DOI:
Original papers
DOI: 10.11152/mu-1234Correlation between Contrast-Enhanced Ultrasound and Microvessel Density via CD31 and CD34 in the rabbit VX2 lung peripheral tumor model
Jin Xing
1, Wen He
1, Yi-Wen Ding
2, Yang Li
2, Yan-Dong Li
31Department of Ultrasound, Beijing Tian Tan Hospital, 2Department of Ultrasound, Fuxing Hospital, 3Department of Pathology, Fuxing Hospital, Capital Medical University, Beijing, China
Received 31.08.2017 Accepted 4.11.2017 Med Ultrason
2018, Vol. 20, No 1, 37-42
Corresponding author: Wen He, PhD
Department of Ultrasound, Beijing Tian Tan Hospital, Capital Medical University, 6 Tian Tan Xi Li Street, Dong Cheng District, Beijing, 100000, China
Phone: +86 01067098885, Fax: +86 01067098885 E-mail: [email protected]
Introduction
Lung cancer is the most common cause of death among malignant tumors around the world. Data has shown that lung cancer accounted for 19.4% of all cancer deaths in 2012 [1]. Its five-year survival rate after treat- ment is as low as 10% [2]. Computed tomography (CT) is the most common method for detecting lung cancer.
Ultrasound has limitations due to the reflection of the gas and ribs block. However, when lung cancer grows close to the pleura, ultrasound (US) can detect tumor struc- tures and blood flow from the rib space [3]. On the so- nogram, most peripheral lung carcinomas are round and polycyclic. These reveal an infiltrating growth or central necrosis. Invasion into adjacent structures of the chest wall is a very reliable sign of malignant growth [4].
Angiogenesis, the formation of new blood vessels from the endothelium of the existing vasculature, is a critical factor for tumor growth and metastasis in many human tumors [5,6]. Various studies have evaluated tu- mor angiogenesis through noninvasive imaging meth- ods such as contrast-enhanced dynamic CT, contrast- enhanced dynamic magnetic resonance imaging (MRI), non-enhanced magnetic resonance angiography, positron emission tomography, and Doppler ultrasonography [7- Abstract
Aim: To evaluate the tumor angiogenesis in lung peripheral VX2 tumor model by contrast-enhanced ultrasound (CEUS) and to determine the correlation between CEUS parameters and microvessel density (MVD) calculated via CD31 and CD34 expression. Materials and methods: VX2 pulmonary tumors were created in eight Japanese white rabbits by implanting a VX2 sarcoma into the lower portion of the right lung through ultrasound guidance. Tumors were allowed to grow for 14-21 days to achieve a diameter of 7-15 mm, and were examined by CEUS using a SonoVue contrast agent. The results were re- corded as digital video images, and the time-intensity curves and hemodynamic parameters were analyzed. Pathological tumor specimens were immediately obtained after the ultrasound examinations. Tumor specimens were stained with hematoxylin and eosin (H&E) and expressed as CD31 and CD34. The different endothelial cell markers were determined by immunohisto- chemical staining. MVD was calculated via CD 31 and CD34, and the relationship between CEUS parameters and MVDs was analyzed. Results: Two distinct types of microvessels were identified in lung peripheral VX2 tumors: differentiated (CD34+) and undifferentiated (CD31+) vessels. A significant correlation was found between CEUS parameters and undifferentiated MVD (CD31+ vessels) in lung peripheral VX2 tumors (p<0.05). A reverse correlation was observed between different MVDs.
Conclusions: Two different degrees of differentiation of vascular endothelial cells (CD31 and CD34) exist in the rabbit lung peripheral VX2 tumor model. CD31 MVD can more effectively evaluate tumor angiogenesis compared with CD34 MVD.
CEUS, as a non-invasive imaging method, can effectively evaluate tumor angiogenesis in rabbit peripheral lung cancer.
Keywords: contrast-enhanced ultrasonography; VX2 tumor; lung cancer; microvessel density; peak intensity
9]. However, these methods are limited to detecting low flow in small vessels. Contrast-enhanced ultrasonogra- phy (CEUS) increases the sensitivity to low flow, offer- ing detailed information on tumor vascularity [10,11].
Microvessel density (MVD) is an independent prog- nostic indicator, and has been considered the gold stand- ard for evaluating tumor angiogenesis [12]. Different blood vessel makers have been used to reveal the dif- ferent aspects and characteristics of tumor vasculature.
CD31 is usually expressed in both differentiated and un- differentiated endothelial cells, and CD34 is expressed in differentiated endothelial cells [13].
The VX2 tumor model is the most widely used ani- mal model at present. Many studies have discussed the application of CT in the pulmonary tumor model [14-16], but few have dwelled on US and CEUS. The correlation of CEUS parameters with different blood vessel markers have not been discussed in rabbit VX2 lung peripheral tumor models. The object of this study was to assess vas- cularity measurements via noninvasive quantified CEUS, and correlate CEUS parameters with different MVDs through immunohistochemical analysis.
Materials and methods
Animals, tumor model and protocol
The protocol for this study was approved by the re- search Ethics Committee of the Capital Medical Uni- versity. Twelve Japanese white rabbits, which were approximately three months old, were implanted with VX2 sarcoma. The average weight of these rabbits was 2.59±0.14 kg. The tumor implantation procedure started with a right intercostal incision between the seventh and eighth ribs. A piece of VX2 tumor tissue (approximately 1 mm3) was inserted retrograde into the distal portion of an 18-gauge injection needle (BARD, USA). The needle was punctured vertically into the chest wall at a depth of approximately 1 cm from intercostal incision. Then, the tumor was ejected into the pulmonary parenchyma. After the needle was withdrawn, the pinhole was immediately pressed with a gauze to prevent bleeding and pneumo- thorax. The animals were scanned by ultrasound at 14 days after implantation to monitor tumor growth. The VX2 tumor nodules were allowed to grow for 14-21 days to achieve an approximate size of 7-15 mm in diameter before these rabbits were used for the study. Successful tumor growth was obtained at 66.7% (8 of 12 rabbits) of the implantation sites.
US and CEUS
US imaging was performed using an Acuson Sequoia 512 machine equipped with a 15L8 linear array probe operated at a transmission frequency of 14 MHz. The
procedure was operated by a single sonographer with 12 years experience. The rabbits were placed in a supine po- sition and anesthetized by injecting 3% pentobarbital so- dium from the ear margin at a dose of 1.0 ml/kg. US pa- rameters including acoustic gain, depth, and focus were optimized for each tumor. The tumors were first scanned by gray scale US, followed by color Doppler US in order to obtain the best imaging plane.
Following standard US the tumors were examined by real-time CEUS after injection of the intravascular tracer SonoVue (Bracco SpA, Milan, Italy), which comprise of phospholipid-stabilized microbubbles of sulfur hex- afluoride with a mean diameter of 2.5 µm and a concen- tration of 1-5×108/mL [17]. The microbubble can freely flow through the tissues of tiny capillaries, but cannot enter the tissue space. Prior to injection, 5 mL of saline solution was added to the lyophilized powder under a sulfur hexafluoride atmosphere, and shaken thoroughly.
The contrast agent (0.2 ml/kg) was injected via an ear margin vein as a bolus within 2-3 seconds, followed by a saline bolus of 2 mL. Continuous scanning was started shortly before injecting the ultrasound contrast agent, which lasted for 90 seconds with a low mechanical index (MI) of 0.16. During the whole process, the operator kept the transducer stable. All CEUS images were recorded as digital video images.
Tumor enhancement was quantified offline using the time–signal intensity curve analysis software installed on the Acuson Sequoia 512. This software can display the signal time-intensity curve in the regions of interest (ROI) during enhancement. Quantitative data were retro- spectively analyzed by two investigators who worked in- dependently. The region of interest (ROI) was manually and independently selected by the investigators to highly enhance the tumor regions and avoid the area of tumor necrosis. The size of the ROI was maintained at approxi- mately 2-3 mm diameters as much as possible. For each tumor, three ROIs were selected for plotting the contrast enhancement time-intensity curves, and the results were averaged. The following parameters of the time–intensity curves were noted: arrival time (AT), time to peak (TTP), baseline intensity (BI) and peak intensity (PI).
Immunohistochemistry
After the CEUS examination, rabbits were sacrificed by injecting an overdose of anesthetics, and the tumors were harvested for further histological investigation. The tumor was marked to make sure that it was consistent with the results of the CEUS analysis. Pathological specimens were fixed in 4% paraformaldehyde for 30 minutes, and blocked with 5% donkey serum for 30 minutes. Next, the sections were incubated overnight at 4°C with mouse anti-human CD31 monoclonal antibody and mouse anti-
Fig 1. Ultrasound aspect of the lung peripheral VX2 tumor. The lung peripheral VX2 tumor was scanned by gray scale ultra- sound (a) and color Doppler ultrasound (b).
Fig 2. Contrast-enhanced ultrasound examination: a) Time-in- tensity curve and b) ultrasound aspect.
Fig 3. Pathological finding of H&E in lung peripheral VX2 tumors:a) Large tumor cell proliferation and adjacent lung tissue atrophy (H&E×10); b) Tumor cells presented in different sizes and shapes, and blood vessels were seen between these tumor cells (H&E×40). Microvessel immunohistochemistry of lung pe- ripheral VX2 tumors: c) CD31 antibody (original magnification
×200); d) CD34 antibody (original magnification ×200).
human CD34 monoclonal antibody, respectively. Then, the sections were colored by diaminobenzidine.
The method for counting the MVD number was based on the study conducted by Weidner et al [18], that is, to first look for high density regions of angiogenesis under the low power field (100× total magnification), and switch to a high power field (200× total magnification) to count the number of microvessels. Five digital images were recorded per high power field. The mean of these five fields was calculated and regarded as the tumor’s MVD. The result was reviewed by a pathologist who was blinded to the experiment.
Statistical analysis
Continuous variables were reported as a mean±standard variation. Spearman correlation analy- sis was used to determine the correlation between PI of CEUS-, and the different MVDs (histological CD31 count or CD34 count). A p value <0.05 was considered statistically significant. Statistics were analyzed using the Statistical Package for the Social Sciences (SPSS 21.0;
SPSS Inc., Chicago, IL, USA).
Results
Ultrasonography and CEUS of lung peripheral VX2 tumors
The average size of the tumors was 14±7 mm × 11±4 mm after 14-21 days from implantation of the VX2 sar- coma through ultrasound guidance in the lower portion of the right lung. The gray scale US demonstrated that tumors with heterogeneous structure and low echo had no envelopes, but the boundaries remained clear. In color Doppler imaging, blood flow signals could be observed
around the tumor (fig 1) and the arterial blood flow spec- trum could be detected. The average peak flow rate was 13.5 cm/s, and the average resistance index (RI) was 0.46.
On CEUS images, lung peripheral VX2 tumors were illustrated as highly enhanced in peripheral tissues, and low or not enhanced in the center tissue. All eight tumor lesions were shown as “fast in’’ and “slow out” in the time-intensity curve. The mean value of PI of the lung peripheral VX2 tumor was 24.01±7.36 dB (fig 2).
Pathological finding of H&E in lung peripheral VX2 tumors
At low magnification (H&E×10), the manifestation revealed a large tumor cell proliferation and adjacent lung tissue atrophy. At high magnification (H&E×40), tumor cells manifested in different sizes and shapes, and blood vessels were observed between the tumor cells (fig 3 a,b).
Microvessels in lung peripheral VX2 tumors Anti-CD31 antibody staining and anti-CD34 anti- body staining were observed in lung peripheral VX2 tu- mors (fig 3 c,d).
Correlation between PI and MVD in lung peripheral VX2 tumors
The mean values for AT, TTP, EI, CD31-MVD and CD34-MVD are listed in Table I. The correlations among these parameters are listed in Table II. The positive corre- lation between PI and CD31 MVD values in lung periph- eral VX2 tumors was statistically significant (r=0.721, p<0.05). However, there was no correlation between PI and CD34 MVD values (r=-0.564, p>0.05). Furthermore, there was an inverse relationship between CD31 MVD and CD34 MVD (r=-0.774, p<0.05; fig 4).
Discussions
At present, the technology to determine MVD on bi- opsy specimens is the immunohistochemical staining of vascular endothelial cells. MVD is an independent prog- nostic indicator, and has been considered the gold stand-
ard for evaluating tumor angiogenesis [12]. Generally, the higher the tumor MVD counts, the higher the degree of malignancy, the worse the prognosis, and the shorter the survival time [13].
It has been confirmed that there are two different degrees of endothelial cells (CD31 and CD34) in the presence of renal tumors. CD31 is presented in undif- ferentiated and differentiated mature endothelial cells, while CD34 is only presented in differentiated mature vascular endothelial cells. Highly-expressed CD31 MVD may suggest a higher degree of malignancy and shorter survival time for patients. In contrast, highly-expressed CD34 MVD may indicate a lower degree of malignancy, but survival time is significantly longer [19-21].
Since the lung of rabbits is similar to humans in ana- tomical structure, rabbits have been widely used in the experimental study of lung cancer. In the present study, we used rabbits to study the angiogenesis in the periph- eral lung cancer model. The US detected the blood flow around the tumor and the arterial blood flow spectrum.
The average of RI was 0.46, which was very low. How- ever, this cannot be used to judge the situation of micro- circulation perfusion.
CEUS is a new method for evaluating microcircula- tion perfusion [22]. It can be used to quantitatively, reli- ably and sensitively measure the blood perfusion of tu- mors [23], and evaluate the angiogenesis and biological behavior of tumors [24]. CEUS has been widely used in the evaluation of angiogenesis in kidney, breast, prostate, and liver tumors [25-28]. In the present study, CEUS was used to investigate the formation of angiogenesis in lung tumors.
Previous studies have always chosen a tumor en- dothelial cell marker, either CD31 or CD34, as a single evaluation of angiogenesis, in order to correlate with CEUS quantitative parameters [29-31]. However, these studies did not further explore the degree of differentia- tion of tumor vascular endothelial cells. The correlation between different MVDs and CEUS in lung cancer needs further studies. In the present study, we selected MVD in two different endothelial markers (CD31 and CD34), and studied the correlation of both markers with the quantita- tive parameters of CEUS, respectively, as well as the cor- relation with CD31 and CD34. A significant correlation between CD31 MVD and CEUS parameter PI (r=0.721, Fig 4. Correlations between PI, CD31 MVD and CD34 MVD
of lung peripheral VX2 tumors. The significant positive corre- lation between PI and CD31 MVD, and the reverse correlation between CD31 MVD and CD34 MVD is shown.
Table I. The mean value of contrast-enhanced parameters and microvessel density
Variable Mean value
AT (s) TTP (s) PI (dB) CD31CD34
1.41±0.95 12.85±6.88 24.15±7.07 25.63±4.69 15.38±2.20 AT – arrival time; TTP – time to peak; PI – peak intensity
Table II. The correlation among contrast-enhanced parameters and microvessel density in lung peripheral VX2 tumors.
Variable AT (s) TTP (s) PI (dB) CD31 CD34
Correlation CD31 0.275* -0.232* 0.721 – -0.774
(r) CD34 -0.471* 0.409* -0.564* -0.774 –
AT – arrival time; TTP – time to peak; PI – peak intensity; * – p>0.05
p<0.05) was demonstrated, while there was a non-linear correlation between CD34 MVD and CEUS parameters PI (r=-0.564, p>0.05). Other CEUS parameters such as AT and TTP had no significant correlation with MVDs.
Statistical analysis results indicate that CEUS quantita- tive parameters can effectively evaluate angiogenesis conditions in rabbit peripheral lung cancer and CD31 MVD better reflect the status of tumor blood vessels.
Another result is that CD31 MVD negatively correlated with CD34 MVD (r=-0.774, p<0.05). This indicates that when differentiated mature endothelial cells dominate (CD34 MVD presents a high expression) in the same tumor tissue, the number of undifferentiated endothelial cells is relatively smaller with a relatively lower degree of malignancy, and vice versa.
The limitation of this experiment was that the VX2 tu- mors were prone to necrosis, which can lead to the reduc- tion of active tumors. Hence, it is important to choose the timing of the examination. Furthermore, the sample size for the present study was slightly small. A larger sample size is needed for further research. Moreover, the prog- nosis of the tumor was not discussed in the present study.
Conclusion
CD31 MVD can more effectively evaluate tumor an- giogenesis compared with CD34 MVD. CEUS, as a non- invasive imaging method, can effectively evaluate tumor angiogenesis in rabbit peripheral lung cancer.
Acknowledgement: The authors are grateful to Pro Dan Zhang and Weidong Chen for technical assistance during this study.
Source of funding: This work was funded by the Na- tional Natural Science Foundation of China as part of the project “Import microwave on curing lung cancer – basic research”.
Conflict of interest: none
References
1. GLOBOCAN 2012: estimated cancer incidence, mortality and prevalence worldwide in 2012. Accessed at: http://glo- bocan.iarc.fr/Pages/fact_sheets_cancer.aspx.
2. Berrino F, De Angelis R, Sant M, et al. Survival for eight major cancers and all cancers combined for European adults diagnosed in 1995-99: results of the EUROCARE-4 study. Lancet Oncol 2007;8:773-783.
3. Yuan A, Chang DB, Yu CJ, Kuo SH, Luh KT, Yang PC.
Color Doppler sonography of benign and malignant pulmo- nary mass. AJR Am J Roentgenol 1994;163:545-549.
4. Mathis G. Thoracic ultrasound in subpleural lung consoli- dations. Atemwegs Lungenkr 2010;36:366-371.
5. Brown LF, Berse B, Jackman RW, et al. Expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in breast cancer. Hum Pathol 1995;26:86-91.
6. Wang Z, Tang J, An L, et al. Contrast-enhanced ultrasonog- raphy for assessment of tumor vascularity in hepatocellular carcinoma. J Ultrasound Med 2007;26:757-762.
7. Wang B, Gao ZQ, Yan X. Correlative study of angiogen- esis and dynamic contrast-enhanced magnetic resonance imaging features of hepatocellular carcinoma. Acta Radiol 2005;46:353-358.
8. Li WW. Tumor angiogenesis: molecular pathology, thera- peutic targeting, and imaging. Acad Radiol 2000;7:800- 9. Meyerowitz CB, Fleischer AC, Pickens DR, et al. Quanti-811.
fication of tumor vascularity and flow with amplitude color Doppler sonography in an experimental model: preliminary results. J Ultrasound Med 1996;15:827-833.
10. Wilson SR, Burns PN, Muradali D, Wilson JA, Lai X. Har- monic hepatic US with microbubble contrast agent: initial experience showing improved characterization of heman- gioma, hepatocellular carcinoma, and metastasis. Radiol- ogy 2000;215:153-161.
11. Masaki T, Ohkawa S, Amano A, Ueno M, Miyakawa K, Tarao K. Noninvasive assessment of tumor vascularity by contrast-enhanced ultrasonography and the prognosis of patients with nonresectable pancreatic carcinoma. Can- cer 2005;103:1026-1035.
12. Folkman J. Angiogenesis and breast cancer. J Clin Oncol 1994;12:441-443.
13. Poblet E, Gonzalez-Palacios F, Jimenez FJ. Different im- munoreactivity of endothelial markers in well and poorly differentiated areas of angiosarcomas. Virchows Arch 1996;428:217-221.
14. Okuma T, Matsuoka T, Okamura T, et al. 18F-FDG small- animal PET for monitoring the therapeutic effect of CT- guided radiofrequency ablation on implanted VX2 lung tumors in rabbits. J Nucl Med 2006;47:1351-1358.
15. Kuroki M, Nakada H, Yamashita A, et al. Loss of cellu- lar viability in areas of ground-glass opacity on computed tomography images immediately after pulmonary radiofre- quency ablation in rabbits. Jpn J Radiol 2012;30:323–330.
16. Jin GY, Han YM, Lim YS, Jang KY, Lee SY, Chung GH.
Percutaneous Radiofrequency Thermal Ablation of Lung VX2 Tumors in a Rabbit Model: Evaluation with Helical CT Findings for the Complete and Partal Ablation. J Ko- rean Radiol Soc 2004;50:343-351.
17. Ferrara KW, Merritt CR, Burns PN, Foster FS, Mattrey RF, Wickline SA. Evaluation of tumor angiogenesis with US: imaging, Doppler, and contrast agents. Acad Radi- ol 2000;7:824-839.
18. Weidner N. Intratumor microvessel density as a prognostic factor in cancer. Am J Pathol 1995;147:9-19.
19. Imao T, Egawa M, Takashima H, Koshida K, Namiki M. In- verse correlation of microvesseldensity with metastasis and
prognosis in renal cell carcinoma. Int J Urol 2004;11:948- 20. Rioux-Leclercq N, Epstein JI, Bansard JY, et al. Clinical 953.
significance of cell proliferation, microvessel density, and CD44 adhesion molecule expression in renal cell carcino- ma. Hum Pathol 2001;32:1209-1215.
21. Yildiz E, Ayan S, Goze F, Gokce G, Gultekin EY. Rela- tion of microvessel density with microvascular inva- sion, metastasis and prognosis in renal cell carcinoma. BJU Int 2008;101:758-764.
22. Metoki R, Moriyasu F, Kamiyama N, et al. Quantification of hepatic Parenchymal blood flow by contrast ultrasonog- raphy with flash-replenishment imaging. Ultrasound Med Biol 2006;32:1459-1466.
23. Zhuang H, Yang ZG, Chen HJ, Peng YL, Li L. Time-inten- sity curve parameters in colorectal tumours measured using double contrast-enhanced ultrasound: correlations with tu- mour angiogenesis. Colorectal Dis 2012;14:181-187.
24. Hwang M, Hariri G, Lyshchik A, Hallahan DE, Fleischer AC. Correlation of quantified contrast-enhanced sonog- raphy with in vivo tumor response. J Ultrasound Med 2010;29:597-607.
25. Kabakci N, Igci E, Secil M, et al. Echo contrast-enhanced power Doppler ultrasonography for assessment of an-
giogenesis in renal cell carcinoma. J Ultrasound Med 2005;24:747-753.
26. Du J, Li FH, Fang H, Xia JG, Zhu CX. Correlation of real- time gray scale contrast-enhanced ultrasonography with microvessel density and vascular endothelial growth factor expression for assessment of angiogenesis in breast lesions.
J Ultrasound Med 2008;27: 821-831.
27. Shigeno K, Igawa M, Shiina H, Kishi H, Urakami S. Tran- srectal colour Doppler uhrasonography for quantifying an- giogenesis in prostatic cancer. BJU Int 2003;91: 223-226.
28. Xia Y, Jiang YX, Dai Q, Xiao Y, Lv K, Wang L. Contrast- enhanced ultrasound of hepatocellular carcinoma: Correla- tion of washout time and angiogenesis. Clin Hemorheol Microcirc 2011;48:265-273.
29. Zhang X, Yang G, Zhang Y, et al. An experimental research into endostatin microbubble combined with focused ultra- sound for anti-tumor angiogenesis in colon cancer. Gastro- enterol Rep (Oxf) 2014;2:44–53.
30. Wang Z, Tang J, An L, et al. Contrast-enhanced ultrasonog- raphy for assessment of tumor vascularity in hepatocellular carcinoma. J Ultrasound Med 2007;26:757-762.
31. Lucidarme O, Kono Y, Corbeil J, et al. Angiogenesis: nonin- vasive quantitative assessment with contrast-enhanced func- tional US in murine model. Radiology 2006;239:730-739.
