Amplification and expression of EGFR and ERBB2 in Wilms tumor

Cancer Genetics and Cytogenetics 194 (2009) 88e95

Amplification and expression of EGFR and ERBB2 in Wilms tumor Mohammad Vaseia,*, Helmout Modjtahedib, Oreineb Ale-booyeha, Ahmad Mosallaeic, Abdol Mohammad Kajbafzadehd, Mehdi Shahriarie, Abbas Ali Ghaderif, Hossein Soleymanpoura, Farid Kosarig, Holger Mochh, Guido Sauteri a

Department of Pathology, Shiraz Medical School and Shiraz Institute of Cancer Research, Shiraz University of Medical Sciences, Shiraz, Iran b School of Life Sciences, Faculty of Science, Kingston University London, Kingston, United Kingdom c Department of Radiation-Oncology, Nemazee Hospital, Shiraz, Iran d Pediatric Urology Research Center, Department of Urology, Tehran University of Medical Sciences, Tehran, Iran e Department of Pediatric Oncology, Nemazee Hospital, Shiraz University of Medical Sciences, Shiraz, Iran f Shiraz Institute of Cancer Research, Shiraz University of Medical Sciences, Shiraz, Iran g Department of Pathology, Tehran University of Medical Sciences, Tehran, Iran h Department of Pathology, University Hospital, Zurich, Switzerland i Institute of Pathology, University Hospital, Hamburg-Eppendorf, Germany Received 23 March 2009; received in revised form 5 June 2009; accepted 8 June 2009

Abstract

Wilms tumor is one of the most common solid tumors in children. We evaluated expression and amplification of a number of genes and their prognostic significance in 45 patients with Wilms tumor, using tissue microarray technology. The expression of EGFR, ERBB2, MDM2, CCND1, MLH1, MSH2, TP53, and ABCB1 (alias MDR1) was studied by immunohistochemistry. Amplification of the EGFR, ERBB2, MDM2, CCND1, CTTN (previously EMS1), RAF1, MYC, FGF3 (previously INT2), WNT1, GLI1, CDK4, and NCOA3 (alias AIB1) genes was assessed by fluorescence in situ hybridization. Expression of EGFR was seen in 17 of the 45 cases (38%) but was not associated with gene amplification. The ERBB2 gene was neither overexpressed nor amplified in any case. Tissue microarray and immunohistochemistry analyses for ERBB2 in whole-tumor sections were also negative in all cases. Strong p53 reactivity was noted in blastemal cells in two cases with an unfavorable outcome. ABCB1 reactivity was seen in five cases with favorable histology and outcome. Only one case showed nuclear cyclin D1 positivity. All tumors showed MLH1 and MSH2 expression. All examined genes showed normal copy numbers. Unfavorable histology correlated with poor prognosis (P 5 0.03). There was no significant association between gene expression and prognosis. Overexpression of the EGFR gene in many Wilms tumor cases warrants further study to determine the therapeutic benefit of EGFR inhibitors in combination with other therapies in Wilms tumor patients. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction Wilms tumor (WT) or nephroblastoma, the most common kidney malignancy in childhood, is responsible for ~6% of all cancers in children. Thanks to recent advances in surgery and combination chemotherapy, overall survival in patients with WT is as high as 85% [1]. Nonetheless, gaps remain in our knowledge of the basic oncogenic events in patients with this tumor [2]. Since the early 1980s, aberrant membranous expression of the type-I growth factor receptor family with tyrosine kinase activity, in particular of EGFR (alias ERBB1, c-erbB1), has * Corresponding author. Tel.: and fax: þ98-711-2301784. E-mail address: [email protected] (M. Vasei). 0165-4608/09/$ e see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.cancergencyto.2009.06.003

been reported in a wide range of adult epithelial tumors, and some studies have associated these expressions with poor prognosis and decreased survival [3e5]. The EGFR and ERBB2 (alias HER2) genes encode two homologous receptors with specific extracellular ligand-binding domains and intracellular tyrosine kinase domains [6]. In recent years, several drugs have received U.S. Food and Drug Administration (FDA) approval for the treatment of patients with EGFRand ERBB2-expressing tumors: tyrosine kinase inhibitors (e.g., gefitinib and erlotinib), which target the catalytic domain of EGFR, and monoclonal antibodies (e.g., cetuximab and trastuzumab), which target the extracellular domain of EGFR and ERBB2, respectively [7e9]. Since its launch in 1998, trastuzumab has become an important therapeutic option for patients with ERBB2-positive breast cancer

M. Vasei et al. / Cancer Genetics and Cytogenetics 194 (2009) 88e95

[9,10]. Few studies have appeared on the expression of ERBB2 and EGFR in WT, however, and no studies to date have addressed the amplification status of ERBB2 and EGFR in WT [11-13]. In tumors that express these two receptors, a key factor is the accurate detection of the two molecules on the cell surface. Immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) are the best methods for detecting the expression and amplification of these genes, respectively [14-17]. We therefore used IHC to investigate the expression of the EGFR and ERBB2 genes and used FISH to investigate their amplification in 45 patients with WT. To expand our knowledge of the possible underlying molecular events in WT, we also evaluated amplification of the MDM2, WNT1, RAF1, MYC, CTTN (previously EMS1), FGF3 (previously INT2), CCND1, GLI1, CDK4, and NCOA3 (alias AIB1) genes and evaluated expression of the TP53, MDR1, MDM2, CCND1, MLH1, and MSH2 genes, using tissue microarray technology (TMA). We then looked for correlations between gene expression and amplification findings and clinical outcome.

2. Materials and methods Hematoxylin and eosinestained slides of tumors from a total of 67 patients with WT were retrieved from the archives of the departments of pathology of hospitals affiliated with the medical school of the Shiraz University of Medical Sciences. Of 67 cases reviewed, tissue blocks suitable for TMA were available from 35 tumors. The tumors were classified as having unfavorable or favorable histology according to previously defined criteria [18]. Clinical data and information about the patients’ subsequent course were collected from medical charts in the radiotherapy and pediatric oncology departments. Paraffin blocks and clinical data from 10 further cases were also collected from other centers in Iran. Sections stained with hematoxylineeosin were labeled to identify three different punch sites to construct TMA blocks. Tissue cylinders 0.6 mm in diameter were then punched from each donor tissue block and were embedded in a recipient paraffin block using a custom-made precision instrument.

89

Eight samples were distributed in five paraffin blocks, each containing one to three microbiopsies from various parts of the tumor. Five-micrometer sections of TMA blocks were transferred to glass slides using a paraffin sectioning aid system with adhesive-coated slides (PSACD4x), adhesive tape, and ultraviolet lamp (Instrumedics, Hackensack, NJ). Approximately 3,100 cores were evaluated by IHC. After paraffin removal and antigen retrieval, TMA sections were incubated with monoclonal antibodies at appropriate dilutions according to their manufacturers’ guidelines (Table 1). The sections were washed and treated with a streptavidin-biotin complex method using Dako universal LSAB kits (Dako, Glostrup, Denmark) or Novostatin super ABC kits (Novocastra, Newcastle, UK). For staining, 3,3’-diaminobenzidine-tetrahydrochloride chromogen (Dako) or aminoethylcarbazole (Sigma, St. Louis, MO) was used. ERBB2 expression was evaluated with using two approaches: the HercepTest (Dako) in five TMA blocks and IHC in whole-tissue sections using an anti-ERBB2 monoclonal antibody (Dako clone PN2A). Slides were scored as positive for ERBB2 membrane overexpression if any tumor cell showed definite membrane staining resulting in a so-called chicken wire pattern. Slides with ambiguous membrane staining (cytoplasmic background) were scored as showing cytoplasmic expression. The ERBB2-expressing tumor cell line SKBR3 was kindly provided by Professor J.P. Mach (Institute of Biochemistry, Lausanne University, Lausanne, Switzerland). Multiple biopsies of SKBR3 xenografts in athymic mice, prepared as described for the WT specimens, were used as the positive control for the HercepTest in TMA and for IHC in whole-tumor sections. The TMA sections were treated according to the paraffin pretreatment reagent kit protocol of the Vysis PathVision assay (Abbott Molecular, Des Plaines, IL) before hybridization. Fluorescence in situ hybridization was performed with Vysis SpectrumOrange-labeled probes for ERBB2, EGFR, MYC, CTTN, CCND1, INT2, WNT1, GLI1, CDK4, MDM2, RAF1, and NCOA3). Vysis SpectrumGreen-labeled centromeric probes for chromosomes 3, 7, 8, 9, 11, 12, 17, and 20 were used as the corresponding references. Hybridization and posthybridization washes were done according to the manufacturer’s procedures. The slides were then

Table 1 Characteristics of the antibodies used in this study Antibody

Sourcea

Clone

Dilution

Antigen retrieval method

MDR p53 MDM2 CCND1 EGFR C-erbB2 MLH1 MSH2

Sanbio Dako Dako Dako Zymed Dako Pharmingen Oncogene

JSB-1 Do-7 SMP14 DCS-6 31G7 PN2A 13291A Ab-2

1:800 1:200 1:50 1:200 1:100 1:2,000 1:200 1:200

without pretreatment microwave 90  C/30 min microwave 90  C/60 min pressure cooker 1 min/citrate buffer pronase 15 min 37  C microwave 98  C/60 min microwave 98  C/60 min microwave 98  C/60 min

a Dako, Glostrup, Denmark; Oncogene, Seattle, WA; Pharmingen, BD Biosciences, San Jose, CA; Sanbio, Uden, the Netherlands; Zymed, South San Francisco, CA.

90

M. Vasei et al. / Cancer Genetics and Cytogenetics 194 (2009) 88e95

counterstained with 125 ng/mL 40 ,6-diamidino-2-phenylindole in an antifade solution. The FISH signals were scored with a Zeiss fluorescence microscope equipped with double-band-pass filters for the simultaneous visualization of green and orange spectrum signals. Amplification was defined as the presence of either target gene at a percentromeric signal ratio > 2, or tight clusters of five gene signals in at least 5% of the tumor cells. Samples of ductal invasive carcinoma with documented positivity for ERBB2 amplification and expression were used as a positive control for the HercepTest and for FISH. The tumors were from 45 patients with a mean age of 3 years for boys and 4.2 years for girls. The tumors were more prevalent in males (M/F 5 1.3) and on the left side (left/right 5 1.5). Follow-up data were available for 41 patients, with duration of postoperative follow-up ranging from 6 months to 19 years (average, 58 months) (Table 2). Follow-up in 36 patients lasted for O3 years, and follow-up in 4 patients lasted for !6 months. The patients were treated with a regimen consisting of vincristine and dactinomycin without abdominal irradiation in stage I, and a regimen of vincristine, dactinomycin, and doxorubicin with abdominal irradiation in stages II and III. The patients also received irradiation to the tumor bed (20 Gy) and the entire spine field. Patients with stage IV disease were treated with vincristine, dactinomycin, doxorubicin, and, in some cases, cyclophosphamide. In patients with lung involvement, abdominal irradiation (20 Gy) and whole-lung irradiation (12 Gy) were performed. For statistical analysis, the patients were divided into two groups: poor prognosis patients, who died or had frequent recurrences or metastases within 3 years after surgery, and good prognosis patients, who had disease-free survival for O3 years. The chi-square test was used to search for correlations between IHC results and prognosis. The primary endpoint for clinical progression was defined as the first recurrence. For survival analysis, the endpoint was defined as tumor-related death or incurable metastasis. Life-table analysis was done by plotting a KaplaneMeier curve with the log-rank test.

Fig. 1. EGFR expression in blastemal cells from Wilms tumor sections. Immunohistochemistry with 3,30 -diaminobenzidine staining; 250.

ERBB2 staining was negative in all tumor samples examined with IHC in TMA blocks. The HercepTest in TMA blocks and IHC with the Dako clone PN2A monoclonal antibody in whole-tumor sections were also negative in all tumors. The ERBB2-positive xenograft and breast cancer samples showed classic positive staining at the cytoplasmic membrane (Fig. 2). A strong p53 reaction was seen in the nuclei of blastemal cells of 2 cases with unfavorable histology. One of these patients developed multiple recurrences and metastases to the lung. In the other case, follow-up information was available only for the first 15 months. Weak reactivity (faintly detectable at high power in blastemal cells) was also seen in six other cases with a good prognosis and favorable histology. Five cases showed diffuse cytoplasmic staining for MDR1. Reactivity was seen in epithelial (three cases) and stromal cells (two cases), and in one case positivity was present in both blastemal and epithelial cells. All of these cases had a favorable outcome and histology. Strong nuclear cyclin D1 staining in O10% of cells was seen in the stromal cells in one case. This tumor had unfavorable histology and the clinical outcome in this patient was poor. Antibodies against MLH1 and MSH2 stained all three elements

3. Results 3.1. Gene expression and amplification Staining for EGFR was seen in 17 out of the 45 cases examined (38%) (Fig. 1). Of these, 11, 15, and 2 cases had EGFR expression in the blastemal, stromal, and epithelial component, respectively. All tumors that showed blastemal positivity also showed stromal positivity. The predominant pattern of positivity was diffuse. Among EGFR-positive cases, 10 patients had complete remission, 4 had a poor prognosis, and 1 was alive after 30 months of follow-up. Long-term follow-up information was not available for 2 of the 45 cases.

Fig. 2. Reactivity of biotinylated trastuzumab on the surface of the breast cancer cell line (SKBR-3) used as a positive control. Immunohistochemistry with aminoethylcarbazole staining; 400.

M. Vasei et al. / Cancer Genetics and Cytogenetics 194 (2009) 88e95

91

Table 2 Clinicopathologic data for 45 pediatric Wilms tumor patients Case

Age

Sex

Stage

Histology

Follow-up, yr

Prognosis (outcome)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

4 yr nd 26 yr 4 yr 7.5 yr 4 yr 8 yr 3 yr 5 yr 4 yr 4 yr 2.5 yr 4.5 yr 2 yr 2 yr 14 mo 3.5 yr 4 yr 6 mo 3 yr 2.5 yr 1 yr 7 yr 2 yr 8 mo 2.5 yr 3 yr 5 yr 14 yr 4 yr 3 yr nd 7 yr 6 yr 1.5 yr 13 mo nd nd nd 2.5 yr 8 yr 8 yr nd nd nd

F nd M M M M M M F F F M F F M M M M M M M M F M M M F F F M M nd F M F F nd nd nd M F M nd nd nd

2 3 4 3 1 2 3 3 2 2 1 1 3 2 1 3 1 2 nd 3 3 2 3 1 1 3 2 2 3 2 2 2 3 1 1 2 3 2 nd 2 2 2 3 3 3

favorable favorable favorable unfavorable favorable favorable unfavorable unfavorable favorable unfavorable favorable favorable unfavorable favorable favorable favorable favorable unfavorable favorable favorable favorable unfavorable favorable favorable favorable unfavorable favorable favorable favorable favorable favorable favorable favorable favorable favorable favorable favorable favorable nd favorable favorable favorable nd favorable unfavorable

6 2 1.5 1 10 9 1 1 13 2 3.5 12 1 2 1 15 11 2 0.5 3 1 1 8 13 7 17 11 5 9 1 8 1.7 1 nd 5 6 7 1 nd 11 7 6 0.5 1.4 3

good poor (died) poor (died) poor (died) good good poor (died) poor (recurrences) good poor (metastases) good good poor (metastases) good poor (recurrences) good good poor (died) lf good poor (died) lf good good good good good good good poor (recurrences) good lf lf lf good good good poor (died) lf good good good lf lf good

Abbreviations: lf, lost to follow-up; F, female; M, male; nd, no data.

in all cases. The reactions ranged from focal to diffuse. Cytoplasmic MDM2 reactivity was seen in the blastemal cells in two cases, but was not considered a positive reaction. For FISH, 1,550 intact tissue cores were evaluated. None of the 12 hybridized genes showed amplification. All corresponding centromeric probes for each gene showed two signals in the hybridized tissues (Fig. 3). 3.2. Correlations with clinicopathological characteristics Of the 45 cases, 24 patients had disease-free survival of O3 years (22 with favorable histopathology and 2 with

unfavorable features), and 12 patients died from their tumor, 6 of whom had unfavorable histology. All of the patients who died had had multiple relapses or metastases within 2.5 years after their surgery. In the unfavorable histology group (10 patients), 6 patients had a poor prognosis, 2 had complete remission, and 2 were alive at the time of this study (!2 years follow-up). Of 33 patients with favorable histology, 22 had complete remission, 6 had died within 3 years, and the rest were alive (after !3 years follow-up). There was a significant direct correlation between histology (favorable vs. unfavorable) and cure rate (P 5 0.03). There was no correlation between EGFR expression and tumor histology, nor was there any significant association between EGFR expression

92

M. Vasei et al. / Cancer Genetics and Cytogenetics 194 (2009) 88e95

Fig. 3. Fluorescence in situ hybridization shows two signals corresponding to ERBB2 (orange) and chromosome 17 centromeric (green) probes in Wilms tumor cells counterstained with 40 ,6-diamidino-2-phenylindole antifade solution.

and tumor progression (P 5 0.90) or overall survival (P 5 0.71). No association was found between expression of other genes and overall survival, nor between tumor stage and prognosis.

4. Discussion EGFR expression or amplification has been reported in many human malignancies. In the urinary tract, overexpression of this gene has been demonstrated in kidney [19] and bladder carcinomas [20], and this in turn has been associated with prognosis. We found that although 38% of the WT cases analyzed here were EGFR-positive, EGFR overexpression was not associated with poor prognosis or gene amplification. Ghanem et al. [11] reported EGFR expression in ~44% of the WT cases they examined, and no correlation with clinical progression. Epidermal growth factor receptor is one of the most important factors in kidney tubule morphogenesis and kidney development [21]. A high level of EGFR in the kidney during embryonic development is a physiological event in nephrogenesis [22,23]. Ghanem et al. [11] also found TGF-a expression in blastemal cells of WT. It has been demonstrated that interaction between EGFR and TGF-a directs renal cell proliferation during fetal life. The expression of EGFR and TGF-a in primitive nephroblastoma cells may thus mimic normal physiological growth factor expression in nephroblasts during kidney development or may also be one of the steps in carcinogenesis in WT. Although Ghanem et al. [11] reported expression of ERBB2 in 34% of WT cases, we did not detect membranous ERBB2 expression in our WT, cases using two different

anti-HER-2 antibodies: the HercepTest (Dako) and FISH in TMA, and IHC in whole-tumor paraffin sections. The problem of obtaining accurate and reproducible interpretations of positive results in IHC for ERBB2 is a controversial issue. By applying 28 different antibodies in a cohort of breast cancers with known levels of ERBB2 gene amplification and expression, Press et al. [24] observed significant variation (2e30%) in detection rates for ERBB2 expression with an IHC method. They also showed that only cell membrane staining correlated with elevated level of ERBB2 gene expression. The anti-ERBB2 antibody used by the Ghanem group was shown to have varying degree of cytoplasmic staining in a large number of the cases of breast cancer in which it was tested [24,25]. Indeed, using antiHER-2 antibody ICR12 (a noncommercial monoclonal antibody), we observed cytoplasmic HER-2 immunostaining in 30% of the cases examined (data not shown). Another methodological consideration is the importance of standardizing the slide scoring system, as was highlighted in relation to the best approach to using ERBB2 status to predict the response to therapy with the antiERBB2 antibody trastuzumab [26]. ERBB2 expression scoring can be improved by avoiding specimen edges and retraction artifacts, and, most importantly, by ignoring tumor cells that lack a complete membranous staining pattern (i.e., cells with the so-called chicken-wire appearance). In one IHC study on 14 paraffin blocks of WT, 13 samples were positive for ERBB2 with a mixed cytoplasmic and membranous pattern [12]. Some studies claimed that cytoplasmic reactivity may be due to cross reaction with a protein other than ERBB2 [24,27]. In our study with the FDA-approved Vysis kit for FISH, there was no amplification of ERBB2, which confirmed our IHC results. Many studies have shown that when a standardized IHC assay is performed on specimens that are carefully fixed, processed, and embedded, there is excellent correlation between gene copy status according to FISH analysis and protein expression level according to IHC [17,28,29]. Tapia et al. [30] showed a high association between 3þ positivity with IHC and gene amplification in a TMA study. Recent studies have favored the FISH approach, not only to confirm 2 þ IHC cases but also to verify 3þ positivity and to prevent the use of expensive and potentially toxic trastuzumab in patients with falsepositive IHC results, who are unlikely to benefit from this therapy [31e33]. In breast cancer treatment, only cases with intense circumferential thick membrane staining show a good response to trastuzumab therapy [31,34]. Cytoplasmic staining for ERBB2 (HER-2) protein has been reported in many tumors such as colon adenocarcinoma, Ewing sarcoma, osteogenic sarcoma, endometrial carcinoma, thyroid tumors, and mesothelioma [35e41]. Nonetheless, the possible correlation in patients with these tumors between the cytoplasmic expression of ERBB2 and survival rate, as well as efficacy of trastuzumab therapy, remains to be clarified.

M. Vasei et al. / Cancer Genetics and Cytogenetics 194 (2009) 88e95

To observe trastuzumab binding on WT cells, we applied biotinylated trastuzumab in 10 paraffin blocks (data not shown). Although the positive controls (paraffin blocks from an SKBR3 xenograft and a breast tumor) showed strong membrane reactivity, there was no antibody binding on the membrane or in the cytoplasm of WT cells. Therefore, although there may be some evidence of cytoplasmic immunoreactivity of ERBB2 (HER2) in WT, these patients may not benefit from immunotherapy with trastuzumab. Further studies, using a wide range of anti-HER-2 antibodies, should clarify the biological and clinical significance of ERBB2 expression in patients with WT. In the present WT series of 45 cases, 8 cases showed p53 reactivity (weak and strong). Strong positivity was seen only in 2 cases with unfavorable histology and outcome; all of the weakly reactive tumors showed a favorable clinical course. Lahoti et al. [42] detected p53 expression in all their WT cases by western blot. Nonetheless, the mutant form of p53 was present only in four patients with a poor prognosis, three of whom showed strong p53 reactivity with IHC. Two other studies also confirmed the presence of a significant difference in survival between patients who showed weak or no p53 reactivity and those with strong reactivity [43,44]. Bardeesy et al. [45] observed that homozygosity for the p53 mutation was restricted to anaplastic areas of WT. The MDM2 gene product forms a tight complex with p53 and inhibits p53 function. When it is overexpressed, it enhances the neoplastic potential of cells [46]. We found that this gene was neither amplified nor overexpressed in WT. The absence of amplification and overexpression of MDM2, whose product binds and sequesters normal p53, may indicate that the accumulation of wild-type p53 is not the basis of intense staining in IHC. We therefore suggest that only strong immunoreactivity of p53 (detectable at low power) should be considered a sign of p53 mutation and a predictor of poor prognosis in WT. An association between MDR1 expression and chemotherapy failure has not been well established in WT. One study showed that MDR1 may be overexpressed but that this overexpression did not necessarily affect the concentration of daunorubicin in the cytoplasm of tumor cells [47]. MDR1 positivity was observed in the cell line of favorable cases, as well as in one unfavorable case of WT [48]. In the present study, MDR1 positivity was seen in both epithelial and stromal cells in five cases with favorable histology and long-term disease-free survival. This result implies that the overexpression of MDR1 cannot be regarded as a sign of resistance to chemotherapy in WT. Impairment of the DNA mismatch repair system has been associated with carcinogenesis in some sporadic and inherited cancers. Immunohistochemical staining for MLH1 and MSH2 is an accurate method for detecting mismatch repair defects [49]. Our results indicate that these DNA mismatch repair proteins are normally expressed in WT. In summary, by studying the expression of a number of genes, we found overexpression of EGFR in the absence of

93

EGFR gene amplification in 38% of the WT cases examined. We detected neither the classic membrane expression of ERBB2 nor amplification of ERBB2, a finding that raises doubts about the effectiveness of trastuzumab therapy in these patients. Our results support the need for further studies on expression patterns and prognostic significance of EGFR and the relative expression of phosphorylated EGFR (pEGFR) and mutated EGFR (e.g. EGFRvIII) in EGFR-positive WT [50]. In particular, further studies are warranted to determine the therapeutic potential of EGFR inhibitors in combination with other forms of therapy in patients with EGFR-positive WT.

Acknowledgments The authors would like to thank the staff of the Molecular Pathology Unit at the Institute of Pathology, Basel University, for their assistance with the FISH and tissue microarray assays, and also Mrs. Martina Mirlacher and Mr. Gharesi Fard, for their help with immunohistochemistry. We also thank Dr. Shokrpour and K. Shashok for their assistance in editing this paper, and Dr. Mehran Karimi for his help in collecting some of the follow-up data. Professor J.P. Mach (Institute of Biochemistry, Lausanne University, Lausanne, Switzerland) kindly provided the ERBB2positive breast cell line SKBR3, as well as trastuzumab. References [1] Gommersall LM, Arya M, Mushtaq I, Duffy P. Current challenges in Wilms’ tumor management. Nat Clin Pract Oncol 2005;2:298e304. [2] Pritchard-Jones K. Controversies and advances in the management of Wilms’ tumour. Arch Dis Child 2002;87:241e4. [3] Nicholson RI, Gee JMW, Harper ME. EGFR and cancer prognosis. Eur J Cancer 2001;37(Suppl. 4):S9eS15. [4] Navolanic PM, Steelman LS, McCubrey JA. EGFR family signaling and its association with breast cancer development and resistance to chemotherapy. Int J Oncol 2003;22:237e52. [5] Rubin I, Yarden Y. The basic biology of HER2. Ann Oncol 2001;12(Suppl. 1):S3e8. [6] Coussens L, Yang-Feng TL, Liao YC, Chen E, Gray A, McGrath J, Seeburg PH, Libermann TA, Schlessinger J, Francke U, Levinson A, Ullrich A. Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with neu oncogene. Science 1985;230:1132e9. [7] Mendelsohn J, Baselga J. Status of epidermal growth factor receptor antagonists in the biology and treatment of cancer. J Clin Oncol 2003;21:2787e99. [8] Modjtahedi H. Molecular therapy of head and neck cancer. Cancer Metastasis Rev 2005;24:129e46. [9] Norum J. The cost effectiveness issue of adjuvant trastuzumab in early breast cancer. Expert Opin Pharmacother 2006;7:1617e25. [10] Hortobagyi GN. Overview of treatment results with trastuzumab (Herceptin) in metastatic breast cancer. Semin Oncol 2001; 28(6 Suppl. 18):43e7. [11] Ghanem MA, Van Der Kwast TH, Den Hollander JC, Sudaryo MK, Mathoera RB, Van den Heuvel MM, Noordzij MA, Nijman RJ, van Steenbrugge GJ. Expression and prognostic value of epidermal growth factor receptor-transforming growth factor-a, and c-erb B-2 in nephroblastoma. Cancer 2001;92:3120e9.

94

M. Vasei et al. / Cancer Genetics and Cytogenetics 194 (2009) 88e95

[12] Pinthus JH, Fridman E, Dekel B, Goldberg I, Kaufman-Francis K, Eshhar Z, Harmelin A, Rechavi G, Mor O, Ramon J, Mor Y. ErbB2 is a tumor associated antigen and a suitable therapeutic target in Wilms tumor. J Urol 2004;172:1644e8. [13] Salem M, Kinoshita Y, Tajiri T, Souzaki R, Tatsuta K, Higashi M, Izaki T, Kohashi K, Tsuneyoshi M, Taguchi T. Association between the HER2 expression and histological differentiation in Wilms tumor. Pediatr Surg Int 2006;22:891e6. [14] Dei Tos AP, Ellis I. Assessing epidermal growth factor receptor expression in tumours: what is the value of current test methods? Eur J Cancer 2005;41:1383e92. [15] Hicks DG, Tubbs RR. Assessment of the HER2 status in breast cancer by fluorescence in situ hybridization: a technical review with interpretive guidelines. Hum Pathol 2005;36:250e61. [16] Ross JS, Fletcher J. HER-2/neu (c-erb-B2) gene and protein in breast cancer. Am J Clin Pathol 1999;112(1 Suppl. 1):S53e67. [17] Schnitt SJ, Jacobs TW. Current status of HER2 testing: caught between a rock and a hard place. Am J Clin Pathol 2001;116:806e10. [18] Argani P, Beckwith BJ. Renal neoplasms of childhood. In: Mills SE, Carter D, Grenson JK, Oberman HA, Reuter V, Stoler MH, editors. Sternberg’s diagnostic surgical pathology. 4th ed. Philadelphia: Lippincott Williams & Wilkins, 2004. pp. 2001e32. [19] Moch H, Sauter G, Gasser TC, Bubendorf L, Richter J, Presti JC Jr, Waldman FM, Mihatsch MJ. EGF-r gene copy number changes in renal cell carcinoma detected by fluorescence in situ hybridization. J Pathol 1998;184:424e9. [20] Neal DE, Sharples L, Smith K, Fennelly J, Hall RR, Harris AL. The epidermal growth factor receptor and the prognosis of bladder cancer. Cancer 1990;65:1619e25. [21] Cybulsky AV, Goodyer PR, McTavish AJ. Epidermal growth factor receptor activation in developing rat kidney. Am J Physiol 1994;267:F428e36. [22] Sakurai H, Tsukamoto T, Kjelsberg CA. EGF receptor ligands are a large fraction of in vitro branching morphogenesis secreted by embryonic kidney. Am J Physiol 1997;273:F463e72. [23] Goodyer PR, Mulligan L, Goodyer CG. Expression of growth-related genes in human fetal kidney. Am J Kidney Dis 1991;17:608e10. [24] Press MF, Hung G, Godolphin W, Slamon DJ. Sensitivity of HER-2/ neu antibodies in archival tissue samples: potential source of error in immunohistochemical studies of oncogene expression. Cancer Res 1994;54:2771e7. [25] Singleton TP, Niehans GA, Gu F, Litz CE, Hagen K, Qiu Q, Kiang DT, Strickler JG. Detection of c-erbB-2 activation in paraffin-embedded tissue by immunohistochemistry. Hum Pathol 1992;23:1141e50. [26] Paik S, Bryant J, Tan-Chiu E, Romond E, Hiller W, Park K, Brown A, Yothers G, Anderson S, Smith R, Wickerham DL, Wolmark N. Realworld performance of HER2 testing: National Surgical Adjuvant Breast and Bowel Project experience. J Natl Cancer Inst 2002;94:852e4. [27] De Potter CR, Quatacker J, Maertens G, Van Daele S, Pauwels C, Verhofstede C, Eechaute W, Roels H. The subcellular localization of the neu protein in human normal and neoplastic cells. Int J Cancer 1989;44:969e74. [28] Dowsett M, Bartlett J, Ellis IO, Salter J, Hills M, Mallon E, Watters AD, Cooke T, Paish C, Wencyk PM, Pinder SE. Correlation between immunohistochemistry (HercepTest) and fluorescence in situ hybridization (FISH) for HER-2 in 426 breast carcinomas from 37 centres. J Pathol 2003;199:418e23. [29] Press MF, Slamon DJ, Flom KJ, Park J, Zhou JY, Bernstein L. Evaluation of HER-2/ neu gene amplification and overexpression: comparison of frequently used assay methods in a molecularly characterized cohort of breast cancer specimens. J Clin Oncol 2002;20: 3095e105. [30] Tapia C, Glatz K, Novotny H, Lugli A, Horcic M, Seemayer CA, Tornillo L, Terracciano L, Spichtin H, Mirlacher M, Simon R, Sauter G. Close association between HER-2 amplification and overexpression in human tumors of non-breast origin. Mod Pathol 2007;20:192e8.

[31] Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, Fleming T, Eiermann W, Wolter J, Pegram M, Baselga J, Norton L. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001;344:783e92. [32] Mass RD, Press MF, Anderson S, Cobleigh MA, Vogel CL, Dybdal N, Leiberman G, Slamon DJ. Evaluation of clinical outcomes according to HER2 detection by fluorescence in situ hybridization in women with metastatic breast cancer treated with trastuzumab. Clin Breast Cancer 2005;6:240e6. [33] Hammock L, Lewis M, Phillips C, Cohen C. Strong HER-2/neu protein overexpression by immunohistochemistry often does not predict oncogene amplification by fluorescence in situ hybridization. Hum Pathol 2003;34:1043e7. [34] Tedesco KL, Thor AD, Johnson DH, Shyr Y, Blum KA, Goldstein LJ, Gradishar WJ, Nicholson BP, Merkel DE, Murrey D, Edgerton S, Sledge GW Jr. Docetaxel combined with trastuzumab is an active regimen in HER-2 3þ overexpressing and fluorescent in situ hybridization-positive metastatic breast cancer: a multi-institutional phase II trial. J Clin Oncol 2004;22:1071e7. [35] Cunningham MP, Essapen S, Thomas H, Green M, Lovell DP, Topham C, Marks C, Modjtahedi H. Coexpression of the IGF-IR, EGFR and HER-2 is common in colorectal cancer patients. Int J Oncol 2006;28:329e35. [36] Essapen S, Thomas H, Green M, De Vries C, Cook MG, Marks C, Topham C, Modjtahedi H. The expression and prognostic significance of HER-2 in colorectal cancer and its relationship with clinicopathological parameters. Int J Oncol 2004;24:241e8. [37] Navarro S, Giraudo P, Karseladze AI, Smirnov A, Petrovichev N, Savelov N, Alvarado-Cabrero I, Llombart-Bosch A. Immunophenotypic profile of biomarkers related to anti-apoptotic and neural development pathways in the Ewing’s family of tumors (EFT) and their therapeutic implications. Anticancer Res 2007;27:2457e63. [38] Kilpatrick SE, Geisinger KR, King TS, Sciarrotta J, Ward WG, Gold SH, Bos GD. Clinicopathologic analysis of HER-2/neu immunoexpression among various histologic subtypes and grades of osteosarcoma. Mod Pathol 2001;14:1277e83. [39] Lambropoulou M, Stefanou D, Alexiadis G, Tamiolakis D, Tripsianis G, Chatzaki E, Vandoros GP, Kiziridou A, Papadopoulou E, Papadopoulos N. Cytoplasmic expression of c-erb-B2 in endometrial carcinomas. Onkologie 2007;30:495e500. [40] Gumurdulu D, Uguz A, Erdogan S, Tuncer I, Demircan O. Expression of c-erbB-2 oncoprotein in different types of thyroid tumors: an immunohistochemical study. Endocr Res 2003;29:465e72. [41] Horvai AE, Li L, Xu Z, Kramer MJ, Jablons DM, Treseler PA. Malignant mesothelioma does not demonstrate overexpression or gene amplification despite cytoplasmic immunohistochemical staining for c-Erb-B2. Arch Pathol Lab Med 2003;127:465e9. [42] Lahoti C, Thorner P, Malkin D, Yeger H. Immunohistochemical detection of p53 in Wilms’ tumors correlates with unfavorable outcome. Am J Pathol 1996;148:1577e89. [43] Cheah PL, Looi LM, Chan LL. Immunohistochemical expression of p53 proteins in Wilms’ tumour: a possible association with the histological prognostic parameter of anaplasia. Histopathology 1996;28: 49e54. [44] Govender D, Harilal P, Hadley GP. p53 protein expression in nephroblastomas: a predictor of poor prognosis. Br J Cancer 1998;77:314e8. [45] Bardeesy N, Beckwith JB, Pelletier J. Clonal expansion and attenuated apoptosis in Wilms’ tumors are associated with p53 gene mutations. Cancer Res 1995;55:215e9. [46] Momand J, Zambetti GP, Olson DC. The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 1992;69:1237e45. [47] Ramachandran C, Melnick SJ, Escalon E, Jhabvala P, Khatib Z, Alamo A, Smith S. Expression of apoptosis, cell proliferation, and drug resistance genes in pediatric Wilms’ tumors. Anticancer Res 2000;20:3759e65.

M. Vasei et al. / Cancer Genetics and Cytogenetics 194 (2009) 88e95 [48] Re GG, Willingham MC, el Bahtimi R, Brownlee NA, HazenMartin DJ, Garvin AJ. Anaplasia and drug selection-independent overexpression of the multidrug resistance gene, MDR1, in Wilms’ tumor. Mod Pathol 1997;10:129e36. [49] Marcus VA, Madlensky L, Gryfe R, Kim H, So K, Millar A, Temple LK, Hsieh E, Hiruki T, Narod S, Bapat BV, Gallinger S, Redston M. Immunohistochemistry for hMLH1 and hMSH2:

95

a practical test for DNA mismatch repair-deficient tumors. Am J Surg Pathol 1999;23:1248e55. [50] Cunningham MP, Essapen S, Thomas H, Green M, Lovell DP, Topham C, Marks C, Modjtahedi H. Coexpression, prognostic significance and predictive value of the EGFR, EGFRvIII and phosphorylated EGFR. in colorectal cancer. Int J Oncol 2005;27: 317e25.