablation AntiTg antibodies (pAntiTgAb), ultrasonography (US), diagnostic whole-body scan, PET/CT, CT and/or MRI. This system is based on a dynamics assessment; as a results, it could potentially modify the initial risk estimates (for example, those with an initial high risk could became low risk patients). Therefore, this dynamic system allows a more personalised approach to treatment, follow-up and prognosis [6]. Thus, our objectives were to determine the response rate based on the ATA dynamic risk assessment system, using a low activity radioiodine ablation in patients with low-risk differentiated thyroid cancer (LRDTC) and second, to assess the influence of clinical factors in the therapeutic response.

Methods

A longitudinal and multicentre study, including 5 hospitals, was performed in accordance with the ethical standards prescribed by the Declaration of Helsinki. Additionally, this study was approved by our institutional Ethics Committee and all patients gave their informed consent for participation in the research study.

Patients

Patients with LRDTC and TT or NTT and both with or without cervical lymph node dissection, referred to our department to receive a low activity ([30 mCi (1.1 GBq)]) for RRA, were consecutively included since November 2013 up to January 2016. Low-risk criteria were defined as follows: (i) pT1a-T3/N0-x/M0 disease stage [7], (ii) macroscopic complete resection of the tumour and (iii) neither aggressive histology (tall cell, columnar cell carcinoma, etc.) nor vascular invasion. Although patients with pT3 initially were included, those with minimal extrathyroidal extension were ultimately excluded for the present analysis. Likewise, patients with suspicion of locoregional or distant metastases on whole body scintigraphy (WBS) after RRA were not included. Patients underwent RRA after thyroid hormone withdrawal (THW) or administration of recombinant human thyroid-stimulating hormone (rhTSH). pTg, pAntiTgAb and preablation TSH (pTSH) levels were measured. Patients were required to have TSH levels > 30 mU/L at the time of RRA.

A WBS was performed 6–8 days after RRA by using a dual-headed-gamma camera equipped with high-energy collimators, obtaining planar images in anterior and posterior views. The WBS speed was 10 cm/min and a 256 × 512 matrix was used. SPECT/CT of the neck, and in some cases, of other anatomical regions, was performed using a 360° orbit (180° for each head), step and shoot mode, at 35 s per view. Immediately after emission image acquisition, a low-dose CT acquisition (120 kV, 2.5 mA, 15 mm slice thickness) was obtained without changing the patient's position. The WBS was classified as negative (absence of cervical uptake) or positive (thyroid bed uptake).

The following variables were collected: age at diagnosis, gender, histology (papillary, follicular or mixture of both), tumour focality (unifocal or multifocal) and laterality (unilateral or bilateral), primary tumour size (T1, T2, T3) and stage grouping (I, II, III) following the TNM 7thed [1], time from surgery to RRA (≤6 months or > 6 months), type of TSH stimulation (THW or rhTSH administration), pTSH, pTg, pAntiTgAb level (positive ≥30 UI/mL) and WBS findings (positive or negative).

Response to Therapy

A response-to-therapy assessment was performed between 6 and 12 months after RRA using 4 categories based on ATA guidelines [6]. For statistical purpose, they were grouped into 2 categories: complete response and incomplete response. Thus, a complete response (an excellent response based on ATA guidelines) was defined as a TSH-stimulated Tiroglobulin (Tg) < 1 ng/mL or suppressed Tg < 0.2 ng/mL in the absence of structural or functional evidence of disease (and in the absence of AntiTgAb).

On the other hand, an incomplete response was defined as one of the following:

• 1. Abnormal suppressed Tg (≥1 ng/mL) and/or stimulated Tg values (≥10 ng/mL) or rising AntiTgAb with negative imaging, which indicated a biochemical incomplete response.

• 2. Persistent or newly identified locoregional or distant metastases on US and/or WBS, defined as structural incomplete response (independently of Tg and AntiTgAb).

• 3. Nonspecific findings on imaging studies (findings that cannot be classified as either benign or malignant), non-stimulated Tg detectable 0.2–1 ng/mL, stimulated Tg detectable 1–10 ng/mL, or AntiTgAb stable or declining in the absence of structural or functional disease, considered indeterminate response.

All patients received levothyroxine to maintain TSH suppressed with an initial TSH goal of 0.1–0.5 mU/L. The evaluation of response to therapy was performed using Tg testing (basal or stimulated) and neck US. In some cases, DxWBS was also performed. The measurement of Tg was performed using immunometric assays calibrated against the international standard CRM 457. The AntiTgAb were assessed in the same sample.

A new dose of 131-I was considered to complete the ablation or treatment of new locoregional or distant metastases.

Statistical Analysis

Quantitative variables were described using the case number, mean and SD. Qualitative variables were reported selecting the case number and percentage. For comparisons of categorical variables between groups, the chi-square was used, as well as the Student t test or Mann-Whitney U test for quantitative variables.

We analysed the influence of all the collected clinical and RRA-related variables in the response using a binary multivariate analysis. It was conducted by using the logistic regression method. The OR and CI were calculated. A p value not exceeding 0.05 was considered statistically significant.

Results

Patients

A total of 120 patients were initially evaluated. Twelve of them were excluded, 5 due to metastasis on WBS, 6 because of low TSH levels before RRA and 1 patient due to lack of WBS. Our study included 108 patients of which 21 men and 87 women. The mean age (range) was 50.0 ± 13.4 (17–82) years. pTSH levels ranged from 31 to 150 mU/L (mean 106 ± 35.5 mU/L) and pTg span was 0.0–72 ng/mL (mean 5.66 ± 14.3 ng/mL). The remaining demographic and clinical features are summarised in Table 1.

Table 1.

Patients' clinical characteristics

Parameters	n (%)
Gender (patients)
Female	87 (80.5)
Male	21 (19.5)
Histology
Papillary	103 (95.3)
Follicular	5 (4.7)
T stage
T1a	56 (52)
T1b	34 (31.4)
T2	13 (12)
T3	5 (4.6)
Stage
I	96 (88.9)
II	8 (7.4)
III	4 (3.7)
Tumor multifocality
Unifocal	68 (63)
Multifocal	40 (37)
Time from surgery to RAI
<6 months	98 (91)
>6 months	10 (9)
Type of TSH stimulation
Thyroid hormone withdrawal	54 (50)
rhTSH administration	54 (50)
pAntiTgAb
Positive	13 (12)
Negative	95 (88)
WBS
Positive	100 (92.6)
Negative	8 (7.4)

RAI, radioiodine remnant ablation; TSH, thyroid-stimulating hormone; rhTSH, recombinant human TSH; pAntiTgAb, preablation serum antiTg antibodies levels; WBS, whole body scintigraphy.

Response to Therapy

Complete response (excellent response) was documented in 86 patients (79.6$). In contrast, 22 patients (20.4$) had incomplete response: 3 due to biochemical incomplete response (2.9$), 6 because of structural incomplete response (5.5$) and 13 with indeterminate response (12$). In the latter group, the criteria were suppressed Tg = 0.5 ng/mL in 1 patient, TSH-stimulated Tg = 2 ng/mL in 1 patient, persistence of positive Anti-TgAb in 6 and nonspecific findings on imaging studies in 5. Three patients with previously positive pAntiTgAb showed a normal level during the follow-up. A total of 6 patients received further dose of 131-I ranging from 1.1 to 4.4 GBq (2 because of an increase in the Tg level, 3 due to persistent thyroid remnant on DxWBS and 1 because of lymph node recurrence).

Predicting Factors of Complete Response

In the univariate analysis, the primary tumour size (χ² = 8.1, p = 0.03) and pAntiTgAb (χ² = 21.7, p < 0.01) were significantly related with the assessment of response. All the results are detailed in Table 2.

Table 2.

Univariate analysis for complete response

Variables	Complete response	In complete response	p value
Age, year
Continuous value	50.4±13.2	48.3±14.3	0.50
Group
<45 years	27	10	0.30
≥45 years	59	12
Gender			1.0
Male	17	4
Female	69	18
Histology			0.58
Papillary	83	20
Follicular	3	2
T stage			0.03
T1a	47	9
T1b	23	11
T2	13	0
T3	3	2
Stage			0.43
I	75	21
II	8	0
III	3	1
Tumor multifocality
Unifocal	55	13	0.80
Multifocal	31	9
Type of TSH stimulation			0.56
THW	47	7
rhTSH administration	39	15
pTSH, mU/L*	105.4±35.2	111.6±37.07	0.46
pTg, ng/mL**	5.1±12.9	11.2±25	0.14
Median	1.6	3.0
Continuous value
Group A			0.56
<1	32	4
≥1	50	9
Group B			0.55
<1	32	4
1–10	40	6
>10	10	3
Group C			0.28
<10	72	10
≥10	10	3
pAcTgAb***			<0.01
Negative	82	13
Positive	4	9
WBS‡			0.08
Positive	82	18
Negative	4	4

^*Preablation thyroid-stimulating hormone.

^**Preablation serum thyroglobulin levels in patients without antiTg antibodies levels.

^***Preablation serum antiTg antibodies levels.

^‡WBS.

TSH, thyroid-stimulating hormone; rhTSH, recombinant human TSH; pTSH, preablation TSH; pTg, preablation thyroglobulin; THW, thyroid hormone withdrawal; WBS, whole body scintigraphy.

In a binary multivariate model for incomplete response, negative pAntiTgAb (OR 0.07, 95$ CI 0.19–0.20, p < 0.01) was identified as the only independent factor related with RRA results.

Discussion

The most important goal of our study was to establish the status of response using low activity radioiodine ablation in patients with LRDTC based on ATA dynamic risk assessment system. We found that a dose of 1.1 GBq was enough to reach a complete response in 79.8$ of patients with LRDTC.

Given that 2015 ATA dynamic risk assessment system is relatively new [6], most of the studies have used the success rate ablation as surrogate clinical end point for response to 131-I [3, 4, 5, 6, 7]. Particularly, 2 large multicentre randomised trials [3, 4] and 1 meta-analysis [5] have shown that low-dose radioiodine (1.1 GBq) is as effective as a high-dose (3.7 GBq) in ablating thyroid remnant in the low and intermediate risk patients. However, some series [3, 4, 8, 9] have reported variable success rates (ranging between 39 and 98$) using a low-dose radioiodine in low/intermediate risk. This can be explained by the variability of inclusion criteria and of success ablation definition. Thus, the definition of success ablation has been based on the combination of different Tg levels, presence or absence of AntiTgAb, WBS findings, and recent US.

Although we used other nomenclature for 131-I response (based on ATA guidelines), our results may be comparable with those of previous studies. Thus, we found that the excellent response represented 79.8$ and was lower than that described by some authors [3, 4], mainly due to differences in definition of response criteria. However, it was higher than those found by Qu et al. [8] and Fallahi et al. [9], possibly because of the exclusion of patients with N1a and N1b stage. Nevertheless, we considered that a single low-dose of 131-I is enough to reach a complete response in a high number of patients with LRDTC.

With regard to 2015 ATA guidelines, the new risk stratification system [6] allows a more personalised approach to treatment, follow-up and prognosis. Thus, based on this system, 86–91$ of patients with low risk reach an excellent response, 11–19$ exhibit a biochemical incomplete response and 12–29$ show an indeterminate response after RRA using different doses [9, 10, 11, 12], showing a general disease-specific death risk of < 1$ [6]. Qu et al. [8], using a low-dose in low and intermediate risk patients reported an excellent response in 80$ and indeterminate response in 20$ of their sample. We found similar results using a low dose in low-risk patients.

On the other hand, we identified that negative pAntiTgAb was an independent predicting factor of ­response after RRA. Our results could be explained for the very low prevalence of positive pAntiTgAb in patients with complete response (4.6$) in comparison with those with incomplete response (40.9$). In the same way, previous studies [13, 14, 15, 16] have focused on the value of the AntiTgAb as risk factor of persistent/recurrent disease, although they referred to levels during the follow-up instead of the previous values to RRA, as was our case. In particular, they have found that the onset or rise of antiTgAb or rising titres imply an increased risk of persistent/recurrent disease. In fact, the 2015 ATA guidelines [6] have used the antiTgAb level as factor of ­response. Additionally, falling levels of antiTgAb could indicate success therapy according to some authors [17, 18]. Spencer et al. [19] suggest that high levels may predict the likelihood of recurrence in patients without Hashimoto thyroiditis. Despite additional factors, as insensitivity and interference from high endogenous Tg concentrations [19, 20, 21], the antiTgAb seem to be a potential tumour marker and predictive response factor, since it may reflect viable malignant thyroid tissue.

We also found that the tumour size was related to response, with better results in microcarcinoma. In fact, due to better prognosis in these patients, the European Association of Nuclear Medicine [22], British Thyroid Association [23] and the most recent ATA guidelines [6] do not recommend RRA in microcarcinoma without risk factors. Thus, there has been an inclination towards managing these patients with a more conservative approach. However, some centres continue using RRA in these patients in order to achieve improved sensitivity of Tg, enabling early recognition of recurrent or metastatic disease. Additionally, there are other risk factors, such as age, tumour multifocality and family history, which could impact the therapeutic approach.

Although pTg did not show statistical significant differences in our study, patients with complete response presented lower pTg than those with incomplete response (5.1 ± 12.9 vs. 11.2 ± 25 ng/mL). Previous studies [24, 25, 26] have showed that stimulated Tg level previous to RRA is a useful factor to predict the presence of persistent and recurrent disease on follow-up with a high negative predictive value using cut-off levels ranging from 1.0 to 18 ng/mL. The meta-analysis by Webb et al. [27] found a 6$ likelihood of having persistent disease in patients with preablation Tg value of less than 10 ng/mL. In the same way, previous studies [28, 29, 30, 31, 32] have showed that preablation-stimulated Tg is a good predictor of successful ablation using different cut-off points.

The type of TSH stimulation in low-risk patients has been a controversial topic in previous literature. Similar to our results, some authors [3, 4, 33, 34, 35] have showed similar success ablations rate with rhTSH and THW. Based on superior short-term quality of life and similar remnant ablation efficacy found in previous studies, ATA considered the rhTSH stimulation an acceptable alternative to THW for achieving remnant ablation, particularly in patients with ATA low-risk and intermediate-risk DTC without extensive lymph node involvement [6]. Likewise, British Thyroid Association [23] recommends the use of rhTSH stimulation for patients with T1–T3, pN0 or Nx or N1, and M0 and R0 (no microscopic residual disease).

Otherwise, some authors [36, 37, 38] have found tumour multifocality as an important prognostic factor for disease recurrence/persistence, although the cohort was different from our study since they included patients with intermediate and high risk. Similar to our results (although with different inclusion criteria), others found no association between multifocality and tumour recurrence [39, 40]. Thus, the prognosis of multifocal nature of the disease remains controversial, although the recently developed ATA guidelines [6] have associated a higher risk of tumour recurrence of 4–6$ with multifocal disease regardless of the tumour size.

Limitations and Advantages

The weaknesses of our research were the retrospective design and the short time of follow-up for establishing response status. Clearly, a period of time between 6 and 12 months is not enough to make generalizations in thyroid cancer; for this reason, we will reassess these patients in other time points during the follow-up. However, despite this, a clear pattern of complete response was found using low dose of 131-I. The measurement of Tg, AntiTgAb and TSH in different centres is apparently the other disadvantage of our study. Nevertheless, all of them were performed according to international standards.

Although previous studies have found that low-dose radioiodine is effective in ablating thyroid remnant in the low and intermediate risk patients, almost all of them have employed the successful ablation as the surrogate clinical end point. However, we used the ATA dynamic risk assessment system for 131-I response, allowing a better approach to prognosis.

Conclusions

A low dose of 131-I was sufficient for reaching a complete response in the majority of patients with LDTBR at 6–12 months after RRA. Tumour size (T) and pAntiTgAb were variables related to therapeutic response, the negative pAntiTgAb being the only independent factor on multivariate analysis. On the contrary, there was no significant difference in response rate in the thyrotropin stimulation method.

Disclosure Statement

The authors declare that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.