Comparison of Fluorescence and Chromogenic In Situ Hybridization for Detection of HER-2/neu Oncogene in Breast Cancer

Transcription

1 Anatomic Pathology / HER-2 DETECTION BY CISH IN BREAST CANCER Comparison of Fluorescence and Chromogenic In Situ Hybridization for Detection of HER-2/neu Oncogene in Breast Cancer Deepali Gupta, MD, 1 Lavinia P. Middleton, MD, 2 Marion J. Whitaker, CT(ASCP), 3 and Jacki Abrams, MD 1,3 Key Words: HER-2/neu; Fluorescence in situ hybridization; Chromogenic in situ hybridization; Breast cancer; CISH; FISH DOI: /P40P2EAD42PUKDMG Abstract Determination of HER-2/neu oncogene amplification has become clinically important in the present management of breast cancer and may have important applications in other areas of clinical oncology and scientific research. In situ hybridization is an extremely accurate and sensitive technique for assessing amplification of HER-2/neu. A new method using a chromogen-labeled probe offers numerous advantages, including the ability to view the morphologic features of the cells of interest using a light microscope, which can be found in every laboratory. We used both techniques to assay 31 cases of infiltrating breast carcinoma; assays were performed in laboratories at 2 institutions. Identical results for both methods were found in 26 cases (10 amplified, 16 nonamplified). One case was misinterpreted as overexpressed by chromogenic in situ hybridization (CISH) because of background precipitate. In 4 cases, CISH suggested low-level amplification. Three of these cases subsequently were found to have chromosome 17 polysomy. In the remaining case, the initial section chosen was suboptimal, showing weak signals by both methods. If a probe for chromosome 17 (now available for CISH) is used in cases of questionable HER-2/neu amplification, CISH seems to be as accurate and more practical than FISH. The HER-2/neu (also known as c-erb-b2) oncogene is a member of the epidermal growth factor receptor family, and its amplification is known to be one of the most common genetic alterations associated with human breast cancer. 1 Detection of HER-2/neu oncogene amplification is necessary for selecting cancer patients for trastuzumab (Herceptin) therapy; it also may be important for treatment purposes in anthracycline-based and hormonal regimens 2 and for providing prognostic information. 3 Erroneous or inconclusive HER-2/neu test results may lead to inappropriate patient management 4 ; therefore, there is a critical need to develop a reliable assay to assess HER-2/neu oncogene status. Early molecular methods for HER-2/neu detection, such as the Southern, Northern, and Western blot techniques, were laborious and impractical for application in surgical pathology. 5 Subsequent immunohistochemical methods offered the advantage of preserving morphologic features and were readily integrated into most surgical pathology laboratories, where immunohistochemical analysis usually is a standard procedure for many other diagnostic applications. In addition, fresh tissue is not required, and the procedure can be performed on archival specimens. Unfortunately, widespread variation in results occurred because many different, heavily marketed, commercial antibodies were used. 6 More than 30 different antibodies, both commercially and privately developed, are being used. 5 In a retrospective study, immunohistochemical analysis was performed on paraffin sections of 187 breast cancers with documented HER-2/neu amplification by the Southern blot method. 6 Seven polyclonal and 21 monoclonal commercially available antibodies were used. Depending on the antibody used, a sensitivity of 6% to 82% was obtained. Am J Clin Pathol 2003;119: DOI: /P40P2EAD42PUKDMG 1

2 Gupta et al / HER-2 DETECTION BY CISH IN BREAST CANCER The standardization hoped for by the development of the US Food and Drug Administration (FDA)-approved HercepTest (DAKO, Carpinteria, CA) was not readily achieved, as the assay has many shortcomings. 4,6 The main problem overall with any immunohistochemical method for detecting HER-2/neu is that a negative result may not be reliable, since there is no positive internal control to check for fixation, processing, or assay problems. At the other end of the spectrum, positive results may not be reliable either, because of possible misinterpretations, such as scoring of the HercepTest, 7 or technical factors that contribute to the variability in the reported overexpression rates, including differences in tissue fixation or processing 8 or antigen retrieval methods. 9 Interobserver variability in the scoring of results has led to marked differences in HER-2/neu reporting. 10 Although the HercepTest is the one commercially available test kit that has received FDA approval, it has not been shown to be superior to other immunohistochemical methods, and it is more expensive. 11 Fluorescence in situ hybridization (FISH) is regarded as the gold standard for detecting HER-2/neu amplification: it has high sensitivity (96.5%) and specificity (100%) for detecting amplification of HER-2/neu. 12 It has the advantage that it can be performed on small tumor samples and on formalin-fixed and paraffin-embedded tissue samples, with tissue preparation having little or no effect on the testing. It also permits direct visualization of the gene amplification in the nuclei and provides an objective count of the genes and chromosomes on a cell-by-cell basis. However, it requires a fluorescence microscope and special training for interpretation. The cost of either of the 2 FDA-approved FISH assays (just reagents) is almost $100, with considerable cost added by the need to purchase a fluorescence microscope and filters. It also may be difficult to visualize the morphologic features of the tumor cells and to separate in situ from invasive carcinoma when evaluating the amplification product with fluorescence. In addition, fluorescence fades quickly and, therefore, is not a permanent record. Chromogenic in situ hybridization (CISH) seems to provide an attractive alternative to FISH. The tissue preparation and probe hybridization procedures of FISH and CISH are similar. They differ in the method of probe detection, which in the case of CISH is based on a peroxidase reaction that can be visualized by light microscopy, obviating the need for fluorescence, and the tissue can be visualized easily along with the amplification product. Thus, the invasive tumor and corresponding histologic or nuclear grade can be correlated immediately with HER-2/neu amplification. CISH also provides a permanent record and is less expensive than FISH. Since CISH is a relatively new technique, the correlation of CISH with FISH has not been fully established in clinical laboratories. The aims of the present study were to establish such a correlation in our hospital-based practice where CISH has been implemented, replacing a previously published immunohistochemical method 11 for detecting HER-2/neu expression in infiltrating breast carcinoma, and to provide practical information for others interested in using this technique. Materials and Methods During a 5-month period, approximately 50 cases of infiltrating breast carcinoma (all histologic types) were assayed for HER-2/neu oncogene amplification by CISH in the anatomic pathology department at St Luke s Episcopal Hospital, Houston, TX. From these, 31 cases were selected for this study. All tumors interpreted as amplified cases and low-level-amplified cases during that period were included, as were 16 cases of carcinoma showing no amplification. These nonamplified cases included 5 well-differentiated tumors, 7 moderately differentiated tumors, and 4 poorly differentiated tumors, using the Elston-Ellis modification of the Scarff-Bloom-Richardson histologic scoring system. 13 Chromogenic In Situ Hybridization All reagents for the assay were obtained from Zymed Laboratories (South San Francisco, CA), and the manufacturer s suggested method 14 was used. Briefly, 5-µm-thick, formalin-fixed, paraffin-embedded sections were deparaffinized with xylene, hydrated, and pretreated with enzyme and heat, as specified. Denaturation was carried out at 95 C, followed by hybridization at 37 C for 16 to 24 hours using a digoxigenin-labeled HER-2/neu probe. After hybridization, stringency washes and a blocking step were performed. Signals were detected using mouse antidigoxigenin and polymerized horseradish peroxidase goat antimouse followed by development with diaminobenzidine. Slides were counterstained with methyl green/alcian blue. A case of infiltrating ductal carcinoma, positive for HER- 2/neu oncoprotein by immunohistochemical analysis 11 and amplified by FISH, was used as a positive control; nonneoplastic breast tissue was used as a negative (or nonamplified) control. In most cases, nonneoplastic elements served as internal controls for each individual case. HER-2/neu oncogene was simultaneously evaluated by 2 of us (J.A., M.J.W.) using a dual-head Olympus BX41 microscope (Olympus America, Melville, NY) under either a 40 dry or 100 oil objective. As described by Tanner et al, 14 amplification was defined as greater than 10 discrete copies per nucleus or as large gene copy clusters (seen as confluent masses of more than 10 signals) in more than 50% of the nuclei evaluated (we evaluated at least 50 cells in all cases). Low-level amplification was defined as 6 to Am J Clin Pathol 2003;119: DOI: /P40P2EAD42PUKDMG

3 Anatomic Pathology / ORIGINAL ARTICLE copies per nucleus in more than 50% of cells. Unaltered gene copy was defined as 1 to 5 copies per nucleus. Fluorescence In Situ Hybridization FISH was performed using the PathVysion HER-2- DNA Probe Kit (Vysis, Downers Grove, IL) according to the manufacturer s instructions. The PathVysion kit applies 2 DNA probes directly labeled with different fluorescent dyes: SpectrumOrange fluorophore-labeled locus-specific identifier HER-2/neu, which is specific for the HER-2/neu gene locus on chromosome 17q , and SpectrumGreen fluorophore-labeled chromosome enumerator probe 17, which is targeted to the alpha satellite DNA sequence located at the centromeric region of chromosome 17. Briefly, 5-µm-thick, formalin-fixed, paraffin-embedded sections were mounted on positively charged slides, deparaffinized in xylene, dehydrated in 100% alcohol, and allowed to air dry. The slides then were pretreated in 0.2N hydrochloric acid, immersed in a 1-mol/L concentration of sodium isothiocyanate at 80 C for 10 minutes and washed, after which they were digested in protease solution for 15 minutes at 37 C. The slides were washed in distilled water for 3 minutes at room temperature and allowed to air dry. The sections were denatured and hybridized with the dual probe in the Hybrite denaturation/hybridization system (Vysis) and placed in a humidity chamber at 37 C overnight. Posthybridization washing was performed at 73 C for 2 minutes using 2 standard saline citrate with a 0.3% solution of NP-40 at room temperature. The slides were dried, counterstained with 4,6-diaminido-2-phenylindole dihydrochloride (DAPI), and coverslipped. Positive controls were purchased from Vysis. The slides were evaluated for HER- 2/neu gene copy number using a Zeiss epifluorescence microscope (Carl Zeiss, Thornwood, NY). Both signals were counted in 60 tumor nuclei according to established criteria. 15 Signals from overlapping nuclei were not counted, and split signals were counted as a single chromosome component. The adjacent ductal epithelial cells served as internal controls. The ratio of total HER-2/neu to the total CEP 17 signals in the 60 nuclei was calculated. The expected ratio of locus-specific identifier HER-2/neu to chromosome enumerator probe 17 is less than 2.0 for normal or nonamplified tumor tissue specimens. A ratio of 2.0 or greater was considered positive for HER-2/neu gene amplification, and the results were reported as amplified or nonamplified. The FISH results were evaluated independently by a pathologist (L.P.M.) who was unaware of the CISH results. Results Table 1 summarizes the results. Of 31 cases analyzed, 26 cases had identical results for both methods, with 10 tumors amplified and 16 nonamplified. Five cases were discordant. In 4 cases, the CISH method suggested low-level amplification Image 1A. With FISH, 3 of these cases were explained by polysomy for chromosome 17 Image 1B. Nonamplified tumors showed 1 to 5 clearly defined dots in the nucleus Image 2. Amplification was seen most often as large copy gene clusters in the majority of nuclei Image 3. Discussion The major aim of our study was to compare CISH with 1 of the 2 FDA-approved FISH assays presently used in diagnostic pathology laboratories and thus validate its use as our routine method for assessing HER-2/neu status on breast cancers and other tumors. At the time of this writing, only 4 studies reporting on this particular CISH technique for the detection of HER-2/neu in breast cancer exist in the diagnostic pathology literature. The first study, by Tanner et al, 14 compared the results from a large series of tumors in which CISH was performed on formalin-fixed, paraffin-embedded sections; and FISH was used on imprints of whole tumor nuclei prepared from fresh tissue samples. A probe for chromosome 17 was used in the FISH assay but not in the CISH assay. Differences in amplification rates (FISH, 23.6%; CISH, 17.2%) were speculated to have resulted from differences in preparations (sectioned vs whole nuclei). That study 14 served to define the criteria for amplification by the CISH method. Although their criteria for amplification by FISH are stated and include a low-level category, it is not entirely clear how Tanner et al 14 determined criteria for low-level Table 1 Correlation of HER-2/neu Analysis by CISH and FISH in 31 Breast Cancers FISH CISH Amplified Nonamplified Chromosome 17 Polysomy Amplification Low-level amplification No amplification CISH, chromogenic in situ hybridization; FISH, fluorescence in situ hybridization. Am J Clin Pathol 2003;119: DOI: /P40P2EAD42PUKDMG 3

4 Gupta et al / HER-2 DETECTION BY CISH IN BREAST CANCER A B Image 1 A, Tumor nuclei with more than 5 discrete brown dots consistent with low-level amplification by chromogenic in situ hybridization ( 1,000). B, Same case showing no amplification of HER-2/neu (ie, 2-4 copies per nucleus) and polysomy for chromosome 17 (inset) by fluorescence in situ hybridization ( 1,000). amplification by CISH or whether low-level amplification by the 2 methods correlated. Both in situ methods also were compared simultaneously with immunohistochemical analysis in the same series of tumors. Subsequently, a study by Kumamoto et al 16 compared the results of CISH with HercepTest results for 18 breast carcinomas. Only 4 of 18 tumors had a definitively positive (3+) immunohistochemical score, and 4 of 18 had the everproblematic and equivocal (2+) score. Four of these 8 cases showed high-level amplification by CISH, and 3 showed low-level amplification. The criteria used by Kumamoto et al 16 for low-level amplification were modified slightly from those used by Tanner et al, 14 and a probe for chromosome 17 was not used to address the possibility of chromosomal polysomy. Kumamoto et al 16 concluded that CISH could be a useful adjunct to immunohistochemical analysis. More recently, Zhao et al 17 reported a series of tumors in which the same DNA probe used in the Zymed CISH assay Image 2 Nonamplified tumor. Chromogenic in situ hybridization showing 1 to 4 clearly defined dots in each nucleus ( 1,000). Image 3 Amplification of HER-2/neu by chromogenic in situ hybridization, seen as large copy gene clusters in the majority of the nuclei ( 1,000). 384 Am J Clin Pathol 2003;119: DOI: /P40P2EAD42PUKDMG

5 Anatomic Pathology / ORIGINAL ARTICLE was first labeled with fluorescence for detection of HER- 2/neu, followed by sequential relabeling with a peroxidasediaminobenzidine detection system. The reported concordance for FISH and CISH of 100% was not surprising, since Zhao et al 17 switched only one detection label for another on the same slide after the DNA hybridization reaction had taken place. They did not actually compare CISH independently with either of the 2 FDA-approved FISH assays. A CISH probe for chromosome 17 was used in all cases in which there were more than 3 HER-2/neu gene copies per nucleus in a majority (>50%) of cells, either in nonamplified tumors or in amplified tumors. Thus, they were able to separate the cases of true amplification from chromosomal polysomy. They also compared the in situ hybridization technique with immunohistochemical analysis using various antibodies and confirmed the validity of the CISH method, concluding that CISH could prove very useful, assuming the consistency of results by other studies. Technical and interpretative aspects also were addressed. Finally, the most recent study, by Dandachi et al, 18 compared results of CISH, the HercepTest, and polymerase chain reaction in a large number of formalin-fixed, paraffin-embedded breast tumor specimens. Discrepant cases (between CISH and immunohistochemical analysis) were then assayed using the FISH PathVysion assay; concordance between CISH and FISH was 100%. The authors concluded that CISH was a promising alternative to immunohistochemical analysis or FISH in the routine diagnostic setting. Another new alternative technique to FISH has recently been reported. This technology has been termed GOLDFISH and uses gold-enhanced autometallography with in situ hybridization instead of the diaminobenzidine-based chromogenic detection system in CISH, also permitting direct visualization of gene amplification and the morphologic features under an ordinary light microscope. 19,20 Further studies comparing CISH and GOLDFISH with respect to cost and ease of applicability may prove beneficial. Since beginning the study with the 31 cases described herein, our experience with CISH at the time of this writing includes more than 150 different tumors. Although they have been infrequent, technical problems have occurred, including no signal or weak signal, overdigestion of tissue, and high background debris. Recognition and awareness of these potential problems are essential for optimal results. Maintaining consistency and routine with respect to tissue fixation and processing in a laboratory is not always possible. Thus, at St Luke s Episcopal Hospital, we usually recommend that HER-2/neu testing be performed on an excisional biopsy specimen, rather than a core biopsy specimen, to ensure optimal sampling and to avoid potential fixation issues, since inpatient material is better controlled. For patients who undergo preoperative chemotherapy, CISH is performed on the core biopsy specimen, and we have not encountered any technical or interpretive problems so far, as has occurred with immunohistochemical assays. We frequently are asked to perform the assay on archival specimens when patients experience relapse and the original assay (done by immunohistochemical analysis) result was negative or the primary tumor occurred before routine performance of HER-2/neu testing. Rarely, we have had to repeat the assay using a longer enzymatic digestion period if the signal was absent or weak, and this has corrected the problem. Early on, we found that using an alcian blue based counterstain provided better color contrast with the brown signal than the hematoxylin used in other studies, 14,16-18 and nuclear morphologic features were still visible. We have used Proteinase K (Sigma, St Louis, MO) successfully as an alternative to the SPOT-light pretreatment enzyme (contained in the Zymed CISH kit) when this kit reagent was depleted. It is critical to titrate the enzyme with respect to concentration and incubation time and to ensure that it is prepared properly; otherwise, underdigestion or overdigestion of tissue may occur, and the signal will be weak or absent. Another critical element in the assay is the oven temperature (95 C) in the denaturation step. As with any immunohistochemical or in situ hybridization assay, the manufacturer s suggested procedure is only a guideline. It is essential that each laboratory customize and optimize the various steps to correct for differences in laboratory conditions, such as fixation and processing, as well as those inherent in the various tissues. Misinterpretation of amplification occurred in our study. When we initially began using CISH, the probe for chromosome 17, where the HER-2/neu oncogene is located, was not available using CISH technology. As we suspected, most (3/4) of our low-level amplification cases showed polysomy for chromosome 17 by FISH, explaining the presence of more than 5 copies per nucleus. In 1 case, the signal was extremely weak and difficult to count accurately, even under oil magnification at 100. Subsequently we repeated CISH on a different block of tumor, and it clearly was nonamplified. In the majority of cases, true amplification is immediately apparent at a magnification of 10 or 20. We recently have begun using the CISH chromosome 17 probe in any case in which large gene copy clusters are not apparent but there seem to be more than 5 copies in the majority of nuclei. Our 1 falsepositive case of amplification, in which there appeared to be more than 10 copies per nucleus, was caused by overinterpretation of signal in the presence of high background, which in this case was due to a high level of endogenous peroxidase. As a result, a methanol hydrogen peroxide Am J Clin Pathol 2003;119: DOI: /P40P2EAD42PUKDMG 5

6 Gupta et al / HER-2 DETECTION BY CISH IN BREAST CANCER quenching step (considered optional by the assay manufacturer) has now been incorporated routinely. (The result of a repeated CISH assay using this quenching agent was negative.) Careful attention to nonneoplastic cells such as lymphoid, ductal, or endothelial cells (usually present near the tumor) should alert the pathologist to this problem, as these cells also will appear to be amplified. For clinical purposes, a valid unequivocal result is ideal. Maintaining a category of low-level amplification by the in situ hybridization method does not solve the problem of the indefinite results obtained by the immunohistochemical method. In many laboratories that use the FISH assay, this problem has been eliminated by using the ratio of HER- 2/neu signals to chromosome 17 signals and considering a ratio of 2.0 or more as amplification. It seems reasonable to implement this same definition, selectively, for amplification by CISH because using chromosome 17 in all cases would add unnecessary cost and time since the majority of breast cancers are negative for HER-2/neu. We suspect that the earlier reported rates of 25% to 30% 21 for HER-2/neu overexpression by immunohistochemical analysis in breast carcinomas may be based on false-positive interpretation owing to differences in scoring systems or on other factors that may affect the immunohistochemical assay 6-10 or, as recently suggested, 21 on chromosomal 17 polysomy. A valid argument could be made for selected HER- 2/neu testing, as certain histologic subtypes of breast cancer, as well as most well-differentiated tumors, are negative; our own experience has corroborated that reported in the literature. 21,22 Unfortunately, commercial marketing and the media have created an almost overinformed public. Pressure from clinicians and patients has virtually necessitated the routine testing of all breast cancers for HER-2/neu expression, and we have found that clinicians and patients alike have learned to question both positive and negative immunohistochemical results, requesting in situ hybridization whenever the result is not what they expected. To that end, we support the findings of the most recent studies on CISH by Dandachi et al 18 and Zhao et al 17 and conclude that CISH, by itself, is a practical, economic (at a material cost similar to that of the HercepTest), and definitive means for routinely assessing HER-2/neu in breast tumors. We have shown that the accuracy of CISH should be identical to that of FISH if proper technical procedure is followed and if careful attention is given to the interpretation of results. From the Departments of Pathology, 1 University of Texas Medical School, 2 M.D. Anderson Cancer Center, and 3 St Luke s Episcopal Hospital and Texas Heart Institute, Houston. Address reprint requests to Dr Abrams: Dept of Pathology, St Luke s Episcopal Hospital, 6720 Bertner St, MC4-265, Houston, TX Acknowledgments: We thank Marianne Mallia-Hughes and staff for their editorial assistance. References 1. Sahin AA. Biologic and clinical significance of HER- 2/neu (c-erbb-2) in breast cancer. Adv Anat Pathol. 2000;7: Baselga J. Herceptin alone or in combination with chemotherapy in the treatment of HER-2 positve metastatic breast cancer: pivotal trials. Oncology. 2001;61(suppl 2): Hynes NE. Amplification and overexpression of the erbb-2 gene in human tumors: its involvement in tumor development, significance as a prognostic factor, and potential as a target for cancer therapy. Semin Cancer Biol. 1993;4: Cell Markers and Cytogenetics Committees College of American Pathologists. Clinical laboratory assays for HER- 2/neu amplification and overexpression: quality assurance, standardization, and proficiency testing. Arch Pathol Lab Med. 2002;126: Hanna W. Testing for HER-2 status. Oncology. 2001;61(suppl 2): Press MF, Hung G, Godolphin W, et al. Sensitivity of HER- 2/neu antibodies in archival tissue samples: potential source of error in immunohistochemical studies of oncogene expression. Cancer Res. 1994;54: Jacobs TW, Gown AM, Yaziji H, et al. Specificity of HercepTest in determining HER-2/neu status of breast cancers using the United States Food and Drug Administration approved scoring system. J Clin Oncol. 1999;17: Penault-Llorca F, Adelaide J, Houvenaeghel G, et al. Optimization of immunohistochemical detection of erbb-2 in human breast cancer: impact of fixation. J Pathol. 1994;173: Farabegoli F, Ceccarelli C, Santini D, et al. c-erbb-2 overexpression in amplified and non-amplified breast carcinoma samples. Int J Cancer. 1999;84: Kay EW, Walsh CJ, Cassidy M, et al. c-erbb-2 immunostaining: problems with interpretation. J Clin Pathol. 1994;47: Vang R, Cooley LD, Harrison WR, et al. Immunohistochemical determination of HER-2/neu expression in invasive breast carcinoma. Am J Clin Pathol. 2000;113: Pauletti G, Godolphin W, Press M, et al. Detection and quantitation of HER-2/neu gene amplification in human breast cancer archival material using fluorescence in situ hybridization. Oncogene. 1996;13: Elston CW, Ellis IO. Pathological prognostic factors in breast cancer, I: the value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology. 1991;19: Tanner M, Gancberg D, Di Leo A, et al. Chromogenic in situ hybridization: a practical alternative for fluorescence in situ hybridization to detect HER-2/neu oncogene amplification in archival breast cancer samples. Am J Pathol. 2000;157: Hopman AH, Ramaekers FC, Raap AK, et al. In situ hybridization as a tool to study numerical chromosome aberrations in solid bladder tumors. Histochemistry. 1988;89: Am J Clin Pathol 2003;119: DOI: /P40P2EAD42PUKDMG

7 Anatomic Pathology / ORIGINAL ARTICLE 16. Kumamoto H, Sasano H, Taniguchi T, et al. Chromogenic in situ hybridization analysis of HER-2/neu status in breast carcinoma: application in screening of patients for trastuzumab (Herceptin) therapy. Pathol Int. 2001;51: Zhao J, Wu R, Au A, et al. Determination of HER-2 gene amplification by chromogenic in situ hybridization (CISH) in archival breast carcinoma. Mod Pathol. 2002;15: Dandachi N, Dietze O, Hauser-Kronberger C. Chromogenic in situ hybridization: a novel approach to a practical and sensitive method for the detection of HER2 oncogene in archival human breast carcinoma. Lab Invest. 2002;82: Tubbs R, Skacel M, Pettay J, et al. Interobserver interpretative reproducibility of GOLDFISH, a first generation goldfacilitated autometallographic bright field in situ hybridization assay for HER-2/neu amplification in invasive mammary carcinoma. Am J Surg Pathol. 2002;26: Tubbs R, Pettay J, Skacel M, et al. Gold-facilitated in situ hybridization: a bright-field autometallographic alternative to fluorescence in situ hybridization for detection of HER-2/neu gene amplification. Am J Pathol. 2002;160: Bose S, Mohammed M, Shintaku P, et al. Her-2/neu gene amplification in low to moderately expressing breast cancers: possible role of chromosome 17/HER-2/neu polysomy. Breast J. 2001;7: Hoff EF, Tubbs RR, Myles JL, et al. HER-2/neu amplification in breast cancer: stratification by tumor type and grade. Am J Clin Pathol. 2002;117: Am J Clin Pathol 2003;119: DOI: /P40P2EAD42PUKDMG 7