MDA435/LCC6^HER2/GFP/Luc (HER2+) and MDA435/LCC6^Vector (HER2−) cell lines, which overexpress and minimally express HER2/neu, respectively, were generous gifts from Dawn Waterhouse (BC Cancer Agency, Vancouver, BC, Canada). The HER2+ cell line constitutively expresses luciferase allowing for bioluminescent imaging in the presence of luciferin. Cells were grown in Dulbecco’s Modified Eagle Media supplemented with 10% fetal bovine serum until 80% confluent. Cells were harvested with Cell Dissociation Buffer (Invitrogen, Carlsbad, CA, USA), washed with PBS, and incubated on ice with 20 μg/mL trastuzumab. After 1 h, cells were washed with PBS and incubated with Alexa Fluor 546 goat anti-human IgG (Invitrogen) for 30 min. HER2 expression was analyzed (30,000 events collected) with the FACSAria Flow Cytometer (Becton Dickinson, Franklin Lakes, NJ, USA).

⁸⁹Zr-oxalate was produced via the ⁸⁹Y(p,n)⁸⁹Zr transmutation reaction on the CS-15 cyclotron (Cyclotron Corporation, Berkeley, CA, USA) as described previously [33,36]. The resulting ⁸⁹Zr-oxalate was produced with a specific-activity of 8.1–15.4 GBq/μmol (220–418 mCi/μmol). Trastuzumab was incubated with Df-Bz-NCS (Macrocyclics, Dallas, TX, USA) in 0.1 M NaHCO₃ buffer pH 9.0 at room temperature for 30 min in a 1:5 molar ratio. The resulting product, Df-Bz-NCS-trastuzumab, was purified via Zeba Spin Desalting Columns (Pierce Biotechnology, Rockford, IL, USA). ⁸⁹Zr was complexed with Df-Bz-NCS-trastuzumab at a ratio of 0.22 MBq/μg (6 μCi/μg) of antibody in 0.5 M HEPES buffer pH 7.0 at 37 °C for 1 h with constant agitation on an orbital mixer. Approximately 100 µg of antibody was used per reaction. ⁸⁹Zr-trastuzumab was purified with Zeba Spin Desalting Columns and radiochemical purity was determined by radio-TLC using a mobile phase of 50 mM DTPA and analytical size-exclusion chromatography (Superose 12 10/300 GL, GE Healthcare, Piscataway, NJ, USA) with 20 mM HEPES and 150 mM NaCl (pH 7.3) eluted at a flow rate of 0.75 mL/min. Millenium 32 software (Waters, Milford, MA, USA) was used to quantify chromatograms by integration.

The immunoreactive fraction of ⁸⁹Zr-trastuzumab was determined using a cell-binding assay, as previously described by Lindmo et al. [37]. HER2+ cells were suspended in microfuge tubes at increasing concentrations ranging from 0.5 to 5 × 10⁶ cells in 500 μL PBS. Approximately 37 kBq/0.25 μg of ⁸⁹Zr-trastuzumab in 50 μL were added to each tube (n = 3) and agitated on an orbital mixer for 60 minutes at 25 °C. Cells were pelleted by centrifugation, washed twice with PBS, and subsequently counted for ⁸⁹Zr activity. The specific binding was calculated as the ratio of bound radioactivity to the total amount of administered activity and was background corrected. To determine binding specificity, similar studies were performed with the addition of 100 μg of non-radiolabeled trastuzumab in HER2+ cells and also in HER2− cells.

4 × 10⁶ HER2+ or HER2− cells were injected into the mammary fat pad of athymic nude mice 6–8 weeks of age. Tumors grew for 6 weeks until they reached approximately 10 mm in greatest dimension. A metastatic model was created by injection of 2 × 10⁶ HER2+ cells into the tail vein of athymic nude mice as previously reported [38]. As the HER2+ cells constitutively express luciferase, bioluminescent imaging was performed 6 weeks after tail vein injection to evaluate the development of metastatic lesions.

Tumors were harvested after completion of MicroPET/CT imaging studies and immediately fixed in 10% formalin. After allowing for the radioactivity to decay to background levels, tumors were embedded in paraffin. Five-micron sections were prepared and blocked in Dako Protein Block (Dako, Carpinteria, CA, USA) for 30 min at room temperature. Antigen retrieval was performed in citrate-based buffer using a pressure cooker (Biocare Medical). The sections were incubated with rabbit polyclonal anti-HER2 primary antibody (1:200, Abcam, Cambridge, MA, USA) overnight at 4 °C and visualized with Alexa Fluor 555-conjugated goat anti-rabbit IgG (1:200, Invitrogen). Sections were mounted were mounted with SlowFade Gold antifade reagent with 4,6-diamidino-2-phenylindole (DAPI) (Invitrogen) and cover slipped. Images were obtained with a Soft Imaging Solutions FVII cooled monochrome digital camera, Peltier cooled to −10 °C (Olympus America, Center Valley, PA, USA).

The unpaired, two-tailed Student’s t-test was utilized for data analysis. Differences at the 95% confidence level (P < 0.05) were considered to be statistically significant.

Trastuzumab was successfully conjugated to Df-Bz-NCS and radiolabeled with ⁸⁹Zr. The radiolabeling efficiency was 78.4% ± 11.3%, the radiochemical purity was 98.7 ± 1.2%, and the specific activity was 136.9 ± 22.2 MBq/mg (3.7 ± 0.6 mCi/mg) (n = 9).

To evaluate the relative expression levels of the HER2 expression on each cell line, flow cytometry was performed. MDA435/LCC6^HER2/GFP/Luc (HER2+) cells demonstrated elevated levels of HER2 expression in comparison to MDA435/LCC6^Vector (HER2−) cells which demonstrated minimal expression (Figure 1).

The immunoreactive fraction of ⁸⁹Zr-trastuzumabwas 95.6% (Figure 2A). The uptake of ⁸⁹Zr-trastuzumab was 8.5-fold greater in the HER2+ cells compared to the HER2− cells (P < 0.01) (Figure 2B). Pre-incubation with excess unlabeled trastuzumab significantly inhibited the uptake of ⁸⁹Zr-trastuzumab in HER2+ cells (Figure 2B).

Biodistribution studies with ⁸⁹Zr-trastuzumab were performed in athymic nude mice bearing HER2+ or HER2− tumors. Uptake of ⁸⁹Zr-trastuzumab in HER2+ tumors was elevated after 24 h (26.9 ± 1.04% ID/g) and continued to remain elevated at 96 h post-injection (28.83 ± 1.33% ID/g) resulting in tumor to muscle ratios (T/M) of 8.7 ± 1.2 and 14.5 ± 2.9, respectively. In contrast, ⁸⁹Zr-trastuzumab uptake in HER2− tumors was significantly lower at 24 h (8.40 ± 0.72% ID/g) and 96 h (7.91 ± 0.96% ID/g) in comparison to HER2+ tumors (P < 0.001 at 24 and 96 h), resulting in T/M of 3.0 ± 0.4 and 3.4 ± 0.6, respectively. The circulating levels of ⁸⁹Zr-trastuzumab in the blood declined from 23.6 ± 2.4% ID/g at 24 h to 14.3 ± 2.1% ID/g at 96 h (P < 0.001). Bone uptake increased from 4.8 ± 0.5%ID/g at 24 h to 10.0 ± 1.4% ID/g at 96 h. The uptake for each selected organ at 24 and 96 h is shown in Figure 3A,B, respectively.

MicroPET/CT imaging studies demonstrated high uptake of ⁸⁹Zr-trastuzumab in HER2+ tumors when compared to HER2− tumors at 24 h and 96 h (Figure 4A,B, respectively). The uptake in HER2− tumors was minimally increased compared to background. The SUV_mean for ⁸⁹Zr-trastuzumab for HER2+ and HER2− were 2.4 ± 0.3 and 1.4 ± 0.3 at 24 h (P = 0.003), and 3.1 ± 0.4 and 1.2 ± 0.1 at 96 h (P < 0.001), respectively. The level of ⁸⁹Zr-trastuzumab in the blood significantly decreased from 24 to 96 h (SUV_mean 1.7 ± 0.2 and 1.2 ± 0.3, respectively) as evidenced by visualization of the decreased radioactivity in the blood pool in the heart and the inferior vena cava. Uptake in the liver (SUV_mean 0.82 ± 0.4) and kidney (SUV_mean 0.58 ± 0.3) were low compared to HER2+ tumors at 96 h.

For the metastatic tumor model, bioluminescent imaging demonstrated the development of metastasis in the jaw (SUV_mean 2.6) and in the mid-thorax region (SUV_mean 2.9). These lesions were readily visualized by microPET/CT demonstrating the potential use of ⁸⁹Zr-trastuzumab for characterizing HER2 expression in metastatic lesions (Figure 5).

To confirm that HER2 expression levels in tumors related to HER2 expression in vitro, immunofluorescence staining was performed with an anti-HER2 antibody. Images were acquired under the same conditions and displayed on the same scale to ensure that the relative brightness observed in images reflected differences in HER2 expression level. A high level of HER2 staining (red) was seen in the HER2+ tumors when compared with the HER2− tumors. DAPI (blue) was used as the counterstain for the nucleus (Figure 6).

In the current study, elevated ⁸⁹Zr-trastuzumab uptake was associated with the HER2 overexpression in breast cancer cells in both orthotopic mammary fat pad and metastatic tumor models. HER2+ tumors were clearly distinguished from HER2− tumors by microPET/CT imaging with this agent. Elevated ⁸⁹Zr-trastuzumab uptake in HER2+ tumors in comparison to low normal tissue uptake resulted in high resolution images with excellent contrast between the tumor and normal tissues. The circulating blood levels of ⁸⁹Zr-trastuzumab continued to decline over 4 days while the uptake in HER2+ tumors remained elevated.

Previous studies utilized a multi-step procedure with a succinylated derivative of desferrioxamine B for conjugation to trastuzumab. Vosjan et al. have reported a simple, efficient method for the radiolabeling of antibodies with ⁸⁹Zr [35]. However, this procedure has not been evaluated for ⁸⁹Zr-radiolabeling of trastuzumab. In the current study, the ⁸⁹Zr-radiolabeling of trastuzumab was completed in approximately 3 h. The relative ease of the two-step procedure used in the current study makes it feasible for incorporation into good manufacturing practice grade production and for widespread clinical use.

Both an orthotopic mammary fat pad and metastatic model was utilized in the current study. ⁸⁹Zr-trastuzumab produced with the simpler method was able to characterize HER2 expression in tumors from the orthopic and metastatic mouse models. While previous studies utilized heterotopic subcutaneous xenograft models for ovarian and breast cancer [24,39,40], the orthotopic model may be more clinically relevant by mimicking the tumor microenviroment and vasculature. Our results are in agreement with Dijkers et al., who reported an uptake value of ~33% ID/g in HER2/neu positive hindlimb tumors. In contrast, Holland et al. reported 2.5-fold higher uptake values (~73–86% ID/g) in subcutaneous shoulder tumors [39]. These differences may be explained by the different levels of HER2/neu expression in the varying cell lines used between the studies and also by differences in tumor perfusion due to the different locations of tumor growth.

The metastatic breast cancer model utilized in the current study was developed via tail vein injection of HER2+ cells. The development of tumors was initially monitored via bioluminescent imaging. This model is more of a measure of tumor invasiveness rather than metastasis as distant sites are not derived from cells migrating from a primary tumor. Despite this caveat, this model was suitable for evaluating the ability of ⁸⁹Zr-trastuzumab to detect HER2 expression in distant sites of disease. The 3.3 day half-life of ⁸⁹Zr is ideal for imaging full, intact antibodies (150 kDa) which require multiple days for equilibration as demonstrated in this study and previous studies [40]. Because ⁸⁹Zr is relatively biologically inert, ⁸⁹Zr-radiolabeled antibodies have low background uptake and improved stability in comparison to other radionuclides such as ⁶⁴Cu, which is readily transchelated by liver superoxide dismutase resulting in elevated liver uptake, or some ¹²⁴I-radiolabeled compounds which are may be dehalogenated and taken up by the thyroid. There is some residual bone uptake of ⁸⁹Zr (5–10%) which may be attributed to a slow rate of intratumoral metabolism of ⁸⁹Zr-trastuzumab leading to transmetalation of ⁸⁹Zr⁴⁺ ions which are sequestered in the bone [39]. Further studies on the metabolism of ⁸⁹Zr-trastuzumab are underway by our group and others.

⁸⁹Zr-trastuzumab has significant clinical potential for the detection and characterization of lesions for HER2 status. Confirmation of HER2 status is necessary prior to initiation of trastuzumab therapy. Immunohistochemical staining and fluorescence-in-situ hybridization techniques from biopsy specimens taken at initial diagnosis are the current standards for determining HER2 expression [13]. However, tumor heterogeneity plays a significant role in breast cancer [41] and HER2 expression and gene amplification may vary across the primary tumor. Moreover, differences in the extent of HER2 expression between the primary tumor and across metastatic lesions have been reported [17,42]. As demonstrated in this study, PET imaging with ⁸⁹Zr-trastuzumab provides a comprehensive view of the HER2 status not only within the primary tumor, but also across metastatic lesions. PET imaging with ⁸⁹Zr-trastuzumab may assist in the stratification of patients by HER2 status for decision-making about trastuzumab therapy. With the increased costs of healthcare, a method to select patients that benefit from trastuzumab therapy, while avoiding the unnecessary treatment of patients with low HER2 expressing tumors could have an impact on cancer treatment.

Although beyond the scope of this study, PET with ⁸⁹Zr-trastuzumab may improve the ability to detect lesions not previously detected on routine staging studies such as CT, MRI, or bone scan. In a feasibility study involving 14 patients with metastatic breast cancer, ⁸⁹Zr-trastuzumab PET imaging provided complementary information to CT, MRI, and bone scan [28]. Of note, ⁸⁹Zr-trastuzumab PET imaging was able to reveal previously undetected brain metastases [28]. Determination of the sensitivity and specificity of ⁸⁹Zr-trastuzumab PET imaging and correlation with HER2 status in pathological human specimens will be challenging as it is unethical to subject patients to the discomfort of additional unnecessary biopsies.

Quantitative pre-therapy imaging studies with ⁸⁹Zr-radiolabeled trastuzumab may also provide individualized pharmacokinetic and biodistribution information necessary to determine the optimal dose of trastuzumab for each patient. Scouting studies with ⁸⁹Zr-radiolabeled trastuzumab may provide information regarding tumor uptake in comparison to critical normal organs allowing for maximal tumor targeting while minimizing normal tissue toxicity. Thus, ⁸⁹Zr-trastuzumab imaging can provide this information in an efficient and safe manner, with fewer patients treated at a suboptimal dose.