Lung Metastases From Papillary Thyroid Cancer With Persistently Negative Thyroglobulin and Elevated Thyroglobulin Antibody Levels During Radioactive Iodine Treatment and Follow-Up: Long-Term Outcomes and Prognostic Indicators

Background: The lung is the most frequent site of distant metastasis from differentiated thyroid cancer (DTC). However, lung metastasis from papillary thyroid cancer (PTC) with persistently negative thyroglobulin (Tg) and elevated Tg antibody (TgAb) levels is an extremely rare entity, and the prognosis is therefore elusive. We investigated the clinical characteristics, long-term outcomes, and prognostic factors of lung metastases in PTC patients with persistently negative thyroglobulin (Tg) and elevated Tg antibody (TgAb) levels during radioactive iodine (131I) treatment and follow-up. Methods: We retrospectively reviewed 10,482 patients with DTC who underwent postoperative 131I treatment from 2007 to 2017 at Shanghai Sixth's People's Hospital. The relationships between progression-free survival (PFS) and several variables were assessed by univariate and multivariate analyses using the Kaplan–Meier method and a Cox proportional hazards model, respectively. Results: Forty-seven patients with PTC were enrolled in this study (4.48‰ of all patients with DTC). The median age at the initial diagnosis of lung metastasis was 39.6 ± 15.4 years, and the patients comprised 14 male and 33 female patients (male: female ratio = 1.00:2.36). Twenty-five patients had 131I avidity and 22 had non-131I avidity. At the end of the 5-years follow-up, 12 patients exhibited progressive disease (PD), and 2 patients had died. At the end of the 10-years follow-up, 21 patients showed PD and five patients had died. The 5- and 10-year PFS rates were 74.47 and 53.32%, respectively; the 5- and 10-years overall survival (OS) rates were 95.74 and 89.36%, respectively. The timing of diagnosis of lung metastases, maximal size of lung metastases, and 131I avidity were significantly associated with the 5-years PFS rate (P = 0.035, P = 0.030, and P<0.001, respectively). Only 131I avidity was associated with the 10-years PFS rate (P < 0.001). The multivariate analyses also showed that non-131I avidity were the independent poor prognostic factors for 10-years PFS at the end of follow-up (P < 0.001). Conclusions: Lung metastases from PTC in patients with persistently negative Tg and elevated TgAb levels had an excellent prognosis and survival rate during 131I treatment and follow-up. The loss of 131I avidity remained the strongest independent predictor of a poor prognosis and survival in these patients.

Results: Forty-seven patients with PTC were enrolled in this study (4.48‰ of all patients with DTC). The median age at the initial diagnosis of lung metastasis was 39.6 ± 15.4 years, and the patients comprised 14 male and 33 female patients (male: female ratio = 1.00:2.36). Twenty-five patients had 131 I avidity and 22 had non-131 I avidity. At the end of the 5-years follow-up, 12 patients exhibited progressive disease (PD), and 2 patients had died. At the end of the 10-years follow-up, 21 patients showed PD and five patients had died. The 5-and 10-year PFS rates were 74.47 and 53.32%, respectively; the 5-and 10-years overall survival (OS) rates were 95.74 and 89.36%, respectively. The timing of diagnosis of lung metastases, maximal size of lung metastases, and 131 I avidity were significantly associated with the 5-years PFS rate (P = 0.035, P = 0.030, and P<0.001, respectively). Only 131 I avidity was associated with the 10-years PFS rate (P < 0.001). The multivariate analyses also showed that non-131 I avidity were the independent poor prognostic factors for 10-years PFS at the end of follow-up (P < 0.001).

INTRODUCTION
Papillary thyroid cancer (PTC) is the most common histological type of differentiated thyroid cancer (DTC), accounting for 90% of thyroid cancers (1). According to the National Cancer Center, PTC is one of the fastest growing malignant tumors in China (2), especially in women under 30 years of age, and has become the most commonly diagnosed cancer. The dramatic rise in thyroid cancer among younger women is similar to the situation in Western countries (3). Despite the increased incidence, patients with PTC usually have a good prognosis with a 5-years overall survival (OS) rate of >98.1% and 10-years OS rate of >85% to 90% from the initial diagnosis (4). However, patients with PTC have a worse prognosis in the presence of distant metastases and/or local recurrence, which seriously affect patients' quality of life and survival time (5).
Distant metastatic lesions from DTC are usually seen in the lungs, followed by the bones. During the past 70 years, radioactive iodine ( 131 I) therapy has been the mainstream and routine treatment strategy for patients with DTC with lung metastasis. During the 131 I treatment period, twothirds of patients with DTC with lung metastases underwent therapeutic 131 I whole-body scanning ( 131 I-WBS), and half of these patients achieved remission; the remaining onethird of the patients did not have 131 I uptake capacity and died prematurely of radioiodine-refractory DTC (RR-DTC) (6). Two recent randomized, multicenter, double-blind, placebocontrolled clinical trials showed significant improvement in progression-free survival (PFS) in patients with progressive RR-DTC treated with sorafenib and lenvatinib, which have been approved by the US Food and Drug Administration for the treatment of progressive RR-DTC (7). Several independent prognostic factors associated with a poor prognosis in patients with DTC with lung metastases have been reported, such as age at the initial diagnosis of lung metastases, 131 I avidity, timing of diagnosis of lung metastases, maximal size of lung metastases at diagnosis, pathological type, and the presence of extrapulmonary distant metastases (8)(9)(10)(11).
The serum thyroglobulin (Tg) level is the most sensitive and reliable marker indicating persistent or recurrent disease during follow-up after total or near-total thyroidectomy and 131 I remnant ablation in patients with DTC. However, several factors may lead to false-negative Tg findings; e.g., a detectable amount of Tg cannot be released, the ability to secrete Tg is lost, structural changes occur in Tg, or artificially low Tg levels are induced by elevated circulating Tg antibody (TgAb) (12,13). Previous studies have shown that TgAb interferes with serum Tg detection in 10-25% of patients with PTC with positive TgAb, especially in patients with increased TgAb levels (14). In this regard, several studies have demonstrated that a progressively increasing TgAb titer can be used as a surrogate marker for predicting persistent or recurrent disease in patients with PTC with a negative serum Tg level (15,16). As reported in the current American Thyroid Association guideline, the criteria for negative Tg in patients with DTC are a stimulated Tg level of <10 ng/mL before 131 I ablation, a Tg level of 0.2 ng/mL after thyroid-stimulating hormone (TSH) suppression, or a stimulated Tg level of <1 ng/mL during followup (3).
In clinical practice, patients with PTC with lung metastases very rarely have persistently negative Tg and elevated TgAb levels during 131 I treatment and follow-up. To the best of our knowledge, only Viola et al. (17) has described a 56-year-old PTC patient with lung metastases who had elevated TgAb and negative serum Tg levels 10 years after initial treatment involving total thyroidectomy and 131 I remnant ablation. Therefore, little is known about the clinical outcomes and prognostic factors of these patients with PTC. To improve our understanding of the natural history of lung metastases in these patients with PTC after total or near-total thyroidectomy with persistently negative Tg and elevated TgAb levels during 131 I treatment and followup, we evaluated these patients' clinical characteristics, clinical outcomes, and independent prognostic factors in a large referral 131 I treatment center in China.

Patients
We retrospectively reviewed the medical records of 10,482 consecutive patients with DTC who underwent 131 I treatment at Shanghai Sixth's People's Hospital after total or near-total thyroidectomy from 2007 to 2017. Of these, the clinical followup data of 503 patients with DTC diagnosed with lung metastasis were evaluated. The inclusion criteria for patients with PTC with lung metastases who had persistently negative Tg and elevated TgAb levels during 131 I treatment and follow-up were (i) histological confirmation of PTC, (ii) treatment with 131 I after total or near-total thyroidectomy, (iii) a negative Tg level [defined as a preablative stimulated Tg level of <10 ng/mL at the time of initial 131 I therapy or a Tg level of <0.2 ng/mL (TSH suppression) or <1 ng/mL (after stimulation) 6 months after the first remnant ablation according to the American Thyroid Association guideline (3)], and (iv) an elevated TgAb level (defined as a TgAb level of ≥80 UI/mL) at the time of diagnosis of lung metastasis from PTC. According to these criteria, 56 patients were identified as having lung metastases from PTC with negative Tg and elevated TgAb levels. Among them, nine patients were excluded from the current study because of poor PTC confirmed by pathological examination at the initial surgery (n = 2), breast cancer that may lead to lung metastases during follow-up (n = 1), insufficient follow-up data (n = 2), and a negative preablative stimulated Tg level at the time of initial 131 I therapy but a TSHsuppression Tg level of >0.2 ng/mL during follow-up (n = 4) (Figure 1).

Collection of Variables
The surgical and histopathologic data evaluated in this study were sex, maximum primary tumor diameter, pathology of primary tumor, number of primary tumors, pathological examination showing Hashimoto's thyroiditis, extrathyroidal invasion, and N stage. Among these, the maximum diameter, N stage, and extrathyroidal invasion of the primary tumor were evaluated according to the 8th edition of the TNM classification system (18). The maximum tumor diameter was divided into <2 cm, 2-4 cm, and >4 cm; the N stage was divided into N0, N1, and N2; and extrathyroidal invasion was divided into none, minimal, and gross. Poor PTC was excluded from the study, and the PTC pathology was divided into classic PTC and follicular variant of PTC.
According to the recommendation in the 7th and 8th editions of the TNM classification system (18), the cut-off age for risk stratification was set at 45 and 55 years, respectively, when lung metastases of PTC were diagnosed.
The following data regarding lung metastases from PTC were collected: timing of diagnosis of lung metastases, maximal size of lung metastases at diagnosis (mm), and 131 I avidity. The timing of discovery of lung metastases of PTC was divided into lung metastases at presentation and delayed lung metastases. The discovery of lung metastases at presentation was defined as the detection of lung metastasis within 6 months before and after initial thyroidectomy, and the discovery of delayed lung metastases was defined as the detection of lung metastasis ≥6 months after initial thyroidectomy (19). According to the maximal diameter of the lung nodules on chest computed tomography (CT), pulmonary metastatic lesions were classified as lung nodules of >1 cm, 0.5-1 cm, or <0.5 cm. 131 I avidity for lung metastases of PTC was defined as visible 131 I uptake on 131 I-WBS after 131 I treatment, and non-131 I avidity was defined as negative 131 I-WBS results after 131 I treatment or 131 I uptake of <10% of multiple lung metastatic lesions seen on 131 I-WBS combined with 131 I single-photon emission CT/CT( 131 I-SPECT/CT) (20).

Diagnostic Criteria for Lung Metastases in Patients With PTC
The diagnosis of lung metastases from PTC was established based on clinical symptoms, serum TgAb levels, chest CT findings before or after 131 I treatment, and 131 I avidity on therapeutic 131 I-WBS and/or 131 I-SPECT/CT. The diagnosis of lung metastases was confirmed by one of the following four approaches: Criterion 1, pathological puncture results confirmed lung metastasis of PTC under CT guidance before 131 I treatment; Criterion 2, lung 131 I avidity was seen on therapeutic 131 I-WBS, and pulmonary nodes were found on chest CT and/or therapeutic 131 I-SPECT/CT in patients with PTC with an elevated serum TgAb level after 131 I treatment; Criterion 3, lung 131 I avidity could be detected on therapeutic 131 I-WBS, but the chest CT findings were negative and the serum TgAb level was elevated before 131 I treatment (21); and Criterion 4, positive pulmonary nodules could be detected on chest CT in PTC patients with non-131 I-avid lung metastases, and the TgAb level increased continuously for at least 2 years during follow-up.

I Treatment of Lung Metastasis
At 1-6 months after total or near-total thyroidectomy combined with neck lymph node dissection, the standard 131 I treatment at our institution was performed in these patients with PTC. Before 131 I treatment, each patient was given a low-iodine diet and began levothyroxine withdrawal for at least 2 weeks to achieve a TSH level of ≥30 mIU/L. The patients underwent routine measurements of free triiodothyronine (FT3), free thyroxine (FT4), TSH, Tg, and TgAb; neck ultrasonography (US); and CT scans before oral administration of 131 I treatment. We subsequently used an empirical active regimen to determine the prescribed activity of 131 I. Specifically, 3.7 GBq (100 mCi) was taken orally for the first time to ablate thyroid remnants; oral doses of 131 I with a standard activity of 5.55-7.40 GBq (150-200 mCi) were used for subsequent treatment of lung metastases of PTC. At 4-7 days after 131 I oral administration, 131 I-WBS and/or 131 I-SPECT/CT fusion imaging was performed for these patients. The repeated 131 I treatment was suspended in patients with non-131 I-avid lung metastases on the 131 I-WBS after completion of residual thyroid tissue ablation.

Tg and TgAb Measurement
The FT3, FT4, TSH, Tg, and TgAb levels were measured before 131 I treatment and during follow-up. A cobas R analyzer (Roche Diagnostics GmbH, Basel, Switzerland) was used to determine the Tg and TgAb levels using the same highsensitivity electrochemiluminescence immunoassay method in the same laboratory of our hospital based on the manufacturer's instructions. Quality control was ensured by assaying the Tg and TgAb levels in control sera in each analytical series, and all sera in which the interassay coefficient of variation exceeded 10% were reassessed. The analytical limit of Tg was 0.1 µg/mL with a detection range of 0.1 to 25,000 µg/L, and the analytical limit of TgAb was 10 UI/mL with a detection range of 10 to 4,000 IU/mL. A persistently negative Tg level was defined as a preablative stimulated Tg level of <10 ng/mL at the time of initial 131 I therapy or a Tg level of <0.2 ng/mL (TSH suppression) or <1 ng/mL (after stimulation) during follow-up (3). An elevated TgAb level was defined as a TgAb level of ≥80 IU/mL according to a previous report (22). TgAb levels were classified into three categories in this study: <1,000 IU/mL, 1,000-4,000 IU/mL, an≥4,000 IU/mL.

Evaluation of Change in Pulmonary Nodules Based on Chest CT Imaging
Progressive disease (PD) was evaluated based on the change in metastatic lung nodules compared with baseline using chest CT according to the Response Evaluation Criteria in Solid Tumors (RECIST v1.1) during 131 I treatment and follow-up and similar to our previous studies (19,20). Chest CT examinations were performed with a helical CT scanner on suspended full inspiration with a slice thickness of 3 mm. The evaluation criteria were complete response (CR), defined as disappearance of all detectable target lesions; partial response (PR), defined as a 30% decrease in the sum of all target lung metastatic lesion volumes; PD, defined as a 20% increase in the sum of all lung metastatic lesion volumes, or the appearance of more than one new lesion; and stable disease (SD), defined as neither PR nor PD.

Follow-Up
After the initial 131 I ablation and treatment, the patients took levothyroxine orally to suppress the TSH level to <0.1 µg/mL. Regular follow-up was performed, including measurement of FT3, FT4, TSH, Tg, and TgAb; neck US; and chest CT. During the follow-up period, the FT3, FT4, TSH, Tg, and TgAb levels were measured and neck US was performed every 3-6 months, and chest CT was performed every 12 months. At the end of followup, OS and PFS were assessed. The OS was defined as the time from the detection of lung metastatic lesions of PTC to death of any cause, and PFS was defined as the time from the diagnosis of lung metastatic lesions of PTC to the detection of PD or death of any cause.

Statistical Analysis
Continuous variables are presented as mean, standard deviation (SD), minimum, and maximum, and categorical variables are presented as number with percentage. The Kaplan-Meier method was applied to calculate the PFS rate of patients with PTC from the date of diagnosis of lung metastasis to the date of PD during follow-up. The 5-and 10-years PFS rates were analyzed with the Kaplan-Meier method, and differences in PFS were compared by the log-rank test with a P < 0.05 regarded as statistically significant. A Cox proportional hazards model was used for the multivariate analysis to investigate the associations between prognostic factors and PFS at the end of followup; again, a P < 0.05 was considered statistically significant. Statistical analyses were performed using MedCalc software version 17.0 (MedCalc Software, Mariakerke, Belgium) and GraphPad Prism 7 (GraphPad Software, San Diego, CA, USA).

Clinical Symptoms and Diagnosis of Lung Metastases
Among all 47 patients, 4 were diagnosed based on Criterion 1, 19 were diagnosed based on Criterion 2, 3 were diagnosed based on Criterion 3, and 21 were diagnosed according to Criterion 4 One patient had synchronous bone and lung metastases, and three patients had synchronous lung and mediastinal metastases. Thirteen patients had locoregional recurrent disease without undergoing surgical resection; of these 13 patients, 10 had a solitary lymph node metastasis, the size of which stabilized during follow-up, and the patients were unwilling to undergo surgery. The remaining three patients had locally invasive lesions that could not be surgically resected. Forty-three patients with lung metastases were asymptomatic, while four had clinical symptoms related to the lung metastatic lesions: breathing difficulty in one, hemoptysis in one, and dry cough in two.

Changes in Serum TgAb Levels
According to the serum TgAb grouping method used in the present study, the TgAb level was <1,000 IU/mL in 27 (57.45%) patients, 1,000-4,000 IU/mL in 13 patients, and >4,000 IU/mL in 7 patients. The changes in the serum TgAb levels are shown in Figure 2.
In the <1,000 IU/mL group, the serum TgAb level decreased in 25.53% (12/47) of patients with an average change of 373 (range, 81-880) to 94.5 (range, 10-182) IU/mL from the diagnosis of pulmonary metastases to the end of follow-up. The Tg level increased in the remaining 31.91% (15/47) of patients with an average change of 493.6 (range, 109-944) to 917.7 (range, 232-2,435) IU/mL.   In the 1,000-4,000 IU/mL group, a decrease in the serum TgAb level was seen in seven (14.9%) patients and an increase was seen in six (12.8%). In the seven patients with a decreased TgAb level, the average TgAb level decreased from 2,021 (range, 1,496-3,391) to 766 (range, 236-1,452) IU/mL. In the six patients with an increased TgAb level, the mean TgAb level was 1687.5 IU/mL (range, 1,008-3,245 IU/mL) before 131 I treatment; at the end of follow-up, five (10.6%) patients' TgAb level had increased to 4,000 IU/mL, and one patient's TgAb level had increased from 1,008 to 1,048 IU/mL.
In the >4,000 IU/mL group, only one patient's serum TgAb gradually decreased from >4,000 to 1,356 U/mL, and in the remaining six patients, the serum TgAb levels were still >4,000 IU/mL after 131 I treatment and follow-up.

Treatment Outcome of Lung Metastases
All 47 patients with PTC patients with lung metastases who had persistently negative Tg and elevated TgAb levels during follow-up underwent 131 I treatment. The 131 I treatment outcome were showed in Table 1. Twenty-five (53.19%) patients had 131 I avidity and 22 (46.81%) had non-131 I avidity on 131 I-WBS after an oral therapeutic dose of 131 I. Among the 25 patients with 131 I uptake of lung metastases, diffuse absorption was seen on 131 I-WBS in 3 patients with elevated TgAb levels after 131 I treatment, but lung nodules could not be detected by chest CT. Of the 22 patients with non-131 I avidity, only 1 patient who developed progressive thyroid cancer was treated with sorafenib. Sorafenib was administered at a dose of 200 mg orally twice a day. A cycle was defined as 4 weeks. This patient developed a severe handfoot reaction and stopped taking the sorafenib after three cycles. Neither chemotherapy nor external beam irradiation was used to treat the lung metastases in this study.

Results of 5-and 10-Years OS and PFS
The follow-up time ranged from 2 to 10 years (mean, 6.9 ± 2.9 years). At the end of follow-up, 42 patients were alive and 5 had died. Of the five patients who died, one died of locally recurrent disease 2.4 years after the diagnosis of lung metastasis; two died of lung metastasis at 5.5 and 9.1 years after the diagnosis of lung metastasis, respectively; and two died of cardiac disease and myeloma at 4.1 and 9.4 years from the initial follow-up, respectively. Based on the RECIST v1.1 criterion using chest CT imaging, at the end of the 5-years follow-up, seven patients had obtained CR, including three with undetectable pulmonary nodules on chest CT; eight patients had PR; 20 had SD; and 12 had PD. At the end of the 10-year follow-up, seven patients had achieved CR, seven had PR, 12 had SD, and 21 had PD. The 5and 10-years OS rates were 95.74 and 89.36%, respectively, and the 5-and 10-years PFS rates were 74.47 and 53.32%, respectively (Figures 3A,B).

Factors Predicting 5-and 10-Year PFS Rate
The 5-years PFS rate and its prognostic factors in the 47 patients with lung metastases as evaluated by the Kaplan-Meier method and log-rank test are listed in Table 2. At the end of the 5years follow-up period, 12 patients exhibited PD and 35 did not. In the univariate analysis, the timing of diagnosis of lung metastases, maximal size of lung metastases, and presence of 131 I  avidity were significantly associated with the 5-years PFS rate. Patients with lung metastatic lesions of <5 mm, delayed lung metastases, and 131 I-avid disease had higher 5-years PFS rates than patients with lung metastatic lesions of 5-10 mm/≥10 mm, lung metastases at initial presentation of PTC, and non-131 I-avid disease, respectively. Nevertheless, other several factors were not significantly associated with the 5-years PFS rate (Figures 3C-E). The 10-years PFS rate and its prognostic factors in the 47 patients with lung metastases as evaluated by the Kaplan-Meier method and log-rank test are listed in Table 3. At the end of the 10-years follow-up period, 21 patients showed PD; the univariate analysis demonstrated that only 131 I avidity was associated with 10-years PFS rate in these patients. Patients without 131 I avidity had more significantly shorter than patients with 131 I avidity. However, there was no significant association between the 10-years survival rate and the remaining several factors (Figure 3F).
A Cox proportional hazards model was used to assess the independent predictors of the 10-years PFS rate ( Table 4). The absence of 131 I avidity (relative risk, 1048.45; 95% confidence interval, 26.600-41324.708; P < 0.001) were independent poor prognostic factors for the 10-years PFS rate in the multivariate analysis. Other several factors were not predictive of the 10-years PFS rate.

DISCUSSION
In the current study, we retrospectively analyzed 47 patients with PTC with lung metastases who had persistently negative serum Tg and elevated serum TgAb levels during 131 I treatment and follow-up. The study showed that the 5-and 10-years OS rates were 95.74 and 89.36% and that the 5-and 10-years PFS rates were 74.47 and 53.32%, respectively. Lung metastatic lesions of 5-10 mm/≥10 mm, lung metastases at initial presentation of PTC, and non-131 I-avid lung metastases predicted a poor 5-years PFS rate, and only patients without 131 I avidity showed a poor 10years PFS rate. The absence of 131 I avidity were the independent poor prognostic factors for the 10-years PFS rate at the end of follow-up.  The median age of the 47 patients with PTC with lung metastases at diagnosis was 39.6 years (range, 12-70 years), which is significantly lower than that reported in patients with DTC with lung metastases. Previous studies have shown that the median age of patients with DTC with lung metastases varies from 45 to 52 years (8-11, 21, 23, 24). Age has been recognized as a predictive factor that is closely associated with survival and recurrence in patients with DTC with lung metastases, but this remains controversial (25,26); however, most studies have suggested that age can be used as an independent prognostic factor for DTC (8-11, 21, 23, 24, 26-28). The cut-off age for risk stratification was set at 45 and 55 years according to the recommendations of the 7th and 8th editions of the TNM classification system, respectively (18). In the present study, 7 (14.89%) patients were ≥55 years of age and 40 patients were <55 years of age, while 16 (34.04%) patients were ≥45 years of age, and 31 patients were <45 years of age. The log-rank test showed that a cut-off age of 55 or 45 years was not significantly associated with the 5-and 10-years PFS rates in these patients in the univariate analysis. Study have showed that a cut-off age of 55 years resulted in higher accuracy for prognosis and recurrence in patients with DTC than a cut-off age of 45 years (18,29); however, the cut-off of 55 years did not provide advantages over the cut-off of 45 years in the present study. A possible explanation is that the sample size of patients aged >55 years was too small for a statistical analysis, producing a deviation (patients ≥55 vs. <55 years: ratio of 1.0:5.7). Our multivariate analysis also showed that age was not an independent prognostic factor for these patients.
In previous reports, the incidence of non-131 I-avid lung metastases from DTC ranged from 5 to 45%, with an average value of about 30% (8-11, 21, 23, 24). In the present study, 131 I was not absorbed by lung metastases in 46.81% (22/47) of patients with PTC with persistently negative Tg and elevated TgAb levels; this incidence is clearly higher than that reported above. This might also suggest differences in sample sizes, selection bias, and/or screening methods used in the referral population in our center and/or in clinical management protocols after DTC diagnosis; all of these factors affect the incidence rate of non-131 I avidity in patients with DTC. 131 I avidity is well-recognized as a prognostic factor for lung metastases from DTC. Several studies have also indicated that DTC patients with 131 I avidity showed longer OS, disease-free survival, or PFS than those with non-131 I-avid lung metastases (8-11, 21, 23, 24). Consistent with previous studies, our study showed that the 5-and 10-years PFS rates were significantly higher at the end of follow-up in patients with than without 131 I avidity (96.00 vs. 50.00% and 76.00 vs. 31.82%, respectively). The multivariate analysis also showed that 131 I avidity is the strongest predictive factor for a better prognosis in these patients. Normally, when the largest lung metastatic lesion of DTC measures <1 cm, it is defined as a micronodule, and a >1cm lesion is defined as a macronodule. Most studies have confirmed that lung metastases composed of micronodules are an important predictor of a better prognosis and longer survival in patients with DTC; studies have shown that compared with macronodules, micronodules are associated with significantly improved OS or PFS in patients with DTC (10,11,30). However, the ratio of micronodules to micronodules and its association with prognosis and survival are not inconsistent among previous studies For example, Sohn et al. (11) showed that among 89 patients with only lung metastases who were treated from 1996 to 2012, the ratio of micronodules to macronodules was 1.00:2.33 (24/56) with the 10-years PFS rate for about 60% and 28%, respectively. Mona et al. (23) reported that among 199 consecutive patients with follicular cell-derived thyroid cancer with lung metastases who underwent 131 I treatment, 75% had a <1-cm nodule as the largest lesion and 25% of patients had a ≥1-cm nodule as the largest lesion, producing a ratio of 3:1 and the 10-years PFS rate was about 37 and 5%, respectively. These differences may be due to the different criteria used to select patients and the different modes of diagnosis and treatment. In the present study, the size of the lung metastases in 40 patients was <1 cm and that in seven patients (14.89%) was >1 cm, for a ratio of 5.71:1.00. Because the number of patients with micronodules and macronodules varies considerably, statistical deviations may occur. To avoid this, we classified the lung metastases as lung nodules of >1 cm, 0.5-1 cm, and <0.5 cm. Our study showed that the size of the lung nodule could affect the 5-years PFS rate but could not affect 10-years PFS rate. The multivariate analysis also showed that the size of the largest lung metastatic lesion was not an independent prognostic factor at the end of the 10years follow-up.
The presence of distant metastatic disease at presentation is relatively rare, with a rate of 3% to 15%. In our study, 22 (46.81%) patients had lung metastases at the initial presentation, and 25 (53.19%) had delayed lung metastases during followup. The presence of lung metastatic disease at presentation was higher in our series than in past reports. Some investigators have reported that delayed metastasis has a worse prognosis than early metastasis (10,11,31). In contrast, our study showed that the 5-years PFS rate in patients with delayed metastasis was significantly higher than that in patients with lung metastasis at presentation; however, there was no significant difference in 10years PFS rate between the two groups. This discrepancy might be explained by the difference in the number of patients with 131 I avidity between these two groups. Stimulated Tg assessment was performed after withdrawal TSH from thyroid replacement therapy or after stimulation with recombinant human TSH. Not only is a high stimulated Tg level of >10 µg/mL considered an indicator of recurrent/metastatic disease in patients with DTC, but a high stimulated Tg level of ≥50 µg/mL is associated with shorter PFS and OS in patients with DTC (32). Because a gradual increase in the TgAb titer after thyroidectomy can be used as a predictor of persistent or recurrent disease in patients with DTC with a negative serum Tg level (15,16), we investigated whether high levels of TgAb are also associated with shorter PFS and OS of patients with PTC with lung metastases. In this study, the TgAb levels were classified into three categories: <1,000 IU/mL, 1,000-4,000 IU/mL, and >4,000 IU/mL, and the percentages of patients in these three groups were 57.45% (27/47), 27.70% (13/47), and 14.89% (7/47), respectively. Our study showed no significant difference in the 5-or 10-years PFS rate among the three groups. The multivariate analysis also showed that the TgAb level was not an important prognostic factor for the 10-years PFS rate among the three groups at the end of follow-up. The use of the TgAb level as a prognostic parameter in patients with DTC remains controversial. Rubello et al. (33) and Adel et al. (34) reported a higher post-ablation local recurrence rate that was associated with a persistently increased TgAb level. However, most authors found no correlation between these parameters (35,36). To the best of our knowledge, the present study is the first to evaluate the relationship between the TgAb level and lung metastasis in these patients with PTC. Larger sample sizes and longer-term follow-up of these patients are needed for further validation.
It is well-known that about 20-30% of PTC patients have Hashimoto's thyroiditis (37)(38)(39); our study showed a 27.66% (13/47) rate PTC patients with Hashimoto's thyroiditis, which was basically accordant with the above study. Whether Hashimoto's thyroiditis is related to the prognosis in PTC patients remains controversial; some anthers have reported that PTC patients with Hashimoto's thyroiditis are associated with better TNM staging and prognosis, with a lower rate of lymph node metastases, distant metastasis and recurrence rate because of its protection (39)(40)(41)(42). Other studies have demonstrated that PTC patients with preexisting Hashimoto's thyroiditis are closely associated with the increased risk of recurrence (37,38,(43)(44)(45). In the present study, we found that Hashimoto's thyroiditis wasn't associated with PFS and this may be related to the relatively limited number of patients, selected deviation and different diagnostic diagnosis and treatment methods and so on.
In addition to age, 131 I avidity and the size of lung metastases may also be associated with a better prognosis and longer survival of patients with lung metastases from DTC, including patients with DTC with only lung metastases, non-extrathyroidal extension, and so on. In the present study, the pathological type was PTC in all 47 patients because follicular thyroid carcinoma with lung metastases rarely causes the TgAb levels to rise. Studies have demonstrated that coexisting bone metastases are more closely associated with the prognosis than are solitary lung metastases of DTC (8)(9)(10)(11). Of our 47 patients with lung metastases, bone metastases were found in only 1 patient. In our series of 47 patients with PTC, the 5-and 10-years OS rates were 95.74 and 89.36%, respectively. These survival rates are obviously higher than those of patients with DTC with lung metastases reported in previous studies (50-93% and 48-85%, respectively). In our study, only five patients had died by the end of the follow-up; therefore, the prognostic factors associated with OS in these patients were not assessed by univariate and multivariate analyses. We speculate that the better prognosis of these patients was mainly related to the above-mentioned factors (age at the diagnosis of lung metastases, 131 I avidity, size of the lung metastases, the presence of only lung metastases, and so on).
The present study has several limitations. First, it has certain inherent limitations associated with its retrospective design. Second, we only registered patients from a single tertiary referral center, and we included a relatively limited number of patients. Nonetheless, this is the first assessment of the clinical outcome in patients with lung metastases from PTC with persistently negative Tg and elevated TgAb levels during 131 I treatment and follow-up. Third, in all eligible patients, the lung nodules were confirmed by a CT scan, significantly increased serum TgAb levels, and 131 I-WBS because it is unrealistic and unethical to obtain histological evidence of diffuse lung lesions in these patients. Fourth, our study lacked a control group in DTC patients with elevated Tg. In our past study, however, we reported the survival and prognosis of 372 DTC patients with lung metastases and elevated Tg, whose 10-years OS rate was significantly lower than that of PTC patients with elevated TgAb and persistently negative Tg level (10). Moreover, only a few other diseases can cause an increase in the TgAb level, such as type 1 diabetes, rheumatoid arthritis, and malignant anemia and so on (46). When PTC and multiple inflammatory lung nodules are present simultaneously, it is difficult to confirm them and they can only be excluded by follow-up, especially for non-131 I-avid lung metastases of PTC.

CONCLUSION
This study is the first to evaluate the clinical characteristics, outcomes, and prognostic factors of lung metastases from PTC in patients with persistently negative Tg and elevated TgAb levels, which represents an extremely rare pattern of invasion of PTC with an incidence of only 4.48‰. In these patients, the 5-and 10-years OS rates were 95.74 and 89.36%, respectively, and the 5-and 10-years PFS rates were 74.47 and 53.32%, respectively. These rates suggest that lung metastasis from DTC has an excellent prognosis and good survival in patients with PTC with persistently negative Tg and elevated TgAb levels during 131 I treatment and follow-up. The timing of diagnosis of lung metastases, maximal size of lung metastases, and 131 I avidity may be associated with the 5-years PFS rate, and only 131 I avidity was associated with the 10-years PFS rate in these patients with lung metastases. The loss of 131 I avidity was the strongest poor independent prognostic factor for prognosis and survival in these patients.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/supplementary material.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Committee of the Sixth People's Hospital Affiliated to Shanghai Jiaotong University. Written informed consent was obtained from the individual(s), and minor(s)' legal guardian/next of kin, for the publication of any potentially identifiable images, or data included in this article.

AUTHOR CONTRIBUTIONS
Z-LQ and Q-YL designed the study. Z-KS and C-TS conducted the statistical analysis. G-QZ and H-JS collected the clinical data. Z-LQ wrote the whole paper. Z-KS and Q-YL supervised and edited the paper. All authors read and approved the final paper.

FUNDING
We declare all sources of funding received for the research being submitted. This work was supported by the National Natural Science Foundation of China (grant number 81771865) and the Shanghai Key Discipline of Medical Imaging (grant number 2017ZZ02005).