Integrin αVβ3‑targeted SPECT/CT for the assessment of Bevacizumab therapy in orthotopic lung cancer xenografts

Introduction

Inhibition of angiogenesis is a promising strategy for breast and lung cancer therapy (1,2). It has been demonstrated that combined therapy with angiogenesis inhibitors may significantly enhance the treatment effect and extend survival in breast and lung cancer patients (3,4). However, inhibitors of angiogenesis tend to be expensive and are not effective in a proportion of cancer patients (5). In addition, the tumor volume changes are much slower than blood supply inhibition. Conventional tools [X-ray and computed tomography (CT)] rely on tumor volume change to evaluate the effect of tumor, which is not suitable for antiangiogenic drugs (6). Therefore, it is critical that a reliable tool is identified to monitor and assess the treatment effect of inhibiting angiogenesis therapy.

Imaging tools have been used to monitor cancer therapy for decades (7,8). In this area, nuclear medicine imaging techniques are promising. 99mTc-3PRGD2 single-photon emission computed tomography (SPECT)/CT has been developed as an imaging modality for evaluating tumor vascular status (9,10). 99mTc-3PRGD2 is a radiolabeled dimeric arginylglycylaspartic acid (RGD) peptide that is being investigated for measurement of the expression of integrin αvβ3 (11). Integrin αvβ3 is widely distributed in newly generated vessels. Additionally, this subtype of integrin has been documented as being associated with tumor angiogenesis and metastasis (12). Therefore, the degree of expression of integrin αvβ3 may reflect the status of tumor angiogenesis.

Bevacizumab is a recombinant humanized monoclonal antibody that blocks angiogenesis by inhibiting vascular endothelial growth factor-A (VEGF-A); it has been used to treat breast and lung cancer since 2004, and has shown promising therapeutic results (3). The present study determined the utility of imaging the uptake of 99mTc-3PRGD2 by tumors as a biomarker for anti-angiogenic treatment with bevacizumab in a lung cancer A549 cell xenograft model, which has high-to-moderate vessel density and exhibits integrin αvβ3 expression, and a PC-3 prostate cancer model, which has low vessel density and also exhibits integrin αvβ3 expression. Following the present study, how changes in tumor uptake of 99mTc-3PRGD2 affect the tumor response to antiangiogenic treatment can be better understood before it can be used clinically to monitor investigational therapy.

Materials and methods

Animal models and treatment protocol

All animal experiments were performed in accordance with the protocol approved by the Institutional Animal Care and Use Committee at Jilin University (Changchun, China). The A549 and PC-3 cell lines were obtained from the American Type Culture Collection (Manassas, VA, USA). These cells were cultured in F-12 medium (Gibco, Life Technologies, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and 1% penicillin and streptomycin (Gibco; Thermo Fisher Scientific, Inc.) solution at 37°C in a humidified atmosphere of 5% CO2. Cells were grown as a monolayer and were harvested or passaged when they reached 90% confluence to maintain exponential growth. A total of 30 male Athymic nu/nu mice were obtained from the Department of Experimental Animals (Peking University) at 4–5 weeks of age. Mice were housed under standard laboratory conditions (temperature, 20–24°C; relative humidity, 50–60%, 12/12 h light/dark cycle) and had food and water available ad libitum. Each mouse was implanted subcutaneously near the shoulder with 5×106 cells. At 4 weeks after inoculation with A549 and PC-3 cells, the mice were divided into three groups, with 9 or 10 mice in each group. All the groups were size-matched, with an average tumor volume of 180±90 mm3 1 day before baseline SPECT/CT imaging. Vehicle (0.15% hydroxypropylmethyl cellulose, 2% ethanol, 5% Tween 80, 20% PEG 400 and 73% saline) or bevacizumab (Genentech, San Francisco, CA, USA) was injected intraperitoneally. The treatment protocol of each group was: A549 Bevacizumab group (A549 model, n=10), 1 mg bevacizumab twice a week from day 0 after baseline imaging; A549 Vehicle group (A549 model, n=10), vehicle with a dose of 1 mg twice a week from day 0 after baseline imaging; and PC-3 Bevacizumab group (PC-3 model, n=9), 1 mg bevacizumab twice a week from day 0 after baseline imaging.

Preparation of 99mTc-3PRGD2 and small animal SPECT/CT

Na99mTcO4 was obtained from a commercial 99Mo-99mTc generator (Beijing Atom High Tech Co., Ltd., Beijing, China). The kit for preparation of 99mTc-3PRGD2 was formulated by containing 20 µg/ml of HYNIC-3P4-RGD2, 5 mg of TPPTS, 6.5 mg of tricine, 40 mg of mannitol, 38.5 mg of disodium succinate hexahydrate and 12.7 mg of succinic acid. Then 1 ml of Na99mTcO4 solution (1110–1850 MBq) in saline was added to each kit vial followed by 20 min incubation at 100°C (11,13). The radiochemical purity of the prepared 99mTc-3PRGD2 was >90%. Helical CT and SPECT scans of rats were obtained using a SPECT/CT system (NanoScan; Mediso Medical Imaging Systems, Budapest, Hungary). Longitudinal SPECT/CT imaging was performed at baseline (−1), 5, and 15 days after treatment initiation. At 1 h prior to SPECT/CT imaging, 37.0–44.4 MBq 99mTc-3P-RGD2 in 0.1–0.2 ml of saline was administered intravenously via the lateral tail vein. Animals were anesthetized with 3% isoflurane inhalation, which was maintained at 1.5% for the duration of scanning, and then placed in a prone position in an air-warmed chamber. For radioactivity quantification, the regions of interest were drawn manually to cover the entire tumor, based on a transverse view of the CT image. For tumor delineation with SPECT, a threshold of ≥50% of the maximum pixel value on the SPECT image was chosen. Tumor volume and radioactivity counts were generated using NanoScan Image Processing software (version 3.306; PMOD Technologies LLC, Zürich, Switzerland) and the amount of radioactivity in each tumor was calculated. The tumor uptake of 99mTc-3PRGD2 was expressed as the percent-injected dose (%ID) and %ID/g. Reference regions of interest were drawn over muscle as background radioactivity for tumor-to-muscle (T/M) ratio calculations.

Tumor immunostaining

Immunofluorescence staining was performed to determine the location and expression of integrin αvβ3. Tumors were sectioned into two pieces for immunostaining and hematoxylin and eosin (H&E) staining. Once tumors were harvested, the tumor sections for immunostaining were immediately snap-frozen in optical cutting temperature solution (99% purity, Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). Tumors were then cut into 5-mm sections. Following thorough drying at room temperature, slides were fixed with ice-cold acetone for 10 min, and air-dried for 20 min at room temperature. The sections were then blocked with 10% goat serum (Abcam, Cambridge, MA, USA) for 30 min at room temperature and then incubated with rat anti-integrin β3 antibody (1:100; cat. no. 181720; BD Biosciences, Franklin Lakes, NJ, USA) and rat anti-CD31 antibody (1:100; cat. no. 551262; BD Biosciences) for 1 h at room temperature. The β3 antibody was chosen to represent αvβ3 as the only other integrin with an αβ3 subunit besides αvβ3 is expressed on platelets. The majority of β3 in the tumor sections is likely to be in the vasculature and tumor cells. After incubating with Cy3-conjugated goat anti-rat (1:100; cat. no. 115-165-003; Jackson ImmunoResearch Europe Ltd., Newmarket, UK) and fluoresceinisothiocyanate-conjugated goat anti-rat secondary antibodies (1:100; cat. no. 115-095-003; Jackson ImmunoResearch Europe Ltd.) at room temperature (25°C) for 4 h, the sections were washed with PBS. Fluorescence was visualized with a Nikon fluorescence microscope at ×200 magnification (Nikon Eclipse E600; Nikon Corporation, Tokyo, Japan).

H&E staining

Histopathological analysis was performed by H&E staining of tumors according to previously published methods (14). Briefly, all the tissues were fixed in 10% neutral buffered formalin at room temperature (25°C) for 4 h. Tissues were embedded in paraffin and 4-mm sections were deparaffinized and rehydrated using a graded alcohol series. Sections were stained with H&E at room temperature (25°C) for 20 min to evaluate the morphology and then examined under a light microscope. Aperio's Image Scope v10.1.3.2028 Viewer (Leica Mircosystems, GmbH, Wetzlar, Germany) was used to visualize the whole-slide digital scans and capture images in 30 fields of view for analysis.

Statistical analysis

All data were expressed as the mean ± standard error. Statistical analyses were performed by two-way analysis of variance followed by the Newman-Keuls test for multiple comparisons to compare treatment groups. P<0.05 was considered to indicate a statistically significant difference. A one-way analysis of variance and Newman-Keuls post-hoc test was performed to determine alterations over time. SPSS 19.0 software package (IBM Corp., Armonk, NY, USA) was used for linear and nonlinear regression analysis.

Results

Tumor volume in bevacizumab-treated tumor models

Fig. 1 depicts the tumor volumes for A549 Bevacizumab, A549 Vehicle and PC-3 Bevacizumab groups. In the PC-3 Bevacizumab and A549 Vehicle group, the tumor volume increased at 5 and 15 days after therapy. The difference in tumor volume was not significantly different prior to and following treatment in the A549 Bevacizumab group (P>0.05).

Figure 1. Comparison of tumor volumes between the A549 Bevacizumab (n=10), A549 Vehicle (n=10) and PC-3 Bevacizumab (n=9) groups in the mouse tumor model. Tumor volumes were determined by caliper measurements.

Tumor uptake in bevacizumab-treated tumor models

Fig. 2 compared the %ID tumor uptake (Fig. 2A), %ID/g tumor uptake (Fig. 2B) and T/M ratios (Fig. 2C) of 99mTc-3PRGD2 in the A549 Bevacizumab, A549 Vehicle and PC-3 Bevacizumab groups. For the PC-3 Bevacizumab group, a lower tumor uptake was observed throughout the study compared with the other 2 groups, and there was no significant alteration prior to and following bevacizumab therapy. After 5 and 15 days therapy, the %ID/g value for A549 Bevacizumab group declined. However, for the A549 Vehicle group, the %ID value increased 5 days after injection, but the %ID/g tumor uptake decreased after 5 and 15 days therapy as the weight of the tumor increased. For the A549 Bevacizumab and A549 Vehicle groups, the T/M ratio decreased 5 and 15 days after injection. Further SPECT/CT studies confirmed that bevacizumab-treated tumors have persistent tumor uptake decrease. Fig. 3 depicted SPECT/CT images for decreased uptake of 99mTc-3PRGD2 of tumors in the A549 Bevacizumab group after treatment, increased uptake in the A549 Vehicle group, and low uptake of 99mTc-3PRGD2 before and after treatment in the PC-3 Bevacizumab group.

Figure 2. Comparison of the (A) %ID tumor uptake, (B) %ID/g tumor uptake and (C) T/M ratios in A549 Bevacizumab (n=10), A549 Vehicle (n=10) and PC-3 Bevacizumab (n=9) groups. Tumor uptake values and T/M ratios were calculated from SPECT/CT images of the mouse models. ID, injected dose; T/M, tumor-to-muscle; SPECT/CT, single photon emission computed tomography/computed tomography.

Figure 3. SPECT/CT images from (A) bevacizumab-treated and (B) vehicle-treated A549 tumor models and (C) bevacizumab-treated PC-3 tumor models. SPECT/CT images were obtained at baseline (−1), and at 5 and 15 days after initiation of bevacizumab therapy. SPECT/CT, single photon emission computed tomography/computed tomography.

Change in microvessel density (MVD) following bevacizumab treatment

Fig. 4 depicts selected histological slices (H&E stained) of tumor tissues from animals before and after 5 and 15 days of A549 Bevacizumab, A549 Vehicle and PC-3 Bevacizumab groups. The PC-3 model exhibited low MVD throughout the study. Prior to bevacizumab therapy, the MVD of the A549 model was moderately high. After 15 days therapy, the MVD of the A549 model was significantly lower than before therapy.

Figure 4. Histological slices of tumor tissues of A549 and PC-3 from animals at baseline (−1), and 5 and 15 days after initiation of bevacizumab therapy, illustrating alterations in vascular density.

Change in integrin αvβ3 following bevacizumab treatment

Fig. 5 depicts overlay images of A549 and PC-3 tumor tissues after immunohistochemical staining for integrin β3 and CD31 in A549 Bevacizumab, A549 Vehicle and PC-3 Bevacizumab groups. The PC-3 model exhibited low integrin β3 levels. At 5 days after the start of therapy, the integrin β3 level in the bevacizumab-treated models decreased. No evident alterations in the levels of CD31 were observed following bevacizumab-treatment. In the vehicle-treated model, the expression level of integrin β3 was elevated. At 15 days after the start of therapy, the integrin β3 and CD31 levels in the bevacizumab-treated models decreased. In the vehicle-treated model, tumor volume increased and integrin β3 and CD31 levels decreased, which may be due to the lack of neovascularization in the deep tumor tissue.

Figure 5. Images of A549 and PC-3 tumor tissues following immunohistochemical staining for integrin β3 and CD31 in vehicle- and bevacizumab-treated groups to illustrate the alterations in integrin β3 expression levels and blood vessel density during therapy. Tumor tissues were harvested at −1, 5 and 15 days after initiation of bevacizumab therapy. CD31 was used as the marker for tumor blood vessels (mature and newly formed), which was visualized with fluorescein isothiocyanate (green color). Integrin β3 was visualized with cyanine 3 (red color). The yellow color in overlay images indicates the integrin β3 expression on new blood vessels. Original magnification ×200. CD31, cluster of differentiation 31.

Discussion

Multiple growth factors, including VEGF and platelet-derived growth factor (PDGF) elevate integrin αvβ3 levels in vitro (15). Consequently, suppressing the expression or signaling of VEGF and PDGF receptors may decrease integrin αvβ3 levels. The target site of bevacizumab is VEGF and PDGF receptors. Therefore, integrin αvβ3 may be deemed to be a relevant biomarker for the evaluation of the biological effects that occur following bevacizumab therapy. Previous enhanced magnetic resonance imaging studies have indicated that, either in preclinical or clinical studies, a tumor vasculature system experiences alterations following bevacizumab therapy (16–18). In the present study, SPECT/CT was used to dynamically observe changes in 99mTc-3PRGD2 uptake at two imaging time points, prior to and following dosing with bevacizumab in xenografts. At 5 days after the start of therapy, a significant decrease in the tumor uptake in the A549 model was observed. At 15 days after the start of therapy, the tumor volume in the A549 model mildly decreased. In comparison, the tumor volume in the PC-3 model and the vehicle-treated model increased. Additionally, the tumor uptake consistently increased in the vehicle-treated model. Therefore, 99mTc-3PRGD2 SPECT/CT may be used in early therapy monitoring in bevacizumab treatment in the A549 model, but not in the PC-3 model.

The present study revealed that the degree of increase in %ID/g and T/M values were evident in the A549 model. Previous studies indicated that the treatment effect of bevacizumab was closely associated with MVD and integrin αvβ3 levels in different tumor types, which may explain differences with the present study (18). Compared with the PC-3 model, which had low MVD, the A549 model, with moderately high MVD, may express higher levels of integrin αvβ3. Further studies using models with poor integrin αvβ3 expression are required to investigate whether SPECT/CT could be used to rule out integrin αvβ3 expression non-invasively in an experimental and clinical setting.

MVD is the number of vessels per unit area of tumor tissue; it directly reflects the capability of formation of new tumor vessels. MVD is positively associated with tumor metastasis and proliferation, and can be used as an index for evaluating therapeutic effects of solid tumors (19,20). Through observing alterations in tumor microvessels via H&E staining, the present study demonstrated that the MVD of the bevacizumab treated model significantly decreased. The immunofluorescence staining results during therapy suggested that at 5 days after the start of therapy, a decrease of integrin αvβ3 on the surface of tumor cell was not evident, whereas the decrease of integrin αvβ3 within the new epithelial cells was. These results indicated the anti-angiogenic effect of bevacizumab and may explain the decrease of 99mTc-3PRGD2 SPECT/CT uptake in tumors. 18F-labeled peptides such as 18F-fluciclatide and 18F-FPPRGD2 have been used to monitor the therapeutic effects of anti-angiogenic agents, including sunitinib, ZD4190 and functional paclitaxel (21–23). The results of the present study are in accordance with these previous studies. Considering the availability of pharmacokinetics, biodistribution, radiation dose and 9mTc-3PRGD2, we hypothesize that 99mTc-3PRGD2 SPECT/CT could be more cost-effective compared with the 18F-labeled analogs. The T/M ratio of tumor RGD uptake was linearly associated with integrin αvβ3 and CD31 expression, as previously reported (13). Quantitative analysis revealed that 99mTc-3PRGD2 SPECT/CT was prominent in future clinical applications (24). However, to understand the association between T/M ratio and the treatment effects better, further research is required.

In conclusion, in the A549 model, 99mTc-3PRGD2 tumor uptake decreased following treatment with bevacizumab. 99mTc-3PRGD2SPECT/CT may be used as a non-invasive tool to evaluate the early biological effects of anti-angiogenic therapy.

Acknowledgements

This study was supported by the Research Fund of Science and Technology Department of Jilin Province (grant nos. 20150520154JH and 20160101064JC), the Foundation of National Health and Family Planning Commission of Jilin Province (grant nos. 2015Q020 and 2016Q038), the Department of Education of Jilin Province for Thirteen-Five Scientific Technique Research [grant no. (2016) 460], the Norman Bethune Program of Jilin University (grant no. 2015437) and Jilin University Funding Project for Young Teacher Cultivation Plan (grant no. 419080500365).

References