Abstract
PET using 18F-FDG has been shown to effectively detect various types of cancer by their increased glucose metabolism. The aim of this study was to evaluate the use of coregistered PET and CT (PET/CT) in patients with suspected thyroid cancer recurrence. Methods: After total thyroidectomy followed by radioiodine ablation, 61 consecutive patients with elevated thyroglobulin levels or a clinical suspicion of recurrent disease underwent 18F-FDG PET/CT. Of these, 59 patients had negative findings on radioiodine (131I) whole-body scintigraphy (WBS). Fifty-three of the 61 patients had both negative 131I WBS findings and elevated thyroglobulin levels. PET/CT images were acquired 60 min after intravenous injection of 400–610 MBq of 18F-FDG using a combined PET/CT scanner. Any increased 18F-FDG uptake was compared with the coregistered CT image to differentiate physiologic from pathologic tracer uptake. 18F-FDG PET/CT findings were correlated with the findings of histology, postradioiodine WBS, ultrasound, or clinical follow-up serving as a reference. The diagnostic accuracy of 18F-FDG PET/CT was evaluated for the entire patient group and for those patients with serum thyroglobulin levels of less than 5, 5–10, and more than 10 ng/mL. Results: Thirty patients had positive findings on 18F-FDG PET/CT; 26 were true-positive and 4 were false-positive. In 2 patients, increased 18F-FDG uptake identified a second primary malignancy. 18F-FDG PET/CT results were true-negative in 19 patients and false-negative in 12 patients. The overall sensitivity, specificity, and accuracy of 18F-FDG PET/CT were 68.4%, 82.4%, and 73.8%, respectively. The sensitivities of 18F-FDG PET/CT at serum thyroglobulin levels of less than 5, 5–10, and more than 10 ng/mL were 60%, 63%, and 72%, respectively. Clinical management changed for 27 (44%) of 61 patients, including surgery, radiation therapy, or chemotherapy. Conclusion: Coregistered 18F-FDG PET/CT can provide precise anatomic localization of recurrent or metastatic thyroid carcinoma, leading to improved diagnostic accuracy, and can guide therapeutic management. In addition, the findings of this study suggest that further assessment of 131I WBS–negative, thyroglobulin-positive patients by 18F-FDG PET/CT may aid in the clinical management of selected cases regardless of the thyroglobulin level.
Serum thyroglobulin and radioiodine whole-body scintigraphy (WBS) play an important role in the postsurgical follow-up of differentiated thyroid cancer (1–3). Elevated thyroglobulin levels are a sensitive marker for residual or recurrent disease (4). Radioiodine WBS using 123I or 131I allows localization of local recurrences or distant metastases and aids in the decision on subsequent radioiodine treatment (4–6). However, radioiodine WBS shows negative findings in 10%–15% of patients with detectable serum thyroglobulin levels (5,7,8). At least 2 factors may account for the discrepancy between serum thyroglobulin levels and radioiodine WBS. First, the tumor volume might be too small to be detected by WBS, and second, tumor cells may lose the ability to trap radioiodine while still retaining the ability to secrete thyroglobulin (8,9).
It is important to localize tumor sites to initiate the appropriate treatment, such as surgery or external-beam radiotherapy (10,11). Various imaging modalities including ultrasonography, CT, bone scintigraphy, and MRI are currently being used for further diagnostic evaluation. In addition, several radiopharmaceuticals such as 201Tl, 99mTc-sestamibi, 99mTc-tetrofosmin, and 111In-octreotide have been evaluated for detection of recurrent or metastatic thyroid cancer (12–15).
Cancer frequently exhibits increased glucose metabolism that can be visualized on 18F-FDG PET. Differentiated thyroid cancer is not generally characterized by a marked increased 18F-FDG uptake (16). Several groups have studied 18F-FDG PET in detecting metastatic or recurrent non–radioiodine-avid lesions (10,11,17–19). However, most of these studies were conducted with relatively small patient sample sizes and exact tumor localization was often difficult because of the lack of anatomic landmarks, particularly in the neck region, and the limited spatial resolution of PET (20,21). Furthermore, variable physiologic 18F-FDG uptake in muscle, brown adipose tissue, and lymphoid tissue or nonmalignant lesions can confound image interpretation (20–22).
The use of combined 18F-FDG PET/CT has the potential to increase the diagnostic accuracy of 18F-FDG PET by providing coregistered metabolic and anatomic information (20). There is limited information available so far describing the role of 18F-FDG PET/CT in differentiated thyroid cancer patients. The aims of this study were to evaluate the use of coregistered 18F-FDG PET and CT (18F-FDG PET/CT) in 131I WBS–negative, thyroglobulin-positive patients with suspected thyroid cancer recurrence and to assess the diagnostic performance of 18F-FDG PET/CT in a subset of patients with thyroglobulin values less than 10 ng/mL or more than 10 ng/mL.
MATERIALS AND METHODS
Patients
18F-FDG PET/CT scans from 61 consecutive patients studied at the University of Pittsburgh Medical Center were retrospectively examined. All data were acquired and managed with the prior approval of the Institutional Review Board at the University of Pittsburgh. All patients had undergone a previous total thyroidectomy for well-differentiated thyroid cancer followed by 131I ablation of residual thyroid tissue. Ablation doses generally ranged from 3.7 to 5.5 GBq (100–150 mCi) of 131I. Fifty-nine patients presented for 18F-FDG PET/CT imaging after negative 131I WBS results (53 of whom had both negative 131I WBS results and elevated thyroglobulin levels) and 2 patients underwent 18F-FDG PET/CT because of suggestive findings on other imaging modalities. Thyroglobulin status at the time of 18F-FDG PET/CT was as follows: Forty-six patients had elevated thyroglobulin (>2 ng/mL) and elevated thyroid-stimulating hormone (TSH) levels upon thyroid hormone withdrawal, 7 patients had elevated thyroglobulin levels (>1 ng/mL) while on thyroid hormone medication with suppressed TSH, and the remaining 8 patients had nondetectable thyroglobulin levels (<1.0 ng/mL), with 18F-FDG PET/CT performed for restaging because of elevated thyroglobulin antibodies or a high clinical suspicion of recurrent disease. For detailed patient demographics, see Table 1.
Patient Demographics (n = 61)
18F-FDG PET/CT Imaging
Patients fasted for at least 6 h before 18F-FDG PET/CT imaging with the exception of liberal water intake. An intravenous catheter was placed for radiopharmaceutical administration, and the blood glucose level was measured before tracer injection. The blood glucose levels of all patients were less than 150 mg/dL (mean, 96 ± 24 mg/dL) at the time that the 18F-FDG was injected. Each patient received 400–610 MBq (11–16.5 mCi) of 18F-FDG intravenously. After tracer injection, the patients rested on a comfortable chair during the 18F-FDG uptake period. PET/CT was initiated 60 min after injection of the 18F-FDG, and a dual-slice lutetium oxyorthosilicate PET/CT scanner (Siemens) was used. CT was performed before acquisition of the PET data in a single step with the patients supine. First, a scout scan was obtained to determine the axial range of the study. The scanning parameters for whole-body CT craniocaudal scanning were 130 kV, 80–120 mAs, 5-mm collimation, and a pitch of 1.6. During the scan, patients were asked to maintain shallow respiration. The subsequent 3-dimensional PET data acquisition included 4–6 bed positions, (4 min per bed position) over the same axial extent. The PET acquisition included a dead-time correction and online delayed coincidence subtraction to correct for random coincidences. The helical CT scan was reconstructed by filtered backprojection into 512 × 512 pixel images with a slice thickness of 2.4 mm to match the PET scan. Images were reconstructed using ordered-subset expectation maximization with 2 full iterations of 8 subsets. Rescaled CT images were used to produce attenuation correction values for the PET emission reconstruction.
18F-FDG PET/CT Analysis
All 18F-FDG PET/CT scans were interpreted by specialized head and neck radiologists in conjunction with nuclear medicine physicians. The 18F-FDG PET portion and the CT portion of PET/CT were jointly interpreted using a dedicated image fusion workstation, and a final consensus was reached for all patients. The leading criterion for 18F-FDG PET/CT interpretation was the presence of focally increased 18F-FDG uptake. Therefore, any increased 18F-FDG uptake was compared with the anatomic finding on CT. Areas of increased 18F-FDG uptake corresponding to normal structures such as salivary glands, muscle, fat, and lymphoid tissue were not recorded. All areas with abnormally increased 18F-FDG uptake corresponding to a CT abnormality (tissue mass or lymph node) were interpreted as positive for recurrent disease. In addition, focally increased 18F-FDG that did not correspond to normal structures or any other structural findings was recorded as positive. Suggestive findings on CT were interpreted as negative if they did not correspond to an area of abnormally increased 18F-FDG uptake.
The results of 18F-FDG PET/CT were correlated with patient follow-up information, which included the results from subsequent imaging modalities such as neck ultrasound, MRI, CT, and postradioiodine treatment scanning; thyroglobulin levels; and histologic examination of surgical specimens. The 18F-FDG PET/CT findings were classified as follows:
Lesions were true-positive if positive findings on 18F-FDG PET/CT were confirmed by the presence of carcinoma on histologic examination or were confirmed by other imaging modalities in the presence of persistent abnormal or increasing thyroglobulin levels.
Lesions were false-positive if biopsy samples of suggestive lesions were negative for carcinoma on histologic examination or the lesions had resolved on subsequent follow-up imaging. The presence of a second primary tumor was also considered a false-positive finding.
Lesions were true-negative if the findings of 18F-FDG PET/CT were negative, if elevated thyroglobulin had normalized without treatment, and if metastatic disease was not evident on subsequent follow-up. Lesions were also true-negative if there was no change in thyroglobulin level (only for levels between 2 and 8 ng/mL) and patients had not received subsequent treatment for at least 12 mo. Follow-up was continued in all patients with true-negative lesions for at least 12 mo.
Lesions were false-negative if the findings of 18F-FDG PET/CT were negative and metastatic thyroid cancer was found at histologic examination of surgical biopsy specimens or if disease progression was seen on other imaging modalities, such as posttreatment radioiodine scans. 18F-FDG PET/CT findings were also false-negative if patients had persistent elevated thyroglobulin levels (>8 ng/mL) or rising thyroglobulin levels.
Statistical Analysis
Sensitivity, specificity, positive and negative predictive values, and accuracy were calculated for 18F-FDG PET/CT for all patients and for subgroups with thyroglobulin levels less than 5, 5–10, or more than 10 ng/mL. All patients in this study underwent follow-up, for a median of 18 mo (range, 12–30 mo).
RESULTS
18F-FDG PET/CT
The findings of 18F-FDG PET/CT were positive in 30 of 61 patients (49.2%) and negative in 31 patients (50.8%). PET/CT findings were true-positive in 26 of 30 studies; 23 of 26 true-positive findings were confirmed by histologic examination of surgical biopsy specimens; 2 of 26 underwent empiric radioiodine treatment and had positive posttreatment scan results, and 1 patient had a further increase in serum thyroglobulin levels and had progression of disease on follow-up imaging. In the true-positive group, 11 of the 26 patients had local recurrences or lymph node metastases in the neck, 9 had both local and distant recurrent disease, and 4 had distant metastases only. Illustrations of true-positive 18F-FDG PET/CT findings are shown in Figures 1 and 2. Detailed 18F-FDG PET/CT results are summarized in Table 2.
A 47-y-old woman with history of papillary thyroid cancer who underwent thyroidectomy and radioiodine ablation. Two years later, patient presented with thyroglobulin level of 6.1 ng/mL (TSH suppressed) and negative findings on 131I WBS. 18F-FDG PET (A) demonstrates small foci of increased 18F-FDG uptake (arrows) corresponding to small lymph nodes in right lower neck on CT (B). These are clearly visualized on fused 18F-FDG PET/CT (C) and were subsequently proven to be thyroid cancer metastases.
A 62-y-old man with history of papillary thyroid cancer. 131I WBS and 18F-FDG PET/CT were performed 1 y previously and had negative findings (thyroglobulin, 6.9 ng/mL). At current presentation, patient has thyroglobulin level of 29 ng/mL and small focal area of increased 18F-FDG uptake in left neck (arrow). Focus was proven by histopathology to be thyroid cancer metastasis. Shown are 18F-FDG PET (A), CT (B), and fused 18F-FDG PET/CT (C) images.
18F-FDG PET/CT Results
The findings of 18F-FDG PET/CT were false-positive in 4 patients. One patient had a histologically confirmed parathyroid adenoma; 1 had a focal abscess that had formed at the area of prior surgery, 1 underwent surgery that revealed only fat and lymphoid tissue on histologic examination, and 1 had a second primary malignant tumor in the lung.
The findings of 18F-FDG PET/CT were true-negative in 19 of 61 patients as validated by thyroglobulin levels or histologic examination of surgical specimens. On follow-up 17 of these 19 patients had decreasing or negative levels of serum thyroglobulin levels (2–8 ng/mL for at least 12 mo was considered stable). Two patients underwent surgery for suggestive findings identified by other imaging modalities, but histologic examinations of the surgical specimens were negative for thyroid carcinoma; these patients also had stable thyroglobulin levels (<8 ng/mL). For at least 12 mo, none of these patients had received a specific treatment.
The findings of 18F-FDG PET/CT were false-negative in 12 of 61 patients. Four patients underwent surgery because of suggestive findings on other imaging modalities, with histologic examination of surgical specimens revealing thyroid carcinoma. Another 4 patients had rising thyroglobulin levels (>8 ng/mL) during follow-up, and the remaining 4 patients underwent empiric radioiodine treatment and had positive findings on posttreatment scans.
18F-FDG PET/CT showed a second primary tumor in 2 patients, a breast carcinoma in one and lung cancer in the other. The findings of 18F-FDG PET/CT in the first patient were classified as true-positive because an area of increased 18F-FDG uptake in the neck turned out to be a local recurrence of papillary thyroid cancer. An additional area of focally increased 18F-FDG uptake in the right breast was moderately differentiated infiltrating ductal carcinoma of the breast. The findings of 18F-FDG PET/CT in the second patient, a 64-y-old woman with a history of Hürthle cell thyroid cancer, were classified as false-positive because the focus of increased 18F-FDG uptake turned out to be a moderately differentiated adenocarcinoma of the lung with mucinous differentiation, which did not explain the increased thyroglobulin level.
Diagnostic Accuracy of 18F-FDG PET/CT and Serum Thyroglobulin Levels
When the sensitivity of 18F-FDG PET/CT was compared for different thyroglobulin levels, the sensitivity increased from 60% for thyroglobulin levels less than 5 ng/mL to 62.5% for thyroglobulin levels ranging from 5 to 10 ng/mL and to 72% for thyroglobulin levels higher than 10 ng/mL (Table 2). The findings of 18F-FDG PET/CT were true-positive in 14%, 45%, and 62% of patients with thyroglobulin levels less than 5, 5–10, and more than 10 ng/mL, respectively. Table 3 shows true-positive 18F-FDG PET/CT findings with respect to serum thyroglobulin and TSH levels.
True-Positive 18F-FDG PET/CT Findings Compared with Serum Thyroglobulin Levels for Stimulated and Suppressed TSH Levels
Change in Treatment After 18F-FDG PET/CT
In 27 of the 61 patients (44%), 18F-FDG PET/CT resulted in subsequent treatment changes. Surgery was performed on 23 patients for whom 18F-FDG PET/CT suggested resectable tumor recurrence. 18F-FDG PET/CT showed local recurrences or metastases in the neck in 20 of 23 patients who underwent selective neck dissection. Eleven of these 20 patients had recurrent disease in the neck only, but 9 patients had additional tumor localizations in the mediastinum (n = 3), lungs (n = 3), and bone (n = 3). The remaining 3 of 23 patients showed no evidence of recurrences or metastases in the neck, and 18F-FDG PET/CT correctly identified distant metastases in the lungs (n = 2) and mediastinum (n = 1). Two of 27 patients had multiple nonresectable metastases identified by 18F-FDG PET/CT and received chemotherapy. Four patients underwent external-beam radiation, and 2 of these 4 underwent this treatment after initial surgery for tumor recurrence. 18F-FDG PET/CT identified a second primary tumor in 2 patients, breast carcinoma in one and lung cancer in the other.
DISCUSSION
In our study of 61 patients, we found an overall sensitivity and specificity of 68.4% and 82.4%, respectively, for detecting and localizing recurrent or metastatic differentiated thyroid cancer on coregistered 18F-FDG PET/CT. An important advantage of PET/CT is its high specificity, compared with previously reported specificities using 18F-FDG PET alone (10,11,17,18,21). There are several potential sources for false-positive findings, including difficult-to-interpret 18F-FDG uptake in muscle and brown fat, the salivary glands, vocal cords, tonsils, and other lymphoid tissues (23). In our study, the coregistered metabolic and anatomic information from PET/CT allowed for improved differentiation between physiologic and pathologic 18F-FDG uptake. A recent study compared the PET results of PET/CT with combined PET/CT and found an increase in specificity from 50% to 89% (20). Other groups reported similar results (24,25). The 4 false-positive cases in our series—a histologically confirmed parathyroid adenoma, a focal abscess in the surgical bed, a specimen containing fat and lymphoid tissue, and a second primary tumor in the lung—are not unexpected in a larger patient sample size. A second primary tumor was classified as false-positive because the positive PET finding did not explain the source of the increased thyroglobulin level.
The sensitivity of 18F-FDG PET/CT in our patient group is comparable to previously reported sensitivities (10,11,17). In one of the largest 18F-FDG PET studies, on 64 patients with suspected recurrent disease (11), the sensitivity was 69.4%, which favorably compares with 68.4% in our study. The sensitivity of 18F-FDG PET is generally determined by the metabolic activity of recurrent or metastatic thyroid cancer. An inverse relationship has been reported between the ability to concentrate radioiodine and the uptake of 18F-FDG in thyroid cancer (8,16,18). In most tumors, 18F-FDG uptake increases with the level of dedifferentiation (16) and was, for example, found to correlate with the expression of the glucose transporter protein 1 (26).
18F-FDG PET is especially helpful for detecting recurrent or metastatic lesions in patients with increased thyroglobulin levels and negative radioiodine WBS findings (10,11,16–19). In our study, 59 of 61 patients presented with negative 131I WBS findings, 53 of whom had both negative WBS findings and elevated thyroglobulin levels. The sensitivities of 18F-FDG PET/CT at serum thyroglobulin levels of less than 5, 5–10, and more than 10 ng/mL were 60%, 63%, and 72%, respectively. In the United States, most insurance providers cover 18F-FDG PET in patients with a thyroglobulin level greater than 10 ng/mL. However, our results suggest that 18F-FDG PET/CT could also be helpful in selected cases of thyroglobulin levels less than 10 ng/mL. The low specificity in the subgroup with thyroglobulin levels less than 10 ng/mL was related to the 3 false-positive findings, which were, however, independent of the thyroglobulin level. As shown in Table 3, we found no obvious difference in the diagnostic performance of 18F-FDG PET/CT carried out at suppressed (<1 U/L) or stimulated (>25 U/L) TSH.
An important observation is that in 20 of 26 true-positive cases, local recurrences or lymph node metastases were located in the neck. Physical palpation is limited in the postsurgical neck, and various imaging modalities are therefore applied to detect recurrent disease. CT of the neck is difficult to interpret without administering intravenous contrast material, and MRI, although a sensitive tool, lacks specificity (27). Ultrasound has an emerging role, with advances in technology such as high-resolution phased-array transducers, color flow Doppler, and power Doppler providing detailed information and improved detection of local recurrences and lymph node metastases (28). In addition, ultrasound offers the advantage of immediate biopsy of suggestive lesions. Future studies will be needed to compare both the clinical effectiveness and the cost effectiveness of neck ultrasound versus 18F-FDG PET/CT. Of note, in 14 patients, 18F-FDG PET/CT identified tumor deposits outside the neck, predominantly in the chest and bone. An invaluable advantage of 18F-FDG PET and 18F-FDG PET/CT is the ability to identify thyroid cancer recurrences and metastases in soft tissue, lymph nodes, liver, lungs, and bone in a single imaging procedure.
18F-FDG PET/CT provided important diagnostic information that changed patient management in 27 (44%) of 61 patients. A crucial advantage of coregistered 18F-FDG PET/CT is the precise localization of local recurrences and distant metastatic disease. This information, for example, improved surgical planning and target definition for external-beam radiation. 18F-FDG PET/CT directed the treatment to neck dissections, thoracotomy, wedge resections of the lungs, radiation therapy, or chemotherapy in cases with multiple metastases. Although the exact contribution of 18F-FDG PET/CT to patient management is difficult to quantify, its clinical relevance is probably best illustrated by the fact that most patients had previous negative findings on radioiodine WBS and underwent an extensive diagnostic work-up before being referred for 18F-FDG PET/CT. Our findings are in line with the findings of other studies: In one, a change in management was found in 22 (67%) of 33 patients (24), and in another, 17 (74%) of 23 patients showed increased 18F-FDG uptake (25).
Although this study reports on one of the largest patient groups examined with 18F-FDG PET or PET/CT, the number of patients is still limited in the subgroup analysis. We did not perform an independent analysis of 18F-FDG PET versus 18F-FDG PET/CT because the limitations of 18F-FDG PET alone for tumor detection are well documented (29). We also did not compare the diagnostic accuracy of 18F-FDG PET versus CT because, on patients with differentiated thyroid cancer, CT is generally performed without intravenous contrast material, which compromises the diagnostic information provided. Follow-up, including histopathologic results, served as the endpoint for assessing the diagnostic accuracy of 18F-FDG PET/CT. After thyroidectomy and radioiodine ablation, rising thyroglobulin levels are generally a reliable indicator of recurrent thyroid cancer. We assumed the presence of disease in patients with persistent elevated (>8 ng/mL) or rising thyroglobulin levels, although this assumption might not be true in all cases. In addition, verification of true-negative findings is difficult in these patients. 18F-FDG PET/CT was considered true-negative in 19 of 61 patients defined by stable (2–8 ng/mL), decreasing, or negative follow-up thyroglobulin levels. Thyroglobulin autoantibodies or heterophilic antibodies (human antimurine antibodies), which could interfere with thyroglobulin measurements, were not available for all patients.
CONCLUSION
Coregistered 18F-FDG PET/CT provided precise anatomic localization of recurrent or metastatic thyroid carcinoma, leading to improved diagnostic accuracy. The use of 18F-FDG PET/CT altered patient management and was also helpful in guiding the site of therapeutic intervention. 18F-FDG PET/CT appears to be useful, specifically in patients with elevated thyroglobulin levels and negative radioiodine whole-bode scintigraphy findings. Our results also show that 18F-FDG PET/CT adds valuable diagnostic information in selected cases regardless of the thyroglobulin level.
Acknowledgments
We very much appreciate the help of Dr. Serguei Roumiantsev in the data collection and analysis of this study.
Footnotes
-
COPYRIGHT © 2007 by the Society of Nuclear Medicine, Inc.
References
- Received for publication August 25, 2006.
- Accepted for publication November 3, 2006.
Jump to section
Related Articles
Cited By...
- The role of FDG-PET in localization of recurrent lesions of differentiated thyroid cancer (DTC) in patients with asymptomatic hyperthyroglobulinemia in a real clinical practice
- IMAGING IN ENDOCRINOLOGY: 2-[18F]-fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography in differentiated thyroid carcinoma: clinical indications and controversies in diagnosis and follow-up