Abstract
BACKGROUND AND PURPOSE:
[18F]-fluoroestradiol (FES) is a novel molecular imaging radiopharmaceutical that has demonstrated promise in the staging of ER-positive brain metastases. Our aim was to determine if hybrid [18F] FES brain PET/MRI is superior to conventional brain MRI alone for detecting estrogen receptor (ER)–positive brain and calvarial metastases, compare limited [18F] FES brain PET with vertex-to-below the knees [18F] FES PET results, and determine the impact of [18F] FES brain PET/MRI on patient management decisions.
MATERIALS AND METHODS:
Following institutional review board approval, a retrospective analysis was conducted of patients who underwent [18F] FES brain PET/CT and contrast-enhanced brain MRI at our institution from January 2024 to November 2024. Intracranial lesions, including parenchymal, dural-based, calvarial, leptomeningeal, and extracranial soft tissue/scalp, were analyzed by an expert neuroradiologist using fused [18F] FES brain PET/MRI, and lesion size, location, and maximum standard uptake value were recorded. Two radiology residents reviewed separately acquired contrast-enhanced brain MRI reports and documented the number and location of intracranial lesions. A nuclear medicine physician reviewed vertex-to-below the knees [18F] FES PET/CT and reported the intracranial lesion number and maximum standard uptake value.
RESULTS:
Ten female patients with ER-positive breast cancer with 90 metastatic brain and calvarial lesions were included in our analysis. The mean age was 64.3 (SD, 7.5) years. Twelve of 90 (13.3%) lesions were occult on contrast-enhanced brain MRI but detected by 18F-FES brain PET/MRI, including 1 parenchymal lesion, 8/19 (42.1%) calvarial lesions, and 5/6 (50.0%) extracranial soft-tissue/scalp lesions. Meanwhile, 16/90 (17.8%) lesions were occult on vertex-to-thigh 18F-FES PET/CT compared with [18F] FES brain PET/MRI, including 13/58 (22.4%) parenchymal lesions and 3/7 (42.9%) dural-based lesions. There was a modification in patient management for 9 of 10 patients (90%) following the [18F] FES brain PET/MRI.
CONCLUSIONS:
By integrating [18F] FES brain PET/CT/MRI into our clinical workflow, we improved the detection of ER-positive lesions, resulting in a substantial impact on clinical management decisions.
Breast cancer is one of the most common malignancies in the US and the second most common cancer to metastasize to the brain.1 Estrogen receptor (ER)-positive breast cancer has a particular predilection for metastasis to the CNS. Contrast-enhanced MRI is currently the criterion standard for detecting intracranial metastatic disease. Although highly sensitive, conventional MRI lacks the ability to provide information on the receptor status of metastases, which is particularly relevant for the management of ER-positive breast cancer. Furthermore, treatment of brain metastatic disease from breast cancer typically consists of surgical resection and radiation therapy. Unfortunately, it can be particularly challenging to distinguish radiation changes from recurrence on treatment-monitoring brain MRI scans. FDG PET is frequently used to differentiate posttreatment changes from viable tumor. However, there are several limitations to FDG PET when used in the brain, particularly the high physiologic uptake inherent to the brain parenchyma.1 Some subtypes of breast cancer, particularly invasive ductal carcinoma and invasive lobular carcinoma, typically ER-positive, demonstrate relatively low FDG avidity.2 In addition, inflammatory tissue can also demonstrate high FDG uptake, further complicating its utility. For these reasons, the diagnostic performance of FDG PET can vary considerably.1 Radiolabeled amino acid and hormone PET tracers are particularly promising for intracranial lesion detection because they demonstrate low physiologic uptake in the brain parenchyma, offering a better tumor-to-background ratio.3
[18F] fluoroestradiol (18F-FES) is a novel molecular imaging radiopharmaceutical that has demonstrated promise in the staging of ER-positive brain metastases. 18F-FES attaches to sex steroid-binding protein, which allows for the noninvasive quantification of tumor ER in primary and metastatic lesions. While initially used in humans as early as 1988, and rats in 1984, the US Food and Drug Administration (FDA) only approved the use of 18F-FES for the detection of ER-positive lesions on May 20, 2024.4,5 18F-FES was the first PET tracer used to image a receptor in malignancy and marked a diagnostic milestone in cancer imaging.4 The use of 18F-FES PET has also been demonstrated in other neoplasms expressing ERs, including uterine and ovarian cancers.5 While 18F-FES PET is particularly valuable when tissue sampling is unsafe, its use in conjunction with biopsy or immunohistochemistry has had a positive impact on clinical, economic, and quality of life metrics.6
The FDA has approved the use of 18F-FES PET for the detection of ER-positive lesions as an adjunct to biopsy in patients with recurrent or metastatic breast cancer. The FDA also indicates that tissue sampling is required to confirm breast cancer recurrence and ER status.2,7 In 2022, the Society of Nuclear Medicine and Molecular Imaging published appropriate-use criteria indicating that 18F-FES PET is most appropriate when endocrine therapy is being considered at initial diagnosis or after progression on endocrine therapy; ER status is difficult or dangerous to determine via tissue sampling; and ER status is inconclusive.2
Compared with 18F-FDG PET, 18F-FES PET confers greater lesion sensitivity and has been shown to alter treatment plans in patients with newly diagnosed ER-positive breast cancer. One study demonstrated 90.8% of lesions detected on 18F-FES PET compared with 82.8% of lesions detected by 18F-FDG PET. This result changed treatment plans in 26.3% of patients.8 In the posttreatment setting, 18F-FES PET has demonstrated superiority to 18F-FDG PET in differentiating ER-positive viable neoplasms from posttreatment change.9
We recently introduced an 18F-FES brain PET/CT/MRI protocol in routine clinical practice for patients with ER-positive metastatic breast cancer and intracranial lesions. The protocol includes a limited 18F-FES-PET of the brain, fusion with a separately acquired contrast-enhanced brain MRI using commercially available software, and also a standard protocol skull vertex to below the knees 18F-FES PET/CT. The purpose of this study was to compare the diagnostic outcomes and clinical impact of 18F-FES brain PET/MRI to conventional structural MRI imaging alone, to vertex-to-below the knees 18F-FES-PET/CT, and to assess for resulting changes in patient management decisions. Notably, our results demonstrated hybrid 18F-FES brain PET/CT/MR imaging’s superior efficacy in identifying occult lesions not well visualized on conventional MRI, which subsequently influenced the clinical management decisions for patients with ER-positive intracranial metastases.
MATERIALS AND METHODS
The methodology proposed by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) initiative was followed in the reporting of this investigation.
Patient Selection
The subjects were consecutive patients scanned at our institution with 18F-FES brain PET from January 2024 to November 2024. Inclusion criteria were patients with a confirmed ER-positive breast cancer diagnosis; known or suspected ER-positive metastatic breast cancer based on clinical symptoms, MRI findings, or previous imaging studies; and a clinical indication for 18F-FES PET/CT to guide treatment decisions, such as differentiating between viable tumor tissue and posttreatment changes. Exclusion criteria included patients who were recently treated with drugs that can interfere with the availability of ERs, such as selective ER modulators (SERMs), selective ER down-regulators (SERDs), and ER antagonists, because an insufficient discontinuation period may lead to false-negative results on 18F-FES PET/CT imaging.
Beginning in June 2023, before scheduling patients with known or suspected ER-positive metastatic breast cancer for 18F-FES PET/CT, clinical notes and structural MR brain imaging were reviewed by a nuclear medicine physician before scheduling. For patients with suspected ER-positive brain metastases, as identified through clinical notes and structural MRI/CT imaging, referrals were Protocolled to include imaging from the skull VX-BTK and a dedicated brain/head acquisition. Additionally, an “Ask at Order Entry” question was added to the VX-BTK FES PET/CT order in our electronic medical record system, prompting the referring provider to indicate concern for ER-positive disease in the brain/head. If “yes,” the system automatically generates an order for additional dedicated brain/head imaging.
Image Acquisition
Our dedicated 18F-FES brain PET/CT protocol is highlighted in Supplemental Data.
Notably, 18F-FES PET/CT was performed 50–60 minutes after IV administration of 222-MBq (6-mCi) 18F-FES using one of the following scanners: Discovery 710HD (GE Healthcare), Omni Legend (GE Healthcare), or Biograph Vision 450 (Siemens). Imaging and reconstruction parameters for these scanners are detailed in the Supplemental Data. Brain PET data were corrected for attenuation using CT and fused with separately-acquired contrast-enhanced MRI of the brain using MIMNeuro software (Version 7.3.2; MIM Software).
Conventional MRI sequences were obtained on the following scanners: Discovery 710HD (GE Healthcare), Omni Legend (GE Healthcare), and Biograph Vision 450 (Siemens). MRI sequences obtained included T1 FLAIR sagittal, T1 FLAIR axial, T2 axial, 3D SWI/3D SWAN or 2D gradient recalled echo T2* axial, DSC perfusion axial, DWI/ADC axial, T2 FLAIR axial, postcontrast 3D T1 MPRAGE/BRAVO (GE Healthcare) axial, and postcontrast 3D T1 FLAIR axial.
Image Analysis
Following Northwell institutional review board (IRB) approval, assessment of intracranial lesions across the imaging modalities was performed, and lesion locations were reported as follows: parenchymal, dural-based, calvarial, leptomeningeal, and extracranial soft tissue/scalp. A board-certified fellowship-trained expert neuroradiologist with 10 years of clinical and research experience in brain PET/MRI analyzed all intracranial lesions on fused 18F-FES brain PET/MR images. A fellowship-trained nuclear medicine physician with 25 years of experience reported all intracranial lesions on the VX-BTK 18F-FES PET/CT. Fellowship-trained neuroradiologists interpreted the conventional MRI brain studies as part of routine clinical care. Two radiology residents separately reviewed these contrast-enhanced MRI brain reports and documented the number and location of intracranial lesions. The patient file, including relevant therapeutic and surgical interventions and pertinent imaging findings, was reviewed during real-life tumor board discussions with a multi disciplinary team consisting of neurosurgeons, radiation oncologists, neuro-oncologists, and neuroradiologists. Subsequent changes in patient management based on these collaborative discussions were recorded.
Statistical Analysis
We began by describing the sample using frequencies and percentage or mean (SD). We stratified lesions by cerebral location and detection on contrast-enhanced brain MRI. Maximum standard uptake value (SUVmax) represents a truncated variable that is abnormally distributed, so we relied on generalized linear modeling to estimate a log-γ model in FES-positive lesions. Log-γ models can reliably estimate biologic processes that are truncated at a floor and may be skewed by pathologic processes. Since biologic affinity values are often biased by individual differences in underlying affinity for a ligand, we used mixed effects models to account for these individual differences using random intercepts at the individual level. One benefit of this approach is that it is robust to the inclusion of covariates at the individual level, including, for example, participant age. A likelihood ratio test was used to determine whether this adjustment improved model fit over a patient-naïve lesion model. Statistical analysis was performed and a P value of < .05 was considered statistically significant. All statistical analyses were completed using Stata 17/MP (StataCorp).
RESULTS
Ten female patients with ER-positive breast cancer (mean age, 64 [SD, 7.5] years) with 92 intracranial lesions were included. Eight patients had luminal type A breast cancer, 3 of whom had HER2+ disease. Because comparisons with brain MRI were impossible with lesions outside the field of view and because these extracranial lesions had unusually high SUVmax values (mean, 10.2 [SD, 5.4]) potentially indicating quantification failures, we excluded two 18F-FES PET-positive extracranial soft-tissue lesions from further analysis.
Of the 90 lesions included in our analysis, 58 were parenchymal lesions; 7, dural-base; 19, calvaria; and 6, extracranial soft-tissue/scalp intracranial lesions (Figs 1–3 and Table 1). Although 18F-FES brain PET can detect leptomeningeal metastases from ER-positive breast cancer as described by Ryu et al10 in 2022, there were no leptomeningeal lesions identified in this patient cohort. Lesion statistics aggregated by unique patients are described in Table 2. Additional lesion-specific contrast-enhanced brain MRI, 18F-FES brain PET/MRI, and VX-BTK 18F-FES PET/CT characteristics are summarized in the Supplemental Data.
FIG 1.
A 57-year-old woman with ER/PR positive breast cancer metastatic to the brain, status post stereotactic radiosurgery. A, Initial 18F-FES brain PET/MRI demonstrates avid FES uptake (SUVmax, 4.12) in a right frontal paramedian homogeneously-enhancing intraparenchymal lesion, consistent with active ER-positive metastatic disease (red arrow). Conventional MRI findings demonstrate a mild increase in size of the 2.6 × 1.7 × 2.6 cm right paramedian frontal lesion (previously, 2.3 × 1.6 × 2.5 cm) with a thick peripheral rim of hyperperfusion, consistent with viable tumor, and central hypoperfusion, consistent with necrosis (not shown). Overall, findings were concerning for progression of tumor. B, Repeat 18F-FES brain PET/MRI following radiation therapy demonstrates resolution of FES uptake associated with the index lesion (red arrow) and interval development of a subcentimeter FES-avid enhancing lesion in the right paramedian cingulate gyrus, SUVmax, 6.38 (white arrow). These findings were corroborated on the brain MRI, which demonstrated a decrease in size of the 1.5 × 1.0 × 2.1 cm right paramedian frontal lesion (previously, 1.8 × 1.3 × 2.8 cm) and marked improvement in the amount of associated edema. Interval development of a smaller enhancing lesion just caudal to the right paramedian frontal lesion in the right cingulate gyrus is also identified. PR indicates progesterone receptor.
FIG 3.
A 67-year-old woman with triple-positive breast cancer. The patient received chemotherapy and immunotherapy. In 2023, she was found to have a left frontal calvarial and dural-based metastatic lesion, treated with stereotactic radiosurgery. 18F-FES brain PET/MRI demonstrates multifocal, increased FES uptake (SUVmax 1.5) in the peripheral enhancing portions of the dominant dural-based lesion in the left frontal lobe, red arrows, concerning for progression of disease. Note prominent white matter background uptake in this patient. Conventional MRI findings were suggestive of disease progression, as demonstrated by an increase in the enhancing soft-tissue component of the left frontal dural-based lesion. Perfusion data suggested hyperperfusion in the areas of enhancement in the peripheral portions of the dominant dural-based mass, while the internal portions demonstrate hypoperfusion (not shown).
Table 1:
Detection of ER-positive intracranial metastatic disease by dedicated [18F]-FES brain PET with MR imaging compared with contrast-enhanced brain MRI and VX-BTK [18F]-FES on PET/CT, and associations between lesion characteristics and SUVs
| Comparison | Patient Characteristics | Lesion Characteristics | Results from Generalized Linear Multilevel Model on SUVmax | ||
|---|---|---|---|---|---|
| All Patients (N = 10) | All Lesions (L = 90) | 18F-FES Negative (L = 37) | 18F-FES Positive (L = 53) | Coef. (SE) P Value | |
| Occult (vs not) | 3 (30%) | 14 (15.6%) | 0 (0%) | 14 (26.4%) | 0.1 (0.24), .702 |
| Location | |||||
| Parenchymal | 9 (90%) | 58 (64.4%) | 34 (91.9%) | 24 (45.3%) | –0.06 (0.19), .781 |
| Dural | 4 (40%) | 7 (7.8%) | 3 (8.1%) | 4 (7.5%) | –0.35 (0.24), .184 |
| Calvarial | 4 (40%) | 19 (21.1%) | 0 (0%) | 19 (35.8%) | 0.13 (0.16), .468 |
| Extracranial | 3 (30%) | 6 (6.7%) | 0 (0%) | 6 (11.3%) | 0.09 (0.2), .684 |
| Multiloci | 6 (60%) | – | – | – | – |
| VX-BTK PET-positive | 8 (80%) | 37 (41.1%) | – | 37 (69.8%) | 0.56 (0.53,) .318 |
| VX-BTK-occult | 4 (40%) | 16 (17.8%) | 0 (0%) | 16 (30.2%) | – |
| SUVmax, mean (SD)a | 3.2 (1.8) | 3.4 (2.3) | – | 3.4 (2.3) | –0.68 (0.16), .003 |
| VX-BTK SUVmax, mean (SD) | 3.2 (1.3) | 3.7 (2) | – | 3.7 (2) | –0.2 (0.02), 0 |
Note:—The SD indicates standard deviation; SUV, standardized uptake value; VX-BTK, vertex to below the knees; PET, positron emission tomography; N, number; L, lesion.
Results from multiple independent generalized linear multilevel models with a γ distribution and a log-link function showing the association between comparison variables and SUVmax after adjusting for age and within-person differences in the propensity for specific node types. Regression coefficients are reported alongside standard errors and P values from t tests.
Table 2:
Patient characteristicsa
| Patient | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| Age range (yr) | 55–59 | 65–69 | 60–64 | 70+ | 55–59 | 60–64 | 55–59 | 65–69 | 70+ | 70+ |
| No. lesions | 3 | 4 | 2 | 3 | 13 | 17 | 6 | 20 | 9 | 13 |
| SUVmax | 5.25 | 0.98 | 2.30 | 0.00 | 5.35 | 1.81 | 1.20 | 0.00 | 4.64 | 3.80 |
| WB SUVmax | 3.60 | 0.00 | 2.00 | 0.00 | 3.94 | 2.46 | 1.50 | 0.00 | 5.21 | 3.48 |
| FES-positive | X | X | X | Y | X | X | X | Y | X | X |
| MRI-positive | X | X | X | X | X | X | X | X | X | X |
| VX-BTK-occult | Y | X | Y | Y | Y | X | X | Y | Y | Y |
| VX-BTK-positive | X | Y | X | Y | X | X | X | Y | X | X |
| MRI-occult | X | X | X | X | Y | X | X | X | Y | Y |
Note:—SUV indicates standardized uptake value; VX-BTK, vertex to below the knees; PET, positron emission tomography.
The table show cases in which individuals had evidence of positive (X) or negative (Y) lesions by specific characteristics.
FIG 2.
A 62-year-old woman with metastatic breast cancer (HER2+) status post lumpectomy, adjuvant chemotherapy, radiation treatment, and immunotherapy. She had a history of intracranial metastases since 2016, treated by multiple stereotactic radiosurgeries and resections in 2021 and 2022. 18F-FES brain PET/MRI demonstrates avid FES uptake (SUVmax, 2.29) associated with a heterogeneously enhancing lesion involving the cerebellar vermis and left mesial superior cerebellar hemisphere, findings consistent with active metastatic disease (red arrow). Notably, there is no evidence of tracer uptake associated with the heterogeneously enhancing lesion in the left anterior inferior temporal lobe (yellow arrow), which was biopsy-proved radiation necrosis. Conventional MRI findings demonstrate an increase in size and degree of enhancement of the 1.4 × 1.9 × 1.3 cm vermian and left mesial superior cerebellar hemisphere lesion (previously, 1.0 × 1.1 × 1.1 cm) with increased surrounding vasogenic edema, concerning for disease progression. There is no interval change in the heterogeneously enhancing left anterior temporal lesion.
In total, 12/90 (13.3%) lesions were occult by contrast-enhanced MRI alone and detected only by 18F-FES brain PET/MRI, including 1/58 (1.7%) parenchymal lesions, 8/19 (42.1%) calvarial lesions, and 3/6 (50.0%) extracranial soft-tissue/scalp lesions. These MRI-occult lesions were not detected by any pre- or postcontrast MRI sequence. Please note that in the instance of the single parenchymal lesion occult by MRI, the FES PET was acquired 1 month following the brain MRI, with the FES-avid parenchymal focus prompting repeat structural imaging. In addition, none of the calvarial and extracranial soft-tissue/scalp lesions were an isolated finding, with each patient also demonstrating intracranial FES-avid metastatic disease.
Meanwhile, 16/90 (17.8%) lesions were occult on VX-BTK 18F-FES PET/CT compared with 18F-FES brain PET/MRI, including 13/58 (22.4%) parenchymal lesions and 3/7 (42.9%) dural-based lesions (Supplemental Data). Furthermore, 53/90 (58.8%) lesions were positive on 18F-FES brain PET, while 37/90 (41.1%) were negative on 18F-FES brain PET, indicating the presence or absence of ER+ metastatic disease, respectively. Two patients had no FES-positive lesions, while 2 patients had no FES-negative lesions (Supplemental Data). On the basis of these results, the sensitivity according to the following variables is as follows: 84% using the MRI-positive variable, 59.8% using the FES-positive variable, 82.6% using the MRI nonoccult variable, and 42.4% using the “VX-BTK PET-positive” variable.
There was statistically significant variability in the mean SUVmax values between patients (χ2 = 26.31, P < .001), supporting the use of a multilevel modeling approach. SUVmax values for 18F-FES brain PET-positive lesions ranged from 0.77 to 11.72, with an average value of 3.4 (SD, 2.3; skewness = 1.3, kurtosis = 5.2). In Table 1, results from log-γ models predicting lesion-level differences in SUVmax revealed that 18F-FES PET values in the brain were strongly associated with 18F-FES PET values on the VX-BTK 18F-FES PET/CT scan (P < .001). We also saw that SUVmax was significantly associated with occult status on the VX-BTK 18F-FES PET/CT scan (P = .003).
Each case was also reviewed by a multidisciplinary team to determine whether the additional information from 18F-FES brain PET/MRI influenced changes in treatment management, including further radiation therapy/systemic therapy or continued surveillance strategies. Notably, there was a modification in the management approach for 9/10 patients (90%) following the 18F-FES brain PET/MRI. Specific changes included adjustments in radiation therapy treatment planning, altered systemic therapy plans, and/or continued surveillance. For example, 18F-FES brain PET/MRI was used to distinguish between tumor recurrence and radiation necrosis in 14 lesions that were deemed equivocal following contrast-enhanced brain MRI and a subsequent multidisciplinary discussion at our institutional tumor board. Of these 14 lesions, 4/14 (28.5%) were identified as representing residual/recurrent ER-positive disease on the basis of 18F-FES brain PET/MRI. These lesions were subsequently treated with a repeat course of radiation or initiation of systemic therapy and demonstrated improved posttreatment structural imaging.
Two patients had 18F-FES PET ordered to evaluate the extent of disease. One of these patients had clinical uncertainty between metastatic breast cancer and meningioma based on contrast-enhanced brain MRI findings of an extra-axial dural-based enhancing lesion. Further evaluation of the lesion with 18F-FES brain PET/MRI revealed a lack of accompanying FES uptake, favoring meningioma rather than metastatic ER-positive breast cancer. 64Cu-DOTATATE brain PET was subsequently recommended to confirm somatostatin receptor–bearing tumor. The second patient presented after whole-brain radiation therapy for intracranial metastatic disease and was noted to have 30-subcentimeter enhancing lesions by brain MRI. Of the 27 enhancing lesions noted on contrast-enhanced brain MRI, 14/27 (51.8%) were FES-avid on 18F-FES brain PET/MRI, indicating persistent ER-positive disease despite prior radiation therapy. In this patient, FES avidity helped differentiate active disease from post-radiation sequelae. Importantly, three 18F-FES-avid active metastatic lesions were occult on structural imaging.
DISCUSSION
Recent advances and increased availability of hybrid PET/CT and PET/MRI systems are revolutionizing neuro-oncologic imaging. Specifically, the use of radiolabeled amino acid and hormone PET tracers, which bind to specific receptors on tumors, offers improved accuracy in defining the tumor-versus-background and guiding treatment strategies.11 In this study, we demonstrated improved detection of ER-positive lesions using hybrid 18F-FES brain PET/CT/MRI compared with contrast-enhanced brain MRI alone. These scans included not only intracranial lesions but improved detection of calvarial and extracranial soft-tissue/scalp lesions. Furthermore, nearly one-fifth of ER-positive intracranial lesions detected on 18F-FES brain PET/MRI were not identified on same-day VX-BTK 18F-FES PET/CT, emphasizing the importance of dedicated brain FES PET imaging for patients with clinical suspicion of intracranial metastatic disease. Importantly, 18F-FES brain PET/MRI altered management decisions in all except 1 patient when reviewed by a multidisciplinary team.
In the management of brain metastases from ER-positive breast cancer, accurate differentiation between viable tumor tissue and treatment-related changes is crucial. Both conventional MRI and 18F-FES PET play important roles in this context. MRI remains the criterion standard for initial detection and anatomic assessment of brain lesions due to its excellent soft-tissue contrast. However, in the postsurgical or postradiation setting, MRI lacks specificity in distinguishing between tumor recurrence and posttreatment effects such as radiation necrosis. 18F-FES PET offers high specificity for identifying viable ER-positive disease by targeting ERs directly, helping differentiate tumor progression from treatment sequelae more effectively than conventional MRI. However, the high specificity of 18F-FES PET is contingent on specific conditions, namely consistent homogeneous ER expression across lesions and discontinuation of SERM before imaging. Limited availability and cost are additional prohibitive factors. By leveraging the strengths and complementary roles of conventional MRI and 18F-FES PET through hybrid 18F-FES PET/MRI, we created a comprehensive, patient-tailored approach in the evaluation of suspected or confirmed brain metastases from ER-positive breast cancer.
Hybrid PET/MRI allows the systematic addition of high-resolution MR imaging to PET, thereby providing precise and consistent anatomic information and helping to overcome difficulties in localization inherent to PET, and may help exclude or identify the presence of multiple pathologies.4 Hybrid 18F-FES PET/MRI leverages the high-resolution anatomic information provided by MRI with the specific receptor-based molecular data conferred by 18F-FES PET to optimize detection and identification of ER-positive metastases. The increased sensitivity and specificity of hybrid 18F-FES PET/MRI for ER-positive lesions contributes to a comprehensive assessment of the anatomic and molecular tumor profile, thereby addressing the limitations of structural imaging. While PET/MRI is superior to either technique alone,12 the availability of simultaneous PET/MRI scanners in clinical practice remains limited. In the absence of a dedicated hybrid PET/MRI system, an FES-PET may be fused with separately-acquired contrast-enhanced brain MRI using commercially available software, as presented here.
In the normal brain, uptake of 18F-FES has been demonstrated in regions of high ER concentration, including the pituitary gland and the hypothalamus, in both rats and healthy postmenopausal women. 18F-FES uptake has also been demonstrated in the white matter, which is not counteracted by the administration of ER antagonists, whereas the physiologic uptake in the pituitary gland is.13 18F-FES PET dosimetry is comparable with other widely used nuclear medicine radiopharmaceuticals and is far below the maximum established whole-body dose limit. One study determined the effective dose equivalent was 0.022 mSv/MBq, with organs receiving the highest dose including the liver, gallbladder, and urinary bladder.14 The consumption of chocolate (used as a fatty meal) between tracer injection and imaging decreased 18F-FES uptake in the gallbladder and stomach compared with fasting or control groups.15
Endocrine therapy using an estrogen blockade is frequently used in patients with ER-positive breast cancer. The presence or absence of 18F-FES uptake may help determine the suitability of such treatment in select patients.9 While aromatase inhibitors such as anastrazole decrease estrogen (ER ligand), the binding of 18F-FES is not altered.16 However, selective R modulators (SERMs) and selective ER down-regulators (SERDs) reduce the uptake of 18F-FES. For patients already on an estrogen blockade, this result can be an obstacle to imaging with 18F-FES PET. According to GE Healhtcare’s patient preparation protocol, SERMs, such as clomiphene and tamoxifen, require an 8-week washout period, and SERDs, such as fulvestrant, require a 28-day washout period before 18F-FES PET. This requirement may not be feasible or appropriate in some patients.
The combination of FDG PET and 18F-FES-PET has been shown to predict progression-free survival, particularly in patients with ER-positive breast cancer on endocrine therapy. FDG-avid tumors with a lower 18F-FES uptake conferred the worst progression-free survival (median of 3.3 months). FDG-avid tumors with a high 18F-FES uptake conferred an intermediate progression-free survival (median, 7.9 months). Patients with tumors demonstrating low FDG avidity had the longest progression-free survival (median, 26.1 months).16
There are several limitations to our study. First, with only 10 patients included, the small sample size limits generalizability. Future studies should aim to recruit a larger cohort, possibly through multicenter collaborations, allowing subgroup analyses to identify additional trends. Second, our PET/MR images were not obtained using a dedicated hybrid machine but rather by the fusion of separately acquired PET and contrast-enhanced MRI. The advantage of such an approach is that it provides a credible characterization of what may be reasonably achieved by a center that does not have access to a dedicated PET/MRI machine, being currently limited in availability. As more PET/MRI systems become integrated into clinical practice, future studies can use these dedicated systems to simultaneously acquire images, reducing alignment issues and improving image co-registration. An important caveat is that as many as 30% of patients can undergo conversion to ER-negative disease in the setting of adjuvant chemotherapy.17 Therefore, while an FES study with positive findings is very informative, a study with negative findings should be carefully interpreted within the context of all the other clinical, pathologic, and imaging information available.18 Future multicenter, longitudinal studies would help provide additional insight into changes in ER status with time, which may impact therapeutic interventions.
CONCLUSIONS
By integrating 18F-FES brain PET/CT/MRI into our clinical workflow, we demonstrated improved detection of ER-positive lesions compared with structural imaging and VX-BTK 18F-FES PET/CT. Notably, 18F-FES brain PET/MRI had a significant impact on patient management. This result demonstrates the potential pivotal role that 18F-FES brain PET/MRI may play in presurgical and radiation therapy planning for patients with ER-positive metastatic breast cancer with CNS metastases. FES PET further demonstrates promise for monitoring treatment response and evaluating tumor recurrence in these patients. The specificity for ER-positive disease offers the potential for theranostics, with significant implications for personalized and precise therapeutic interventions.
ABBREVIATIONS:
- ER
estrogen receptor
- FES
fluoroestradiol
- SERD
selective ER down-regulator
- SERM
selective ER modulator
- SUVmax
maximum standard uptake value
- VX-BTK
vertex to below the knees
Footnotes
Disclosure forms provided by the authors are available with the full text and PDF of this article at www.ajnr.org.
REFERENCES
- 1.Galldiks N, Langen KJ, Albert NL, et al. PET imaging in patients with brain metastasis-report of the RANO/PET group. Neuro Oncol 2019;21:585–95 10.1093/neuonc/noz003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ulaner GA, Mankoff DA, Clark AS, et al. Summary: appropriate use criteria for estrogen receptor-targeted PET imaging with 16α-18F-Fluoro-17β-fluoroestradiol. J Nucl Med 2023;64:351–54 10.2967/jnumed.123.265420 [DOI] [PubMed] [Google Scholar]
- 3.Urso L, Bonatto E, Nieri A, et al. The role of molecular imaging in patients with brain metastases: a literature review. Cancers (Basel) 2023;15:2184 10.3390/cancers15072184 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Katzenellenbogen JA. The quest for improving the management of breast cancer by functional imaging: the discovery and development of 16α-[18F]fluoroestradiol (FES), a PET radiotracer for the estrogen receptor, a historical review. Nucl Med Biol 2021;92:24–37 10.1016/j.nucmedbio.2020.02.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Huang YT, Chen TW, Chen LY, et al. The application of 18 F-FES PET in clinical cancer care: a systematic review. Clin Nucl Med 2023;48:785–95 10.1097/RLU.0000000000004760 [DOI] [PubMed] [Google Scholar]
- 6.Munter-Young R, Fuentes-Alburo A, DiGregorio N, et al. Clinical and economic outcomes of adding [18F]FES PET/CT in estrogen receptor status identification in metastatic and recurrent breast cancer in the US. PLoS One 2024;19:e0302486 10.1371/journal.pone.0302486 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Food and Drug Administration (FDA). CERIANNA (fluoroestradiol F 18). Available from https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/212155s000lbl.pdf. Accessed March 15, 2025
- 8.Liu C, Gong C, Liu S, et al. 18F-FES PET/CT influences the staging and management of patients with newly diagnosed estrogen receptor-positive breast cancer: a retrospective comparative study with 18F-FDG PET/CT. Oncologist 2019;24:e1277–85 10.1634/theoncologist.2019-0096 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ivanidze J, Sharbatdaran A, McCalla A, et al. [18F]-Fluoroestradiol (FES) brain PET in the evaluation of patients with estrogen receptor positive breast cancer and known or suspected brain metastases. Eur J Radiol 2024;181:111791 10.1016/j.ejrad.2024.111791 [DOI] [PubMed] [Google Scholar]
- 10.Ryu J, Jeong JH, Moon DH, et al. Determination of the estrogen receptor status of leptomeningeal metastasis in patients with metastatic breast cancer using [18F]-FES PET/CT: a case report. Nucl Med Mol Imaging 2022;56:105–09 10.1007/s13139-022-00736-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Pimlott SL, Sutherland A. Molecular tracers for the PET and SPECT imaging of disease. Chem Soc Rev 2011;40:149–62 10.1039/b922628c [DOI] [PubMed] [Google Scholar]
- 12.Catana C. Principles of simultaneous PET/MR imaging. Magn Reson Imaging Clin N Am 2017;25:231–43 10.1016/j.mric.2017.01.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Iqbal R, Menke-van der Houven van Oordt CW, Oprea-Lager DE, et al. [18F] FES uptake in the pituitary gland and white matter of the brain. Eur J Nucl Med Mol Imaging 2021;48:3009–10 10.1007/s00259-021-05281-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Mankoff DA, Peterson LM, Tewson TJ, et al. [18F] fluoroestradiol radiation dosimetry in human PET studies. J Nucl Med 2001;42:679–84 [PubMed] [Google Scholar]
- 15.Boers J, Giatagana K, Schröder CP, et al. Image quality and interpretation of [18F]-FES-PET: is there any effect of food intake? Diagnostics (Basel) 2020;10:756 10.3390/diagnostics10100756 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kurland BF, Peterson LM, Lee JH, et al. Estrogen receptor binding (18F-FES PET) and glycolytic activity (18F-FDG PET) predict progression-free survival on endocrine therapy in patients with ER+ breast cancer. Clin Cancer Res 2017;23:407–15 10.1158/1078-0432.CCR-16-0362 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Djahansouzi S, Hanstein B, Rein D, et al. The rate of estrogen receptor-conversion associated with tumor progression in estrogen receptor-positive breast cancer patients following adjuvant tamoxifen administration. Cancer Rep (Hoboken) 2022;5:e1431 10.1002/cnr2.1431 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ivanidze J, Subramanian K, Youn T, et al. Utility of [18F]-fluoroestradiol (FES) PET/CT with dedicated brain acquisition in differentiating brain metastases from posttreatment change in estrogen receptor-positive breast cancer. Neurooncol Adv 2021;3:vdab178 10.1093/noajnl/vdab178 [DOI] [PMC free article] [PubMed] [Google Scholar]