Jolanta Kunikowska M.D., Nuclear Medicine Department, Medical University of Warsaw, Banacha St. 1a, 02-097 Warsaw, Poland, tel.: +48 22 599 22 70, fax: +48 22 599 11 70, e-mail: jolanta.kunikowskaigwum.edu.pl

Introduction

Somatostatin receptors (SSTR1 to SSTR5) play an important role not only in the maintenance of normal thyroid physiology, but also in pathologic conditions. Somatostatin (SST) and SSTR1-5 are expressed in medullary thyroid carcinoma (MTC) and differentiated thyroid cancers (DTC), but the role of SSTR subtypes is still not clear [1-3]. Some authors reported that SSTR2 has been found to be the predominant subtype in DTC with a high expression particularly in papillary cancer [4]. SSTR2 also seems to play an important role in the pathophysiology of benign thyroid diseases [5]. Recently published studies suggest that SSTR2 was the most common receptor expressed in benign diseases like multinodular goitre, Graves’ disease, Hashimoto disease, or follicular adenoma [6].

On the other hand, 68Ga labelled somatostatin (SST) analogues (like: DOTATATE and DOTATOC) have high affinity to SSTR2. Significant correlation was found between SSTR2 expression and the maximum Standardised Uptake Value (SUVrnax) as a semiquantitative parameter on tire PET/CT [7]. The visualisation of SSTR2 status in vivo with radiolabelled SST analogues help to define the role of SSTR2 in thyroid diseases.

The incidental 68Ga-DOTATATE or 68Ga-DOTATOC uptake seen in the thyroid gland is not an occasional finding on diagnostic PET/CT scans during diagnosis of neuroendocrine neoplasia (NEN). In the published data the problem of unexpected increased uptake of 68Ga-labelled SST analogues in thyroid gland and clinical significance of such findings has not been presented yet and needs further evaluation.

The aim of the study was to evaluate the prevalence of incidental thyroid uptake of 68Ga-DOTATATE PET/CT in patients referred to the Nuclear Medicine Department for evaluation of NEN.

Material and methods

The study was approved by the Ethical Committee of the Medical University of Warsaw. Written informed consent was obtained from all patients.

We retrospectively evaluated 1150 patients with NEN, who underwent 68Ga-DOTATATE PET/CT imaging between January 2010 and March 2013.

Clinical history, serum TSH levels, thyroid antibodies (TAb), ultrasonography, and cytological assessment of the material from fine-needle aspiration biopsy (FNA) of the thyroid lesion were investigated. The cytology examination was described by Bethesda classification [8].

Laboratory examination: serum thyroid-stimulating hormone (TSH), free thyroxine (fT4), free triiodothyronine (fT3), serum thyroperoxidase antibody (ATPO), and thyroglobulin antibody (ATG) were measured using the chemiluminescence method (Immulite 1000 Siemens Healthcare Diagnostics, Architect ci 8200 Abott).

The normal range for TSH level was defined as 0.4-4.0 μIU/mL, for fT4 0.89-1.76 ng/dL, for fT3 1.5-4.1 pg/mL, ATPO < 35.0 IU/mL, and ATG < 115.0 IU/mL.

68Ga -DOTATATE PET/CT scans were performed not earlier than five weeks after the last injection of long-acting SST analogues – octreotide and seven weeks after lanreotide.

Sites of thyroid pathological uptake were noted and confirmed on ultrasound (US) (Voluson E8, linear probe 6 cnyl,0/25Hz) and FNA.

PET/CT imaging

Preparation of radiopharmaceutical 68Ga -DOTATATE and the 68Ga -DOTATATE PET/CT was performed using a previously-described procedure [9].

Image analysis

Image analysis was performed using a Multi-Modality Work Station (MMWS), Syngo (TrueD) (Siemens Medical Solutions). Attenuation-corrected PET images and PET/CT images were analysed. The PET images without attenuation correction were also reviewed. The PET/CT fusion images were used for evaluation of uptake using volume regions of interest (ROIs). The SUV maximum and mean were calculated in all ROIs using the following formula:

SUVrnax = Maximum Activity Concentration at time from pixel values/(Injected Dose/Patient Weight)

The thyroid uptake of 68Ga-DOTATATE was analysed: firstly qualitative-visual analysis of distribution pattern and semiquantitative with measurement of SUVrnax.

Statistical methods

Patient characteristics, thyroid SUVrnax, TAb level, and serum TSH were characterised by the mean (± SD) and median.

To assess the normal distribution data we used the Shapiro-Wilk test and graphical analysis by inspection of the histograms. Spearman’s correlation coefficient was used to determine the association of thyroid SUVmax with ATPO and TSH. The tests were two-sided and a p-value of < 0.05 was considered statistically significant.

Calculations were performed using Statistica software (6.0 Stat Soft, Inc. 1984-2002) and Excel (Microsoft Office Professional Edition 2013).

Results

Labelling and imaging procedure

68Ga-DOTATATE was successfully labelled with a radiochemical purity of over 99%. For PET/CT examination a mean injected activity of 68Ga-DOTATATE of 156.4 ± 18.1 MBq, (range: 120-200 MBq) was administered without any major side effects.

Thyroid uptake

Among 1150 patients with histological confirmation of NEN (G1 – 42% and G2 – 58%), referred for staging or restaging after surgery, we revealed abnormalities in the thyroid in 46 (4.1%) patients (12 males and 34 females, mean age 63.8 years ± 10.8; range 25-85 years) with the prevalence significantly higher in women 3.0% than in men 1.1%.

The incidental increased uptake of 68Ga-DOTATATE in the thyroid gland was defined as focal or diffuse, inhomogeneous. A diffuse uptake was noted in 74% of patients and focal in 26% of patients.

Subsequently we analysed the clinical significance of abnormal radiotracer distribution pattern in the thyroid gland and the impact of such a finding on the clinical management of those patients. We compared it with ultrasound, TSH level, TAb, and FNA.

Thyroid diffuse uptake

Increased diffuse, inhomogeneous 68Ga-DOTATATE thyroid uptake was seen in 34/46 patients (8 males, 26 females, mean age 49.5 years ± 16.7; range 25-85). Laboratory tests and ultrasonography revealed: Hashimoto disease in 13 patients (38.2%), multinodular goitre in 9 patients (26.6%), and multinodular goitre coexisting with Hashimoto disease in 2 patients (5.8%) (Figs. 1 and 2). In the remaining 10 patients (29.4%) no abnormalities of laboratory tests and ultrasound were seen.

Figure 1. PET/CT with ⁶⁸ Ga-DOTATATE in patients with pancreatic NEN G2 with diffuse, homogenous uptake in thyroid gland (SUVmax 9.8). US as well, laboratory tests revealed Hashimoto disease (Fusion PET/CT, PET, CT)
Rycina 1. Badanie ⁶⁸ Ga-DOTATATE PET/CT u pacjenta z NEN G2 trzustki, z widocznym rozlanym, jednorodnym gromadzeniem znacznika w gruczole tarczowym (SUVmax 9.8). W badaniu USG oraz w badaniach laboratoryjnych potwierdzono chorobę Hashimoto (Fuzja PET/CT, PET, CT)

Figure 2. PET/CT with ⁶⁸ Ga-DOTATATE in patients with ileal NEN G1 with diffuse, nonhomogeneous uptake (SUVmax 3.3). US as well, laboratory test revealed Hashimoto disease (Fusion PET/CT, PET, CT)
Rycina 2. Badanie ⁶⁸ Ga-DOTATATE PET/CT u pacjenta z NEN G1 jelita cienkiego, z widocznym rozlanym, niejednorodnym gromadzeniem znacznika w gruczole tarczowym (SUVmax 3.3). W badaniu USG oraz w badaniach laboratoryjnych potwierdzono chorobę Hashimoto (Fuzja PET/CT, PET, CT)

The mean SUVmax was 4.6 ± 1.6, and median SUVrnax was 5.3 (range 2.0-10.9).

The mean TSH level was 1.8 ± 1.2 mlU/L and median 1.7 mlU/L (range: 0.005-5.3 mlU/L). In patients with Hashimoto disease the mean ATPO level was 327.0 ± 733.3 IU/mL, median 67.5 IU/ ml (range: 17.3– 2821 IU/mL), the mean ATG level was 99.6 ± 84.0 IU/mL, and median 65.0 IU/mL (range: 19.2-282.0 IU/mL).

Before 68Ga-DOTATATE PET/CT imaging 15/34 (44%) patients had no prior history of thyroid disease. Among them, laboratory test and ultrasound revealed Hashimoto disease in 5 patients, and the remaining 10 patients did not reveal pathology within 6-12-month follow-up.

19/34 (66%) patients had previous history of thyroid disease; 10 patients had long-term supplementation with L-thyroxine, 2 patients Thiamazole therapy.

Spearman correlation coefficient showed no correlation between thyroid SUVmax value and the ATPO level (rho = -0.107, p = 0.547) and SUVmax and TSH value (rho = 0.017, p = 0.924).

Focal thyroid uptake

12/46 patients (4 males, 8 females, mean age 55.5 ± 17.2 years, range 35-78) had focal uptake in the thyroid. All patients had prior history of thyroid diseases: one patient had Hashimoto disease, six patients had multinodular goitre coexisting with Hahimoto disease, and five patients had multinodular goitre. Among them three patients were receiving substitutive therapy with L-thyroxine and one patient with multinodular goitre Thiamazole therapy. The ultrasound revealed hypoechogenic lesions in 10 patients and irregular echogenicity in 2 patients.

The mean thyroid SUVmax was 7.3 ± 3.1, and median SUVmax 6.6 (range 2.8-12.6). The mean TSH level was 3.0 ± 4.9 mlU/L, and the median was 1.47 mlU/L (range 0.06-19.2 mlU/L).

Based on histopathology 2/12 patients after thyroidectomy revealed the following: one patient had papillary thyroid cancer (SUVmax 11.6) (Fig. 3), and one patient had benign multinodular goitre. Among the remaining 9/12 patients managed with a 12-month follow-up all had category II lesion (Fig. 4). One patient’s (with category III lesion) follow-up data was unavailable.

Figure 3. PET/CT with ⁶⁸ Ga-DOTATATE in patients with gastric NEN G1 with focal uptake in the right lobe (SUVmax 11.6). ENA biopsy as well as post-surgical examination revealed papillary carcinoma (Fusion PET/CT, PET, CT)
Rycina 3. Badanie ⁶⁸ Ga-DOTATATE PET/CT u pacjenta z NEN G1 żołądka, z widocznym ogiuskiem gromadzenia znacznika w prawym płacie tarczycy (SUVmax 11.6). W biopsji cienkoigłowej oraz w pooperacyjnym badaniu histopatologicznym – rak brodawkowaty tarczycy. (Fuzja PET/CT, PET, CT)

Figure 4. PET/CT with ⁶⁸ Ga-DOTATATE in patients with unknown primary NEN with focal uptake in the right lobe (SUVmax 4.2). US and ENA biopsy examination revealed category II lesion in multinodular goitre (Fusion PET/CT, PET, CT)
Rycina 4. Badanie ⁶⁸ Ga-DOTATATE PET/CT u pacjenta z NEN o nieznanym punkcie wyjścia z widocznym ogniskiem gromadzenia znacznika w prawym płacie tarczycy (SUVmax 4.2). W badaniu USC oraz biopsji cienkoigłowej – zmiana kategorii II w wolu wieloguzkawym (Fuzja PET/CT, PET, CT)

Discussion

The spread of 68Ga-DOTA-somatostatin analogues PET/CT availability as a new standard imaging for the diagnosis of patients with NEN forces us to clarify the clinical role of incidental radiotracer uptake seen in the thyroid gland. In addition, no guidelines have been drawn up for the clinical assessment and management of such abnormalities.

In visual analysis normal 68Ga-DOTATATE uptake tends to be bilateral, diffuse, homogenous, and only slightly visible on maximum intensity projection (MIP), with mean SUVmax 2.9 ( ± 1.2) [9]. Knowledge of normal 68Ga-DOTATATE distribution in the thyroid gland is crucial for accurate image interpretation, which should not be confused with any pathology. The assessment of the functional SSTR2 receptor status in vivo in normal thyroid by 68Ga-DOTATATE PET/CT seems to be partially consistent with the results obtained in vitro, but published studies are still very limited and often controversial [3, 5, 6, 10]. In the normal thyroid, increased expression of SSTR3 and SSTR5 have been documented by molecular methods, whilst SSTR2 expression was at very low levels [1]. In immunochemistry (IHC) some authors found that in the normal thyroid tissue SSTR2 was expressed in 13%, SSTR3 in 24%, and SSTR5 only in 2%, without expression of SSTRlor SSTR4 [10]. Conversely, in a recently published study SSTR expression was highly expressed in normal thyroid tissue for SSTR1, SSTR3, SSTR4, and SSTR5, whilst SSTR 2a was not expressed at all [6]. Results depend on the investigation method used – expression of receptors at the level of rnRNA SSTRs or IHC [1, 6, 10-12].

A study of the Polish population revealed thyroid dysfunctions in over 10% of older patients, but 80% of them were subclinical [13]. In our study an unexpected increase of 68Ga-DOTATATE uptake in the thyroid gland was seen in 4.1% of patients.

Most of them presented a diffuse pattern as predominantly seen in Hashimoto disease and less frequently in multinodular goitre and normal thyroid gland. Data in the literature regarding SSTR2 expression in vivo in benign thyroid pathologies are still limited with a very small sample size [14-20].

In patients with NEN, who underwent 111In-octreotide scintigraphy for tumour localisation, the increased uptake of radiotracer was seen predominantly in endemic goitre with or without thyroid autonomy [14]. The diffuse, bilateral uptake of 111In-octreotide or 111In-pentetreotide as a dominant pattern in patients with autoimmune thyroiditis (AITD): Graves’ disease and Hashimoto disease [15-18], was reported by other authors. Thyroid 111In-pentetreotide accumulation in Graves’ disease patients was increased seven fold compared to the control group and was almost absent after radioiodine-induced hypothyroidism [15]. In untreated hyperthyroid patients a positive correlation between 111In-pentetreotide uptake in the thyroid and TSH, fT4, fT3, as well receptor autoantibodies levels was found. These correlations were absent after treatment with Thiamazole [17].

Currently, because of its higher sensitivity, PET/CT with 68Ga-labelled somatostatin analogues is the preferred diagnostic method for detection of SSTR2 in vivo [24]. The majority of published studies focus on the clinical value 68Ga-labelled somatostatin analogues PET/CT in patients with non-radioiodine-avid DTC and high thyroglobulin levels [21-23].

The both studies suggest that increased 68Ga-labelled SST analogues uptake in thyroid in patients with AITD might be related to the activity of autoimmune diseases [18, 19]. Both studies suggest that increased 68Ga-labelled SST analogues uptake in thyroid in patients with AITD might be related to the activity of autoimmune disease. Intense tracer uptake was found in some cases (4.8%) of structurally and functionally normal thyroid lobes, but follow-up after 6-14 months did not show any thyroid pathology [20].

Normal thyroids as well as ones with non-toxic nodules and goitres showed a clearly detectable radiotracer uptake with large variability [19]. The radiotracer uptake seen in thyroid can reflect a dense infiltration of lymphocytes that express SSTR2 on their plasma membrane [16, 17, 19, 20].

Contrary to published studies, we found the diffuse pattern with nearly the same frequency for multinodular goitre as for normal thyroid gland. In our series about 30% of patients with increased, inhomogeneous 68Ga-DOTATATE uptake had normal thyroid gland, and 12-month follow-up did not reveal any pathology. Longer follow-up studies are needed to determine whether increased 68Ga-DOTATATE uptake in these patients can predict future autoimmune thyroiditis or demonstrate normal distribution pattern.

A fundamental problem is differential diagnosis between malignant and benign lesions in focal 68Ga-DOTATATE uptake in thyroid gland, which was noted in 26% of patients.

The prevalence of malignancy among this group was 8% (papillary thyroid cancer). Subsequent biopsy recommended after 68Ga-DOTATATE PET/CT critically changed in the management of this patients. Interestingly, in published studies the focal pattern was associated mainly with hot thyroid nodule (in 99mTc) and less frequently with benign cold nodules [19, 20]. Due to the high expression of SSTR2 reported in papillary thyroid cancer and benign thyroid nodules in vitro and in vivo studies is probable that the high value of SUVmax could be seen in both conditions [5, 6, 19, 20]. Therefore, in the case of focal uptake of 68Ga-DOTATATE, the patient should undergo further examination, including ultrasonography and FNA, to exclude a thyroid carcinoma.

Our study, for the first time, underlined the clinical significance of an unexpected increase in 68Ga-DOTATATE uptake in the thyroid gland seen on PET/CT. One of the main limitations of this study is its retrospective design, which may have introduced potential selection bias. For some of the variables the follow-up data were scarce. Another important limitation is that the study population included a small percentage with incidental, abnormal thyroid uptake seen in all performed examinations. In particular, due to the small number of patients, the methods of statistical analysis used were limited. For all of the above-mentioned reasons the results should be interpreted with appropriate caution.

Conclusions

Patients with focal 68Ga -DOTATATE uptake should undergo further examination due to potential risk of malignancy. Diffuse 68Ga-DOTATATE uptake was predominantly associated with active autoimmune thyroiditis or benign goitre.

Both focal and diffuse thyroid pathologies were often seen in women.

Acknowledgments

Special thanks to Michal Najdzik for graphic processing.