ELSEVIER
Contents lists available at ScienceDirect
Journal of Pediatric Surgery
journal homepage: www.elsevier.com/locate/jpedsurg
2
Journal of Pediatric Surgery
@
Correlation between selected angiogenic markers and prognosis in pediatric adrenocortical tumors Angiogenic markers and prognosis in pediatric ACTs
CrossMark
André Ivan Bradley dos Santos Dias a,*, 1,2, Camila Girardi Fachin a,2, Lucimar Retto Silva Avó b, Caio Vinicius Gonçalves Frazão ª, Eliana Maria Monteiro Caran , Sérgio Tomaz Schettini ª, Maria Teresa Seixas Alves b, Raul C. Ribeiro ª, Simone de Campos Vieira Abib c
a Pediatric Surgery, Department of Surgery, Universidade Federal de São Paulo/ Escola Paulista de Medicina - UNIFESP/EPM, Rua Coronel Lisboa, 687, São Paulo-SP, 04020-041, Brasil
b Department of Pathology, Universidade Federal de São Paulo/ Escola Paulista de Medicina - UNIFESP/EPM, Rua Botucatu, 740, São Paulo-SP, 04023-062, Brasil
” Instituto de Oncologia Pediátrica, Universidade Federal de São Paulo/ Escola Paulista de Medicina - UNIFESP/EPM, Rua Botucatu, 743, São Paulo, SP, 04023-062, Brasil
d St. Jude Children’s Research Hospital, 262 Danny Thomas Place, Memphis, TN 38105-3678
ARTICLE INFO
Article history: Received 6 August 2014 Received in revised form 7 November 2014 Accepted 7 December 2014
Key words: Adrenocortical tumor Immunohistochemistry Prognosis Angiogenesis and children
ABSTRACT
Background/Purpose: Pediatric adrenocortical tumor (ACT) remains a challenging disease. Tumor weight and disease stage are still the most used indicators to prognosis and guidance of clinical decisions. Histology has not added meaningful data for risk stratification and management. ACT is metabolically active, highly vascularized, locally invasive and has the propensity to produce distant metastasis. Our objective was to correlate the expression of vascular endothelial growth factor (VEGF) and intratumoral microvessel density (MVD) with clinical and prognostic aspects in pediatric ACT.
Procedure: In 27 tumors, immunohistochemical expression of VEGF, CD105 (endoglin) and CD34 was analyzed. MVD was determined by CD34 and CD105 antibodies. MVD and VEGF expression was correlated with clinical characteristics and outcome. Normal pediatric glands were used as controls.
Results: Endoglin MVD was significantly higher and CD34 MVD was significantly lower in ACT than control. The VEGF expression did not differ between groups. Cytoplasmic staining for endoglin was correlated with hypertension in ACT. Endoglin MVD greater than 1 mv/field, CD34 MVD less than 32 mv/field and VEGF expression levels above 4.8% were associated with clinical and biological indicators of poor prognosis.
Conclusions: Endoglin and CD34 MVD values are potential histological markers to refine the histologic classification of pediatric ACT.
@ 2015 Elsevier Inc. All rights reserved.
Adrenocortical tumors (ACTs) are rare, comprising 0.2% of all pediatric malignancies [1]. The estimated worldwide incidence is about 0.3 new case per million individuals per year [2]. However, in the south and south- east regions of Brazil, ACT incidence rates are 15-20 times higher than those described in other countries. This ACT cluster in Brazil is due to the presence of a founder TP53 mutation in this population [3,4]. Surgery is the only curative treatment modality in ACT. Patients with incomplete resection or metastatic disease have a dismal prognosis.
Other prognostic factors have not been firmly established for pedia- tric ACT [5-7]. Currently, tumor size and disease stage remain the pri- mary prognostic factors of this disease [5,8,9]. However, small tumors can behave aggressively suggesting that more specific prognostic factors should be investigated. Histologically, ACTs are classified as adenoma
and carcinoma. Adenoma histology is associated with excellent prognosis, but only about 20% of pediatric ACTs are classified as that. Moreover the distinction between adenoma and carcinoma is difficult [7] and many times tumors are considered of undefined histology. Current immunohis- tochemistry analysis although corroborating the morphologic diagnosis of ACT has not contributed to prognostic analysis. We reasoned that the existing histological classification of pediatric ACT could be improved by including markers of angiogenesis.
Angiogenesis refers to new blood vessel formation from a pre- existing vascular bed via the proliferation, differentiation and migration of endothelial cells [10]. Pathological angiogenesis is characterized by the persistent proliferation of endothelial cells (ECs). A direct relation- ship has been demonstrated between tumor angiogenesis and the po- tential for growth and metastatic spread [10-12]. These processes are controlled by angiogenic factors including vascular endothelial growth factor (VEGF), which is one of the key cytokines involved in normal and pathological angiogenic processes. VEGF expression has been described as a prognostic factor in lung cancer [13], liver cancer [14] and pulmonary metastasis in colorectal cancer [15].
* Corresponding author. E-mail address: andrebradleymd@gmail.com (A.I.B.S. Dias).
1 Present adress: Av. Prof. Noé Azevedo, 208, cj 112, São Paulo-SP, 04117-000, Brasil.
2 The two first authors contributed equally to this paper.
Tumor angiogenesis can also be quantified by assessing the intratumor microvessel density (MVD). MVD has previously been described as an independent prognostic factor for survival in breast [16], prostate [17] and other malignancies [18,19]. Although significant heterogeneity exists among patient populations evaluated and tech- niques used to determine MVD, most studies have demonstrated in- verse relations between tumor vascularity and prognosis [18].
Some studies on cervical, colon, endometrial and breast carcinomas [19-21] have suggested that endoglin represents a more sensitive and specific marker for newly formed blood vessels than other pan- endothelial markers, such as CD31, CD34 and factor VIII. In some tumors, the prognostic value of MVD has been demonstrated via endoglin ex- pression, while other pan-endothelial markers have failed to produce such results [19,20].
IHC analysis of CD34 and factor VIII (both pan-endothelial markers) expression often reveals more intense staining in larger vessels, while such staining is diminished or absent in the microvessels of many tumors [22].
No reports on the prognostic value of angiogenic markers in pediatric ACT have been reported. However, this topic is relevant because of the lack of robust prognostic marker and relative high incidence of pediatric ACT in the south and southeast regions of Brazil.
The objective of this study was to determine the clinical and biological correlates of the expression for VEGF and MVD in children with ACT; verifying their relevance as prognostic markers as well as refining the existent histological classification.
1. Material and methods
This study was approved by the ethics committee of the Paulista Medical School of the Federal University of São Paulo (UNIFESP/EPM). The requirement for informed consent was waived in cases of patient death or inability to contact patients who had completed follow up. The clinical data and pathological tissue samples from the primary tu- mors were retrospectively retrieved from the records maintained at the Pediatric Oncology Institute (IOP-GRAAC) and the Department of Pathology, which are both located at UNIFESP/EPM, for the period from July 1997 to November 2011.
1.1. Patients
Of the 33 pediatric patients with primary ACT who were admitted to the hospital during this period, 27 were included in this study. Patients older than 20 and those whose tumor samples could not be retrieved were excluded. All patients underwent surgical resection by the same surgical team, in which surgeons had at least 10 years of experience in Pediatric Oncology. Patients’ data are shown in Table 1.
The follow-up time was calculated from the date of admission to either the time of death or their last clinical appointment. The criterion for cure was established as four years of follow-up with no evidence of active disease. Of the 13 patients who died, 11 deaths occurred before 24 months, one was recorded at 30 months and another occurred at 46 months of follow-up. No deaths occurred after 46 months.
As for pathology, four cases were classified as adenoma, 19 carcinoma, and four of undetermined classification. Eleven pediatric adrenal glands removed together with renal tumors were used as controls and were examined using the same IHC and morphometric analysis protocols used for the tumor samples.
1.2. Immunohistochemistry (IHC)
The tissues isolated from the clinical specimens were fixed in 10% buffered formalin, processed and stained with hematoxylin and eosin. For IHC analysis, one paraffin block containing the maximum amount of tumor with proper fixation was selected for each case.
| Age at diagnosis (months), mean (SD) | 65 (66) |
|---|---|
| Sex | |
| Female | 15 (55.6%) |
| Male | 12 (44.4%) |
| Disease stage - COG | |
| I | 11 (41%) |
| II | 2 (7%) |
| III | 9 (33%) |
| IV | 5 (19%) |
| Metastasis at diagnosis | 5 (19%) |
| Hepatic | 3 (11%) |
| Hepatic + Pulmonary | 2 (7%) |
| Surgical data | |
| Total resection | 23/25 (92%) |
| Abdominal recurrence | 11/25 (44%) |
| Intraoperative rupture | 10/25 (40%) |
| Tumor weight (g), mean (SD) | 224 (269) |
| Tumor diameter (cm), mean (SD) | 8.8 (4.9) |
| Pathological diagnosis | |
| Adenoma | 4 (15%) |
| Borderline | 4 (15%) |
| Carcinoma | 19 (70%) |
| Pathological data | |
| Capsule invasion | 19 (70.4%) |
| Necrosis | 18 (66.7%) |
| Free surgical margins | 13 (48.1%) |
| Mitotic index (mitosis/50 hpf), mean (SD) | 67.4 (70.8) |
| CD34 MVD (MV/field), mean (SD) | 30.7 (17.6) |
| CD105 MVD (MV/field), mean (SD) | 1.12 (0.92) |
| VEGF (%), mean (SD) | 11.40 (17.22) |
| Event-free survival analysis | |
| Relapse | |
| Death | |
| Abandonment | none |
| Alive without disease | 8 (29.6%) |
| Alive with disease | 6 (22.2%) |
| Dead | 13 (48.2%) |
Abbreviations: SD, standard deviation; COG, Children’s Oncology Group; hpf, high power field; MVD, microvessel density; MV, mcrovessel.
Four-micrometer-thick sections from the paraffin-embedded tissue blocks were stained for endoglin, CD34 and VEGF. The tissue sections were mounted on coated slides and dried for 1 h at 57 ℃. The sections were deparaffinized in xylene and rehydrated in a descending ethanol series. Heat-induced epitope retrieval (30 min in the microwave) was employed. For all antigens, endogenous peroxidase was quenched by immersion in 3% hydrogen peroxide five times for 5 min each.
The primary antibodies used included a rabbit polyclonal antibody against VEGF (Santa Cruz Biotechnology, Santa Cruz, CA, USA), a mouse monoclonal antibody against CD34 (Cell Marque Corp., Rocklin, CA, USA) and a mouse anti-human monoclonal antibody against endoglin (clone SN6H, Dako Corp., Carpinteria, CA, USA). Each antibody was used at a 1:100 dilution.
After overnight primary antibody incubation at 4 ℃, CD34 and VEGF staining was performed using Novolink™ detection with the Compact Polymer™ system kit (Leica Biosystems, UK), whereas endoglin staining was performed using Dako’s CSA™ kit. The appropriate positive and negative controls were examined for each batch of slides.
1.3. MVD morphometric analysis
CD34 MVD and endoglin MVD was quantified using light microscopy (Olympus™ BX51 microscope; Olympus Corporation, Shinjuku Monolith, 2-3-1 Nishi-Shinjuku, Shinjuku-ku, Tokyo, Japan) without the operator’s knowledge of the patient details. After scanning the immunostained sec- tions at a low magnification (40x), the most vascular area of the tumor were located. This area was then evaluated at 200x magnification, and
10 contiguous field images were digitized (Sony™ camera model DXC- 107A; Sony Corporation, 1-7-1 Konan, Minato-ku, Tokyo, Japan).
Two independent observers (both experienced histology board- certified pathologists from the UNIFESP/EPM, with expertise in Pediatric Oncology), who were blinded to patient identity, counted the microvessels (MVs). Any single cell or spot stained by the IHC marker was counted as an MV. A visible vascular lumen was not required, as in previous reports [20,23]. The highest number of MVs counted in each field was recorded. Any discrepancy in the MV count of either ob- server greater than 10% was resolved by recounting the MVs. The mean mv/field in 10 contiguous fields was used for the statistical analysis.
1.4. Cytoplasmic staining morphometric analysis
VEGF cytoplasmic staining was quantified for 10 contiguous micro- scopic fields at 400x magnification at the area of most intense staining [17]. The images were digitized and analyzed using image-editing soft- ware, as previously described [24]. The stained areas were quantified as a percent ratio between the immunolabeling area in the analyzed image and the total area of the digitized field. The cytoplasm of some cortical cells, both normal and tumorous, was stained by endoglin IHC, and the quantification was performed using the same method as that employed for VEGF.
1.5. Control group
Eleven pediatric adrenal glands, which had been removed together with renal tumors, but not infiltrated by it, were used as controls and were examined using the same protocols described above and com- pared to the tumor group.
1.6. Statistical analysis
The comparison of IHC features between the tumor and control groups and the correlation of clinical, surgical and pathological findings with the IHC data were performed using the Student’s t-test and an ANOVA for normally distributed variables, whereas the Mann-Whitney, Kruskal-Wallis and Spearman’s correlation tests were used for variables with non-normal distributions.
A survival analysis was performed using Cox uni- and multivariate regression analyses and Kaplan-Meier curves according to the log- rank test. The IHC data were treated as a continuous variable, although it was divided into medians for the purpose of performing a survival analysis with Kaplan-Meier curves.
All statistical analyses were performed using SPSS (Statistical Package for Social Sciences, SPSS Inc., Chicago, IL) version 17 (November 2008).
2. Results
Only CD34 MVD was a normally distributed variable; all the others had non-normal distributions.
2.1. CD34, CD105 and VEGF expression in ACT and normal adrenal cortex
The number of CD34 MVD was significantly higher in normal adrenal cortex samples than in ACT samples (mean = 58.3 ± 26.4 mv/field vs.30.7 ± 17.6 mv/field; p = 0.001). Fig. 1A shows IHC photomicrograph staining for CD34 (200x) in pediatric ACT and control tissue. CD34, a pan endothelial marker, has affinity for non angiogenic vessels that are expected to be more frequent in normal glands than tumors. However, the number of endoglin MVD was significantly higher in the tumor samples than in normal adrenal cortex (median = 1, IIQ = 0.4-1.7 versus median = 0, IIQ = 0-0 mv/field; p < 0.001), as shown in Fig. 1B. These data can be explained by the observation that endoglin has a higher affinity for endothelial cells in newly formed vessels.
During the evaluation of endoglin expression in the endothelium, areas of cytoplasmic staining were observed in both the tumor and con- trol samples (Fig. 1C). This parameter was referred to as CD105 CITO. Although there have been reports of this type of cytoplasmic expression in other tumor types [25,26], there are no reports in the literature concerning the cytoplasmic expression of endoglin in pediatric ACT. However, there was no statistically significant difference in expression rate CD105 CITO between the ACT and control samples (mean = 32.9, SD = 20.5 versus mean = 22.34, SD = 13.2% expression; p = 0.122).
The VEGF cytoplasmic expression ratio observed in case of ACT and control samples is illustrated in Fig. 1D. Unexpectedly, there was no sta- tistically significant difference between the groups (median = 4.82, IIQ = 1.62-11.13 versus median = 7.72, IIQ = 0.65-20.45% expression; p < 0.001), which was likely due to the extensive vascularization of the normal adrenal cortex. Comparisons between groups regarding the IHC data are summarized in Table 2.
2.2. Relationship between the IHC and clinical data
In the tumor group, the relationship between the IHC data and some previously described prognostic factors was analyzed. The potential prognostic factors were selected based on various clinical, surgical and pathological parameters reported, including age [9,27], tumor weight [5,28] and diameter [5], the time interval between the onset of signs and symptoms and diagnosis [9,29,30] and mitotic index [5,9].
Several statistically significant associations were observed between the IHC parameters and clinical. In a univariate analysis, CD34 MVD values correlated inversely with age (p < 0.001, r = - 0.63), tumor weight (p<0.016, r =- 0.48) and diameter (p<0.018, r = - 0.47); endoglin MVD correlated with tumor diameter (p < 0.018, r = 0.47), mitotic index (p = 0.003, r = 0.55) and stage (median 0.5 for I and II versus 1.7 mv/field for III and IV, p = 0.001). The greater the tumor diameter and stage, the greater the endoglin MVD. Moreover, endoglin MVD was associated with capsule invasion (median 0.3 versus 1.3 mv/field, p = 0.010) and tumor intraoperative rupture (median 0.5 versus 1.55 mv/field, p = 0.010). VEGF expression correlated with tumor weight (p = 0.02, r = 0.461) and diameter (p = 0.006, r = 0.532). The bigger the tumor, the greater the VEGF expression. VEGF expression was also associated with metastasis at diagnose (median 3.11% versus 15.91% expression). Those with metastasis had greater VEGF expression than those which didn’t. CD105 CITO was associated with arterial hypertension (mean 16.9% for “no” versus 39.6% expression for “yes”, p = 0.010) and inversely with acne (mean 46% for “no” versus 24% ex- pression for “yes”, p = 0.004). The tumor that had capsule invasion and ruptured intraoperatively had bigger endoglin MVD than those which didn’t.
Of the patients that had pathologically compromised surgical mar- gins, 64% (9 patients) developed local recurrence, and all died but one that had tumor CD34 MVD higher than 32, CD105 MVD lower than 1 and VEFG expression lower than 4.8%; 22% (3 patients) did not develop local recurrence, and only one is dead due to pulmonary metastasis; in 14% (2 patients) local recurrence cannot be evaluated - one was a par- tial resection that died 2 months after surgery and the other patient died during the surgery.
Pathologic free surgical margins had significant correlation with CD105 MVD (p = 0.02).
2.3. Survival analysis
The impact of the IHC parameters on survival was first analyzed using Kaplan-Meier curves compared with the log-rank test, in which the pa- rameters were categorized according to whether the values were above or below the median. As shown in Fig. 2, CD34 MVD (inversely), endoglin MVD and VEGF expression were significantly associated with poor outcome (p < 0.05).
CONTROL
TUMOR
A
CD34
B
Endoglin (CD105)
C
CD105 CITO
D
VEGF
Cox univariate models were constructed for all clinical and IHC parameters. Among the clinical parameters evaluated, tumor weight demonstrated to be a risk factor for death (HR: 1.005, CI: 1.002-1.008, p < 0.001). The results for the IHC parameters are shown in Table 3. These parameters were also categorized above or below the median for further analysis. The significant ones included endoglin MVD (p = 0.009) and VEGF expression (p = 0.007). CD34 MVD reached a p-value close to 0.05 (p = 0.051), and the group that scored below the median (up to 31.9 mv/field) demonstrated higher mortality rates when the results were categorized (p = 0.031).
A Cox multivariate model was constructed according to tumor weight, endoglin MVD, VEGF expression and CD34 MVD. In this model, in addition to the tumor weight (HR: 1.004, CI: 1.001-1.007, p =
| IHC Parameter | Control (n = 11) | ACT (n = 27) | p-value |
|---|---|---|---|
| CD34 MVD Mean (SD) | 58.3 (26.4) | 30.7 (17.6) | 0.001 T* |
| VEGF Median (IIQ) | 7.72 (0.65-20.45) | 4.82 (1.62-11.13) | 0.910 MW |
| Endoglin MVD Median (IIQ) | 0 (0-0) | 1 (0.4-1.7) | <0.001 MW* |
| CD105 CITO Mean (SD) | 22.34 (13.28) | 32.98 (20.52) | 0.122 ₸ |
Abbreviations: CD34 MVD, microvascular density using CD34 IHC; Endoglin MVD, micro- vascular density using endoglin IHC; VEGF, cytoplasmic expression of vascular endothelial growth factor; CD105 CITO, cytoplasmic expression of endoglin; SD, standard deviation; IIQ, second quartile; MW, Mann-Whitney test; T, Student’s t-test.
⁎ p < 0.05
0.007), the endoglin MVD (HR: 1.913, CI: 1.057-3.464, p = 0.032) and VEGF expression (HR: 1.061, CI: 1.011-1.114, p = 0.016) represented independent risk factors for death.
3. Discussion
Prognostic factor analysis in pediatric ACT has been limited because of the tumor rarity and biological heterogeneity and lack of uniform treatment. Tumor size and disease stage, but not histology, have been associated with prognosis. Pediatric ACT grows very rapidly and is highly vascularized. We reasoned that markers of vascular status correlate with its clinical behavior and prognosis. This malignancy is very rare, thus the difficulty of assembling an adequate amount of patients for a proper prognostic study. Our sample size (27 patients in 14 years) appears small but is expressive when compared to others (29 patients in 90 years) [9]. It’s also composed by very advanced tumors and is there- fore biased with a lower survival rate than literature. Consequently, prognostic markers can be only suggested, but not defined.
The difficulty to determine whether an adrenocortical tumor is benign or malignant remains in the literature and is typical of these tumors, with weight being one of the guidelines to that dilemma. Even the standard IHC was not yet known until now and a control of normal adrenal glands was needed to establish some parameters. Undetermined is a tumor in which not all the criteria were fulfilled either for carcinoma or for ade- noma, in a definitive way.
The three most frequently used histopathological classification sys- tems are Hough et al. [31], Weiss [32,33] modified by Bugg et al. [34] and Aubert et al. [35]; and van Slooten et al. [36]. They have good predic- tive value for adults, but not for pediatric patients [5,37]. According to
Survival Probability
1.0
CD34 MVD
1.0
Log Rank
Endoglin MVD
1.0
Log Rank
VEGF expression
Log Rank
p= 0.019
P< 0.001
p= 0.048
0.8
0.8
n=11
n= 14
0.8
n= 14
0.6
≥ 31.9 mv/field
0.6
≤ 1 mv/field
0.6
≤ 4.82% expression
0.4
0.4
0.4
n= 16
n= 13
0.2
0.2
n= 13
< 31.9 mv/field
0.2
> 4.82% expression
A
> 1 mv/field
0
0
B
C
9
0
12
24
36
48
60
72
84
96 108
20
32
1441
21 .14
56 6 168
0
12
24
36
48
60
72
84
96
108
120 132 144 156 168
0
12
24
36
48
60
72
84
96
108
120
2 144
156 168
FOLLOW-UP (months)
these criteria, many children classified as having carcinoma have a benign outcome [6].
The CD34 and endoglin MVDs were evaluated in both pediatric his- tologically normal adrenal cortex and ACT samples. The findings indi- cated that adrenal cortex malignancy in children correlated with an increased number of vessels with endoglin endothelial staining but a decreased number of vessels expressing CD34.
Moreover, as also demonstrated by the present study, CD34 MVD has been inversely correlated with tumor size in a breast cancer series [19]. This finding demonstrates that small tumors have a proportionally greater number of pre-existing vessels entrapped by the tumor that are, therefore, more likely to be stained by pan-endothelial markers. This re- sult is supported by an experimental study that investigated the early vasculature in adenocarcinomas in rats and found that tumor vascula- ture was initially acquired via the infiltration of a pre-existing vascular network [38].
Concerning the reviewed literature, this work is the first to demon- strate cytoplasmic endoglin expression (via IHC) in both cancerous and normal pediatric adrenal cortex cells. In ACT, this cytoplasmic expres- sion was associated with acne and arterial hypertension.
Transforming growth factor-ß (TGF-B) is a major regulator of tissue morphogenesis that generally inhibits the cell cycle by slowing or blocking the G1 phase [39]. Specifically for cells of the adrenal cortex, TGF-B acts as a potent inhibitor of autocrine and/or paracrine steroidogenic functions [40,41].
| IHC Parameters | HR+ | CI 95%+ | p-value |
|---|---|---|---|
| CD34 MVD | 0.967 | 0.934-1.000 | 0.051 |
| CD34 MVD below the median (≤31.9 vessels/field) | 4.179 | 1.136-15.367 | 0.031* |
| Endoglin MVD | 1.781 | 1.158-2.739 | 0.009* |
| Endoglin MVD above the median (>1 vessel/field) | 7.464 | 1.973-28.245 | 0.003* |
| VEGF | 1.039 | 1.010-1.068 | 0.007* |
| VEGF above the median (>4.8% of area expression) | 3.089 | 0.945-10.095 | 0.062 |
| CD105CITO | 1.011 | 0.988-1.034 | 0.343 |
| CD105CITO above the median (>2.8% of area expression) | 2.797 | 0.853-9.165 | 0.089 |
Abbreviations: CD34 MVD, microvascular density using CD34 IHC; Endoglin MVD, microvas- cular density using endoglin IHC; VEGF, cytoplasmic expression of vascular endothelial growth factor; CD105 CITO, cytoplasmic expression of endoglin; HR, hazard ratio; CI, confidence interval.
* p < 0.05.
Endoglin expressed by tumors interacts with TGF-B, for it is an ac- cessory item in the TGF-B receptor complex [42-44], and this may ex- plain the observed increased steroidogenic production. Additionally, endoglin may block the inhibitory effect of TGF-ß on cell proliferation, which may represent an important pathway for pediatric adrenal cortex carcinogenesis and therefore constitute a potential therapeutic target.
However, further studies employing additional methods are re- quired to determine the point of synthesis and release of endoglin in pe- diatric ACT, as well as the relationship between endoglin expression and TGF-ß production in terms of hormonal production, cancer prognosis, arterial hypertension and carcinogenesis.
MVD in tumors has previously been assessed by staining tissues with pan-endothelial antibodies against antigens such as CD34, CD31 and the von Willebrand factor. However, ECs are highly heterogeneous, which raises the question of whether a pan-endothelial marker is the ideal re- agent to evaluate tumor MVD [19].
Previous papers have reported that anti-endoglin antibodies, unlike pan-endothelial markers, preferably bind to “activated” EC in vitro and in tissues participating in angiogenesis [22,45-47]. The present findings corroborate these reports and also show that endoglin MVD is directly associated whereas CD34 MVD is inversely associated with patient prognosis.
VEGF expression was similar between the groups evaluated. In addi- tion, the regulation of VEGF and its receptors by adrenocorticotropic hor- mone (ACTH) in the adrenal cortex has been demonstrated in rats [48]. Thus, VEGF may be a local regulator of adrenal cortical angiogenesis and an important mediator of the tropic action of ACTH, thereby ensuring the coordination of ACTH-stimulated cortical growth and vascularization [49]. These data suggest that ACTH may serve as a major regulator of the adrenal cortex by acting on both the steroidogenic and endothelial compartments, which may explain the elevated VEGF expression levels observed in normal tissue.
VEGF expression has previously been described as a prognostic factor in lung cancer [13], liver cancer [14] and pulmonary metastases in colo- rectal cancer [15]. The results of this study show that VEGF expression in pediatric ACT was also associated with prognosis, although no difference in VEGF immunostaining was found between the ACT and normal cortex samples.
Establishing a prognosis for pediatric ACT remains a challenge. The main contribution of this study is the introduction of IHC prognostic parameters - CD34 MVD, endoglin MVD and VEGF expression - that demonstrated predictive value for tumor evolution.
These markers should be considered for clinical application. Al- though these conclusions are based on retrospective data, a multicenter
prospective study is necessary to confirm them. Further studies should also include metastatic samples. On the other hand, it is a cheaper method when compared to genetics, and thus can be useful in different hospital settings. Thus, the present data contribute to better define un- determined tumors. It can also help to select a group of patients with more aggressive disease who may be candidates to receive more ad- vanced treatment protocols. Because children with metastatic adreno- cortical carcinoma have such a poor prognosis with conventional therapeutic approaches including chemotherapy and or radiation thera- py, they would be the logical group of patients to be treated with novel therapies. In our sample, even though patients with abdominal recur- rence received adjuvant therapy (radio and/or chemotherapy), all of them died, except for one whose IHC angiogenic markers indicated good prognosis.
In conclusion, endoglin MVD was higher in the ACT group than in the control group, while CD34 MVD was relatively lower in the ACT group compared to the controls. Endoglin MVD greater than 1 mv/field, CD34 MVD less than 32 mv/field and VEGF expression above 4.8% were asso- ciated with more aggressive pediatric ACT. Combined, VEGF expression, endoglin MVD and CD34 MVD were able to define which patient with undetermined histology tumor would have a less aggressive disease or not. Therefore, VEGF expression, endoglin MVD and CD34 MVD (inversely) may refine the histological classification in pediatric ACT.
References
[1] Liou LS, Kay R. Adrenocortical carcinoma in children. Review and recent innovations. Urol Clin North Am 2000;27:403-21.
[2] Ribeiro RC, Michalkiewicz EL, Figueiredo BC, et al. Adrenocortical tumors in children. Braz J Med Biol Res 2000;33:1225-34.
[3] Ribeiro RC, Sandrini F, Figueiredo B, et al. An inherited p53 mutation that contributes in a tissue-specific manner to pediatric adrenal cortical carcinoma. Proc Natl Acad Sci U S A 2001;98:9330-5.
[4] Pinto EM, Billerbeck AEC, Villares MCBF, et al. Founder effect for the highly prevalent R337H mutation of tumor suppressor p53 in Brazilian patients with adrenocortical tumors. Arq Bras Endocrinol Metabol 2004;48:647-50.
[5] Wieneke JA, Thompson LD, Heffess CS. Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol 2003;27:867-81.
[6] Mendonca BB, Lucon AM, Menezes CA, et al. Clinical, hormonal and pathological findings in a comparative study of adrenocortical neoplasms in childhood and adulthood. J Urol 1995;154:2004-9.
[7] Teinturier C, Pauchard MS, Brugieres L, et al. Clinical and prognostic aspects of adre- nocortical neoplasms in childhood. Med Pediatr Oncol 1999;32:106-11.
[8] Hanna AM, Pham TH, Askegard-Giesmann JR, et al. Outcome of adrenocortical tumors in children. J Pediatr Surg 2008;43:843-9.
[9] Klein JD, Turner CG, Gray FL, et al. Adrenal cortical tumors in children: factors associated with poor outcome. J Pediatr Surg 2011;46:1201-7.
[10] Folkman J, Shing Y. Angiogenesis. J Biol Chem 1992;267:10931-4.
[11] Sasano H, Ohashi Y, Suzuki T, et al. Vascularity in human adrenal cortex. Mod Pathol 1998;11:329-33.
[12] Fox SB. Tumour angiogenesis and prognosis. Histopathology 1997;30:294-301.
[13] Parra ER, Park JY, Saito DM, et al. Prognostic index expression of cyclin-D1, cerbB-2 and VEGF: metastases vs corresponding primary cancers and metastatic vs non- metastatic adenocarcinomas. Histol Histopathol 2008;23:987-93.
[14] Chen J, Li Q, Wang C, et al. Prognostic significance of c-erbB-2 and vascular endothelial growth factor in colorectal liver metastases. Ann Surg Oncol 2010;17:1555-63.
[15] Tamura M, Oda M, Tsunezuka Y, et al. Vascular endothelial growth factor expression in metastatic pulmonary tumor from colorectal carcinoma: utility as a prognostic factor. J Thorac Cardiovasc Surg 2004; 128:517-22.
[16] Weidner N, Folkman J, Pozza F, et al. Tumor angiogenesis: a new significant and in- dependent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst 1992;84:1875-87.
[17] Weidner N, Carroll PR, Flax J, et al. Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. Am J Pathol 1993;143:401-9.
[18] Fox SB, Gatter KC, Harris AL. Tumour angiogenesis. J Pathol 1996; 179:232-7.
[19] Kumar S, Ghellal A, Li C, et al. Breast carcinoma: vascular density determined using CD105 antibody correlates with tumor prognosis. Cancer Res 1999;59:856-61.
[20] Saad RS, Jasnosz KM, Tung MY, et al. Endoglin (CD105) expression in endometrial carcinoma. Int J Gynecol Pathol 2003;22:248-53.
[21] Brewer CA, Setterdahl J, Li MJ, et al. Endoglin expression as a measure of microvessel density in cervical cancer. Obstet Gynecol 2000;96:224-8.
[22] Wang JM, Kumar S, Pye D, et al. Breast carcinoma: comparative study of tumor vas- culature using two endothelial cell markers. J Natl Cancer Inst 1994;86:386-8.
[23] Abulafia O, Triest WE, Sherer DM. Angiogenesis in squamous cell carcinoma in situ and microinvasive carcinoma of the uterine cervix. Obstet Gynecol 1996;88:927-32.
[24] Lopes NN, Plapler H, Chavantes MC, et al. Cyclooxygenase-2 and vascular endothelial growth factor expression in 5- fluorouracil-induced oral mucositis in hamsters: evalua- tion of two low-intensity laser protocols. Support Care Cancer 2009;17:1409-15.
[25] Bialas M, Papla B, Bulanda A. Immunohistochemical investigation of selected endo- thelial markers in pulmonary epithelioid haemangioendothelioma. Pol J Pathol 2011;62:236-40.
[26] Minhajat R, Mori D, Yamasaki F, et al. Organspecific endoglin (CD105) expression in the angiogenesis of human cancers. Pathol Int 2006;56:717-23.
[27] Tucci Jr S, Martins AC, Suaid HJ, et al. The impact of tumor stage on prognosis in children with adrenocortical carcinoma. J Urol 2005;174:2338-42 [discussion 42].
[28] Michalkiewicz E, Sandrini R, Figueiredo B, et al. Clinical and outcome characteristics of children with adrenocortical tumors: a report from the International Pediatric Adrenocortical Tumor Registry. J Clin Oncol 2004;22:838-45.
[29] Ribeiro RC, Sandrini Neto RS, Schell MJ, et al. Adrenocortical carcinoma in children: a study of 40 cases. J Clin Oncol 1990;8:67-74.
[30] Sabbaga CC, Avilla SG, Schulz C, et al. Adrenocortical carcinoma in children: clinical aspects and prognosis. J Pediatr Surg 1993;28:841-3.
[31] Hough AJ, Hollifield JW, Page DL, et al. Prognostic factors in adrenal cortical tumors. A mathematical analysis of clinical and morphologic data. Am J Clin Pathol 1979; 72(3):390-9.
[32] Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 1984;8(3):163-9.
[33] Weiss LM, Medeiros LJ, Vickery Jr AL. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13(3):202-6.
[34] Bugg MF, Ribeiro RC, Roberson PK, et al. Correlation of pathologic features with clinical outcome in pediatric adrenocortical neoplasia. A study of a Brazilian population. Brazilian Group for Treatment of Childhood Adrenocortical Tumors. Am J Clin Pathol 1994;101(5):625-9.
[35] Aubert S, Wacrenier A, Leroy X, et al. Weiss system revisited: a clinicopathologic and immunohistochemical study of 49 adrenocortical tumors. Am J Surg Pathol 2002; 26(12):1612-9.
[36] van Slooten H, Schaberg A, Smeenk D, et al. Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 1985;55(4):766-73.
[37] Cagle PT, Hough AJ, Pysher TJ, et al. Comparison of adrenal cortical tumors in children and adults. Cancer 1986;57(11):2235-7.
[38] Thompson WD, Shiach KJ, Fraser RA, et al. Tumours acquire their vasculature by vessel incorporation, not vessel ingrowth. J Pathol 1987;151:323-32.
[39] Laiho M, DeCaprio JA, Ludlow JW, et al. Growth inhibition by TGF-beta linked to suppression of retinoblastoma protein phosphorylation. Cell 1990;62:175-85.
[40] Hotta M, Baird A. Differential effects of transforming growth factor type beta on the growth and function of adrenocortical cells in vitro. Proc Natl Acad Sci U S A 1986; 83:7795-9.
[41] Feige JJ, Cochet C, Savona C, et al. Transforming growth factor beta 1: an autocrine regulator of adrenocortical steroidogenesis. Endocr Res 1991;17:267-79.
[42] Lastres P, Letamendia A, Zhang H, et al. Endoglin modulates cellular responses to TGF-beta 1. J Cell Biol 1996; 133(5):1109-21.
[43] Barbara NP, Wrana JL, Letarte M. Endoglin is an accessory protein that interacts with the signaling receptor complex of multiple members of the transforming growth factor-beta superfamily. J Biol Chem 1999;274(2):584-94.
[44] Cheifetz S, Bellon T, Cales C, et al. Endoglin is a component of the transforming growth factor-beta receptor system in human endothelial cells. J Biol Chem 1992; 267(27):19027-30.
[45] Seon BK, Matsuno F, Haruta Y, et al. Long-lasting complete inhibition of human solid tumors in SCID mice by targeting endothelial cells of tumor vasculature with antihuman endoglin immunotoxin. Clin Cancer Res 1997;3:1031-44.
[46] Wang JM, Kumar S, Pye D, et al. A monoclonal antibody detects heterogeneity in vascular endothelium of tumours and normal tissues. Int J Cancer 1993;54:363-70.
[47] Westphal JR, Willems HW, Schalkwijk CJ, et al. Characteristics and possible function of endoglin, a TGF-beta binding protein. Behring Inst Mitt 1993:15-22.
[48] Mallet C, Feraud O, Ouengue-Mbele G, et al. Differential expression of VEGF receptors in adrenal atrophy induced by dexamethasone: a protective role of ACTH. Am J Physiol Endocrinol Metab 2003;284:E156-67.
[49] Shifren JL, Mesiano S, Taylor RN, et al. Corticotropin regulates vascular endothelial growth factor expression in human fetal adrenal cortical cells. J Clin Endocrinol Metab 1998;83:1342-7.