We reviewed a database of patients and healthy subjects who underwent ¹⁸F-FDG PET/CT between March 2010 and February 2012 at our institute; breast cancer patients were enrolled through the breast care center and healthy subjects through our health promotion center. During the study period, 531 female patients with primary breast cancer and 801 healthy subjects were evaluated with ¹⁸F-FDG PET/CT. The following exclusion criteria for breast cancer patients or healthy subjects were utilized in this study: (1) male sex, (2) distant metastasis at the time of diagnosis, (3) treatment with neoadjuvant chemotherapy, (4) history of thyroid disease, and (5) previous malignancy. We excluded 42 breast cancer patients and 515 healthy male subjects who met the above exclusion criteria. In total, we analyzed 489 patients with breast cancer and 286 healthy women for this study. The Institutional Review Board of our institute approved this study (129792-2014-021).

All patients fasted and rested for at least 6 hours before undergoing ¹⁸F-FDG PET/CT (Biograph 64; Siemens, Erlangen, Germany) scanning. Serum glucose levels were less than 200 mg/dL before ¹⁸F-FDG administration. A total of 4.5 MBq/kg of ¹⁸F-FDG was injected into a vein contralateral to the tumor side. At 60 minutes after ¹⁸F-FDG injection, a 3-minute static emission study was performed at the transmission position (early phase), another study for the delayed phase was performed 120 minutes after the injection, and whole body imaging was performed from the base of the skull to the mid thighs. Lowdose whole body CT was performed before the emission scan for CT attenuation correction instead of a transmission scan. These data were also used for anatomic localization of the PET emission image. The attenuation-corrected PET images were reconstructed with an ordered subset expectation maximization iterative algorithm. The resulting in-plane image resolution of the axial view had a full width of 5.8 mm at half maximum. Imaging data were obtained with a Biograph Sensation 16 (Biograph 64) scanner, which produces transaxial, coronal, and sagittal reconstructions of CT, PET, and fusion of PET/CT data for interpretation. The Biograph scanner combines a 16-detector-row spiral CT scanner (Somatom Emotion; Siemens) and a high resolution PET scanner with 4.5-mm spatial resolution and three-dimensional image acquisition. A multimodality computer platform was used for image review and manipulation. After scatter and decay correction and reorientation in axial, sagittal, and coronal planes, the ¹⁸F-FDG PET/CT images were reviewed by an experienced nuclear physician. All patients were studied in the supine position, oriented slightly obliquely and contralateral to the tumor side. To calculate the maximum standardized uptake value (SUVmax), circular regions of interest (ROI) were manually drawn on the attenuationcorrected emission images throughout the axial planes in which a suspicious lesion could be delineated.

On ¹⁸F-FDG PET/CT images, thyroid uptake was considered to be present when there was increased uptake in the thyroid gland compared to the physiological background. For the visual analysis, abnormal FDG uptake was defined as substantially greater activity than in the aortic blood on attenuation-corrected images. Metabolic activity within the thyroid glands was determined. Focal thyroid uptake was defined as uniform distribution of the tracer above the background throughout one lobe. All PET/CT images were read directly from the screen of the computer workstation, and CT findings were recorded for abnormalities, including the heterogeneity of the thyroid parenchyma (homogenous or heterogeneous) and the pattern of calcification (nodular, globular, amorphous, or linear) based on the CT portion of the PET/CT scan. When a space-occupying lesion was present, its exact size and the solid or cystic pattern were confirmed. Semiquantitative evaluation was performed using a Syngo MultiModality Workplace (Siemens AG, Berlin, Germany). A ROI was drawn on the fused PET/CT image to measure the SUV of the tumor. The SUV is defined as follows: (peak kBq/mL in ROI)/(injected activity/g of body weight). To calculate the SUVmax, manually defined circular ROIs were drawn on the attenuation-corrected emission images throughout the axial plane on which a suspicious lesion could be delineated. We defined ¹⁸F-FDG PET/CT-detected focal thyroid incidentalomas as the presence of focal FDG uptake in the thyroid including very tiny foci of FDG uptake with a definite nodule on the CT images (Figure 1).

Immunohistological staining for estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) was performed using arrayed tissue blocks. Briefly, 5 μm-thick, formalin-fixed, paraffin-embedded tissue sections were obtained using a microtome, transferred to adhesive slides, and dried at 62°C for 30 minutes. Sections were incubated, using a benchmark automatic immunostaining device (Ventana Medical Systems, Tucson, USA), with primary antibodies against ER (1:50 dilution; Diona, Seoul, Korea), PR (1:100 dilution; Diona), and HER2 (1:250 dilution; DAKO, Glostrup, Denmark), followed sequentially by incubation with biotinylated antimouse immunoglobulin and peroxidase-labeled streptavidin, using 3.3´-diaminobenzidine chromogenic as the substrate. ER and PR staining were scored using the Allred scoring system on a scale from 0 to 8 for nuclear staining only. The intensity score and percentage score were added together to obtain the Allred score. Allred scores of 0 to 2 were classified as negative and Allred scores of 3 to 8 were classified as positive. HER2 was scored using the current American Society of Clinical Oncology/College of American Pathologist guidelines. Membranous immunoreactivity was scored (0 and 1+ indicates negative; 2+, indeterminate; and 3+, positive for overexpression) and the percentage of tumor cells staining positive was shown as raw score ranging from 0% to 100% [9]. Samples were considered positive for HER2 when strong (3+) membranous staining was observed in at least 10% of tumor cells, whereas those with staining of grades 0 to 2+ were regarded as negative. Tissue samples of HER2 2+ breast cancers underwent silver in situ hybridization analysis. For practical considerations in a hospital setting, as a substitute for gene expression profiling, the use of immunohistochemical surrogate panels of ER, PR, and HER2 has been proposed to potentially differentiate between the subtypes [10].

Fine-needle aspiration biopsy (FNAB) is recommended, regardless of the size of the nodule, if it presents sonographic signs suggestive of a malignancy according to the guidelines of the National Cancer Institute [11].

FNAB was performed with a 21-gauge needle on a 10 mL syringe under ultrasound guidance. A cytological diagnosis was made by experienced cytopathologists in our institution. Cytological diagnosis was as follows: (1) nondiagnostic (applies to specimens that are unsatisfactory due to blood, overlaid thick or air-dried smears, or an inadequate number of follicular cells); (2) benign; (3) atypia of undetermined significance (results of proliferating cell nuclear antigen cytology were not easily classified into benign, suspicious, or malignant categories); (4) follicular neoplasm (findings of high cellularity and scant or absent colloid); (5) suspicious for malignancy (only one or two characteristic features of papillary thyroid carcinoma were present, or a malignant diagnosis could not be made with certainty; (6) malignant (whenever the morphologic features were conclusive for malignancy) [12].

The following data were recorded for all breast cancer patients: age, body mass index, number of lymph node metastases, tumor size, histological type, lymph vascular invasion (yes/no), extensive intraductal component (yes/no), p53 expression (yes/no), Ki-67 expression >14% (yes/no), ER status, PR status, HER2 status (positive/negative), thyroid ultrasonographic findings, cytological results of focal thyroid incidentalomas, and SUVmax. We also collected the medical records of healthy women with focal thyroid incidentalomas, such as thyroid ultrasonographic findings, cytological results of the focal thyroid incidentaloma, and SUVmax. We compared the incidence of ¹⁸F-FDG PET-detected focal thyroid incidentalomas between patients with breast cancer and healthy women. We compared the clinicopathologic findings between patients with and without ¹⁸F-FDG PET-detected focal thyroid incidentalomas.

The statistical analyses were performed using SAS version 8.2 (SAS Institute Inc., Cary, USA). Clinical and demographic characteristics were compared using Pearson chi-squared test. SUVmax values were compared using the Mann-Whitney test. All p-values reported are two-tailed, and a p-value of less than 0.05 was considered statistically significant.

In 489 breast cancer patients, 46 patients (9.4%) had a focal thyroid incidentaloma detected by pretreatment ¹⁸F-FDG PET/CT. Seventeen healthy women (5.9%) had an ¹⁸F-FDG PET/CT-detected focal thyroid incidentaloma. The incidence of focal thyroid incidentalomas detected on ¹⁸F-FDG PET/CT was higher in breast cancer patients than in healthy women (p<0.01).

The breast cancer patients with focal thyroid incidentalomas (mean age, 52.9 years) was similar to them without focal thyroid incidentaloma (mean age, 51.2 years) (p=0.27). The breast cancer histologic types in the group with thyroid incidentalomas were more often ductal carcinoma in situ compared with the group without thyroid incidentalomas (19.8% vs. 14.5%, p=0.03) (Table 1). The group with thyroid incidentalomas had more ER positivity (56.5% vs. 50.0%), PR positivity (50% vs. 39.3%), and Ki-67 positivity (95.6% vs. 50.2%) than the group without thyroid incidentalomas (p<0.01) (Table 1). HER2 overexpression was higher in the group without focal thyroid incidentalomas than in the group with focal thyroid incidentalomas (23.2% vs. 15.2%, p<0.03). Tumor size, number of metastatic lymph nodes, lymphovascular invasion, extent of the intraductal component, and p53 expression were not different between the two groups (Table 1).

Aspiration was performed twice for each nodule. FNAB was performed in 23 of 37 focal thyroid incidentalomas (62.1%) in breast cancer and 5 of 12 thyroid incidentalomas (41%) in healthy subjects. The other 21 thyroid incidentalomas were diagnosed as benign nodular hyperplasia by high-resolution thyroid ultrasonography.

In 46 breast cancer patients with focal thyroid incidentalomas, 35 patients underwent thyroid ultrasonography. The results were as follows: six benign, 11 indeterminate, 11 suspected malignancies, and seven diffuse thyroiditis. Among them, 23 patients underwent ultrasound-guided FNAB, resulting in the diagnosis of 12 benign follicular neoplasms and 11 papillary carcinomas (11/46, 23.9%). The mean SUVmax of the focal incidentalomas confirmed as malignant thyroid carcinomas was 7.1±10.7, which was not significantly different from that of the benign focal incidentalomas (7.1± 6.2), (p=0.99). In 17 healthy women with focal thyroid incidentalomas, 12 underwent thyroid ultrasonography, which identified three benign, six indeterminate, and two suspicious malignant nodules. Among them, six patients underwent ultrasound-guided FNAB, resulting in the diagnosis of one benign follicular nodule, three atypia of indeterminate significance, and two suspicious malignant nodules (2/17, 11.7%). All of the malignant thyroid incidentalomas in breast cancer patients and healthy women were diagnosed as papillary thyroid cancer.