Background
Central nervous system (CNS) tumors are the second most common group of malignancies among children; leukemias as a group are the most common. However, CNS tumors are the most common form of solid tumors in children. The overall average annual incidence rate for pediatric CNS tumors (ages 0–19 years) is 5.26 per 100,000 [
1]. Embryonal tumors are the most common CNS neoplasms in infants less than 36 months of age and are described by the World Health Organization (WHO) classification scheme as undifferentiated small round cell tumors with divergent patterns of differentiation [
2]. Embryonal Tumors include primitive neuroectodermal tumor (PNET), medulloblastoma (MB), atypical teratoid/rhabdoid tumor, and several other histology types. Though those tumors in this category are histologically similar, they have different patterns of incidence and survival, so it is important to look at them individually.
In USA (United States of America), the incident rate of CNS embryonal tumors under 14 years old is 0.8 per 100,000 and the median age is 9. In the age 0 ~ 4, the embryonal tumors are the most common histology, and between age 5 ~ 14, the embryonal tumors are still the third most common subtype compared with pilocytic astrocytoma [
1]. Conversely, there were few reports related with the large-scale follow-up or prognostic analysis of pediatrics CNS embryonal tumors in china.
MB comprises up to 20% of all pediatric brain tumors and is currently treated with surgical resection, radiation therapy, and chemotherapy [
3]. Molecular genetic parameters, being associated with poorer prognosis of MB, include overexpressed ERBB-2, high MYCC expression, and possibly p53 accumulation [
4]. The single most-predictive clinical factor is extent of disease at the time of diagnosis, patients with disseminated disease fare less well [
5]. Especially ERBB-2, belonging to the human epidermal growth factor receptor (EGFR) family, is overexpressed in 40% of MBs and its expression correlates with poor outcome [
6]. However, ERBB-2 expression in PNET is unclear and should be invested. PNET is histologically similar to classic MB and constitutes 2% of all childhood brain tumors. The most common sites of PNET onset are the cerebrum, suprasellar, or pineal region of children in their first decade of life [
6]. Because MB and PNET share the aggressive biological behavior, it is crucial to determine the prognostic factors for guiding the individual treatment.
Glioblastoma multiforme (GBM) is the second most frequently reported malignancy in CNS, which account for 15.6% of all primary brain tumors in adults. GBMs are more common in older adults and are uncommon in children [
1]. GBM is frustratingly chemoresistant and follows a highly aggressive course, with an average survival of roughly 1 year. Although small cells are common in GBM, they are predominant or exclusive in a subset known as small cell GBM [
7]. Small cell GBM is a histological subtype of GBM with characteristic features of highly proliferative, monotonous small glial cells with high nuclear cytoplasm ratio. In this study, we also focused on the prognostic research for small cell astrocytoma/GBM for the reason that it shares some similar features with embryonal tumors. The cytogenetical investigations for IDH1/2 mutation, 1p/19q loss, and PTEN alteration are strongly supportive methods for the differential diagnosis of small cell GBM [
8]. PTEN also represents a putative tumor suppressor gene in MB because loss of PTEN function would contribute to an over-activation of the PI3K/AKT signaling pathway, which is activated in MB [
9]. Mutations in the IDH1/2 genes are similarly detected in patients with non-glial tumors with the exception of PNET, which suggesting the unique mechanism of PNET shared with small cell GBM [
10]. Therefore, small cell GBM shares the similar genetics and histopathology features with MB and PNET, and we group them together as pediatrics CNS small cell tumors in this study.
The members of the EGFR family have been linked to the astrocytic tumors malignant transformation. This receptor family consists of four tyrosine kinase receptors, ERBB1-4, and seems to be involved in tumor cell proliferation, differentiation and cell survival [
11]. Due to overexpression of the ERBB1-4 proteins on the surface of neoplastic astrocytes, they are candidates for targeted therapy [
12]. Such treatment, however, requires reliable detection systems for these receptor proteins in tumor tissue. EGFR gene amplification can now simply be evaluated by means of fluorescence in situ hybridization (FISH) [
13]. Several studies have shown a varying degree amplification of the EGFR (ERBB1) gene, located on chromosome 7, in GBM [
14]. EGFR gene amplification distinguishes small cell GBM from anaplastic oligodendrogliomas [
7], nevertheless the spectrum of clinicopathologic and prognostic features has not been explored fully in small cell GBM. And the other members of EGFR family expression levels and their correlation with prognosis are also unclear.
In china, according our investigation, most hospitals even only offer patient surgical resection without radiation or chemotherapy. It is important to raise the neurosurgeons and oncologic doctors attention for combining various forms treatment to increase young patients disease-free survival time and improve the quality of life. For pathologists’ responsibility, finding the prognostic factors of pediatrics CNS small cell tumors becomes the important task to give the physicians suggestion. This study was also designed to investigate the clinical features and the extent of ERBB-1 ~ 4 gene expression in the small cell GBM, PNET and MB. Further and most importantly, we assumed to explore the prognosis factors in children small cell CNS tumors.
Data collection and methods
Clinical data collection
All 71 pediatric (≤16 year-old) CNS primary small cell tumors out of 383 children CNS primary tumors (18.54%) were operated at the Department of Neurosurgery, Affiliated Provincial Hospital, Shandong University, Jinan, China, and consecutively collected in the time period 2000 to 2012, including 44 cases medulloblastoma (MB), 8 cases primitive neuroectodermal tumor (PNET) and 19 cases small cell glioblastoma (GBM). A statistical analysis was performed to collect demographic and clinical data that included age, sex, tumor localization, treatment modalities, and postoperative survival (Additional file
1).
Craniotomies were performed under general anesthesia, and all patients underwent magnetic resonance imaging (MRI) a few days before and within 72 hours after surgery. The extent of tumor resection was determined by the postoperative MRI scans, defined as gross total resection or partial resection (residual volume exceeding 2 cm). The data and tumor specimens were retrieved and revised in 2013 by neurosurgeons and pathologists in Provincial Hospital according to the 2007 WHO Classification of Tumors of the Central Nervous System [
2]. All specimens were taken during the patients’ first surgery.
Immunohistochemistry (IHC)
Expression of ERBB1-4 receptor proteins was determined by IHC using commercial monoclonal antibodies EGFR (1:50, Santa Cruz), ERBB-2 (1:50, Santa Cruz), ERBB-3 (1:50, Santa Cruz) and ERBB-4 (1:50, Santa Cruz). The proliferation of tumor cells was detected by ki-67 IHC staining. Formalin-fixed and paraffin-embedded sections, 4 μm thick, with representative tumor tissue, were incubated with primary antibodies after antigen retrieval by pressure autoclaving. An automatized histostainer was used for the immunohistochemcial procedures (Dako Autostainer, Denmark). For visualization of immunoreactivity, DAKO EnVision system was used with diaminobenzidin as chromogene. Sections were counterstained with haematoxylin. Positive controls were included in each staining run.
The immunoreactivity was assessed by means of intensity and percentage of immunoreactive tumor cells. Intensity was recorded as 0 (no reaction) to 3 (strong reaction). Fraction of immunoreactive tumor cells was recorded as 0 (no positive cells), 1 (<10% positive cells), 2 (10-50% positive cells), or 3 (>50% positive cells). A staining index was calculated as the product of intensity and fraction of positive tumor cells [
15].
FISH (Fluorescence in situ hybridization)
The copy number of EGFR and ERBB-2 can be determined by direct fluorescence in situ hybridization (FISH) [
16,
17].
EGFR/CEN-7 FISH Probe Mix (DakoCytomation) and the Histology FISH Accessory Kit (Dako) were applied for gene copy number detection of the EGFR gene located on chromosome 7p11.2 and for copy number detection of the chromosome 7 centromere region (chromosome 7 copy number detection). The Red-labeled DNA probe (EGFR) binds to a 196 kb segment containing the EGFR gene on chromosome 7q11.2. The fluorescein Green-labeled probe (CEN-7) binds to the centromeric region of chromosome 7. The PathVysion ERBB-2/HER-2 probe kit (Abbott Molecular) was used for the FISH analysis. The protocol was similar with EGFR gene FISH staining. Slides were hybridized with prewarmed probes for the ERBB-2 gene (orange) and chromosome 17 centromere (Her2/neu/CEP17 SG probe, Vysis) overnight at 37°C.
In brief, the sections were de-paraffinised using xylene, rehydrated, and pretreated using DAKO solution kit (Histology FISH Accessory kit). The probes were added to the sections, coverslipped, sealed with rubber cement, and placed in a DAKO Hybridizer. The sections and probes were co-denatured for 5 min at 73°C, followed by annealing at 37°C over night. After hybridization slides were washed in 0.4 × SSC (with detergent) at 73°C for two minutes followed by one minutes in 2 × SSC at room temperature. Then the sections underwent dehydrating in ethanol three times for 3 min. The slides were counterstained and embedded with a 4,6-diamidino-2-phenylindole/antifade solution (DAKO). FISH signals for each locus-specific FISH probe were assessed under a Nikon Eclipse 90i microscope (Nikon, Tokyo, Japan) equipped with a triple-pass filter (DAPI/Green/Orange).
The signal enumeration was performed under high magnification (×1,000). The entire area of tumor tissue was evaluated in each case, and all nuclei were assessed for the orange (marker) and green (reference) signals. For a patient to be included, 100 evaluable cells were to be assessed. Tumors were interpreted as amplified when the ratio of HER-2/EGFR signals divided by chromosome 17/7 centromere signals was equal to or greater than 2.2 and the normal specimens showed a ratio of <1.8. Results at or near the cutoff point (1.8 – 2.2) were repeated with a fresh specimen slide.
Statistics and follow-up analysis
Statistical analyses were made using SPSS version 16.0 (SPSS Inc., Chicago, IL). Survival time was calculated from date of surgery to date of death or the termination of observation. Multiple Cox regression analyze was used to study the association between sex (categorical variable; male versus female), age (continuous variable), tumor size (continuous variable), extent of surgical resection (categorical variable; gross total resection versus partial resection), the treatment (categorical variable; no radiotherapy versus radiotherapy), ki-67 (Staining Index, continuous variable), ERBB-1 ~ 4 protein expression (Staining Index, continuous variable) and survival prognosis. The Kaplan-Meier method was applied to draw the survival curves and the log-rank test was used for survival analysis. The association between results from the FISH investigations (categorical variables; positive versus negative) and survival were studied in the same manner. The relationship of IHC and FISH staining between the different tumor groups was analyzed by non-parametric Kruskal-Wallis H method. Two-sided P-values less than 0.05 were regarded as statistically significant.
The study was approved by the Committee for Medical Research Ethics of Affiliated Provincial Hospital, Shandong University.
Discussion
In this research, we concentrated on the prognostic factors for the pediatrics CNS small cell tumors contained small cell GBM, PNET and MB. In the Background part, we already indicated the reason why we grouped them together for the similar genetics or histopathology characters. EGFR protein overexpression was determined as the negative prognostic factor for all small cell CNS tumors in our research. EGFR gene amplification was evaluated necessarily by means of FISH in MB. Even for the small cell GBM and PNET, the disease-free survival time of negative FISH signals was longer than the one of positive signals. We assumed that the limited number of cases was the reason for that the Cox analyses excluded the EGFR amplification from the prognostic factors of small cell GBM and PNET.
The relationship has been correlated the presence of small cell architecture in primary GBM with EGFR amplification [
18]. Combination of an anti-EGFR agent Iressa and a JAK2/STAT3 inhibitor synergistically suppressed STAT3 activation and potently killed GBM cell lines that expressed EGFR [
19]. However, Alterations of CDKN2A, EGFR, CDK4, and MDM2 genes, commonly implicated in gliomagenesis, were not identified in any PNET, which was opposite with our results [
20]. Aberrant activation of Hedgehog (HH) signaling has been identified as a key etiologic factor in MB [
21]. Frank Götschel et al. identified a novel crosstalk mechanism whereby EGFR signaling silences proteins acting as negative regulators of HH signaling [
22]. The effects of the EGFR inhibitor gefitinib on cell growth and signaling were evaluated in three MB cell lines (D283, D341, Daoy), and gefitinib induced G0/G1 arrest in all lines, indicating that gefitinib might be a molecularly targeted agent for the treatment of MB [
23]. As those data indicated, the mechanism of EGFR over-activation in CNS small cell tumors is still ambiguous. The deepen study combining clinical bed to bench is required. At the meantime, the molecular inhibitors of EGFR tyrosine kinases (erlotinib or gefitinib) should be considered as the effective individualized treatment and start the consortium study in China for pediatrics CNS small cell tumors.
According the previous results, ERBB-2 and ERBB-3 mRNAs were detected only in a few high-grade gliomas, while ERBB-4 expression was most pronounced in low-grade gliomas [
24]. It was also reported that ERBB-2 and ERBB-4 are highly expressed in aggressive forms of medulloblastoma [
25]. However, ERBB-4 expression was downregulated in HH signaling-induced MBs from mice. According to the animal experiments, HH signalling inhibited ErbB-4 expression in mouse cerebellar granule progenitors and human MB cells [
26]. And the other researchers also suggested that expression of ERBB-2 is much lower on MB tumor cells than on breast cancer cells, so they are not susceptible to ERBB-2 monoclonal antibodies, like trastuzumab (Herceptin) [
5]. The expression and amplification of ERBB-2 and ERBB-4 were not pronounced in MB, PNET and small cell GBM detected by FISH and IHC according our results. Expanding the number of cases will offer more evidence for the ERBB-2 and ERBB-4 expression. It is now recognized that MB is a collection of heterogeneous entities with disparate demographics, transcriptomes, genetics, and clinical outcomes [
27]. According to international consensus, the principle subgroups of MB are WNT, SHH, Group 3, and Group 4 [
28]. Because our prognostic study did not account for these subgroups, we hypothesized that the difference of EGFR family members expression in MB could have resulted from differential subgroup representation among studies.
We have found that all children patients involved in this study did not receive any chemotherapy and only 39.44% of patients underwent the radiotherapy after surgery. Other researchers have reported a median survival of 14.3 months for small cell GBM [
29]. With present means of surgery, craniospinal radiotherapy, and chemotherapy, between 75% and 90% of children greater than 3 years of age with nondisseminated MB are likely to be survivors 5 years after treatment [
4]. Compared with those promising prolonged survival time in western countries, the median time of MB and small cell GBM in our hospital was only 2 years and 8 months. We assumed that the short disease-free survival time was affected by the less of radiotherapy. Infants and children with supratentorial PNET and MB are special compared with adult patients for long-term neurocognitive development being considered. This is not only due to the whole-brain radiation therapy these children are often treated with, but also due to the local effects of the tumor and the need for higher-dose boost radiotherapy. In the future, the children cognitive levels should be involved when considering the treatment and prognosis, which will lead more challenge for neurosurgeons and social work.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
WL and QP made substantial contributions to conception and design, and analysis and interpretation of data; WL, TG and LZ carried out the immunohistochemistry. QC and LZ participated in the FISH staining. WL, SZ, LZ, QC, JW and TG participated in the design of the study and performed the statistical analysis. WL, SZ and JW were involved in drafting the manuscript or revising it critically for important intellectual content; QP have given final approval of the version to be published; and WL and QP agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All authors read and approved the final manuscript.