64Cu-SAR-Bombesin PET-CT Imaging in the Staging of Estrogen/Progesterone Receptor Positive, HER2 Negative Metastatic Breast Cancer Patients: Safety, Dosimetry and Feasibility in a Phase I Trial

Breast cancers are most frequently oestrogen receptor (ER) and progesterone receptor (PR) positive and [18F]Fluorodeoxyglucose PET-CT (FDG) has lower sensitivity for these subtypes. The gastrin-releasing peptide receptor (GRPR) is overexpressed in ER+/PR+ breast cancers. This study assessed the safety and potential of [64Cu]Cu-Sarcophagine (SAR)-Bombesin PET/CT (BBN) in re-staging metastatic ER+/PR+/human epidermal growth-factor-2-negative (HER2-) breast cancer. Seven patients with metastatic ER+/PR+/HER2- breast cancer undergoing staging underwent [64Cu]Cu-SAR-BBN PET-CT. Bloods, vital signs and electrocardiogram, blood tracer-clearance and dosimetry were undertaken. GRPR status was assessed in available metastatic biopsy samples. Staging with conventional imaging ([18F]FDG, bone scan and diagnostic CT) was within 3 weeks of [64Cu]Cu-SAR-BBN PET/CT. PET scans were assessed visually and quantitatively. Seven patients underwent imaging. One of the seven had de-novo metastatic breast cancer and six of the seven recurrent metastatic disease. Two of the seven had lobular subtype. No adverse events were reported. All seven patients were positive on conventional imaging (six of seven on FDG). [64Cu]Cu-SAR-BBN imaging was positive in five of the seven. Both [64Cu]Cu-SAR-BBN-negative patients had disease identified on [18F]FDG. One patient was [64Cu]Cu-SAR-BBN positive/[18F]FDG negative. Four of seven patients were [64Cu]Cu-SAR-BBN positive/[18F]FDG positive. In these four, mean SUVmax was higher for [64Cu]Cu-SAR-BBN than [18F]FDG (SUVmax 15 vs. 12). In the classical lobular subtype (two of seven), [64Cu]Cu-SAR-BBN was more avid compared to [18F]FDG (SUVmax 20 vs. 11, and 20 vs. <3). Dosimetry calculations estimated whole-body effective dose for 200 MBq of [64Cu]Cu-SAR-BBN to be 1.9 mSv. [64Cu]Cu-SAR-BBN PET/CT appears safe and may have diagnostic value in metastatic ER+/PR+/HER2- breast cancer, particularly the lobular subtype. Further evaluation is warranted.


Introduction
Metastatic breast cancer carries a poor prognosis with an overall five-year survival of 32% [1]. Detection of breast cancer metastases is essential for accurate staging and treatment planning. [ 18 F]Fluorodeoxyglucose (FDG) PET-CT is now the standard of care for the staging of breast cancer in Australia. However, the heterogeneous nature of metastatic breast cancer means a targeted molecular tracer appropriate for one subtype of breast cancer may not be effective in another. Breast cancers that are oestrogen receptor (ER) positive, progesterone receptor (PR) positive and human epidermal growth factor receptor 2 (HER2) negative account for up to 80% of all breast cancers, and yet ER+/PR+ cancers have significantly lower avidity and sensitivity on FDG PET compared to triple-negative breast cancers (ER-/PR-/HER2-) [2][3][4].
The gastrin-releasing peptide receptor (GRPR) is overexpressed in breast cancer, particularly ER+/PR+ breast cancers and their metastases [5]. This makes GRPR a potential imaging and theranostic target in the diagnosis and treatment of metastatic breast cancer [5][6][7]. Bombesin is a tetra-decapeptide with high binding affinity to GRPR, and its uptake in breast cancer has been demonstrated in human trials [8,9].
Copper-64 is an alternative PET radioisotope and forms a chemically identical theranostic pair with Copper-67, a beta particle emitter. A GRPR antagonist conjugated to a sarcophagine derivative and radiolabelled with either copper-64 ([ 64 Cu]Cu-SAR-BBN) or copper-67 ([ 67 Cu]Cu-SAR-BBN) demonstrated high affinity for GRPR in animal studies [10,11]. This first human study aims to evaluate the safety and feasibility of [ 64 Cu]Cu-SAR-BBN as a theranostic tool in women with metastatic ER+/PR+/HER2-breast cancer.

Patient Characteristics
The patient characteristics are detailed in Tables 1 and 2. In total, nine patients were enrolled between July 2020 and July 2021. All enrolled patients had proven ER+/PR+/HER2breast cancer on histopathology prior to enrolment. Two of nine patients were enrolled but not imaged with [ 64 Cu]Cu-SAR-BBN, due to clinical deterioration. Seven patients underwent all imaging, including [ 64 Cu]Cu-SAR-BBN PET-CT. Median age was 60 (38-81), and six of seven patients had received previous lines of treatment, which included chemotherapy (three of seven), endocrine therapy (five of seven) and targeted therapy (six of seven). One of seven had newly diagnosed de-novo metastatic breast cancer. Two of seven patients had invasive lobular type and five of seven invasive ductal type on initial biopsy. Six of seven patients had metastatic disease in bone, four of seven in liver and four of seven in nodes on conventional imaging.

Safety and Adverse Events
No adverse events or change in vital signs or ECGs were reported. There were no changes to the haematological, biochemical or coagulation profiles between baseline and at 1 h post injection.

Dosimetry and Blood Clearance
The [ 64 Cu]Cu-SAR-BBN PET-CT scans and blood clearance data were acquired at three time points for five of seven patients and at two time points for two patients (for one patient, the 24 h and for the other patient the 3 h scan and blood data were not acquired). Dosimetry calculations estimated the whole-body effective dose to be 0.0095 mSv/MBq or 1.9 mSv for a 200 MBq injection of [ 64 Cu]Cu-SAR-BBN (Table 3). From all organs, the pancreas demonstrated the highest uptake but with good clearance over 24 h. Doses to critical organs, such as the bone marrow (mean ± SD: 0.013 ± 0.007 mGy/MBq) and kidney (mean ± SD: 0.036 ± 0.009 mGy/MBq), were acceptable, compared to other PET diagnostics.  Table 4. Blood data showed rapid clearance of blood pool activity and relatively rapid clearance of renal and hepatic activity.          Table 5 and combined patient findings detailed in Table 6. [ 64 Cu]Cu-SAR-BBN detected a higher mean total tumour volume (TTV) (413 vs. 164 mL) and mean SUVmax (SUVmax 11 vs. 9) across the seven patients combined compared to FDG. SUVmean across the total seven patients did not significantly differ between [ 64 Cu]Cu-SAR-BBN and FDG (mean SUVmean 4 vs. 4). Overall, [ 64 Cu]Cu-SAR-BBN identified a higher mean number of lesions compared to FDG (49 on BBN vs. 9 on FDG).

Biopsy of Metastatic Lesions and Receptor Status
GRPR receptor staining was performed on four metastatic biopsy specimens in four of seven patients (Table 7). GRPR staining was positive in one of four cases. This positive case was a sternal/chest wall lesion (patient 4), intensely avid on both [ 64 Cu]Cu-SAR-BBN-PET and FDG PET performed 7 weeks following biopsy. The other three cases were negative on GRPR staining, including a lesion positive on 64 Cu-SAR-BBN-PET. One lesion was negative on both 64 Cu-SAR-BBN-PET and GRPR receptor staining.

Discussion
We reported the first in-human trial evaluating [ 64 Cu]Cu-SAR-BBN as a potential theranostic tool in patients with ER/PR+ HER2-negative metastatic breast cancer. In addition, we describe the first in-human comparison of GRPR-targeted PET with current the standard of imaging, inclusive of FDG PET, in the same subset of patients.
The heterogeneity of breast cancer, both between and within individual patients, poses unique challenges in molecular imaging, making a "one size fits all" targeted approach not currently feasible. As a result, novel targeted radiotracers have been investigated, including [ 18 F]fluoro-17ß-estradiol (FES) targeting estrogen receptors, [ 64 Cu]Cu-DOTAtranstuzumab targeting HER 2 and [ 68 Ga]Ga-PSMA targeting endothelial expression of PSMA [15]. Targeting GRPR is of particular interest, as Morgat et al. observed discordant binding of FDG and GRPR on breast cancer samples, where binding of [ 68 Ga]Ga-RM2, a GRPR antagonist, was significantly higher in samples with low FDG binding [16]. This suggests GRPR PET may be complimentary to FDG PET in evaluating ER-positive breast cancer. In our trial, the fact that whilst there were patients negative on FDG or [ 64 Cu]Cu-SAR-BBN PET alone, but no patient was negative on both, supports this proposal.
[ 64 Cu]Cu-SAR-BBN PET was not superior to FDG in all ER/PR-positive breast cancer patients, but was particularly useful in the lobular carcinoma subtype (ILC). Lower intensity of ILC on FDG PET compared to invasive ductal (IDC) subtypes has been reported previously [17][18][19][20]. This has been attributed to ILC's infiltrative pattern, reduced cell density, low GLUT1 expression and decreased cell proliferation rate [17]. In contrast, high GRPR expression within ILC specimens was shown to be not inferior to IDC (80.7% vs. 74.6%, respectively) [5]. This is supported by our study, as both ILC patients had greater intensity and tumour volume on [ 64 Cu]Cu-SAR-BBN compared to FDG PET.
Only small numbers of ILC patients evaluated with PET agents targeting GRPR (GRPR PET) have been previously described.
Zang et al. and Stoykow et al. reported one of one and three of three ILC patients positive on GRPR PET in the setting of primary staging [21,22]. Zhang et al. reported two ILC cases; however, this was in the setting of a PET heterodimer targeting both GRPR and αvβ3 receptors [23]. Of note, Michalski et al. reported two of two ER-positive metastatic ILC patients negative on GRPR PET. [24] Both patients reported current endocrine therapy for a duration of at least 18 months and one of two patients subsequently had loss of ER expression on examination of metastatic breast cancer cells. In our study, neither patient with ILC was on endocrine therapy at the time of imaging. Loss of ER expression following treatment has been previously described and high GRPR expression is reported in only 12% of ER-negative breast cancer specimens [5].
Two patients in this cohort had negative [ 64 Cu]Cu-SAR BBN PET findings. Whilst biopsy specimens for these patients were not assessed via GRPR staining, approximately 17% of ER-positive tumors have previously been shown to have low GRPR expression [5]. Further work in larger trials is required to better define the frequency of negative [ 64 Cu]Cu-SAR BBN PET findings in the ER/PR-positive breast cancer population to determine its value as a diagnostic and staging tool.
Targeted radionuclide therapy involves linking a radioactive atom to a targeting molecule and delivering ionizing radiation directly to cancer cells or their microenvironment [25]. Recent investigations into GRPR-targeted radionuclide therapy in the setting of metastatic prostate cancer have been promising. [ 177 Lu]Lu-RM2 was well tolerated in a dosimetry study amongst prostate cancer patients with high tumor doses with rapid clearance from normal organs [26]. [ 177 Lu]Lu-NeoBOMB1 in animal models demonstrated excellent tumor uptake in prostate cancer cells and favourable pharmacokinetics [27]. A limitation of our study was the small number of participants and single-institution design, meaning the findings from this study are hypothesis generating only. However, this is the first human trial that suggests GRPR-PET may be superior to FDG-PET in patients with ER/PR-positive metastatic breast cancer of the lobular subtype. Further work is needed to confirm these findings in larger studies.

Study Design
This prospective phase 1 study was undertaken at a single Australian Institution (St. Vincent's Hospital, Sydney, Australia). The study protocol was approved by the St. Vincent's Hospital institutional review board (HREC/2019/SVH/ETH12700), and patients provided informed and written consent. The study was registered with ANZCTR (ACTRN12619001383156).

Patient Enrolment
Women were considered eligible for the study if they had biopsy-proven metastatic ER+/PR+/HER2-breast cancer requiring staging. Women were excluded if they had an ECOG status higher than 2, or an active second malignancy. Clinical information including whether the patient was de novo or metastatic at diagnosis, history of prior treatments, and both diagnostic and metastatic histopathology including receptor status were all documented. All patients imaged underwent [ 64 Cu]Cu-SAR-BBN PET-CT, diagnostic CT of the chest, abdomen and pelvis, [ 99 Tc]Tc-MDP bone scan and [ 18 F]FDG PET-CT within a 3-week period.

[ 64 Cu]Cu-SAR-BBN Production and Quality Control
[ 64 Cu]Cu-SAR-BBN was prepared by the South Australian Health & Medical Research Institute SAHMRI (Adelaide, Australia) using a modified version of a previously described method [11]. [ 64 Cu]CuCl 2 (1000-2100 MBq, 500 µL, 0.05 M HCl; SAHMRI) was added to SAR-BBN (Auspep Pty Ltd., Melbourne, Australia) (60 µg) in 5 mL of a solution of sodium phosphate buffer (0.1 M, pH 6.5-7.0) containing sodium gentisate (4 mg). The reaction mixture was incubated for 25 min at room temperature, after which the reaction mixture was filtered through a 0.22 µM filter into a sterile product vial. The reaction was quenched by addition of 10 mL of a 5% aqueous ethanol solution containing sodium ascorbate (750 mg) via the 0.22 µM filter into the sterile product vial. [ 64 Cu]Cu-SAR-BBN was produced in an average 84% radiochemical yield with more than 95% radiochemical purity. [ 64 Cu]Cu-SAR-BBN was injected within 24 h of production in all patients. to mid-thigh) was preceded by a non-contrast low-dose CT scan using the following CT parameters: slice thickness of 2 mm, with 2 mm increment, soft-tissue reconstruction kernel, 120 kV and 50 mAs, pitch of 0.8 and a 440 matrix. Emission data were corrected for randoms, scatter and decay using the Siemens body.xml reconstruction protocol. Blood sugar levels (BSL) were less than 13 mmol/L (median BSL 5.1 mmol/L) prior to injection. Subsequently, 60 min after injection (median time 62 min), non-contrast low-dose CT scan was obtained followed by whole-body PET scan.

Bone Scan and Diagnostic CT
Bone scan was performed using a clinical protocol using a dual-head gamma camera (Discovery 670) equipped with a low-energy general-purpose collimator. Patients were injected with 800 MBq of [ 99m Tc]Tc-MDP with whole-body imaging at 3 h. Diagnostic CT scan of the chest, abdomen and pelvis with intravenous contrast was performed as per standard of care.

Image Interpretation
All PET-CT images were reported using the SIEMENS FUSION Viewer (Syngo). All studies were interpreted by credentialed nuclear medicine physicians with experience in reporting PET-CT images. Data for all scans were analysed both visually and quantitatively. Identification of positive sites of disease on PET was determined clinically and quantitatively. Quantitative analysis was undertaken to calculate total tumour burden as well as overall tumour versus individual lesion SUV max and SUV mean data (MIM Software, Cleveland) using an SUV max minimum value of 3 and volume 0.2 mL. Bone and CT scans were interpreted by the same nuclear medicine physicians using RECIST criteria.

Safety Assessment
Haematological, biochemical and coagulation assays were collected at baseline. Vital signs and safety physical examinations were conducted at baseline, and at 1, 3 and 24 h post injection. Electrocardiogram (ECG) was performed at 1 h pre and 1 h post injection.

Dosimetry
Radiation dosimetry of [ 64 Cu]Cu-SAR-BBN was determined using standard methods established by the Committee on Medical Internal Radiation Dose (MIRD) and implemented inside the 3D-RD-S software package. Time-activity curves were generated from volumesof-interest defined for source organs within the images and estimated absorbed dose to all International Commission on Radiological Protection target tissues were calculated using the MIRD S-value methodology. Once the absorbed dose for all target tissue was obtained, the effective dose was calculated as the average absorbed dose over all tissues with each tissue dose weighted by the corresponding tissue-weighting factor. Blood tracer clearance was determined radio-metrically as described below.

Blood Activity
In order to investigate radiopharmaceutical clearance, 2 × 1 mL blood samples were taken at baseline, 1 h, 3 h and 24 h post injection. These were counted on a Perkin Elmer Wizard2 gamma counter together with 5 sets of 2 × 1 mL standards of approximately 0.6, 1.0, 2.5, 5.0 and 8.0 kBq/mL. The standards were prepared within 1 h from the collection of the blood sample at 3 h post acquisition. The counts to activity conversion factor was estimated using all standards and blood activity concentration of each sample was decay corrected to the time of injection. All patient data were fitted to a bi-exponential function using R (R CORE TEAM 2021).

Metastatic Biopsy and GRPR Staining
Histopathology records for metastatic biopsy were sourced. Available tissue samples underwent GRPR receptor status staining. Tissue was formalin fixed and paraffin embedded. Mild antigen retrieval was performed using Cell Conditioning for 36 min at 95 • C. Polyclonal antibody TA316872 (OriGene) to GRPR was used at a dilution of 1/200 and incubated at 36 • C for 32 min. Ultraview was used for detection. Slides were scored by a consultant pathologist with cytoplasmic staining as positive.

Conclusions
[ 64 Cu]Cu-SAR-Bombesin appears safe and may have a diagnostic and theranostic utility in subtypes of metastatic ER+/PR+/HER2-breast cancer patients, especially the lobular subtype. Further investigation is warranted. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. The study was registered with ANZCTR (ACTRN12619001383156).