Original Article: Clinical Investigation
Vasculogenic mimicry plays an important role in adrenocortical carcinoma
Faming Zhang,1,1 Hao Lin,1,f Kaiyuan Cao,2 Hua Wang,1 Jincheng Pan,1 Jintao Zhuang,1 Xu Chen,1 Bin Huang,1 Daohu Wang1 and Shaopeng Qiu1
1Department of Urology, The First Affiliated Hospital, and 2Research Center for Clinical Laboratory Standard, Zhongshan Medical School, Sun Yat-sen University, Guangzhou, Guangdong, China
Abbreviations & Acronyms
ACC = adrenocortical carcinoma
CCK-8 = Cell Counting Kit-8 OS = overall survival PAS = periodic acid-Schiff PBS = phosphate-buffered saline qPCR = quantitative polymerase chain reaction siRNA = small interfering ribonucleic acid TKI = tyrosine kinase inhibitor
VEGF = vascular endothelial growth factor VEGFR-2 = vascular endothelial growth factor receptor 2 VM = vasculogenic mimicry
Correspondence: Shaopeng Qiu M.D., Ph.D., Department of Urology, The First Affiliated Hospital, Sun Yat-sen University, No. 58, Zhongshan 2nd Road, Guangzhou, Guangdong 510080, China. Email: qiushp@mail.sysu.edu.cn
¡These authors contributed equally to this work.
Received 9 April 2015; accepted 20 January 2016. Online publication 24 February 2016
Objectives: To determine the prognostic role of vasculogenic mimicry in adrenocortical carcinoma, and to explore its relationship with vascular endothelial growth factor receptor 2 expression.
Methods: A total of 46 samples of adrenocortical carcinoma were collected and reviewed. Vasculogenic mimicry and vascular endothelial growth factor receptor 2 were detected by immunohistochemistry and double staining. Survival analysis was carried out to access pronostic significance. Three-dimensional culture method was applied to test the ability of vasculogenic mimicry formation by adrenocortical carcinoma cell lines SW- 13 and H295R. Quantitative polymerase chain reaction and western blotting were used to monitor the expression of vascular endothelial growth factor receptor 2 in SW-13 and H295R. After being treated with specific inhibitor or small interfering ribonucleic acid to downregulate expression of vascular endothelial growth factor receptor 2, vasculogenic mimicry formation and cell proliferation of SW-13 cells were evaluated by 3-D culture and Cell Counting Kit-8 methods.
Results: Vasculogenic mimicry was observed in 19 of the 46 (41.30%) adrenocortical carcinoma samples. Both vasculogenic mimicry and vascular endothelial growth factor receptor 2 expressions showed a positive association with Weiss score and TNM stage, whereas vascular endothelial growth factor receptor 2 was also associated with tumor size (all P < 0.05). Vasculogenic mimicry was closely correlated with vascular endothelial growth factor receptor 2 expressions (r = 0.470, P < 0.01). The median overall survival of patients with vasculogenicmimicry-positive or vascular endothelial growth factor receptor 2-positive was shorter than that of patients with vasculogenic mimicry-negative or vascular endothelial growth factor receptor 2-negative (P = 0.001 and 0.028, respectively). The vasculogenic mimicry-forming SW-13 cells expressed higher levels of vascular endothelial growth factor receptor 2 than that of H295R, which was unable to form vasculogenic mimicry on Matrigel. However, downregulation of vascular endothelial growth factor receptor 2 only decreased cell proliferation, but not vasculogenic mimicry formation by SW-13 cells.
Conclusions: Vasculogenic mimicry and overexpression of vascular endothelial growth factor receptor 2 seem to correlate with poor prognostic outcomes in adrenocortical carcinoma. Anti-angiogenesis treatments targeting vascular endothelial growth factor receptor 2 should be combined with therapies targeting vasculogenic mimicry in adrenocortical carcinoma.
Key words: adrenocortical carcinoma, anti-angiogenesis therapy, mitotane, vascular endothelial growth factor receptor 2, vasculogenic mimicry.
Introduction
Primary ACC is rare, but highly aggressive, with an incidence of approximately 1-2/ 100 000.1 At present, ACC can only be potentially cured by operation, whereas a certain pro- portion of patients lose their chance to receive radical excision as a result of disseminated dis- eases.2 Furthermore, most cases relapse even after complete tumor removal.3 Currently, although combination chemotherapy with mitotane is recommended for the treatment of advanced ACC, the efficacy achieved is not satisfactory, and more effective treatment approaches are urgently required.
Angiogenesis refers to new blood capillaries originating from pre-existing microvessels. Abnormal regulation of angiogenesis was found to be involved in the pathogenesis of several disorders including cancer.4 Anti-angiogenesis therapy therefore, was considered a promising treatment strategy for cancer. Although good tolerance and significant survival pro- longation of anti-angiogenesis treatments in patients with ad- vanced renal, lung, and breast cancers have been reported,5-7 the preliminary results of regimens targeting neovessels in ACC were largely disappointing,8,9 and the underling mecha- nisms remained mostly unknown.
VM was first described by Maniotis et al. as a new blood supply system formed independently of normal vessel endothelial cells in melanoma.1º Subsequently, VM was found in a spectrum of cancers and associated with poor prognosis in some of them.11 van der Schaft et al. even showed that VM might be one of the alternative mecha- nisms helping tumor cells escape from anti-angiogenesis treatment.12 However, research regarding VM in ACC is currently unavailable.
Recently, Scully et al. showed that VEGFR-2 was one of the most important factors that govern the process of VM and tumor development in glioblastomas.13 Furthermore, VEGFR-2 was found to be overexpressed in ACC.14 There- fore, the present study was designed to determine the role of VM in the clinicopathology and prognosis of ACC, and to explore the relationship between VM and VEGFR-2 in ACC.
Methods
Patient selection and tissue sample collection
The present study was approved by the medical ethics com- mittee of Sun Yat-sen University, and written informed con- sent was obtained from each patient for surgery and research purposes. Patients were restricted to those who have underwent complete tumor resection and postoperative pathological diagnosis of ACC between January 1999 and March 2014 in the First Affiliated Hospital of Sun Yat-sen University (n = 46, median age 51 years). The information of patients’ age, sex, hormone secretion, tumor location, pri- mary tumor size, Weiss score,15 stage at diagnosis16 and follow-up data were collected for analysis. Archived ACC tissue samples from these patients were used for VM and VEGFR-2 detection by immunohistochemistry.
Patients were followed up by clinic interview or phone call. The total followed up period was from 1 to 80 months (median time 13.5 months). OS time was cacu- lated as the duration from the date of surgery to the date of death.
Cell culture
Two human ACC cell lines, SW-13 and H295R, were pur- chased from American Type Culture Collection (Manassas, VA, USA) and maintained in Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY, USA) containing 10% fetal bovine serum (Gibco) and 4.5 g of glucose at 37°℃ in a 5% CO2 atmosphere.
Immunohistochemical staining
Paraffin-embedded specimens were cut into 5-um serial sec- tions, dewaxed twice in xylene, rehydrated by serial concentra- tions of ethanol and then rinsed in PBS. After being heated for 15 min for antigen retrieval, the sections were incubated with endogenous peroxidase sealing fluid to inactivate endogenous peroxidase, and with 10% goat serum at room temperature for 10 min to block non-specific reactions. The sections were incu- bated with polyclonal rabbit anti-human VEGFR-2 antibody (1:50 dilution, Abcam, Cambridge, MA, USA) or monoclonal rabbit anti-human CD34 antibody (1:100 dilution; Abcam) at 4℃ for 12 h. The sections were washed with PBS and incubated by SP Rabbit & Mouse HRP kit (CWBIO, Beijing, China) at 37°℃ for 30 min, then developed in diaminobenzidine substrate, counterstained in hematoxylin, and dehydrated in ethanol and xylene before being mounted. The sections incubated with PBS, instead of primary antibodies, were used as the negative control.
Two independent pathologists observed the distribution, staining intensity and positive ratio of VEGFR-2/CD34 expres- sion. The staining was classified as follows: no staining scored 0; faint or moderate staining in ≤25% of tumor cells scored 1; moderate or strong staining in 25-50% of tumor cells scored 2; and strong staining in ≥50% of tumor cells scored 3. For each sample, four randomly selected areas were observed under high magnification, and 100 tumor cells in each area were counted to calculate the proportion of positive cells. Positively high expres- sion was defined as staining index ≥3. Low expression was defined as staining index <3, accordingly.
CD34/PAS double staining
CD34 immunohistochemical staining was applied, then the sections were treated with 0.5% periodic acid solution for 10 min and rinsed in distilled water for 3 min. In a dark chamber, these sections were treated with Schiff solution for 30 min. After distilled water rinsing, the sections were coun- terstained with hematoxylin. In accordance with Zhang et al., specimens with the presence of typical VM structures were considered VM-positive by whole slide examination.17
3-D culture
SW-13 and H295R cells were dissociated and resuspended at 5 × 104 cells/mL. Each well of 96-well tissue culture plates was coated with 50 uL Matrigel, which was allowed to poly- merase at 37℃ for 30 min. The indicated cell suspension was then plated at 100 µL/well onto the surface of Matrigel, and incubated at 37℃ for 1-4 days. The tube formation of cells was observed under light microscope with a magnification of ×200. The numbers of complete tubular structures from three randomly chosen fields were counted. The mean value of the three read- ings was used as the final reading of the well.
Inhibition of VEGFR-2 by VEGFR TKI II
Cells were starved overnight. Then cells were harvested and resuspended with Dulbecco’s modified Eagle’s medium
with different concentrations of VEGFR TKI II (1-40 nmol/L; Millipore, Billerica, MA, USA) and seeded on wells coated with Matrigel at a density mentioned previ- ously. The control groups were treated with 0.1% dimethylsulfoxide instead.
RNA interference to knock down VEGFR-2
SW-13 cells were transfected with siRNA against VEGFR-2 or negative control siRNA with a concentration of 100 nmol/ L by Lipofectamine 2000 and Opti-MEM I (Gibco) according to the manufacturer’s protocol. All siRNA were purchased from RiboBio, Guangzhou, China. Briefly, 2 × 105 cells were plated in each well of six-well plates and cultured to reach a 90% confluence. Cells were then transfected with siRNA by using the transfection reagent in serum-free med- ium. Total RNA and protein were isolated at 36 and 72 h after transfection, respectively. VEGFR-2 expression was monitored by qPCR and western blotting.
qPCR
Total RNA was isolated using a column total RNA extraction kit (Invitrogen, Grand Island, NY, USA) according to the manufac- turer’s protocol, and subjected to reverse transcription using reverse transcriptase of First Strand cDNA Synthesis kit (Invitrogen). The RNA concentration was 1000 ng/uL. Then amplification was car- ried out in a total volume of 10 µL using SYBR Premix Ex Taq kit (Takara, Shiga, Japan). Primer sequences were as followed: 5’- GCTTTGGCCCAATAATCA GA-3’ (forward) and 5’-AC ACGACTCCATGTTGGTCA-3’ (reverse) for VEGFR-2; 5’-TGT- TCGTCATGGGTGTGAAC-3’ (forward) and 5’-ATGGCATG- GACTGTGGT CAT-3’ (reverse) for the internal control glycer- aldehydes-3-phosphatedehydrogenase. PCR was repeated in tripli- cate.
Western blotting
Briefly, total proteins were extracted from the cells and dena- tured in a sodium dodecyl sulfate sample buffer, then equally loaded onto 8% polyacrylamide gel. After electrophoresis, the proteins were transferred to a polyvinylidene difluoride mem- brane. Blots were incubated with primary rabbit polyclonal anti-VEGFR-2 antibody (1:1000 dilution) and anti-ß-actin (1:2000 dilution) overnight at 4℃, and detected with horse- radish peroxidase-conjugated secondary antibody (dilution 1:10 000). All antibodies were purchased from Abcam.
Cell proliferation assay
Cell proliferation was tested by CCK-8 (Dojindo, Kumamoto, Japan). SW-13 cells were seeded into 96-well plates at 4000 cells/100 µL/well and transfected with siVEGFR-2 or negative control, followed by incubation at 37°℃ in an envi- ronment with 5% CO2 for 48 h, 10 µL of CCK-8 solutions were added to each well, followed by incubation for 2 h. Optical density was measured at 450 nm with a microplate reader. All experiments were carried out in triplicate.
Statistical analysis
Statistical analysis was carried out using the spss version 21.0 (SPSS, Chicago, IL, USA). Comparisons of quantitative data were analyzed by t-test or ANOVA. The association between VM and the clinicopathological features of ACC was ana- lyzed by Pearson’s x2-test or Fisher’s exact test. Kaplan- Meier survival plots were generated, and comparisons between survival curves were made with the log-rank statis- tic. All tests were two-sided. P-values <0.05 were considered statistically significant.
Results
Evidence of VM in ACC
Different from those vessels with endothelial cells stained positive for CD34 (Fig. 1a), the existence of VM was defined as CD34-negative/PAS-positive structures sur- rounded by tumor cells (not endothelial cells) with or with- out red blood cells in it through whole slide examination. A total of 46 ACC patients were included in this study and VM was observed in 19 (41.30%) of the 46 ACC cases (Fig. 1b-d).
Clinical significance of VM and VEGFR-2 in ACC
The association of VM and VEGFR-2 with clinicopathological parameters is summarized in Table 1. In the present study, VM was significantly associated with clinical stage and Weiss Score (all P < 0.05). A total of 35 out of 46 specimens (76.09%) were classified as VEGFR-2-positive (Fig. 1e). A significant associa- tion was observed between expressions of VEGFR-2 and TNM stage, Weiss Score and tumor size (all P < 0.05).
However, both VM and expressions of VEGFR-2 showed no association with sex, age, location or hormone secretion (all P > 0.05). In the 46 ACC samples, all the VM-positive samples were, meanwhile, VEGFR-2-positive (19/35, 54.29%), whereas none of the VEGFR-2-negative specimens (Fig. 1f) presented the CD34-negative/PAS-positive loops. Furthermore, the correlation analysis showed a significant positive correlation between VEGFR-2 expressions and VM (r = 0.470, P < 0.01).
Survival analysis of VEGFR-2 and VM
In the present study, survival analysis showed that patients who were VEGFR-2-positive showed shorter OS than those who were VEGFR-2-negative (P = 0.028; Fig. 1g). The median OS time was 11 ± 1.38 months (95% CI 8.30-13.70 months) for VEGFR-2-positive patients, whereas the corresponding OS time was 23 ± 5.39 months (95% CI 12.43-33.6 months) for VEGFR2-negative patients. Consistently, patients with VM showed poorer prognosis for OS than those without VM (P = 0.001; Fig. 1h). The median OS time was 5.00 ± 2.90 months (95% CI 0.00-10.69 months) for VM-positive patients and 23.00 ± 3.75 months (95% CI 15.66-30.34 months) for VM negative patients, respectively.
(a)
(b)
Survival proportions
(g)
1.5
- VEGFR2 (+)
- L - VEGFR2 (-)
Fraction survival
1.0
0.5
(c)
(d)
0.0
0
20
40
60
80
100
Survival time (months)
(h)
Survival proportions
1.5
- VM (+)
(e)
(f)
Fraction survival
- L - VM (-)
1.0
0.5
0.0
0
20
40
60
80
100
Survival time (months)
VM formation and VEGFR-2 expressions in ACC cell lines SW-13 and H295R
To further explore the relationship between VM and VEGFR- 2 in ACC, we used two human ACC cell lines, SW-13 and H295R, in our subsequent in vitro experiments. By culturing on Matrigel, SW-13 cells formed a vasculogenic structure of cord- like networks, whereas H295R cells were unable to form the tubular networks (Fig. 2a,b). Those typical VM structures formed by SW-13 cells were visible 3 h after seeding under an inverted microscope, and persistently existed for up to 4 days.
Furthermore, the level of VEGFR-2 mRNA in SW-13 was much higher than that in H295R cells (Fig. 2c). Further investigation of the differential expression of VEGFR-2 genes was achieved by western blot analysis, as shown in Fig- ure 2d. Western blot analysis validated the VEGFR-2 expres- sion profile of SW-13 cell lines.
Effects of VEGFR-2 on VM formation and cell proliferation in SW-13 cells
In order to determine the role of VEGFR-2 in VM formation in ACC, SW-13 was first treated with VEGFR TKI II, which
inhibited VEGFR-2 at a concentration of 20 nmol/L accord- ing to the instructions. Interestingly, we found even 40 nmol/ L of inhibitor cannot disrupt VM structures formed by SW- 13 cells (Fig. 3a,b). To avoid the off-target effects of the inhibitor, we further designed siRNA to downregulate the expression of VEGFR-2 in SW-13. Efficiencies of the siRNA were tested by qPCR and western blotting (Fig. 4a,b). In accordance with the inhibitor, downregulation of VEGFR-2 by siRNA showed no effects on VM formation by SW-13 cells on Matrigel (Fig. 4c,e,f). However, siVEGFR-2 signifi- cantly decreased cell proliferation at 48 h after transfection (P < 0.05; Fig. 4d).
Discussion
Blood supply is essential for tumor growth and metastasis. Different from the widely accepted mechanisms, such as angiogenesis and vasculogenesis, VM refers to de novo gen- eration of a tumor circulation system independent of vessel endothelial cells. Since its discovery, VM has been observed, and has been shown to be novel and significant in a variety of tumors.11 However, to our knowledge, VM in ACC has
| Table 1 Associations between clinicopathological features of ACC and VM formation and VEGFR-2 expression | |||||||
|---|---|---|---|---|---|---|---|
| Characteristic | VEGFR-2 | VM | |||||
| n | Negative | Positve | P-value | Negative | Positve | P-value | |
| Sex | 0.883 | 0.878 | |||||
| Men | 26 | 6 | 20 | 15 | 11 | ||
| Women | 20 | 5 | 15 | 12 | 8 | ||
| Age (years) | 0.514 | 0.766 | |||||
| ≤48 | 17 | 5 | 12 | 11 | 6 | ||
| >48 | 29 | 6 | 23 | 16 | 13 | ||
| Tumor location | 0.270 | 0.273 | |||||
| Left | 18 | 3 | 15 | 9 | 9 | ||
| Right | 26 | 7 | 19 | 16 | 10 | ||
| Bilateral | 2 | 1 | 1 | 2 | 0 | ||
| Hormone secretion | 0.210 | 0.930 | |||||
| Present | 7 | 3 | 4 | 4 | 3 | ||
| Absent | 39 | 8 | 31 | 23 | 16 | ||
| Tumor size (cm) | 0.029 | 0.182 | |||||
| 3-7 | 10 | 5 | 5 | 4 | 6 | ||
| >7 | 36 | 6 | 30 | 23 | 13 | ||
| Weiss score | 0.002 | 0.003 | |||||
| 3-5 | 16 | 8 | 8 | 14 | 2 | ||
| >5 | 30 | 3 | 27 | 13 | 17 | ||
| TNM stage | 0.000 | 0.023 | |||||
| I + II | 10 | 8 | 2 | 9 | 1 | ||
| III + IV | 36 | 3 | 33 | 18 | 18 | ||
(a)
(b)
SW-13
H295R
(c)
1.5
(d)
SW-13
H295R
Arbitray units
1.0
1
VEGFR-2
0.5
**
0.1313
B-actin
0.0
SW-13
H295R
not been reported. In the present study, we confirmed the occurrence of VM in ACC. Furthermore, we found the inci- dence of VM was highly associated with tumor stage and Weiss Score. This finding might show that VM was more prevalent in ACC with higher malignancy, which was similar to that in other tumors.18-20 Actually, VM has been reported to be a signature of poor prognosis in malignant tumors, such as ovarian cancer, glioma, renal cell carcinoma and hepato- cellular carcinoma.17,21-23 In the present retrospective study of 46 patients with ACC, our findings were also consistent
with the aforementioned studies. Patients with VM formation showed a poorer clinical outcome than those without it; the median OS time was significantly shorter in the VM-positive group than that in the VM-negative group.
The recurrence in ACC is frequent and therefore, there is no doubt that effective adjuvant therapies would be of great benefit. However, current treatment regimens with mitotane failed in many patients with advanced ACC. Targeting at tumor neovessels has attracted more and more attention in ACC since the discovery of high expression of VEGF and its
(a)
(b)
0.1%DMSO
VEGFR-2 Kinase Inhibitor 40 nm
(a)
(b)
1.5
Arbitrary units
1.0
1
0.8971
NC
SİRNA#1
SiRNA#2 siRNA#3
0.5
VEGFR-2
0.2692
0.2986
0.0
ß-actin
NC SiRNA-1 siRNA-2 SiRNA-3 Expression of VEGFR2 after knockdown
(d) 0.6
0.4525
(c)
Cell viability
0.4
**
**
0.3285
0.3379
0.2
0.0
Control
Control
SİRNA#1
SİRNA#3
(e)
(f)
VEGFR-2 siRNA#1
VEGFR-2 siRNA#3
receptor VEGFR-2 in ACC.14 In accordance with that, we also found VEGFR-2 was highly expressed in 76.09% of ACC cases, and was more evident in the progressive stage of ACC than in the primary stage. Furthermore, patients with VEGFR-2-positive expressions also showed shorter OS time than those with VEGFR-2 negative expressions.
Surprisingly, very slight antitumor effects were observed in ACC by using those VEGFR-targeting TKI with adequate serum drug levels.9 Owing to the lack of endothelial-cell struc- ture and conjunction with the pre-existing circulation system, VM might facilitate disseminations as a new route for cancer cells, and was believed to cause the off-target effect of anti- angiogenesis therapy.24 Accordingly, we hypothesized that VM appearing in advanced ACC might be associated with TKI resistance. Exploration of VM in ACC should be of beneficial for finding improved treatment options for ACC.
To date, the mechanisms underlining VM formation are yet to be determined. VEGFR-2, as an essential angiogenic mediator to trigger signaling cascades induced by VEGF, was
reported to participate in forming VM structures in glioblas- toma.13 According to our research, VEGFR-2 was positively correlated with the formation of VM in ACC. However, it is noteworthy that just 46 patients were enrolled in our trial. All the VM-positive patients accidently fell into the group of patients with overexpression of VEGFR-2. Therefore, bias should be taken in consideration on the correlation within VM and VEGFR-2 in our ACC cohort. More cases would be required to validate their relationships and clinical signifi- cance in ACC.
To further confirm the relationship between VEGFR-2 and VM, we used two ACC cell lines in our in vitro experiments. Interestingly, although the more aggressive cell line, SW-13, was able to form a VM structure on Matrigel and expressed a higher level of VEGFR-2, the less aggressive cell line, H295R, did not show the VM-forming ability and exerted a lower expression of VEGFR-2. These results showed that VEGFR-2 might truly play an important part in VM forma- tion by ACC cells.
If VEGFR-2 was involved in VM formation in ACC, then TKI mainly targeted at VEGFR-2 should also be effective for disruption of VM. However, when downregulation of VEGFR-2 occurred by either a specific inhibitor or siRNA in SW-13, VM patterns were not disrupted, even though cell proliferation was significantly decreased. This finding was similar to that Vartanian et al. had reported in melanoma.25 They found VEGFR-2-specific inhibitor did not affect VM formed by melanoma cells. Accordingly, we deduced that VEGFR-2 might participate in angiogenesis, and thus show a poor prognosis in ACC; downregulation of VEGFR-2 expres- sion in SW-13 cells can abrogate cell proliferation, which was consistent with the results of reduced tumor growth in some of the ACC patients treated with TKI.9 However, VEGFR-2 might not be involved in VM formation by ACC cells. Therefore, VM in ACC could still serve as an alterna- tive nutrition supply system for tumor progression, even though VEGFR-2 was inhibited by TKI.
In conclusion, the VM presented in our ACC patients was associated with grade malignancy and disease progression as overexpression of VEGFR-2. Downregulation of VEGFR-2 in ACC could reduce cell proliferation, but not VM formation, which showed that therapies targeting VM should be considered in addition to anti-angiogenesis treatment for ACC.
Acknowledgments
This work was supported by grants from the China National Natural Science Foundation (81272809 and 81402116); Science and Technology Planning Project of Guangdong (2011B050400021); The Guangdong province of integration of production and research projects (2012B090600021), and Guangdong Key Laboratory of Urology; Medical Science Foundation of Guangdong province (A2012603); and The First Affiliated Hospital of Guangzhou Medical University (2010A060801016). The funders had no role in study design, data collection and analysis, decision to publish or prepara- tion of the manuscript.
Conflict of interest
None declared.
References
1 Golden SH, Robinson KA, Saldanha I, Anton B, Ladenson PW. Prevalence and incidence of endocrine and metabolic disorders in the United States: a comprehensive review. J. Clin. Endocrinol. Metab. 2009; 94: 1853-78.
2 Allolio B, Fassnacht M. Adrenocortical carcinoma: clinical update. J. Clin. Endocrinol. Metab. 2006; 91: 2027-37.
3 Fassnacht M, Allolio B. What is the best approach to an apparently non- metastatic adrenocortical carcinoma? Clin. Endocrinol. 2010; 73: 561-5.
4 Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat. Med. 1995; 1: 27-31.
5 Conti A, Santoni M, Amantini C et al. Progress of molecular targeted thera- pies for advanced renal cell carcinoma. Biomed Res. Int. 2013; 2013: 419176.
6 Sandler A, Gray R, Perry MC et al. Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N. Engl. J. Med. 2006; 355: 2542-50.
7 Ramaswamy B, Elias AD, Kelbick NT et al. Phase II trial of bevacizumab in combination with weekly docetaxel in metastatic breast cancer patients. Clin. Cancer Res. 2006; 12: 3124-9.
8 Kroiss M, Quinkler M, Johanssen S et al. Sunitinib in refractory adrenocorti- cal carcinoma: a phase II, single-arm, open-label trial. J. Clin. Endocrinol. Metab. 2012; 97: 3495-503.
9 O’Sullivan C, Edgerly M, Velarde M et al. The VEGF inhibitor axitinib has limited effectiveness as a therapy for adrenocortical cancer. J. Clin. Endocri- nol. Metab. 2014; 99: 1291-7.
10 Maniotis AJ, Folberg R, Hess A et al. Vascular channel formation by human melanoma cells in vivo and in vitro: vasculogenic mimicry. Am. J. Pathol. 1999; 155: 739-52.
11 Folberg R, Maniotis AJ. Vasculogenic mimicry. APMIS 2004; 112: 508-25.
12 van der Schaft DW, Seftor RE, Seftor EA et al. Effects of angiogenesis inhi- bitors on vascular network formation by human endothelial and melanoma cells. J. Natl Cancer Inst. 2004; 96: 1473-7.
13 Scully S, Francescone R, Faibish M et al. Transdifferentiation of glioblas- toma stem-like cells into mural cells drives vasculogenic mimicry in glioblas- tomas. J. Neurosci. 2012; 32: 12950-60.
14 Xu YZ, Zhu Y, Shen ZJ et al. Significance of heparanase-1 and vascular endothelial growth factor in adrenocortical carcinoma angiogenesis: potential for therapy. Endocrine 2011; 40: 445-51.
15 Aubert S, Wacrenier A, Leroy X et al. Weiss system revisited: a clinico- pathologic and immunohistochemical study of 49 adrenocortical tumors. Am. J. Surg. Pathol. 2002; 26: 1612-9.
16 Fassnacht M, Johanssen S, Quinkler M et al. Limited prognostic value of the 2004 international union against cancer staging classification for adrenocorti- cal carcinoma. Cancer 2009; 115: 243-50.
17 Zhang Y, Sun B, Zhao X et al. Clinical significances and prognostic value of cancer stem-like cells markers and vasculogenic mimicry in renal cell carci- noma. J. Surg. Oncol. 2013; 108: 414-9.
18 Wang SY, Ke YQ, Lu GH et al. Vasculogenic mimicry is a prognostic factor for postoperative survival in patients with glioblastoma. J. Neurooncol. 2013; 112: 339-45.
19 Luo F, Yang K, Liu RL et al. Formation of vasculogenic mimicry in bone metastasis of prostate cancer: correlation with cell apoptosis and senescence regulation pathways. Pathol. Res. Pract. 2014; 210: 291-5.
20 Zhang Z, Han Y, Zhang KK, Teng LZ. Investigation of vasculogenic mimi- cry in intracranial hemangiopericytoma. Mol. Med. Rep. 2011; 4: 1295-8.
21 Sun QM, Zou X, Zhang T et al. The role of miR-200a in vasculogenic mimi- cry and its clinical significance in ovarian cancer. Gynecol. Oncol. 2014; 132: 730-8.
22 Chen Y-S, Chen Z-P. Vasculogenic mimicry: a novel target for glioma ther- apy. Chin. J. Cancer 2014; 33: 74-9.
23 Sun D, Sun BC, Liu TJ et al. Slug promoted vasculogenic mimicry in hepa- tocellular carcinoma. J. Cell Mol. Med. 2013; 17: 1038-47.
24 Xu Y, Li Q, Li XY et al. Short-term anti-vascular endothelial growth factor treatment elicits vasculogenic mimicry formation of tumors to accelerate metastasis. J. Exp. Clin. Cancer Res. 2012; 31: 16.
25 Vartanian A, Stepanova E, Grigorieva I et al. VEGFR1 and PKCalpha signal- ing control melanoma vasculogenic mimicry in a VEGFR2 kinase-indepen- dent manner. Melanoma Res. 2011; 21: 91-8.