Introduction

Differentiated thyroid cancers (DTC’s) are generally malignant endocrine tumors with good prognoses, although a small percentage has an unfavorable prognosis. The prognosis can vary depending on the histological tumor type and stage. Five to 20% of patients with DTC’s have local or regional recurrences. The most frequent sites of recurrence are regional lymph nodes (cervical and mediastinal) and the thyroid bed [1, 2].

Most DTC cases are treated by total or near-total thyroidectomy followed by I-131 radioiodine ablation to eradicate any residual thyroid tissue and to improve the accuracy of diagnostic whole body scanning (DxWBS) and serum thyroglobulin (Tg) in the long-term follow-up. In addition, I-131 ablation has been shown to decrease long-term recurrence and mortality rates [3, 4]. However, some DTCs do not trap iodine, resulting in a negative DxWBS.

Alternatively, fluorine-18-fluorodeoxyglucose positron emission tomography (FDG-PET) has proven to be a valuable diagnostic technique for detecting many types of malignant tumors and metastases. Several studies have indicated that the use of PET and PET/CT to evaluate metastasis and the recurrence of DTC is useful, even when DxWBS is negative [5–10]. In addition, some studies suggest that the sensitivity of FDG-PET is improved when combined with elevated TSH levels [11, 12]. Because thyroid ablation is usually performed under TSH stimulation, FDG-PET in conjunction with initial ablation is effective for detecting metastatic lesions. However, there have been no previous reports on FDG-PET performed concurrently with ablation. Therefore, in this study, we investigated the effectiveness of FDG-PET and PET/CT for detecting lymph node and distant metastatic DTC’s concurrent with initial I-131 ablation.

Materials and methods

Patients

This retrospective study included all patients with histologically proven differentiated thyroid cancer who underwent both postoperative FDG-PET and subsequent I-131 ablation in our institution from 2003 to 2010. The criterion for I-131 ablation was initial therapy to ablate thyroid tissue remnants after total or near-total thyroidectomy. Patients with distant metastases detected by conventional imaging (CT, MRI, I-123 and/or bone scintigraphy) were excluded. Patients with high concentration of an anti-thyroglobulin antibody (TgAb) (>20 UI/ml) were also excluded because their serum thyroglobulin values may have been altered [13].

A total of 54 patients (16 males and 38 females; median age = 50 years; age range 15–70 years) were identified via a search of medical records. Fifty-three patients had papillary carcinomas and one patient had a follicular carcinoma. Following total thyroidectomy, patients were referred to our institution for the purposes of ablation. Their pathological TNM staging is summarized in Table 1.

Table 1 Pathological TNM classifications at the time of diagnosis
Full size table

This study was approved by our institutional review board (IRB) and all patients provided written informed consent for I-131 ablation and the FDG-PET study.

I-131 ablation

All patients were prepared for I-131 ablation by withdrawal of levothyroxine (T4) for 4 weeks and replacement with triiodothyronine (T3) for the first 2 weeks of this period. Patients were instructed to follow a low-iodine diet for 2 weeks prior to this treatment. Recombinant human thyroid stimulating hormone (rhTSH) was not administered. A dose of 3.7 GBq I-131 was administered to 51 patients, 2.96 GBq was administered to 1 patient, and 2.22 GBq was administered to 2 patients. Post-treatment whole body scans (TxWBS) were obtained at 3–4 days and at 1 week after therapy using a large field of view gamma camera (E-CAM; Toshiba Medical Systems, Tokyo, Japan) on a photo peak of 364 keV with a high-energy collimator.

FDG-PET

FDG-PET was performed 3–4 days prior to I-131 ablation. Patients were required to fast for at least 6 h prior to imaging. Fifty minutes after the intravenous injection of 185–370 MBq of FDG, emission scans of the area between the proximal femur and the base of the skull were acquired. Prior to August 2006, 20 patients were scanned using a PET scanner (HEADTOME-V, Shimadzu, Kyoto, Japan), and after November 2006, 34 patients were scanned using a PET/CT scanner (Biography 16, Siemens Medical Solutions, Erlangen, Germany).

Image interpretation

All FDG-PET scans and TxWBS were simultaneously interpreted by consensus of at least 2 experienced radiologists with over 15 years of experience in nuclear medicine image interpretation. Accumulation of FDG and radioiodine was evaluated in the thyroid bed, cervical lymph nodes, mediastinal lymph nodes, and distant metastases (lung, bone, and other sites).

Measurements of serum thyroglobulin levels

Serum thyroid stimulating hormone (TSH) and thyroglobulin (Tg) levels were measured on the treatment day with TSH stimulation and 3–6 months after ablation without TSH stimulation. TSH and Tg were measured by chemiluminescent immunoassay (CLIA) and electro-chemiluminescent immunoassay (ECLIA), respectively. Tg was considered not detectable (ND) when <1.5 ng/ml.

Statistical analysis

We compared the characteristics of PET-positive patients with those who were PET-negative by Chi-square test (gender), unpaired t test (age and serum TSH levels), and Mann–Whitney U test (serum Tg levels). The success rates for negative Tg levels after ablation were compared by Chi-square test. We also divided patients into four groups based on the presence or absence of lymph node accumulation on FDG-PET and TxWBS, and compared their serum Tg levels by Mann–Whitney U test. Excel 2007 (Microsoft Corp., Redmond, WA, USA) was used for analyses. A p value <0.05 was considered statistically significant.

Results

FDG-PET

FDG-PET imaging concurrent with I-131 ablation was positive in 25 sites of 18 patients (33%) and negative in 36 patients (67%; Table 2). FDG uptake regions included the thyroid bed (n = 9), cervical lymph nodes (n = 12), mediastinal lymph nodes (n = 3), and axillary lymph nodes (n = 1). Fourteen of 54 patients (26%) showed FDG accumulation beyond the thyroid bed. Lung and bone metastases were not observed. For these 2 groups of patients, there were no significant differences for sex, age, or their serum TSH levels on the day of treatment (Table 3). In the PET-negative group, 6 of 36 patients had negative serum Tg levels on the treatment day, while all 18 patients in the PET-positive group were positive for serum Tg. In addition, Tg levels on the treatment day were significantly higher in PET-positive patients (median = 114.1 ng/ml) than in those who were PET-negative (median = 13.8 ng/ml; p < 0.001). Five of 9 patients with an FDG accumulation in the thyroid bed showed local residual tumors in their surgical findings.

Table 2 Sites of FDG accumulation concurrent with thyroid ablation
Full size table
Table 3 Clinical features of patients in PET-positive and negative groups
Full size table

I-131 TxWBS

TxWBS were positive in 73 sites of 54 patients (98%) and negative in one patient (2%). The radioiodine uptake sites included the thyroid bed (n = 53), cervical lymph nodes (n = 12), mediastinal lymph nodes (n = 6), and lung (n = 2). Chest CT scans of 2 patients whose TxWBS showed diffuse accumulations in bilateral lungs did not detect lung metastases. The Tg values of these 2 patients were 30.0 and 289.2 ng/ml. Bone metastases were not seen.

Comparison of FDG-PET with TxWBS

Seven of 12 cervical lymph nodes (58%) that were PET-positive were negative on TxWBS (Table 4a). All 3 mediastinal lymph nodes and 1 axillary lymph node that were PET-positive were negative on TxWBS (Table 4b). Figures 1, 2 and 3 show 3 representative pre- and post-treatment scans.

Table 4 Number of cervical (a) and mediastinal (b) lymph nodes on FDG-PET and TxWBS
Full size table
Fig. 1
figure 1

A 52-year-old male patient with follicular carcinoma. a PET/CT scan shows FDG accumulation in a retropharyngeal lymph node (white arrow). b TxWBS shows radioiodine accumulation in the retropharyngeal lymph node (black arrow). Radioiodine accumulations were seen in the thyroid bed and in the midline just superior to the thyroid bed. c Cervical MRI shows a swollen lymph node (open arrow). d. Three months after ablation, the lymph node decreased in size on MRI (open arrow), although serum Tg persisted at 7.9 ng/ml. Therefore, additional radioiodine therapy was performed

Full size image
Fig. 2
figure 2

A 70-year-old female patient with papillary carcinoma. a. PET/CT shows FDG accumulations in the thyroid bed (black arrow) and mediastinal lymph nodes (open arrow). b. TxWBS shows radioiodine accumulation only in the thyroid bed (black arrow). Three months after ablation, serum Tg was 7.3 ng/ml

Full size image
Fig. 3
figure 3

A 67-year-old female patient with papillary carcinoma. a. PET/CT shows FDG accumulation in a right retropharyngeal lymph node (white arrow). b. TxWBS shows radioiodine accumulation in the thyroid bed and a retropharyngeal lymph node (open arrow). However, 6 months after ablation, serum Tg persisted at 20.9 ng/ml. Therefore, additional radioiodine therapy was performed

Full size image

Follow-up serum Tg after I-131 ablation

One patient with negative FDG-PET dropped out prior to follow-up; hence, 53 patient Tg values were evaluated after ablation. Twenty-one of 35 patients (60%) in the PET-negative group were negative for serum Tg after ablation. However, only 5 of 18 patients (28%) in the PET-positive group were negative for Tg after ablation. The success rate of a negative Tg level after ablation was significantly lower for patients with PET-positive findings than for those with negative findings (p = 0.026). In addition, the median Tg level after ablation was significantly higher in patients with PET-positive findings (median = 7.6 ng/ml; range <1.5–615.8 ng/ml) than in PET-negative patients (median = < 1.5 ng/ml; range <1.5–33.6 ng/ml; p = 0.023).

Table 5 shows the serum Tg levels before and after ablation for lymph node accumulation on FDG-PET and I-131 TxWBS (two patients with lung metastases detected on TxWBS were excluded from this analysis). For PET-negative patients, the median Tg level on the treatment day for TxWBS-negative patients was significantly lower than that for TxWBS-positive patients (6.0 vs. 44.9 ng/ml; p = 0.028), and the median Tg level after ablation for TxWBS-negative patients was also significantly lower than that for TxWBS-negative patients (<1.5 vs. 2.9 ng/ml; p = 0.007). In contrast, for PET-positive patients, there was no difference in the median Tg values on the treatment day between TxWBS-negative and TxWBS-positive patients (176.8 vs. 291.4 ng/ml; p = 0.641), and there was no difference between median values for TxWBS-negative and TxWBS-positive patients (7.3 vs. 9.6 ng/ml; p = 0.739).

Table 5 Comparisons of serum Tg levels based on presence or absence of lymph node accumulation on FDG-PET and TxWBS
Full size table

Discussion

FDG-PET imaging concurrent with initial I-131 ablation was positive for 33% of our patients after total thyroidectomy, with unexpectedly high rates of accumulation in cervical and mediastinal lymph nodes. All patients with PET-positive findings had positive Tg values on their treatment day. In addition, PET-positive patients had higher Tg values than those who were PET-negative. After initial ablation, PET-positive patients also had higher Tg values than those who were PET-negative, regardless of their results on TxWBS. Serum thyroglobulin is a sensitive tumor marker for differentiated thyroid cancer (DTC) after total thyroidectomy and ablation. Therefore, we suggest that the majority of abnormal FDG accumulations on PET and PET/CT represent residual local remnant tumors and lymph node metastases.

However, the concordance rate in lymph nodes between FDG-PET and TxWBS was low. In our study, only 5 of 16 sites (31%) of lymph node metastasis that showed positive on FDG-PET were positive on TxWBS. This disagreement between PET results and radioiodine imaging is commonly observed [5, 14–16]. Lesions that are PET-positive and TxWBS-negative are metastases of high-grade cancer cells [9]. Cells of poorly differentiated thyroid cancers usually lose some of their differentiated functions, such as the uptake of iodine, and therefore are likely to be missed on TxWBS. Therefore, for these metastases, additional I-131 treatment may not be valid.

Mazzaferri et al. [17] showed that initial ablation decreased long-term recurrence and mortality rates, although some patients had recurrence after ablation. The results of this study suggest that these recurrent cases may include poorly differentiated thyroid cancer cells, which cannot absorb radioiodine. FDG-PET concurrent with ablation can detect these patients in advance and may affect the choice and intensity of management options.

In the present study, we did not consider the effect of ablation therapy on lymph node metastases because the primary purpose of initial ablation is not to treat metastases, but to destroy residual thyroid tissue. In order to treat metastatic lesions, additional I-131 therapy or higher dose therapy (≥5.55 GBq) is necessary [18–20]. However, it is important when planning a subsequent therapeutic strategy that we can recognize the presence of any residual tumor tissue and whether or not there is accumulation of I-131 at the time of initial ablation.

In this regard, FDG-PET concurrent with I-131 ablation could have an impact on management of DTC patients. Because FDG-PET can determine both the location and extent of lymph node metastasis, it can facilitate the choice between repeated radioiodine therapy, surgery, and radiation therapy. For lymph node metastases, for which both PET and TxWBS indicate positive, repeat radioiodine therapy is necessary, as opposed to a ‘watch and see’ policy. In the present study, in PET-positive patients, there were no differences in Tg levels after ablation between TxWBS-negative and TxWBS-positive patients. We suggest that a reason was an insufficient I-131 dose to treat lymph node metastases. Therefore, for PET-positive cases, a higher activity of I-131 should be administered for retreatment (a maximum ablation dose of 3.7 GBq was administered in the present study). However, for lymph node metastases in which PET is positive and TxWBS is negative, additional surgery or irradiation should be considered. In addition, when both FDG-PET and TxWBS are negative, repeated high-dose I-131 treatment may not be necessary.

Patients with distant metastases detected by conventional imaging were excluded from our analysis. The I-131 dose was determined based on the clinical status and the treatment purpose in our institution. The I-131 dose ranged from 2.22 to 3.7 GBq only for thyroid ablations. If there was evidence of distant metastases to the lungs or bones, the activity was between 5.55 and 7.4 GBq. Therefore, all patients were screened by chest CT and DxWBS before I-131 ablation. In addition, patients with findings suspicious of bone metastases were examined by MRI or bone scintigraphy. Most distant metastases, especially lung metastases, can be detected by conventional diagnostic imaging. However, we expected that PET and PET/CT scans could detect more lesions not detected by conventional imaging. Therefore, we excluded patients in advance from this analysis who had known distant metastases. Thus, as we anticipated, unexpected residual tumors were detected by PET and PET/CT.

A limitation of this study was that no Tg level data under baseline conditions unstimulated by TSH were available. For many cases, TSH levels were not suppressed substantially before ablation because the period between total thyroidectomy and ablation was short. Therefore, we could not review whether FDG-PET should be performed for patients with undetectable Tg under baseline conditions. However, we recommend that PET should be performed at least for patients who are positive for Tg before initial I-131 ablation because a Tg-positive result, even under baseline conditions, is indicative of Tg-positive under TSH stimulation. Another limitation was that no follow-up morphological evaluations for PET-positive and TxWBS-negative metastatic lymph nodes were available. Also, not all FDG-PET findings were biopsy confirmed, which may have resulted in a small number of false positives due to inflammatory lymph nodes or postoperative inflammation.

Recently, the application of SPECT/CT for I-131 TxWBS has been reported [21–24]. SPECT/CT might be able to more accurately indicate the position of any remnant thyroid tissue and metastatic lymph nodes compared to a planar image.

In conclusion, FDG-PET concurrent with I-131 ablation after total thyroidectomy could detect unexpected abnormal FDG accumulations in 33% of our patients, which might have represented local remnant tumors and lymph node metastases. All of these patients had positive serum thyroglobulin results. Among these patients, only 31% of their lymph node metastases were positive on TxWBS. Therefore, FDG-PET concurrent with ablation can identify lymph node metastases that do not take up radioiodine and may influence the choice and intensity of management options.