Abstract
Background:
According to American Thyroid Association (ATA), all patients with papillary thyroid carcinoma (PTC) should initially be classified regarding the risk of tumor recurrence. If a very high postoperative thyroglobulin (Tg) classifies patients as high risk of recurrence, it is reasonable to hypothesize that, at the other extreme, a low Tg may reclassify patients from intermediate to low risk. The objective of this study was to evaluate the rate of persistent/recurrent disease in intermediate-risk patients with low postoperative (before 131I) Tg, thereby evaluating whether these patients can be reclassified as low risk based on this finding already at the time of initial therapy rather than one to two years after radioiodine therapy.
Methods:
A total of 181 patients with the following characteristics were evaluated: (i) diagnosis of PTC; (ii) submitted to total thyroidectomy with lymph node dissection in the case of a suspicion of metastases based on preoperative ultrasonography (US) or perioperative evaluation (cN1); (iii) apparently without persistent tumor after surgery; (iv) a postoperative stimulated Tg (sTg) ≤2 ng/mL and negative anti-Tg antibodies (TgAb); and (v) considered to be at intermediate risk by ATA criteria.
Results:
When evaluated 9–12 months after radioiodine therapy, 170 patients (94%) had a sTg <1 ng/mL and negative TgAb and a negative neck US (excellent response). Ten patients (5.5%) had a sTg >1 ng/mL (≤2 ng/mL) in the absence of apparent disease detected by imaging methods (indeterminate response). US detected cervical lymph node metastases (not detected at the time of 131I therapy) in one patient (incomplete structural response). During follow-up, recurrence was observed in 5/180 (2.7%) patients without apparent disease in the initial assessment. Thus, persistent or recurrent disease occurred in only 3.3% of cases. The 175 patients without tumor persistence/recurrence were not submitted to any additional therapy, and all had a Tg/thryoxine <1 ng/mL in the last assessment. There was no case of death related to the disease.
Conclusions:
This study suggests that intermediate-risk patients (according to ATA criteria) with a postoperative sTg ≤2 ng/mL can be reclassified as low risk already at the time of initial therapy.
Introduction
T
A reclassification already at the time of initial therapy has several implications. For example, until an “excellent response” to initial therapy is documented, thyrotropin (TSH) suppression (<0.1 mIU/L) is recommended for intermediate-risk patients, but is not necessary for low-risk patients (1,7,8). Also, after radioiodine therapy, among patients with negative basal Tg, TgAb, and neck ultrasonography (US), determination of a stimulated Tg (sTg; with or without diagnostic WBS) remains recommended for the intermediate-risk group, but is not necessary in patients who are at low risk (8 –12). It should be noted that 25–60% of patients with PTC have a low Tg (without TgAb) after total thyroidectomy, even in the presence of histological data that suggest a higher risk of tumor persistence/recurrence (13 –17), demonstrating that a significant number of patients could probably be reclassified from intermediate to low risk.
In order to validate the reclassification at initial therapy, it is necessary to know the negative predictive value of the postoperative Tg specifically in the intermediate-risk group. Several previous studies that evaluated the predictive value of postoperative stimulated Tg levels included unselected patients, many of them low risk. As a consequence, the high negative predictive value reported for a low Tg may be lower if only intermediate-risk patients were studied. Indeed, in several series, not only serum Tg but findings such as lymph node metastases, extrathyroidal extension, and aggressive histology, which classify the patient as intermediate risk (1,2), were also predictors of recurrence (13 –16,18,19). Furthermore, although different classifications have been validated, considering its widespread use, it appears ideal to use the American Thyroid Association (ATA) risk classification (1,2) rather than extrapolating results obtained with other systems.
The objective of the present study was to evaluate the rate of tumor persistence/recurrence specifically in patients at intermediate risk according to ATA criteria (1,2) who had a low postoperative Tg (before 131I), thereby evaluating whether these patients can be reclassified as low risk based on this finding already at the time of initial therapy rather than one to two years after radioiodine therapy.
Patients and Methods
Patients
Patients with the following characteristics were first selected: (i) diagnosis of PTC; (ii) submitted to total thyroidectomy with lymph node dissection in the case of a suspicion of metastases based on preoperative US or perioperative evaluation (clinical N1 [cN1]); (iii) apparently complete tumor resection; and (iv) a sTg ≤2 ng/mL and negative TgAb before the administration of 131I. High-risk patients, that is, patients with tumors with extensive extrathyroidal invasion (pT4), extensive lymph node involvement (20), or distant metastases (M1), were excluded. Finally, only patients classified as intermediate risk according to ATA criteria (1,2) were included, that is, patients with at least one of the following findings: (i) vascular invasion, (ii) clinically apparent lymph node metastases (cN1), (iii) invasion of the tumor into the perithyroidal soft tissues (pT3), and (iv) aggressive histological subtype. Although considered to be of intermediate risk (1,2), the following patients were excluded: patients with PTC with minimal extrathyroidal invasion (pT3) when the tumor was ≤2 cm, classical, without vascular invasion, and without clinically apparent lymph node metastases (cN0) (9); and patients with PTC of the aggressive histological subtype when it was a microcarcinoma restricted to the thyroid.
Adjuvant 131I therapy
Approximately four months after thyroidectomy, all patients were submitted to adjuvant therapy with 131I (1.1–5.5 GBq [30–150 mCi]) after withdrawal of levothyroxine (L-T4) for four weeks and use of a low-iodine diet 10–14 days before therapy. A TSH >30 mIU/L prior to therapy was confirmed in all patients. Tg and TgAb were measured immediately before 131I administration. Anterior and posterior whole-body images were obtained seven days after 131I (RxWBS), and the scans were analyzed by experienced nuclear medicine professionals.
Initial evaluation after 131I therapy
The patients were evaluated 9–12 months after initial therapy by sTg and TgAb measurements and neck US. Diagnostic WBS (DxWBS) combined with sTg was obtained in most patients.
Late follow-up
Patients without apparent disease at the initial assessment were followed up by annual measurement of Tg during L-T4 therapy (Tg/T4) and TgAb and neck US. Imaging methods other than US (chest and mediastinal CT, FDG-PET/CT, and RxWBS) were performed when Tg/T4 converted to levels >1 ng/mL. Patients without apparent disease were maintained on an L-T4 dose sufficient to obtain a serum TSH of 0.1–0.5 mIU/L. Diagnosis of a tumor in lesions detected by the imaging methods was made by cytology or histology, and/or unequivocal ectopic uptake (excluding false-positive results) on RxWBS or FDG-PET/CT.
The study was approved by the Research Ethics Committee at the authors' institution.
Assays
Chemiluminescent assays were used for the measurement of Tg (Access Thyroglobulin Assay, Beckman Coulter, Fullerton, CA [functional sensitivity of 0.1 ng/mL]) and TgAb (Immulite 2000, Diagnostic Products Corporation, Los Angeles, CA [reference value of up to 40 IU/mL] or ARCHITECT Anti-Tg, Abbott Laboratories, IL [reference value of up to 4.11 IU/mL]).
Imaging methods
US was performed with a linear multifrequency transducer for morphological analysis (B-mode) and for power Doppler evaluation. US was defined as negative when it did not reveal suspicious lesions (21,22) or, if lesions were present, when cytology and Tg measurement in the needle washout fluid obtained by US-guided fine-needle aspiration were negative. DxWBS was performed with a tracer activity of 131I (185 MBq) after L-T4 withdrawal for four weeks or stimulation with recombinant human TSH (rhTSH). The patients received instructions regarding limiting exposure to environmental iodine and using a low-iodine diet for 10 days before 131I administration. Anterior and posterior whole-body images were obtained three days after 131I administration. Chest and mediastinal CT was performed with 5 mm sequential sections. FDG-PET/CT was carried out after stimulation with rhTSH.
Statistical analysis
The chi-square test and multivariate logistic regression were used, and a p-value of <0.05 was considered significant.
Results
Characteristics of the patients
A total of 181 patients were studied, and their characteristics are shown in Table 1. According to the risk classification of the Brazilian Society of Endocrinology and Metabolism (9), British Thyroid Association (10), and ATA (1,2), the patients were at intermediate risk (selection criterion). According to the risk classification of the European Thyroid Association (2), Latin American Thyroid Society (23), and European Society for Medical Oncology (24), the patients were classified as high risk. According to the American Joint Committee on Cancer (25), 54.1% of the patients were at high risk for mortality (stage III or IVA).
With vascular invasion.
All cN1.
With vascular invasion or aggressive subtype.
Extrathyroid invasion with primary tumor >2 cm or aggressive subtype or vascular invasion.
cN1, clinical lymph node metastases.
Initial assessment after surgery and radioiodine therapy
When evaluated 9–12 months after radioiodine therapy, 170 patients (94%) had a sTg <1 ng/mL and negative TgAb and neck US (excellent response). Ten patients (5.5%) had a sTg >1 ng/mL (≤2 ng/mL after rhTSH) in the absence of apparent disease detected by imaging methods (indeterminate response). US detected cervical lymph node metastases (not detected at the time of 131I therapy) in one patient (incomplete structural response). DxWBS was available for 140 patients. None of the patients had a positive DxWBS (abnormal foci of uptake outside the thyroid bed or uptake in the thyroid bed of >0.5%).
Late follow-up
The time of follow-up ranged from 24 to 96 months (median 60 months). During follow-up, recurrence was observed in 5/180 (2.7%) patients without apparent disease at the initial assessment (lymph node metastases in 4 [detected by US in 3 and by FDG-PET/CT in 1] and pulmonary metastases in 1 [detected by CT and new RxWBS]). Tumor recurrence was observed in 4/170 (2.3%) patients with an excellent response and in 1/10 with an indeterminate response in the first year after radioiodine therapy.
Considering the patient in whom apparent disease persisted 9–12 months after radioiodine therapy and the five patients with subsequent recurrence, tumors were diagnosed in only 3.3% of cases.
Last assessment
The 175 patients without tumor persistence/recurrence were not submitted to any additional therapy. All patients had a Tg/T4 <1 ng/mL in the last assessment (154 had an undetectable Tg/T4, 12 had a detectable Tg/T4 ≤0.27 ng/mL (26), 6 had a Tg/T4 between 0.3 and 0.5 ng/mL, and 3 had a Tg/T4 between 0.5 and 1 ng/mL). None of these nine patients with a Tg/T4 >0.27 ng/mL showed an increase in Tg concentration.
In the last assessment, three of the six patients with persistent/recurrent disease achieved remission after surgical treatment and/or radioiodine therapy. Two patients continued to present elevated but declining Tg levels in the absence of apparent metastases; metastases were detected by imaging in the others. There was no case of death related to the disease.
Predictive factors of tumor
Specifically in the patients with apparently complete tumor resection, a postoperative sTg ≤2 ng/mL, and RxWBS without distant metastases, none of the variables analyzed (sex, age, tumor size, multicentricity, vascular invasion, histological subtype, lymph node metastases [cN1], and extrathyroid invasion [pT3]) was a predictor of tumor persistence/recurrence.
Discussion
The ATA (1,2) and ETA (2) already include a postoperative Tg in the risk classification of recurrence, defining patients with very high Tg, but without findings on RxWBS that would explain these levels, as high risk. It is proposed here that a low postoperative Tg also reclassifies patients from intermediate to low risk. The concept that this reclassification can be done at any time after therapy, as along as low stimulated Tg is achieved, is shared by other authors (27). So far, this reclassification approach has only been tested with Tg levels obtained one to two years after surgery and radioiodine therapy (3 –6,27). In order to validate early reclassification after initial surgery, the present study was conducted to demonstrate that the frequency of tumor persistence/recurrence is similar in intermediate-risk patients with a low postoperative sTg compared with low-risk patients.
It is worth mentioning some of the characteristics of the study. Meeting the objective, specifically patients at intermediate risk who had a postoperative sTg ≤2 ng/mL (without TgAb) were evaluated, and a quite large number of these patients were included. It should be noted that many patients with discrete lymph node involvement (cN0pN1a), who were previously classified as intermediate risk, would now be classified as low risk (2,20). In the present study, the patients were not submitted to elective dissection of the central neck compartment. Therefore, if present, lymph node involvement was clinically apparent (cN1), a feature defining intermediate risk even according to the current ATA classification (1,2,20). None of the patients received a second dose of 131I only because of an elevated Tg, and the Tg concentrations at the last assessment of patients without structural disease therefore reflect the effect of the initial therapy. The frequency of persistent or recurrent tumors found does not seem to be underestimated. First, the investigation protocol followed current guidelines (1,2,8 –10,12,23,24). Second, the time of follow-up was at least two years, with a median of five years, and it is known that three out of four recurrences occur in these first years (26,28,29). Third, the results of the sTg obtained after initial therapy (<2 ng/mL in all patients) and of the basal Tg at the last assessment (<1 ng/mL in all patients and ≤0.27 ng/mL in 95%; no increase in any case) render the occurrence of a relevant number of additional long-term recurrences unlikely in these patients who have not developed structural disease so far (1,10,12,23,24,26).
Previous studies reported persistent structural (tumor detected by imaging methods during initial assessment after initial therapy) or recurrent disease (apparent tumor or Tg elevation during follow-up in patients initially with negative Tg and imaging methods) in 3% (3,4), 7% (15), and 12% (5,6) of low-risk patients treated with total thyroidectomy followed by 131I ablation. In the present study, this frequency was 3.5% among intermediate-risk patients with a postoperative sTg ≤2 ng/mL. In those series, low-risk patients who relapsed and those with persistently elevated Tg or apparent tumor during follow-up after initial therapy corresponded to approximately 10% (4), 15% (3,5,15), and 20% (6). This frequency was about 3% in the present series (five patients with structural recurrence). These comparisons indicate that the evolution of patients initially classified as intermediate risk, but with a postoperative sTg ≤2 ng/mL, seems to be the same as that of patients classified as low risk by ATA criteria. Finally, there was no case of progressive disease or death related to the disease.
Furthermore, the current concept is that once an “excellent response to initial therapy” is achieved, the risk of long-term recurrence becomes low in patients initially classified as intermediate risk. Indeed, this risk was approximately 3% in previous series (3 –5,15), a value similar to the initial probability of persistent or recurrent structural disease found in the present study (3.5%).
Measurement of sTg with or without DxWBS continues to be recommended for intermediate-risk patients with negative basal Tg, TgAb, and US after 131I therapy (8 –12). In the present study, all patients with a postoperative sTg ≤2 ng/mL and no tumor detected by US after initial therapy had a new sTg <2 ng/mL. Consistently, in previous studies, 98.5% of patients with T3 and/or N1 tumors (14,16,30) or those at intermediate risk (31,32) with a postoperative sTg <1–2 ng/mL had a sTg <1–2 ng/mL on repetition after radioiodine therapy. Hence, as recommended for low-risk patients with negative basal Tg and US (8 –12), sTg measurement would not be necessary in these intermediate-risk patients who already have a low postoperative sTg.
In the case of small nonmetastatic tumors with nonaggressive histology, which were adequately resected, minimal extrathyroidal extension (ETE) does not seem to have an impact on the prognosis (9,33 –35), and this finding alone is not an indication for ablation with 131I (1,9,33 –36). There is no agreement that, in the absence of other risk factors (ETE, vascular invasion, metastases, incomplete tumor resection), the histological subtype of PTC impacts the prognosis, especially in the case of microcarcinomas (37 –41). There is no consensus whether ablation with 131I is indicated in the uncommon cases of small and papillary thyroid cancers restricted to the gland with an aggressive subtype and without evidence of persistent disease after surgery (34). Hence, although considered to be of intermediate risk (1,2), patients with PTC with minimal ETE were excluded if the tumor was ≤2 cm, classical histology, without vascular invasion, and cN0. Similarly, intrathyroidal microcarcinomas were excluded if they showed an aggressive subtype. If these patients had been included, the frequency of persistent or recurrent disease would be the same or even lower than that observed here, and their inclusion would therefore not alter the conclusion of the study.
In the same way patients with a very high postoperative Tg are considered to be high risk (1,2), the results of the present study suggest that patients at intermediate risk can be reclassified as low risk already at the time of initial therapy when the postoperative sTg is ≤2 ng/mL.
Footnotes
Author Disclosure Statement
The authors declare that no competing financial interests exist.
