The aim of this study was to investigate the clinical value of 18F-fluoroestradiol (18F-FES) PET/CT in the assessment of the estrogen receptor (ER) and its intratumoral heterogeneity in breast cancer patients.Methods
Forty-six female patients (50 lesions) with histologically confirmed invasive breast cancer who underwent both 18F-FES and 18F-FDG PET/CT in our center were retrospectively included. All the patients enrolled were scheduled to undergo biopsy. The 18F-FES and FDG uptakes were compared with pathological features (tumor size, ER, progesterone receptor, human epidermal growth factor receptor 2, and Ki67%). The optimal threshold to discriminate ER-positive and ER-negative lesions was determined by receiver operating characteristic curve analysis. Furthermore, we observed the intratumoral heterogeneity by a heterogeneity index (SUVmax/SUVmean) and compared the results with the Chang-Gung Image Texture Analysis.Results
There was good agreement between 18F-FES uptake and ER, progesterone receptor, and human epidermal growth factor receptor 2 expression (P < 0.001), and the use of SUVmean instead of SUVmax can provide a slightly better correlation. The optimal threshold for 18F-FES PET/CT to discriminate between ER-positive and ER-negative lesions, as determined by receiver operating characteristic curve analysis, was an SUVmax of 1.82 (sensitivity = 88.2% and specificity = 87.5%) and SUVmean of 1.21 (sensitivity = 85.3% and specificity = 93.7). Our simplified heterogeneity index–FES can easily observe ER heterogeneity. In addition, our results suggested that recurrent/metastatic patients and lesions located other than breast might have greater heterogeneity.Conclusions
18F-FES PET/CT is a feasible, noninvasive method for assessing ER expression in breast cancer patients. Because intratumoral heterogeneity exists, 18F-FES PET/CT might better reflect the ER expression, especially in metastatic patients after treatment, thus assisting in making individualized treatment decisions.