Quantitative Imaging of Estrogen Receptor Expression in Breast Cancer with PET and 18F-Fluoroestradiol

developed and tested as an agent for the imaging of estrogen receptor (ER) expression in vivo. 18F-FES uptake has been shown to correlate with ER expression assayed in vitro by radio-ligand binding; however, immunohistochemistry (IHC) rather than radioligand binding is used most often to measure ER ex-pression in clinical practice. We therefore compared 18F-FES uptake with ER expression assayed in vitro by IHC with both qualitative and semiquantitative measures. Methods: Seven-teen patients with primary or metastatic breast cancer were studied with dynamic 18F-FES PET; cancer tissue samples, col-lected close to the time of imaging, were assayed for ER expres-sion by IHC. For each tumor, partial-volume-correctedmeasures of 18F-FES uptakewere comparedwith ER expressionmeasured by 3 different ER scoringmethods: qualitative scoring (0–31), the Allred score (0–10), and a computerized IHC index. Results: There was excellent agreement (r5 0.99) between observers us-ing IHC as well as the different methods of measuring ER content (P, 0.001). ER-negative tumors had 18F-FES partial-volume-corrected standardized uptake values of less than 1.0, whereas ER-positive tumors had values above 1.1. Correlation coeffi-cients for the different measures of ER content and the different measures of 18F-FES uptake ranged from 0.57 to 0.73, with the best correlation being between the computerized IHC index and 18F-FES partial-volume-corrected standardized uptake values. Conclusion: Our results showed good agreement be-tween 18F-FES PET and ER expression measured by IHC. 18F-FES imaging may be a useful tool for aiding in the assessment of ER status, especially in patients with multiple tumors or for tu-mors that are difficult to biopsy