Xiaoxia Xu, Teli Liu, Fei Liu, Xiaoyi Guo, Lei Xia, Qing Xie, Nan Li, Haifeng Huang,2,Xianteng Yang,2, Yangchun Xin, Hua Zhu, Zhi Yang

1Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Nuclear Medicine, Peking University Cancer Hospital & Institute, Beijing 100142, China;

2Department of Orthopaedics, Guizhou Provincial People’s Hospital, Guiyang 550002, China;

3Katzin Diagnostic & Research PET/MR Center, Nemours/Alfred I. DuPont Hospital for Children, Wilmington, DE 19803, USA Correspondence to: Hua Zhu, PhD. Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Nuclear Medicine, Peking University Cancer Hospital & Institute, Beijing 100142, China. Email: zhuhuananjing@163.com; Zhi Yang, PhD. Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Nuclear Medicine, Peking University Cancer Hospital & Institute, Beijing 100142, China. Email: pekyz@163.com.

Abstract Objective: Epidermal growth factor receptor (EGFR) is overexpressed in a wide variety of solid tumors, serving as a well-characterized target for cancer imaging or therapy. In this study, we aimed to design and synthesize a radiotracer, 64Cu-NOTA-C225, targeting EGFR for tumor positron emission tomography (PET) imaging.Methods: Cetuximab (C225) was conjugated to a bifunctional chelator, p-isothiocyanatobenzyl-1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA), and further radiolabeled with copper-64 for PET imaging. 64Cu-NOTA-IgG and Cy5.5-C225 were also synthesized as control probes. A431 and A549 mouse models were established for micro-PET and/or near-infrared fluorescence (NIRF) imaging.Results: 64Cu-NOTA-C225 exhibited stability in vivo and in vitro up to 24 h and 50 h post-injection,respectively. A431 tumors with average standard uptake values (SUVs) of 5.61±0.69, 6.68±1.14, 7.80±1.51 at 6, 18 and 36 h post-injection, respectively, which were significantly higher than that of moderate EGFR expressing tumors (A549), with SUVs of 0.89±0.16, 4.70±0.81, 2.01±0.50 at 6, 18 and 36 h post-injection, respectively. The expression levels of A431 and A549 were confirmed by western blotting. Additionally, the tracer uptake in A431 tumors can be blocked by unlabeled cetuximab, suggesting that tracer uptake by tumors was receptor-mediated.Furthermore, NIRF imaging using Cy5.5-C225 showed that the fluorescence intensity in tumors increased with time, with a maximal intensity of 8.17E+10 (p/s/cm2/sr)/(μW/cm2) at 48 h post-injection, which is consistent with the paradigm from micro-PET imaging in A431 tumor-bearing mice.Conclusions: The 64Cu-NOTA-C225 PET imaging may be able to specifically and sensitively differentiate tumor models with different EGFR expression levels. It offers potentials as a PET radiotracer for imaging of tracer EGFR-positive tumors.

Keywords: Epidermal growth factor receptor; immuno-PET imaging; copper-64; monoclonal antibody

Introduction

Epidermal growth factor receptor (EGFR) is expressed in a wide variety of human cancers, including breast cancer,lung cancer, glioblastoma, head and neck cancer, ovarian carcinoma and colorectal cancer (1). Evidence suggests that EGFR is involved in the pathogenesis and progression of different carcinoma types (2). Of several therapeutic approaches targeting EGFR, inhibition of EGFR with mAbs was the first used in clinical settings (3).

Cetuximab (C225, Erbitux?), a recombinant human/mouse chimeric mAb, binds with the extracellular domain of human EGFR with high affinity and specificity(4-9). It has been approved by the Food and Drug Administration (FDA) for the treatment of patients with EGFR-expressing carcinoma, such as colorectal carcinoma and non-small-cell lung cancer; however, only a subset of patients responds to this treatment (4). The efficacy of cetuximab treatment depends on the level of EGFR expression in the specific tumors of patients. Previously, a clinical study with 1,125 patients suggested that the survival benefit associated with chemotherapy that included cetuximab for non-small-cell lung cancer expressing high levels of EGFR was not limited by EGFR mutation status(10). The expression level of EGFR in most solid tumors was usually determined by immunohistochemistry,fluorescence in situ hybridization and gene copy number ex vivo; these techniques are invasive, do not account for spatiotemporal heterogeneity, and most notably do not differentiate EGFR expression between primary tumors and metastases. Thus, the current approach to determine the level of EGFR expression within tumors is not sufficient to guide clinical treatment (11,12). More quantitative and reliable methods to measure the expression level of EGFR are required (13). Noninvasive molecular imaging technique to measure tumor EGFR overexpression in vivo and to provide more biological information had shown great potential to guide the EGFR-targeting cancer treatment plans (14,15). In this study, we synthesized64Cu-NOTA-C225; the NOTA chelator was used to connect cetuximab and copper (II) to simplify conjugate production and increase the stability of the copper (II) complex. The stability was evaluated in vitro and in vivo, and micropositron emission tomography (micro-PET) imaging was performed in tumor models with different EGFR expression levels.

Materials and methods

General equipment, reagents, cell lines and animal preparation are shown in Supplementary Table S1. All animal studies were performed according to a protocol approved by the Animal Care and Use Committee of Peking University Cancer Hospital.

Synthesis, radiolabeling and quality control of 64Cu-NOTA-C225/IgG and Cy5.5-C225

C225/IgG was conjugated to NOTA at a molar ratio of 1:5 at pH 8.4 for 2 h, as previously reported (16). The final products, NOTA-C225 and NOTA-IgG, were purified by PD-10 size-exclusion columns using metal-free phosphate buffer solution (PBS) as the mobile phase to remove unbound NOTA. The molecular weights of the conjugated antibodies (C225, NOTA-C225, IgG and NOTA-IgG)and the number of NOTA chelator molecules per antibody were measured and calculated using matrix-assisted laser desorption ionization time of flight mass spectrometry(MALDI-TOF-MS). For64Cu radiolabeling, C225/IgG was incubated with64CuCl2in 400 μL of sodium acetate buffer (0.1 mol/L, pH 5.5) at room temperature for 20 min(Figure 1A). Cy5.5-C225 was synthesized according to reported methods with some modifications (17)(Supplementary materials).

In vitro and in vivo stability study

The in vitro and in vivo stability of64Cu-NOTA-C225 was measured according to a published method (18,19)(Supplementary materials).

Biodistribution and pharmacokinetics study of 64Cu-NOTA-C225 in normal Kunming mice

Two experimental groups of normal Kunming mice were utilized (n=4 for each group).64Cu-NOTA-C225 (7.4 MBq) was administered via tail vein injection. At 18 and 36 h post-injection, retroorbital blood samples were collected before the normal Kunming mice were sacrificed. All tissues of interest (heart, liver, spleen, lung, kidney,stomach, intestine, bone, muscle and brain) were weighed,and their radioactivity was measured using a γ-counter.The percentage of injected dose per gram of tissue(%ID/g) was calculated by comparison of the radioactivity for each sample with that from a 1:100 diluted standard dose (20). The pharmacokinetics study of64Cu-NOTAC225 was performed according to a published procedure(21). Normal Kunming mice (n=4) were used to measure time-radioactivity curves. At various predetermined intervals (1, 3, 5, 10 and 15 min and 0.5, 1, 1.5, 2, 4, 12, 24 and 36 h), 10 μL of blood was drawn from the tail vein, and the radioactivity of each blood sample was measured. The pharmacokinetic parameters were determined by a threecompartment model using GraphPad Prism software(Version 6; GraphPad Software, San Diego, CA, USA).

Figure 1 Synthesis of 64Cu-NOTA-C225. (A) Synthetic scheme of 64Cu-NOTA-C225; (B) Instant thin-layer chromatography (ITLC) of Cu2+ and 64Cu-NOTA-C225; (C) In vitro stability of 64Cu-NOTA-C225 in 0.01 mol/L phosphate buffer solution (PBS) and 5% human serum albumin (HSA); (D) In vivo stability of 64Cu-NOTA-C225 in blood and liver.

Micro-PET and near-infrared fluorescence (NIRF)imaging of tumor models with different EGFR expressions

For micro-PET imaging studies, no obesity diabetes and severe combined immune deficiency (NOD-SCID) mice and/or BALB/c nude mice bearing subcutaneously inoculated A431 or A549 xenografts were administered 18.5±0.5 MBq of64Cu-NOTA-C225/64Cu-NOTA-IgG via tail vein injection (n=4 for each group). Animals were anesthetized with 2.5% isoflurane in 0.5 L/min oxygen flow and maintained with 1.5% isoflurane during the PET scan. Micro-PET imaging was performed on a SuperArgus PET scanner (Sedecal, Spain). Static images were collected at 6, 18 and 36 h post-injection. The images were reconstructed using a three-dimensional ordered subsets expectation maximum algorithm without attenuation correction. The blocking study was performed on A431 tumor-bearing mice (n=3) by co-injection of the radiotracer with cetuximab at the dose of 25 mg/kg. Semi-quantitative analysis of the radioactivity uptake for tumor, heart, liver and muscle, was performed by the average standardized uptake values (SUVs). The SUV was defined as the ratio between the counts per second per pixel in a region of interest (ROI) encompassing the entire organ and the total counts per second per pixel in the mouse (22). The tumorto-heart (T/H), tumor-to-liver (T/L) and tumor-to-muscle(T/M) ratios were calculated according to the SUV values.

For NRIF imaging, a Xenogen IVISTM200 small animal imaging system with a Cy5.5 filter set (excitation 615 to 655 nm; emission 695 to 770 nm) was used. A431 tumor mice were anesthetized with 2.5% isoflurane and an oxygen flow of 0.5 L/min and maintained in the anesthetized state with 1.0% isoflurane during imaging. Each mouse (n=3)was injected with 1.0 nmol Cy5.5-C225 and was imaged at various time points (2, 24, 48, 72, 96, 160 and 180 h postinjection). All of the fluorescence images were acquired using a 5 s exposure and were normalized by dividing the fluorescence images by the reference illumination images.ROIs were drawn on tumor sites, and the average radiant efficiency [presented as thein the unit of(p/s/cm2/sr)/(μW/cm2)] within the ROI was used for subsequent quantitative analysis.

Western blotting and autoradiograph

The EGFR expression in A431 and A549 tumor cells was measured using western blotting analysis. Western blotting was performed as described previously (21).

Statistical analysis

Results

Conjugation of cetuximab

The conjugates, NOTA-C225 and NOTA-IgG, were characterized by MALDI-TOF-MS, and the average number of NOTA chelators attached to each antibody molecule was calculated by the mass difference between the conjugate and starting antibody. The results for NOTAC225 showed two major peaks centered at 152,663 Da(unconjugated mAb) and 153,141 Da (conjugated mAb).The mass difference equates to an average of 1.0 NOTA group per molecule of cetuximab, and no additional peaks were observed. MALDI-TOF-MS results for NOTA-IgG showed two major peaks centered at 148,066 Da(unconjugated mAb) and 149,651 Da (conjugated mAb).The mass difference equates to an average of 3.3 NOTA groups per molecule of IgG.

Radiosynthesis and quality control of 64Cu-NOTAC225/IgG

Both64Cu-NOTA-C225 and64Cu-NOTA-IgG were obtained with high radiochemical yields (＞80%). After purification with a PD-10 column, the radiochemical purities of radiotracers were above 99%. No free64Cu was found64Cu-NOTA-C225 (Figure 1B) and64Cu-NOTAIgG. The specific activity of64Cu-NOTA-C225/IgG was about 2.96 GBq/μmol.

In vitro and in vivo stability study

Both64Cu-NOTA-C225 and64Cu-NOTA-IgG were stable in vitro [in both PBS and human serum albumin(HSA)] at 37 °C for more than 50 h (Figure 1C).64Cu-NOTA-C225 was also stable in vivo (Figure 1D), as determined in the liver and blood of Kunming mice at 1, 2,12 and 24 h post-injection. These results support that both64Cu-NOTA-C225 and64Cu-NOTA-IgG exhibit sufficient stability for preclinical study.

Biodistribution of 64Cu-NOTA-C225 in normal mice

The biodistribution data showed that the liver exhibited relatively moderate uptake of 5.36±0.82 %ID/g at 12 h post-injection and 3.66±0.35 %ID/g at 24 h post-injection.The radioactivity of64Cu-NOTA-C225 in the liver at 12 h post-injection was significantly lower than that of previously reported64Cu-DOTA-C225 and64Cu-PCTAC225 (the %ID/g was greater than 15 and 12, respectively)(23,24).64Cu-NOTA-C225 showed moderate blood clearance in normal mice, 6.99±0.96 %ID/g at 18 h postinjection and 5.42±0.36 %ID/g at 36 h post-injection(Figure 2A), which is a common feature of intact mAbs (14).

Pharmacokinetics of 64Cu-NOTA-C225 in normal mice

The blood metabolism of64Cu-NOTA-C225 follows a three-phase decay model as described below (Figure 2B):

Figure 2 Biodistribution and pharmacokinetics study of 64Cu-NOTA-C225 in normal Kunming mice. (A) Biodistribution of 64Cu-NOTAC225 in normal Kunming mice at 18 h and 36 h post-injection (n=4); (B) Blood time-radioactivity curve of 64Cu-NOTA-C225 (n=4).

Y0 is the Y value when X (time) is zero. Plateau is the Y value at infinite time, expressed in the same units as Y.KFast, KMediumand KSloware the rate constants. Half-lives(Fast, Medium and Slow) are computed as the ratio between ln2 and the corresponding rate constants.PercentFastis the percentage of the area (from Y0 to Plateau) covered by the fastest of the three components.PercentSlowis the percentage of the area (from Y0 to Plateau) covered by the slowest of the three components.The following are equations for64Cu-NOTA-C225 in normal Kunming mice:

The half-lives (Fast, Medium and Slow) were 0.002809,0.2990 and 6.885 h, respectively.

Micro-PET imaging and autoradiography

The uptake of64Cu-NOTA-C225 by tumors was significantly higher in the human epidermoid A431 mouse model (high tumor EGFR overexpression) than in the adenocarcinoma A549 mouse model (moderate tumor EGFR overexpression) at all time points (Figure 3). The64Cu-NOTA-C225 uptake in A431 tumors increased steadily over time from 6 to 36 h post-injection (Figure 3A),while that in A549 tumors showed moderate uptake at 18 h post-injection only (Figure 3B). The expression levels of EGFR in A431 and A549 tumor tissues were also determined ex vivo by western blotting (Figure 3C) using βactin protein as the internal standard.

Semi-quantitative analyses of micro-PET tumor images showed that the SUVs of A431 tumors were 5.61±0.69,6.68±1.14 and 7.80±1.51 at 6, 18 and 36 h, respectively,while those of A549 tumors were 0.89±0.16, 4.70±0.81 and 2.01±0.50, respectively (Figure 4A), with the former always higher than the latter (P＜0.01). The SUVs of64Cu-NOTA-C225 in other major organs in A431 and A549 mice are shown in Figure 4B,C. The T/H, T/M and T/L ratios of A431 mice were 3.89±3.82, 9.64±4.26 and 3.05±1.43 at 36 h post-injection, respectively, while those of A549 mice were 0.73±0.29, 0.43±0.35 and 4.34±1.94,respectively (Figure 4D). The radioactivity uptake values for the non-specific radioligand64Cu-NOTA-IgG in A431 tumors were significantly lower than those for the specific radioligand64Cu-NOTA-C225 at each time point (Figure 5A,B), supporting the specific binding of64Cu-NOTAC225 within tumors. Moreover, when excess cetuximab was co-injected with the specific radioligand64Cu-NOTAC225, the tumor radioactivity uptake was significantly reduced in A431 tumors, with overall PET images matching those of the non-specific radioligand64Cu-NOTA-IgG (Figure 5A,B). The specific binding of64Cu-NOTA-C225 in tumors observed by PET images was further confirmed by autoradiography ex vivo (Figure 5C).

NIRF imaging of Cy5.5-C225 in A431 xenotransplantation mice

In vivo fluorescence imaging over extended time periods showed that Cy5.5-C225 gradually accumulated in A431 tumors transplanted into NOD-SCID mice at 0, 24, 48 and 72 h post-injection (Figure 6A). When the fluorescence intensities in tumors were plotted against time postinjection, a bell-shaped curve with maximal intensity[8.17E+10(p/s/cm2/sr)/(μW/cm2)] at 48 h post-injection appears (Figure 6B). When both time-activity curves of PET imaging and fluorescence imaging were combined,the general patterns match each other, each confirming the conclusion of the other (Figure 6C).

Figure 3 Micro-positron emission tomography (Micro-PET) of tumor models with different epidermal growth factor receptor(EGFR) expressions. (A) Micro-PET images of 64Cu-NOTAC225 in A431 xenograft no obesity diabetes and severe combined immune deficiency (NOD-SCID) mice at 6, 18 and 36 h after injection; (B) Micro-PET images of 64Cu-NOTA-C225 in A549 xenograft NOD-SCID mice at 6, 18 and 36 h after injection.White arrows indicate tumor; (C) Relative EGFR expressions in A431 (left column) and A549 cancer cells (right column), and representative EGFR expressions were presented as the ratios of EGFR vs. β-actin expressions by western blotting analysis.

Discussion

The development of EGFR-targeted molecular probes is greatly needed to measure the level of EGFR expression and the heterogeneity of different lesions in vivo. The retrospective analysis of a large randomized trial study showed that non-small-cell lung carcinoma patients with high tumor EGFR expression would most likely benefit from the addition of cetuximab to platinum-based first-line chemotherapy, independently of tumor histology (25).Cetuximab antibody blocks the natural ligand from binding to EGFR, leading to a decrease in receptor dimerization,autophosphorylation and activation of signaling pathways(26). EGFR protein expression has the potential to serve as a biomarker to predict favorable outcomes from cetuximab administration in multiple cancer types. However, it is rather disappointing that the methods for the detection of EGFR positivity as well as other potential biomarkers are by no means standardized, which necessitates substantial improvements (27). The mainstream method to detect EGFR expression in tumors is to analyze the EGFR RNA in tumor tissue samples collected from surgery or puncture.However, the spatiotemporal heterogeneity of EGFR in tumor tissue cannot be analyzed. This limits the application of the in vitro measurement of EGFR (28). In addition,Laskin et al. (29) found evidence to demonstrate that the quantification of EGFR by immunohistochemistry is also unreliable.

With this study, we are the first to report64Cu-NOTAC225 as a micro-PET imaging radioligand in human epidermoid (A431) and adenocarcinoma (A549) mouse models. The critical feature of our64Cu-NOTA-C225 radioligand is in vivo and in vitro stability. Previous studies indicated that64Cu labeled DOTA-cetuximab showed promise for PET imaging of EGFR-positive tumors;however, in vivo instability of this compound was observed(30). It has been established that DOTA is not the ideal chelator for64Cu. Bifunctional chelators bearing crossbridged macrocycles demonstrate much greater stability with64Cu, such as 2,2’-[1,4,8,11-tetraazabicyclo(6.6.2) hexadecane-4,11-diyl] diacetic acid, and can reduce copper (II) dissociation and non-specific copper (II) uptake(31). Unfortunately, the radiolabeling requires harsher labeling conditions (95 °C for 60 min and pH greater than 8) (32,33), which are incompatible for proteins.

We selected NOTA as the linker to synthesize64Cu-NOTA-C225 as our radioligand.64Cu-NOTA-C225 is convenient for radiolabeling and stable both in vitro and in vivo. The complete radiolabeling required only 20 min at room temperature, and the64Cu-NOTA-C225 product is stable in vitro and in vivo for up to 50 h/24 h; this stability is sufficient for application in clinical studies. The radioactivity uptake of64Cu-NOTA-C225 in the liver was significantly lower than that of previously reported64Cu-DOTA-C225 and64Cu-PCTA-C225 (%ID/g was above 15 and 12, respectively) (23,24), which supports the observation of decreased dissociation of copper (II) from the radioligand in vivo.

The anti-EGFR antibodies radiolabeled with technetium-99m, indium-111, lutetium-177 and zirconium-89 have been studied previously (13,34,35).64Cu is a beneficial and favorable radionuclide among the other copper radionuclides, as it emits low energy positrons with the endpoint energy of 653 keV and average energy of 278 keV. This radionuclide is also characterized by a medium energy beta particle, gamma radiation and an appropriate half-life (t1/2=12.7 h), allowing the combination of PET scan with diagnosis and radiotherapy (36). Our result showed that micro-PET imaging using64Cu-NOTA-C225 can measure EGFR expression in tumor models. For the EGFR-overexpressing tumor, A431, the PET signal gradually increases within the tumor at 6, 18 and 36 h post-injection. However, for the moderate EGFR-expressing tumor, A549, tracer uptake by the tumor was lower at all time points. The amount of EGFR expression was measured ex vivo by western blotting,confirming that the EGFR expression in A431 is much greater than that in A549. A similar paradigm was observed in the Cy5.5-C225 fluorescence imaging probe study.

Figure 4 Semi-quantitative analyses of micro-positron emission tomography (micro-PET) images. (A) Comparison of the tumor radioactivity uptake of 64Cu-NOTA-C225 standard uptake value (SUV) between A431 and A549 tumors; (B) SUV of 64Cu-NOTA-C225 in A431 tumors and major organs; (C) SUV of 64Cu-NOTA-C225 in A549 tumors and major organs; (D) Tumor-to-heart (T/H), tumor-toliver (T/L) and tumor-to-muscle (T/M) of 64Cu-NOTA-C225 in A431 and A549 mice. All data were represented as (n=4).

Conclusions

We have synthesized a new radioligand,64Cu-NOTAC225, with high chemical and radiochemical yield and purity. The radioligand displays sufficient in vitro/vivo stability. PET imaging with64Cu-NOTA-C225 showed that the radioligand can specifically and sensitively bind with EGFR receptors in EGFR-positive tumors. This study demonstrates the potential of64Cu-NOTA-C225 for further clinical application in patients with EGFRoverexpressing tumors and proper selection of patients for cetuximab therapy, monitoring the effects of therapies,following disease progression, and eventually enabling individualized treatment.

Acknowledgements

This study was supported by Natural Science Foundation of Beijing Municipality (No. 7162041), Beijing Nova program (No. Z171100001117020), Beijing Municipal Commission of Health and Family Planning (215 backbone program No. 2015-3-072) Beijing talent project (No.2017000021223ZK33).

Footnote

Conflicts of Interest: The authors have no conflicts of interest to declare.

Figure 5 Micro-positron emission tomography (Micro-PET) imaging and autoradiography. (A) Micro-PET images of 64Cu-NOTA-C225,64Cu-NOTA-C225 + Block and 64Cu-NOTA-IgG in A431 xenograft nude mice at 6, 18 and 36 h post injection; (B) Standard uptake values(SUVs) of specific radioligand 64Cu-NOTA-C225 + Block and those of non-specific radioligand 64Cu-NOTA-IgG; (C) Autoradiography images of tumor sections (25 μm) obtained from A431 tumor 60 h after injection with and without blocking.

Figure 6 Near-infrared fluorescence (NIRF) imaging of Cy5.5-C225 and semi-quantitative analyses. (A) NIRF of Cy5.5-C225 in A431 xenograft mice; (B) Radiant efficiency curve of tumors changed over time in optical imaging; (C) Semi-quantitative analysis of micropositron emission tomography (micro-PET) tumor uptake (blue line) and optical radiant efficiency in A431 tumors (red line).