Differentiated thyroid cancer (DTC) is the most common endocrine cancer1 with a yearly incidence of 8.7 per 100,000 in the United States.2 Patients with DTC usually have an excellent prognosis; the mortality of thyroid cancer (TC) is 0.5 per 100,000,2 and the 10-year cause-specific survival rate is 85%.3 However, a small fraction of patients with DTC shows a poor prognosis. Factors negatively affecting patient survival include age, tumor size, and the presence of metastases at the time of initial diagnosis.4
Radioiodine therapy with 131I (RIT) is a well-established treatment modality for adjuvant postsurgical ablation of thyroid remnant tissue and therapy of nonresectable local recurrences, lymph node, and distant metastases.3
18F-FDG PET is considered a valuable tool for the detection of radioiodine-negative metastases in the follow-up of patients with DTC.5–7 It has been observed that dedifferentiated TC lesions tend to lose radioiodine affinity, but instead show an increased glucose metabolism.8 Histological studies9 and imaging studies10–12 provide evidence that increased glucose uptake correlates with a higher grade of malignancy and a worse prognosis.
Most studies evaluated FDG PET in the later course of therapy in patients with DTC when the development of non–radioiodine-avid tumor lesions had been suspected. However, data regarding the clinical value of FDG PET at the time of thyroid remnant ablation are limited. Studies evaluating FDG PET in this scenario have shown its usefulness for detection of additional metastases,13 a correlation with ablation success rates,14 remission status,15 and an impact on patient management.13,16 Recently, an association of FDG uptake of the primary tumor (before surgery) with disease persistence/progression has been described in patients with DTC incidentally detected by PET/CT.17 However, to our knowledge, no data are available regarding the prognostic value of FDG PET at the time of thyroid remnant ablation for long-term survival of patients.
We therefore retrospectively analyzed high-risk patients with DTC who received FDG PET at thyroid remnant ablation regarding biochemical response and survival. To our knowledge, this is the first study to evaluate the predictive value of FDG PET at remnant ablation for survival of patients with DTC.
PATIENTS AND METHODS
The records of 141 patients with well-differentiated papillary or follicular TC who underwent thyroid remnant ablation after total thyroidectomy, received at least 1 further course of RIT due to suspected or proven metastases, and underwent FDG PET before remnant ablation were retrospectively analyzed. The determining factor for considering a patient for an FDG PET scan was a substantiated clinical suspicion or the proven existence of metastases. All patients were naive for prior therapeutic administrations of 131I. Patients with poorly differentiated TC were excluded.
Patients received a neck ultrasound and a CT scan of the thorax at each RIT. Survival data were acquired by routine follow-up examinations and by telephone interviews. The follow-up period ranged from 6.3 to 124.1 months (mean 54.1 ± 33.0 months).
Informed consent was given by all patients for RIT, FDG PET, and CT. This study was approved by the institutional ethics committee of Hokkaido University.
Radioiodine Therapy and Scintigraphy
Thyroid-stimulating hormone (TSH) was stimulated endogenously by withdrawal of thyroid hormone medication for 3 weeks before RIT. Administered activities were 5.0 ± 0.8 GBq (range, 3.3–5.6 GBq) at remnant ablation and 5.2 ± 0.6 GBq (range, 3.6–5.6 GBq) at second RIT (FUJIFILM RI Pharma Co Ltd, Tokyo, Japan). Patients who failed to achieve a sufficient endogenous stimulation of TSH (>30 mIU/L) were injected with TRH IV before radioiodine administration (n = 20). Posttherapy whole-body scintigraphy was performed at a mean of 4.2 ± 1.6 days after radioiodine administration using a γ-camera equipped with high-energy collimators (Millennium MG; General Electric, Elgems, Tirat Carmel, Israel). Pretreatment radioiodine scans were not performed due to concerns of potential thyroid stunning.18
FDG PET or PET/CT was performed at mean 24.6 ± 37.7 days before RIT. A total of 4.5 MBq/kg of FDG was injected intravenously. Patients fasted at least 6 hours before FDG injection. Blood glucose levels were measured in all patients before FDG injection (blood glucose exceeded 150 mg/dL in 3 cases: 173 mg/dL, 181 mg/dL, and 184 mg/dL). Static PET emission data (head to pelvis) were acquired 60 minutes after injection. PET acquisitions were performed on a Siemens EXACT HR+ scanner (Asahi-Siemens, Tokyo, Japan), with a 2-minute emission scan and a 2-minute transmission scan at each bed position. Images were reconstructed iteratively (FORE OSEM, 1 iteration, 30 subsets). PET/CT examinations were performed on a Siemens TruePoint Biograph Scanner with TrueV Option (Siemens, Japan); emission data were acquired for 3 minutes at each bed position, and a low-dose CT was acquired for attenuation correction. PET/CT images were reconstructed iteratively using OSEM (TrueX), 2 iterations, 21 subsets.
Posttherapy 131I whole-body scans and FDG PET scans were analyzed by consensus interpretation by 2 experienced nuclear medicine physicians. In case of discordant results, consensus was reached by discussion.
131I whole-body scans and FDG PET scans were evaluated on an organ system basis. Images were analyzed visually for tracer uptake in the thyroid bed, cervical and mediastinal lymph nodes, lung, bone, and uptake in other extrathyroidal organs. Patients with pathologic tracer uptake (defined by a localized area of higher uptake compared with the surrounding normal tissue, except for physiologic uptake) in any region other than the thyroid bed were categorized as radioiodine positive and/or FDG positive, respectively.
Biochemical Response Assessment
Thyroglobulin-antibody (Tg-Ab) levels were determined at the day of patient admission. Thyroid-stimulating hormone and Tg levels were measured on the day of radioiodine administration. Thyroglobulin was measured by a radioimmunoassay until 2008 and by an enzyme-linked immunosorbent assay thereafter. Assessment of biochemical response was based on the change of TSH-stimulated thyroglobulin levels (ΔTg = Tg2 ÷ Tg1). Patients with detectable Tg-Ab or who failed to achieve a sufficient endogenous TSH stimulation at first or second RIT (TSH >30 mIU/L) were excluded from Tg assessment.
Statistical analysis was performed using SPSS Statistics (Version 17; IBM) and JMP 10 (SAS Institute Japan Co Ltd). The 2-tailed Mann-Whitney U test was used to calculate the significance levels between 2 independent samples. Differences were rated as statistically significant when P was less than 0.05.
Survival was analyzed by generation of Kaplan-Meier survival curves. Survival curves were compared by the log-rank test and generalized Wilcoxon test. We used these tests for detecting differences in patterns of survival curves because it is well known that the log-rank test will be relatively more effective at detecting differences in the right tails of the curves, whereas the Wilcoxon test will be more sensitive to early differences.19
Mean patient age at the time of thyroid remnant ablation was 58.6 ± 14.2 years. The mean interval between total thyroidectomy and remnant ablation was 2.4 ± 1.1 months. The mean interval between first and second RIT was 12.2 ± 11.1 months. Mean TSH was 88.8 ± 48.7 mIU/L at thyroid remnant ablation and 96.7 ± 56.7 mIU/L at second RIT.
Further details on histology, sex, presence and location of metastases, radioiodine, and FDG uptake are listed in Table 1.
Thyroglobulin Levels at Thyroid Remnant Ablation and Biochemical Response
Overall, 80 patients were eligible for Tg analysis because patients who failed to achieve a sufficient endogenous stimulation of TSH at RIT and patients with detectable Tg-Ab were excluded from Tg analysis (insufficient TSH-stimulation, 15; detectable Tg-Ab, 41; both, 5). Detailed results of initial Tg and biochemical response are listed in Table 2.
There were no significant differences in initial Tg levels based on sex or age. Initial Tg was significantly lower in the subgroup of patients without evidence of metastases compared with metastasized patients. Thyroglobulin was noticeably (but not significantly) higher in patients with iodine-positive metastases compared with patients with iodine-negative metastases. Patients with FDG-positive metastases had significantly higher Tg levels compared with patients with FDG-negative metastases.
Regarding biochemical response, Tg showed a median reduction between first and second RIT (ΔTg) to 74.5% of the initial Tg level in all patients. There was no significant dependence of ΔTg on sex, age, and the presence of metastases, although there was a tendency for a lower ΔTg in female patients (P was only slightly above 0.05) and in patients younger than 45 years. Patients with iodine-positive metastases showed a marked decrease in Tg between first and second RIT, whereas an increase in Tg was observed in patients with iodine-negative metastases, this difference was highly significant (P < 0.01). In addition, patients with FDG-negative metastases showed a significantly better biochemical response than patients with FDG-positive metastases (P < 0.05).
Survival data could be obtained for 88 patients. Overall, 10 patients (11.4%) died during the follow-up period. All deaths were disease-related. According to the Kaplan-Meier curve, overall survival was 61.8% at the end of the observation period (124 months). Detailed results are listed in Table 3, the corresponding survival curves are presented in Figure 1.
There were no sex-dependent differences in survival. Five patients died in each group of male and female patients, resulting in an overall survival of 59.7% for male and 62.3% for female patients (P > 0.05, Fig. 1A). Regarding age and the initial presence of distant metastases, no deaths occurred in the group of young patients and in nonmetastasized patients. Overall survival was 60.6% in the group of patients 45 years or older and 57.2% in the group of patients who initially presented with distant metastases; however, the differences did not reach significance (P > 0.05, Figs. 1B, C).
The dependency of survival on radioiodine and FDG uptake was analyzed in patients with evidence of lymph node or distant metastases at thyroid remnant ablation (n = 80). Regarding radioiodine uptake, overall survival rates were 61.3% (iodine-positive) and 58.2% (iodine-negative). The difference between the survival curves was highly significant using the generalized Wilcoxon test (P < 0.01), but not using the log-rank test (P > 0.05, Fig. 1D). Regarding FDG uptake, no deaths occurred in the group of patients with FDG-negative metastases, whereas overall survival was only 48.5% in FDG-positive patients. This difference was significant using the log-rank test (P < 0.05), however, not using the generalized Wilcoxon test (P > 0.05, Fig. 1E). Subdividing the FDG-positive group into FDG+ I+ and FDG+ I– patients, the overall survival was 50.1% and 48.2%, respectively, which did not differ significantly using the log-rank test (P > 0.05). However, deaths occurred earlier in the FDG+ I– group compared with the FDG+ I+ group, which lead to a significant difference using the generalized Wilcoxon test (P < 0.05, Fig. 1F). A case example of a patient with initially FDG+ I+ metastases who died in the late time course (82 months after first RIT) is shown in Figure 2. Figure 3 depicts another case example of an initially FDG+ I– patient who died in the early time course (32 months after first RIT).
Our data suggest that FDG PET performed at thyroid remnant ablation delivers important prognostic information regarding long-term survival of patients with metastasized DTC, whereas iodine uptake status of the metastases is more important for therapy response in the early stage.
In high-risk DTC patients who presented with metastases at the time of initial diagnosis, the FDG uptake status of the metastases was of high impact for long-term survival. In our patients, FDG PET was the only significant predictive factor for survival in the long term by univariate analysis (log-rank test). These results are in line with data previously published by Alzahrani et al15 who showed that FDG PET at thyroid remnant ablation correlates with long-term remission status (up to 48 months). Our current study corroborates and complements these observations with long-term survival data up to 10 years. Furthermore, it has been shown recently that FDG PET performed at thyroid remnant ablation reveals previously unknown metastases of TC in about 14% of patients.13 Together with our current results, we therefore conclude that FDG PET should be considered at thyroid remnant ablation in high-risk patients for detection of previously unknown metastases, as well as for risk stratification and estimation of long-term prognosis.
In contrast, radioiodine uptake capacity of the metastases seems to be of higher importance for therapy response in the short term. Biochemical response (ΔTg) between first and second RIT showed a highly significant dependence on radioiodine uptake and to a less pronounced extent also on FDG uptake. The dependence of ΔTg on radioiodine uptake is expected because this represents the molecular basis of RIT and underlines its effectiveness. The less pronounced, but nevertheless significant, dependency of ΔTg on FDG uptake is consistent with the hypothesis that FDG uptake indicates the presence of less-differentiated tumor tissue,8 which exhibits a poorer response to RIT due to loss of NIS expression,20 decreasing radioiodine uptake, and thus compromising therapy efficiency.
We hypothesize that the presence of less-differentiated tumor tissue, which can be detected by FDG PET,8 is the dominating factor for outcome and survival in the long term. Well-differentiated tumor parts showing intense radioiodine uptake can be effectively controlled by RIT in the early phase; however, long-term survival is to a larger part determined by the less-differentiated portions of the tumor. This is also reflected by the time-dependent differences of the survival curves of patients with iodine-positive and iodine-negative metastases (Fig. 1D), as well as the FDG+ I+ and FDG+ I– groups (Fig. 1F). In these groups, a significant difference in survival was observed using the generalized Wilcoxon test, which emphasizes events occurring in the early time course. This underlines the effectiveness of RIT in the early phase, with deaths occurring later in iodine-positive patients compared with iodine-negative patients. However, the convergence of the survival curves at later time points, resulting in a nonsignificant difference using the log-rank test (which puts more emphasis on late occurring events), suggests a diminished antitumor effect of RIT in the later treatment course.
Robbins et al10 previously evaluated the predictive value of FDG PET in a large collective of patients with DTC in the later therapy course. In line with our current results, they also observed a high impact of FDG uptake on survival of patients with metastases. However, they did not observe differences in survival between iodine-positive and iodine-negative patients. This observation does not contradict our current results because it is most likely caused by differences in the patient collectives examined. We currently evaluated FDG PET in the early course of treatment of DTC in patients who were naive for RIT, whereas the study of Robbins et al10 mostly included patients in the later treatment course. Because it is suggested by the time-dependent difference in survival of iodine-positive and iodine-negative patients in our current data, the efficacy of RIT might be compromised in the later therapy course, thus no longer showing a significant effect on survival in the patient collective studied by Robbins et al.10
In line with our current results, Wang et al21 also observed a prognostic value of FDG PET in TC patients, whereas iodine uptake did not correlate with survival. Wang et al21 examined a mixed collective of TC patients (about 80% had received up to 5 RIT before inclusion); our current study confirms these results for the first time in high-risk patients presenting for thyroid remnant ablation after thyroidectomy.
In our patient collective, deaths only occurred in the group of older patients (≥45 years) and in the group of patients who initially presented with distant metastases; however, the factors age and presence of distant metastases at the time of initial diagnosis did not reach significance regarding survival. Reasons that these 2 factors, which are well known to have an impact on the prognosis of patients with DTC, did not reach significance in our current analysis may be the limited number of patients on the one hand, as well as our inclusion criteria on the other hand, which distinctively reduced the number of patients at lower risk in the groups of young patients and patients without metastases, thus complicating comparability regarding these factors with other studies. Therefore, our current results do not generally contradict the established prognostic importance of age and initial presence of metastases.
It is a limitation of our study that only univariate analysis of survival is available. Because DTC in general has a favorable outcome, overall, only 10 patients died during follow-up. This limited number of events would result in a low statistical power; therefore multivariate analysis was not performed. This topic needs to be addressed in future studies including larger patient collectives, preferably in prospective settings.
Our data demonstrate that FDG PET performed at the time of thyroid remnant ablation in high-risk patients with DTC is highly predictive for survival. FDG uptake had a significantly larger impact on long-term overall survival than radioiodine uptake status. On the other hand, short-term biochemical response between the first and second RIT, as well as survival in the early therapy course, was more dependent on radioiodine uptake status.
We conclude that FDG PET represents a valuable tool for risk stratification in patients with DTC. Its use at thyroid remnant ablation might be considered in high-risk patients for detection of radioiodine-negative metastases and prediction of long-term outcome. Further studies in a larger patient collective, preferably in a prospective setting, are warranted to evaluate FDG PET at thyroid remnant ablation.
1. Pellegriti G, Frasca F, Regalbuto C, et al. Worldwide increasing incidence of thyroid cancer: update on epidemiology and risk factors. J Cancer Epidemiol
2. Davies L, Welch HG. Increasing incidence of thyroid cancer in the United States, 1973-2002. JAMA
3. Luster M, Clarke SE, Dietlein M, et al. Guidelines for radioiodine therapy
of differentiated thyroid cancer
. Eur J Nucl Med Mol Imaging
4. Stokkel MP, Duchateau CS, Dragoiescu C. The value of FDG-PET in the follow-up of differentiated thyroid cancer
: a review of the literature. Q J Nucl Med Mol Imaging
5. Reske SN, Kotzerke J. FDG-PET for clinical use. Results of the 3rd German Interdisciplinary Consensus Conference, “Onko-PET III”, 21 July and 19 September 2000. Eur J Nucl Med
6. Nanni C, Rubello D, Fanti S, et al. Role of 18
F-FDG-PET and PET/CT imaging in thyroid cancer. Biomed Pharmacother
7. Shiga T, Tsukamoto E, Nakada K, et al. Comparison of (18
)I-Na, and (201
)Tl in diagnosis of recurrent or metastatic thyroid carcinoma. J Nucl Med
8. Feine U, Lietzenmayer R, Hanke JP, et al. 18
FDG whole-body PET in differentiated thyroid carcinoma. Flipflop in uptake patterns of 18
FDG and 131
I [in German]. Nuklearmedizin
9. Schonberger J, Ruschoff J, Grimm D, et al. Glucose transporter 1 gene expression is related to thyroid neoplasms with an unfavorable prognosis: an immunohistochemical study. Thyroid
10. Robbins RJ, Wan Q, Grewal RK, et al. Real-time prognosis for metastatic thyroid carcinoma based on 2-[18
F]fluoro-2-deoxy-D-glucose-positron emission tomography scanning. J Clin Endocrinol Metab
11. Deandreis D, Al Ghuzlan A, Leboulleux S, et al. Do histological, immunohistochemical, and metabolic (radioiodine and fluorodeoxyglucose uptakes) patterns of metastatic thyroid cancer correlate with patient outcome? Endocr Relat Cancer
12. Yoshio K, Sato S, Okumura Y, et al. The local efficacy of I-131 for F-18 FDG PET
positive lesions in patients with recurrent or metastatic thyroid carcinomas. Clin Nucl Med
13. Lee JW, Lee SM, Lee DH, et al. Clinical utility of 18
/CT concurrent with 131
I therapy in intermediate-to-high-risk patients with differentiated thyroid cancer
: dual-center experience with 286 patients. J Nucl Med
14. Iwano S, Kato K, Ito S, et al. FDG-PET performed concurrently with initial I-131 ablation for differentiated thyroid cancer
. Ann Nucl Med
15. Alzahrani AS, Abouzied ME, Salam SA, et al. The role of F-18-fluorodeoxyglucose positron emission tomography in the postoperative evaluation of differentiated thyroid cancer
. Eur J Endocrinol
16. Rosenbaum-Krumme SJ, Gorges R, Bockisch A, et al. (1)(8)F-FDG PET
/CT changes therapy management in high-risk DTC after first radioiodine therapy
. Eur J Nucl Med Mol Imaging
17. Piccardo A, Puntoni M, Bertagna F, et al. (1)(8)F-FDG uptake as a prognostic variable in primary differentiated thyroid cancer
incidentally detected by PET/CT: a multicentre study. Eur J Nucl Med Mol Imaging
18. McDougall IR, Iagaru A. Thyroid stunning: fact or fiction? Semin Nucl Med
19. Lawless JF. Statistical Models and Methods for Lifetime Data
. 2nd ed. Hoboken, NJ: Wiley-Interscience; 2003.
20. Moon SH, Oh YL, Choi JY, et al. Comparison of 18
F-fluorodeoxyglucose uptake with the expressions of glucose transporter type 1 and Na+/I− symporter in patients with untreated papillary thyroid carcinoma. Endocr Res
21. Wang W, Larson SM, Fazzari M, et al. Prognostic value of [18
F]fluorodeoxyglucose positron emission tomographic scanning in patients with thyroid cancer. J Clin Endocrinol Metab