Abstract
Background and purpose
Brain [18F] fluoroestradiol (FES) PET enables noninvasive assessment of estrogen receptor (ER) expression and may improve the detection of ER-positive (ER+) breast cancer brain metastases (BC-BM), with pilot studies demonstrating clinical utility. Our aim was to leverage our dynamic brain [18F] FES PET acquisition protocol to characterize BC-BM uptake kinetics across different postinjection intervals, with the goal of improving lesion detectability.Materials and methods
Patients with ER+ primary disease and BC-BM were enrolled prospectively and underwent dynamic [18F] FES PET/CT and contemporaneous standard-of-care contrast-enhanced brain MRI. PET was acquired during 90 minutes using a dynamic protocol (34 frames). Four 15-minute static time windows were reconstructed, including 30-45, 45-60, 60-75, and 75-90 minutes after injection, respectively. MRI was used as the criterion standard for clinical lesion characterization and segmentation, with subsequent [18F] FES PET co-registration. The time window-specific PET MRI lesion matching rate (LMR) and mean standardized uptake value ratio (SUVR, normalized to cerebellar cortex) were calculated.Results
Eleven patients with a total of 92 MRI-confirmed lesions were included (73 parenchymal, 8 leptomeningeal, 6 calvarial, 4 dural-based, and 1 pituitary stalk). Time window-3 yielded the highest LMR at 68% (63/92 lesions), followed by time window-2 (62%), time window-4 (60%), and time window-1 (57%). The mean SUVR increased across time, with means of 1.83 (standard deviation [SD], 1.46), 2.08 (SD, 1.88), 2.29 (SD, 2.37), and 2.38 (SD, 2.50) for time window-1, -2, -3, and -4, respectively. Despite a higher SUVR at later time points, 8 lesions (6 parenchymal, 1 leptomeningeal, and 1 calvarial) that were visible at 60-75 minutes were undetectable at 75-90 minutes.Conclusions
Time window-3 at 60-75 minutes after injection provided optimal LMR for brain [18F] FES PET in ER+ BC and suspected BM. Imaging beyond this interval may not further improve diagnostic yield and may reduce visibility. Limitations of this study include the small cohort size, population heterogeneity regarding prior treatment, and lack of pathologic confirmation. Our findings suggest that 60-75 minutes after injection may represent the optimal acquisition time window and highlight the importance of timing standardization for neuro-oncologic applications of brain [18F] FES PET.