Does Recombinant Human Thyrotropin-Stimulated Positron Emission Tomography with [18F]Fluoro-2-Deoxy- d -Glucose Improve Detection of Recurrence of Well-Differentiated Thyroid Carcinoma in Patients with Low Serum Thyroglobulin?

Abstract

Background:

Thyrotropin (TSH) stimulates thyrocyte metabolism, glucose transport, and glycolysis. The interest in using recombinant human TSH (rhTSH) stimulation of fluoro-2-deoxy-d-glucose (FDG) with positron emission tomography (PET) has been shown, but mainly for patients with high serum thyroglobulin (Tg) concentration. We evaluated the use of rhTSH-stimulated PET-FDG in patients with low serum Tg concentration.

Methods:

Sixty-one PET/computed tomography (CT)-FDG (Biograph Sensation 16; Siemens Medical Solutions, Knoxville, TN) were performed in 44 patients (28 women and 16 men; 51 ± 16 years) with positive Tg levels, negative or no contributive iodine-131 whole-body scintigraphy results, and no contributive morphological imaging results (ultrasound, magnetic resonance imaging, and CT). Thirty-eight patients had papillary carcinoma and six had follicular thyroid carcinoma. All patients had previously undergone total thyroidectomy and postoperative iodine ablation of thyroid bed remnant tissue. The rhTSH-stimulated PET/CT-FDG (5 MBq/kg) was performed after two 0.9 mg intramuscular doses of rhTSH (Thyrogen^®; Genzyme) which were administered 48 and 24 hours before imaging, while patients continued levothyroxine (LT₄). Blood sampling was performed immediately before FDG injection for measurement of serum TSH and Tg concentrations (TSH₁ and Tg₁) and after 48 hours (TSH₂ and Tg₂). PET/CT-FDG findings were compared with the Tg: (i) at the initial iodine treatment during T₄ withdrawal (Tg_ini), (ii) under T₄ (Tg_T4) within 3 months before the PET/CT-FDG, (iii) with Tg₁, and (iv) with Tg₂. PET/CT-FDG findings were correlated with the findings of histology, iodine-131 whole-body scintigraphy, morphological imaging, or clinical follow-up.

Results:

The mean Tg_ini was 785 ± 2707 μg/L for a TSH of 73 ± 64 mU/L. The mean Tg_T4 was 7 ± 15 μg/L (T₄ = 195 ± 59 μg/day; mean TSH of 0.24 ± 0.57 mU/L). Among the 44 patients, PET/CT-FDG findings were positive in 20 and negative in 24. Among the 61 PET/CT-FDG, 25 PET/CT-FDG were positive (41%). Among the 25 positive PET, the Tg_T4 values were less than 10 μg/L for 19, including 9 true-positive patients (20% of the 44 patients). There was no difference of PET/CT-FDG results (positive vs. negative) as related to the serum Tg concentrations (p = 0.99 for Tg_ini, p = 0.95 for Tg_T4, p = 0.07 for Tg₁, and p = 0.42 for Tg₂). No relation was observed with PET/CT-FDG results and initial tumor size (p = 0.52) or node metastasis (p = 0.14).

Conclusion:

In the diagnosis of recurrent disease in patients with differentiated thyroid carcinoma and low Tg level, the sensitivity of rhTSH-stimulated PET/CT-FDG seems to be low and no correlation was observed between PET/CT-FDG findings and Tg level. However, positive PET-FDG results have been found in 9/44 (20%) patients with serum Tg levels lower than 10 μg/L. Therefore, this series shows that a cutoff value of 10 μg/L for the Tg under T₄ is probably not the best criteria to select patient candidates for PET/CT-FDG examination to detect the recurrence of differentiated thyroid carcinoma.

Introduction

S erum thyroglobulin (Tg) measurements and iodine-131 whole-body scintigraphy (WBS) play an important role in the postsurgical follow-up of patients with well-differentiated thyroid carcinoma (1 –3). However, WBSs are negative in 10–15% of patients with detectable serum Tg (4,5).

The value of positron emission tomography (PET) using fluoro-2-deoxy-d-glucose (FDG) in the diagnosis of recurrent and metastatic disease, particularly when WBS is negative, has been assessed by many authors (Table 1). Most importantly, we were interested in performing the present study because the performance of PET-FDG had been described to be related to the serum Tg level; the cutoff value was considered to be 10 μg/L, with good performance of PET-FDG only in patients who had serum Tg concentrations greater than 10 μg/L while on levothyroxine (LT₄) therapy (6,7).

Table 1.

Review of Literature from 1996 to 2008 That Studied the Interest in Using Positron Emission Tomography–Fluoro-2-Deoxy-d-Glucose for Thyroid Cancer Detection (Series That Expressed Diagnostic Performance and/or Mean Thyroglobulin Levels)

Authors	n	Se (%)	Sp (%)	Tg (μg/L) on T₄	Tg (μg/L) on rhTSH	Previous treatment	Reason for PET-FDG [delay (months)]	Delay between FDG injection and imaging	Duration of imaging
Feine (30)	41	95 (FDG and WBS)				Thyroidectomy + iodine ablation	Suspicion of recurrent disease	45 min	5–7 min/bed
Dietlein et al. (31)	58	86 (FDG and WBS)				Thyroidectomy + iodine ablation	3–24 months since last ablation
Wang et al. (32)	37			6500 ± 36,132		Total thyroidectomy, 28/37 Subtotal thyroidectomy, 3/37 Hemithyroidectomy, 6/37 Iodine ablation, 34/37	83 months since diagnosis [2.5–456]	45–60 min	6 min/bed
Chung et al. (22)	54	94	95	195 ± 183 [47–427]	>10	Thyroidectomy + iodine ablation	Suspicion of recurrent disease	60 min	30 min
Grunwald et al. (33)	222	75 (85 if WBS−)	90			Thyroidectomy + iodine ablation	Suspicion of recurrent disease	30 min
Muros et al. (34)	10				210 ± 20 [20.6–568]	Thyroidectomy + iodine ablation	Suspicion of recurrent disease	45 min	20–30 min
Moog et al. (8)	10			2982 ± 5085 [27–16,868]	6301 ± 7519 [124–21,780]	Thyroidectomy + iodine ablation	46 months since surgery	76 min	7 min/bed
Wang et al. (24)	125					Total thyroidectomy, 92/125 Subtotal thyroidectomy, 13/125 Hemithyroidectomy, 17/125 No surgery, 3/125 Iodine ablation, 101/125	63 months since diagnosis [1.5–448.5]	45–60 min	6 min/bed
Helal et al. (23)	37			199 ± 563	1923 ± 4961	Thyroidectomy + iodine ablation	Suspicion of recurrent disease	50–60 min	7 min/bed
Schluter et al. (6)	64						Suspicion of recurrent disease	60 min	8 min/bed
Frilling et al. (35)	24	95	25			Thyroidectomy + iodine ablation	Suspicion of recurrent disease	60 min	6 min/bed
Petrich et al. (12)	30			540 ± 2749 [<0.3–15,080]	626 ± 3013 [<0.3–15,080]		4.9 years since treatment [0.1–12.1]	90 min
Lind et al. (15)	161	77				Thyroidectomy + iodine ablation	Suspicion of recurrent disease
Chin et al. (13)	7			7 ± 7 [0.9–19.3]	21 ± 28 [1–80]	Thyroidectomy + iodine ablation	Suspicion of recurrent disease	45–50 min
Iwata et al. (21)	19			2263 ± 4563 [6.1–19,668]	13,777 ± 22,397 [71–80,424]	Thyroidectomy, 17/19 Iodine ablation, 2/19	Before ablation (diagnosis)	60 min
Nahas et al. (37)	33	66	100		126 [4.5–372]50 [5.2–123]10 [1.2–41.6]	Thyroidectomy + iodine ablation	Suspicion of recurrent disease
Ong et al. (38)	17	88				Thyroidectomy + iodine ablation	Done few times after the first ablation	50–70 min
Hooft et al. (28)	19			196 ± 425 [2–1455]	1264 ± 3548 [9–14,300]		7–107 months since diagnosis	60 min	5 min/bed
Palmedo et al. (39)	40	95 (PET/CT)79 (PET)	91 (PET/CT)76 (PET)			Thyroidectomy + iodine ablation	Suspicion of recurrent disease	60–90 min
Saab et al. (27)	15			59 ± 165 [<1–630]	234 ± 626 [5–2470]	Thyroidectomy + iodine ablation	Suspicion of recurrent disease	60 min
Iagaru et al. (40)	76	87	89					60 min	4 min/bed
Freudenberg et al. (41)	21	70			22 ± 38 [<0.3–156]	Thyroidectomy + iodine ablation	21–75 months since last ablation	60 min
Mirallié et al. (25)	45	57–67	77		11/24/22 [0.4–1978]	Thyroidectomy + iodine ablation	Suspicion of recurrent disease	60 min
Shammas et al. (7)	61	60–72	25–93			Thyroidectomy + iodine ablation	Suspicion of recurrent disease	60 min	4 min/bed
Zuijdwijk et al. (42)	39	84–100	88–100		≥3	Thyroidectomy + iodine ablation	Persisting elevated Tg; restaging in patients with known recurrence; clinically suspected residual disease	60 min
Present study	44			7 ± 15 [0–89]	34 ± 85 [0–476]	Thyroidectomy + iodine ablation	Suspicion of recurrent disease	60 min	3 min/bed

All studies were performed for suspicion of recurrent disease, or staging (19). Number of patients (n), sensitivity (Se) and specificity (Sp), mean ± standard deviation [range] of Tg under LT₄ or after rhTSH, previous treatment before PET-FDG, reason for PET-FDG, delay between FDG injection and imaging, and duration of PET examination are shown. For Mirallié et al. (25) and Nahas et al. (37), mean Tg level are expressed regarding three different groups of patients.

CT, computed tomography; LT4, levothyroxine; PET-FDG, positron emission tomography–fluoro-2-deoxy-d-glucose; rhTSH, recombinant human thyrotropin; Tg, thyroglobulin; WBS, iodine-131 whole-body scintigraphy.

Moog et al. (8) suggested that the sensitivity of PET-FDG could be increased by discontinuing LT₄ treatment in patients who had undergone thyroidectomy for differentiated thyroid cancer. This conclusion was supported by experimental studies which showed that thyrotropin (TSH) stimulates glucose transport and Glut1 expression in cultured thyroid cells (9 –11). In 30 patients, with Tg levels superior to 0.3 μg/L in combination with negative or equivocal WBS and/or morphological imaging, Petrich et al. (12) showed an increase of FDG uptake measured by PET after recombinant human TSH (rhTSH) administration, in local recurrence and distant metastasis, but not in inflammatory lesions or background. Chin et al. (13) have shown in seven patients that rhTSH stimulation improves the detectability of occult thyroid metastases with PET-FDG, compared with scans performed on TSH suppression.

In view of these results, we hypothesized that rhTSH-stimulated PET-FDG may allow detection of locoregional recurrence or distant metastasis in patients who have had thyroid surgery for thyroid cancer and whose serum Tg is detectable but relatively low (mean, 7 ± 15 μg/L in our series).

Materials and Methods

Patients

Forty-four patients (28 females and 16 males) with well-differentiated thyroid carcinoma were prospectively enrolled in the study. Their age was 51 ± 16 years (range, 9–83; median, 52). The inclusion criteria were Tg after rhTSH > 2 μg/L during the rhTSH test (threshold of our laboratory) in combination with negative or no contributive WBS and no contributive morphological imaging (neck ultrasound, enhanced computed tomography [CT], or magnetic resonance imaging). The study was reviewed by our local committee, and informed consent was obtained from all patients. No criterion was required on Tg before rhTSH administration. Exclusion criteria were uncontrolled diabetes (glycemia ≥ 10 mM) and pregnancy or nursing.

Thirty-eight patients had papillary and six had follicular thyroid carcinoma. The TNM stages ranged from pT1N0M0 to pT4pN1_bpM1 (International Union Against Cancer, 2005). The World Health Organization classification was stages I–IV. The mean time between thyroid surgery and the study phase was 6 ± 5 years (range, 1–23; median, 4). All patients had previously undergone total or near total thyroidectomy and postoperative iodine ablation of thyroid bed remnant tissue (3 ± 3 treatments [range, 1–19; median, 3]), with administered dose of 13,323 ± 10,264 MBq [range, 3700–67,710; median, 11,100].

Laboratory tests

TSH was determined using chemoluminescence tests (Nichols Advantage until March 2006, Liaison DiaSorin after [Saluggia, Italy]). Tg dosages were determined with radioimmunoassay (Brahms, Henningsdorf, Germany) until February 2003, then by immunofluorimetry in homogeneous phase (Kryptor^®; Brahms). Results were expressed with respect to the international standard CRM 457. Analytic sensitivity was 0.34 μg/L, and the functional sensitivity was 1.6 μg/L. The reproducibility with Kryptor was 6.7% for a concentration of 5.3 μg/L and 5.6% for a concentration of 84.6 μg/L (n = 150). A recovery test was systematically performed to emphasize the interference related to the anti-Tg antibody. A recovery superior to 75% was considered satisfactory. In cases of low recovery, specific research for anti-Tg antibodies was performed.

The Tg levels were measured at the initial iodine treatment during LT₄ withdrawal (Tg_ini) and during T₄ administration (Tg_T4) within 3 months before PET/CT-FDG.

PET-FDG protocol

The rhTSH-stimulated PET/CT-FDG was performed after two 0.9 mg intramuscular doses of rhTSH (TSH alfa, Thyrogen^®; Genzyme, Cambridge, MA), which were administered 24 and 48 hours before imaging, while patients continued the same LT₄ dose. Blood sampling for measurement of the serum TSH and Tg concentrations was performed immediately before FDG injection (TSH₁ and Tg₁) and 48 hours after PET (TSH₂ and Tg₂) (Fig. 1). We used the highest value of Tg (Tg_max).

FIG. 1.

Study protocol for a week. All patients remained under levothyroxine treatment. PET/CT-FDG, positron emission tomography/computed tomography–fluoro-2-deoxy-d-glucose; rhTSH, recombinant human thyrotropin; Tg, thyroglobulin.

PET/CT studies were performed on a Biograph^® LSO Sensation 16 (Siemens Medical Solutions, Knoxville, TN). Five MBq/kg of FDG was injected after 30 minutes of rest. Sixty minutes later (±5 minutes), acquisitions began with CT, in the craniocaudal direction. Scan parameters were set to 120 kV and 100–150 mA s (based on patient's weight) using the dose reduction software (CareDose^®; Siemens Medical Solutions) which finally yielded a mean effective mAs of 89.1 ± 6.7. The patient's arms were positioned along the body, and acquisition was performed in free breathing, with a 0.75-mm collimation. PET image acquisitions followed, in the caudocranial direction, within a scan period based on 3 minutes per bed position. Acquisition time was actually adapted as a function of the injected dose (regarding 5 MBq/kg as the standard) and the time between injection and acquisition (standardized to 60 minutes), to maintain a normalized count rate for all patients. Six to eight bed positions per patients were acquired; the axial field of view for one bed position was 162 mm, with a bed overlap of 25% (plane spacing: 2 mm). The transverse spatial resolution reached 4.4 mm (centered point source in air).

When necessary, an additional step was performed on the neck immediately after the whole-body PET. The patient's arms were positioned along the body, and a CT acquisition was performed in free breathing, with a 0.75-mm collimation. Then, the PET images were performed, considering only one bed position. The acquisition time was set to 10 minutes whatever the postinjection delay.

Image reconstruction

All CT images were reconstructed in 512 × 512 matrices, and all PET images were reconstructed in 168 × 168 matrices. CT images were successively reconstructed for attenuation correction and for anatomical localization. For attenuation correction, images were reconstructed in 5-mm contiguous slices. For anatomical localization, images were reconstructed in 3-mm slices every 2 mm. PET images were iteratively reconstructed using the Fourier Rebinning and Attenuation Weighted Ordered Subset Expectation Maximization clinical software on a clinical Leonardo^® workstation (Siemens Medical Solutions), with four iterations and eight subsets. Images were corrected for random coincidences, scatter, and attenuation using the CT data. PET images were finally smoothed with a Gaussian filter (full width at half maximum = 5 mm).

Image interpretation

PET/CT-FDG images were visually evaluated by three board-certified nuclear physicians (P.V., A.E.S., and A.H.) on a clinical Leonardo workstation (Siemens Medical Solutions). When necessary, images were analyzed in consensus. Therefore, any increased FDG uptake was compared with the anatomic finding on CT. Areas of increased FDG uptake corresponding to normal structure such as muscles, salivary glands, vocals cords, tonsils, brown fat, and lymphoid tissues were recorded. The criterion for PET/CT-FDG interpretation was the presence of focal FDG uptake particularly located at sites typical for thyroid recurrences or metastases (neck, mediatinum, lungs, and bones). All areas with abnormal FDG uptake corresponding to a CT abnormality (tissue mass or lymph node) were interpreted as positive for recurrent disease. In addition, focally FDG increased which did not correspond to normal structures, or no other structural findings were recorded as positive. Suggestive findings on CT were interpreted as negative if they did not correspond to an area of abnormal increase of FDG uptake.

Diagnostic performances were analyzed as previously described (7). The results of PET/CT-FDG were correlated with patient follow-up information, which included the results of imaging modalities (WBS, neck ultrasound, enhanced CT, and magnetic resonance imaging) and postradioiodine treatment (Tg_T4 or Tg under rhTSH and histological findings). The PET/CT-FDG findings were classified as follows:

Lesions were true-positive if positive findings on PET/CT-FDG were confirmed by the presence of carcinoma on histological examination. Lesions were possibly true-positive (true-positive?) if PET/CT-FDG findings were confirmed by other imaging modalities or the presence of persistent abnormal or increasing Tg levels during the follow-up.

Lesions were false-positive if biopsy samples of suggestive lesions (by PET) were negative for carcinoma on histological examination. Lesions were possibly false-positive (false-positive?) if the lesions had resolved on subsequent follow-up imaging (no second primary tumor has been found in our series).

Lesions were true-negative if the findings of PET/CT-FDG were negative, elevated Tg had normalized without treatment, and metastatic disease was not seen on subsequent follow-up. Lesions were also probably true-negative (true-negative?) if there was no change in Tg level (only for levels between 2 and 5 μg/L after rhTSH) at 6-month interval. Follow-up was continued in all patients with true-negative lesions for at least 19 months.

Lesions were false-negative if the findings of PET/CT-FDG were negative and metastatic thyroid cancer was found at histological examination. Lesions were probably false-negative (false-negative?) if the findings of PET/CT-FDG were negative, with evident progression in the disease as was seen on other imaging modalities. PET/CT-FDG findings were also probably false-negative if patients had persistent elevated Tg levels (>5 μg/L) or rising Tg levels during the follow-up.

Statistics

All data were expressed as the mean ± standard deviation, range, and median. All patients in this study underwent a follow-up, with a median of 32 months (±13 months). Tg levels were compared with the PET/CT-FDG results (positive or negative) by nonparametric tests. Relation between PET/CT-FDG and histological results (size and node status) was evaluated with χ ² tests. A threshold at 0.05 was considered as significant.

Results

All the Tg dosage recovery tests were negative. At the initial treatment (LT₄ withdrawal), the mean serum Tg (Tg_ini) was 785 ± 2707 μg/L (range, 1–18,000; median, 58), for a TSH of 73 ± 64 mU/L (range, 8–426; median, 52). In the 3 months before the PET/CT-FDG, the patients were receiving LT₄ (195 ± 59 μg/day [range, 125–350; median, 175]) during which time they had a mean TSH of 0.24 ± 0.57 mU/L (range, 0.01–2.6; median, 0.025) and a mean Tg (Tg_T4) of 7 ± 15 μg/L (range, 0–89; median, 4) (Table 2).

Table 2.

Mean Value, Standard Deviation, Range, and Median for the Thyrotropin (mU/L) and Thyroglobulin (mg/L) Concentration During the Positron Emission Tomography/Computed Tomography–Fluoro-2-Deoxy-d-Glucose Examination

	Mean ± SD (n = 61)	Range	Median
Tg_T4	7 ± 15	0–89	4
TSH₁ (PET)	128 ± 66	28–320	116
TSH₂	25 ± 28	3–137	17
Tg₁ (PET)	25 ± 68	0–414	8
Tg₂	34 ± 85	0–476	10
Tg_max	35 ± 85

Tg_T4 were performed during LT₄ administration within the 3 months before PET/CT-FDG. TSH₁ and Tg₁ were performed the day of the PET/CT-FDG. TSH₂ and Tg₂ were performed 48 hours later. Tg_max is the maximum of the two Tg (Tg₂ and Tg₁).

SD, standard deviation.

Sixty-one PET/CT-FDG studies were performed in the 44 patients. The results of serum TSH and Tg concentrations measured immediately before FDG injection (TSH₁ and Tg₁) and 48 hours after PET (TSH₂ and Tg₂) are shown in Table 2.

Among the 44 patients, the last WBS was negative in 31/44 (70%) and noncontributive for the other 13 patients (30%). In these 13 patients, radioactive iodine uptake was discretely concentrated consistent with a small remnant in the thyroid bed in 7 (16% of the 44 patients), neck nodes in 3 (7% of the 44 patients), and neck and lung nodes in 3 (7% of the 44 patients). For these six patients with neck nodes or lung uptake, PET/CT-FDG had been performed for the diagnosis and presurgical evaluation of the disease.

Among the 44 patients, PET/CT-FDG findings were positive for 20 patients and negative for 24 patients.

For the 20 patients with positive PET, 10 of 20 were confirmed by histological examination (true-positive), 7 of 20 had a further increase in Tg level and progression of disease on follow-up imaging (true-positive?), and 1 of 20 underwent empiric radioiodine treatment with positive posttreatment scan result (true positive?). One of the 20 patients underwent surgery that revealed benign lymph nodes (false-positive), and another one underwent biopsy and had histological benign lymphoid proliferation (false-positive).

For the 24 patients with negative PET, 10 of 24 patients had Tg levels that remained negative or decreasing during follow-up, and all imaging remained negative during the follow-up (true-negative?). One of 24 had positive histopathology after surgery (false-negative) and 13 of 24 had rising Tg levels during follow-up (false-negative?).

Among the 61 PET/CT-FDG studies, 25 were positive and 36 were negative (Table 3). FDG uptake was located in the neck for 12 of 25 patients (48%), in the mediastinum for 6 of 25 (24%), in the neck and lung for 4 of 25 (16%), in the lung for 2 of 25 (8%), and in the bone for 1 of 25 (4%). Among the 25 positive PET/CT-FDG studies, the Tg_T4 values were less than 10 μg/L in 19 cases (76%) and greater than 10 μg/L in 6 cases (24%). Of these 19 patients, 9 (47%) were classified as true-positive. Among the 36 negative PET/CT-FDG cases, the serum Tg_T4 was lower than 10 μg/L in 31 and higher than 10 μg/L in 5 (χ ² = 0.90; p = 0.4983, NS for the comparison of PET/CT-FDG results and TgT4 levels; Table 3).

Table 3.

Comparison of Positron Emission Tomography/Computed Tomography–Fluoro-2-Deoxy-d-Glucose Results and Thyroglobulin Level Under Levothyroxine (n = 61; p = 0.4983, Not Significant)

	PET/CT-FDG+	PET/CT-FDG−
Tg_T4 ≤ 10 μg/L	19 (31%)	31 (51%)
Tg_T4 > 10 μg/L	6 (10%)	5 (8%)
	25	36

Sixty-one PET studies were performed in 44 patients (1–3 per patients [mean, 1.4]). Therefore, 17 additional PET/CT-FDG studies were performed in 11 patients (2 additional studies in 5 patients, and 1 additional study in 7 patients). The results of the original study and the additional one or two PET/CT-FDG studies were similar when performed in the same patient in 10 of 11 patients. In eight patients, the PET/CT-FDG studies were negative when repeated. For two patients, the initial and repeated PET/CT-FDG examinations were positive, with intense FDG uptake in the neck nodes. Histological examinations were obtained for these two patients. One was false-positive (lymphoid proliferation), and one was true-positive (neck recurrence). In the one patient in whom there was a discrepancy between the two PET/CT-FDG studies, the first PET/CT-FDG was positive (one false-positive) and the second was negative.

There was no relationship between the positivity or negativity of the PET/CT-FDG results and the serum Tg concentrations (Table 4), the size of the initial tumor size (p = 0.52, not significant), or the presence of lymph node metastasis (p = 0.14, not significant).

Table 4.

Comparison of Thyroglobulin Results (mg/L) with Respect to the Positron Emission Tomography Results (n = 61)

	PET positive (n = 25)	PET negative (n = 36)	p
Tg_ini	244 ± 566	1131 ± 3410	0.9999 (NS)
Tg_T4	8 ± 13	8 ± 17	0.9557 (NS)
Tg₁ (PET)	33 ± 85	20 ± 53	0.0740 (NS)
Tg₂	42 ± 99	28 ± 73	0.4204 (NS)
Tg_max	43 ± 99	29 ± 74	0.3351 (NS)

Values are given as mean ± standard deviation. Tg_ini, Tg under LT₄ withdrawal at the initial treatment; Tg_T4, Tg under LT₄ (within) 3 months before the PET-FDG. Tg₁ was performed the day of the PET/CT-FDG. Tg₂ was performed 48 hours after PET/CT-FDG. Tg_max is the maximum of the two Tg (Tg₂ and Tg₁).

NS, not significant.

Discussion

We performed 61 PET/CT-FDG studies after rhTSH stimulation in 44 patients with a history of thyroid surgery for thyroid cancer who while on LT₄ had a mean Tg of 7 ± 15 μg/L (median, 4 μg/L). This series shows that a cutoff value of 10 μg/L for Tg under LT₄ is probably not the best criteria to select patient candidates for PET/CT-FDG examination to detect the recurrence of differentiated thyroid carcinoma.

The interest in using PET-FDG to detect locoregional recurrence or distant metastasis is evident from the fact that there are more than 30 studies with about 1200 patients in the literature on this topic (Table 1). There are at least five reviews on this subject, focusing on progressive dedifferentiation of thyroid carcinoma cells leading to a loss of iodine-concentration capacity (14 –18). In such cases the sensitivity and specificity of PET-FDG ranges from 66% to 100% and from 25% to 100%, respectively. Some false-positive results have also been described, however, relating to physiological FDG uptake in muscles, brown fat, salivary glands, vocal cords, tonsils, or lymphoid tissues (19). Nevertheless, the utility of PET-FDG is powerful when compared with ^99mTc-methoxyisobutyl-isonitrile (^99mTc-MIBI) and thallium 64 (20,21) and iodine scintigraphy (22) for the detection of recurrent cervical metastasis (23,24).

Many authors (7,25,26) have shown that the performances of PET-FDG were lower in patients with low Tg_T4 level compared with patients with Tg_T4 greater than 10 μg/L. This may be partially explained by the fact that patient with low Tg_T4 level usually have a small mass of tumor tissue, and PET-FDG does not have a high degree of sensitivity for detecting small masses of tumor tissue. In the United States, most insurance providers cover PET-FDG only in patients with a Tg level under LT₄ greater than 10 μg/L. However, the interest is in using an imaging technique for detection of very early recurrences with small residual masses.

In the literature, evaluation of PET-FDG has been performed in patients with high Tg level (Table 1), with mean levels of Tg under LT₄ ranging from 7 to 2982 μg/L and Tg under rhTSH ranging from 21 to 13,777 μg/L. At these levels of Tg, PET-FDG appears frequently less interesting than other morphological imaging processes that have already localized recurrence or metastasis. The major interest in using PET-FDG would be for very early detection of these recurrences or metastases. Four studies have tested the use of PET-FDG after rhTSH. In these four studies, two were performed in patients with high Tg levels (3,12) and one was only performed in 7 of the 15 patients of the series (27). Only Chin et al. (13) has tested the use of PET-FDG after rhTSH in patients with low Tg level, but only in seven patients. In our study a large series of 61 PET-FDG with rhTSH were performed prospectively in 44 patients with a mean Tg_T4 of 7 ± 15 μg/L (median 4 μg/L), which is the lowest value in the literature.

Experimental studies have shown that TSH stimulated glucose transport and Glut1 expression in cultured thyroid cells (9,10). Based on these results, Moog et al. (8) have shown in 10 patients, an average of 63% of the FDG tumor uptake after TSH stimulation. Similar results were found by others (12,13,28).

Based on the literature, we assumed that PET-FDG would be more accurate after rhTSH administration in patients who were taking LT₄ and having serum Tg levels lower than 10 μg/L. This would allow detection of early recurrence or small metastasis of thyroid carcinoma. PET-FDG under rhTSH was performed as previously described (12). It would have been more interesting to make paired PET-FDG under LT₄ and with rhTSH in the same patients, but this was not done to avoid double examination and irradiation of the patients.

In our series, the mean Tg levels under LT₄ was 7 ± 15 μg/L. After rhTSH injection, the serum Tg levels reached to 25 ± 68 and 34 ± 85 μg/L, the day of the PET/CT-FDG and 2 days later, respectively, and TSH level reached 128 ± 66 and 25 ± 28 mU/L, respectively.

For the 20 patients with positive PET/CT-FDG, recurrence was confirmed in 10 after histological examination, and 8 were probably true-positive (increase of Tg level, progression of disease on follow-up imaging, and/or positive WBS). Two PET/CT-FDG were false-positive. For the 24 patients with negative PET/CT-FDG, 10 remained negative during the follow-up and are probably true-negative, but 14 were false-negative (1/14) or probably false-negative (13 showed increased Tg level). Therefore, the sensitivity of PET/CT-FDG is probably low in a population with relatively low Tg level under LT₄.

However, positive PET/CT-FDG was not found particularly in patients with Tg_T4 > 10 μg/L. We found no significant difference between the PET/CT-FDG result and the Tg levels, whatever the level of Tg tested: at initial treatment, under LT₄, or after rhTSH (Table 4). There was a tendency for Tg₁ to be higher in patients with positive PET-FDG (33 ± 85 vs. 20 ± 53 μg/L), but this result was not significant (p = 0.07). The Tg at initial treatment appears to be higher (1131 μg/L) in patients with negative than in patients with positive (244 μg/L) PET/CT-FDG, but the difference was not significant. It is important to note that the Tg at initial ablation (Tg_ini) was quite high in our population, which is probably an indication that patients in our series had significant disease at diagnosis. So, our results suggest that, in patients with history of thyroid carcinoma and relatively low Tg level while taking LT₄, there is no relation between the PET results and serum Tg levels.

Even if the diagnostic performances were low in our population, it is important to note that the serum Tg under LT₄ therapy was less than 10 μg/L in 19/25 cases with positive PET/CT-FDG results (Table 3), including 9 true-positive patients. Consequently, PET/CT-FDG after rhTSH administration allowed an early detection in these nine patients (i.e., 20% of our 44 patients). Notably, we did not find any link between PET findings and age, sex, histology, tumor size, or nodal involvement.

Hooft et al. (29) have shown that in 19 patients with recurrent differentiated thyroid carcinoma, who benefited from PET-FDG, the positive PET-FDG, present in 13 patients, was associated with the expression of hexokinase type I in the primary tumor. Therefore, the authors suggested that lack of hexokinase I in the primary tumor could be responsible for lower PET positivity. This phenomenon may partially explain our results, but only a study of the presence and functionalities of hexokinase type I in the primary tumor of our patients could validate this hypothesis.

In conclusion, for the diagnosis of recurrent disease in patients with thyroid carcinoma, the sensitivity of rhTSH-stimulated PET/CT-FDG seems to be low. No correlation was observed between PET/CT-FDG findings and Tg levels. However, positive PET/CT-FDG results have been found in a significant number of patients with Tg level lower than 10 μg/L. Therefore, rhTSH-stimulated PET/CT-FDG seems to be worthwhile in some patients whose serum Tg while taking LT₄ was less than 10 μg/L. Further investigations are necessary, however, to better identify those patients in whom PET/CT-FDG would achieve early detection of recurrent tumor.

Footnotes

Acknowledgments

The authors thank Béatrice Guilbert for her help in collecting and verifying biological data. The authors also thank the technologists of the Department of Nuclear Medicine of Rouen for their help in managing the patients for this study.

Disclosure Statement

The authors declare that no competing financial interests exist.

References

Schlumberger

, Berg

, Cohen

, Duntas

, Jamar

, Jarzab

, Limbert

, Lind

, Pacini

, Reiners

, Sanchez

, Toft

, Wiersinga

. 2004. Follow-up of low-risk patients with differentiated thyroid carcinoma: a European perspective. Eur J Endocrinol, 150:105–112.

Pacini

, Schlumberger

, Harmer

, Berg

, Cohen

, Duntas

, Jamar

, Jarzab

, Limbert

, Lind

, Reiners

, Sanchez

, Smit

, Wiersinga

. 2005. Post-surgical use of radioiodine (131I) in patients with papillary and follicular thyroid cancer and the issue of remnant ablation: a consensus report. Eur J Endocrinol, 153:651–659.

Cooper

, Doherty

, Haugen

, Kloos

, Lee

, Mandel

, Mazzaferri

, McIver

, Sherman

, Tuttle

. 2006. Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid, 16:109–142.

Pacini

, Agate

, Elisei

, Capezzone

, Ceccarelli

, Lippi

, Molinaro

, Pinchera

. 2001. Outcome of differentiated thyroid cancer with detectable serum Tg and negative diagnostic (131)I whole body scan: comparison of patients treated with high (131)I activities versus untreated patients. J Clin Endocrinol Metab, 86:4092–4097.

Schlumberger

, Arcangioli

, Piekarski

, Tubiana

, Parmentier

. 1988. Detection and treatment of lung metastases of differentiated thyroid carcinoma in patients with normal chest X-rays. J Nucl Med, 29:1790–1794.

Schluter

, Bohuslavizki

, Beyer

, Plotkin

, Buchert

, Clausen

. 2001. Impact of FDG PET on patients with differentiated thyroid cancer who present with elevated thyroglobulin and negative 131I scan. J Nucl Med, 42:71–76.

Shammas

, Degirmenci

, Mountz

, McCook

, Branstetter

, Bencherif

, Joyce

, Carty

, Kuffner

, Avril

. 2007. 18F-FDG PET/CT in patients with suspected recurrent or metastatic well-differentiated thyroid cancer. J Nucl Med, 48:221–226.

Moog

, Linke

, Manthey

, Tiling

, Knesewitsch

, Tatsch

, Hahn

. 2000. Influence of thyroid-stimulating hormone levels on uptake of FDG in recurrent and metastatic differentiated thyroid carcinoma. J Nucl Med, 41:1989–1995.

Filetti

, Damante

, Foti

. 1987. Thyrotropin stimulates glucose transport in cultured rat thyroid cells. Endocrinology, 120:2576–2581.

10.

Hosaka

, Tawata

, Kurihara

, Ohtaka

, Endo

, Onaya

. 1992. The regulation of two distinct glucose transporter (GLUT1 and GLUT4) gene expressions in cultured rat thyroid cells by thyrotropin. Endocrinology, 131:159–165.

11.

Russo

, Damante

, Foti

, Costante

, Filetti

. 1994. Different molecular mechanisms are involved in the multihormonal control of glucose transport in FRTL5 rat thyroid cells. J Endocrinol Invest, 17:323–327.

12.

Petrich

, Borner

, Otto

, Hofmann

, Knapp

. 2002. Influence of rhTSH on [(18)F]fluorodeoxyglucose uptake by differentiated thyroid carcinoma. Eur J Nucl Med Mol Imaging, 29:641–647.

13.

Chin

, Patel

, Cohade

, Ewertz

, Wahl

, Ladenson

. 2004. Recombinant human thyrotropin stimulation of fluoro-D-glucose positron emission tomography uptake in well-differentiated thyroid carcinoma. J Clin Endocrinol Metab, 89:91–95.

14.

Nahhas A

, Khan

, Gogbashian

, Banti

, Rampin

, Rubello

. 2008. 18 F-FDG PET in the diagnosis and follow-up of thyroid malignancy. In Vivo, 22:109–114(review).

15.

Lind

, Kohlfurst

. 2006. Respective roles of thyroglobulin, radioiodine imaging, and positron emission tomography in the assessment of thyroid cancer. Semin Nucl Med, 36:194–205.

16.

Nanni

, Rubello

, Fanti

, Farsad

, Ambrosini

, Rampin

, Banti

, Carpi

, Muzzio

, Franchi

. 2006. Role of 18F-FDG-PET and PET/CT imaging in thyroid cancer. Biomed Pharmacother, 60:409–413.

17.

Reske

, Kotzerke

. 2001. 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, 28:1707–1723.

18.

Schoder

, Yeung

. 2004. Positron emission imaging of head and neck cancer, including thyroid carcinoma. Semin Nucl Med, 34:180–197.

19.

Jabour

, Choi

, Hoh

, Rege

, Soong

, Lufkin

, Hanafee

, Maddahi

, Chaiken

, Bailet

. 1993. Extracranial head and neck: PET imaging with 2-[F-18]fluoro-2-deoxy-D-glucose and MR imaging correlation. Radiology, 186:27–35.

20.

Grunwald

, Briele

, Biersack

. 1999. Non-131I-scintigraphy in the treatment and follow-up of thyroid cancer. Single-photon-emitters or FDG-PET? Q J Nucl Med, 43:195–206.

21.

Iwata

, Kasagi

, Misaki

, Matsumoto

, Iida

, Ishimori

, Nakamoto

, Higashi

, Saga

, Konishi

. 2004. Comparison of whole-body 18F-FDG PET, 99mTc-MIBI SPET, and post-therapeutic 131I-Na scintigraphy in the detection of metastatic thyroid cancer. Eur J Nucl Med Mol Imaging, 31:491–498.

22.

Chung

, So

, Lee

, Choi

, Lim

, Lee

, Hong

, Youn

, Lee

, Cho

. 1999. Value of FDG PET in papillary thyroid carcinoma with negative 131I whole-body scan. J Nucl Med, 40:986–992.

23.

Helal

, Merlet

, Toubert

, Franc

, Schvartz

, Gauthier-Koelesnikov

, Prigent

, Syrota

. 2001. Clinical impact of (18)F-FDG PET in thyroid carcinoma patients with elevated thyroglobulin levels and negative (131)I scanning results after therapy. J Nucl Med, 42:1464–1469.

24.

Wang

, Larson

, Fazzari

, Tickoo

, Kolbert

, Sgouros

, Yeung

, Macapinlac

, Rosai

, Robbins

. 2000. Prognostic value of [18F]fluorodeoxyglucose positron emission tomographic scanning in patients with thyroid cancer. J Clin Endocrinol Metab, 85:1107–1113.

25.

Mirallié

, Guillan

, Bridji

, Resche

, Rousseau

, Ansquer

, Bodet-Milin

, Curtet

, Carnaille

, Murat

, Charbonnel

, Kraeber-Bodere

. 2007. Therapeutic impact of 18FDG-PET/CT in the management of iodine-negative recurrence of differentiated thyroid carcinoma. Surgery, 142:952–958.

26.

Schluter

, Grimm-Riepe

, Beyer

, Lubeck

, Schirren-Bumann

, Clausen

. 1998. Histological verification of positive fluorine-18 fluorodeoxyglucose findings in patients with differentiated thyroid cancer. Langenbecks Arch Surg, 383:187–189.

27.

Saab

, Driedger

, Pavlosky

, McDonald

, Wong

, Yoo

, Urbain

. 2006. Thyroid-stimulating hormone-stimulated fused positron emission tomography/computed tomography in the evaluation of recurrence in 131I-negative papillary thyroid carcinoma. Thyroid, 16:267–272.

28.

van Tol

, Jager

, Piers

, Pruim

, de Vries

, Dullaart

, Links

. 2002. Better yield of (18)fluorodeoxyglucose–positron emission tomography in patients with metastatic differentiated thyroid carcinoma during thyrotropin stimulation. Thyroid, 12:381–387.

29.

Hooft

, van der Veldt

, van Diest

, Hoekstra

, Berkhof

, Teule

, Molthoff

. 2005. [18F]fluorodeoxyglucose uptake in recurrent thyroid cancer is related to hexokinase I expression in the primary tumor. J Clin Endocrinol Metab, 90:328–334.

30.

Feine

. 1998. Fluor-18-deoxyglucose positron emission tomography in differentiated thyroid cancer. Eur J Endocrinol, 138:492–496.

31.

Dietlein

, Scheidhauer

, Voth

, Theissen

, Schicha

. 1997. Fluorine-18 fluorodeoxyglucose positron emission tomography and iodine-131 whole-body scintigraphy in the follow-up of differentiated thyroid cancer. Eur J Nucl Med, 24:1342–1348.

32.

Wang

, Macapinlac

, Larson

, Yeh

, Akhurst

, Finn

, Rosai

, Robbins

. 1999. [18F]-2-fluoro-2-deoxy-d-glucose positron emission tomography localizes residual thyroid cancer in patients with negative diagnostic (131)I whole body scans and elevated serum thyroglobulin levels. J Clin Endocrinol Metab, 84:2291–2302.

33.

Grunwald

, Kalicke

, Feine

, Lietzenmayer

, Scheidhauer

, Dietlein

, Schober

, Lerch

, Brandt-Mainz

, Burchert

, Hiltermann

, Cremerius

, Biersack

. 1999. Fluorine-18 fluorodeoxyglucose positron emission tomography in thyroid cancer: results of a multicentre study. Eur J Nucl Med, 26:1547–1552.

34.

Muros

, Llamas-Elvira

, Ramirez-Navarro

, Gomez

, Rodriguez-Fernandez

, Muros

Lopez dlT, Becerra

, Carreras

. 2000. Utility of fluorine-18-fluorodeoxyglucose positron emission tomography in differentiated thyroid carcinoma with negative radioiodine scans and elevated serum thyroglobulin levels. Am J Surg, 179:457–461.

35.

Frilling

, Tecklenborg

, Gorges

, Weber

, Clausen

, Broelsch

. 2001. Preoperative diagnostic value of [(18)F] fluorodeoxyglucose positron emission tomography in patients with radioiodine-negative recurrent well-differentiated thyroid carcinoma. Ann Surg, 234:804–811.

36.

Lind

, Kresnik

, Kumnig

, Gallowitsch

, Igerc

, Matschnig

, Gomez

. 2003. 18F-FDG-PET in the follow-up of thyroid cancer. Acta Med Austriaca, 30:17–21.

37.

Nahas

, Goldenberg

, Fakhry

, Ewertz

, Zeiger

, Ladenson

, Wahl

, Tufano

. 2005. The role of positron emission tomography/computed tomography in the management of recurrent papillary thyroid carcinoma. Laryngoscope, 115:237–243.

38.

Ong

, Ng

, Sundram

. 2005. Initial experience in use of fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography in thyroid carcinoma patients with elevated serum thyroglobulin but negative iodine-131 whole body scans. Singapore Med J, 46:297–301.

39.

Palmedo

, Bucerius

, Joe

, Strunk

, Hortling

, Meyka

, Roedel

, Wolff

, Wardelmann

, Biersack

, Jaeger

. 2006. Integrated PET/CT in differentiated thyroid cancer: diagnostic accuracy and impact on patient management. J Nucl Med, 47:616–624.

40.

Iagaru

, Kalinyak

, McDougall

. 2007. F-18 FDG PET/CT in the management of thyroid cancer. Clin Nucl Med, 32:690–695.

41.

Freudenberg

, Frilling

, Kuhl

, Muller

, Jentzen

, Bockisch

, Antoch

. 2007. Dual-modality FDG-PET/CT in follow-up of patients with recurrent iodine-negative differentiated thyroid cancer. Eur Radiol, 17:3139–3147.

42.

Zuijdwijk

, Vogel

, Corstens

, Oyen

. 2008. Utility of fluorodeoxyglucose-PET in patients with differentiated thyroid carcinoma. Nucl Med Commun, 29:636–641.