|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology HER-2/neu Analysis in Archival Tissue Samples of Human Breast Cancer: Comparison of Immunohistochemistry and Fluorescence In Situ HybridizationPathologisches Institut and Klinik für Frauenheilkunde und Geburtshilfe Klinikum-GroBhadern, Ludwig-Maximilians-Universität, München, Germany. Address reprint requests to Annette Lebeau, MD, Pathologisches Institut der Universität, Thalkirchner Str 36, 80337 München, Germany; email a.lebeau{at}lrz.uni-muenchen.de
PURPOSE: The objective of our study was to compare the methods used in the literature to analyze HER-2/neu status on archival breast cancer tissue. Therefore, a series of antibodies was evaluated to assess their immunohistochemical (IHC) sensitivity in correlation to gene amplification determined by fluorescence in situ hybridization (FISH). MATERIALS AND METHODS: HER-2/neu overexpression was studied on paraffin sections of 85 invasive breast cancers using a panel of five monoclonal (9G6, 3B5, CB11, TAB250, GSF-HER2) and two polyclonal antibodies (A8010, A0485) in addition to the HercepTest (DAKO, Glostrup, Denmark). HER-2/neu gene amplification was determined by FISH using a dual-color probe (PathVysion; Vysis, Stuttgart-Fasanenhof, Germany). RESULTS: HER-2/neu overexpression was demonstrated in 26% (9G6, TAB250, GSF-HER2), 27% (3B5, CB11), 33% (A8010) and 42% (A0485, HercepTest) of the tumors. FISH on paraffin sections identified gene amplification in 28% of the tumors. Strongly positive IHC results (3+) were always associated with gene amplification. Among the 16 tumors presented with weakly positive IHC results (2+) using the HercepTest, 12 (75%) lacked gene amplification. CONCLUSION: The comparison of IHC and FISH demonstrated an excellent correlation of high-level HER-2/neu overexpression (3+) with gene amplification; ie, FISH does not provide further information in these tumors. However, weakly positive IHC results (2+) obtained with the HercepTest share only a minor association with gene amplification.
THE HER-2/ NEU proto-oncogene encodes a membrane receptor that is homologous to the epidermal growth factor receptor. Roughly 25% of human invasive breast cancers have increased levels of HER-2/neu on their cell surface.1,2 The main cause of HER-2/neu overexpression seems to be gene amplification. However, it should be noted that overexpression has been observed in some breast cancer specimens despite a lack of gene amplification.3-6 HER-2/neu oncogene amplification and overexpression have been correlated with poor prognosis in women with breast cancer. Recently, HER-2/neu assumed therapeutic implications when the United States Food and Drug Administration (FDA) approved trastuzumab (Herceptin; Genentech, Inc., South San Francisco, CA), a humanized monoclonal antibody directed against the HER-2/neu protein, for the therapy of metastatic breast cancer. Trastuzumab binds to the extracellular domain of HER-2/neu and is capable of blocking cell proliferation of tumor cells that overexpress HER-2/neu in both in vitro assays and in animals.7-9 It also induces antibody-dependent cellular toxicity against tumor cells.10 Therefore, it is important that tumor tissue is evaluated to identify patients who might benefit from this therapeutical approach. An overview of published immunohistochemical studies on overexpression of HER-2/neu shows a wide range of overexpression rates, varying from 9% at the lowest to 60% at the highest.1,2,11-14 However, a wide variety of different antibodies have been used in these studies. Most pathologists are aware that immunohistochemistry (IHC) has its own challenges regarding specimen handling (eg, fixation and staining) that can affect the quality of the results. In addition, this approach requires a uniform method for interpreting and scoring the results. The HercepTest (DAKO, Glostrup, Denmark) approved by the FDA represents an immunohistochemical test kit using a rabbit polyclonal antibody with standardized procedures and evaluation criteria. However, this antibody has not been used in the clinical trials of Herceptin (trastuzumab) that have been published thus far15,16; therefore, ongoing studies that are using the HercepTest for entry must clarify its predictive value. The same is true for most of the commercially available antibodies for IHC. Some argue that detection of HER-2/neu gene amplification in breast cancer cells by the fluorescence in situ hybridization (FISH) technique is the best approach. Criticisms of this technique include its need for higher technical prerequisites and costs. It also requires a reasonable amount of experience in interpreting the results. Moreover, it is unknown yet whether the FISH technique fails to identify those patients with carcinomas that overexpress HER-2/neu in the absence of gene amplification who might profit from Herceptin therapy. There are, however, only limited data currently available on the frequency of this phenomenon.3,17 The latest publication on this issue has been presented by Mass et al18 at the 2000 Annual Meeting of the American Society of Clinical Oncology. The objective of this investigation was to compare the methods used to analyze HER-2/neu status in archival breast cancer tissue. The main intention of our study was to identify the most reliable procedure for determination of HER-2/neu status with regard to Herceptin therapy. Therefore, a series of antibodies were evaluated to assess their immunohistochemical reactivity in correlation to gene amplification determined by FISH. Paraffin-embedded samples of 85 breast cancers were analyzed for HER-2/neu overexpression using five monoclonal and two polyclonal antibodies according to our own preparation procedures. In addition, the HercepTest was applied on the same cases. To find out the more practicable and valuable procedure, FISH was performed in each case both on intact paraffin tissue sections and on isolated nuclei extracted from paraffin-embedded tissue after microdissection applying a directly labelled HER-2/neu gene probe.
Tumor Specimen Paraffin-embedded tissue blocks from 85 patients with invasive breast cancer and an average age of 50 years (range, 25 to 63 years) were obtained from institutional archives. Patients underwent surgery from January 1993 to November 1997. The tumor specimens were fixed in neutral-buffered formalin and embedded in paraffin according to the same standard procedure. All cases were of a consecutive series of lymph node–positive breast cancers, predominantly with metastases to 10 or more lymph nodes and/or extension of the tumor beyond the capsule of a lymph node. The histopathologic characteristics are listed in Table 1.
Slide Preparation IHC and FISH were performed on sections prepared from the same archival paraffin block. For the immunohistochemical studies, serial 1- to 2-µm tissue sections were prepared; for FISH analysis on intact paraffin sections, 4-µm sections were prepared. The sections were placed on Superfrost Plus slides (Menzel-Gläser, Braunschweig, Germany). One- to two-micrometer sections were chosen for IHC instead of 4-µm sections to minimize superposition of cell membranes of different cells and therefore to achieve easier evaluation. However, testing of six negative (score of 0/1+) and positive tumors (score of 2+/3+) with the HercepTest interpreted by two of the authors (A.L. and U.L.) in a blinded fashion indicated that variation in section thickness did not result in different scoring. For these six cases, 2-, 4-, and 6-µm sections were cut from each tumor block and stained in the same run. For FISH on isolated nuclei extracted from paraffin-embedded tissue, invasive tumor areas were microdissected from the paraffin blocks and re-embedded in paraffin, and 25-µm tissue sections were prepared from these tumor blocks and placed in plastic cups. A last slide stained with hematoxylin and eosin was prepared from each block and used for pathologic confirmation of breast cancer.
IHC
IHC technique. Pretreatment of the slides, optimal dilutions of the antibodies, and optimal detection techniques were determined by IHC testing and/or use of conditions as described in the commercial information. In brief, paraffin sections were dewaxed in xylene and rehydrated in decreasing concentrations of ethanol. Sections were pretreated by the detergent Saponin (Sigma, Deisenhofen, Germany) for 30 minutes at room temperature for immunostaining with 9G6, 3B5 (Oncogene Science, Calbiochem-Novabiochem, Germany), CB11 (Novocastra, Newcastle, United Kingdom), and GSF-HER2 (Dr Lindhofer, Institut für Immunologie der GSF, München, Germany) or by Ficin (Digest All; Zymed Lab Inc, South San Francisco, CA) for 10 minutes at 37°C for TAB250. Microwave pretreatment was performed for immunostaining with A0485 (DAKO) for 30 minutes in tissue retrieval solution (DAKO). Primary antibodies were diluted and incubated as listed in Table 2. Immunodetection was performed with the APAAP technique (DAKO) or the LSAB2 Kit (DAKO). The HercepTest was applied exactly according to the commercial instructions for use. For negative control, corresponding sections of each tissue block were processed in the same way, except the primary antibody was omitted (buffer substitution). Positive controls were included in each staining run and consisted of freshly cut paraffin sections of a case of invasive ductal carcinoma known to overexpress HER-2/neu. For HercepTest staining, the control slides supplied in the kit were used, consisting of three pelleted, formalin-fixed, paraffin-embedded human breast cancer cell lines with staining intensity scores of 0, 1+, and 3+. Interpretation of IHC results. For the determination of HER-2/neu protein overexpression, only the membrane staining pattern and intensity of invasive tumor cells were scored. Interpretation was independently performed by two of the authors (A.L. and U.L.) each blinded to one another and to the results of the other assays. The proportion of positively stained invasive tumor cells on a given slide was estimated as a continuous variable. To compare these data, we used cutoff points according to the literature. For the monoclonal antibodies 9G6 and 3B5, each distinctive membrane staining was determined positive according to previous publications.23,24 For each of the remaining antibodies, immunostaining was scored according to the criteria specified by DAKO for the interpretation of the HercepTest, which have also been used in the published Herceptin trials.15,16 Immunoreaction was determined as weakly positive (2+) if more than 10% of the tumor cells showed weak to moderate complete membrane staining or as strong positive (3+) if a strong complete membrane staining was observed in more than 10% of the tumor cells. All other staining patterns were interpreted as negative (0/1+). The investigators came generally to similar estimates of the stained cells on the slides (< 5% discrepancy) and to identical scoring results. The monoclonal antibodies never showed staining of normal ductal epithelial cells. Nonneoplastic epithelial cells were exceptionally scored as weak and incomplete at the cell membrane when the polyclonal antibodies were used. Only two cases indicated a membrane staining of normal ductal epithelial cells by the HercepTest equivalent to a score of 1+.
FISH FISH on isolated nuclei extracted from paraffin-embedded tissue. Twenty-five micrometer–thick sections were dewaxed in xylene for 40 minutes and rehydrated in descending concentrations of ethanol followed by two washes in H2O. Disaggregation of the nuclei was performed in a solution of protease type 24 (1 mg/mL) for 60 minutes at 37°C followed by an incubation in a solution of RNAse (0.1 mg/mL). Tissue fragments were filtered through nylon mesh (60 µm) and nuclei were put on slides by centrifugation of 100 µL of the suspension in a Cytospin 3 centrifuge (Shandon, Life Sciences International-Europe-Ltd, Cheshire, United Kingdom) for 5 minutes at 800 rounds/min. Pretreatment of the slides was performed by incubation in 4x sodium chloride/sodium citrate (SSC), 0.1% Triton X for 2 hours at 37°C followed by an incubation in 1 mol/L of NaSCN for 30 minutes at 56°C and an incubation in a pepsin solution (2mg/mL H2O; pH 2.0) for 60 minutes at 37°C. Before denaturation, air-dried slides were warmed for 20 minutes at 80°C. Two microliters of the hybridization mixture (PathVysion) were applied onto the isolated nuclei on the slides. The DNA and the probes were simultaneously denatured at 80°C for 10 minutes. Hybridization was carried out overnight at 37°C under a coverslip in a moist chamber. Washes were performed at 42°C three times in 0.05x SSC (1 SSC is 0.15 mol/L of NaCl and 0.015 mol/L of sodium citrate, pH7) for 10 minutes each. Tissue sections were counterstained with 4,6-diaminidino-2-phenyl-indole (DAPI) Vectashield (Vector Laboratories, Inc, Burlingame, CA) (antifade). FISH on paraffin sections. The slide-mounted tissue sections were air-dried and baked overnight at 56°C. Slides were dewaxed in xylene for 15 minutes x 2, followed by immersion in 100% ethanol for 10 minutes. Air-dried tissue sections were subsequently treated in 1 mol/L of NaSCN for 30 minutes at 80°C, washed in Aqua bidest. and treated in a pepsin solution (8mg/mL H2O; pH 2.0) for 30 minutes at 37°C. Slides were washed in Aqua bidest and air-dried. Before denaturation, slides were warmed at 80°C for 20 minutes and 10 to 20 µL of the hybridization mixture (PathVysion) was applied onto the tissue sections. The amount of the probe used was adjusted according to the size of the tissue section to be covered. The sections and the probes were simultaneously denatured at 80°C for 10 minutes. Hybridization was carried out overnight at 37°C under a coverslip in a moist chamber. Washes were performed at 42°C three times in 0.05x SSC for 10 minutes each. Tissue sections were counterstained with DAPI/Vectashield. Positive controls were included in each preparative run and consisted of freshly cut paraffin sections of cases known to be amplified for the HER-2/neu gene by FISH. Scoring criteria. Slides were evaluated for HER-2/neu gene amplification using a Zeiss Axioscope fluorescence microscope (Carl Zeiss Inc., Jena, Germany) equipped with specially designed filter combinations for spectrum green and spectrum orange detection (Vysis, Inc., Downers Grove, IL.). Interpretation was performed by two of the authors (A.L. and D.D.) who were blinded to one another and to the results of the other assays. Only the invasive tumor component was scored for FISH on paraffin sections. The interobserver variability was low (< 5%), not least because FISH was repeated in those cases with insufficient preparation. In the remaining questionable cases, the evaluation of a higher number of nuclei (up to 300) led to corresponding results. Cases that proved to be nonassessable for technical reasons were excluded (see Results). FISH was considered successful if evaluation of the nuclei met the following criteria: (1) no detectable DNA loss had occurred as determined by DAPI staining within the nuclei, (2) hybridization was uniform throughout the isolated nuclei or the tissue sections, and (3) single-copy status could be identified and not assumed to be a consequence of lack of signals. Signal enumeration was conducted at 630x or 1,000x magnification with the appropriate filter combination. Nuclear boundaries were determined by DAPI excitation at the same magnification, and overlapping nuclei were not scored. In cases with an obvious gene amplification (> 10 signals/nucleus), a minimum of 40 nuclei were analyzed. Precise signal enumeration was not possible in some cases because nuclei with high levels of gene amplification often exhibit signal clusters. For all other cases, a minimum of 200 randomly selected isolated nuclei and in the tissue sections a minimum of 80 to 100 randomly selected invasive tumor cell nuclei were analyzed. Tumors containing up to four HER-2/neu signals per nucleus in more than 90% of the analyzed nuclei were considered to be not amplified. This accounts for the cells that are disomic for the HER-2/neu gene but that are in the G2 or M phase of the cell cycle. According to the criteria of Pauletti et al,5 the contribution of chromosome 17 polysomy to HER-2/neu copy number was determined in all tumors with multiple copies of the gene (> four signals per nucleus). Therefore, the ratio of HER-2/neu signals per nucleus relative to chromosome 17 centromere signals per nucleus was calculated. A tumor was considered to be amplified if this ratio was more than two. Taking this criteria into account, amplified tumors were further distinguished in low and high amplified. Tumors with more than four but fewer than 10 HER-2/neu signals per nucleus in more than 10% of analyzed nuclei were considered to be low amplified. If more than 10% of the nuclei showed more than 10 HER-2/neu signals per nucleus, a tumor was considered to be high amplified.
IHC The results of immunostaining 85 primary breast cancers for HER-2/neu overexpression with the various antibodies are listed in Table 3. Protein overexpression was demonstrated in 26% to 42% of the 85 tumors, depending on the antibody used. While the monoclonal antibodies identified almost the same tumors as positive, the two polyclonal antibodies (A8010, A0485) detected protein overexpression in additional cases. A0485 demonstrated the highest rate of positivity (42%).
Comparing the results of the polyclonal antibody A0485 according to our own staining protocol with the HercepTest, which uses the same antibody in prediluted form, all 84 tumors under study showed identical immunoreactivity also with regard to the level of overexpression (Table 4).
When comparing our results for A0485 to CB11 as a representative of the monoclonal antibodies, the level of concordance was 85% (72 of 85) for positive (2+/3+) versus negative (0/1+) results. Discrepancies between the two antibodies were represented by tumors detected as weakly positive (2+) by the A0485 and negative (0/1+) by CB11 (Table 5).
FISH FISH on isolated nuclei extracted from paraffin-embedded tissue. FISH on isolated nuclei extracted from paraffin-embedded tissue was considered successful in 82 (97%) of the 85 cases. Hybridization results in three cases didn’t meet the evaluation criteria. Twenty-three percent (19 of 82) of the informative tumors demonstrated high gene amplification (> 10 gene signals per nucleus). Low gene amplification was identified in one case (1%; 4 < x  10 gene signals per nucleus). Among the remaining tumors, three demonstrated more than four gene and centromere signals per nucleus but with a ratio of gene signals relative to centromere signals less than two. Therefore, these tumors were considered to be not amplified but polysomic for chromosome 17. FISH on paraffin sections. FISH on paraffin sections was considered successful on 79 (93%) of the 85 primary breast cancers. Pretreatment and hybridization results in six cases didn’t meet the evaluation criteria. Twenty-five percent (20 of 79) of the informative tumors demonstrated high gene amplification. Low gene amplification was identified in two tumors (3%). Three of the remaining cases demonstrated more than four gene and centromere signals per nucleus, but with a ratio of gene signals relative to centromere signals less than two. Therefore, these tumors were considered to be not amplified but polysomic for chromosome 17. Comparison of FISH techniques. The results of the two FISH preparation techniques were compared in the 79 cases that were informative for both methods (Table 6). Results were concordant in 96% (76 of 79) of the tumors. Discrepancies were due to three cases considered to be not amplified by FISH on isolated nuclei and to be amplified by FISH on paraffin sections (low amplified, n = 1; high amplified, n = 2). In these cases, FISH on isolated nuclei demonstrated some nuclei with more than four gene signals but less than 10% of the scored nuclei; these cases did not meet the criteria for gene amplification.
Correlation of IHC to FISH Because of the underestimation of gene amplification in some cases studied by FISH on isolated nuclei, we compared the IHC results with the gene amplification rates demonstrated by FISH on paraffin sections in 79 breast cancers. The relationship between gene amplification and protein overexpression of HER-2/neu is listed in Table 7 without regard to the degree of positivity. All tumors with high gene amplification demonstrated overexpression of HER-2/neu protein with the commercially available antibodies. Two tumors with high gene amplification reacted immunohistochemically negative with the GSF-HER2 antibody. Among the two tumors with low gene amplification, protein overexpression was detected exclusively by the A0485/HercepTest (two of two). The comparison of the results in the group of tumors not showing gene amplification indicated a variable rate of positive immunoreaction depending on the antibody used. The highest rate of positively reacting cases was detected by the A0485/HercepTest. Twenty-three percent (13 of 57) of the tumors without gene amplification showed immunohistochemical evidence of protein overexpression by using the A0485.
The following results were achieved when FISH results were compared to the degree of immunohistochemical reactivity for those antibodies scored according to the HercepTest interpretation criteria (Table 8): strongly positive reactivity (3+) was always associated with high gene amplification regardless of the antibody used. In those cases exhibiting no or only low gene amplification by FISH but HER-2/neu overexpression by IHC, the degree of positivity was exclusively weak (2+). One of the tumors indicating chromosome 17 polysomy showed weakly positive immunohistochemically staining (2+) using the polyclonal antibodies (A8010, A0485/HercepTest). Seventy-five percent (12 of 16) of the tumors with weak protein overexpression (2+) using the HercepTest did not show gene amplification. The relationship between the level of HER-2/neu overexpression detected by HercepTest and gene amplification analyzed by FISH using a dual-color probe is illustrated in Figs 1 and 2.
Since the FDA approved trastuzumab (Herceptin) for the therapy of metastatic breast cancer, there has been a need to evaluate HER-2/neu status of breast cancer specimens to identify those patients who might benefit from this therapeutical approach. In the Herceptin clinical trials published thus far, patient eligibility was determined by testing tumor specimens for overexpression of HER-2/neu protein.16 Specimens were tested with a research use–only immunohistochemical assay (referred to as the Clinical Trial Assay [CTA]) that consisted of the combination of the antibodies 4D5 (murine) and CB11 and were scored as 0, 1+, 2+, or 3+. Only patients with 2+ or 3+ positive tumors were eligible. Data from the trials suggest that the beneficial treatment effects were largely limited to patients with the highest levels of HER-2/neu protein overexpression (3+; Herceptin complete prescribing information, Genentech). With regard to the trastuzumab (Herceptin) application, the FDA approved the HercepTest as a standardized immunohistochemical assay in combination with the HercepTest scoring system to identify those patients who would be candidates for this therapy. But the HercepTest was not used in the clinical trials that have been published so far. Consequently, we neither know the predictive value of the HercepTest nor do we know of most of the commercially available antibodies used in the literature to determine HER-2/neu overexpression. The situation becomes even more complicated by using a variety of tissue fixatives, preparation protocols, and cutoffs in different laboratories. In addition, a few previous studies have suggested that the interpretation of IHC results is subject to interobserver variability.25-27 Perhaps the most widely cited study in the discussion about sensitivity and specificity of IHC versus techniques that determine gene copy number in paraffin-embedded tissue is the study of Press et al.27 The results of this study suggest that IHC for HER-2/neu lacks sensitivity. However, there are several limitations to the IHC assays used in that study. It is difficult to assign the conclusions of Press et al27 to the current situation for methodological reasons. Particularly, antigen retrieval was not performed with one exception, and the immunodetection system used is of lower sensitivity than current systems. In our study, we compared the methods to analyze HER-2/neu status in archival breast cancer tissue, thereby using material that is usually available. The analyzed cases consisted of a consecutive series of node-positive breast cancers, predominantly with metastases to 10 or more lymph nodes and/or perinodal tumor growth. We chose those characteristics because they indicate that a tumor is at high risk to spread distant metastases, being a potential candidate for trastuzumab therapy. A series of commercially available antibodies was evaluated to assess their sensitivity and characterize their reactivity in correlation to gene amplification detected by FISH on paraffin-embedded tissue. As no internationally accepted standard exists, we applied the scoring proposed by the HercepTest for the interpretation of the immunoreactivity of CB11, TAB250, A8010, A0485, and HercepTest. The decision to consider each distinctive membrane staining positive for the monoclonal antibodies 3B5 and 9G6 was based on investigations that indicated a high level of concordance between immunohistochemically detected overexpression and gene amplification using these criteria.23,24 The rates of HER-2/neu overexpression found in this study were 26% to 42%, depending on the antibody used. The monoclonal antibodies identified almost the same tumors as positive-indicating protein overexpression in 26% (9G6, TAB250, GSF-HER2) or 27% (3B5, CB11). The two polyclonal antibodies A8010 and A0485 detected overexpression in 33% and 42% of the tumors, respectively. The polyclonal antibody A0485 according to our own staining protocol and the HercepTest according to the manufacturer’s guidelines gave identical results also with regard to the level of overexpression (2+ or 3+). The high rate of concordant results obtained by the monoclonal antibodies as shown in this study suggests that discrepancies in published overexpression levels using these antibodies might be due more to variations in preparation techniques and scoring criteria than to differences in the nature of the antibodies.1,2,11-14,28 Compared with the monoclonal antibodies, HER-2/neu overexpression was detected in additional tumors by the polyclonal antibodies used in this study. The highest level of overexpression of 42% was detected by the DAKO A0485/HercepTest, which is not in keeping with the range of 20% to 30% reported in the main literature based on application of different antibodies.1,2 One explanation for this finding could be that the A0485/HercepTest has higher sensitivity than other commercially available antibodies. The comparison of the results for the A0485 to the CB11 showed a level of concordance of 85% for positive (2+/3+) versus negative results. Discrepancies were exclusively represented by tumors detected as weakly positive (2+) by A0485 and negative by CB11. This observation is at least partly in agreement with the recent findings of Roche and Ingle,14 who observed concordant results in 60% of the tumors. Why Roche and Ingle14 observed a higher rate of 57% positive cases by application of the HercepTest remains unclear, possibly due to differences in technical procedures (eg, fixation). The importance of this issue becomes evident in the recently published study of Jacobs et al29 on specimens that had been fixed initially in alcoholic formalin followed by fixation in 10% neutral-buffered formalin. It might be possible that the use of a fixative that contained alcohol instead of neutral-buffered formalin as recommended by the manufacturers contributed to the high level of reactivity even in normal epithelium. Unlike the results of Jacobs et al,29 nonneoplastic epithelial cells in the formalin-fixed specimens of our study did not show a positive membrane staining equivalent to a score of 2+ or 3+ by the Hercep-Test. Therefore, with regard to therapeutically relevant results, any application of the HercepTest must take into account that the reactivity of this IHC assay depends essentially on the processing of the tissue. Benz mentioned in his presentation at the 21st San Antonio Breast Cancer Symposium in 1998 that an amplification of the gene copy number of HER-2/neu up to three or four times results in an expression of 106 HER-2/neu receptor molecules per cell. He pointed out that the HercepTest might detect even lower levels of protein expression,30 but he did not comment on the associated gene copy number. To examine this issue more closely, we analyzed the gene copy number of all immunohistochemically studied cases by FISH. This was also done to find out whether the routine use of FISH for the determination of HER-2/neu status is necessary. Principally, FISH on paraffin-embedded tissue can be performed on isolated nuclei extracted from the paraffin blocks or on paraffin sections. Previous studies have determined gene amplification either on cytologic preparations4,31,32 or on paraffin sections.5,33,34 We compared these techniques directly to identify the more reliable procedure. FISH on isolated nuclei was informative in 97% (82 of 85) of the analyzed tumors, compared with 93% (79 of 85) by FISH on paraffin sections. High gene amplification was demonstrated in 23% (19 of 82) and 25% (20 of 79) of the informative cases, respectively. Low gene amplification was identified in one case (1%) by FISH on isolated nuclei and on two cases (3%) on paraffin sections. Therefore, the overall amplification rate was 24% and 28%, respectively, which is in keeping with the literature.1,2 Because it is assumed that chromosome 17 centromere signals correspond to copies of intact chromosome 17 in tumor cells as they do in normal cells, chromosome 17 polysomy (with a corresponding increase of HER-2/neu gene copy number) was identified in three cases with each preparation technique among the remaining tumors considered as not amplified. To avoid misinterpretation of gene amplification in our study, we used a dual-color probe that contains a mixture of a HER-2/neu gene probe and a centromere control for chromosome 17. Only those nuclei with more than four centromere signals were counted as polysomic for chromosome 17 to exclude those cells with premitotic chromosome reduplication. Because the biologic significance of chromosome 17 polysomy is not clear, we applied the criteria suggested by Pauletti et al5 to discriminate true low-level amplification from polysomy. Low-level amplification was assessed only when more than four but fewer than 10 signals per nucleus were determined and the ratio of gene signals relative to centromere signals was more than two. As the criteria to discriminate chromosomal polysomy from gene amplification are not applied in the literature in the same way,5,6,33-36 it is difficult to compare our own results with those of the published studies. The reported frequency of chromosome 17 polysomy up to 21% in the literature35 and low-level amplification up to 30%36 might be at least partly explained by the fact that these studies included cases with three and four HER-2/neu signals per nucleus, which have not been considered as polysomic or amplified in our study. The comparison of the results of the two FISH preparation techniques indicated concordant results in 96% (76 of 79) of the tumors that were informative for both techniques. The re-examination of the three cases that were considered not amplified by FISH on isolated nuclei but amplified by FISH on paraffin sections (low gene amplification, n = 1; high gene amplification, n = 2) revealed that the discrepancies were due to an underestimation of nuclei with gene amplification by FISH on isolated nuclei. These results indicate that the microdissection of the invasive tumor areas that preceded the extraction of the nuclei from the paraffin material did not completely exclude the risk of an underestimation of tumor cells in the nuclei suspensions. Therefore, this method bears the risk of loss of important information because the histologic context of the nuclei under study is not regarded. Remarkably, FISH on paraffin sections did not result in a reduced rate of low-level amplification as a potential result of unavoidable nuclear truncation. As a consequence of the underestimation of gene amplification in some cases studied by FISH on isolated nuclei, we compared the IHC results with the FISH results on paraffin sections. All tumors with high gene amplification (gene signals > 10 per nucleus) demonstrated overexpression of the HER-2/neu protein with the commercially available antibodies. Among the two tumors with low gene amplification, protein overexpression was detected exclusively by the A0485/HercepTest. Therefore, 3% of the cases were interpreted as negative by the other antibodies but scored as low-amplified by FISH. Similar findings have been reported in 6% or 9% of cases in other studies.3,17,33 It might be possible that in cases with low gene amplification, gene transcription and posttranscriptional and translational events could be abnormal or downregulated, which leads to a low or abnormal protein or epitope production. But this would not explain the reactivity of the A0485/HercepTest in these cases. One interpretation for these results is that the A0485/HercepTest is more sensitive to a low level of HER-2/neu expression, as discussed above. Twenty-one percent (12 of 56) of the tumors without gene amplification showed positive immunostaining by the HercepTest. Contrary to these results, the monoclonal antibodies detected HER-2/neu overexpression in the absence of gene amplification in 2% to 5% of the cases. This result is in keeping with a published range of 3% to 10% in other studies and could represent single-copy overexpression at the transcriptional level and/or beyond.3-6,33 The level of IHC positivity was exclusively weak (2+) in those tumors categorized to overexpress HER-2/neu in the absence of gene amplification. Strong positive reactivity (3+) was always associated with high gene amplification, regardless of the antibody used. Therefore, we conclude that FISH does not provide further information in these tumors. But among the tumors presenting with weakly positive results (2+) using the HercepTest, 75% (12 of 16) lacked gene amplification. These data are in keeping with the results of Mass et al,18 who studied the concordance between the CTA and FISH in the Herceptin pivotal trials. They observed an overall concordance of 82% and gene amplification in 24% of the IHC (CTA) 2+ cases. Additionally, these investigators analyzed the benefit of Herceptin therapy based on gene amplification status detected by FISH. The authors concluded that patients who are IHC 2+ and FISH-positive have a similar clinical benefit from Herceptin as those who are IHC 3+. However, the conclusion drawn by the authors that HER-2/neu gene amplification by FISH seems to be a more precise predictor of clinical benefit from Herceptin therapy than overexpression by IHC seems debatable with regard to the data presented for the following reasons: (1) there are no data available concerning response to Herceptin therapy in IHC-negative and FISH-positive patients; (2) the suggested better discrimination of median duration of survival by FISH score in the trial H0648 (first-line chemotherapy ± Herceptin) was not underscored by statistical analysis; (3) only the clinical trial H0649 (second- or third-line Herceptin monotherapy) was evaluated by discriminating between IHC 2+ and 3+ cases; (4) one patient in the trial H0650 (first-line Herceptin monotherapy) who was FISH-negative but IHC 3+ responded to Herceptin—all responders in this study had tumors that overexpressed HER-2/neu at the 3+ level,37 and no data were presented considering the question of whether FISH provided further information in this subgroup; (5) the given data are based on a comparison between FISH and the CTA, which is not currently available for IHC or applied. In conclusion, the comparison of IHC and FISH in our study demonstrated an excellent correlation of high level HER-2/neu protein overexpression (3+) with gene amplification; ie, FISH does not provide further information in these tumors. The analyzed antibodies differ in their reactivity in cases with no or low gene amplification. The highest rate of HER-2/neu overexpression was identified by the DAKO A0485/HercepTest. Seventy-five percent of the tumors that were scored as 2+ by using the HercepTest did not show gene amplification. With regard to trastuzumab (Herceptin) therapy, considering the data currently available (see Mass et al18), it seems reasonable to apply FISH analysis to those cases especially.
We thank Dr Horst Lindhofer (Institut für Immunologie der GSF, München) for the generous gift of his antibody against the HER-2/neu protein. We also thank DAKO A/S, Glostrup, Denmark, for generously supplying the HercepTest.
This study contains parts of the doctoral thesis of D.D.
1. Révillion F, Bonneterre J, Peyrat JP: ERBB2 oncogene in human breast cancer and its clinical significance. Eur J Cancer 34: 791-808, 1998 2. Ross JS, Fletcher JA: HER-2/neu (c-erb-B2) gene and protein in breast cancer. Am J Clin Pathol 112: S53-S67, 1999 (suppl 1)[Medline]
3.
Slamon DJ, Godolphin W, Jones LA, et al: Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244: 707-712, 1989
4.
Kallioniemi O-P, Kallioniemi A, Kurisu W, et al: ERBB2 amplification in breast cancer analyzed by fluorescence in situ hybridization. Proc Natl Acad Sci U–A 89: 5321-5325, 1992 5. Pauletti G, Godolphin W, Press MF, et al: Detection and quantitation of HER-2/neu gene amplification in human breast cancer archival material using fluorescence in situ hybridization. Oncogene 13: 63-72, 1996[Medline] 6. Persons DL, Borelli KA, Hsu PH: Quantitation of HER-2/neu and c-myc gene amplification in human breast cancer archival material using fluorescence in situ hybridization. Mod Pathol 10: 720-727, 1997[Medline]
7.
Hudziak RM, Lewis GD, Winget M, et al: p185HER2 monoclonal antibody has antiproliferative effects in vitro and sensitizes human breast tumor cells to tumor necrosis factor. Mol Cell Biol 9: 1165-1172, 1989 8. Lewis GD, Figari I, Fendly B, et al: Differential responses of human tumor cell lines to anti-p185HER2 monoclonal antibodies. Cancer Immunol Immunother 37: 255-263, 1993[Medline]
9.
Baselga J, Norton L, Albanell J, et al: Recombinant humanized anti-HER2 antibody (Herceptin) enhances the antitumor activity of paclitaxel and doxorubicin against HER2/neu overexpressing human breast cancer xenografts. Cancer Res 58: 2825-2831, 1998 10. Pegram MD, Baly D, Wirth C, et al: Antibody dependant cell-mediated cytotoxicity in breast cancer patients in phase III clinical trials of a humanized anti-HER2 antibody. Proc Am Assoc Cancer Res 38: 602, 1997 (abstr) 11. Elledge RM, Green S, Ciocca D, et al: HER-2 expression and response to tamoxifen in estrogen receptor-positive breast cancer: A Southwest Oncology Group study. Clin Cancer Res 4: 7-12, 1998[Abstract]
12.
Paik S, Bryant J, Park C, et al: erbB-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
13.
Thor AD, Berry DA, Budman DR, et al: erbB-2, p53, and efficacy of adjuvant therapy in lymph node-positive breast cancer. J Natl Cancer Inst 90: 1346-1360, 1998
14.
Roche PC, Ingle JN: Increased HER2 with U. S. Food and Drug Administration-approved antibody. J Clin Oncol 17: 434, 1999 (letter) 15. Slamon DJ, Leyland-Jones B, Shak S, et al: Addition of Herceptin (humanized anti-HER2 antibody) to first line chemotherapy for HER2 overexpressing metastatic breast cancer (HER+/MBC) markedly increases anticancer activity: A randomized, multinational controlled phase III trial. Proc Am Soc Clin Oncol 17: 98A, 1998 (abstr 377)
16.
Cobleigh MA, Vogel CL, 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
17.
Ciocca DR, Fujimura FK, Tandon AK, et al: Correlation of HER-2/neu amplification with expression and with other prognostic factors in 1103 breast cancers. J Natl Cancer Inst 84: 1279-1282, 1992 18. Mass R, Sanders C, Kasian C, et al: The concordance between the clinical trials assay (CTA) and fluorescence in situ hybridization in the Herceptin pivotal trials. Proc Am Soc Clin Oncol 19: 75a, 2000 (abstr 291) 19. Elston CW, Ellis IO: Pathological prognostic factors in breast cancer: Experience from a large study with long-term follow-up. Histopathology 19: 403-410, 1991[Medline] 20. Van de Vijver MJ, Peterse JL, Mooi WJ, et al: Oncogene activations in human breast cancer. Cancer Cells 7: 385-391, 1989 21. Corbett IP, Henry JA, Angus B, et al: NCL-CB11, a new monoclonal antibody recognizing the internal domain of the c-erbB-2 oncogene effective for use on formalin-fixed, paraffin-embedded tissue. J Pathol 161: 15-25, 1990[Medline] 22. Kallioniemi O-P, Holli K, Visakorpi T, et al: Association of c-erbB-2 protein over-expression with high rate of cell proliferation, increased risk of visceral metastasis and poor long-term survival in breast cancer. Int J Cancer 49: 650-655, 1991[Medline] 23. De Potter CR, Van Daele S, Van de Vijver MJ, et al: The expression of the neu oncogene product in breast lesions and in normal fetal and adult human tissues. Histopathology 15: 351-362, 1989[Medline] 24. Singleton TP, Niehans GA, Gu F, et al: Detection of c-erbB-2 activation in paraffin-embedded tissue by immunohistochemistry. Hum Pathol 23: 1141-1150, 1992[Medline] 25. Busmanis I, Feleppa F, Jones A: Analysis of cerbB2 expression using a panel of 6 commercially available antibodies. Pathology 26: 261-267, 1994[Medline]
26.
Kay EW, Walsh CJ, Cassidy M, et al: C-erbB-2 immunostaining: Problems with interpretation. J Clin Pathol 47: 816-822, 1994
27.
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 54: 2771-2777, 1994 28. Ravdin PM, Green S, Albain KS, et al: Initial report of the SWOG biological correlative study of c-erbB-2 expression as a predictor of outcome in a trial comparing adjuvant CAF T with tamoxifen (T) alone. Proc Am Soc Clin Oncol 17: 97a, 1998 (abstr 374)
29.
Jacobs TW, Gown AM, Yaziji H, et al: Specificity of Hercep-Test 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 30. Benz CC: The molecular biology of HER-2. Presented at 21st San Antonio Breast Cancer Symposium, Mini-Symposium I, San Antonio, Texas, December 12-15, 1998
31.
Szöllösi J, Balázs M, Feuerstein BG, et al: ERBB-2 (HER-2/neu) gene copy number, p185HER-2 overexpression, and intratumor heterogeneity in human breast cancer. Cancer Res 55: 5400-5407, 1995 32. Farabegoli F, Ceccarelli C, Santini D, et al: c-erbB-2 over-expression in amplified and non-amplified breast carcinoma samples. Int J Cancer 84: 273-277, 1999[Medline]
33.
Jacobs TW, Gown AM, 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 34. Press MF, Bernstein L, Thomas PA, et al: HER-2/neu gene amplification characterized by fluorescence in situ hybridization: Poor prognosis in node-negative breast carcinomas. J Clin Oncol 15: 2894-2904, 1997[Abstract] 35. Jimenez RE, Wallis T, Tabasczka BS, et al: Determination of HER-2/neu status in breast carcinoma: Comparative analysis of immunohistochemistry and fluorescent in situ hybridization. Mod Pathol 13: 37-45, 2000[Medline] 36. Xing WR, Gilchrist KW, Harris CP, et al: FISH detection of HER-2/neu oncogene amplification in early onset breast cancer. Breast Cancer Res Treat 39: 203-212, 1996[Medline] 37. Vogel C, Cobleigh M, Tripathy D, et al: First-line, non-hormonal, treatment of women with HER2 overexpressing metastatic breast cancer with Herceptin (trastuzumab, humanized anti-HER2 antibody). Proc Am Soc Clin Oncol 19: 71a, 2000 (abstr 275) Submitted October 18, 1999; accepted August 25, 2000.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
