We included 1,051 potentially eligible patients with differentiated thyroid cancer who underwent their first RAI therapy at 3 to 12 months after conventional open total thyroidectomy from April 2013 to January 2015. After excluding patients with follicular thyroid cancer, remnant cancer after surgery, confirmed distant metastasis, reoperation for recurrent cancer, and ATAb-positive patients, 680 patients with PTC were enrolled in this retrospective study. Disease stage and the risk assessment were based on the 7th American Joint Committee on Cancer/tumor, node, metastasis (AJCC/TNM) staging system [8] and the modified 2009 risk stratification system of the American Thyroid Association (ATA) [1]. The ATA risk stratification system was classified into low risk, intermediate risk, and high risk of recurrence.

Doses of RAI ranged from 1.11 GBq (30 mCi) to 6.66 GBq (180 mCi) based on tumor size, extra-thyroidal invasion, and lymph node involvement. A total of 425 patients (62.5%) were prepared for RAI therapy by withdrawing thyroid hormone, and 255 patients (37.5%) received recombinant human TSH. Patients were asked to follow a low-iodine diet for 2 weeks. Serum TSH levels were > 30 mIU/L in all patients at the pre-Tg check point. Serum TSH, pre-Tg, and ATAb levels were measured prior to RAI administration on the day of RAI therapy, and TSH, post-Tg, and ATAb, as well as post-treatment whole-body RAI (RxWBS) scans, were performed 7 days after RAI therapy. Thyroid bed uptake on RxWBS was graded from I to III by visual assessment: I, overt remnant thyroid (star sign); II, focal uptake on the thyroid bed; and III, no uptake on the thyroid bed. All images were analyzed separately by two experienced nuclear medicine physicians who were blinded to the clinical findings. To verify the response to therapy, we reclassified the patients into four categories: excellent response (ER), biochemical incomplete response (BCIR), structural incomplete response (SIR), and indeterminate response (IR) according to suppressed or TSH-stimulated Tg and ultrasonography with or without a diagnostic wholebody scan (DxWBS), obtained 6 to 18 months after RAI therapy. This is known as dynamic risk stratification (DRS) [1]. Definitions for each category are as follows: ER, clinical, biochemical, or structural evidence of disease-free conditions; BCIR, abnormal Tg or rising ATAb levels in the absence of localizable disease; SIR, persistent or newly identified loco-regional or distant metastases; and IR, nonspecific biochemical or structural findings that cannot be confidently classified as either benign or malignant. The IR group included patients with stable or declining ATAb levels without definitive structural evidence of disease. Additionally, clinical outcomes were estimated as no evidence of disease (NED), recurrence, or persistence at the final follow-up visit, which took place a mean of 27.8 months after the surgery. NED status was defined as a negative Tg (suppressed Tg < 0.2 ng/mL or TSH-stimulated Tg < 1.0 ng/mL) with NED in the imaging study. Recurrence was defined as a disease that developed again after NED, and persistent disease refers to structural and/or biochemical evidence of disease until the final follow-up without NED status.

Serum Tg and ATAb were measured with immunoradiometric assays (BRAHAMS AG, Hennigsdorf, Germany). The assays had measurement ranges of 0.1 to 1,000 ng/mL and 10 to 4,000 IU/mL, respectively. Serum TSH levels were measured using the TSH-CTK-3 immunoradiometric assay (DiaSorin SpA, Saluggia, Italy).

This study was reviewed and approved by the Institutional Review Board of Chonnam National University Hwasun Hospital, Hwasun, Korea (CNUHH-2017-059). Informed consent was waived due to the study's retrospective design.

All statistical analyses were performed using SPSS software version 23.0 (IBM Co., Armonk, NY, USA). Descriptive quantitative data are expressed as mean values and standard deviations, and qualitative data are expressed as percentages. A univariate analysis was performed prior to the multivariate analysis using the chi-square test for categorical variables and Student’s t test for continuous variables. Multivariate analyses were performed to identify independently significant prognostic variables in DRS or recurrent/persistent disease; all variables (except DRS or outcome) with a p value below 0.05 in univariate analysis were included. A p < 0.05 was considered significant.

Of the 680 enrolled patients with PTC (168 males and 512 females), the mean age at diagnosis was 49.2 ± 11.6 years. Most of the patients had classical PTC (96.0%), and the mean tumor size was 1.1 ± 0.7 cm (range, 0.2 to 5.3). In total, 255 patients (37.5%) were in the ATA low risk group, and 292 (42.9%) and 133 (19.6%) were in the ATA intermediate and high risk groups, respectively, on the initial risk stratification. After a mean 108.2 mCi of RAI therapy, 514 patients (75.6%) had an ER, 34 (5.0%) had a BCIR, 13 had a (2.0%) SIR, and 119 (17.5%) had an IR.

In the final analysis, 572 out of 680 patients (84.1%) showed NED status; six patients showed recurrence during follow-up, and 102 had persistent disease. Univariate analysis suggested that sex, ATA risk, tumor size, AJCC stage (T and N status), RAI dosage, pre-Tg, post-Tg and DRS were significant predictors of recurrent/persistent disease. Sex, ATA risk, tumor size, N status and pre-Tg remained as statistically significant independent factors in multivariate analysis (Table 1). There was no difference in outcome between the overall patient cohort and the individual patient groups, regardless of whether they were receiving recombinant human thyrotropin (rhTSH).

Among the four groups, the SIR group had the largest tumor size and worst prognostic factors on the initial postoperative risk stratification (ATA initial risk stratification, T and N stage, and AJCC stage) followed, in order, by the BCIR, IR, and ER groups. Although RAI dose was lowest in the ER group and highest in the SIR group, the final outcome showed that NED was 99.4% in the ER group and zero in the SIR group. Pre-Tg levels were also significantly different between the groups, with the highest level being in the SIR, followed by the BCIR, IR, and ER groups. The size of the SIR group increased significantly with pre-Tg levels (0.5%, 2.6% and 17.2% of patients with pre-Tg < 1, 1 to 10, and > 10 ng/dL, respectively; p < 0.001). However, post-Tg levels were not different. Multivariate analysis revealed that sex, RAI dose, and pre-Tg affected the DRS (Table 2).

The mean post-Tg level (13.8 ± 32.2 ng/mL) increased significantly after RAI therapy in all 680 patients compared to the mean pre-Tg level (2.5 ± 8.9 ng/mL). Serum TSH levels decreased from 93.1 ± 33.9 mIU/dL at pre-ablation to 33.4 ± 16.7 mIU/dL at 7 days after RAI therapy. Pre-Tg levels were < 1 ng/mL (mean 0.3 ± 0.3 ng/mL) in 422 patients, which increased to 4.2 ± 10.3 ng/mL after ablation. A total of 205 patients with pre-Tg < 1 ng/mL had post-Tg levels < 1 ng/mL (group 1); the post-Tg level increased to 1 to 10 ng/mL in 167 patients (group 2) and to > 10 ng/mL in 50 patients (group 3). No differences in the postoperative risk stratification (ATA initial risk stratification, T and N stage, or AJCC stage), response to therapy (DRS), or clinical outcome were observed among the three groups. RAI doses and uptake patterns were different among the three groups. No uptake was seen in 8.3% of group 1 patients, and uptake with the star pattern was seen in only 3.8% of these patients. All group 2 and 3 patients showed some uptake, and the rate of star sign uptake was significantly higher compared to group 1 (Table 3). There was no significant difference between the 255 patients on rhTSH and the 425 patients with T4 withdrawal.