Article Tools
SPECIAL ARTICLES
Article Tools
Guidelines for Human Epidermal Growth Factor Receptor 2 Testing: Biologic and Methodologic Considerations
Abstract
The goal of this review is to systematically address a number of issues raised in the American Society of Clinical Oncology–College of American Pathologists (ASCO-CAP) guidelines on testing for the human epidermal growth factor receptor 2 (HER-2) alteration. A group of investigators who are experienced in the conduct and interpretation of HER-2 assay methods reviewed the ASCO-CAP guidelines and address several areas of the HER-2 testing guidelines with a particular emphasis on biologic and methodologic considerations. Although HER-2 status determined by immunohistochemistry (IHC) and the status determined by fluorescent in situ hybridization (FISH) are significantly correlated, we feel that standard considerations of laboratory testing, including test accuracy, reproducibility, and precision, as well as the current data favor FISH over IHC assay methods for determining HER-2 status. These considerations are clearly important in clinical practice because HER2 amplification is directly linked to protein expression levels in breast cancer. However, this protein is not consistently analyzed in formalin-fixed tissues as a result of variability in fixation methods and times and the impact of fixation on HER-2 protein antigenicity. Conversely, gene amplification and FISH are significantly less dependent on tissue fixation methods, making this assay more reproducible between central and peripheral laboratories than IHC. Moreover, review of the existing data demonstrate that FISH is more strongly correlated with responsiveness to either trastuzumab or lapatinib treatment. Until other methods achieve similar test accuracy, reproducibility, and predictive value, we suggest FISH as the primary HER-2 testing modality for women with breast cancer who are candidates for HER-2–targeted therapies.
The human epidermal growth factor receptor 2 (HER2, CERBB2) gene is amplified in 20% to 25% of human breast cancers.1 HER2 amplification and overexpression are recognized as important markers for aggressive disease and are the molecular targets for specific therapies, such as trastuzumab (Herceptin; Genentech, South San Francisco, CA) and lapatinib (GlaxoSmithKline, London, United Kingdom).
Trastuzumab is a monoclonal humanized antibody that binds the extracellular domain of HER-2 and interferes with the signal transduction cascade initiated by HER-2 overexpression and, possibly, stimulates an immune response to tumor cells overexpressing this receptor.2 In addition, some cytotoxic agents have been shown to interact with trastuzumab in a synergistic fashion.3 Lapatinib is a small-molecule inhibitor of the tyrosine kinase activity of both HER-2 and epidermal growth factor receptors (HER-1).4 Considerable data indicate that HER-2 overexpression is required for responsiveness to trastuzumab therapy.5–8 Preclinical and initial clinical data demonstrate that the same is true for lapatinib.9,10
Both trastuzumab and lapatinib have been approved by the US Food and Drug Administration (FDA) for the treatment of HER-2–positive breast cancer and are clearly associated with improved clinical outcomes in metastatic5,7–9 and, for trastuzumab, early node-positive11,12 and node-negative13 HER-2–positive invasive breast cancers. However, considerable debate exists regarding what test represents the best assessment of HER-2 status. The FDA has approved two immunohistochemical (IHC) assay methods (HercepTest; DAKO, Carpinteria, CA; Pathway; Ventana Medical Systems, Tucson, AZ) and three fluorescent in situ hybridization (FISH) assays (PathVysion; Abbott Laboratories, Abbott Park, IL; INFORM; Ventana Medical Systems, Tucson, AZ; and PHarmDx; DAKO, Glostrup, Denmark) for this purpose. The American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) recently created a set of joint guidelines for the laboratory evaluation of HER-2 status.14,15 Although we agree with many of the points raised in the ASCO-CAP guidelines and agree with the need for increased standardization of HER-2 testing in the clinical setting, we feel that a critical review of these procedures and recommendations is required. Ultimately, HER-2 diagnostic testing should permit the most accurate determination of whether or not an individual tumor specimen contains this alteration. Consequently, we disagree with some of the recommendations of the ASCO-CAP guidelines and summarize our perspective based on a review of published data, biologic considerations, and extensive practical experience.
The ASCO-CAP guidelines describe in detail the methodologic requirements for IHC and FISH analyses. They recommend either using IHC assays for initial evaluation of HER-2 status followed by reflex testing by FISH of some IHC categories or primary use of FISH in initial testing. Although the ASCO-CAP guidelines identify many of the technical shortcomings of both assay methods (Tables 3, 5, and 6),14,15 discussion of important biologic and methodologic aspects of these shortcomings is limited. Considerable published data exist regarding these issues, and we believe that pathologists performing HER-2 tests and oncologists using these data should understand the potential errors inherent in these methods. The biologic and technologic considerations behind HER-2 testing methods are the focus of this perspective.
Key criteria for validation of diagnostic tests should be applied to the various approaches to HER-2 testing. Clinical diagnostic tests must satisfy a number of evaluation criteria, including accurately determining concentrations/levels of the analyte and reproducibility and precision across multiple sites and users. Finally, if assays are used to select patients for biologic or targeted therapies, information about the relative response rates of patients selected using different assay methods should influence diagnostic approaches. When these key primary criteria are met, secondary considerations, such as cost, availability, and ease of application, although important, should never override objective measurements of either test performance or predictive/prognostic value of the test method. We reviewed the biologic rationale as well as the technical limitations for HER-2 testing and conclude that inherent technical properties strongly argue for primary HER-2 FISH testing. Specific reasons for this include the following: HER2 gene amplification is directly linked to protein expression levels in breast cancers; HER-2 protein is not consistently analyzable in formalin-fixed tissues because of nonstandardized fixation methods; and FISH is relatively independent of tissue fixation methods. Accordingly, FISH is more reproducible between central and peripheral laboratories than IHC, and FISH is more accurate for HER-2 measurement and more strongly correlated with responsiveness to trastuzumab or lapatinib treatment. We also show that statements indicating that FISH testing leads to increased costs are incorrect.
Much of the discussion on HER-2 overexpression in the absence of gene amplification begins with early work from two of our groups.16 In a study of the HER2 amplification and expression status in 187 frozen breast cancer specimens, approximately 10% had HER-2 overexpression at the mRNA and protein levels without apparent HER2 amplification as determined by Southern hybridization.16 All remaining specimens showed complete concordance between HER2 amplification and expression status. We referred to the 10% of breast cancer specimens lacking amplification by Southern hybridization but having overexpression as single-copy overexpressors or nonamplified overexpressors.16 We allowed that these specimens may reflect “alterations that occur in control mechanisms for gene expression,” or “alternatively, may represent instances of true gene amplification which are missed because of dilutional artifacts” resulting from admixture of DNA from nonmalignant cells within the tissue.16 Subsequent reanalysis of these same specimens for HER2 amplification by FISH confirmed the latter possibility.17 The overwhelming majority of the single-copy overexpression specimens contained HER2 amplification by FISH. We concluded that the breast cancer specimens showed almost complete agreement between the HER2 amplification status and expression status. Similar results were obtained in a separate study from another one of our laboratories.18 The 1:1 relationship between HER2 gene amplification and HER-2 overexpression is also strongly supported by the fact that no alternative mechanisms giving equivalent expression levels of HER-2 have yet been actually demonstrated. Similar studies have been conducted in other types of primary cancer (eg, prostate, bladder, lung, and GI cancers), and to date, there are no data suggesting mechanisms other than gene amplification as being associated with pathologic HER-2 overexpression.
Analysis of frozen breast cancer samples not only demonstrates that HER2 gene amplification is associated with pathologic levels of protein overexpression, but also demonstrates that the full range of HER-2 expression is not continuous. There are clear and consistent data demonstrating a dichotomous separation between the pathologic overexpression levels in breast cancers with gene amplification and lower expression levels in specimens lacking HER2 amplification.19 Expression levels are substantially higher in amplified specimens, ranging from 500,000 to more than 2,000,000 receptors per tumor cell, whereas nonamplified expression ranges from 25,000 to 185,000 receptors per tumor cell.19 These observations are supported by more recent and extensive transcript expression array data available in the public domain (eg, the Rosetta/Netherlands Cancer Institute expression array data). These data also demonstrate that the dynamic range of HER-2 expression is discontinuous, with the discontinuity occurring between nonamplified and amplified breast cancers provided that samples with intact macromolecules (eg, mRNA) are used for these analyses. Conversely, if the measured RNA is degraded to any degree or the protein has been subjected to fixation artifacts, then a false lowering of the mRNA and/or protein levels results, giving the appearance of amplified specimens with lower levels of expression.
Although the concordance between HER-2 IHC and FISH results has been shown by us20–22 and others23–27 to be statistically significant, the difficulties of standardizing IHC in paraffin-embedded tissue specimens are clearly a problem. Initial reported frequencies of IHC positivity ranged from 2% to almost 50%.28–31 It was thought that the use of a standardized test with standardized control cell lines would solve this problem; however, this approach did not circumvent scoring errors.24,32,33 Frequencies of HER-2 positivity ranging from 30% to 60% in large cohort studies have been reported in studies using the FDA-approved HercepTest (DAKO).29
It is now clear that variable fixation, especially ethanol exposure, and antigen retrieval methods can lead to incorrect IHC results.24,33 Although it may seem that ethanol fixation can be avoided, secondary ethanol fixation does occur in processing of incompletely formalin-fixed tissues during tissue dehydration procedures. This may occur especially if formalin fixation is performed for too short a time or if insufficient quantities of formalin are used in the fixation process. The ASCO-CAP guidelines seek to address this problem by recommending the use of formalin fixation and fixation times of at least 6 hours. However, no data are provided to demonstrate that these times are optimal. The guidelines (Appendix G) also address other IHC interpretation issues and pitfalls related to IHC, including nonspecific binding of HER-2 IHC antibody to areas of tissue altered by crush artifact (eg, needle biopsy specimens), tissue borders, and cautery artifact.
To compensate for false positivity as a result of fixation, tissue processing problems, and antigen retrieval techniques, the ASCO-CAP panel recommended avoiding interpretation of IHC results in cancers with “strong staining of normal breast ducts.”15 Others suggest subtracting the amount of staining observed in normal breast epithelium from the amount of staining observed in breast cancer cells in the same section.27,33 This leads to subtraction of staining intensity in normal tissue from what is already a subjective assessment in the target tissue, making this arbitrary at best. In addition, there are no recommendations on what procedure should be followed if no normal breast epithelium is present in the biopsy specimen. Furthermore, there are no recommendations about the acceptable distance between normal glands and tumor cells for this to function as an effective internal control. This is critical because the degree of secondary ethanol fixation in tissues is unlikely to be uniform across a poorly fixed surgical specimen. Finally, it is unclear what the ASCO-CAP guidelines mean by “strong staining of normal breast ducts” because this is a subjective assessment and epithelial cells of normal ducts do not have HER2 gene amplification, express only normal levels of HER-2, and show only low levels of membrane staining, not strong staining.
Although the ASCO-CAP panel is obviously aware of the possibility of false-positive IHC results, they convey the impression that this preanalysis problem can be solved by following the recommendations they list in Table 5 of their publication (entitled “Sample Exclusion Criteria to Perform or Interpret a HER-2 IHC Assay”).15 However, from a practical standpoint, fixation conditions and times are difficult to standardize in pathology laboratories on a global basis. The fixation conditions of specimens sent to reference laboratories are, in general, unknown. In addition, fixation is dependent on numerous variables beyond just fixation times, including tissue composition (fat v stroma v cancer cells), the ratio of tissue to formalin, and time to gross inspection and dissection of specimens. Furthermore, the quality of fixation may vary across the tissue depending on when formalin and/or ethanol have reached a specific region at a particular concentration.
False-negative HER-2 IHC results are also well documented and vary in frequency from antibody to antibody.18,34 In a study using tissue microarrays, we recently found an IHC HER-2 positivity (2+/3+) rate of 15.6% in samples that were interpretable by FISH but only a 7.6% IHC positivity rate in samples that were not interpretable by FISH (Fig 1).35 Because it is unlikely that true macromolecular differences exist between cancers that demonstrate good or poor FISH signals, we conclude that tissue damage, leading to noninterpretable FISH, also causes decreased HER-2 IHC immunostaining in a considerable fraction of these samples. The use of slides that are not freshly cut represents another potential problem for IHC. In one study, substantial discrepancies were observed in HER-2 IHC results obtained on freshly cut sections compared with those that had been cut and stored for 6 months at 4°C (Fig 2).36 However, the decay of HER-2 protein antigenicity is likely to start even earlier. After only 2 weeks of slide storage, there is a 50% reduction of immunostaining intensity on sections seen with multiple antibodies.37 These results clearly show that several preanalytic issues can cause erroneous IHC results. Unfortunately, such preanalytic problems in test samples cannot be avoided by the use of control cell lines, interlaboratory comparisons, experienced interpreters, or automated quantitation of staining intensity. Because there is no consistent epithelial internal positive control for IHC within the tissue, these problems are not easily recognized. In contrast, normal cells present in breast cancer biopsies, including fibroblasts/lymphocytes, do contain two copies of the HER2 gene that serve as an internal control for the FISH procedure.
Fig 1. Human epidermal growth factor receptor 2 protein (HER2) expression in fluorescent in situ hybridization (FISH) interpretable and noninterpretable breast carcinomas. Note that the observed high fraction of FISH noninterpretable breast carcinomas (26%) in this tissue microarray (TMA) study does not reflect the situation in diagnostic large-section analysis, where only less than 5% of analyses fail. The reduced performance of TMA sections for FISH studies is caused by limitation of the analysis to a 0.6-mm tissue spot and the need to process all of the TMA tissues together under the same hybridization conditions without any variation in proteinase digestion times, as might be required by variable preanalytic tissue fixation conditions. IHC, immunohistochemistry. Data adapted.35
Fig 2. Immunohistochemistry (IHC) staining results in old and fresh tissue microarray sections. These images show examples of simultaneous IHC staining obtained on consecutive sections from individual tumors. (A) Human epidermal growth factor receptor 2 (HER-2) staining on freshly cut section. (B) HER-2 staining on a 6-month-old section (same section as in A). Reprinted with permission.36
As stated earlier, the critical area of subjectivity in IHC interpretation represents a major problem for IHC analyses, and numerous studies document this. Even when experienced pathologists are involved, κ statistics of only 0.67 and 0.74 were achieved for two IHC-based tests, whereas a κ of 0.97 for FISH was observed in the same material.18 The use of automated analysis systems could partially reduce observer subjectivity; however, the selection of the representative area to score remains subjective. Although the current ASCO-CAP guidelines recommend reflex FISH analysis in equivocal IHC 2+ breast cancers to identify the approximately 15% to 48% of these breast cancers that have HER2 gene amplification,20,21,24–27,38,39 the guidelines make no recommendation about identification of the critical 2% to 8% of IHC 0/1+ breast cancers that also have HER2 gene amplification or the 5% to 22% of IHC 3+ breast cancers that lack HER2 gene amplification.20,21,24–27,38,39 The guidelines apparently consider the misclassification of the IHC 0/1+ breast cancers that are FISH amplified as unaltered for HER-2 to be an insignificant issue, as long as the clinical laboratory has 95% concordance with FISH. However, we consider this to be an important problem, especially because these patients can and do show significant responsiveness to HER-2–targeted therapy.40 Moreover, given that the majority of human breast cancers (approximately 75%) are not altered for HER-2 status, the small false-negative IHC rate (2% to 8% that are truly amplified) in this larger group represents a substantial proportion (8% to 25%) of breast cancers with the HER2 gene alteration (because amplification/overexpression is only found in 25% of breast cancers). Equally problematic are IHC 3+ breast cancers that lack HER2 amplification (approximately 5% to 22% false-positive IHC rate) that are, nevertheless, considered candidates for trastuzumab treatment with the associated increased risk of cardiotoxicity and attendant substantial treatment costs, despite the fact that published data demonstrate that these patients have a low probability of responding to HER-2–targeted therapy.6,10
The ASCO-CAP guidelines deal with FISH in a parallel fashion to the recommendations for IHC. Although the tissue processing recommendations in the guidelines, to use only tissues fixed in formalin for 6 to 48 hours, correspond to the recommendations of the FDA, no data have been provided to support restricting FISH assays to tissues fixed for this length of time. Considerable circumstantial evidence suggests that FISH assays, contrary to IHC assays, may be accurately performed on tissues fixed for variable lengths of time and in other fixatives. Although FISH fails in approximately 1% to 5% of tissues because of presumed fixation and/or hybridization problems, these faulty specimens are identified as (hybridization) assay failures, and false results are not reported. The FDA approval for FISH requires the use of formalin-fixed, paraffin-embedded (FFPE) tissue sections; however, the range of fixatives and fixation times compatible with obtaining a successful FISH result are not defined. Our experience and the experience of others with large clinical trials evaluating consecutive patient samples from many countries, with a wide range of fixatives and fixation times, is that the FISH success rate is relatively high, slightly exceeding 98%.21 This high success rate across all tissue samples from approximately 44 countries suggests that the FDA and ASCO-CAP restrictions limiting FISH evaluations to FFPE tissues may be overly restrictive. This is not surprising because, in general, DNA is the most stable of the macromolecules being evaluated (ie, DNA, RNA, and protein).
In FISH analyses, each copy of the HER2 gene and its centromere 17 (CEP17) reference are visible and can be counted in the tissue section (Fig 3). To define amplification, the presence of at least twice as many HER2 signals as CEP17 signals per tumor cell is recommended by us34,41 and others42 based on previous studies with Southern blot analyses.16 This cutoff value, later used in clinical trials, was subsequently accepted by the FDA as the value to differentiate HER2-amplified from HER2-nonamplified breast cancers. Use of this cutoff has correlated well with overexpression of the gene,16,17,43 with more aggressive disease behavior,1,44–46 and with responsiveness to HER-2–targeted therapy.6–8 However, one FDA-approved FISH assay (Oncor/Ventana INFORM) evaluates only the HER2 gene copy number in nuclei. Although we recommend the use of an internal control, such as CEP17, we have shown that similar conclusions are generally reached with either the HER2-to-CEP17 ratio, as originally formulated with the Oncor INFORM FISH assay, or the average HER2 copy number alone, as subsequently used in the Oncor/Ventana INFORM FISH assay.46 However, there are some breast cancers that may be misclassified using this approach. Those breast cancers (approximately 2% to 9% of breast cancers) with increased CEP17 copy number (> four copies per tumor cell) but without HER2 gene amplification may be incorrectly considered as HER2 amplified (false positive) if no CEP17 control is included in the FISH assay.18,47 Despite the fact that these different FISH methods generally yield similar results, for these less frequent breast cancers, the use of an internal control probe from the same chromosome but outside the HER2 amplicon is essential. The FISH ratio is also useful to assist with decision making near the cutoff of 2.0. For example, near the cutoff ratio of 2.0, the average HER2 copy number (numerator) should be increased above an average of 4.0 copies per tumor cell nucleus.46 In addition, the use of a ratio with a control gene on the same chromosome is consistent with the approach used to originally assess gene amplification by Southern hybridization.1,16,45 We consider the use of both HER2 and CEP17 copy numbers to determine a FISH ratio as the optimal and most biologically appropriate approach to select HER2-amplified tumors.
Fig 3. Patterns of human epidermal growth factor receptor 2 (HER2) gene copy alterations as detected by fluorescent in situ hybridization. Red dots indicate the HER2 gene probe, and green signals represent the CEP17 reference probe. (A) Classical amplification with clusters of red HER2 signals. (B) Normal HER2/CEP17 copy numbers. (C) Low-level increase of HER2 relative to CEP17. (D) Coamplification of HER2 and CEP17. More than 95% of breast cancers correspond to categories A and B, either clearly HER2 amplified or clearly not amplified.
The ASCO-CAP guidelines recommend changes in the FISH scoring criteria that differ from the FDA-approved FISH scoring. These guidelines propose to define a range of 1.8 to 2.2 for HER2-to-CEP17 ratios and of 4.0 to 6.0 for average HER2 copy number per tumor cell nucleus as equivocal for HER2 gene amplification. Approximately 2% of breast cancers have HER2 FISH ratios in the 1.8 to 2.2 range (Fig 4).6,21 However, because HER-2 status is used by clinicians and patients to make a bimodal decision (to treat or not to treat), the recommendation to have three diagnostic ranges (not amplified, equivocal, and amplified) creates patient management challenges in the context of a dichotomous treatment decision. We do not believe that creating an equivocal range is currently justified because no data exist demonstrating that the current FDA-approved evaluation criteria, already recommended by the manufacturer, are insufficient.14,15 Rather, we suggest that these FDA-approved recommendations be followed near the cutoff ratio of 2.0 (or average HER2 copy number of 4.0). These recommendations are that when the HER2-to-CEP17 ratio is between 1.80 and 2.20, a minimum of 20 additional cells are scored by the initial scorer and a second scorer counts a minimum of an additional 40 cells. When ratios from these two individuals are in agreement, the FISH result is reported. If the ratios are not in agreement, the entire assay is repeated, and the specimen is rescored. This additional attention is warranted because the increased number of tumor cells evaluated by two individuals will increase the precision of the estimated ratio and improve diagnostic accuracy at this critical cut point.48
Fig 4. Human epidermal growth factor receptor 2 (HER-2) fluorescent in situ hybridization (FISH) ratios in 2,502 consecutive patients (Cases) screened for entry onto Breast Cancer International Research Group clinical trials (trials 005, 006, and 007), as described elsewhere in detail.21 These patients have been ordered (x-axis) according to their HER-2 FISH ratios (y-axis). Please note the marked change in slope that occurs at approximately 2.0 as the breast cancers transition from HER2 not amplified to HER2 amplified. The breast cancer patients with HER2 FISH ratios between 1.80 and 2.20 are identified by vertical lines to demonstrate the approximately 1.8% of breast cancer patients who are within ± 10% of the 2.0 ratio cutoff for HER2 amplification and are referred to in the ASCO-CAP guidelines as indeterminate.
Interpretation of both IHC and FISH assays requires a high level of training, experience, and attention to detail. Essential to the maintenance of any diagnostic standard is the use of internal and external quality audit schemes. These provide a measure of individual laboratory performance and can direct selection of appropriate methodologies for diagnostic use. Taken collectively, issues surrounding fixation, selection of different procedures for testing, and observer bias in analysis provide a key challenge to the wider application of IHC. There is now considerable evidence that IHC performance is poorly controlled in the real world.15,21,26,49,50 The ASCO-CAP guidelines draw attention to this with the alarming claim that “20% of HER-2 assays performed in the field were incorrect.”15 Conversely, the latest evidence on FISH external quality assurance schemes suggests that FISH is much more consistent in external quality audits. The United Kingdom National External Quality Assurance Scheme51 documents performance of diagnostic laboratories within the United Kingdom and across Europe and Asia and includes participants from the United States. Data from this scheme show a marked difference between the levels of acceptable performance for IHC- and FISH-based assays. Remarkably, only between 57% and 65% of participants using the DAKO HercepTest demonstrated acceptable performance, whereas performance of laboratories performing FISH testing have been consistently above this level, with 89% to 96% of laboratories demonstrating acceptable performance for this approach in the most recent United Kingdom National External Quality Assurance Scheme analysis (J. Bartlett, personal communication, October 2007).
Studies evaluating interlaboratory agreement for FISH assays generally show high concordance rates, ranging from 92%20,21 to 99%,49 if unselected groups of patients are analyzed. Our own experience includes 2,502 samples previously screened as whole tissue sections using an FDA-approved HER-2 FISH assay for gene amplification.21 When these same samples were re-evaluated in a blinded fashion as tissue microarrays using a different FISH assay to simultaneously assess HER2, topoisomerase II-α (TOP2A), and CEP17 status, the agreement between the laboratories for HER2 status was 99.4% (unpublished data). High concordance rates are expected in consecutive patient sets based on the approximately 95% straightforward cases in unselected tumor series. Therefore, it is not surprising that the CAP proficiency testing survey results also showed that FISH assay comparisons between clinical laboratories demonstrated the highest level of interlaboratory agreement that CAP had observed for a test performed as part of its proficiency testing program.32,52,53 When CAP published the initial results of its proficiency testing program for HER-2 FISH assays, they found that 100% of the laboratories participating in the program correctly classified the unknown samples using FISH.52 This initial publication reported results from the first 2 survey years with 35 and 63 laboratories participating. Moreover, subsequent published reports from the CAP proficiency testing program have consistently shown a high rate of agreement among the 139 laboratories participating in the FISH testing program.53 Similar findings have been reported for the proficiency testing programs in the United Kingdom.48
However, the ASCO-CAP guidelines also mention three FISH studies with markedly lower concordance rates.26,50,54 From these three studies, they conclude that no gold standard exists for HER-2 testing because both tests (IHC and FISH) suffer from comparable problems. We feel strongly that users of HER-2 testing consider these three specific studies in greater detail. All three studies come from the same laboratory, and all three suffer from the problem that laboratories using FISH for retesting of problematic IHC samples enrich their data set for borderline FISH samples for which scoring is subject to higher error rates. The largest of these studies included 2,535 patients, 829 of whom were initially analyzed by FISH in local laboratories and subsequently reanalyzed in a central laboratory.26 Because inclusion criteria for this study included “women whose tumors stain 0 to 2+ by local community-based immunohistochemistry (IHC) but demonstrate amplification by local community-based FISH,”50 the composition of this patient sample set is likely skewed. Importantly, despite this probable selection bias, FISH showed a significantly higher agreement rate (88%) between laboratories than IHC (82%).26 The other studies used as an example of poor FISH reproducibility in the ASCO-CAP guideline article are from the same laboratory and include subsets of the same patients.54,55 Again, we consider it possible that some of the nine patients (which in itself represented an extremely small number) analyzed by FISH and reported in the first of the three studies as representative of FISH nonreproducibility were selected as IHC 2+.50 For their own quality assessment, laboratories using FISH primarily for reflex testing of problematic IHC 2+ samples should be aware that they may be accumulating a disproportionate number of borderline (1.8 to 2.2 ratio) FISH samples.
In the ASCO-CAP guidelines, the authors stated that no gold standard exists for HER-2 testing.14,15 This conclusion is apparently based on the contradictory results reported in the literature comparing HER-2 status determined by FISH and IHC assays in FFPE tissue specimens (> 100 studies in Medline since 2000). Given the technical issues outlined earlier, such discrepancies are expected in studies using FFPE tissues where nonstandardized tissue processing greatly affects IHC results. Realistically, only FFPE tissue specimens, not frozen tissue specimens, will be available for evaluation of HER-2 status in the clinical setting, so we consider it important to assess the accuracy (not concordance) of assay methods used to analyze HER-2 status in this type of material. For this purpose, it is imperative to thoroughly validate HER-2 results obtained from FFPE tissues with HER-2 levels in unfixed reference tissues analyzed with alternative quantitative assays and/or to compare HER-2 results with clinical outcomes.
Only a few studies have validated the accuracy of HER-2 testing in tissues with known HER-2 status. Two independent studies comparing HER-2 FISH and IHC data obtained on FFPE tissues with frozen tissue data from those same specimens have been performed, and both arrived at similar conclusions.18,43 In one study, the HercepTest, the CB11 immunohistochemical assay, and FISH were used to analyze a series of 191 FFPE breast carcinomas that were previously characterized for HER-2 expression levels by quantitative radioimmunohistochemistry on frozen specimens. This comparative evaluation of reference tissues clearly demonstrated that FISH was the most accurate, reproducible, and precise assay for assessment of known HER-2–overexpressing breast cancers.18
In another study, a separate group of 117 FFPE breast cancer specimens with known HER-2 status were used to compare results from the HercepTest IHC assay, the Pathway IHC assay, the Ventana FISH assay, and the PathVysion FISH assay.43 The HER-2 status of these reference breast cancers was first determined by fully molecularly characterizing fresh-frozen tissues from the same specimens using Southern hybridization to determine HER2 amplification levels, Northern hybridization to determine HER2 mRNA levels, and Western immunoblot of tissue lysates as well as frozen section immunohistochemistry to determine HER-2 protein levels in each specimen. When FFPE tissue sections from these exact same specimens were evaluated, the correct HER-2 status was consistently more accurately determined using FISH rather than IHC, regardless of whether the FDA-approved IHC HercepTest assay or the FDA-approved IHC Pathway assay was used for assessing HER-2 status.43,56
Although work in frozen tissue specimens shows that the HER-2 status can be equally well determined at either the DNA level or protein level,16 this same concordance level has not been attained in FFPE specimens for the reasons enumerated earlier. We believe that the technical issues related to HER-2 IHC in FFPE tissues have not yet been overcome and, therefore, strongly suggest the use of FISH as the primary method for assessment of all FFPE breast cancer specimens to determine HER-2 status.
Few studies have compared the utility of IHC and FISH tests for prediction of response to trastuzumab therapy. Reanalysis of response data by IHC and FISH results from three different pivotal clinical trials of trastuzumab in women with metastatic breast cancer5,7,8 again demonstrates that FISH is superior to IHC in predicting response to trastuzumab treatment.6 Entry onto these clinical trials required 2+ or 3+ IHC staining according to a clinical trials IHC assay. This clinical trials IHC assay led to a substantial enrichment of patients entering the trials whose breast cancers had the HER-2 alteration; however, approximately 25% of these patients had breast cancers that did not contain HER2 gene amplification despite 2+ or 3+ IHC. Tumors from 765 (96%) of 799 of these patients were reanalyzed by FISH. Only one (1.5%) of 65 patients whose breast cancers lacked HER2 gene amplification (HER2 FISH ratio < 2.0) responded to single-agent trastuzumab (Table 1).6 Conversely, women whose breast cancers had HER2 amplification by FISH (HER2 FISH ratio ≥ 2.0) showed a 24% response rate to single-agent trastuzumab (Table 1). This included women who had breast cancers with HER2 FISH ratios between 2.0 and 3.0.57 Recent data from the Herceptin Adjuvant clinical trial presented at the 30th Annual San Antonio Breast Cancer Symposium in 2007 showed that women treated in the adjuvant setting whose breast cancers have HER2 FISH ratios between 2.0 and 4.0 have similar responses to trastuzumab as women whose breast cancers have HER2 FISH ratios of 4.0 to 6.0 and 6.0 to 8.0.58 Similar observations have also been made for lapatinib in two large clinical trials of women with metastatic breast cancer.40 In addition, in women with metastatic breast cancer enrolled onto the H0648 clinical trial, which resulted in the initial approval for trastuzumab, there was no significant improvement with trastuzumab therapy in either response rate or overall survival in women whose breast cancers lacked HER2 amplification by FISH even if they had been scored as 3+ by IHC.6 A similar lack of responsiveness to a lapatinib-based regimen compared with chemotherapy alone was observed for women with metastatic disease whose breast cancers lacked HER2 amplification by FISH.10 For trastuzumab, only women whose breast cancers showed HER2 amplification by FISH demonstrated a significant improvement in both response rate and overall survival when treated with chemotherapy and trastuzumab compared with chemotherapy alone6–8 (Fig 5). Similar observations have been made in other clinical trials for both trastuzumab59 and lapatinib.10 On the basis of these combined data, FISH clearly seems to be the best method of correctly identifying patients with metastatic breast cancer who are likely to respond to trastuzumab therapy. Importantly, these distinctions between assay methods and treatment response rates are not as apparent if all HER-2–positive patients by each assay method (IHC or FISH) are pooled and compared for response to treatment (Fig 5C). However, when the discrepant patients (eg, FISH-negative, IHC-positive patients) are plotted separately, it becomes apparent that trastuzumab provides no additional survival benefit beyond that of chemotherapy alone in the FISH-negative patients (Fig 5B).
|
| Study and HER2 Amplification | No. of Assessable Patients | Objective Response (CR plus PR) | Time to Progression (months) | Survival Time (months) | |||
|---|---|---|---|---|---|---|---|
| No. | % | Median | Range | Median | Range | ||
| H0649g | |||||||
| FISH positive | 173 | 33 | 19 | 3.2 | 2.6-3.5 | 14.2 | 12.4-18.1 |
| FISH negative | 36 | 0 | 0 | 1.9 | 1.5-2.8 | 8.8 | 5.8-15.6 |
| H0650g | |||||||
| FISH positive | 82 | 28 | 34 | 4.9 | 3.5-6.3 | 24.5 | 17.4-36.1 |
| FISH negative | 29 | 1* | 3.5 | 1.7 | 1.5-3.3 | 24.4 | 10.8-34.7 |
NOTE. Data adapted.(6)
Abbreviations: HER2, human epidermal growth factor receptor 2; CR, complete response; PR, partial response; FISH, fluorescent in situ hybridization.
*The original article included two nonamplified patients as partial responders. However, as stated in the discussion of this article, “Because the data from the H0649g trial suggested no benefit of trastuzumab in the HER2-negative cohort, further supplemental analysis of these 2 tissue specimens revealed the first of these samples to actually have HER2 amplification with a HER2:CEP17 ratio of 5.70, whereas the second specimen remained nonamplified, with a ratio of 1.67.”(6)
Fig 5. Overall survival time by human epidermal growth factor receptor 2 (HER2) amplification status for patients with metastatic immunohistochemistry (IHC)–positive (IHC 2+/3+) breast cancers receiving chemotherapy and trastuzumab versus chemotherapy alone in the pivotal H0648 registrational clinical trial for trastuzumab. (A) Patients with fluorescent in situ hybridization (FISH)–positive tumors. (B) Patients with FISH-negative tumors. (C) Comparison of all patients in the H0648 trial (all had IHC-positive [2+/3+] breast cancers) and FISH-positive breast cancer patients by treatment group and overall survival. Note that IHC positive (2+/3+) includes both HER-2 FISH-positive and FISH-negative patients. H, trastuzumab; CT, chemotherapy. Adapted with permission.6
Some have argued that FISH is a more expensive method than IHC for HER-2 determination.27 However, an analysis of total cost and benefits supports primary FISH testing as the most cost-effective approach for patient management. To best inform policy makers, HER-2 testing guidelines should also consider an economic evaluation that assesses the clinical benefits and adverse effects of therapy measured in quality-adjusted life-years; the observed response to therapy with various testing methods; and the cost of testing, monitoring, and treatment. This is especially important when considering that the cost of trastuzumab is approximately $1,000 per treatment once a week for 52 weeks and the cost of lapatinib is approximately $120 per daily treatment for 6 months. In addition, trastuzumab administered either concurrently with or after anthracycline chemotherapy has significantly increased anthracycline-associated cardiac toxicity rates.7,11–13 In a recent review of economic models, adjuvant therapy seems cost effective, except possibly in some subpopulations of patients.60 The cost effectiveness of alternative HER-2 testing methods was evaluated for trastuzumab for metastatic disease, and it was demonstrated that either primary FISH testing or evaluation of all IHC-positive patients with FISH assay was most cost effective. However, these conclusions are sensitive to the accuracy of the testing regimen.61
Calculation of net changes in HER-2 status can be performed by comparison of IHC with FISH assessments in cohorts of patient samples (Tables 2, 3, and 4). The disparities in IHC and FISH test results are listed for several recent publications (Table 2). The referenced studies include English language articles published since January 2001 that provide a direct comparison of IHC and FISH from consecutive samples. To allow comparison of a primary FISH versus retest of IHC 2+ regimens, the published results had to include both negative and positive IHC and FISH observations. The table lists the number of expected trastuzumab candidates if primary FISH is the test regimen (FISH-positive patients are trastuzumab candidates) versus a regimen of retesting IHC 2+ with FISH (IHC 3+ and IHC 2+/FISH-positive patients are trastuzumab candidates). The total number of discordant observations and change in expected number of trastuzumab or lapatinib candidates are listed. Discordant results range from 1.9% to 9.0% of patients evaluated, whereas the net change in candidates ranges from a 40% decrease to a 14.4% increase. Evaluation of HER-2 status should consider costs of treatment and testing as well as potential benefits of targeted therapy. The cost of the diagnostic test is minimal compared with the substantial cost of the therapy. More accurate assignment of patients to treatment more than offsets the costs of erroneously treating women whose breast cancers lack HER2 amplification. In addition, there is a human cost of failing to treat women whose breast cancers have HER2 amplification.
|
| Study | No. of Patients | No. of New Trastuzumab Candidates (IHC 0 or 1+, FISH positive) | No. of Reduced Trastuzumab Candidates (IHC 3+, FISH negative) | Discordant Observations | Net Change in Candidates | ||
|---|---|---|---|---|---|---|---|
| No. | % | No. | % | ||||
| Dowsett et al(38) | 426 | 2 | 6 | 8 | 1.9 | −4 | −3.1 |
| Elkin et al(61) | 1,527 | 34 | 19 | 53 | 3.5 | 15 | 3.6 |
| Press et al(56) | 1,190 | 38 | 11 | 49 | 4.1 | 27 | 14.4 |
| Dybdal et al(20) | 529 | 9 | 21 | 30 | 5.7 | −12 | −5.5 |
| Press et al(21)* | 842 | 19 | 57 | 76 | 9.0 | −38 | −13.5 |
| Press et al(21)† | 1,407 | 35 | 89 | 124 | 8.8 | −54 | −12.3 |
| Owens et al(25) | 6,556 | 163 | 59 | 222 | 3.4 | 104 | 7.2 |
| Rossi et al(62) | 81 | 0 | 6 | 6 | 7.4 | −6 | −18.8 |
| Ciampa et al(23) | 108 | 2 | 20 | 22 | 20.4 | −18 | −40.0 |
| Yaziji et al(27) | 2,913 | 40 | 28 | 68 | 2.3 | 12 | 2.3 |
NOTE. To demonstrate how the results were calculated, the observations reported in Press et al,(21) where FISH was performed at a central laboratory, are compared in Tables 3 and 4 to DAKO HercepTest performed at local and community hospital laboratories for 842 observations.
Abbreviations: FISH, fluorescent in situ hybridization; IHC, immunohistochemistry.
*Local testing using DAKO HercepTest.
†Local testing using any IHC.
|
| FISH in Central Laboratory | IHC With DAKO Herceptest in Local/Community Hospital Labs (No. of samples) | ||||
|---|---|---|---|---|---|
| 0 | 1+ | 2+ | 3+ | Total | |
| Positive | 11 (a) | 8 (b) | 21 (c) | 204 (d) | 244 (e) |
| Negative | 296 (f) | 142 (g) | 103 (h) | 57 (i) | 598 (j) |
Abbreviations: FISH, fluorescent in situ hybridization; IHC, immunohistochemistry.
|
| Value | Definition | Example Calculations |
|---|---|---|
| New trastuzumab candidates | IHC 0 or 1+, FISH positive | a + b = 11 + 8 = 19 |
| Reduced trastuzumab candidates | IHC 3+, FISH negative | i = 57 |
| Discordant observations (No. and %) | Sum of above two values | a + b + i = 19 + 57 = 76; 76/842 = 9.0% |
| Trastuzumab candidates with retest of IHC 2+ | IHC 3+ or IHC 2+ and FISH positive | d + i + c = 204 + 57 + 21 = 282 |
| Trastuzumab candidates with primary FISH | FISH positive | a + b + c + d = e = 244 |
| Net change in candidates using primary FISH (No. and %) | Difference in above two values | 244 – 282 = −38;−38/282 = −13.5% |
NOTE. See Table 3 for values of letters.
Abbreviations: IHC, immunohistochemistry; FISH, fluorescent in situ hybridization.
Although HER-2 IHC status is strongly correlated with HER-2 amplification status determined by FISH, IHC is, nevertheless, significantly less accurate in assigning tumors with known HER-2 status to their correct classification. In addition, FISH-negative status in metastatic breast cancer is associated with lack of responsiveness to either trastuzumab or lapatinib. Moreover, only women with FISH-positive metastatic breast cancers have shown significant improvement in outcomes when receiving trastuzumab- or lapatinib-based combination regimens. IHC is associated with a number of both false-negative and false-positive results that are related to a number of critical factors including preanalytic tissue processing, lack of antibodies that function consistently in variably processed FFPE tissue, increased staining related to antigen retrieval techniques, and subjective scoring of membrane staining. IHC errors have not been sufficiently reduced by use of cell line controls, interlaboratory comparisons, experienced interpreters, or automated immunostaining and quantitation. So far, HER2 gene amplification is the only established mechanism for pathologic HER-2 protein overexpression seen in human breast cancers. DNA is less affected by tissue processing artifacts, and therefore, experimental errors occur less frequently in HER2 gene copy number assessment by FISH. For these reasons, we believe that FISH is the preferable method for HER-2 analysis, although the protein (and not the DNA) is the target for treatment. The main problem with HER-2 FISH is that approximately 1% to 2% of breast cancers have FISH ratios at or near the cutoff for gene amplification according to the existing definitions. However, we are aware of no published data demonstrating that patients with borderline HER2 amplification respond less favorably to trastuzumab or lapatinib. Until such data are available, we do not advocate changing the FISH criteria for classification of HER2 amplification because the decision to treat or not to treat with HER-2–targeted therapies is a dichotomous decision. Given substantial sales of trastuzumab and of lapatinib each year, a large number of such borderline patients should be available for studies and should permit demonstration of the optimal cutoff based on outcome data. Because the use of FISH already reduces the number of patients to be treated compared with IHC analysis, it is hoped that additional studies will provide more clarity on the molecular determinants of trastuzumab and/or lapatinib responsiveness and lead to appropriate treatment of the correct patients, resulting in improved efficacy of HER-2–targeted therapies.
Supported in part by the Breast Cancer International Research Group, grants from the Breast Cancer Research Foundation and Expedition Inspiration, and Grant No. CA48780 from the National Cancer Institute.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment or Leadership Position: None Consultant or Advisory Role: James Lee, Abbott Laboratories (C); Michael F. Press, Genentech (C), GlaxoSmithKline (C) Stock Ownership: None Honoraria: Guido Sauter, Abbott Laboratories; John M.S. Bartlett, Abbott Laboratories, Dako; Dennis J. Slamon, Aventis Pharmaceuticals, Genentech BioOncology, Roche Pharmaceuticals; Michael F. Press, Genentech, GlaxoSmithKline Research Funding: James Lee, Abbott Laboratories; Michael F. Press, Genentech, GlaxoSmithKline Expert Testimony: None Other Remuneration: None
Conception and design: Guido Sauter, James Lee, John M.S. Bartlett, Dennis J. Slamon, Michael F. Press
Provision of study materials or patients: Guido Sauter, Michael F. Press
Collection and assembly of data: Guido Sauter, James Lee, Dennis J. Slamon, Michael F. Press
Data analysis and interpretation: Guido Sauter, James Lee, John M.S. Bartlett, Dennis J. Slamon, Michael F. Press
Manuscript writing: Guido Sauter, James Lee, John M.S. Bartlett, Dennis J. Slamon, Michael F. Press
Final approval of manuscript: Guido Sauter, James Lee, John M.S. Bartlett, Dennis J. Slamon, Michael F. Press
| 1. | D Slamon, G Clark, S Wong , etal: Human breast cancer: Correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235: 177– 182,1987 Crossref, Medline, Google Scholar |
| 2. | MM Goldenberg: Trastuzumab, a recombinant DNA-derived humanized monoclonal antibody, a novel agent for the treatment of metastatic breast cancer. Clin Ther 21: 309– 318,1999 Crossref, Medline, Google Scholar |
| 3. | A Kataoka, M Ishida, S Murakami , etal: Sensitization of chemotherapy by anti-HER. Breast Cancer 11: 105– 115,2004 Crossref, Medline, Google Scholar |
| 4. | TE Kim, JR Murren: Lapatinib ditosylate GlaxoSmithKline. IDrugs 6: 886– 893,2003 Medline, Google Scholar |
| 5. | M Cobleigh, C Vogel, D Tripathy , etal: 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 Link, Google Scholar |
| 6. | R Mass, M Press, S Anderson , etal: Evaluation of clinical outcomes according to HER2 detection by fluorescence in situ hybridization in women with metastatic breast cancer treated with trastuzumab. Clinical Breast Cancer 6: 240– 246,2005 Crossref, Medline, Google Scholar |
| 7. | D Slamon, B Leyland-Jones, S Shak , etal: Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344: 783– 792,2001 Crossref, Medline, Google Scholar |
| 8. | C Vogel, M Cobleigh, D Tripathy , etal: 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 Link, Google Scholar |
| 9. | C Geyer, J Forster, D Lindquist , etal: Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med 355: 2733– 2743,2006 Crossref, Medline, Google Scholar |
| 10. | A Di Leo, H Gomez, Z Aziz , etal: Phase III, double-blind, randomized study comparing lapatinib plus paclitaxel with placebo plus paclitaxel as first-line treatment for metastatic breast cancer. J Clin Oncol 26: 5544– 5552,2008 Link, Google Scholar |
| 11. | M Piccart-Gebhart, M Procter, B Leyland-Jones , etal: Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 353: 1659– 1672,2005 Crossref, Medline, Google Scholar |
| 12. | E Romond, E Perez, J Bryant , etal: Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 353: 1673– 1684,2005 Crossref, Medline, Google Scholar |
| 13. | D Slamon, W Eiermann, N Robert , etal: Phase III trial comparing AC-T with AC-TH and with TCH in the adjuvant treatment of HER2 positive early breast cancer patients: Second interim efficacy analysis. Presented at the 29th Annual San Antonio Breast Cancer Symposium December 14-17, 2006 San Antonio, TX Google Scholar |
| 14. | A Wolff, M Hammond, J Schwartz , etal: American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. Arch Pathol Lab Med 131: 18,2007 Medline, Google Scholar |
| 15. | A Wolff, M Hammond, J Schwartz , etal: American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol 25: 118– 145,2007 Link, Google Scholar |
| 16. | D Slamon, W Godolphin, L Jones , etal: Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244: 707– 712,1989 Crossref, Medline, Google Scholar |
| 17. | G Pauletti, W Godolphin, M Press , etal: 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, Google Scholar |
| 18. | J Bartlett, J Going, E Mallon , etal: Evaluating HER2 amplification and overexpression in breast cancer. J Pathol 195: 422– 428,2001 Crossref, Medline, Google Scholar |
| 19. | G Konecny, G Pauletti, M Pegram , etal: Quantitative association between HER-2/neu and steroid hormone receptors in hormone receptor-positive primary breast cancer. J Natl Cancer Inst 95: 142– 153,2003 Crossref, Medline, Google Scholar |
| 20. | N Dybdal, G Leiberman, S Anderson , etal: Determination of HER2 gene amplification by fluorescence in situ hybridization and concordance with the clinical trials immunohistochemical assay in women with metastatic breast cancer evaluated for treatment with trastuzumab. Breast Cancer Res Treat 93: 3– 11,2005 Crossref, Medline, Google Scholar |
| 21. | M Press, G Sauter, L Bernstein , etal: Diagnostic evaluation of HER-2 as a molecular target: An assessment of accuracy and reproducibility of laboratory testing in large, prospective, randomized clinical trials. Clin Cancer Res 11: 6598– 6607,2005 Crossref, Medline, Google Scholar |
| 22. | G Pauletti, S Dandekar, H Rong , etal: Assessment of methods for tissue-based detection of the HER-2/neu alteration in human breast cancer: A direct comparison of fluorescence in situ hybridization and immunohistochemistry. J Clin Oncol 18: 3651– 3664,2000 Link, Google Scholar |
| 23. | A Ciampa, B Xu, G Ayata , etal: HER-2 status in breast cancer: Correlation of gene amplification by FISH with immunohistochemistry expression using advanced cellular imaging system. Appl Immunohistochem Mol Morphol 14: 132– 137,2006 Crossref, Medline, Google Scholar |
| 24. | T Jacobs, A Gown, H Yaziji , etal: 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 Link, Google Scholar |
| 25. | MA Owens, BC Horten, MM Da Silva: HER2 amplification ratios by fluorescence in situ hybridization and correlation with immunohistochemistry in a cohort of 6556 breast cancer tissues. Clin Breast Cancer 5: 63– 69,2004 Crossref, Medline, Google Scholar |
| 26. | E Perez, V Suman, N Davidson , etal: HER2 testing by local, central, and reference laboratories in specimens from the North Central Cancer Treatment Group N9831 intergroup adjuvant trial. J Clin Oncol 24: 3032– 3038,2006 Link, Google Scholar |
| 27. | H Yaziji, L Goldstein, T Barry , etal: HER-2 testing in breast cancer using parallel tissue-based methods. JAMA 291: 1972– 1977,2004 Crossref, Medline, Google Scholar |
| 28. | D Barnes, G Lammie, R Millis , etal: An immunohistochemical evaluation of c-erbB-2 expression in human breast carcinoma. Br J Cancer 58: 448– 452,1988 Crossref, Medline, Google Scholar |
| 29. | P Roche, J Ingle: Increased HER2 with U.S. Food and Drug Administration-approved antibody. J Clin Oncol 17: 434,1999 Link, Google Scholar |
| 30. | C De Potter, C Beghin, A Makar , etal: The neu-oncogene protein as a predictive factor for haematogenous metastases in breast cancer patients. Int J Cancer 45: 55– 58,1990 Crossref, Medline, Google Scholar |
| 31. | J Baselga, D Tripathy, J Mendelsohn , etal: Phase II study of weekly intravenous trastuzumab (Herceptin) in patients with HER2/neu-overexpressing metastatic breast cancer. Semin Oncol 26: 78– 83,1999 Medline, Google Scholar |
| 32. | R Tubbs, J Pettay, P Roche , etal: Discrepancies in clinical laboratory testing of eligibility for trastuzumab therapy: Apparent immunohistochemical false-positives do not get the message. J Clin Oncol 19: 2714– 2721,2001 Link, Google Scholar |
| 33. | T Jacobs, A Gown, H Yaziji , etal: 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 Link, Google Scholar |
| 34. | M Press, G Hung, W Godolphin , etal: 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 Medline, Google Scholar |
| 35. | C Tapia, P Schraml, R Simon , etal: HER2 analysis in breast cancer: Reduced immunoreactivity in FISH non-informative cancer biopsies. Int J Oncol 25: 1551– 1557,2004 Medline, Google Scholar |
| 36. | M Mirlacher, M Kasper, M Storz , etal: Influence of slide aging on results of translational research studies using immunohistochemistry. Mod Pathol 17: 1414– 1420,2004 Crossref, Medline, Google Scholar |
| 37. | TW Jacobs, JE Prioleau, IE Stillman , etal: Loss of tumor marker-immunostaining intensity on stored paraffin slides of breast cancer. J Natl Cancer Inst 88: 1054– 1059,1996 Crossref, Medline, Google Scholar |
| 38. | M Dowsett, J Bartlett, I Ellis , etal: Correlation between immunohistochemistry (HercepTest) and fluorescence in situ hybridization (FISH) for HER-2 in 426 breast carcinomas from 37 centres. J Pathol 199: 418– 423,2003 Crossref, Medline, Google Scholar |
| 39. | P Lal, P Salazar, C Hudis , etal: HER-2 testing in breast cancer using immunohistochemical analysis and fluorescence in situ hybridization: A single-institution experience of 2,279 cases and comparison of dual-color and single-color scoring. Am J Clin Pathol 121: 631– 636,2004 Crossref, Medline, Google Scholar |
| 40. | M Press, R Finn, D Cameron , etal: HER2 gene amplification, HER2 and EGFR messenger RNA and protein expression and lapatinib efficacy in women with metastatic breast cancer. Clin Cancer Res 14: 7861– 7870,2008 Crossref, Medline, Google Scholar |
| 41. | G Sauter, H Moch, D Moore , etal: Heterogeneity of erbB-2 gene amplification in bladder cancer. Cancer Res 53: 2199– 2203,1993 Medline, Google Scholar |
| 42. | OP Kallioniemi, A Kallioniemi, W Kurisu , etal: ERBB2 amplification in breast cancer analyzed by fluorescence in situ hybridization. Proc Natl Acad Sci U S A 89: 5321– 5325,1992 Crossref, Medline, Google Scholar |
| 43. | M Press, D Slamon, K Flom , etal: Evaluation of HER-2/neu gene amplification and overexpression: Comparison of frequently used assay methods in a molecularly characterized cohort of breast cancer specimens. J Clin Oncol 20: 3095– 3105,2002 Link, Google Scholar |
| 44. | M Press, M Pike, V Chazin , etal: 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 Medline, Google Scholar |
| 45. | D Slamon, M Press, W Godolphin , etal: Studies of the HER-2/neu proto-oncogene in human breast cancer. Cancer Cells 7: 371– 380,1989 Google Scholar |
| 46. | M Press, L Bernstein, P Thomas , etal: 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 Link, Google Scholar |
| 47. | AD Watters, JJ Going, TG Cooke , etal: Chromosome 17 aneusomy is associated with poor prognostic factors in invasive breast carcinoma. Breast Cancer Res Treat 77: 109– 114,2003 Crossref, Medline, Google Scholar |
| 48. | I Ellis, J Bartlett, M Dowsett , etal: Updated recommendations for HER2 testing in the UK. J Clin Pathol 57: 233– 237,2004 Crossref, Medline, Google Scholar |
| 49. | S Paik, J Bryant, E Tan-Chiu , etal: Real-world performance of HER2 testing: National Surgical Adjuvant Breast and Bowel Project experience. J Natl Cancer Inst 94: 852– 854,2002 Crossref, Medline, Google Scholar |
| 50. | P Roche, V Suman, R Jenkins , etal: Concordance between local and central laboratory HER2 testing in the breast intergroup trial N9831. J Natl Cancer Inst 94: 855– 857,2002 Crossref, Medline, Google Scholar |
| 51. | UKNEQAS Immunocytochemistry Journal. United Kingdom National External Quality Assurance Scheme http://www.ukneqasicc.ucl.ac.uk/neqasicc.shtml Google Scholar |
| 52. | Clinical laboratory assays for HER-2/neu amplification and overexpression: Quality assurance, standardization, and proficiency testing. Arch Pathol Lab Med 126: 803– 808,2002 Cell Markers and Cytogenetics Committees College of American Pathologists Medline, Google Scholar |
| 53. | D Persons, R Tubbs, L Cooley , etal: HER-2 fluorescence in situ hybridization: Results from the survey program of the College of American Pathologists. Arch Pathol Lab Med 130: 325– 331,2006 Medline, Google Scholar |
| 54. | E Perez, V Suman, N Davidson , etal: HER2 testing by local, central, and reference laboratories in the NCCTG N9831 Intergroup adjuvant trial. J Clin Oncol 22: 18s,2004 suppl abstr 567 Crossref, Google Scholar |
| 55. | EA Perez, PC Roche, RB Jenkins , etal: HER2 testing in patients with breast cancer: Poor correlation between weak positivity by immunohistochemistry and gene amplification by fluorescence in situ hybridization. Mayo Clin Proc 77: 148– 154,2002 Crossref, Medline, Google Scholar |
| 56. | M Press, S Anderson, N Dybdal , etal: Fluorescence in situ hybridization (FISH) is superior to immunohistochemistry (IHC) for determining HER2 status: The Herceptin experience. Lab Invest 82: 47A,2002 abstr 186 Medline, Google Scholar |
| 57. | M Press, L Bernstein, J-Y Zhou , etal: Low-level HER-2/neu gene amplification by fluorescence in situ hybridization and responsiveness to single agent Herceptin immunotherapy among women with metastatic breast cancer. Presented at the 31st Annual San Antonio Breast Cancer Symposium December 10-14, 2004 San Antonio, TX Google Scholar |
| 58. | W McCaskill-Stevens, M Procter, E Azambuja , etal: Disease-free survival according to local immunohistochemistry for HER2 and central FISH for patients treated with adjuvant chemotherapy with/without trastuzumab in the HERA trial. 30th Annual San Antonio Breast Cancer Symposium December 13-16, 2007 San Antonio, TX abstr 1069 Google Scholar |
| 59. | A Seidman, M Fornier, F Esteva , etal: Weekly trastuzumab and paclitaxel therapy for metastatic breast cancer with analysis of efficacy by HER2 immunophenotype and gene amplification. J Clin Oncol 19: 2587– 2595,2001 Link, Google Scholar |
| 60. | J Norum: The cost effectiveness issue of adjuvant trastuzumab in early breast cancer. Expert Opin Pharmacother 7: 1617– 1625,2006 Crossref, Medline, Google Scholar |
| 61. | E Elkin, M Weinstein, E Winer , etal: HER-2 testing and trastuzumab therapy for metastatic breast cancer: A cost-effectiveness analysis. J Clin Oncol 22: 854– 863,2004 Link, Google Scholar |
| 62. | E Rossi, A Ubiali, M Cadei , etal: HER-2/neu in breast cancer: A comparative study between histology, immunohistochemistry, and molecular technique (FISH). Appl Immunohistochem Mol Morphol 14: 127– 131,2006 Crossref, Medline, Google Scholar |
Acknowledgment
We thank Ronald Simon, MD, and Ivonne E. Villalobos, MHA, for assistance in the preparation of this article.
