Evaluation of the anti-angiogenic effect of bevacizumab on rat C6 glioma by spectral computed tomography

Abstract

Background

Anti-angiogenic drugs have become a research hotspot in recent years. However, dynamically observing their therapeutic effect at different time points during treatment is a clinical problem.

Purpose

To explore the feasibility of the quantitative parameters of spectral computed tomography (CT) in evaluating the anti-angiogenic effect of bevacizumab on rat C6 glioma.

Material and Methods

Twenty-six male Sprague-Dawley rats were used to establish the C6 glioma model. The rats were randomly divided into the experimental group (n = 13) and control group (n = 13). The experimental group was intraperitoneally injected with 0.2 µL/g bevacizumab every day, whereas the control group was injected with the same dose of normal saline every day for one week. Spectral CT scanning was performed on the 4th and 8th days after treatment; meanwhile, the brain tissues were collected by heart perfusion for H&E staining, and VEGF and HIF-1α immunohistochemical staining.

Results

On the 4th and 8th days, significant differences in the 70-keV single-energy CT value, slope of the energy spectrum curve, and iodine concentration were found between the experimental group and the control group. Correlation analysis between immunohistochemistry and quantitative parameters of spectral CT showed that the single energy CT value of 70 keV, slope of the energy spectrum curve, and concentration of iodine were positively correlated with VEGF and HIF-1α at different time points in the experimental group and the control group.

Conclusion

Spectral CT multi-parameter imaging can be employed as a new method to evaluate the anti-angiogenic effect of bevacizumab on rat C6 glioma.

Keywords

Spectral computed tomography glioma quantitative parameters bevacizumab anti-angiogenesis

Introduction

Glioma is one of the most common primary tumors of the central nervous system. Patients with this disorder have a median survival time of only 14.6 months (1). Malignant glioma has a high incidence as well as a high mortality and serves as a serious threat to human health and life safety (2,3). Some studies have shown that the occurrence and development of tumors depend on the neovascularization sent by the host, neovascularization is a characteristic manifestation of glioma. Tumor cell division, proliferation, invasion, and distant metastasis are dependent on neovascularization (4). Neovascularization is closely related to the prognosis and tumor grade of patients (5), a process that the formation of neovascularization is driven by vascular markers (6,7), e.g. vascular endothelial growth factor (VEGF) and hypoxia inducible factor (HIF-1α). VEGF is an important factor affecting angiogenesis, especially high expression in gliomas (8,9). It plays an important role in the process of tumor neovascularization and can increase the chances of tumor invasion and metastasis. HIF-1α widely exists in tumor cells and can adapt cancer cells to hypoxia environment by regulating different target genes such as VEGF (10). HIF-1α can regulate the expression of related genes such as energy metabolism, cell proliferation, and angiogenesis. It also provides energy for cells with tumor hypoxia under radiotherapy to enhance the survival ability of tumor cells.

The higher the expression of vascular-related markers, the higher the malignant degree of the tumor, the stronger the degree of invasion and the ability of distant metastasis, and the worse the therapeutic effect and prognosis of the patients. Anti-angiogenic drugs have thus become a research hotspot in recent years. Drug targeting therapy can reduce neovascularization, normalize blood vessels, and improve tissue oxidation after the treatment period. The corresponding tumor microenvironment will also be improved and can enhance the efficacy of radiotherapy and chemotherapy (11,12). However, dynamically observing their therapeutic effect at different time points during the treatment process is a clinical problem. It is very important to discover a non-invasive and dynamic method to monitor the efficacy of anti-angiogenic drugs for glioma treatment. Previous studies have shown that dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters can dynamically evaluate the therapeutic effect of anti-tumor growth and anti-angiogenesis in rat glioma model (13,14) and 18F-FLT MicroPET/computed tomography (CT) imaging can monitor the efficacy of sunitinib in anti-angiogenic treatment of gliomas (15). However, it is not clear whether spectral CT can accurately evaluate the efficacy of anti-angiogenic drugs. Therefore, the present study established a rat C6 glioma model, observed the correlation between the quantitative parameters of spectral CT and anti-angiogenesis, and explored the feasibility and value of the spectral CT quantitative parameters for evaluating anti-angiogenesis.

Material and Methods

The study protocol was approved by the Ethics Committees of the Lanzhou University Second Hospital, Lanzhou, PR China (D2017‑030).

C6 glioma cell culture

The C6 cell line was purchased from the primary cell biopharmaceuticals of Qi’s biotechnology (Jiangyin, PR China) and cultured at 37°C in a 5% CO₂ incubator maintained at constant temperature. DMEM containing 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) was used as the medium and was changed every other day. The medium was often given to the second to third generations. The cells were digested with 10% trypsin for 3–5 min when they reached the logarithmic growth phase. After digestion, counting, centrifugation, addition of the serum-free medium, Hank’s solution blowing, and washing once or twice to form the cell suspension, the inoculation concentration was adjusted to 1 × 10⁶/mL–10⁸/mL for use.

Rat model of glioblastoma and drug therapy

Thirty Sprague-Dawley male rats (SPF grade, 180–230 g) were used to establish the C6 glioma model. Rats were anesthetized with 2% pentobarbital sodium (0.2 mL/100 g) and then the head was fixed on the brain stereotactic apparatus (RWD Life Science Co., Shenzhen, PR China). The right side of the sagittal suture of the skull was opened 4 mm, a point of 1 mm was used for the coronal suture. A 10 µL microinjector (Hamilton [10 µL flat head], Yuanquan Biotechnology Co., Ltd., Shanghai, PR China) was used to extract 10 µL of the C6 glioma cell suspension. The injection rate was 1 µL/min and injection time was 10 min. After the injection, the needle was retained for 5 min and then pulled out slowly and vertically. The needle eyes were closed with bone wax and the routine operation area was sutured and sterilized after operation. Two weeks after modeling, all rats were scanned by spectral CT (HD750 CT scanner; GE Healthcare, Little Chalfont, UK), and tumor formation was observed. After scanning, rats were randomly divided into the experimental and control group.

The experimental group was intraperitoneally injected 0.2 μl/g bevacizumab (Roche, Shanghai, PR China) every day, while the control group was injected the same dose of normal saline (0.9% NaCl) every day for one week. The pre-experimental results showed that bevacizumab had a certain anti-angiogenic effect on the 4th and 8th days after injection. Therefore, spectral CT scanning was performed on the 4th and 8th days after treatment. Quantitative parameters and pseudo-color images were obtained using the postprocessing software of spectral CT.

CT scanning

Two weeks after tumor implantation, rats were anesthetized with 2% pentobarbital sodium (0.2 mL/100 g) and fixed. The tail vein was indwelled with a single intravenous infusion needle (0.45 × 13.5 RWLB; Weigao Medical Polymer Co., Ltd., Weihai, PR China). The prone position was used and the head was advanced for scanning. Thereafter, the contrast medium was injected into the high-pressure syringe (Ulrich, Germany) for spectral CT scanning in the spectral CT scanning mode: axial scanning mode: Iohexol (300 mgI/mL; Yangtze River Pharmaceutical Group, Taizhou, PR China) = 2.5 mL/kg; flow rate = 0.2 mL/s; rotating speed = 0.5 s; tube voltage = 80/140 keV instantaneous switching; tube current = 630 mA fixed tube current; layer thickness = 0.625 mm; detector width = 20 mm; matrix = 512 × 512; and delay time = 30 s; scan field of view (SFOV) = small head; display field of view (DFOV) = 9 cm; scanning range = 4 cm; reconstruction algorithm = Stnd; 30% adaptive statistical iterative reconstruction (ASIR).

Analysis of spectral CT data

All images and data were transferred to the AW4.6 workstation for image postprocessing. Application of GSI-Viewer, a unique analysis software of spectral CT was used for data measurement and analysis. The energy spectrum scanning image was adjusted to optimal single-energy 70 keV to place the region of interest (ROI) and the data were saved. The ROI was the solid area of the tumor. The ROI was designated as a circle with a diameter of 0.5 mm. When the ROI was indicated, necrotic cysts and large blood vessels were avoided, and the contralateral mirror area avoided the lateral ventricle. Three ROIs were set up for each case and the average value was statistically analyzed. The single-energy CT value of 40–140 keV, the concentration of iodine and the slope of the energy spectrum curve were recorded. The slope of the energy spectrum curve was calculated according to the corresponding longitudinal coordinates of 40 keV and 90 keV. Thus, the slope of the energy spectrum curve = (CT40 keV– CT90 keV)/(90–40).

Pathological examination

For hematoxylin and eosin (H&E) and immunohistochemical staining of the tumor tissue, three rats were randomly selected from the experimental group and the control group at the end of each spectral CT scanning. The brain was perfused under deep anesthesia and whole brain was fixed with 4% paraformaldehyde for 24 h. The position sampling that corresponded to the spectral CT measurement was then recorded. The section, which was 4–5 µm thick and was stained with H&E, VEGF, and HIF-1α. Immunohistochemical staining was carried out by the En Vision method. DAB staining was performed and the nucleus was re-stained. Negative and positive controls were derived for each staining.

In the present study, we selected the level of the maximum diameter of the tumor in axial to perform CT image-related measurements on tumors. Meanwhile, histological sections and pathological staining were also performed according to the above criteria. Two senior pathologists independently analyzed the results of pathological sections according to the blind principle and calculated the number of VEGF and HIF-1α positive cells in each high-power field.

Statistical analysis

The experimental data were analyzed by GraphPad Prism7 and SPSS 23.0. Satisfying the normal distribution, data were expressed as mean ± SD. Independent samples t test was employed to compare the quantitative parameters of spectral CT with or without statistical significance. The correlation between the parameters of spectral CT and expression of VEGF and HIF-1α positive protein at different time points were analyzed by Pearson’s correlation test. The test level for α = 0.05. When the P < 0.05, the difference in time was statistically significant.

Results

Experimental animal condition

Of the 30 Sprague-Dawley rats, two did not have tumors and, after scanning, two unexpectedly died. Thus, 26 rats were included in the final analysis.

Results of tumor volume measurement by enhanced scanning

Tumor could be observed on the 14th day after successful establishment of the model (Fig. 1a–c). Tumor volume in the experimental group was approximately 84.28 ± 13.52 mm³, 80.80 ± 12.12 mm³, and 77.25 ± 4.12 mm³ on days 0 (before treatment), 4, and 8, respectively, while that in the control group was approximately 83.33 ± 14.65 mm³, 99.82 ± 13.15 mm³, and 113.25 ± 20.52 mm³, respectively. In the control group, the edge of the tumor was evidently enhanced, whereas the degree of central enhancement (i.e. the necrotic area after enhanced scanning) was decreased. In the experimental group, the parenchyma of the tumor was significantly enhanced after contrast-enhanced scanning, the center appeared as the necrotic area, and tumor volume was reduced during continuous administration.

Fig. 1.

Spectral CT images of the tumor. (a–c) Pseudo-color images of the tumor located in the axial, sagittal, and coronal positions of the right basal ganglia region after successful modeling; L1 represents the position and size of the ROI placed on the tumor. (d–f) Tumor volume gradually increased on days 0, 4, and 8 in the control group. (g–i) Tumor volume gradually decreased on days 0, 4, and 8 in the experimental group. CT, computed tomography.

The optimal contrast-to-noise ratio (CNR) of spectral CT imaging

The optimal CNR curve of 40–140-keV single-energy imaging of tumor and adjacent brain tissue was drawn by AW4.6 workstation energy spectrum imaging optimal CNR analysis software. The results showed that 22/26 samples obtained the optimal CNR at 70 keV and the remaining four cases achieved optimal CNR at 60 keV and 65 keV, respectively. But good tissue contrast was also obtained at 70-keV single-energy imaging.

Quantitative parameters of spectral CT

There were no significant differences in the 70-keV single-energy CT value, energy spectrum curve slope, and iodine concentration between the control group and the experimental group before treatment. The 70-keV single-energy CT values of the experimental group on the 4th and 8th days after treatment were 80.58 ± 5.57 HU and 75.51 ± 5.42 HU, respectively. The slopes of the energy spectrum curve were 1.42 ± 0.38 and 1.54 ± 0.39, and iodine concentrations were 11.91 ± 3.23 mg/cm³ and 9.51 ± 1.36 mg/cm³. In the control group, the 70-keV single-energy CT values were 90.77 ± 9.67 HU and 92.08 ± 7.42 HU, slopes of the energy spectrum curve were 1.09 ± 0.29 and 1.05 ± 0.21, and iodine concentrations were 22.32 ± 5.35 mg/cm³ and 27.12 ± 5.06 mg/cm³ on the 4th and 8th days, respectively. All parameters obeyed the normal distribution; the difference between groups was statistically significant (Table 1). Spectral CT images of the control group and the experimental group were shown in Fig. 1d–i.

Table 1.

Quantitative parameters of spectral CT at different time points.

Group	Day 0			Day 4			Day 8
Group	70 keV (HU)	K*	I (mg/cm³)	70 keV (HU)	K*	I (mg/cm³)	70 keV (HU)	K*	I (mg/cm³)
Experimentalgroup	85.97 ± 9.04	1.27 ± 0.43	17.48 ± 2.07	80.58 ± 5.57	1.42 ± 0.38	11.91 ± 3.23	75.51 ± 5.42	1.54 ± 0.39	9.51 ± 1.36
Control group	81.35 ± 9.98	1.19 ± 0.31	18.21 ± 1.43	90.77 ± 9.67	1.09 ± 0.29	22.32 ± 5.35	92.08 ± 7.42	1.05 ± 0.21	27.12 ± 5.06
T	1.350	0.563	–0.991	–3.562	2.623	–6.426	–6.426	4.225	–13.005
P	0.188	0.579	0.331	<0.05	<0.05	<0.001	<0.001	<0.001	<0.001

Values are given as mean ± SD.

*K is the slope of the energy spectrum curve.

Pathological results

H&E staining was used to observe the dense arrangement, irregular morphology, large nucleus, and deep staining of tumor cells in the solid area of the tumor. The present study found that the proportion of nucleoplasma increased and as treatment time prolonged, small punctate or patchy necrotic areas gradually appeared in the tumor in the experimental group. Tumor cells were also arranged relatively loosely, the cell structure collapsed, and inflammatory cell infiltration was observed around the necrotic area. In the control group, the area of tumor tissue necrosis was relatively minor. As observation time extended, the number of VEGF and HIF-1α positive cells in the experimental group was found to be lower than that in the control group (Fig. 2).

Fig. 2.

The expression of VEGF, HIF-1ɑ positive cells in the control group and experimental group at different time points. (a–c) The expression of VEGF positive cells and VEGF increased on days 0, 4, and 8 in the control group; (d–f) the expression of VEGF positive cells and VEGF decreased on days 0, 4, and 8 in the experimental group. (g–i) The expression of HIF-1α positive cells and HIF-1α increased on days 0, 4, and 8 in the control group; (j–l) the expression of HIF-1α positive cells and HIF-1α decreased on days 0, 4, and 8 in the experimental group.

Correlation analysis between the quantitative parameters of spectral CT and immunohistochemical staining results

As the observation time extended, the 70-keV single-energy CT value and iodine concentration decreased, whereas slope of the energy spectrum curve increased in the experimental group. In contrast, the 70-keV single-energy CT value and iodine concentration increased while the slope of the energy spectrum curve decreased in the control group. The correlation between the quantitative parameters of spectral CT and the positive results of immunohistochemical staining were analyzed by the Pearson’s correlation test. The results of the correlation analysis showed that the single energy CT value of 70 keV, the slope of the energy spectrum curve, and the concentration of iodine were positively correlated with VEGF and HIF-1α at different time points in the control group and the experimental group (Table 2, Figs. 3 and 4).

Table 2.

Correlation analysis between the quantitative parameters of spectral CT and VEGF, HIF-1α at different time points.

Parameters	Day 0		Day 4		Day 8
Parameters	r	P	r	P	r	P
VEGF
Experimental group
70 keV (HU)	0.737	0.002*	0.856	0.000^†	0.663	0.007*
K	0.554	0.032^‡	0.770	0.001*	0.715	0.003*
I (mg/cm³)	0.555	0.032^‡	0.813	0.000^†	0.763	0.001*
Control group
70 keV (HU)	0.737	0.000^†	0.668	0.006*	0.609	0.016^‡
K	0.610	0.012^‡	0.611	0.016^‡	0.646	0.009*
I (mg/cm³)	0.563	0.023^‡	0.702	0.004*	0.558	0.031^‡
HIF-1α
Experimental group
70 keV (HU)	0.734	0.002*	0.851	0.000^†	0.767	0.001*
K	0.719	0.004*	0.763	0.001*	0.792	0.000^†
I (mg/cm³)	0.575	0.025^‡	0.798	0.000^†	0.813	0.000^†
Control group
70 keV (HU)	0.780	0.001*	0.702	0.004*	0.616	0.014^‡
K	0.692	0.003*	0.631	0.012^‡	0.676	0.006*
I (mg/cm³)	0.611	0.012^‡	0.685	0.005*	0.605	0.017^‡

*P < 0.01 compared with VEGF and HIF-1α.

^†P < 0.001 compare with VEGF and HIF-1α.

^‡P < 0.05 compared with VEGF and HIF-1α.

r, correlation coefficient.

Fig. 3.

Correlation analysis between the quantitative parameters of spectral CT and VEGF at different time points in the experimental group and control group. The single-energy CT value of 70 keV, iodine concentration, and the slope of the energy spectrum curve were moderately and highly correlated with VEGF on days 0, 4, and 8, respectively. CT, computed tomography.

Fig. 4.

Correlation analysis between the quantitative parameters of spectral CT and HIF-1ɑ at different time points in the experimental group and control group. The single-energy CT value of 70 keV, iodine concentration, and the slope of the energy spectrum curve were moderately and highly correlated with HIF-1ɑ on days 0, 4, and 8, respectively. CT, computed tomography.

Discussion

Previous studies in the evaluation of tumor neovascularization mainly focused on directly detecting the morphological and functional changes of tumor blood vessels, and only stayed on the changes of tumor blood perfusion after drug action and ignored the changes of tumor microenvironment. Although pathological examination is the gold standard for evaluating tumor angiogenesis, these related measurements are susceptible to the location of the material taken and the experience of the observer. In addition, immunohistochemistry and immunofluorescence are invasive, so it is difficult to repeatedly and dynamically monitor the efficacy of targeted anti-angiogenic drugs. Therefore, their clinical application is limited.

Spectral CT is a new imaging technique that uses the attenuation coefficient of a material irradiated by X-ray at different energy levels for projection reconstruction, enabling the transformation into the corresponding image (16). Spectral CT can quantitatively detect the tissue structure and functional state of the lesion by using the effective atomic number, single-energy CT value, energy spectrum curve, and material separation technique. Spectral CT can also improve the contrast and detection rate of the lesion (17). The single-energy image provided by spectral CT can remove the X-ray hardening artifact (18) and accurately reflect the change in tissue density. As X-ray attenuation at different energy levels has different characteristics, it is necessary to determine the best single-energy level based on different detection purposes (19). Some researchers (20) have found that low energy can better display and diagnose pancreatic cancer. In fact, Tang et al. (21) compared 62 cases of abdominal traditional CT images to energy spectrum CT images. They found that the single-energy CT images could clearly display gastrocolic ligaments while the 50–70-keV single-energy CT images had the optimal CNR. In the present study, the spectral CT images of 26 Sprague-Dawley rats were used to plot the optimal CNR curve of tumor using the 40–140-keV single-energy imaging with AW4.6 workstation spectral imaging analysis software. The results showed that 22/26 samples obtained the optimal CNR at 70 keV and the remaining four cases achieved the optimal CNR at 60 keV and 65 keV, respectively. But good tissue contrast was also obtained at 70-keV single-energy imaging. Therefore, this study identified 70-keV single-energy imaging was the optimal single energy image to display tumor and peri-tumor tissue.

Bevacizumab is a recombinant humanized VEGF monoclonal IgG1 antibody (22) that competitively inhibits the binding of VEGF to its receptor on the cell surface by binding to VEGF-A, ultimately inhibiting tumor angiogenesis (23,24). In this study, after treatment with the anti-angiogenic drug, bevacizumab, the 70-keV single-energy CT value and iodine concentration decreased while the slope of the energy spectrum curve increased in the experimental group. This is because VEGF-A is overexpressed in gliomas of the central nervous system and many neovascularization exists in the tumor. The greater the neovascularization, the higher the microvessel density and the higher the degree of enhancement. In addition, the greater the iodine content in the lesion, the higher the corresponding single energy CT value and the iodine concentration value. After treatment with bevacizumab and competitive inhibition of VEGF binding to the corresponding VEGFR on the cell surface, the expression of VEGF and HIF-1α decreased to achieve anti-angiogenesis, the corresponding single-energy CT value and iodine concentration decreased, and the slope of the energy spectrum curve increased.

The present study had several limitations. First, the sample size was relatively small (n = 26). At the end of each scan, three rats from each group were killed, which may have led to a statistical bias. Therefore, it is necessary to increase the sample size for further verification. Second, the ROI selected in this study may contain pathological microcapsule regions. Owing to the low soft-tissue resolution of CT, it is difficult to display the internal microstructure of the tumor, resulting in statistical differences. Finally, because the tumor volume was small, the diameter of the ROI was 0.5 mm, which may be affected by the volume effect. Hence, some differences exist in the detection value.

In conclusion, the quantitative parameters for spectral CT, namely the 70-keV single-energy CT value, slope of the energy spectrum curve, and iodine concentration, can reflect the curative effect of tumor anti-angiogenic drugs. Therefore, spectral CT can serve as a new method for non-invasive and dynamic evaluation of the early efficacy of anti-angiogenic drugs.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received the following financial support for the research, authorship, and/or publication of this article: This study has received funding by National Natural Science Foundation of China (No. 81772006).

ORCID iDs

Lei Han

Xiaoyu Huang

References

Wen

Kesari

Malignant gliomas in adults. N Engl J Med 2008; 359:492–507.

Kebir

Weber

Lazaridis

, et al. Hybrid 11C-MET PET/MRI combined with “machine learning” in glioma diagnosis according to the revised glioma WHO classification 2016. Clin Nucl Med 2019; 44:214–220.

Liu

Wang

, et al. Near-infrared fluorescent peptides with high tumor selectivity: novel probes for image-guided surgical resection of orthotopic glioma. Mol Pharm 2019; 16:108–117.

Teicher

Holden

Ara

, et al. Influence of an anti-angiogenic treatment on 9L gliosarcoma: oxygenation and response to cytotoxic therapy. Int J Cancer 1995; 61:732–737.

Chang

Zhang

Guo

, et al. Multimodal imaging patterns predict survival in recurrent glioblastoma patients treated with bevacizumab. Neuro Oncol 2016; 18:1680–1687.

Matsumoto

Batra

Saito

, et al. Antiangiogenic agent sunitinib transiently increases tumor oxygenation and suppresses cycling hypoxia. Cancer Res 2011; 71:6350–6359.

Winkler

Kozin

Tong

, et al. Kinetics of vascular normalization by VEGFR2 blockade governs brain tumor response to radiation: role of oxygenation, angiopoietin-1, and matrix metalloproteinases. Cancer Cell 2004; 6:553–563.

Van Cutsem

Lambrechts

Prenen

, et al. Lessons from the adjuvant bevacizumab trial on colon cancer: what next? J Clin Oncol 2011; 29:1–4.

Chaudhry

O’Donovan

Brenchley

, et al. Vascular endothelial growth factor expression correlates with tumour grade and vascularity in gliomas. Histopathology 2001; 39:409–415.

10.

Amelio

Inoue

Markert

, et al. TAp73 opposes tumor angiogenesis by promoting hypoxia-inducible factor 1alpha degradation. Proc Natl Acad Sci U S A 2015; 112:226–231.

11.

Degroot

Fuller

Kumar

, et al. Tumor invasion after treatment of glioblastoma with bevacizumab: radiographic and pathologic correlation in humans and mice. Neuro Oncol 2010; 12:233–242.

12.

Gerstner

Emblem

Sorensen

GA.

Vascular magnetic resonance imaging in brain tumors during antiangiogenic therapy-are we there yet?

Cancer J 2015; 21:337–342.

13.

Wei

, et al. The antitumour growth and antiangiogenesis effects of xanthatin in murine glioma dynamically evaluated by dynamic contrast-enhanced magnetic resonance imaging. Phytother Res 2019; 33:149–158.

14.

Hou

Pan

, et al. Dynamic contrast-enhanced magnetic resonance imaging for monitoring the anti-angiogenesis efficacy in a C6 glioma rat model. Acta Radiol 2019. DOI: 10.1177/0284185119887598

15.

Bao

Wang

Zhang

, et al. Early monitoring antiangiogenesis treatment response of Sunitinib in U87MG Tumor Xenograft by (18)F-FLT MicroPET/CT imaging. Biomed Res Int 2014; 2014:218578.

16.

Zhao

Liu

, et al. Gemstone spectral imaging dual-energy computed tomography: a novel technique to determine urinary stone composition. Urology 2013; 81:727–730.

17.

Liang

Ren

Gao

, et al. Iodine concentration in spectral ct: assessment of prognostic determinants in patients with gastric adenocarcinoma. AJR Am J Roentgenol 2017; 209:1033–1038.

18.

Yamada

Jinzaki

Tanami

, et al. Virtual monochromatic spectral imaging for the evaluation of hypovascular hepatic metastases: the optimal monochromatic level with fast kilovoltage switching dual-energy computed tomography. Invest Radiol 2012; 47:292–298.

19.

Pomerantz

Kamalian

Zhang

, et al. Virtual monochromatic reconstruction of dual-energy unenhanced head CT at 65-75 keV maximizes image quality compared with conventional polychromatic CT. Radiology 2013; 266:318–325.

20.

Huang

Chen

, et al. Spectral CT evaluation of interstitial brachytherapy in pancreatic carcinoma xenografts: preliminary animal experience. Eur Radiol 2014; 24:2167–2173.

21.

Tang

Zhang

Sun

, et al. Spectral CT in the demonstration of the gastrocolic ligament: a comparison study. Surg Radiol Anat 2013; 35:539–545.

22.

Dong

Meng

, et al. Risk of adverse vascular events in patients with malignant glioma treated with bevacizumab plus irinotecan: a systematic review and meta-analysis. World Neurosurg 2019;130:e236–e243.

23.

Cárdenas

Howison

Pagel

MD.

A linear algorithm of the reference region model for DCE-MRI is robust and relaxes requirements for temporal resolution. Magn Reson Imaging 2013; 31:497–507.

24.

Moestue

Huuse

Lindholm

, et al. Low-molecular contrast agent dynamic contrast-enhanced (DCE)-MRI and diffusion-weighted (DW)-MRI in early assessment of bevacizumab treatment in breast cancer xenografts. J Magn Reson Imaging 2013; 38:1043–1053.