|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Evaluation of HER-2/neu Gene Amplification and Overexpression: Comparison of Frequently Used Assay Methods in a Molecularly Characterized Cohort of Breast Cancer SpecimensByFrom the Breast Cancer Research Program of the Lee Breast Center, Department of Pathology and Department of Preventive Medicine, Norris Comprehensive Cancer Center, University of Southern California; Division of Hematology and Oncology, University of California Los Angeles School of Medicine, Los Angeles, CA; and Vysis, Inc, Downers Grove, IL. Address reprint requests to: Michael F. Press, MD, PhD, Norris Topping Tower Room 5409, 1441 Eastlake Ave, U.S.C./Norris Comprehensive Cancer Center, Los Angeles, CA 90033; email: press@ hsc.usc.edu.
PURPOSE: To compare and evaluate HER-2/neu clinical assay methods. MATERIALS AND METHODS: One hundred seventeen breast cancer specimens with known HER-2/neu amplification and overexpression status were assayed with four different immunohistochemical assays and two different fluorescence in situ hybridization (FISH) assays. RESULTS: The accuracy of the FISH assays for HER-2/neu gene amplification was high, 97.4% for the Vysis PathVision assay (Vysis, Inc, Downers Grove, IL) and 95.7% for the the Ventana INFORM assay (Ventana, Medical Systems, Inc, Tucson, AZ). The immunohistochemical assay with the highest accuracy for HER-2/neu overexpression was obtained with R60 polyclonal antibody (96.6%), followed by immunohistochemical assays performed with 10H8 monoclonal antibody (95.7%), the Ventana CB11 monoclonal antibody (89.7%), and the DAKO HercepTest (88.9%; Dako, Corp, Carpinteria, CA). Only the sensitivities, and therefore, overall accuracy, of the DAKO Herceptest and Ventana CB11 immunohistochemical assays were significantly different from the more sensitive FISH assay. CONCLUSION: Based on these findings, the FISH assays were highly accurate, with immunohistochemical assays performed with R60 and 10H8 nearly as accurate. The DAKO HercepTest and the Ventana CB11 immunohistochemical assay were statistically significantly different from the Vysis FISH assay in evaluating these previously molecularly characterized breast cancer specimens.
A LARGE NUMBER of molecular genetic alterations have now been identified in human breast cancers. These include both inherited and acquired genetic alterations. Among acquired alterations HER-2/neu gene amplification has received a great deal of attention recently. This is because of its potential clinical utility as a prognostic marker and as a predictor of responsiveness to treatment. HER-2/neu (c-erbB-2) gene amplification, identified in 20% to 30% of breast cancers, is a prognostic marker of poor clinical outcome in node-negative and node-positive breast cancer and a predictor of lack of responsiveness to tamoxifen antiestrogen therapy and responsiveness to adjuvant doxorubicin chemotherapy. The development of an immunotherapeutic, trastuzamab (Herceptin; Genentech, Inc, South San Francisco, CA), to the extracellular domain of HER-2/neu has provided a new treatment for women with metastatic, HER-2/neu-overexpressing breast cancer. The importance of HER-2/neu in breast cancer treatment decision making has focused attention on the ability of various clinical assays to correctly assign HER-2/neu amplification and overexpression status. One of us (D.J.S.) first identified HER-2/neu gene amplification in breast cancers 15 years ago using Southern hybridization of frozen breast cancer specimens.1 Subsequently, we have shown a correlation between gene amplification and an increased level of HER-2/neu expression, referred to as overexpression, at the mRNA and protein product levels.2 Overexpression of HER-2/neu in cultured cells and transgenic mice is associated with malignant transformation in the cells overexpressing this gene.3-5 Although initially controversial, we and others have shown that HER-2/neu gene amplification or overexpression predicts shorter disease-free survival or shorter overall survival in both axillary lymph node–negative6-17 and node-positive breast cancer.1,2,8,18,19 In addition, HER-2/neu overexpression has been associated with responsiveness to cytotoxic chemotherapy12,20-25 and resistance to tamoxifen antiestrogen therapy.26-29 Experimental work with animal models and clinical trials with women having HER-2/neu-overexpressing breast cancer have shown tumor regression in response to anti–HER-2/neu antibody therapy. Recently, the United States Food and Drug Administration (FDA) approved an immunotherapeutic agent for women with metastatic, HER-2/neu-overexpressing breast cancers. This immunotherapeutic is a recombinant, humanized anti–HER-2/neu monoclonal antibody, trastuzamab. The FDA has also approved two fluorescence in situ hybridization (FISH) assays for clinical testing of HER-2/neu gene amplification (Ventana Medical Systems, Inc, Tucson, Az, and Vysis, Inc, Downers Grove, IL) and two immunohistochemical assays for HER-2/neu overexpression (DAKO, Corp, Carpinteria, CA, and Ventana Medical Systems, Inc). The FDA approval for gene amplification tests was originally for identification of women with node-negative disease at high-risk for recurrence or disease-related death or for selection to doxorubicin chemotherapy. A recently approved application to the FDA broadened the indications for FISH testing to include selection of women with metastatic breast cancer for trastuzamab therapy. Both immunohistochemistry assays are approved for selection of women with metastatic breast cancer to receive trastuzamab treatment. Previous molecular analyses of frozen breast cancer specimens indicate that amplification is closely associated with overexpression;2,18 however, the detection rates for immunohistochemistry reported by clinical laboratories are different from one another and, for at least one of the immunohistochemical assays, higher than expected.30,31 This apparent discrepancy has generated considerable confusion among clinicians, pathologists, and patients regarding appropriate methods for HER-2/neu testing. To address this issue, we have used each of the four FDA-approved assays and two additional immunohistochemical assays to test for HER-2/neu alterations in breast cancer specimens in which we previously determined amplification and overexpression levels using solid matrix blotting techniques.
Breast Cancer Specimens Breast cancer specimens were selected for inclusion in this study based on the following criteria: (1) specimens were previously molecularly characterized for HER-2/neu gene amplification and overexpression using solid matrix blotting techniques,2,6,18,32 and (2) the analyses of amplification and expression showed agreement between amplification status and expression status. To minimize inclusion of tumors that were misclassified because of dilutional artifacts associated with solid matrix blotting methods, samples selected as representative of gene amplification were required to show both amplification and overexpression, whereas cases selected as representative of nonamplified samples were required to show neither gene amplification nor overexpression by solid matrix blotting techniques. Tumors showing a disagreement between the amplification and expression status were excluded from the analysis. HER-2/neu gene amplification was previously determined by solid matrix blotting (either Southern hybridization or slot/dot blot hybridization) using DNA extracted from breast cancer specimens.2,11 HER-2/neu expression was assessed by either solid matrix blotting of RNA or protein extracted from breast cancer specimens (Northern hybridization, dot blot hybridization, and Western immunoblot) or immunohistochemistry. If expression was assessed by more than one method, then the expression level was assigned according to the classification determined by the majority of assay methods. Ninety-eight cases had expression data available from two or more assay methods.2,18 The remaining 19 cases had expression data available from only a single assay method.6 Breast cancer specimens with known HER-2/neu gene copy and expression levels were evaluated in this study as archival, paraffin-embedded tissue specimens. The specimens were arrayed in two paraffin-embedded, multitumor specimen blocks, as described elsewhere in detail.32 The histopathology of all specimens remaining in the multispecimen blocks was reviewed, and only specimens containing breast carcinoma cells were included in the study. One hundred seventeen cases met all of the above criteria and were included in this study with 43 breast cancers showing both amplification and overexpression and 74 breast cancers showing neither amplification nor overexpression. The multitumor specimen block approach permitted the use of small quantities of reagents and, more importantly, assured equal exposure of each breast cancer to all of the reagents in the assay system.
FISH for Gene Amplification Using Both HER-2/neu and Chromosome 17 Centromere Probes FISH was performed according to the PathVysion (Vysis, Inc) protocol, which is described in the package insert as approved by the United States FDA. In brief, the PathVysion protocol involved rehydration of paraffin-embedded, 4-µm thick, multitumor tissue sections through Hemo-De clearing agent (Vysis, Inc) and a graded series of alcohols. The sections were air dried, pretreated (80°C, 30 minutes), and digested with protease (37°C, 10 to 20 minutes) before hybridizing with fluorescent-labeled probes for HER-2/neu gene and alpha-satellite DNA for chromosome 17. The locus specific identifier (LSI) HER-2/neu DNA probe is a 190-Kb SpectrumOrange (Abbott-Vysis, Inc) directly labeled fluorescent DNA probe specific for the HER-2/neu gene locus (17q11.2-ql2). The chromosome enumeration probe (CEP) 17 DNA probe is a 5.4-Kb SpectrumGreen (Abbott-Vysis, Inc) directly labeled fluorescent DNA probe specific for the alpha satellite DNA sequence at the centromeric region of chromosome 17 (17p11.1-q11.1). The probes were premixed and predenatured in hybridization buffer for ease of use. The nuclei were routinely counterstained with an intercalating fluorescent counterstain, 4'-6'-diamidino-2'-phenylindole (DAPI). Staining was visualized with a Zeiss Axioskop 20 fluorescence microscope (Carl Zeiss, Inc, Oberkochen, Germany). SK-BR-3 human breast cancer cells, known to have HER-2/neu gene amplification, and MDA-MB-468 human breast cancer cells, known to lack HER-2/neu gene amplification, were used as control cells (American Type Culture Collection, HTB30 and HTB132)34 (unpublished data). For each tumor, 60 tumor cell nuclei were identified and scored for both HER-2/neu and chromosome 17 centromere number. Using criteria established for Southern hybridization,2,16,18 HER-2/neu gene amplification was defined as a HER-2/neu-to-chromosome 17 ratio greater than 2.0 as required by the manufacturer.
FISH to Evaluate Gene Amplification Using Only a HER-2/neu Probe For each specimen, gene copy level was assessed in two areas of 20 nonoverlapping tumor cell nuclei. A tumor was amplified if there were more than four copies of HER-2/neu per cell, as described elsewhere7 and as required by the manufacturer (Ventana, Inc). Staining was visualized with a Zeiss fluorescence microscope. SK-BR-3 human breast cancer cells, known to have HER-2/neu gene amplification, and MDA-MB-468 human breast cancer cells, known to lack HER-2/neu gene amplification, were used as control cells.
Immunohistochemical Assay for HER-2/neu Protein Using the R60 Polyclonal Antibody
Immunohistochemical Assay for HER-2/neu Protein Using the 10H8 Monoclonal Antibody
DAKO HercepTest for Immunoenzymatic Staining to Detect HER-2/neu Protein
Immunohistochemical Assay for HER-2/neu Protein Using the CB11 Monoclonal Antibody
Statistical Analyses
One hundred seventeen breast cancer specimens with known HER-2/neu amplification and overexpression status determined by solid matrix blotting techniques and immunohistochemistry in previous studies2,6,11 were reanalyzed as paraffin-embedded tissue sections for gene amplification using two FISH methods and for HER-2/neu overexpression using four different immunohistochemical assays.
FISH for Gene Amplification Using Both HER-2/neu and Chromosome 17 Centromere Probes
FISH to Evaluate Gene Amplification Using Only a HER-2/neu Probe The number of HER-2/neu gene copies was determined in 20 tumor cell nuclei for each of two microscopic fields in the 117 breast cancer specimens using FISH with a digoxigenin-labeled cosmid probe (Ventana, Inc) (Fig 1). The average number of HER-2/neu gene copies per tumor cell nucleus ranged from 1.28 to 25.50; values greater than or equal to 4.0 demonstrate gene amplification. The sensitivity of this FISH assay was 95.4% (41 of 43 breast cancers); the specificity was 96.0% (71 of 74); and the accuracy was 95.7% (112 of 117) (Table 1). The degree of concordance (kappa) was 0.909 (95% CI, 0.830 to 0.987). Forty-four of the 117 breast cancers had more than 4.0 HER-2/neu gene copies per tumor cell nucleus, and 73 had less than 4.0 HER-2/neu gene copies per tumor cell nucleus. One of the three cases with a false-positive result was the same case as the single false-positive result found with the Vysis FISH assay (average of 8.30 HER-2/neu gene copies per tumor cell nucleus). The other two false-positive results were above the cutoff of 4.0 (4.93 and 4.38). The two false-negative cases had averages of 3.58 and 3.97 HER-2/neu gene copies per tumor cell nucleus. These were the same two cases identified as false-negatives with the Vysis FISH assay (above).
Immunohistochemical Assay for HER-2/neu Protein Using the R60 Polyclonal Antibody
Immunohistochemical Assay for HER-2/neu Protein Using the 10H8 Monoclonal Antibody
DAKO HercepTest for Immunoenzymatic Staining to Detect HER-2/neu Protein
Immunohistochemical Assay for HER-2/neu Protein Using the CB11 Monoclonal Antibody
Comparison of FISH and Immunohistochemical Assays The same two cases, known to be amplified by previous analyses, were not scored as amplified with either the Vysis FISH assay (ratios less than 2.0; ie, 1.98 and 1.89 by the Vysis PathVision assay system) or the Ventana FISH assay (HER-2/neu gene copy levels less than 4.0; ie, 3.58 and 3.97). Both of these cases showed moderate (2+) immunostaining with the polyclonal R60 immunohistochemical assay but not with any of the other immunohistochemical assays. One breast cancer, known to lack amplification by previous analyses, was evaluated as amplified with both FISH assays (HER-2/neu-to-chromosome 17 ratio of 4.06 by the Vysis assay and a HER-2/neu gene copy number of 8.30 by the Ventana assay). Membrane immunostaining was detected in this case by 10H8 monoclonal antibody and the HercepTest (1+) but was not sufficiently strong to be scored as overexpression by either assay. A second breast cancer, known to lack amplification by previous analyses, was evaluated as amplified with only one of the two FISH assays (a HER-2/neu gene copy number of 4.93 by the Ventana assay). None of the immunohistochemical assays demonstrated any membrane immunostaining in this breast cancer.
Detection at Various Amplification/Overexpression Levels
The potential role of HER-2/neu gene amplification and/or overexpression in the clinical management of women with breast cancer has been controversial since the first report.1 The initial debate was whether HER-2/neu alterations correlated with poor clinical outcome in any group of women with breast cancer.42-45 Subsequently, HER-2/neu overexpression was accepted as a predictor of disease-free survival or overall survival in women with node-positive breast cancer.46 Although some studies clearly demonstrate that node-negative patients with HER-2/neu amplification/overexpression also have shorter disease-free survival or overall survival,6-9,16 others have disputed this association.23,45,47,48 Some investigators have suggested that the differences in clinical outcome observed among patients with HER-2/neu-overexpressing node-positive breast cancer are associated not with biologic behavior of the breast cancer cells but rather with responsiveness to treatment,46 accounting for the differences observed in node-negative and node-positive breast cancers. However, several studies of clinical outcome among women with node-negative breast cancer treated solely by surgery demonstrated significant survival benefits associated with low expression of HER-2/neu compared with HER-2/neu overexpression.6,7,14,16 Lack of agreement among these studies may also be related to the differing sensitivities and specificities of the assay methods used to assess HER-2/neu as a prognostic marker6,32 and to lack of statistical power of some of the studies.6,8 Although HER-2/neu gene amplification or overexpression seems to be a marker of poor prognosis in both node-negative and node-positive breast cancer, the major interest in this alteration is related to its utility as a predictive marker of responsiveness to treatment. Some studies have shown HER-2/neu overexpression predicts responsiveness to high-dose doxorubicin chemotherapy,20,25 lack of responsiveness to cyclophosphamide, methotrexate, and fluorouracil chemotherapy,12,23 and lack of responsiveness to tamoxifen therapy.26,28 However, HER-2/neu’s role as a predictor of responsiveness to therapy has been questioned because these initial observations have not been confirmed in some subsequent investigations.24,49 Recently, a recombinant humanized, anti–HER-2/neu antibody, known as trastuzamab, was approved by the FDA for treatment of women with metastatic breast cancer whose tumors overexpress HER-2/neu. Although trastuzamab has shown efficacy in this clinical setting, the response rate has been variable in the initial clinical trials.50-52 A major problem when interpreting the published studies of HER-2/neu in the clinical setting is that different methods have been used to evaluate HER-2/neu alterations. These methodologic differences likely contribute substantially to these conflicting results and, therefore, conflicting conclusions related to HER-2/neu. The primary method used in the majority of these published studies is immunohistochemistry using archival, paraffin-embedded tissue blocks. However, this method is not performed consistently across laboratories. Furthermore, few of these laboratories have validated their immunohistochemical staining assessments using standardized specimens with known molecular changes. We previously demonstrated that available antibodies commonly used for immunohistochemistry have a wide range of sensitivities when used in molecularly characterized, archival, paraffin-embedded specimens.32 None of the tested antibodies identified all of the known HER-2/neu–amplified, overexpression breast cancers. These antibodies were developed from animals immunized with native, nonfixed HER-2/neu protein as antigen. Therefore, it is not surprising that HER-2/neu would not be recognized in some formalin-fixed breast cancer specimens because none of the HER-2/neu antibodies were produced using formalin-fixed HER-2/neu protein as antigen. Our previous evaluation of HER-2/neu antibodies32 has been criticized because epitope retrieval methods were not used in our evaluation, the peroxidase antiperoxidase technique was used with all antibodies, and small samples of each tissue specimen were used.31,53,54 The primary goal of our previous study was to demonstrate that differences in antibody sensitivity could account for the differing conclusions about HER-2/neu’s clinical utility described among published articles. At the time our study was conducted, none of the published articles had used heat-induced epitope retrieval as part of the immunohistochemical procedure. As we described in our article, "although the sensitivities of various antibodies have been characterized under the described routine conditions permitting more rational selection of immunostaining reagents, it is anticipated that there will be further improvements in HER-2/neu sensitivity in paraffin-embedded tissues as more antibodies become available for testing and as techniques for immunostaining improve. For example, the sensitivity of several of these antibodies is enhanced by ‘antigen retrieval’ with microwave pretreatment of the tissue sections".32 We consider the use of a single immunostaining detection system in the previous study to be important for comparison between different antibodies on an equal basis. Therefore, the identified differences are due to differences in the primary antibodies and not due to differences in detection systems. Finally, we consider the use of small specimens to be an important asset in the study design because it ensured that the same area of the specimen was evaluated with each antibody, ensuring consistency in the comparison areas. Our primary objective in the current study is different from the previous study. It is well-recognized that differences in the methods used to evaluate HER-2/neu can lead to substantial differences in conclusions regarding clinical utility. Our goal in this study is to compare FDA-approved assay methods with one another and with our own in-house immunohistochemistry assays to determine which of these is the most accurate. Variability in sensitivity is expected because formalin-fixation and paraffin-embedding adversely effect antigenicity and immunostaining of the vast majority of protein antigens and the ability of antibodies to recognize their epitope varies from antibody to antibody. Hence, one expects that formalin-fixation and paraffin-embedding will be associated with reduced immunostaining relative to that observed in frozen tissue sections from the same specimen.2 Many believe that the loss of antigenicity in paraffin-embedded specimens can be corrected by application of heat-induced antigen retrieval to the paraffin-embedded tissue sections such as is performed with the DAKO Herceptest. However, once again, this approach has not been validated before introduction of these reagents and methods by systematic evaluation of clinical breast cancer specimens with known amplification/overexpression levels. In September 1998, the DAKO Herceptest was approved by the FDA to evaluate HER-2/neu expression for patient eligibility to receive trastuzamab treatment. The evaluation consisted of a comparison of immunostaining obtained with the Herceptest to immunostaining obtained with the "Clinical Trials Assay" (CTA) previously used by Genentech, Inc, to screen patient tumors for entry to their pivotal trastuzamab clinical trials (H0648, H0649, and H0650).36-38 The CTA used CB11 and 4D5 mouse monoclonal antibodies with an indirect avidin-biotin technique to detect HER-2/neu membrane receptor protein. Immunostaining with the Herceptest was found to be in agreement with the immunostaining result obtained from the CTA assay in 79% of 548 paraffin-embedded breast cancer specimens. The Ventana Pathway CB11 immunohistochemical assay was approved by the FDA as a diagnostic assay to aid in the selection of patients likely to benefit from trastuzamab therapy in November, 2000. Approval of the CB11 Pathway assay was based on concordance analysis between the DAKO Herceptest and Pathway CB11 assays. Immunostaining with Pathway CB11 was found to be in agreement with the Herceptest in 86% of 450 paraffin-embedded breast cancer specimens. Neither of these immunohistochemical assay comparisons was performed with specimens molecularly characterized for HER-2/neu gene amplification or overexpression. None of the specimens used for comparison had clinical outcome information available nor were any of these women part of a trastuzamab clinical trial. In the past, few investigators have provided data related to the screening or evaluation of HER-2/neu antibodies for use as immunohistochemical reagents, especially in formalin-fixed, paraffin-embedded tissue sections before using those antibodies to analyze study cohorts.6,32,55 Consequently, the conclusions of prognostic marker evaluations and predictive marker studies performed using these uncharacterized immunohistochemical assays are open to criticism. Currently, several groups have reported immunohistochemical staining with primarily the DAKO HercepTest or the Ventana immunohistochemical assay using CB11 monoclonal antibody. Some groups have compared immunohistochemical staining with these reagents to gene amplification determined by FISH with either the Vysis assay or the Ventana assay systems.31,54 In our current study, FISH analyses were compared with immunohistochemical assay evaluations of a group of molecularly characterized breast cancer cases. FISH of the HER-2/neu gene involves hybridization of a specific DNA probe with sequences complementary to the gene to permit assessment of the number of copies of this gene in individual tumor cell nuclei. FISH technology combines the advantages of molecular methods permitting localization of HER-2/neu gene copies in morphologically identified tumor cell nuclei. Immunohistochemistry involves recognition of antigenic determinants of p185HER-2/neu in the tumor cell membranes and permits a cell-by-cell assessment of protein expression. However, HER-2/neu immunohistochemistry suffers from a lack of characterization of which fixatives or fixation times are suitable for preservation of antigen immunoreactivity, lack of adequate internal immunostaining controls, and subjective assessment of membrane immunostaining. Although the range of fixatives most appropriate for FISH are also not characterized, the use of normal cells, present in each breast cancer specimen, as internal hybridization controls to confirm localization of an average of two HER-2/neu gene copies and two chromosome 17 centromeres per normal cell nucleus permits an assessment of the appropriateness of the results. Both FISH and immunohistochemistry are applicable to tumors of all sizes and can be performed on sections from the original specimen blocks used for diagnosis. Further, if carcinoma cells were localized rather than diffusely distributed within a tumor, amplification would be detectable by FISH but, in stroma-rich breast cancers, could be diluted below detectable limits by solid matrix blotting of DNA extracted from the tumor due to abundant noncarcinoma cellular DNA. Because of confusion about the reliability of various HER-2/neu assays for entry onto clinical trials evaluating responsiveness to experimental treatments with trastuzamab and because accurate assignment of women with breast cancer for clinical management decisions is now so important, we evaluated two FISH assay methods for gene amplification and a group of HER-2/neu immunohistochemical assays that are either in wide-spread use for clinical testing (Ventana CB11 and DAKO HercepTest) or have been used in our own laboratory (R60 and 10H8) for assessment of HER-2/neu overexpression, comparing results with these assays to the results of HER-2/neu gene amplification/overexpression obtained using standard molecular biologic methods. Our findings demonstrated that the FISH assays have higher sensitivity and higher accuracy and more frequently correctly identify altered HER-2/neu status (amplification/overexpression) in previously molecularly characterized specimens than did the FDA-approved immunohistochemistry assays interpreted manually. The in-house immunohistochemical assays most frequently used in our laboratory had a sensitivity that was slightly lower but did not differ significantly from the FISH assay results. Based on these findings, if one’s selection of HER-2/neu assays is restricted to commercially available assays, we recommend the use of FISH assays until the short-comings of tissue processing, limited antibody sensitivity for fixed antigen, and variable immunohistochemical scoring can be satisfactorily resolved. At the time when these issues are adequately resolved assay comparability should be reassessed. Standardized thresholds for interpretation of immunohistochemical staining that can be exported from laboratory to laboratory and pathologist to pathologist are needed. Although manual interpretation of HER-2/neu immunostaining in our laboratory and in the laboratories of some other investigators53 has been performed in a standardized fashion, the results obtained by others in other laboratories can be quite different. Our approach to manual or subjective interpretation of HER-2/neu immunostaining has been different from that of several other investigators. Few have used our in-house reagents because they have been produced in our laboratories and are not commercially available.2,18,35 Our approach to scoring immunostaining also differs from that of other investigators. Some have used the percentage of immunostained tumor cells to score overexpression with membrane immunostaining with a certain percentage of positively stained tumor cells considered overexpression,20,24,28,29,56,57 whereas we consider membrane intensity, not percentage of stained cells, to be the most important criterion. Our experience with frozen tissues demonstrates relatively little cell-to-cell heterogeneity within a given cancer. Cancers with amplification and overexpression have overexpression in essentially all tumor cells to a similar level.2,18 However, formalin-fixation and paraffin-embedding of tumor tissue from breast cancers previously evaluated as frozen specimens demonstrates both a loss of immunoreactivity and the introduction of heterogeneity that was not present in the original material.2,6 In contrast to the experience of some investigators,30,31 in this study we had an exceptionally low rate of false-positive results observed with heat-induced antigen retrieval performed with the DAKO HercepTest immunohistochemical assay. However our false-negative rate was significant. There were no characteristics of the immunostains (or the tumor morphology) that permitted us to predict which cases would be false-negatives with any of the immunohistochemical stains. This high false-negative rate and low false-positive rate may be related to differences in tissue processing because our tissues were processed in buffered formalin not in alcoholic formalin.30,31 However, the problem of wide variability in interpretation of HER-2/neu immunostaining has become increasingly apparent. The high false-positive results observed by some investigators may not be only due to differences in tissue fixation and specimen processing, they may be related to differences in interpretation of the immunostained specimens. For example, in our clinical reference laboratory we have observed substantial variability in the interpretation of HER-2/neu status for immunostained slides or tissue samples provided to obtain a second or third opinion (unpublished data). These preliminary results showed agreement of our laboratory assay results and interpretations with the immunohistochemistry assays performed and interpreted by outside pathologists in only approximately 53% of cases with regard to interpretation of HER-2/neu overexpression versus low expression status (2+/3+ v 0/1+). Similarly, a review of the first 680 cases analyzed in our centralized Breast Cancer International Research Group (BCIRG) laboratory for evaluation of HER-2/neu gene amplification as part of the entry criteria for BCIRG005 and BCIRG006 clinical trials demonstrated concordance with the outside or local immunohistochemistry staining result in only approximately 83% of breast cancers (unpublished data). Most of these discrepancies with outside laboratories in both our reference laboratory specimens and our BCIRG Central Laboratory specimens were false-positive results in the outside laboratory. In addition to differences in tissue processing, assay reagents used in immunohistochemistry and differences in performing the immunohistochemical assay with or without antigen retrieval and differences in subjective interpretation of the assay can lead to substantial differences in assessment of overexpression in immunohistochemical assays. Such striking variability in interpretation of FISH results has not been demonstrated. The College of American Pathologists (CAP) reported the CAP survey results for proficiency in HER-2/neu testing.58 In contrast to the results from immunohistochemistry laboratories, complete concordance was observed among the results obtained from laboratories performing FISH assays.58 Our experience has been similar with FISH in our centralized BCIRG laboratory compared with FISH performed in referring laboratories showing a concordance rate of 96% (unpublished data). In conclusion, FISH assays proved to be highly sensitive and specific for detecting HER-2/neu gene amplification. Although less sensitive, our interpretations of in-house immunohistochemical assays were not significantly different from the FISH assays for identification of overexpression. However, two commercially available, FDA-approved immunohistochemical assays were significantly less sensitive at detecting breast cancers with overexpression and were, therefore, less accurate.
Supported by grant no. CA48780 from the National Cancer Institute, grant no. NO1-HD-3-3175 from the National Institute for Child Health and Human Development, Bethesda, MD, and grant no. BCRP 5EB-3101 from the California Breast Cancer Research Program, Oakland, CA. J.P. was supported by grant no. DAMD17-97-1-7161 from the United States Army Medical Research and Materiel Command, Fort Detrick, MD. We thank Ivonne Villalobos for assistance in the preparation of this manuscript, Craig Sawtelle from Ventana Medical Systems, Inc, for participating in CB11 immunostaining, and Karen Petrosyan, MD, for technical assistance.
1. Slamon D, Clark G, Wong S, et al: Human breast cancer: Correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235: 177-182, 1987
2. Slamon D, Godolphin W, Jones L, et al: Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244: 707-712, 1989 3. Muller W, Sinn E, Pattengale P, et al: Single-step induction of mammary adenocarcinoma in transgenic mice bearing the activated c-neu oncogene. Cell 54: 105-115, 1988[CrossRef][Medline] 4. Chazin V, Kaleko M, Miller A, et al: Transformation mediated by the human HER-2 gene independent of the epidermal growth factor receptor. Oncogene 7: 1859-1866, 1992[Medline]
5. DiFiore P, Pierce J, Kraus M, et al: erb B-2 is a potent oncogene when overexpressed in NIH/3T3 cells. Science 237: 178-182, 1987
6. Press M, Pike M, Chazin V, et al: HER-2/neu expression in node-negative breast cancers: Direct tissue quantitation by computerized image analysis and association of overexpression with increased risk of recurrent disease. Cancer Res 53: 4960-4970, 1993 7. Press M, Bernstein L, Thomas P, et al: HER-2/neu gene amplification by fluorescence in situ hybridization: Evaluation of archival specimens and utility as a marker of poor prognosis in node-negative invasive breast carcinomas. J Clin Oncol 15: 2894-2904, 1997[Abstract] 8. Gullick W, Love S, Wright C, et al: c-erb B-2 protein over-expression in breast cancer is a risk factor in patients with involved and uninvolved lymph nodes. Br J Cancer 63: 434-438, 1991[Medline] 9. Kallioniemi O-P, Holli K, Visakorpi T, et al: Association of c-erb B-2 oncogene overexpression with high rate of cell proliferation, increased risk for visceral metastasis and poor long-term survival in breast cancer. Int. J Cancer 49: 650-655, 1991 10. Winstanley J, Cooke T, Murray G, et al: The long term prognostic significance of c-erb B-2 in primary breast cancer. Br J Cancer 63: 447-450, 1991[Medline]
11. Paterson M, Dietrich K, Danyluk J, et al: Correlation between c-erb B-2 amplification and risk of early relapse in node-negative breast cancer. Cancer Res 51: 556-567, 1991 12. Gusterson B, Gelber R, Goldhirsch A, et al: Prognostic importance of c-erbB-2 expression in breast cancer: International (Ludwig) Breast Cancer Study Group. J Clin Oncol 10: 1049-1056, 1992[Abstract] 13. Quenel N, Wafflart J, Bonichon F, et al: The prognostic value of c-erb B2 in primary breast carcinoma: A study on 942 cases. Breast Cancer Res Treat 35: 283-291, 1995[CrossRef][Medline]
14. Seshadri R, Firgaira F, Horsfall D, et al: Clinical significance of HER-2/neu oncogene amplification in primary breast cancer: The South Australian Breast Cancer Study Group. J Clin Oncol 11: 1936-1942, 1993 15. Albanell J, Bellmunt J, Molina R, et al: Node-negative breast cancers with p53(-)/HER2-neu(-) status may identify women with very good prognosis. Anticancer Res 16: 1027-1032, 1996[Medline]
16. Ro JS, El-Naggar A, Ro JY, et al: c-erb B-2 amplification in node-negative breast cancer. Cancer Res 49: 6941-6944, 1989
17. Andrulis I, Bull S, Blakstein M, et al: neu/erbB-2 amplification identifies a poor-prognosis group of women with node-negative breast cancer: Toronto Breast Cancer Study Group. J Clin Oncol 16: 1340-1349, 1998 18. Slamon D, Press M, Godolphin W, et al: Studies of the HER-2/neu proto-oncogene in human breast cancer. Cancer Cells 7: 371-380, 1989 19. Rilke F, Colnaghi M, Cascinelli N, et al: Prognostic significance of HER-2/neu expression in breast cancer and its relationship to other prognostic factors. Int J Cancer 49: 44-49, 1991[Medline]
20. Muss H, Thor A, Berry D, et al: cerb B-2 expression and response to adjuvant therapy in women with node-positive early breast cancer. N Engl J Med 330: 1260-1266, 1994 21. Tetu B, Brisson J: Prognostic significance of HER-2/neu oncoprotein expression in node-positive breast cancer: The influence of the pattern of immunostaining and adjuvant therapy. Cancer 73: 2359-2365, 1994[CrossRef][Medline] 22. Tetu B, Brisson J, Plante V, et al: p53 and c-erb B-2 as markers of resistance to adjuvant chemjotherapy in breast cancer. Mod Pathol 11: 823-830, 1998[Medline]
23. Allred D, Lark G, Tandon A, et al: HER-2/neu in node-negative breast cancer: Prognostic significance of overexpression influenced by the presence of in situ carcinoma. J Clin Oncol 10: 599-605, 1992
24. Thor A, Berry D, Budman D, et al: erb B-2, p53, and efficacy of adjuvant therapy in lymph node-positive breast cancer. J Natl Cancer Inst 90: 1346-1360, 1998
25. Paik S, Bryant J, Park C, et al: erb B-2 and response to doxorubicin in patients with axillary lymph node-positive, hormone receptor-negative breast cancer. J Natl Cancer Inst 90: 1361-1370, 1998 26. Wright C, Nicholson S, Angus B, et al: Relationship between c-erb B-2 protein product expression and response to endocrine therapy in advanced breast cancer. Br J Cancer 65: 118-121, 1992[Medline] 27. Borg A, Baldetorp B, Ferno M, et al: erb B2 amplification is associated with tamoxifen resistance in steroid-receptor positive breast cancer. Cancer Lett 81: 137-144, 1994[CrossRef][Medline]
28. Carlomagno C, Perrone F, Gallo C, et al: c-erb B2 overexpression decreases the benefit of adjuvant tamoxifen in early-stage breast cancer without axillary lymph node metastases. J Clin Oncol 14: 2702-2708, 1996 29. Sjogren S, Inganas M, Lindgren A, et al: Prognostic and predictive value of c-erbB-2 overexpression in primary breast cancer, alone and in combination with other prognostic markers. J Clin Oncol 16: 462-469, 1998[Abstract]
30. Roche P, Ingle J: Increased HER2 with U.S. Food and Drug Administration-approved antibody. J Clin Oncol 17: 434, 1999
31. Jacobs T, Gown A, 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 17: 1983-1987, 1999
32. Press M, Hung G, Godolphin W, et al: Sensitivity of HER-2/neu antibodies in archival tissue samples: Potential source of error in immunohistochemical studies of expression. Cancer Res 54: 2771-2777, 1994
33. Coussens L, Yang-Feng T, Chen Y-CLE, et al: Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with neu oncogene. Science 230: 1132-1139, 1985
34. Kallioniemi O-P, Kallioniemi A, Kurisu W, et al: ERBB2 amplification in breast cancer analyzed by fluorescence in situ hybridization. Proc Natl Acad Sci USA 89: 5321-5325, 1992 35. Park J, Yang X, Park J, et al: Assessment of novel anti-p185HER-2 monoclonal antibodies for internalization-dependent therapies. Hybridoma 18: 487-495, 1999[Medline]
36. Cobleigh M, Vogel C, Tripathy D, et al: Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2-overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 17: 2639-2648, 1999
37. Slamon D, Leyland-Jones B, Shak S, et al: Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344: 783-792, 2001
38. Vogel C, Cobleigh M, Tripathy D, et al: Efficacy and safety of trastuzumab (Herceptin) as a single agent in first-line treatment of HER2-overexpressing metastatic breast cancer. J Clin Oncol 20: 719-726, 2002 39. Fleiss J: Statistical methods for rates and proportions. New York, NY, John Wiley & Sons, 1981 40. Landis J, Koch G: The measurement of observer agreement for categorical data. Biometrics 33: 159-174, 1977[CrossRef][Medline]
41. Potter H, Linklater J, Allen A, et al: Magnetic resonance imaging of articular cartilage in the knee: An evaluation with use of fast-spin-echo imaging. J bone Joint Surg 80A: 1276-1284, 1998
42. Ali IU, Campbell G, Lidereau R, et al: Amplification of c-erb B-2 and aggressive human breast tumors? Science 240: 1795-1798, 1988 43. Ali IU, Campbell G, Lidereau R, et al: Lack of evidence for the prognostic significance of c-erb B-2 amplification in human breast carcinoma. Oncogene Res 3: 139-146, 1988[Medline] 44. Zhou D-J, Ahuja H, Cline M: Proto-oncogene abnormalities in human breast cancer: c-erb B-2 amplification does not correlate with recurrence of disease. Oncogene 4: 105-108, 1989[Medline]
45. Clark G, McGuire W: Follow-up study of HER-2/neu amplification in primary breast cancer. Cancer Res 51: 944-948, 1991 46. Hayes D: Tumor markers for breast cancer: Current utilities and future prospects, in Hayes D (ed): Hematology/Oncology Clinics of North America: Tumor Markers in Adult Solid Malignancies. Philadelphia, PA, Saunders, 1994, pp 485-506
47. Borg A, Tandon A, Sigurdsson H, et al: HER-2/neu amplification predicts poor survival in node-positive breast cancer. Cancer Res 50: 4332-4337, 1990 48. Tandon A, Clark G, Chamness G, et al: HER-2/neu oncogene protein and prognosis in breast cancer. J Clin Oncol 7: 1120-1128, 1989[Abstract] 49. Ravdin P: Should HER2 status be routinely measured for all breast cancer patients? Semin Oncol 26: 117-123, 1999[Medline] 50. Baselga J, Seidman A, Rosen P, et al: HER2 overexpression and paclitaxel sensitivity in breast cancer: Therapeutic implications. Oncol 2: 43-48, 1997 (suppl) 51. Pegram M, Lipton A, Hayes D, et al: Phase II study of receptor-enhanced chemosensitivity using recombinant humanized anti-p185HER2/neu monoclonal antibody plus cisplatin in patients with HER2/neu-overexpressing metastatic breast cancer refractory to chemotherapy treatment. J Clin Oncol 16: 2659-2671, 1998[Abstract] 52. Pegram M, Slamon D: Combination therapy with trastuzumab (Herceptin) and cisplatin for chemoresistant metastatic breast cancer: Evidence for receptor-enhanced chemosensitivity. Semin Oncol 26: 89-95, 1999 (suppl 12)[Medline] 53. Jacobs T, Gown A, Yaziji H, et al: HER-2/neu protein expression in breast cancer evaluated by immunohistochemistry: A study of interlaboratory agreement. Anat Pathol 113: 251-258, 2000
54. Jacobs T, Gown A, Yaziji H, et al: Comparison of fluorescence in situ hybridization and immunohistochemistry for the evaluation of HER-2/neu in breast cancer. J Clin Oncol 17: 1974-1982, 1999 55. Van de Vijver M, Peterse J, Mooi W, et al: neu -protein overexpression in breast cancer: Association with comedo-type ductal carcinoma in situ and limited prognostic value in stage II breast cancer. N Engl J Med 319: 1239-1245, 1988[Abstract]
56. van Diest P, van Dam P, Henzen-Logmans S, et al: A scoring system for immunohistochemical staining: Consensus report of the task force for basic research of the EORTC-GCCG. J Clin Pathol 50: 801-804, 1997
57. Nelson N: Experts debate value of HER2 testing methods. J Natl Cancer Inst 92: 292-294, 2000 58. Nagle RB, Tubbs RR, Roche PC, et al: Clinical laboratory assays for HER-2/neu amplification and overexpression: Quality assurance, standardization, and proficiency testing. Arch Pathol Lab Med (in press) Submitted September 21, 2001; accepted April 23, 2002.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
