Noninvasive Evaluation of EGFR Expression of Digestive Tumors Using 99mTc-MAG3-Cet-F(ab′)2-Based SPECT/CT Imaging

Purpose This study is aimed at investigating the feasibility of cetuximab (Cet) F(ab′)2 fragment- (Cet-F(ab′)2-) based single photon emission tomography/computed tomography (SPECT/CT) for assessing the epidermal growth factor receptor (EGFR) expression in digestive tumor mouse models. Methods Cet-F(ab′)2 was synthesized using immunoglobulin G-degrading enzyme of Streptococcus pyogenes (IdeS) protease and purified with protein A beads. The product and its in vitro stability in normal saline and 1% bovine serum albumin were analyzed with sodium dodecyl sulfate–polyacrylamide gel electrophoresis. The EGFR expression in the human colon tumor cell line HT29 and the human stomach tumor cell line MGC803 were verified using western blotting and immunocytochemistry. Cet-F(ab′)2 was conjugated with 5(6)-carboxytetramethylrhodamine succinimidyl ester to demonstrate its binding ability to the MGC803 and HT29 cells. Cet-F(ab′)2 was conjugated with NHS-MAG3 for 99mTc radiolabeling. The best imaging time was determined using a biodistribution assay at 1, 4, 16, and 24 h after injection of the 99mTc-MAG3-Cet-F(ab′)2 tracer. Furthermore, 99mTc-MAG3-Cet-F(ab′)2 SPECT/CT was performed on MGC803 and HT29 tumor-bearing nude mice. Results HT29 cells had low EGFR expression while MGC803 cell exhibited the high EGFR expression. Cet-F(ab′)2 and intact cetuximab showed similar high binding ability to MGC803 cells but not to HT29 cells. Cet-F(ab′)2 and 99mTc-MAG3-Cet-F(ab′)2 showed excellent in vitro stability. The biodistribution assay showed that the target to nontarget ratio was the highest at 16 h (17.29 ± 5.72, n = 4) after tracer injection. The 99mTc-MAG3-Cet-F(ab′)2-based SPECT/CT imaging revealed rapid and sustained tracer uptake in MGC803 tumors rather than in HT29 tumors with high image contrast, which was consistent with the results in vitro. Conclusion SPECT/CT imaging using 99mTc-MAG3-Cet-F(ab′)2 enables the evaluation of the EGFR expression in murine EGFR-positive tumors, indicating the potential utility for noninvasive evaluation of the EGFR expression in tumors.


Introduction
Gastric cancer and colon cancer are common digestive system tumors, with incidence rates ranked fifth [1] and third [2] among that of all tumors, respectively. Gastric cancer and colon cancer are both among the top five causes of tumor-related death. Many patients are already at the advanced stages when diagnosed and are therefore usually unsuitable for radical surgery. Advances in the assessment of the target point in targeted therapy have contributed to increased treatment effectiveness and improved survival of patients with cancer over the past decades. Epidermal growth factor receptor (EGFR), the receptor for EGF cell proliferation and signal transduction, is related to the inhibition of tumor cell proliferation, angiogenesis, tumor invasion, metastasis, and apoptosis [3,4]. Monoclonal antibody (mAb) treatment targeting EGFR has demonstrated high therapeutic efficacy in the clinic [5][6][7]. However, treatment response is always achieved only in patients with cancer with high EGFR expression.
Traditional computed tomography (CT) and magnetic resonance imaging (MRI) in tumor diagnosis and staging mainly reveal anatomic changes. The expression of specific molecules in a tumor is difficult to demonstrate. Single photon emission computed tomography/computed tomography (SPECT/CT) based on immunological probes (immuno-SPECT/CT) is a common noninvasive molecular imaging method that utilizes a radiolabeled antibody to visualize a specific marker [8][9][10]. The therapeutic effect of EGFR targeted treatment depends highly on the EGFR expression of the tumor. Although pathological results are the gold standard, the means of obtaining samples are typically invasive and inconvenient. Cetuximab (Cet), a US Food and Drug Administration-(FDA-) approved mAb, is widely used for treating digestive tumors with high EGFR expression. However, noninvasive methods that can efficiently classify patients with high-EGFR expression tumors for intensive EGFR targeted treatment are rare.
The intact antibody commonly has a molecular weight of about 150 kDa, which makes its metabolism in the blood very slow (its biological half-life (T 1/2 ) is always >3 days) [11]. Therefore, it presents significant radiation problems for nuclear medicine immunoimaging. Enzymatic digestion can produce F(ab ′ ) 2 fragments (about 100 kDa) from an intact antibody to reduce the molecular weight but nevertheless retain the antigen-binding site and immunological binding activity of the intact antibody [9,12]. In the present study, we fabricated 99m Tc-labeled cetuximab F(ab ′ ) 2 fragments (Cet-F(ab ′ ) 2 ) as a probe for biodistribution and SPECT/CT imaging assessment of the EGFR expression in murine models of digestive tumors.

Western
Blotting. The MGC803 and HT29 cells were seeded in 6-well plates and grown to 70% confluence. The cells were lysed using radioimmunoprecipitation assay lysis buffer plus 1 mM PMSF (Servicebio, Wuhan, China) at 4°C for 30 min. The supernatant was collected, and the protein concentration was quantified by a spectrophotometer. Then, the proteins were denatured with protein loading buffer at 100°C for 10 min. Total protein (20 μg) was loaded into gels (EpiZyme, Shanghai, China) with Muticolor Prestained Protein Ladder (EpiZyme, Shanghai, China). Electrophoresis was performed at 80 V for 30 min and then 120 V for 60 min. All proteins were then transferred to a PVDF membrane. The membrane was blocked with skim milk (5%) blocking buffer for 2 h at room temperature (20°C) and incubated overnight at 4°C with rabbit anti-EGFR antibodies (1 : 1000 dilution, Abcam) and rabbit anti-β-actin mAb (1 : 10,000 dilution, Cell Signaling Technology). Next, the membrane was washed three times with TBS-Tween 20 and incubated with goat anti-rabbit IgG antibodies (1 : 5000, Servicebio, Wuhan, China) for 2 h at room temperature. The washed membrane was scanned and quantitatively analyzed using a Tanon 4200 imaging system (Tanon, Shanghai, China). The EGFR expression was analyzed and normalized to β-actin protein for comparison between the MGC803 and HT29 cells using ImageJ 1.44p (National Institutes of Health, Bethesda, MA, USA).

Preparation of
Cet-F(ab′) 2 . Cetuximab was incubated with IdeS protease for 30 min at 37°C in digestion buffer (50 mM sodium phosphate, 150 mM NaCl, pH 6.6). The digested products were incubated with protein A beads for 1 h and centrifuged. The Fc portion attached to the beads was removed in the sediment while the purified Cet-F(ab ′ ) 2 remained in the supernatant. The Cet-F(ab ′ ) 2 and cetuximab were evaluated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) on a 4-15% gel under 150 V for 1 h. The stability of Cet-F(ab ′ ) 2 in PBS and 1%BSA was evaluated through SDS-PAGE.
2.6. Preparation of 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 . Figure 1 shows the synthetic route of 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 . In brief, 2 Molecular Imaging cetuximab was incubated with MAG 3 for 2 h at room temperature in carbonate buffer (pH 9.0). The molar ratio of cetuximab to MAG 3 was 1 : 5, according to previous described methods [13][14][15]. The MAG 3 -Cet was purified using a Zeba Spin Desalting Column 7 K MWCO (Thermo Fisher Scientific, Waltham, MA, USA). MAG 3 -Cet-F(ab ′ ) 2 was prepared using IdeS protease and purified by removing the Fc portion using protein A beads (Epizyme Biomedical Technology, Shanghai, China). Figure S1 shows the characterization of MAG 3 -Cet and MAG 3 -Cet-F(ab′) 2 by SDS-PAGE (Supplementary File). The cheator-to-antibody ratio of the product was determined by liquid chromatography-mass spectrometry (LC-MS) (Bioaccord, Waters, Milford, USA). In brief, MAG 3 -Cet was incubated with IdeS protease for 30 min at 37°C in digestion buffer (50 mM sodium phosphate, 150 mM NaCl, pH 6.6). The digested products were incubated with protein A beads for 1 h and centrifuged. The Fc portion attached to the beads was removed in the sediment while the purified MAG 3 -Cet-F(ab ′ ) 2 remained in the supernatant. Size-exclusion high performance liquid chromatography (SEC-HPLC) was performed to determine the radio purity of the product using an  3 Molecular Imaging temperature for 1 h under vortexing. 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 was purified from unlabeled reduced 99m Tc with PD-10 desalting columns. Radio-HPLC was performed to determine the radiochemical purity of 99m Tc-MAG 3 -cet-F(ab ′ ) 2 with the same machine, mobile phase, and flow rate as mentioned above.
2.7. Stability, Competition and Binding Assay of 99m Tc-MAG 3 -Cet-F(ab′) 2 . The stability of 99m Tc-MAG 3 -Cet-F(ab′ ) 2 was determined in 1× phosphate-buffered saline (PBS) and 1% bovine serum albumin (BSA) for 1, 6, 12, and 24 h. The unfolding agent was a 1 : 2 (v/v) mixture of 0.9% saline and methanol. For the competition assay, 2 × 10 5 MGC803 cells were seeded in 24-well plates with 1.25-1280 nM unlabeled cetuximab and 10 nM 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 and incubated at 37°C for 60 min. The supernatant was discarded, and the cells were washed twice with iced 1× PBS and harvested for determination of radioactivity using a gamma counter. For the binding assay, 2 × 10 5 MGC803 cells and HT29 cells each were seeded in 24-well plates. 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 (0.1-4 nM) in PBS solution was added to the plates and incubated at room temperature for 1 h. Then, the supernatant was discarded while the cells were washed twice with iced 1× PBS and harvested for determination of radioactivity. The binding results including the maximum binding ability (B max ) and the dissociation constant (K d ) were obtained via GraphPad Prism (GraphPad Inc., La Jolla, CA, USA).

Mouse Model Preparation.
All animal studies were performed in accordance with protocols approved by the Animals Ethics Committee of Zhongshan Hospital, Fudan University. Subcutaneous MGC803 and HT29 tumors were induced in 6-week-old male nude mice, by injecting their lower right flanks with 1 × 10 6 tumor cells suspended in 200 μL PBS. The tumors were monitored every other day.
2.9. Biodistribution and Micro-SPECT/CT Imaging. The MGC803 tumor-bearing mice (n = 16) were injected with 18.5 MBq (50 μg) 99m Tc-MAG 3 -Cet-F(ab′) 2 . After 1, 6, 16, and 24 h, four mice from each group were sacrificed and dissected. Tumors, blood, and major tissues/organs (including heart, lung, liver, kidney, spleen, colon, stomach, bone, and muscle) were harvested and weighed. Sample tissue radioactivity was measured using a gamma counter. The radioactivity concentration of the tissue was expressed as the percentage injected dose per g (%ID/g), and the tumor to muscle (T/M) ratio was defined as the ratio of radioactivity that had accumulated in tumors to that in the contralateral muscle. The experiments were repeated three times.
2.10. Immunofluorescence. After deparaffinization and hydration, the HT29 and MGC803 tumor slices were incubated in 5% BSA in PBS buffer for 1 h. Then, the slices were incubated in rabbit anti-EGFR antibody (1 : 200) overnight at 4°C. After washing in PBS, the slices were stained with CY3-conjugated goat anti-rabbit antibody (1 : 100) for 1 h. Following three PBS washes, the nuclei were stained using DAPI. Images were captured using an Olympus imaging system.

Statistical
Analysis. Data are presented as the means ± standard deviation (SD) derived from at least three independent experiments. The Student t-test was applied for intergroup comparisons using GraphPad Prism. All tests were 2-sided, and a statistical P value of <0.05 was considered statistically significant.

Discussion
EGFR is widely expressed in nearly every cancer type, and its high expression in tumors correlates with poor patient out-come [16]. Several studies have demonstrated that patients with digestive tumors benefited from EGFR-targeted therapy using cetuximab [17][18][19]. For patients with unresectable tumors, it is difficult to evaluate the EGFR expression through pathology. Therefore, noninvasive evaluation of the EGFR expression is particularly important. Cetuximab, an EGFR inhibitor widely used in clinical practice, is suitable for noninvasive evaluation of the EGFR expression [20,21].
Immuno-SPECT combines the high specificity of antibodies with the high sensitivity of SPECT imaging [22,23]. Compared with PET, one advantage of SPECT is the price. Typically, achieving the best imaging contrast for a radionuclide-labeled intact antibody (about 150 kDa) requires ≥48 h [8,10,24]. Therefore, this renders it 7 Molecular Imaging unsuitable for imaging radionuclides with short half-lives. For example, for 99m Tc (T 1/2 ≈ 6 h), which is most widely used in SPECT imaging, it would be impossible to achieve the best imaging time before dramatic decay when labelling an intact antibody. If a nuclide with a long half-life was used for labelling an intact antibody, which would allow achieve-ment of the best imaging time (>48 h) before dramatic decay, issues regarding high radiation caused by delayed peak tumor uptake and slow clearance would arise, which would hinder the clinical translation. Regarding the radiation issue, the pretargeting imaging strategy is advantageous, allowing the injection of modified mAbs first with a  predictable duration for its accumulation to the target site.
Then, a small molecule radioligand that can conjugate to the pretargeted mAb is injected for imaging while the redundant radioligand is cleared quickly [25,26]. The most promising pretargeting methodology is based on inverse electron demand (4 + 2) Diels-Alder (IEDDA) cycloaddition between 1,2,4,5-terazine (Tz) and transcyclooctene (TCO), which has been widely used in tumor imaging studies [26,27]. Recently, the Tz/TCO-based and cetuximab pretargeted imaging strategy was successfully used for assessing the EGFR expression in colorectal cancer [28]. However, the expensive cost of synthesizing the Tz and TCO molecules and the inconvenience caused by two injections hinder its clinical use. Therefore, the synthesis of a new probe with high specificity and relatively small molecular weight is necessary.
van Dijk et al. [29,30] prepared Cet-F(ab ′ ) 2 fragment through pepsin digestion and successfully used it for the imaging EGFR expression of head and neck cancer. Here, we used the IdeS digestion to obtain the Cet-F(ab ′ ) 2 , which is entirely different from pepsin digestion. Compared with pepsin, IdeS is a unique cysteine protease that digests antibodies at a single amino acid site below the hinge region, which is suitable for antibodies from multiple sources (e.g., human, mouse, rabbit, monkey, sheep, chimeric IgG, and Fc fusion protein), with high specificity and rapid reaction time (within 30-60 minutes). van Dijk et al. [29] used pepsin to obtain Cet-F(ab ′ ) 2 required with a longer reaction time (4 h) and stricter reaction conditions (pH = 3:8). In addition, pepsin would digest many more restriction sites compared to IdeS [31,32].
The molecular weight of Cet-F(ab ′ ) 2 is about 100 kDa, which is smaller and therefore leads to quicker clearance, compared to its intact counterpart. Yamaguchi et al. [24] and Perk et al. [33] found that the best imaging time point for intact cetuximab-targeted imaging was 48-72 h, which they confirmed with biodistribution assay and PET imaging. In contrast, the best time point of 99m Tc-MAG 3 -Cet-F(ab′ ) 2 -based imaging as identified through biodistribution assay in the present study was 16 h. Although the molecular   Molecular Imaging weight was reduced only by approximately one-third (150 kDa to 100 kDa), the clearance rate was significantly improved. Therefore, 99m Tc is suitable for labeling this tracer, which would certainly reduce patient radiation exposure if translated to the clinic in the future. In addition, immunocytochemistry and SDS-PAGE showed that the Cet-F(ab ′ ) 2 had similar ability to intact cetuximab to bind to EGFR on tumor cells. Furthermore, the Cet-F(ab ′ ) 2 had excellent stability in both NS and 1% BSA. In vitro, the Western blotting and the immunocytochemistry assays revealed higher expression of EGFR on the MGC803 cells than the HT29 cells, while the fabricated 99m Tc-MAG 3 -Cet-F(ab′) 2 presented higher affinity to the MGC803 cells with a higher B max (5:68 × 10 −19 mol ligands/cell) and a lower K d (0.6147 nM), compared to the HT29 cells (B max = 1:66 × 10 −19 mol ligands/cell, Kd = 1:008 nM). These results indicated a good affinity and targeting ability of 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 to EGFR. The biodistribution studies showed that the T/M ratio peaked at approximately 17:29 ± 5:72 at 16 h after 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 injection. Therefore, we consider 16 h the best imaging time point. In SPECT/CT imaging, the MGC803 tumor had significantly higher 99m Tc-MAG 3 -Cet-F(ab ′ ) 2 uptake than HT29 tumor, which was consistent with in vitro results.

Conclusion
SPECT/CT imaging using 99m Tc-MAG 3 -Cet-F(ab′) 2 showed rapid and sustained high radionuclide-uptake in EGFRpositive digestive tumors with high image contrast, which indicates the potential for noninvasive evaluation of EGFR expression in tumors.

Data Availability
The data used to support the findings of this study are available from the corresponding authors upon reasonable request.

Ethical Approval
All applicable institutional and/or national guidelines for the care and use of animals were followed.