| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Institutes of Pathology,
*
the Department of Clinical Biochemistry,
and the Department of Clinical Genetics,
Aarhus University Hospital, Aarhus, Denmark
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HER-2 status can be analyzed at the DNA-, the mRNA-, or the protein level. Various techniques are available for assessing these different target molecules, each with benefits and drawbacks.3 For practical reasons, immunohistochemistry (IHC) using an anti-HER-2 antibody is by far the most frequently used method for assessing HER-2 status. IHC is available as a standard technique in pathology laboratories, and can be easily used on archival FFPE tissues. Furthermore, it is tumor cell-specific, the observer being able to distinguish signals from neoplastic and non-neoplastic cells. As a result, reports published on the clinical relevance of HER-2 expression have almost all used IHC. However, major disadvantages of HER-2 IHC assessment are that it is at best a semi-quantitative method with considerable inter-observer variations reported in some studies.4 Furthermore, the many available HER-2 antibodies show variable sensitivities and specificities5 that, together with differences in tumor fixation, antigen retrieval method, and IHC technique, contribute to considerable assay imprecision. While this imprecision is improved by the use of standardized IHC tests (such as the HercepTest), there is widespread agreement that IHC HER-2 scores at the 2+ level need to be confirmed with another technique.6, 7
Fluorescence in situ hybridization (FISH) is a common alternative method for assessing HER-2 status. It is a sensitive and apparently highly specific technique for measuring HER-2 gene amplification at the single-cell level,8 a change that appears to be closely correlated with strong protein expression9 and thus of clinical relevance. In contrast to IHC, FISH can give a more objective and quantitative measure of HER-2 abnormalities, and is widely regarded as the gold standard for assaying HER-2 status. However, the technique is more labor-intensive and expensive, and it requires both more specialized equipment and expertise compared with IHC.
Blotting techniques (Southern, Northern, and Western) can be used to measure HER-2 molecules, but they are technically difficult, require large amounts of fresh tissue, and are impractical for routine screening purposes. In addition, these techniques are not tumor cell-specific. Thus, the HER-2 status will be modified by the dilutional effect caused by the often large numbers of non-neoplastic cells (inflammatory, stromal, and normal) found in all tumors, resulting in an underestimation of gene amplification or expression. This problem can be reduced if analysates are enriched for target cells of interest, eg, by histological microdissection. Using laser-assisted microdissection (LAM), homogeneous tumor cell samples can be procured rapidly and precisely from both fresh/frozen or from archival FFPE tissues.10, 11, 12 This allows the detection of tumor-specific genetic alterations in otherwise heterogeneous samples, and enables a more correct detection and quantification of tumor DNA and mRNA.
Polymerase chain reaction (PCR)-based assays can determine changes in both HER-2 gene number and expression. By using real-time quantitative PCR, the target can be both amplified and detected simultaneously, making the assay a potentially sensitive, specific, reliable, and cost-effective way of measuring HER-2 changes. Routine methods of histological fixation and processing either destroy or damage RNA, or cause such extensive tissue cross-linking as to make it inaccessible to analysis.13 However, the high sensitivity of real-time quantitative PCR means that even minute amounts of DNA or RNA may be detectable in FFPE tissues14 , opening up the possibility of performing retrospective clinical and molecular studies on the large specimen archives stored in pathology institutes.
In this study, we focused on the feasibility of using real-time quantitative PCR analysis of DNA and mRNA from laser-microdissected FFPE archival breast cancer tissues to determine HER-2 status, and compared these results with those obtained using IHC and FISH. We show that HER-2 amplification and mRNA expression can be reliably detected by real-time quantitative PCR in archival microdissected FFPE tissues. In particular, our results suggest that PCR analysis of HER-2 DNA amplification in microdissected cells using the LightCycler HER2/neu DNA Quantification kit (Roche Molecular Biochemicals, Mannheim, Germany) may be used as an alternative to the traditional tumor cell-specific assays of HER-2 status.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunohistochemistry
We used the semi-quantitative immunocytochemical HercepTest (DakoCytomation, Copenhagen, Denmark) to assess HER-2 protein status, according to the manufacturer’s instructions. In brief, deparaffinized sections (4 µm) mounted on poly-L-lysine coated slides were heated in diluted Epitope Retrieval Solution (10 mmol/L citrate buffer) for 45 minutes using a water bath at 95°C, followed by cooling at room temperature for 20 minutes. After rinsing and blocking endogenous peroxidase, the sections were covered with primary polyclonal antibody (rabbit anti-human HER2 protein) in a humid chamber for 30 minutes. Slides were washed and then incubated in Visualization Reagent (dextran polymer conjugated with horseradish peroxidase and goat anti-rabbit immunoglobulins, EnVision, DakoCytomation) for 30 minutes at room temperature, and washed again before color development in 3,3'-diaminobenzidine substrate-chromogen (DAB) solution. The sections were counter-stained in Mayer’s hematoxylin. To validate the staining procedure, we simultaneously processed negative and positive controls provided with the kit consisting of sections of FFPE breast carcinoma cell lines representing various levels of Her-2 expression. The same technician performed the staining procedure for all cases. We scored tumor cells for Her-2 in accordance with the manufacturer’s instructions: no visible staining or membrane staining in less than 10% of cells, 0; faint, incomplete membrane staining detected in more than 10% of cells, 1+; weak to moderate complete membrane staining in more than 10% of cells, 2+; strong and complete membrane staining in more than 10% of malignant cells, 3+. Scoring was performed twice (blinded) by the same experienced pathologist.
Fluorescence In Situ Hybridization
Paraffin sections were hybridized using the Vysis PathVysion HER-2/neu DNA probe kit (Abbott Laboratories, Abbott Park, IL) in accordance with the manufacturer’s protocol. For each case, two sections (2 µm) were mounted on coated-glass slides (Superfrost Plus, Menzel, Germany). One section was stained with hematoxylin and eosin and was used as a microscopic control for locating dense tumor cell populations suitable for FISH analysis. The second section was used for FISH. The slide was deparaffinized, dehydrated in graded ethanols, denatured in 70% formamide, 2X standard saline citrate (SSC; 20X SSC = 3 mol/L NaCl2, 0.3 mol/L Na citrate, pH 5.3) at 70°C. Hybridization was carried out overnight in a humid chamber at 37°C in the presence of SpectrumOrange-labeled HER-2 DNA probe and SpectrumGreen-labeled centromere 17 reference probe. After washing in 2X SSC/0.3% NP-40, the nuclei were counter-stained with DAPI (0.2 µmol/L 4.6-diamino-2-phenyldole). Slides were examined using a fluorescence microscope at x100 magnification. At least 120 non-overlapping nuclei were counted in each patient, the counting area blindly selected by the same technician within the rather large marked area on the control H&E stain. These areas were carefully chosen to include large homogeneous tumor regions, with as few intervening inflammatory or stromal non-neoplastic cells as possible. In the tumors showing invasive carcinoma, areas with DCIS were excluded. The signals for Her-2/neu and centromere 17 were recorded and the ratio calculated for each slide by the same technician, without knowledge of the IHC HER-2 status of the patients. In accordance with the manufacturer’s protocol, we did not score nuclei with no signals, with signals of only one color, or with signals in overlapping areas. Nuclei with insufficient counter-stain to determine the nuclear border were not counted. We regarded a ratio 
2.2 as being truly amplified and a ratio <1.8 as not being amplified. Cases falling in the cut-off area between 1.8 to 2.2 were repeated.
Laser Microbeam Microdissection (LMM)
Sections were cut under RNase-free conditions and, mounted for laser-assisted microdissection using a modification of the MOMeNT technique.15
Microdissection of FFPE tissues was carried out using a high-resolution UV-laser system (µ-Cut, Molecular Machines & Industries, Glattbrug, Switzerland) as previously described.12
In brief, the sections were deparaffinized in xylene, rehydrated in graded ethanols, stained with hematoxylin, rinsed in tap water, and finally immersed in 99% ethanol. After drying the sections, the areas of interest were identified. To increase the quality of morphology, we used Cytotec (Schuco International, London, UK) as an optical medium.12
For each analysis, five homogeneous tumor areas were laser-cut (
300 cells each) and then captured using a sterile disposable needle. We dissected neighboring tumor areas to minimize the influence of intratumoral heterogeneity on the results. A naked area of the membrane was cut as a negative control. The tissue-fragments were put in an Eppendorf tube containing 200 µl RNA digestion buffer (20 mmol/L Tris, pH 7.5, 20 mmol/L EDTA, 1% SDS in sterile water) and 5 µl proteinase K (20 mg/ml, Roche Molecular Biochemicals), for RNA extraction, incubated overnight at 60°C, inactivated at 95°C for 10 minutes, and stored at -80°C.16
This procedure was repeated for DNA extraction using the same membrane-mounted slides. The cut fragments were transferred to an Eppendorf tube with 30 µl DNA extraction buffer containing 50 mmol/L Tris, pH. 8.1, 1 mmol/L EDTA, pH.8.0, and 0.5% Tween 20 and 0.24% proteinase K (20 mg/ml), incubated overnight at 55°C, and inactivated for 10 minutes at 95°C.17
The samples were kept at -20°C.
DNA Extraction, Total RNA Extraction, and Reverse Transcription
DNA extraction samples were used without any precipitation steps. Total RNA was extracted using the TRIzol-reagent according to the manufacturer (Life Technologies, Gaithersburg, MD), followed by precipitation with an equal volume of isopropanol, 0.1 volume 3 mol/L sodium acetate (pH 5.2) and 1 µl glycogen (20 mg/ml, Roche). The pellet was washed in 70% ethanol, dried and resuspended in 10 µl RNase-free water. Total RNA from the human bladder cell line HCV29 cell line18
was isolated with the Qiagen Total RNA Isolation kit (Qiagen Inc., Valencia, CA) and used for generating the calibration curve. Reverse transcription (RT) took place in a final reaction volume of 20 µl containing 50 units MuLV reverse transcriptase (Applied Biosystems, Foster City, CA), 10X PCR buffer, 5 pmol specific antisense primer (also used in PCR) for either Her-2/neu or ß-actin, 4 U RNase inhibitor (Roche), 4 mmol/L of each dNTPs, 6.25 mmol/L MgCl2, and 2 µl template. The RT step was performed on a Perkin Elmer 9700 Thermocycler at 42°C for 30 minutes followed by 94°C for 5 minutes.
Real-Time Quantitative RT-PCR Analysis of HER-2 Gene Expression
Analysis of HER-2/neu and of housekeeping gene ß-actin expression was carried out using the LightCycler system (Roche Molecular Biochemicals), following the manufacturer’s instructions. The theory and principles of real-time quantitative PCR are described in detail elsewhere.19
We used intron-spanning PCR amplicons and hydrolysis probes.16, 20
A standard curve for gene expression of Her-2 and ß-actin was constructed using total RNA from a human urothelial cell line HCV 29, as described in detail previously.14
In brief, 10-fold dilutions of RNA in water were made to give 10 different calibrators, each with a known amount of HCV 29 RNA. For both products a standard curve was generated and included in each PCR run. PCR amplification for analyzing gene expression was performed in a 20-µl reaction volume. A reaction mixture (Lightcycler DNA Master Hybridization Probes mixture, Roche Molecular Biochemicals) containing reaction buffer, nucleotides, and Taq polymerase was used for the PCR according to the supplier’s instructions. Probe, sense primer, antisense primer, 3.2 µl H2O and 8.0 µl cDNA were added to this reaction mixture (see Table 1
for sequence and concentration of primers and probes). Quantification was initiated by incubation for 1 minute at 94°C, followed by 40 cycles with the following profile: specific annealing temperature for 2 seconds, extension at 72°C for 10 seconds and denaturation at 94°C, immediately followed by the next amplification cycle. We calculated the relative quantification using the ratio between HER-2/neu and ß-actin mRNA, to eliminate the constant differences between genes and gene transcripts, that might be generated during the process (eg, differences due to variations in tissue preparation, template extraction, mRNA integrity, and PCR17
). To ensure that no DNA contamination had occurred, we included 20 samples without reverse transcriptase.
|
2) should be regarded as positive for HER-2/neu DNA amplification.
Statistical Analysis
Statistical analysis was performed using SPSS Version 11, setting the statistical significance at P < 0.05. The data obtained from real-time quantitative RT-PCR were analyzed with a linear mixed model for repeated measurements, where group is the fixed effect and patient is the random effect. This statistical model furthermore allows for analysis of total variation contributed by variations between the patients in the same group (var1) and the variation secondary to tumor heterogeneity and/or measurement inaccuracies (var2). The same statistical model was used to analyze data obtained with the DNA quantification kit. CV was calculated for inter-assay variation on the data obtained by the DNA quantification kit. Results were compared using scatter plots.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FISH Analysis of HER-2 Using PathVysion DNA Probe Kit
All cases were suitable for FISH analysis. Nine of 30 tumors (30%) were amplified for the HER-2/neu gene. Eight were carcinomas, all of which also showed strong expression by the HercepTest. One case (case no. 29) of DCIS was also amplified. The remaining 19 carcinomas and two DCIS cases were not amplified at the DNA level (Table 2)
. In the amplified cases, HER-2 signals were mostly bright and clustered, although two carcinomas showed innumerable signals. In some cases, signals were difficult to interpret, and hybridization had to be extended over 2 nights. Five of the 8 tumors, which were amplified in both IHC and FISH assays (62.5%), contained signals for more than two cep-17 regions in at least 25% of the tumor cells. In comparison, more than two signals for cep-17 were found in all 8 IHC 2+ cases, in 7 of 8 IHC 1+ cases, and in 2 of 3 IHC 0 cases. In the DCIS lesions, only the single FISH-amplified case, had more than two signals for cep-17 in some tumor nuclei. In general, neighboring tumor cells in most cases displayed distinct heterogeneity, with regard to both Her-2 and cep-17 signals.
|
HER-2 mRNA could be analyzed by real-time quantitative RT-PCR in all 30 cases (Table 2)
, although in one carcinoma (no. 3) there was insufficient cDNA available for subsequent analysis of ß-actin. In all cases, at least three of the five microdissected samples from each paraffin section gave satisfactory results for HER-2 mRNA; in the majority of cases, all five samples were informative. In total, 141 of 150 samples (94%) gave informative results.
Based on the IHC and histological findings, test cases were assigned to five groups: group A, 3+ amplification; group B, 2+ amplification; group C, 1+ amplification; group D, no amplification; group E, DCIS cases only. RT-PCR data were analyzed in relation to these groups (Table 2)
. We performed statistical analysis for repeated measurements on the ratio between HER-2 and ß-actin, as well as on the data obtained from HER-2 and ß-actin alone.
After normalization of the HER-2 results to the reference gene, we found a highly significant difference between group A and the remaining groups (P = 0.0001; Table 3
). The total variation was composed partly by the variation between patients in the same group (Table 3
; var1 = 73%) and partly by the variation that resulted from tumor heterogeneity and/or measurement inaccuracies (Table 3
; var2 = 27%). Similar results were found for HER-2 alone and after normalization to the number of cells in each experiment (n 24; Table 3
). The same test was performed on the data obtained from ß-actin alone, and showed no systematic difference between the groups (P = 0.530). In this case, the values for var1 and var2 were 81% and 19%, respectively. A similar result was found after normalization to the number of cells. Overall, the data obtained exhibited a distinct spread of values for the five samples per case in many of the patients (Table 2)
. Moreover, there was a distinct overlap between data in the different groups (Figure 1)
.
|
|
and Figure 2
).
|
Comparison of IHC, FISH, and Real-Time Quantitative PCR Assays of HER-2 mRNA Expression
Most cases showing amplification in FISH and strong protein expression in IHC (3+), had distinctly high expression of HER-2 mRNA measured by real-time quantitative RT-PCR (Figure 1)
. Cases with only moderate or lesser levels of HER-2 protein expression, showed mild to moderate expression of HER-2 mRNA. There was, however, a distinct overlap between these cases and the cases with strong protein expression (Figure 1)
.
Comparison of IHC, FISH, and LightCycler HER2/neu DNA Quantification Kit Assays of HER-2 Status
Cases amplified in FISH and with strong expression of HER-2 protein (IHC 3+) were all found to be amplified using the LightCycler kit (Tables 4
and 5
, Figure 2
). In addition, the single FISH-positive, IHC-negative DCIS case, was also found to be amplified by DNA-PCR (Tables 4
and 5)
. With the exception of this case, all other IHC 0 and 1+ tumors were not amplified by PCR. Five of the IHC 2+ cases were amplified in PCR, but were negative with FISH (Table 4
and Figure 2
). The degree of polysomy 17 found by FISH in these five cases was not significantly different to that found in the other IHC 2+ cases.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
IHC assays of HER-2 protein expression are currently the most widely used methods for assessing HER-2 status. The main advantages of these techniques, are that they are rapid and simple, they can be performed in most pathology laboratories, and that (when compared with other assays) they are relatively inexpensive. Furthermore, strong IHC-determined HER-2 protein expression is linked to clinical outcome and treatment response. However, IHC assay reliability has been questioned, because variations in tissue fixation, handling and processing, as well as differences in antibody clone used,21 objective scoring criteria, controls and inter-observer variability in interpretation can all influence the precision of the results. These causes of IHC assay variability help explain the widely differing rates of HER-2 expression in breast carcinomas reported previously.
The technical difficulties associated with IHC, have prompted the development of FISH assays for quantifying HER-2 gene amplification, particularly in cases found to be borderline by HER-2 IHC.22, 23 FISH has the advantage of being a quantitative method that correlates well with the results of IHC assays,9, 24 and it is generally regarded as the gold standard technique for evaluating HER-2 status. Major drawbacks are that FISH is technically difficult to carry out, laborious to set up, time consuming to perform, and costly compared to IHC. Moreover, this technique is currently only available in a minority of pathology laboratories.
A key advantage shared by IHC and FISH techniques for HER-2 evaluation is that they are based on microscopic analysis. In principle, this allows for changes in gene copy number and gene product expression to be assessed specifically in the tumor cell population. While this is an integral part of IHC assays such as the HercepTest, it is in practice only a theoretical advantage in most FISH analysis, since simultaneous morphological assessment of tissues is difficult in most routinely applied fluorescence-based in situ hybridization protocols. More precise morphological assessment is possible using chromogenic in situ hybridization (CISH).25, 26 This technique uses non-fluorescence-labeled probes in a paraffin section in situ hybridization protocol. This allows HER-2 status to be assessed in breast carcinomas on a cell-to-cell basis using a bright-field microscope. However, the sensitivity, precision, clinical correlates and usefulness of this technique remain to be established. A possible source of error with CISH, is that current protocols are based on a single-color detection of one probe only, masking possible HER-2 pseudo-amplification caused by chromosome 17 polysomy.
The inability to relate molecular changes in test tissues to specific histological components (eg, tumor cells) is a major disadvantage of blotting and PCR techniques based on analysis of extracted nucleic acids. Earlier published PCR-based methods to detect HER-2 amplification have been essentially semi-quantitative, based on fixed-cycle PCR with analysis by gel electrophoresis.27 However, the development of real-time PCR methods makes it possible to perform more precise quantitative analysis of gene amplification. These methods are easy and rapid to perform, can be used to test multiple samples, have a low risk of contamination (no post-amplification manipulation of amplicons), and can be automated, making them potentially useful methods for screening tumors for HER-2 amplification in a routine clinical setting. However, dilution of tumor genomic material with nucleic acids from non-neoplastic tissue components is an important potential source of imprecision in such quantitative assays in many cancers, not least in breast carcinoma in which non-malignant cells may account for more than 50% of the tumor.28 In the setting of HER-2 analysis, this will tend to increase the number of false-negative tumors, an effect that is undesirable and possible unacceptable in a screening procedure. In keeping with this, a recent study has shown, that low-level HER-2 gene amplification was no longer detectable by real-time PCR if the breast carcinoma cells compromised less than 30% of the tumor specimen.29
To reduce the problem of tumor tissue heterogeneity, we performed histological microdissection before PCR analysis. LAM has emerged recently as a key methodology for this purpose, allowing rapid and precise collection of homogeneous tumor cells from complex tissues suitable for a variety of downstream molecular applications.12
In this study, LAM proved to be a reliable technique that, once learned, was quick and as easy to perform as FISH. Tumor cell specimens obtained in this way gave satisfactory results with both real-time quantitative RT-PCR and PCR in almost all samples, even though the actual number of cells in individual analyses was very small (24 in RT-PCR and 50 in DNA-PCR). Formalin-fixation and paraffin-embedding is highly destructive of RNA, with the result that mRNA levels may be decreased by up to 99% in such tissues, with residual RNA being extensively degraded.13, 14
As a result, it has been considered essentially impossible to perform quantitative RT-PCR on archival paraffin block material. Despite this, several groups have recently published studies, in which the small amounts of degraded RNA in FFPE tissues can be successfully amplified and detected using highly sensitive real-time RT-PCR techniques.14, 16, 20, 29
Our study contributes further to these findings. We were able to show a statistically significant difference between the different tumor groups with regard to both the ratio between HER-2 and ß-actin mRNA, and HER-2 mRNA alone. In contrast, no significant differences were found between groups in levels of mRNA from the housekeeping gene ß-actin. We found considerable variation in HER-2 mRNA levels comparing different patients, a higher degree of variation than was due to differences in the HER-2 data, as compared with ß-actin. The rather large spread of mRNA levels within individual patients comparing the five parallel analyses, which contributed to the total variation, also calls for attention (Table 2)
. We believe this to be an expression of both tumor heterogeneity and of technical inaccuracies. Although we found statistically significant differences in HER-2 mRNA levels in our test tumors, there was a clear overlap of gene expression levels in the different groups. This suggests that HER-2 gene expression data from FFPE tissue studies should be used with caution, especially in the clinical setting. As discussed by Bustin,30
there are several other potentially important problems to consider with regard to the evaluation of real-time quantitative RT-PCR. The use of housekeeping genes is difficult in in vivo studies, as the mRNA level of these genes may vary considerable, and in LAM specimens it is not possible to quantitate total RNA. However, we found that ß-actin mRNA expression was similar in all groups, while there were statistically significant differences in HER-2 mRNA levels comparing different groups with regard to HER-2 normalized to ß-actin, HER-2 alone, and HER-2 normalized to cell number, suggesting that it was appropriate to normalize HER-2 levels to this housekeeping gene.
Compared to our real-time quantitative PCR analysis of HER-2 gene expression, evaluation of gene amplification using the LightCycler HER2/neu DNA Quantification kit (Roche Molecular Biochemicals) showed an even better agreement with both IHC and FISH data. Although DNA undergoes substantial changes during formalin-fixation, it is more stable than RNA. While DNA also undergoes degeneration in FFPE tissues, the short amplicon size used in the DNA quantification kit allows for reliable amplification of the fragmented DNA available in our microdissected samples. Compared to the data for HER-2 gene expression, we found less variation in levels of gene amplification in the different groups (Table 2)
. We believe that this is a result of DNA being relatively less degraded than RNA in FFPE material. The data also reveals a degree of tumor heterogeneity (Table 3
; 34% (var2-CV)) in spite our attempts to minimize this phenomenon by dissecting adjacent tumor areas. We suggest that in similar future studies, at least three repeated measurements should be conducted on tumor tissue from each patient. Furthermore, the total variance was contributed less by the differences between patients in the same group (Table 3
; var1 = 54%). The results from FISH analysis show that this should come as no surprise. Twenty-three of thirty cases (76.7%) showed polysomy of cep-17 and an even higher number (27 of 30) showed heterogeneity of HER-2 signals.
Using the LightCycler HER2/neu DNA Quantification kit, we found 14 tumors in which the mean HER-2 score was above the cut-off value of 2. Eight of these cases strongly overexpressed HER-2 protein by the HercepTest. Interestingly, five of the remaining six tumors scored +2 in the IHC assay. These cases were not FISH amplified. We have considered whether these may in fact be showing true amplification at the DNA level, although clearly the numbers included in our study are too small to allow any firm conclusions. Furthermore, it might be more appropriate for the cut-off value of the LightCycler kit (2 indicates Her-2 gene amplification) to be given as a range of values (analogous to the FISH assay), instead of a single value. Indeed, it should be stressed that the mean value for some of the cases came very close to the cut-off value. However, the precision of FISH analysis is also dependent on the degree of tumor heterogeneity. While we were able to use LAM to help enrich our PCR samples for tumor cells, it should be remembered that tissues examined by FISH may contain a variable number of non-malignant cells within the counting area. Using routine FISH protocols with DAPI counter-stain, fine morphological analysis is not possible, giving a potential source of error. Thus, although FISH is regarded as the gold standard to be used when assessing other methods for HER-2 analysis, it is not necessarily true that this technique will always give the most precise results. In light of this, our results raise the possibility that real-time PCR quantification of HER-2 gene amplification in microdissected breast carcinoma tissue may be as good, or even better than established techniques for determining HER-2 status.
While the advantages of IHC analysis make it likely that this technique will continue to be used as the most widespread method for primary screening of HER-2 status in breast carcinomas, microdissection with real-time PCR may be a useful supplement or even alternative to FISH for quantitative confirmation of HER-2 gene amplification. At first sight, the logistics of performing microdissection and real-time PCR might seem to be against their more widespread use. However, while the equipment and expertise for performing these techniques are not found in most routine laboratories, they are becoming increasingly available. Thus, they are already used in specialized laboratories, including many HER-2 reference pathology laboratories. FISH is also a methodology that requires special equipment and expertise that are not available in many routine laboratories. Furthermore, FISH is still quite a laborious and slow assay to use. In most routine protocols, FISH takes at least 2 days to perform. In our set up, procurement and preparation of microdissected specimens (including sectioning, deparaffinization, staining, microdissection, and protein-digestion) takes 1 to 2 days. If necessary, deparaffinization times can be further reduced. Thus, in a routine clinical setting we believe it would be possible to perform HER-2 analysis on microdissected samples within 1 to 2 days in most cases. While the microdissection is likely to be a rate-limiting step in high-throughput HER-2 testing, this is to a great extent counterbalanced by the suitability of PCR for batch analysis. For example, the LightCycler HER2/neu DNA Quantification kit analyzes 32 samples (ie, 10 tumors in triplicate) in a single run, thus making it possible to perform high-throughput analyses.
Clearly, before considering the use of microdissection with real-time quantitative PCR for HER-2 analysis, larger studies will be required to confirm our findings, and to correlate the results of the assay with clinical outcome of the patients. However, we believe that this technique may have a useful role to play in the assessment of HER-2 gene amplification in both research and clinical settings.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Supported by grants from the Clinical Research Unit of the Danish Cancer Society, the Fritz, Georg and Marie Cecile Glud’s Foundation, the Johan and Lise Boserup Foundation, the Hans and Nora Buchard’s Foundation, the Astrid Thaysen Foundation, the Lily Benthine Lund Foundation of 1/6–78, and the Danish Medical Research Council.
Accepted for publication November 17, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J Bergqvist, J. Ohd, J Smeds, S Klaar, J Isola, H Nordgren, G. Elmberger, H Hellborg, J Bjohle, A-L Borg, et al. Quantitative real-time PCR analysis and microarray-based RNA expression of HER2 in relation to outcome Ann. Onc., May 1, 2007; 18(5): 845 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Stevens, I. Almanaseer, M. Gonzalez, D. Caglar, R. A. Knudson, R. P. Ketterling, D. S. Schrock, T. A. Seemayer, and J. A. Bridge Analysis of HER2 Gene Amplification Using an Automated Fluorescence in Situ Hybridization Signal Enumeration System J. Mol. Diagn., April 1, 2007; 9(2): 144 - 150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Decock How Accurate Is the Antiprimer Quenching-Based Real-Time PCR for Detection of Her2/neu in Clinical Cancer Samples? Clin. Chem., July 1, 2006; 52(7): 1438 - 1439. [Full Text] [PDF]
|
|
|
|
|
|
C. Delfour, P. Roger, C. Bret, M.-L. Berthe, P. Rochaix, N. Kalfa, P. Raynaud, F. Bibeau, T. Maudelonde, and N. Boulle RCL2, a New Fixative, Preserves Morphology and Nucleic Acid Integrity in Paraffin-Embedded Breast Carcinoma and Microdissected Breast Tumor Cells J. Mol. Diagn., May 1, 2006; 8(2): 157 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Buhl, J. Jurlander, F. S. Jorgensen, A. M. Ottesen, J. B. Cowland, L. M. Gjerdrum, B. V. Hansen, and H. Leffers Identification of a gene on chromosome 12q22 uniquely overexpressed in chronic lymphocytic leukemia Blood, April 1, 2006; 107(7): 2904 - 2911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Hood, M. M. Darfler, T. G. Guiel, B. Furusato, D. A. Lucas, B. R. Ringeisen, I. A. Sesterhenn, T. P. Conrads, T. D. Veenstra, and D. B. Krizman Proteomic Analysis of Formalin-fixed Prostate Cancer Tissue Mol. Cell. Proteomics, November 1, 2005; 4(11): 1741 - 1753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Jiang, N. P. Caraway, B. Nebiyou Bekele, H.-Z. Zhang, A. Khanna, H. Wang, R. Li, R. L. Fernandez, T. M. Zaidi, D. A. Johnston, et al. Surfactant Protein A Gene Deletion and Prognostics for Patients with Stage I Non-Small Cell Lung Cancer Clin. Cancer Res., August 1, 2005; 11(15): 5417 - 5424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Tse, D. Brault, J. Gligorov, M. Antoine, R. Neumann, J.-P. Lotz, and J. Capeau Evaluation of the Quantitative Analytical Methods Real-Time PCR for HER-2 Gene Quantification and ELISA of Serum HER-2 Protein and Comparison with Fluorescence in Situ Hybridization and Immunohistochemistry for Determining HER-2 Status in Breast Cancer Patients Clin. Chem., July 1, 2005; 51(7): 1093 - 1101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. P. Brandt Microgenomics: gene expression analysis at the tissue-specific and single-cell levels J. Exp. Bot., February 1, 2005; 56(412): 495 - 505. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
