Multiple-factor analysis of the first radioactive iodine therapy in post-operative patients with differentiated thyroid cancer for achieving a disease-free status

131I treatment is an important management method for patients with differentiated thyroid cancer (DTC). Unsuccessful 131I ablation drastically affects the prognosis of the patients. This study aimed to analyze potential predictive factors influencing the achievement of a disease-free status following the first 131I therapy. This retrospective review included 315 DTC patients, and multiple factors were analyzed. Tumor size, pathological tumor stage, lymph node (LN) metastasis, distant metastasis, American Thyroid Association recommended risks, pre-ablation thyroglobulin (Tg), and thyroid stimulating hormone (TSH) displayed significant differences between unsuccessful and successful group. Cutoff values of Tg and TSH to predict a successful outcome were 3.525 ng/mL and 99.700 uIU/ml by receiver operating characteristic curves analysis. Binary logistic regression analysis showed that tumor stage T3 or T4, LN metastasis to N1b station, intermediate and high risks, pre-ablation Tg ≥ 3.525 ng/ml and TSH <99.700 μIU/mL were significantly associated with unsuccessful outcomes. Logistic regression equation for achieving a disease-free status could be rendered as: y (successful treatment) = −0.270–0.503 X1 (LN metastasis) −0.236 X2 (Tg) + 0.015 X3 (TSH). This study demonstrated LN metastasis, pre-ablation Tg and TSH were the most powerful predictors for achieving a disease-free status by the first 131I therapy.

In recent years, the incidence of thyroid cancer is rapidly increasing, and about 90% of which is differentiated thyroid cancer (DTC), including papillary thyroid cancer (PTC) and follicular thyroid cancer (FTC) 1 . For the optimal management of DTC, total thyroidectomy, post-operative radioactive iodine ( 131 I) therapy and thyroid stimulating hormone (TSH) suppression treatment are the three most essential methods [2][3][4][5] . Theoretically, 131 I therapy is thought to reduce the recurrence and mortality rate in DTC patients by destroying microscopic residual tumor foci, which also facilitates early detection of recurrence by enhancing the sensitivity and reliability of serum thyroglobulin (Tg) and 131 I whole-body scan (WBS) [5][6][7] . However, there is a lack of consensus among guidelines about the optimal 131 I dose for ablation [3][4][5] , although the most recent meta-analysis favored the empiric 100 mCi 6,7 .
Although 131 I therapy is generally considered safe, some potential complications could happen. For instance, a large dose of 131 I could increase the risk of dysfunction of several body organs, including gastrointestinal, pulmonary, and hematopoietic systems, salivary glands, and gonads 4,5,8 . Hakala et al. 9 even reported that for young patients (< 40 years), the risk of secondary malignancies would dramatically increase. More importantly, the use of 131 I for remnant ablation to reduce the risk of recurrence following thyroidectomy in low risk DTC patients remained controversial 10,11 . On the other hand, recent evidence showed that for low risk DTC patients, low dose of 131 I therapy could successfully achieve complete ablation 12,13 . Therefore, it is imperative that clinicians should utilize individualized and optimized 131 I dose for the management of DTC patients to balance the risks and benefits of this important therapy.
Predictors of DTC prognosis have been investigated by several studies, although the values of different parameters are inconsistent among them. For instance, pre-ablation stimulated Tg is known to be a good predictor of successful ablation in DTC patients 14 . Pre-ablation Tg has also been demonstrated as an important indicator for predicting persistence or recurrence during follow-up [15][16][17] . However, cut-off values of Tg for successful ablation or recurrence prediction were quite different among the above mentioned studies. A second good example is about TSH. Guidelines recommended that DTC patients receiving 131 I therapy should have a serum TSH level of higher than 30 μ IU/mL [2][3][4][5] . However, Vrachimis et al. 18 demonstrated that endogenous TSH levels at the time of 131 I ablation did not influence ablation success, recurrence-free survival or differentiated thyroid cancer-related mortality.
Hasbek et al. 19 also found that a high TSH level alone was not a factor for the success of ablation. So, a comprehensive research to assess possible predictive factors would be valuable to the subject.
The goal of this study is to analyze potential predictive values of various clinical factors to achieve a disease-free status by the first 131 I therapy in post-operative DTC patients.

Diagnostic values of Tg and TSH for a successful therapy.
For continuous parameters (pre-ablation Tg and TSH), receiver operating characteristic (ROC) curves were conducted (Fig. 1). A cutoff value of 3.525 ng/ml for pre-ablation Tg could be used to differentiate successful and unsuccessful ablation with a sensitivity of 77.12%, specificity of 74.12%, PPV (positive predictive value) of 64.09%, and negative predictive value (NPV) of 84.39%. A cutoff value of 99.700 μ IU/ml for TSH could distinguish successful and unsuccessful ablation with a sensitivity of 57.63%, specificity of 57.46%, PPV of 44.74% and NPV of 69.33% (Table 2).
Risk assessments for a therapeutic failure. Next, we used a binary logistic regression analysis to identify risk factors associated with therapeutic failure. No risks were found among age, gender, capsular invasion, TNM stage and time interval between surgery and 131 I. In contrast, tumor stage T3 or T4, LN metastasis to lateral cervical region (N1b) station, intermediate and high risks, pre-ablation Tg ≥ 3.525 ng/ml and TSH < 99.700 μ IU/mL were significantly associated with a therapeutic failure (Table 3).

Equation for achieving a disease-free status.
Using an unconditioned logistic regression analysis, LN metastasis, pre-ablation Tg level and pre-ablation TSH level showed significant associations with the disease-free status and were included in the regression model. The equation for achieving a disease-free status by the first 131 I therapy could be rendered as: y (successful therapeutic outcome) = − 0.270-0.503 X 1 (LN metastasis) − 0.236 X 2 (pre-ablation Tg level) + 0.015 X 3 (pre-ablation TSH level).

Discussion
Although a number of studies investigated the predictors of DTC prognosis, inconsistency existed about values of various predictors [14][15][16][17][18][19][20] . The merit of the current study is that it examined all the predictors within a comprehensive statistical framework. This retrospective study demonstrated that LN metastasis, pre-ablation Tg and TSH were the most powerful predictive factors for a successful 131 I ablative outcome. Existence of LN metastasis, high pre-ablation Tg level and low pre-ablation TSH level in patients with DTC were associated with increased risk for therapeutic failure.
The incidence of PTC is usually much higher than FTC in clinic. PTC is prone to LN metastasis at an early stage with no definite relationship with primary lesion size. Even when the primary tumor is ≤ 1 cm, LN metastasis could still happen often in PTC 21,22 . Most scholars believe that the first station of LN metastases is the central region (N1a), and then the second station to the N1b. However, lateral LN metastasis without central LN involvement could also occur in patients with PTC 23 . Several studies showed that for patients (especially age ≥ 45 years), LN metastasis was an indicative factor for high recurrence rate and low survival rate. Compared with N1a station, the 5 year survival rate was significantly lower in N1b station [24][25][26] . Moreover, previous studies 20,27,28 , including one case report from our institution 22 , found that extensive metastases would affect the overall therapeutic effectiveness of 131 I. In our study, after multiple-factor analysis, we found that compared with DTC patients without LN metastasis (N0), no significant difference of disease-free status rate was achieved in the N1a patients [odds ratio (OR) = 1.406, P = 0.185]. However, significantly higher risk of unsuccessful therapeutic outcome was found in N1b patients (OR = 4.783, P < 0.001). In the logistic regression analysis, LN metastasis had a coefficient of − 0.503, which indicated positive LN metastasis would increase the risk of therapeutic failure. Moreover, in our study, 14 DTC patients had distant metastases, none of them achieved a disease-free status by the first 131 I therapy.
The clinical value of Tg has been reported in many studies, especially with regard to DTC disease progression or recurrence [14][15][16][17]29 . Some studies have shown the relationship between pre-ablation Tg and successful ablation. For example, Lim et al. 29 analyzed various predictors for successful ablation and disease-free status using univariate and multivariate analyses, and determined that a Tg value greater than 5 ng/mL was the most powerful predictor for ablation failure. A meta-analysis including 3947 patients demonstrated that the pre-ablation Tg was a useful negative predictor for persistent and recurrent DTC, in particular NPV was 94% if pre-ablation Tg level was less than 10 ng/mL 17 . Gonzalez et al. 15 conducted an investigation on 133 DTC patients and revealed that a pre-ablation Tg of less than 8.55 ng/mL could predict remission of disease in 18 to 24 months after 131 I therapy with a sensitivity of 88%, specificity of 72%, PPV of 47% and NPV of 95%, and area under the ROC curve of 0.872. In our study, a baseline-stimulated Tg value of 3.525 ng/mL was identified as the optimum cut-off by the ROC analysis. If Tg was lower than 3.525 ng/mL, the disease-free status by the first 131 I therapy could be predicted with an accuracy of 75.24%. In the logistic regression equation, Tg had a coefficient of − 0.236, which indicates the low pre-ablation Tg level could predict the possible disease-free status. Hence, measurement the pre-ablation Tg in patients with DTC is important.
We also found that pre-ablation TSH level was an important factor for complete remission by the first 131 30 that DTC patients who failed to produce a TSH level of > 30 μ IU/mL by a postoperative levo-thyroxine (LT4) withdrawal preparation showed much lower rates of successful 131 I ablation. However, several recent studies showed that TSH stimulation for postoperative 131 I ablation was not important and did not influence the ablation success rate 18,19 . In our study protocol, all patients were ideally required to have a TSH level up to 30 μ IU/mL before treatment, yet there were 19 patients who didn't meet this requirement. We gave them 100 mCi of 131 I, nonetheless. As a result, in these 19 patients, only 2 achieved a disease-free status after the first 131 I therapy. In the 17 patients who were not successfully ablated, 2 had distant metastases, and 4 were in N1b stage. This result indicates that remaining functional benign and malignant thyrocytes could be one main reason why their TSH levels failed to reach the required threshold. Then how to explain this phenomenon after all? We believed that for one thing, it was possible that thyroid remnant size and/or DTC metastatic lesions could be major factors influencing TSH levels and ablation success rates. For another, several other factors could also influence TSH. For instance, different durations of LT4 withdrawal could affect TSH level. It could be difficult to assure that all patients had the same duration of LT4 withdrawal, since not all of them were very compliant. It is also possible that older patients tend to have a lower ability to produce TSH. In fact, TSH response was found to diminish with age in DTC patients under LT4 therapy 31 . In our study, we found pre-ablation TSH had a threshold value of 99.7 μ IU/mL to predict successful ablation with a sensitivity of 57.63% and a specificity of 57.46%. This diagnostic capability was not sufficient, and there was a large overlap in pre-ablation TSH in different groups of patients. Therefore, based on the above evidence, we believe TSH should be considered a second line indicative value and it should not be classified as a strong predictor for successful ablation.
This study has several limitations. First, our study was retrospective in nature. Second, we had only a small sample of 315 patients who fulfilled our inclusion criteria. Third, factors such as distant metastasis and FTC were not analyzed in detail due to small sample scale. Further large-scale prospective studies are needed.

Conclusion
This study revealed that LN metastasis, pre-ablation Tg and TSH were the most powerful predictors for achieving a disease-free status by the first 131 I therapy. Patients with LN metastasis, high pre-ablation Tg level and low pre-ablation TSH would be unlikely to achieve a disease-free status.

Materials and Methods
Patients. We retrospectively reviewed the DTC patients' database from the year 2010 to 2015 archive of the Department of Nuclear Medicine, Tianjin Medical University General Hospital. All included patients received   Table 2. Diagnostic and predictive values of pre-ablation Tg* and TSH* for a successful therapeutic outcome. *Tg = thyroglobulin, TSH = thyroid stimulating hormone, PPV = positive predictive value, NPV = negative predictive value, AUC = area under the curve, CI = confidence interval.
total thyroidectomy by our specialized thyroid surgeons. Central neck lymph node removal was conducted in all DTC patients. Unilateral or bilateral neck lymph node dissection of the lateral cervical compartments was performed if the following two conditions were met: 1) clinically or sonographically suspicious lateral lymph nodes were known, 2) pre-operative biopsy or intra-operative excision showed lymph node metastases. After thyroidectomy, the patients came to our department for 131 I treatment. The included patients had the first 131 I therapeutic dose between 100 to 120 mCi. Our exclusion criteria were: 1) less than total thyroidectomy, including subtotal thyroidectomy and lobectomy; 2) thyroglobulin antibody (TgAb) > 40 IU/mL prior to the first 131 I therapy and post-treatment 131 I whole-body scan, avoiding TgAb's influence on the veracity of Tg 32 ; 3) diagnostic 131 I whole-body scan was performed before the first 131 I therapy, ruling out possible stunning effect; 4) patients without adequate data for analysis.
The Institutional Review Board of Tianjin Medical University General Hospital approved the ethical, methodological and protocol aspects of this investigation. All DTC patients provided their written informed consents. We confirm that all methods were carried out in accordance with the relevant guidelines and regulations.  Classification. For the purpose of our study, we defined risk constellations based on the ATA risk classification 3,5 . Specifically, patients with large primary tumor (> 4 cm) or extensive extrathyroidal invasion (T4a or T4b in accordance with the TNM system) or with distant metastases were classified as high risk. Patients with non-metastasized intrathyroidal tumors not exceeding a diameter of 4 cm and with < 5 central compartment LN metastases and limited extrathyroidal invasion were classified as low risk.
Protocol. All patients were treated with 131 I after a preparation of LT4 withdrawal. The patients were asked to keep a strict low iodine diet, and stop using any drug (such as amiodarone) or contrast with iodine. Routine examinations, including serum TSH, Tg, and TgAb levels and cervical ultrasound, were performed. Most of the patients received 100 mCi 131 I, only 5 patients received 120 mCi due to imaging confirmed existence of metastasis before 131 I treatment. WBS was performed 4 to 6 days later. In the follow-up treatment, all patients continued TSH suppression therapy, and regular monitoring was conducted in clinic. Further assessments of TSH, Tg, TgAb, cervical ultrasound, and post-treatment whole body scan or diagnostic whole body scan were conducted after 6 to 7 months. Serum parameter evaluation. By chemiluminescent reaction principle, free triiodothyronine (reference 3.50-6.50 pmol/L), free thyroxine (reference 11.50-23.50 pmol/L) and TSH (reference 0.30-5.00 μ IU/mL, maximum 150.00 μ IU/mL) assays were performed on a fully automated ADVIA Centaur analyzer (Siemens Healthcare Diagnostics, New York, USA). Tg (reference 0-55.00 ng/mL, maximum 300.00 ng/mL) and TgAb (reference 0-40.00 IU/mL, maximum 3000.00 IU/mL) were also assessed by chemiluminescent reaction on a fully automated IMMULITE 2000 analyzer (Siemens Healthcare Diagnostics, Los Angeles, USA).

Predictors and definition.
We assessed the following variables as possible influential factors for 131 I ablation: age, gender, primary tumor size, pathological type, capsular invasion, pathological tumor stage, LN metastasis, distant metastasis, TNM stage, ATA risks, time interval between surgery and 131 I therapy, pre-ablation Tg, and pre-ablation TSH. Disease-free status was defined as stimulated Tg < 1 ng/mL, negative TgAb, and no evidence of tumor on cervical ultrasound and WBS 6 to 7 months after first 131 I therapy 5 .
Date analysis. Analysis was performed using Statistic Package for Social Science (SPSS) version 17.0.
Comparative analyses were performed using independent sample's t test or chi-square tests. Cutoff values of pre-ablation serum Tg and TSH for diagnosis were determined using ROC analysis. OR with 95% confidence interval was calculated by adopting binary logistic regression. Logistic regression (Forward Wald method) was performed to generate an equation to check parameters' power for predicting a successful disease-free status. A P value of < 0.05 was considered to be significant.