Abstract
The clinical utility of a human epidermal growth factor receptor 2 (HER2)–targeting Affibody molecule for detection and characterization of HER2-positive lesions was investigated in patients with recurrent metastatic breast cancer. Methods: Three patients received 111In- or 68Ga-labeled DOTA0-ZHER2:342-pep2 (ABY-002). γ-Camera, SPECT, or PET/CT images were compared with earlier 18F-FDG PET/CT results. Results: Administration of radiolabeled ABY-002 was well tolerated. Blood kinetics of radiolabeled ABY-002 showed a first half-life of 4–14 min, second half-life of 1–4 h, and third half-life of 12–18 h. Radiolabeled ABY-002 detected 9 of 11 18F-FDG–positive metastases as early as 2–3 h after injection. Conclusion: Molecular imaging using 111In- or 68Ga-labeled ABY-002 has the potential to localize metastatic lesions in vivo, adds qualitative information not available today by conventional imaging techniques, and may allow the HER2 status to be determined for metastases not amenable to biopsy. To our knowledge, this is the first report on clinical imaging data obtained with a non–immunoglobulin-based scaffold protein.
Functional imaging with SPECT or PET is used to characterize biologic processes at the molecular level. Most of these investigations are performed using 18F-FDG PET, which has proven value for the detection of metastatic and recurrent disease and early treatment response (1,2). However, 18F-FDG PET shows increased glucose metabolism without providing information on the presence of cancer-specific cell-surface receptors and therefore does not provide guidance for receptor-targeted therapy.
The human epidermal growth factor receptor 2 (HER2) is overexpressed in several cancer types characterized by more aggressive tumors and decreased overall survival but is limited or absent in normal adult tissue (3). Determining the presence of HER2 in potentially responsive patients is increasingly important given the growing arsenal of HER2-targeted therapies (4). Diagnosis of HER2 overexpression is recommended for all newly diagnosed breast carcinomas (5,6). However, approximately 10%−20% of current HER2 assessments may be inaccurate. Thus, some eligible patients do not receive HER2-targeted therapy whereas others are exposed to unnecessary therapy (6,7).
Discordance in HER2 expression between the primary tumor and metastases complicates diagnosis of recurrent disease (8,9). Thus, a complementary in vivo diagnostic method might be beneficial for improved prediction and monitoring of therapy response—for example, a molecular imaging agent with high specificity and affinity for HER2, enabling detection and localization of all HER2-positive tumor lesions, including those not amenable to biopsy.
Affibody molecules (Affibody AB), small non–immunoglobulin-affinity proteins, are proven tracers for molecular imaging (10). Preclinical characterization of the 111In-labeled Affibody molecule DOTA0-ZHER2:342-pep2 (ABY-002) in mice showed specific tumor targeting and allowed imaging as soon as 1 h after injection (11).
Here we describe the first in-humans administration of 111In-ABY-002 and 68Ga-ABY-002 in patients with recurrent breast cancer.
MATERIALS AND METHODS
The 58-amino-acid long peptide ABY-002 was synthesized (Bachem) and labeled with 111In or 68Ga at a good-manufacturing-practice–certified radiopharmacy (Zentral Klinik Bad Berka). Three patients with recurrent metastatic breast cancer and known lesions identified by CT or 18F-FDG PET/CT received approximately 80–90 μg of ABY-002. The exploratory research was approved by the local ethics committee. Patients were imaged between November 2005 and April 2006 in accordance with the German regulations concerning administration of radiolabeled substances to humans. Details about the radiolabeling and imaging are described in the supplemental materials (available online only at http://jnm.snmjournals.org).
RESULTS
Three breast cancer patients received radiolabeled ABY-002. Two patients received both 111In-ABY-002 and 68Ga-ABY-002, and 1 patient received 68Ga-ABY-002 only. The 5 administrations (in total) of 111In-ABY-002 or 68Ga-ABY-002 were well tolerated. No acute adverse drug reactions and no clinically significant changes in blood pressure or pulse occurred during a follow-up period of 3 h after injection. Patient characteristics are summarized in Tables 1 and 2.
Patient Characteristics and Imaging Schedules
Standardized Uptake Value in Lesions on PET Scans of Patients 2 and 3
Patient 1 was diagnosed with mammary carcinoma 11 y earlier. The HER2 status of the primary tumor was unknown, and the patient had not been treated with trastuzumab before the 111In-ABY-002 SPECT investigation. A lymph node dissection in the left axilla performed 1 y earlier showed histologically 10 lymph node metastases with overexpression of HER2 (HercepTest [DakoCytomation] score, 3+) in 60% of these nodes. One year after lymph node dissection, an 18F-FDG PET/CT scan was obtained because of suspected progression of disease. One month after that scan, 111In-ABY-002 was administered and its distribution was followed by dynamic scans for 30 min (Supplemental Fig. 1). The subclavian vein and the blood pool radioactivity in the heart were visible on the 1-min image but not on later images. The left kidney and the liver could be identified after 2 min. Image intensity increased until 5 min and remained constant for the remaining 25 min. Whole-body scans, performed at 1, 2, and 4 h after injection, showed strong uptake of radioactivity in the liver and kidneys and weak uptake in the gastrointestinal tract and in the parotid and submandibular glands. At 2 and 4 h after injection, additional uptake of radioactivity was visible in the mediastinum and the lateral left thorax (Fig. 1). High-quality SPECT images were obtained 3 h after injection, enabling detection of 3 lymph node metastases in the anterior left mediastinum (near the aortic arch, paraaortally and retrosternally) and a small (diameter, 12–14 mm) lesion within the latissimus dorsi muscle at the lateral left thorax. Biopsy and HercepTest analysis of the latter confirmed a HER2 score of 3+. The same 4 lesions were detected by 18F-FDG PET/CT (Fig. 1). An additional lesion identified by 18F-FDG PET/CT and interpreted as either lymph node or adrenal metastasis was not resolved by 111In-ABY-002 imaging because of overlying radioactivity in the left kidney.
Images of patient 1 showing localization of metastases (solid arrows), kidney (dotted arrows), and liver (dashed arrows). γ-camera imaging was performed after intravenous injection of 123 MBq of 111In-ABY-002 with peptide mass dose of less than 100 μg. Whole-body scan 4 h after injection (left) shows uptake predominantly in liver and kidney, with additional uptake in parotid and submandibular glands, thorax, and gastrointestinal tract. SPECT images 3 h after injection (middle) show the localization of thoracic wall metastasis (upper coronal image and sagittal image) and mediastinal lymph node metastases (lower coronal image). Findings indicate HER2-positive metastases in mediastinal lymph nodes and thoracic wall, as was confirmed for the thoracic wall metastasis by HercepTest score of 3+. PET was performed after intravenous injection of 293 MBq of 18F-FDG. Coronal maximum-intensity-projection image 90 min after injection (right) shows localization of thoracic wall metastasis and mediastinal lymph node metastases.
After surgical resection of the thoracic wall metastasis and chemotherapy with 2 doses of vinorelbine (30 mg/m2/d at 19 and 26 d after 111In-ABY-002 imaging), none of these lesions were visible by 68Ga-ABY-002 PET/CT performed 39 d later. This finding is consistent with the significant reduction of the 2 tumor markers carcinoembryonic antigen (cutoff, 3.4) and CA 15-3 (cutoff, <25), which fell from 8.2 ng/mL and 83.3 U/mL, respectively, before therapy to 4.8 ng/mL and 43.4 U/mL, respectively, on the day of imaging.
Patient 2 was diagnosed with HER2-positive mammary carcinoma 18 mo earlier. 18F-FDG uptake was observed in 1 liver metastasis and in a mediastinal lymph node (suggestive of a metastasis). 68Ga-ABY-002 images could not provide unambiguous evidence of a HER2 receptor–positive liver metastasis because of high background uptake of the tracer in the liver. The mediastinal lymph node showed no 68Ga-ABY-002 uptake (a detailed description is provided in the supplemental material, including Supplemental Fig. 2).
Patient 3 was diagnosed 7 y earlier with HER2-positive breast cancer. Recurrent breast cancer with 5 hypermetabolic areas was identified by 18F-FDG PET/CT. Of these, only the metastasis in the iliac spine could clearly be identified on CT. The soft-tissue metastasis could be identified on CT only in connection with the 18F-FDG PET images, whereas the 3 other metastases were not identified on CT. MRI revealed changes in the sartorius muscle, but it was not possible to differentiate between changes based on inflammation and changes based on tumor infiltration. 111In-ABY-002 SPECT 20 h after injection and 68Ga-ABY-002 PET/CT approximately 2 h after injection clearly visualized the 5 lesions in the pelvic area (Fig. 2 and Supplemental Fig. 3). Comparison of the 68Ga-ABY-002 and 18F-FDG PET/CT images revealed 1 additional lesion in the right anterolateral chest wall near the axilla with increased 68Ga-ABY-002 uptake but no hypermetabolic uptake of 18F-FDG (Fig. 2A; Table 2).
Images of patient 3 obtained 85 min after intravenous injection of 278 MBq of 18F-FDG and 135 min after intravenous injection of 267 MBq of 68Ga-ABY-002 with peptide mass dose of less than 100 μg. (A) Potential metastasis (arrows) in chest wall near the axilla is seen with 68Ga-ABY-002 on transverse PET (top), CT (middle), and PET/CT (bottom) images and on coronal maximum-intensity-projection images (in between). This metastasis was not visible with 18F-FDG. (B) 18F-FDG (left column) and 68Ga-ABY-002 (right column) transverse images of identified lesions (arrows) in pelvic area.
Biopsies taken from the quadriceps muscle 4 mo later confirmed solid carcinoma metastasis. Histologically, the growth pattern was compatible with mammary carcinoma. More than 50% of the tumor cells showed a strong and mostly complete membrane staining for HER2 (mean HercepTest score, 2+; minimal value, 0; maximal value, 3+).
At the time of imaging, the patient was on trastuzumab therapy and had a serum concentration of shed HER2 extracellular domain in the range of about 40 ng/mL (40.5 ng/mL 1 mo before and 43.2 ng/mL at the time of imaging).
For all 3 patients, the blood kinetics of radiolabeled ABY-002 were assessed by following the radioactivity in blood from 2 min until 94 h (Fig. 3).
Blood kinetics in the 3 patients. Samples were collected at predetermined times after injection of radiolabeled ABY-002; radioactivity was measured with γ-counter (Isomed 2000; Nuklear Medizintechnik Dresden) and plotted over time; and values were applied for kinetic analysis using Origin Pro 7G (OriginLab) and exponential curve fit function ExpDec 1/2. (A) Blood kinetics of 111In-ABY-002 in patients 1 (administered dose, 123 MBq [▴]) and 3 (101.3 MBq [•]). Time points between 22 and 94 h after injection are not included. (B) Blood kinetics of 68Ga-ABY-002 in patients 1 (110 MBq [▴]), 2 (187 MBq [
]), and 3 (267 MBq [•]).
DISCUSSION
The presented results from 3 breast cancer patients demonstrate that the radiolabeled Affibody molecule ABY-002 has the potential to visualize HER2-expressing metastatic lesions in patients with metastatic breast cancer.
Analysis of the blood kinetics revealed rapid blood clearance of radiolabeled ABY-002 in humans with a first half-life of 4–11 min for 111In-ABY-002 and 10–14 min for 68Ga-ABY-002, in agreement with preclinical data on mice (11). The rapid kinetics enabled high-contrast SPECT or PET with radiolabeled ABY-002 to be performed in humans as soon as 2–3 h after injection. High uptake of radiolabeled ABY-002 was seen in the metastases, the kidneys, and the liver. Most of the previously identified lesions with 18F-FDG PET/CT were visualized. Four of 5 known lesions in patient 1 were visualized with 111In-ABY-002, and all 5 known lesions in patient 3 showed positive uptake of both 111In-ABY-002 and 68Ga-ABY-002. However, 2 lesions positive on 18F-FDG PET/CT scans could not be visualized with radiolabeled ABY-002: 1 lesion in patient 1 and 1 lesion in patient 2. In patient 1, the known metastasis close to the kidney was not resolved with 111In-ABY-002 because of overlying radioactivity in the kidney, and in patient 2 the liver metastasis was not resolved with 68Ga-ABY-002 because of a high background level in surrounding nonmalignant liver tissue.
In patient 3, both 18F-FDG and 68Ga-ABY-002 improved the diagnosis of metastases because CT alone could not identify all the lesions. However, imaging with 68Ga-ABY-002 added further information. The lesion in the sartorius muscle was differentially diagnosed as an inflammatory alteration or tumor infiltration by MRI. The ABY-002 images show overexpression of HER2 indicating the presence of tumor tissue. This distinction could not be made with 18F-FDG because both processes are associated with increased glucose uptake. Whereas glucose metabolism was equally elevated in both soft-tissue metastases and bone metastases, uptake of 68Ga-ABY-002 was approximately 50% lower in bone metastases than in the soft-tissue lesion. One hypothesis is that the HER2 expression level varies between metastases in different locations, possibly explaining the difference in hip-region SUVs, which ranged from 4.3 to 2.1 (50% lower). Although this observation is preliminary and in need of further validation, it could allow the use of ABY-002 to characterize different levels of HER2 expression in tumors and to detect different degrees of response to receptor-targeted therapies. Recent preclinical data have demonstrated that this use of HER2-binding Affibody molecules is possible in mice bearing tumors with different HER2 expression levels (12). The uptake of ABY-002 in a single potential lesion without significantly increased glucose metabolism is an interesting finding. Because biopsy data to confirm this potential receptor-versus-metabolism mismatch are not available, further clinical studies are warranted to validate this phenomenon.
Immunohistochemistry analysis (HercepTest) of 2 metastases amenable to surgery or biopsy confirmed that the lesions detected with radiolabeled ABY-002 were HER2-positive. The thoracic wall metastasis from patient 1 had a HercepTest score of 3+. Biopsies taken from the quadriceps muscle metastasis from patient 3 revealed heterogeneous HER2 expression with scores ranging from 0 to 3+ and a mean score of 2+. Thus, the biopsy data provide additional support for the hypothesis that the difference in the standardized uptake values in the hip region could be due to heterogeneous HER2 expression pattern in different lesions.
Adequate imaging results were obtained for patient 3 even though this patient was on trastuzumab therapy, confirming the extensive preclinical data that HER2-binding Affibody molecules—including ABY-002—and trastuzumab target separate epitopes on the extracellular domain of HER2 (11,13). Thus, potential interference from competitive target binding is not expected. A serum concentration of shed HER2 extracellular domain in the range of about 40 ng/mL did not preclude tumor imaging.
One potential limitation is the high level of liver and kidney uptake seen with 111In-ABY-002 and 68Ga-ABY-002. Whereas high kidney uptake of radiolabeled ABY-002 was known from preclinical studies and is expected for a small protein with predominantly renal clearance, the liver uptake seen in patients was never observed in normal or tumor-bearing mice (10,11). One possible explanation could be that HER2 expression in human liver leads to specific uptake although data on HER2 expression in human liver are inconsistent (14,15).
The high detection rate of known lesions and the short time until the image acquisition after radiolabeled ABY-002 administration is in contrast to results from the few clinical studies performed with 111In-diethylenetriaminepentaacetic acid (DTPA)-trastuzumab, 68Ga-F(ab′)2-trastuzumab, and 89Zr-trastuzumab. 111In-DTPA-trastuzumab was used in 3 studies aiming to identify patients at risk for trastuzumab-related cardiotoxicity (16–18). Only 45% of the known metastases were detected using 111In-DTPA-trastuzumab (17). 68Ga-F(ab′)2-trastuzumab showed no uptake in 3 patients and showed measurable uptake in at least 1 lesion in 3 other patients with HER2-positive metastatic breast cancer, but multiple other lesions were not visualized (19). Recently, clinical application of 89Zr-trastuzumab allowed detection of known and unknown lesions, with an optimal image-acquisition time point of 4–5 d after administration (20).
CONCLUSION
Imaging with ABY-002 seems to be well adapted for the clinical practice where tracer administration and image acquisition on the same day are preferred over modalities that extend over 2 or more days, such as is needed with antibodies or fragments thereof. Further clinical studies are warranted to assess the optimal time point, peptide and radioactivity doses for image acquisition, and sensitivity and specificity.
Acknowledgments
We thank Lars Abrahmsén for critically reading the manuscript.
Footnotes
-
COPYRIGHT © 2010 by the Society of Nuclear Medicine, Inc.
References
- Received for publication November 26, 2009.
- Accepted for publication February 17, 2010.
Jump to section
Related Articles
Cited By...
- Radioiodinated Small-Molecule Tyrosine Kinase Inhibitor for HER2-Selective SPECT Imaging
- In Vivo Imaging of the Programmed Death Ligand 1 by 18F PET
- Preclinical and clinical applications of specific molecular imaging for HER2-positive breast cancer
- Immuno-PET Using Anticarcinoembryonic Antigen Bispecific Antibody and 68Ga-Labeled Peptide in Metastatic Medullary Thyroid Carcinoma: Clinical Optimization of the Pretargeting Parameters in a First-in-Human Trial
- Targeting the Human Epidermal Growth Factor Receptors with Immuno-PET: Imaging Biomarkers from Bench to Bedside
- Imaging Diagnostic and Therapeutic Targets: Human Epidermal Growth Factor Receptor 2
- Phase I Study of 68Ga-HER2-Nanobody for PET/CT Assessment of HER2 Expression in Breast Carcinoma
- Targeting diverse protein-protein interaction interfaces with {alpha}/{beta}-peptides derived from the Z-domain scaffold
- An engineered affibody molecule with pH-dependent binding to FcRn mediates extended circulatory half-life of a fusion protein
- 188Re-ZHER2:V2, a Promising Affibody-Based Targeting Agent Against HER2-Expressing Tumors: Preclinical Assessment
- First-in-Human Molecular Imaging of HER2 Expression in Breast Cancer Metastases Using the 111In-ABY-025 Affibody Molecule
- Novel Ubiquitin-derived High Affinity Binding Proteins with Tumor Targeting Properties
- Positron Emission Tomography Imaging with 18F-Labeled ZHER2:2891 Affibody for Detection of HER2 Expression and Pharmacodynamic Response to HER2-Modulating Therapies
- Imaging of Platelet-Derived Growth Factor Receptor {beta} Expression in Glioblastoma Xenografts Using Affibody Molecule 111In-DOTA-Z09591
- Three Methods for 18F Labeling of the HER2-Binding Affibody Molecule ZHER2:2891 Including Preclinical Assessment
- Site-Specific Radiometal Labeling and Improved Biodistribution Using ABY-027, A Novel HER2-Targeting Affibody Molecule-Albumin-Binding Domain Fusion Protein
- Synthesis, Preclinical Validation, Dosimetry, and Toxicity of 68Ga-NOTA-Anti-HER2 Nanobodies for iPET Imaging of HER2 Receptor Expression in Cancer
- HER2-Positive Tumors Imaged Within 1 Hour Using a Site-Specifically 11C-Labeled Sel-Tagged Affibody Molecule
- PET of EGFR Expression with an 18F-Labeled Affibody Molecule
- Tumor Targeting Using Affibody Molecules: Interplay of Affinity, Target Expression Level, and Binding Site Composition
- Optical Imaging with Her2-Targeted Affibody Molecules Can Monitor Hsp90 Treatment Response in a Breast Cancer Xenograft Mouse Model
- Imaging of Insulinlike Growth Factor Type 1 Receptor in Prostate Cancer Xenografts Using the Affibody Molecule 111In-DOTA-ZIGF1R:4551
- Imaging of Human Epidermal Growth Factor Receptor Type 2 Expression with 18F-Labeled Affibody Molecule ZHER2:2395 in a Mouse Model for Ovarian Cancer
- Immuno-PET of Cancer: A Revival of Antibody Imaging
- Optimization of Hapten-Peptide Labeling for Pretargeted ImmunoPET of Bispecific Antibody Using Generator-Produced 68Ga
- Molecular Design and Optimization of 99mTc-Labeled Recombinant Affibody Molecules Improves Their Biodistribution and Imaging Properties
- Structural basis for high-affinity HER2 receptor binding by an engineered protein
- Targeting of HER2-Expressing Tumors Using 111In-ABY-025, a Second-Generation Affibody Molecule with a Fundamentally Reengineered Scaffold