Abstract
Background:
Patients with differentiated thyroid carcinoma are submitted to a first assessment several months after initial therapy to evaluate their response to treatment. At that assessment, measurement of basal thyroglobulin (Tg) and antithyroglobulin antibodies (TgAb) and neck ultrasonography (US) are recommended. Serum Tg may be falsely negative in the presence of TgAb, and the management of patients with negative Tg but positive TgAb represents a challenge. The objective of this study was to correlate the variation in TgAb concentrations (comparison before and after ablation with 131I) with the risk of structural disease.
Methods:
The sample consisted of 116 low- or intermediate-risk patients who had undetectable Tg, negative US at initial assessment, and positive TgAb 8–12 months after thyroidectomy and ablation with 131I.
Results:
Comparison of TgAb concentrations before and after ablation with 131I showed a reduction of >50% in 56 patients (group A), a reduction of <50% in 35 patients (group B), and an increase in 25 patients (group C). Metastases were detected in 5/116 (4.3%) patients during initial assessment (lymph nodes in two, pulmonary in two, and bone in one). They were diagnosed in 0/56, 2/35 (5.7%), and 3/25 (12%) patients of groups A, B, and C, respectively. During follow-up, metastases were detected in 7/111 (6.3%) patients (lymph nodes in six, and pulmonary in one). They occurred in 1/56 (1.8%), 3/33 (9%), and 3/22 (13.6%) patients of groups A, B, and C, respectively. Thus, structural disease was found in 1.8%, 14.3%, and 24% of groups A, B, and C, respectively. This rate was 4% in low-risk patients and 15.4% in intermediate-risk patients.
Conclusions:
Among patients with undetectable Tg, negative US, and positive TgAb after ablation with 131I, the frequency of structural disease was <5% in patients, with >50% reduction in TgAb. Among patients without a significant reduction in TgAb, approximately 10% of low-risk patients and >20% of intermediate-risk patients had structural disease. These results help define the indication for imaging methods other than US during initial assessment and long-term follow-up.
Introduction
P
Greater controversy exists in the case of patients at low or intermediate risk with undetectable Tg and negative US but positive TgAb during initial assessment after ablation with 131I. The use of diagnostic whole-body scanning (DxWBS) on that occasion is questioned by some authors (11), others consider performing it only if there is no reduction in TgAb (5,9,13), and many recommend it for all patients with persistent TgAb after ablation with 131I (6,8,12). Some authors consider other imaging methods unnecessary when a DxWBS is obtained and this exam and US do not reveal disease (12,13). However, others recommend to extend the investigation using chest computed tomography (CT) and fluorodeoxyglucose positron emission tomography (FDG-PET) scans if TgAb concentrations are not reduced (6,9), and even empirical 131I therapy if TgAb are increased (8). These considerations indicate that a consensus is lacking.
Management based on the trend of TgAb levels appears to be the most common recommendation today. Comparison with previous concentrations (rather than postponing and comparing them with subsequent measurements) has the advantage of estimating the risk of disease early. Thus, already on the first occasion after ablation with 131I, some patients could be spared unnecessary investigation, others should be submitted to more extensive investigation for the early detection of persistent tumors, and thyrotropin (TSH) suppression can already be individualized (10,14). Serum measurement of Tg and TgAb are not recommended before surgery (5,10,13,14). The first time point recommended for their measurement is a few weeks after thyroidectomy before the administration of radioiodine (5,10,13,14). Hence, in current practice, TgAb concentrations obtained on that occasion serve as baseline for subsequent comparisons. Considering the serum half-life of TgAb (15,16), this interval between ablation with 131I and initial assessment to evaluate the response to therapy (≥6 months) is adequate.
To the authors' knowledge, comparison of TgAb immediately before versus during initial assessment after ablation with 131I has been done in only one study involving 52 patients (17). In that study (17), a DxWBS was first obtained in all patients, and only those without ectopic uptake or uptake in the thyroid bed were included; US was not performed during initial assessment. Thus, the results of that study (17) cannot be fully extrapolated to current practice. In other series (18 –21), TgAb were measured on different occasions, and extrapolation of the results to the situation in question may be inadequate. Another demonstration that this comparison (TgAb immediately before vs. during initial assessment after ablation with 131I) needs to be better investigated is the management of patients whose TgAb concentrations are not increased, but are also not significantly reduced. Many authors recommend the same management as for patients with elevated TgAb levels (5,6,9,13). In contrast, others recommend the same management as for patients who show a significant reduction in TgAb levels (8,10,14,22).
In view of the lack of consensus and the importance of the topic, considering that up to one-fourth of patients with DTC have TgAb, all authors recognize the need for further studies that include a larger number of patients (6,9,17).
The present study included patients at low and intermediate risk according to the American Thyroid Association (ATA) guidelines (10,14), that is, the group of patients whose management is the most controversial. To permit application of the results to current practice, patients with undetectable Tg, negative US, and positive TgAb (tests that are always recommended) during initial assessment after ablation with 131I were included. TgAb concentrations were compared to those obtained several weeks after thyroidectomy before the administration of radioiodine. The objective was to correlate the variation in TgAb concentrations with the risk of structural disease.
Patients and Methods
Design
This was a prospective study. The study was approved by the Research Ethics Committee of the authors' Institution.
Patients
First, patients with DTC submitted to total thyroidectomy (and lymph node dissection if pre- or perioperative evaluation showed apparent involvement [cN1]), with apparently complete tumor resection, followed by ablation with 131I (1.1 GBq for T1–T2Nx tumors with nonaggressive histology; 3.7 GBq for T1–T2Nx tumors with aggressive histology and T3Nx or T1–T2N1 tumors; and 5.5 GBq for T3N1 tumors) were included. Patients with papillary microcarcinomas restricted to the thyroid, patients with the noninvasive encapsulated follicular variant of papillary carcinoma (without indication for ablation with 131I) (23), and patients classified as high risk according to the ATA (10,14) were excluded. Serum Tg and TgAb were measured immediately before the administration of radioiodine, and post-therapy whole-body scanning (RxWBS) was obtained 7 days after treatment. Patients whose RxWBS revealed unequivocal ectopic uptake (metastases) were excluded. In addition to clinical examination, serum Tg and TgAb and US were obtained 8–12 months (M = 10 months) after ablation with 131I. The group of interest consisted of patients classified as low or intermediate risk according to the ATA guidelines (10,14) whose clinical examination showed no abnormalities and who had a Tg <1 ng/mL and positive TgAb (24). Finally, for this study, patients who had apparent disease on US were excluded (Table 1).
Group B, reduction <50% in TgAb concentration; group C, increase in TgAb concentration (6,7,9,13,17,19).
Diameter of the largest metastatic LN.
F, female; M, male; PTC, papillary thyroid cancer; TgAb, antithyroglobulin antibodies; LN, lymph node.
Groups
The concentrations of TgAb 8–12 months after initial therapy were compared to those obtained immediately before ablation with 131I. The patients were divided into three groups according to the variation observed: a reduction >50% (group A), a reduction <50% (group B), and an increase (group C) (6,7,9,13,17,19). Based on the interassay precision of the method (25), an increase of ≥10% in TgAb concentrations was required for inclusion in group C.
Protocol
Stimulated Tg (sTg) and DxWBS were obtained from all patients during initial assessment (24). Patients with a sTg >1 ng/mL without disease on DxWBS were evaluated by chest and mediastinal CT, technetium-99m-methoxyisobutylisonitrile (99mTc-MIBI) scintigraphy, and FDG-PET/CT (24). In the case of patients showing no disease on this first assessment, nonstimulated Tg, TgAb, and US were obtained at intervals of 6 months. In addition to US at intervals of 6 months, chest CT was performed annually in patients with tumors >4 cm or at intermediate risk, and every 2 years in patients at low risk with tumors ≤4 cm, while the nonstimulated Tg remained <1 ng/mL and TgAb continued to be positive. If a TgAb elevation or a nonstimulated Tg >1 ng/mL was observed at any time during follow-up, chest CT, 99mTc-MIBI scans, and FDG-PET/CT were performed. TSH was maintained ≤0.5 mIU/L. The time of follow-up ranged from 60 to 140 months (Mdn = 96 months).
Tg and TgAb measurement
Immediately before and 8–12 months after ablation with 131I, serum Tg was measured by a radioimmunometric assay (ELSAhTG, CIS Bio International), with a functional sensitivity of 1 ng/mL, and TgAb were determined by a chemiluminescent assay (Immulite, Diagnostic Products Corp.), with a reference value of up to 40 IU/mL.
Imaging methods
WBS was performed with a tracer (185 MBq) or therapeutic (1.1–5.5 GBq) activity of 131I and a low-iodine diet during the 10 days preceding radioiodine administration. Anterior and posterior whole-body images were obtained 3 (DxWBS) or 7 (RxWBS) days after 131I administration. US was performed with a linear multifrequency transducer for morphological analysis (B-mode) and for power Doppler evaluation. All suspected lesions apparent on US scans (26,27) were evaluated by US-guided fine-needle aspiration biopsy. Chest and mediastinal CT was performed on 5-mm thick sequential sections. 99mTc-MIBI scans were performed during levothyroxine therapy using a tracer dose of 720–925 MBq, and whole-body images were obtained during the early (20 min) and late period (6 h). FDG-PET/CT was carried out after stimulation with recombinant human TSH.
The diagnosis of a tumor in lesions detected by the imaging methods was made by cytology or histology, and/or based on unequivocal ectopic uptake (excluding false-positive results) on DxWBS, RxWBS, or FDG-PET/CT.
Statistical analysis
Means were compared between groups by the Student's t-test or the nonparametric Mann–Whitney U-test. Fisher's exact test or chi-square test was used to detect differences in the proportion of cases. A p-value of <0.05 was considered significant.
Results
Characteristics of the patients
The sample consisted of 116 patients (106 female), ranging in age from 16 to 74 years (Mdn = 49 years); 109 patients had papillary carcinoma, and seven had follicular carcinoma. According to the initial risk classification outlined in the ATA guidelines (10,14), 51 patients were low risk and 65 were intermediate risk (inclusion criterion). According to the American Joint Committee on Cancer (28), 65 patients were stage I, 12 were stage II, 25 were stage III, and 14 were stage IVA. The 131I activity administered was 1.1 GBq in 43 patients, 3.7 GBq in 56 patients, and 5.5 GBq in 17 patients.
TgAb
The concentrations of TgAb 8–12 months after initial therapy ranged from 65 to 2845 IU/mL. When compared to the concentrations obtained before 131I therapy, 56 patients exhibited a reduction >50% (group A), 35 exhibited a reduction <50% (group B), and 25 exhibited an increase (group C). These groups were similar in terms of sex, age, histology, initial risk classification, tumor stage, and 131I activity received (Table 2).
Structural disease upon assessment 8–12 months after initial therapy
Metastases were detected in 5/116 patients (4.3%) during initial assessment. DxWBS revealed ectopic uptake in two patients, and CT showed lymph nodes (the largest with 15 and 18 mm, respectively) in the topography of ectopic uptake. These lymph node metastases not detected by neck US were mediastinal in one patient and cervical (level VI) in the other. In another patient, DxWBS showed pulmonary metastases, but a chest CT was normal. The DxWBS was negative in two patients, but pulmonary metastases were detected by chest CT in one and bone metastases were detected by FDG-PET/CT in the other. A tumor was detected in 0/51 low-risk patients and in 5/65 intermediate-risk patients (7.7%). Considering the variation in TgAb concentrations, metastases were diagnosed in 0/56, 2/35 (5.7%), and 3/25 (12%) patients of groups A, B and C, respectively (Table 3).
Structural disease in patients without apparent tumor in the first year after initial therapy
Metastases were detected in 7/111 (6.3%) patients during long-term follow-up. US showed lymph node metastases in five patients (the largest ranging from 8 to 20 mm), CT revealed pulmonary metastases in one, and FDG-PET/CT was positive in another patient (lymph node metastases; the largest measuring 22 mm). A tumor was detected in 2/51 (4%) low-risk patients and in 5/60 (8.3%) intermediate-risk patients. Considering the variation in TgAb concentrations, metastases occurred in 1/56 (1.8%), 3/33 (9%), and 3/22 (13.6%) patients of groups A, B, and C, respectively (Table 3).
Structural disease upon initial assessment and during long-term follow-up
An apparent tumor was observed in 4% of low-risk patients and in 15.4% of intermediate-risk patients (p = 0.03). This frequency was 1.8%, 14.3%, and 24% in groups A, B, and C, respectively (A vs. B: p = 0.03; A vs. C: p = 0.003; and B vs. C: not significant). Considering the 12 patients with and the 104 patients without disease, TgAb elevation (group C) showed a sensitivity of 50% and a specificity of 81.7%. A nonsignificant reduction in TgAb (groups B and C) exhibited a sensitivity of 91.6% and a specificity of 52.9%.
Table 3 shows the structural disease rates according to the initial ATA risk classification scheme combined with the variation in TgAb concentrations (before and after ablation with 131I). To demonstrate the impact of these results, if they were considered to define investigation in patients with TgAb, undetectable basal Tg and negative US after ablation with 131I, 56 (group A) or 81 (group A and groups B and C of low risk) patients would have been spared additional imaging (such as DxWBS). Conversely, the use of imaging methods other than US led to the detection of metastases in five (distant metastases in 3) of 35 patients (groups B and C of intermediate risk), that is, one patient with structural disease was identified per seven patients investigated.
Additional information
There was no significant difference in TgAb concentrations 8–12 months after initial therapy between patients with and without structural disease, including those with apparent tumors on initial assessment and tumors detected during late follow-up.
Of note, in patients of groups B and C without apparent tumor (n = 49), DxWBS did not reveal visible uptake in the neck, indicating total thyroid ablation.
Among the 12 patients with structural disease detected at some point during follow-up who were treated (surgery, 131I, or external radiotherapy) and maintained under TSH suppression, structural disease persisted in six, and the other six no longer exhibited apparent disease on the imaging methods in the last assessment. In the former patients, TgAb remained positive in all, while four of the latter patients already exhibited negative TgAb (with basal Tg <1 ng/mL).
Discussion
This was a prospective study that included a large number of patients, which was higher than that of previous series (16 –18,20,29). The design was consistent with current clinical practice. TgAb concentrations 8–12 months after initial therapy were compared to those obtained after thyroidectomy and before ablation with 131I, which is typically the first time they are measured (5,10,13,14). Undetectable basal Tg and negative US were inclusion criteria, since they should initially be obtained in all patients. In the only study that performed a similar comparison (17), DxWBS was first obtained, and only patients without ectopic uptake or uptake in the thyroid bed were included. Positive findings on a DxWBS can be present in patients with TgAb and indicate tumor persistence (18,24,29,30). Furthermore, US was not obtained during initial assessment in that series (17). In the present study, patients with microcarcinomas restricted to the thyroid or high-risk patients were excluded, since their management is less controversial. Additionally, the patients included in the study were divided based not only on TgAb variation, but also on the risk of recurrence according to the ATA guidelines (10,14). Indeed, a difference in the rate of persistent or recurrent disease was observed between low- and intermediate-risk patients. Finally, the time of follow-up was ≥5 years in all patients (Mdn = 9 years), and it is known that more than three-quarters of recurrences occur in these first years (17,31,32). In the present series, 91% of the patients were women. This female predominance in the group of patients with positive TgAb at a frequency that is even higher than that observed in the population with DTC has been demonstrated in all studies, ranging from 85% to 95% (15 –21).
A significant reduction in TgAb is rarely observed in patients with persistent or recurrent tumors. In the present study, comparing TgAb concentrations immediately before and 8–12 months after ablation with 131I, a reduction of >50% was not observed in any of the patients with persistent structural disease on initial assessment and in only one with disease detected during long-term follow-up. Making the same comparison, Kim et al. (17) found no TgAb reduction of >50% in patients who relapsed. In subsequent measurements, a reduction of >50% was not observed in patients with TgAb in the first year after thyroidectomy alone who relapsed (21). Sustained reduction has also been reported in only 10% of patients with TgAb one year after thyroidectomy followed by ablation with 131I who progressed to apparent disease (20). Thus, the risk of apparent structural disease is low in patients with undetectable Tg and negative US, but with positive TgAb after initial therapy, when TgAb concentrations are significantly decreasing.
In patients without persistence of normal thyroid tissue or tumor, a TgAb elevation is not expected. However, 18.3% of patients in the current study without an apparent tumor or thyroid remnants exhibited an increase in TgAb concentration 8–12 months after ablation with 131I. This percentage was 26% in a previous study (17). Transient increases in TgAb concentration can occur after 131I therapy (15), and a prolonged period may be necessary for their return to baseline concentrations (15,16). However, this would not explain the increase of TgAb concentrations reported after initial therapy in 6–15% of patients without apparent disease who were not submitted to any intervention (surgery, radioiodine, or external radiotherapy) between measurements (18,20,21). In these cases, a TgAb elevation may have preceded late recurrences not detected until the end of the study, or represent autoimmune dysregulation, with the production of antibodies irrespective of the presence of the antigen (in this case, Tg).
In the two studies in which TgAb were measured immediately before and several months after ablation with 131I, a less intense reduction was observed in 30% (17) and 41.6% (present series) of the patients with apparent tumor, but also in 28% of patients who were free of disease (17, present series). Even in the absence of normal thyroid tissue or tumor, many months may be necessary for a reduction of >50% in TgAb concentrations in some patients (15,16), especially those who exhibited a transient increase in TgAb after therapy with 131I (15). Disease may not be detected, even when TgAb concentrations do not decrease after years of follow-up (18,20). Nevertheless, patients without a significant reduction in TgAb concentration have a higher risk of disease than those exhibiting an important reduction (14.3% vs. 1.8% in this series, and 19% vs. 0% in Kim et al.) (17). A less significant difference was observed when compared to patients exhibiting TgAb elevations (14.3% vs. 24% in this series, and 19% vs. 37% in Kim et al.) (17). It therefore does not seem adequate to manage patients exhibiting only a mild TgAb reduction in the same way as patients with a significant reduction in TgAb levels, as recommended by some authors (8,10,14,22).
One limitation of this study is the fact that although there was no change in the follow-up protocol (see Methods), the Tg and TgAb assays were changed during late follow-up. It is therefore not possible to evaluate the behavior of TgAb in subsequent measurements (>1 year after initial therapy) as a predictor of late recurrence in patients without persistent tumors on initial assessment. This fact does not compromise the objective of the study, which was to correlate the variation in TgAb concentrations before and 8–12 months after ablation with 131I (the same assays were used for these measurements in individual patients) with the risk of structural disease. It should be emphasized that this comparison provides an early definition (already at the first assessment after initial therapy) of the risk of disease and was made in only one previous study (17) that included a smaller number of patients and selection criteria not fully concordant with current practice (as discussed earlier). In contrast, the comparison of TgAb concentrations in the first year after initial therapy with subsequent measurements as a predictor of recurrence has been studied more extensively (1,18,20,21).
In conclusion, in patients with persistent TgAb levels several months after ablation with 131I, who have undetectable Tg and negative US, the ATA risk classification combined with the comparison of TgAb concentrations before radioiodine therapy is a predictor of structural disease. Less than 5% of patients with a >50% reduction in TgAb had structural disease. Among patients without a significant reduction in TgAb, approximately 10% of low-risk patients and >20% of intermediate-risk patients had apparent structural disease. These rates help define the need for ancillary studies: some patients could be spared unnecessary imaging studies, while others should be submitted to more extensive evaluation for the early detection of persistent disease, a more stringent TSH suppression, and more intense surveillance.
Footnotes
Author Disclosure Statement
The authors declare that no competing financial interests exist.