Abstract
Purpose
The evaluation of ERBB2 gene copy numbers and ERBB-2 protein expression in invasive duct carcinomas of the mammary gland (IDC) has been introduced as a part of a regular investigation protocol. Amplification of the ERBB2 gene detected by fluorescence in situ hybridization on interphasic nuclei (I-FISH) is used as a criterion for Herceptin administration. To improve characterization of cases with a borderline ERBB2 gene amplification, we applied a modified evaluation of the ERBB2 gene copy numbers. The results were compared with a commonly used HER-2/CEP17 ratio calculation.
Methods
We investigated 175 patients with primary IDCs in histological sections from paraffin embedded tissue with PathVysion HER-2 DNA Probe Kit (Vysis). Tumor cells of each case were sorted according to the number of ERBB2 signals into groups of tumor cells without amplification, with moderate amplification (≤10 signals) and with strong amplification (>10 signals). If >10% of tumor cells had moderate or strong amplification of the ERBB2 gene, the case was reported as “moderately” or “strongly” amplified.
Results
The groups of patients classified by the proposed system as “without” and as with “strong” amplification had the HER-2/CEP17 ratio <1.91 (median 1.22, n = 33) and >2.3 (median 6.85, n = 115), respectively. Thus, the findings using the two systems of evaluation yielded similar results for these groups of patients. In cases, classified as with “moderate” amplification, the HER-2/CEP17 ratio varied from 1.3 to 4.77 (median 2.11, n = 27). Twelve of these 27 patients were according to the HER-2/CEP17 ratio system “not amplified” (1.3–1.93, median 1.72). Median percentage of the tumor cells with amplification of the ERBB2 gene was 14.5% in this subgroup. Of these 12 cases, 10 were ERBB-2 protein positive, and would be candidates for Herceptin therapy.
Conclusion
The criteria for evaluation of the ERBB2 gene copy number proposed for this study separate gray zone cases with a borderline HER-2/CEP17 finding. The proposed system is easy to apply and characterizes a heterogeneity of the ERBB2 gene copy number in IDC tumor cell population more precisely than the currently used HER-2/CEP17 ratio system.
Keywords: Breast carcinoma, HER-2, ERBB2 gene amplification, I-FISH evaluation
Introduction
Based on evaluation of the ERBB-2 protein expression by immunohistochemical techniques (IHC) and counting of the ERBB2 gene copy numbers using fluorescence in situ hybridization on interphasic nuclei (I-FISH) therapy with a monoclonal antibody against the ERBB-2 protein (Herceptin–trastuzumab, Genentech) is administered to patients with metastatic invasive duct carcinoma of the mammary gland (IDC) (full prescribing information for Herceptin–Trastuzumab, 2003). According to the HUGO Gene Nomenclature Committee (http://www.gene.ucl.ac.uk/nomenclature/) the gene is named ERBB2, but many still use synonyms such as c-erbB-2, HER-2, NEU.
I-FISH is helpful in identification of genetic changes in relation to the tissue and/or cell morphology, and because of that it is routinely used on histological sections from formalin fixed/paraffin embedded material. Cutting off the nuclei due to preparation of histological slides leads to enumeration of the signals in a reduced volume of the tumor cell nuclei, but it gives an advantage of preserved tissue morphology and a possibility to investigate archive materials.
A common principle of reporting I-FISH results in evaluation of the ERBB2 gene copy numbers is based on establishing “HER-2/CEP17 ratio”. A total number of the ERBB2 gene signals is divided by a total number of chromosome 17 centromeric signals (CEP17) in all evaluated tumor cell nuclei (FDA approved PathVysion HER-2 DNA Probe Kit—Vysis/Abbott or HER2 FISH pharmDx Kit—DakoCytomation). Based on the results obtained using HER-2/CEP17 ratio system, a presence or an absence of the ERBB2 gene amplification is reported. However, such an approach may obscure some findings in heterogeneous tumor cell populations. For example, a minor tumor cell clone with the ERBB2 gene amplification may not be reflected in the total count, and calculating the HER-2/CEP17 ratio may hide numerical changes of chromosome 17 [the number of CEP17 signals vary from case to case and also within individual cells of the same tumor (Adeyinka et al. 1999; Marinho et al. 2000)].
Some investigators suggested that a more detailed characterization of I-FISH results may be useful (Fernandez et al. 1996; Klijanienko et al. 1999). To our knowledge, there is no study, which would compare more complex evaluation of I-FISH results with the system based on the HER-2/CEP17 ratio. The aim of the present report was to apply a modified evaluation of I-FISH results, which would characterize the heterogeneity of the tumor cell populations in a greater detail, and to compare the proposed system with the evaluation using the HER-2/CEP17 ratio. In 175 patients with invasive duct carcinoma of the mammary gland histological sections were used for evaluation of the ERBB2 gene amplification focusing mainly on ERBB-2 protein positive cases.
Materials and methods
A group of 175 patients with primary IDC (in a majority of cases ERBB-2 protein positive) was investigated by I-FISH in formalin fixed/paraffin embedded tissue.
Formalin fixed material (buffered formalin, 18–24 h) was processed by a routine histological technique, cut and stained with hematoxylin and eosin to identify the most appropriate sample for IHC and I-FISH on histological sections.
Immunohistochemistry (IHC)
First of all, expression of the ERBB-2 protein was evaluated on formalin fixed/paraffin embedded tissue sections using HercepTest (DakoCytomation) and scored according to a protocol recommended by the producer. On the basis of HercepTest results, a group of 175 cases of IDC was further analyzed. Scoring was performed by two observers independently. Discordant results were discussed and a final scoring was made on a consensual basis.
Fluorescence in situ hybridization on interphasic nuclei (I-FISH)
A directly labeled mixture of LSI ERBB2 (Spectrum Orange) and CEP17 (Spectrum Green) probes (PathVysion HER-2 DNA Probe Kit, Vysis/Abbott) was used to evaluate a copy number of the ERBB2 gene and a number of signals identifying centromeres of chromosome 17. The histological sections were processed according to the protocol distributed with the PathVysion kit.
I-FISH signals were evaluated using fluorescence microscope Provis AX70 with dual and triple bandpass filters (Olympus). For a precise analysis of different layers in a histological section Axioplan 2 microscope equipped with ApoTome slider module (Carl Zeiss) was used. Image analysis was performed with AxioVision 4.2 (Carl Zeiss). The number of signals of ERBB2 gene and CEP17 was counted in at least one hundred adequately preserved non-overlapping tumor cell nuclei. Considering the tumor tissue heterogeneity, all fields of each sample were reviewed.
Proposed criteria for this study and commonly used system of I-FISH evaluation
Proposed criteria for evaluation of copy numbers of ERBB2 gene: individual tumor cells in each investigated case were sorted into three groups according to the I-FISH signal pattern. If a tumor cell has two signals of CEP17 and ≤4 signals of the ERBB2 gene the cell fits into a subgroup without amplification. This approach considers formation of sister chromatids in the cell cycle during the segregation of the chromosomes in the G2 or M phase of the cell cycle (4 signals of the ERBB2 and 2 signals of CEP17) (Lebeau et al. 2001; Hicks and Tubbs 2005). Likewise, if a tumor cell has three signals of CEP17, the cell is placed into the subgroup without amplification if the number of ERBB2 signals is ≤6. If the number of ERBB2 signals (in presence of 2 CEP17 signals) is >4 and ≤10, the cell is categorized as with a moderate amplification. If the number of ERBB2 signals is >10 such a cell is placed in a subgroup with a strong amplification. In cases with an increased number of CEP17 signals, the number of chromosomes was considered in respect of formation of sister chromatids during the cell cycle. Exact ranking of cells with increased number of CEP17 signals is showed in Fig. 1. Final evaluation of the I-FISH result in each patient is reported as a case with a “moderate amplification” or as a case with a “strong amplification” if the count of tumor cells with a moderate or with a strong amplification is observed in >10% of the tumor cell population. The principle of a 10% result of I-FISH was used also by other authors (Ginestier et al. 2004; Pellikainen et al. 2004). A similar system is recommended for evaluation of IHC results of the ERBB-2 protein positivity using the FDA approved HercepTest. To document the simplicity of the system Fig. 1 shows a form used for writing down the I-FISH results and their final evaluation.
Fig. 1.
A form for I-FISH evaluation using criteria proposed in this study. In the left part of the form, the number of CEP17 and ERBB2 signals in each tumor cell is written during the evaluation of the I-FISH result. In the upper right part of the form, each tumor cell is categorized according to its signal pattern into three groups: without amplification of the ERBB2 gene (light gray), with moderate amplification of the ERBB2 gene (mid-tone gray) and with strong amplification of the ERBB2 gene (dark gray). In the column “other findings” the tumor cells with >4 signals of CEP17 are written down. A total number of tumor cells in each group is counted and also a total count of tumor cells with >2 signals of CEP17 (polysomy of chromosome 17) is calculated. The case is evaluated as “without amplification” of the ERBB2 gene if <10% of tumor cells had moderate amplification and <10% of tumor cells had a strong ERBB2 gene amplification. The case is evaluated as “moderately amplified” if >10% of tumor cells had moderate amplification and <10% of tumor cells had strong amplification. The case is evaluated as “strongly amplified” if >10% of tumor cells had strong amplification of the ERBB2 gene
Commonly used criteria for evaluation of the ERBB2 gene amplification in patients with IDC are based on calculation of HER-2/CEP17 ratio. A total count of ERBB2 gene signals in all evaluated tumor cell nuclei is divided by a total count of CEP17 signals. The number of evaluated tumor cell nuclei recommended by the PathVysion protocol is 20. If the HER-2/CEP17 ratio is <2, the case is reported as “not amplified”. If the HER-2/CEP17 ratio is ≥2, the case is referred to as “amplified”. Thus, according to the PathVysion system, designed for formalin fixed and paraffin embedded tissue sections, a cutoff point for ERBB2 gene amplification is assigned as the HER-2/CEP17 ratio 2.0. If the ratio reaches a borderline value (1.8–2.2), evaluation of further 20 tumor cells is recommended.
Results
Criteria for I-FISH evaluation proposed for the study
We classified 33 patients as having IDC “without amplification” of the ERBB2 gene (Group A). Twenty-seven patients were classified as having IDC with “moderate amplification” (Group B), and 115 patients had “strong amplification” of the ERBB2 gene (Group C). The proportion of individual groups of patients does not reflect an expected rate of findings in patients with IDC in a random investigation of surgical samples because the I-FISH was performed preferentially on cases with positive HercepTest (2+ and 3+). The distribution of patients in the three groups and the percentage of cells with moderate and strong amplification of the ERBB2 gene is specified in Table 1. The tumor cells with strong amplification of the ERBB2 gene had the ERBB2 signals organized frequently in clusters. Figure 2 shows a sequence of pictures of a tumor cell nucleus with a strong ERBB2 gene amplification investigated in a histological section taken at 11 different levels of focusing, revealing distribution of the ERBB2 and CEP17 signals at different layers.
Table 1.
Characterization of I-FISH signal pattern in 175 patients with IDC evaluated according to the criteria proposed for the study
| Group Aa (33 cases) | Group Bb (27 cases) | Group Cc (115 cases) | ||||||
|---|---|---|---|---|---|---|---|---|
| Median | Mean | Range | Median | Mean | Range | Median | Mean | Range |
| Tumor cells without amplification of the ERBB2 gene (%) | ||||||||
| 98 | 97.1 | 91–100 | 64 | 59.3 | 15–90 | 2 | 4.3 | 0–32 |
| Tumor cells with moderate amplification of the ERBB2 gene (%) | ||||||||
| 2 | 2.9 | 0–9 | 36 | 39.1 | 10–79 | 17 | 25.9 | 0–89 |
| Tumor cells with strong amplification of the ERBB2 gene(%) | ||||||||
| 0 | 0.1 | 0–2 | 0 | 1.7 | 0–9 | 80 | 69.8 | 11–100 |
aGroup of patients “without amplification” of the ERBB2 gene
bGroup of patients with “moderate amplification” of the ERBB2 gene
cGroup of patients with “strong amplification” of the ERBB2 gene
Fig. 2.
A tumor cell nucleus with strong amplification of the ERBB2 gene. Result of fluorescence in situ hybridization on interphasic nuclei (I-FISH) with PathVysion HER-2 DNA Probe Kit (Vysis/Abbott) on histological section from formalin fixed/paraffin embedded tissue (signals of the ERBB2 gene—Spectrum Orange, signals of centromeres of chromosome 17—Spectrum Green). Pictures 1 to 11 represent a sequence of one tumor cell nucleus at 11 different levels of focusing using Axioplan 2 microscope and ApoTome slider module (Carl Zeiss). The distribution of signals through all layers of the tumor cell nucleus is apparent. Picture 1-11 shows digitally superimposed layers (>20 signals of the ERBB2 gene and 4 signals of chromosome 17 centromeres)
Commonly used criteria for I-FISH evaluation based on establishing HER-2/CEP17 ratio
According to the evaluation based on HER-2/CEP17 ratio, 45 patients with IDC were not amplified (HER-2/CEP17 ratio <2; range 0.84–1.93; median 1.31; mean 1.35) and 130 patients had amplification of the ERBB2 gene (HER-2/CEP17 ratio ≥2; range 2–15.52; median 6.05; mean 6.63). Despite the fact that we evaluated 100 tumor cells in each case (compared to 20 tumor cells, and in borderline cases additional 20 cells, recommended by the PathVysion protocol), six of the 175 patients had borderline findings (HER-2/CEP17 ratio 1.8–2.2) and categorizing these patients as “amplified” or “not amplified” was difficult using the HER-2/CEP17 ratio evaluation system (according to the criteria proposed in this study, five of these six patients were classified as having “moderate amplification” of the ERBB2 gene and one patient was “without amplification” of the ERBB2 gene).
Number of CEP17 signals
Table 2 summarizes the distribution of the number of CEP17 signals in groups of patients categorized according to the proposed criteria (Groups A, B and C), and commonly used criteria (Groups <2 and ≥2). Differences in median of percentage of cells with a given number of CEP17 signals were not statistically significant in Groups A, B and C (one-way Anova). Impact of an increased number of CEP17 signals (>2 signals per tumor cell nucleus) on the resulting HER-2/CEP17 ratio is documented in Table 3 in which the group of patients with “strong amplification” of the ERBB2 gene is shown (Group C). In patients with “strong amplification” of the ERBB2 gene and with >50% of tumor cells with polysomy of CEP17 the HER-2/CEP17 ratio is about twofold lower than in patients with “strong amplification” of the ERBB2 gene and with <10% of tumor cells with polysomy of CEP17. The percentage of tumor cells with strong amplification of the ERBB2 gene, which might influence the HER-2/CEP17 ratio, was comparable in both groups (>50 and <10%).
Table 2.
Number of CEP17 signals in groups of patients with IDC distributed according to proposed criteria and commonly used criteria
| Proposed evaluation | Commonly used criteria | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Group Aa (33 cases) | Group Bb (27 cases) | Group Cc (115 cases) | Group <2d (45 cases) | Group ≥2e (130 cases) | ||||||||||
| Median | Mean | Range | Median | Mean | Range | Median | Mean | Range | Median | Mean | Range | Median | Mean | Range |
| 1 signal CEP17 | ||||||||||||||
| 27 | 31.3 | 0–97 | 30 | 31 | 3–74 | 28 | 28.8 | 0–95 | 26 | 29.1 | 0–97 | 29.5 | 29.8 | 0–95 |
| 2 signals CEP17 | ||||||||||||||
| 50 | 51.1 | 3–87 | 44 | 47.9 | 24–90 | 44 | 42.9 | 5–100 | 49 | 50.2 | 3–87 | 44 | 43.5 | 5–100 |
| 3 signals CEP17 | ||||||||||||||
| 9 | 12.6 | 0–36 | 12 | 14.9 | 0–40 | 14 | 16.9 | 0–58 | 10 | 14.9 | 0–40 | 14 | 16.1 | 0–58 |
| 4 signals CEP17 | ||||||||||||||
| 1 | 4 | 0–28 | 6 | 5.6 | 0–20 | 3 | 8.1 | 0–43 | 2 | 4.9 | 0–28 | 3 | 7.7 | 0–43 |
| >4 signals CEP17 | ||||||||||||||
| 0 | 1.1 | 0–8 | 0 | 0.6 | 0–4 | 0 | 2.7 | 0–38 | 0 | 1 | 0–8 | 0 | 2.4 | 0–38 |
aGroup of patients “without amplification” of the ERBB2 gene
bGroup of patients with “moderate amplification” of the ERBB2 gene
cGroup of patients with “strong amplification” of the ERBB2 gene
dPatients with HER-2/CEP17 ratio <2
ePatients with HER-2/CEP17 ratio ≥2
Table 3.
Effect of polysomy of chromosome 17 on the HER-2/CEP17 ratio in 115 patients with IDC with “strong amplification” of the ERBB2 gene (Group C)
| Percentage of tumor cells with >2 signals of CEP17 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| <10% (37 cases) | 10–25% (28 cases) | 26–50% (28 cases) | >50% (22 cases) | ||||||||
| Median | Mean | Range | Median | Mean | Range | Median | Mean | Range | Median | Mean | Range |
| The HER-2/CEP17 ratio | |||||||||||
| 8.05 | 8.47 | 3.18–15.52 | 7.48 | 7.32 | 3.63–11.63 | 6.22 | 6.82 | 2.93–15.32 | 4.59 | 4.91 | 2.35–9.62 |
| % of tumor cells without amplification of the ERBB2 gene | |||||||||||
| 0 | 3.2 | 0–32 | 2 | 5.4 | 0–26 | 1 | 4.8 | 0–32 | 2.5 | 4.1 | 0–24 |
| % of tumor cells with moderate amplification of the ERBB2 gene | |||||||||||
| 20 | 30.8 | 0–89 | 23.5 | 29 | 2–69 | 12 | 20.1 | 0–75 | 16 | 21 | 0–64 |
| % of tumor cells with strong amplification of the ERBB2 gene | |||||||||||
| 76 | 66 | 11–100 | 71 | 65.6 | 11–98 | 87 | 75.1 | 16–100 | 81.5 | 74.9 | 20–100 |
A comparison of evaluation criteria for I-FISH proposed in this study and criteria using HER-2/CEP17 ratio
In a group of 45 patients with HER-2/CEP17 ratio <2 (“not amplified”) there were, according to the proposed criteria, 33 cases “without amplification” of the ERBB2 gene and 12 IDCs were classified as with “moderate amplification” of the ERBB2 gene (specification bellow). In a group of 130 patients with HER-2/CEP17 ratio ≥2 (“amplified”) there were 115 cases classified according to the proposed criteria as with “strong amplification” of the ERBB2 gene and 15 IDCs as with “moderate amplification” of the ERBB2 gene. Four of these 15 cases had >70% of tumor cells with moderate amplification of the ERBB2 gene and <10% of tumor cells with strong amplification of the ERBB2 gene. All four cases had <10% of tumor cells with >2 CEP17 signals, and HER-2/CEP17 ratio ranged from 3.42 to 4.77. Three of them expressed the ERBB-2 protein 3+, one 2+. These four cases were, according to the proposed criteria, classified as with “moderate amplification” but verge on the group with “strong amplification” of the ERBB2 gene.
Table 4 compares both approaches for I-FISH evaluation. In patients classified according to the criteria outlined for this study as cases “without amplification” of the ERBB2 gene (Group A), the HER-2/CEP17 ratio was <1.6 in all patients but one. The latter patient reached a borderline ratio 1.9. She had 67% of the tumor cells with one signal of CEP17 and with two signals of the ERBB2 gene resulting in an increased HER-2/CEP17 ratio.
Table 4.
Comparison of results in 175 IDCs evaluated with proposed and commonly used evaluation criteria for I-FISH
| Proposed evaluation | Commonly used criteria | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Group Aa (33 cases) | Group Bb (27 cases) | Group Cc (115 cases) | Group <2d (45 cases) | Group ≥2e (130 cases) | ||||||||||
| Median | Mean | Range | Median | Mean | Range | Median | Mean | Range | Median | Mean | Range | Median | Mean | Range |
| HER-2/CEP17 ratio | ||||||||||||||
| 1.22 | 1.23 | 0.84–1.9 | 2.11 | 2.39 | 1.3–4.77 | 6.85 | 7.11 | 2.35–15.52 | 1.31 | 1.35 | 0.84–1.93 | 6.05 | 6.63 | 2–15.52 |
| IHC negative, score 0; number of cases | ||||||||||||||
| 0 | 1 | 0 | 0 | 1 | ||||||||||
| IHC negative, score 1+; number of cases | ||||||||||||||
| 5 | 3 | 0 | 7 | 1 | ||||||||||
| IHC positive, score 2+; number of cases | ||||||||||||||
| 28 | 16 | 12 | 36 | 20 | ||||||||||
| IHC positive, score 3+; number of cases | ||||||||||||||
| 0 | 7 | 103 | 2 | 108 | ||||||||||
aGroup of patients “without amplification” of the ERBB2 gene
bGroup of patients with “moderate amplification” of the ERBB2 gene
cGroup of patients with “strong amplification” of the ERBB2 gene
dPatients with HER-2/CEP17 ratio <2
ePatients with HER-2/CEP17 ratio ≥2
In a group of patients with “moderate amplification” of the ERBB2 gene (Group B) the HER-2/CEP17 ratio varied from 1.3 to 4.77. In the group of patients evaluated as with “strong amplification” of the ERBB2 gene (Group C), the HER-2/CEP17 ratio was >2.3. In a majority of cases with “strong amplification”, the tumor cells had more than 20 signals of ERBB2 gene per a nucleus.
We found a good correlation between score “3+” IHC positivity of the ERBB-2 protein and a copy number of the ERBB2 gene evaluated as amplified (moderately or strong) by criteria proposed in this study (Table 4). None of the patients categorized with the proposed criteria as being “without amplification” of the ERBB2 gene had the ERBB-2 protein expressed 3+. Similarly, a good correlation was found between IHC negative cases (score “0” and “1+”) and an absence of the strong amplification of the ERBB2 gene.
According to criteria based on HER-2/CEP17 ratio system two patients with IDCs with 3+ expression of the ERBB-2 protein had HER-2/CEP17 ratio 1.41 and 1.84, and were classified as “not amplified”. According to proposed criteria they were “moderately amplified”. Similarly, according to criteria based on the HER-2/CEP17 ratio two patients with IDCs ERBB-2 protein negative had HER-2/CEP17 ratio 2.11 and 2.35 and were classified as “amplified”. According to the proposed criteria they were both “moderately amplified”.
Cases discordantly classified according to the proposed criteria and commonly used criteria
In a group of 175 IDCs, 12 patients (6.9%) with the HER-2/CEP17 ratio <2 will be, according to the proposed evaluation, classified as with “moderate amplification” of the ERBB2 gene (Table 5). In nine of them the number of tumor cells with moderate amplification ranged from 10 to 16%, in three cases from 22 to 36%. Median percentage of tumor cells with a moderate amplification in these 12 patients was 14.5. When the status of the ERBB2 gene was compared with the ERBB-2 protein expression, 10 of these 12 patients were IHC positive (eight cases 2+, two cases 3+).
Table 5.
Discordances in the proposed criteria and commonly used criteria of evaluation. Twelve cases with the HER-2/CEP17 ratio <2 (“not amplified”) are classified according to the proposed system as with “moderate amplification” of the ERBB2 gene
| Median | Mean | Range |
|---|---|---|
| Tumor cells without amplification of the ERBB2 gene (%) | ||
| 85.5 | 82.7 | 64–90 |
| Tumor cells with moderate amplification of the ERBB2 gene (%) | ||
| 14.5 | 17.2 | 10–36 |
| Tumor cells with strong amplification of the ERBB2 gene (%) | ||
| 0 | 0.2 | 0–2 |
| The HER-2/CEP17 ratio | ||
| 1.72 | 1.67 | 1.3–1.93 |
| IHC ERBB2 positive: 10 of 12 (83%) | ||
These cases represent IDCs with moderate amplification of the ERBB2 gene present as a minor clone of the tumor cell population. A low percentage of the tumor cells with moderate amplification of the ERBB2 gene remain hidden by using HER-2/CEP17 ratio system, which resulted in classification of these cases as “not amplified”.
Our results point out that it is reasonable to separate patients with an increased number of the ERBB2 gene copies up to 10 copies as a separate subgroup of “moderately amplified cases”. Establishing a group of patients with “moderate amplification” of the ERBB2 gene adds more objectivity to the I-FISH evaluation, and it may better explain occasional discrepancies between the evaluation of IHC results and the “ratio system” of reporting the I-FISH findings.
Discussion
Despite the fact, that the significance of ERBB2 in patients with carcinoma of the mammary gland was extensively reviewed in the literature, problems with evaluation of “borderline” cases and “discrepancies” between IHC and I-FISH results still persist, and reviews of the ERBB2 evaluation are a subject of recent studies (Hicks and Tubbs 2005).
For detecting copy numbers of the ERBB2 gene in histological sections, I-FISH has become widely used, supporting the interpretation of the ERBB-2 protein status and helping to plan the treatment strategy with Herceptin in patients with IDC. An advantage of I-FISH rests in a quantitative assessment of the results in contrast to a subjective evaluation of the intensity of the IHC precipitate on the tumor cell membranes using immunohistochemistry (Hoang et al. 2000; Hicks and Tubbs 2005).
When performing a routine I-FISH examination on IDCs we gradually realized that the commonly used criteria for I-FISH evaluation based on the HER-2/CEP17 ratio may hide important findings, such as a minor tumor cell population with the ERBB2 gene amplification. As a consequence, such carcinomas may be classified as not amplified by a simple calculation of the HER-2/CEP17 ratio. In investigated group of 175 patients with IDCs, 12 cases (6.9%) had HER-2/CEP17 ratio <2 (classified as “not amplified”) but according to the criteria proposed for the current study these patients were evaluated as with “moderate amplification” of the ERBB2 gene. To increase the objectivity of mathematical evaluation we made an attempt to modify the evaluation of the ERBB2 gene copy numbers. In each case the tumor cells were stratified into three categories—without amplification, with moderate amplification and with strong amplification of the ERBB2 gene, according to the ERBB2 and CEP17 signal pattern. A similar approach was used also by others, reflecting a different level of the gene copy number in individual cases (Fernandez et al. 1996; Klijanienko et al. 1999). If more than 10% of tumor cells had >10 signals of the ERBB2 gene, the result was interpreted as a case with “strong amplification”, in a similar approach as used in some other studies (Pauletti et al. 1996; Lebeau et al. 2001).
All 33 patients classified as “without amplification” of the ERBB2 gene had the HER-2/CEP17 ratio <2, and all 115 cases classified as “strongly amplified” had the ratio >2. Of 27 women with IDC classified as “moderately amplified”, 12 patients had the HER-2/CEP17 ratio <2, and would be reported as not amplified. The percentage of the tumor cells with a moderate ERBB2 gene amplification varied in these cases from more than 10% up to 36%. In eight of these patients the expression of the ERBB-2 protein was scored “2+” and two patients had the expression “3+”. As it is known, patients scored “2+” represent a relatively heterogeneous subset necessitating a modification of scoring criteria for the immunohistochemically revealed expression of the ERBB-2 protein. Using a combination of primary antibodies may help to predict the ERBB2 gene status more precisely in such cases (Ginestier et al. 2004).
Patients with amplification of the ERBB2 gene in a proportion of the tumor cells (at the initial examination) might become eligible for Herceptin therapy during the disease progression later if clones with amplification prevail or migrate to form metastatic loci. Other authors also recommend to describe findings in tumors with a heterogeneous cell population with a caution and to report the maximal ERBB2 value with a note, that an investigated tumor demonstrates a heterogeneity of the ERBB2 amplification (Hicks and Tubbs 2005). Intratumoral cytogenetic heterogeneity in breast carcinomas is a known fact confirmed by an analysis using laser-microdissection of the tumor cell clusters (Aubele et al. 1999).
The other 15 patients of our series classified as “moderately amplified” had the HER-2/CEP17 ratio ≥2, and would be classified as amplified according to the HER-2/CEP17 ratio system of evaluation. If histological sections are evaluated, one must consider losses of signals caused by the tissue sectioning. The situation is frequently complicated by a polysomy of chromosome 17. Therefore, an assessment of chromosome 17 copy number is important to distinguish a moderate ERBB2 gene amplification from an increase in chromosome 17 numbers (Pauletti et al. 1996). Moreover, marked polysomies of chromosome 17 without the ERBB2 gene amplification may lead to an overexpression of the ERBB-2 protein due to an increased dose of ERBB2 alleles (Tsuda et al. 2001). The importance of including chromosome 17 count into evaluating the ERBB2 gene amplification in each individual tumor cell is further highlighted by the fact that trisomy/polysomy of chromosome 17 is a frequent finding in patients with IDC (Mendelin et al. 1999).
There are some cases, in which it is difficult to decide, whether there is “moderate amplification” or “strong amplification” of the ERBB2 gene. In a group of 27 patients classified as with “moderate amplification” of the ERBB2 gene, four patients had >70% of the tumor cells with moderate amplification and <10% of the tumor cells with strong amplification. In this type of finding, the correlation with a standardized immunohistochemistry is especially crucial. In a group of patients with “moderate amplification” of the ERBB2 gene 59% of cases had the ERBB-2 protein evaluated as score “2+” and 26% was scored “3+”. There is no doubt that in a majority of cases with the HER-2/CEP17 ratio >2 the amplification of the ERBB2 gene is present. However, the evaluation system based on the HER-2/CEP17 ratio may hide a minor tumor cell clone with the ERBB2 gene amplification (in our series 12 cases of 175 IDCs), or the HER-2/CEP17 ratio may be diminished by the presence of tumor cells with polysomies of chromosome 17. In general, calculating an overall HER-2/CEP17 ratio hides differences in a heterogeneous tumor cell population, which may have therapeutic consequences in patients with borderline findings.
We believe that a contribution of using criteria proposed in this study rests in the following points. Each tumor cell is evaluated individually—the number of ERBB2 gene signals and the number of CEP17 signals are calculated for each evaluated cell. One hundred of tumor cells is evaluated, which contributes to a higher degree of certainty in evaluation of heterogeneous tumor cell populations. After evaluation of one hundred tumor cells, tumor cell by tumor cell is classified according to a number of ERBB2 and CEP17 signals into three categories: tumor cells with strong amplification of the ERBB2 gene, tumor cells with moderate amplification, and tumor cells without amplification of the gene. Such a stratification describes the heterogeneity of the tumor cell population in each individual patients and adds more objectivity into the evaluation system.
The evaluation criteria proposed in this study are easy to apply, and they characterize the tumor cell populations more precisely than the currently used system of the HER-2/CEP17 ratio. An impact of using the proposed system for evaluation of the ERBB2 gene copy number for patients with a minor tumor clone bearing the ERBB2 gene amplification should be verified in clinical studies.
Acknowledgments
This study was supported by the Research Project of the Ministry of Health and Faculty Hospital Motol MZO 00064203 (6704). The authors thank the Mrs. M. Neradova for excellent technical assistance.
Abbreviations
- CEP17
Chromosome 17 centromeric enumeration probe
- ERBB2 (HER-2)
Gene
- ERBB-2
Protein
- IDC
Invasive duct carcinoma of the mammary gland
- I-FISH
Fluorescence in situ hybridization on interphasic nuclei
- IHC
Immunohistochemistry
- LSI
Locus specific identifier
References
- Adeyinka A, Mertens F, Idvall I, Bondeson L, Pandis N (1999) Multiple polysomies in breast carcinomas: preferential gain of chromosomes 1, 5, 6, 7, 12, 16, 17, 18, and 19. Cancer Genet Cytogenet 111(2):144–148 [DOI] [PubMed] [Google Scholar]
- Aubele M, Mattis A, Zitzelsberger H, Walch A, Kremer M, Hutzler P, Hofler H, Werner M (1999) Intratumoral heterogeneity in breast carcinoma revealed by laser-microdissection and comparative genomic hybridization. Cancer Genet Cytogenet 110(2):94–102 [DOI] [PubMed] [Google Scholar]
- Fernandez JL, Goyanes V, Lopez-Fernandez C, Buno I, Gosalvez J (1996) Quantification of C-ERB-B2 gene amplification in breast cancer cells using fluorescence in situ hybridization and digital image analysis. Cancer Genet Cytogenet 86(1):18–21 [DOI] [PubMed] [Google Scholar]
- Ginestier C, Charafe-Jauffret E, Penault-Llorca F, Geneix J, Adelaide J, Chaffanet M, Mozziconacci MJ, Hassoun J, Viens P, Birnbaum D, et al (2004) Comparative multi-methodological measurement of ERBB2 status in breast cancer. J Pathol 202(3):286–298 [DOI] [PubMed] [Google Scholar]
- Hicks DG, Tubbs RR (2005) Assessment of the HER2 status in breast cancer by fluorescence in situ hybridization: a technical review with interpretive guidelines. Hum Pathol 36(3):250–261 [DOI] [PubMed] [Google Scholar]
- Hoang MP, Sahin AA, Ordonez NG, Sneige N (2000) HER-2/neu gene amplification compared with HER-2/neu protein overexpression and interobserver reproducibility in invasive breast carcinoma. Am J Clin Pathol 113(6):852–859 [DOI] [PubMed] [Google Scholar]
- Klijanienko J, Couturier J, Galut M, El-Naggar AK, Maciorowski Z, Padoy E, Mosseri V, Vielh P (1999) Detection and quantitation by fluorescence in situ hybridization (FISH) and image analysis of HER-2/neu gene amplification in breast cancer fine-needle samples. Cancer 87(5):312–318 [DOI] [PubMed] [Google Scholar]
- Lebeau A, Deimling D, Kaltz C, Sendelhofert A, Iff A, Luthardt B, Untch M, Lohrs U (2001) Her-2/neu analysis in archival tissue samples of human breast cancer: comparison of immunohistochemistry and fluorescence in situ hybridization. J Clin Oncol 19(2):354–363 [DOI] [PubMed] [Google Scholar]
- Marinho AF, Botelho M, Schmitt FC (2000) Evaluation of numerical abnormalities of chromosomes 1 and 17 in proliferative epithelial breast lesions using fluorescence in situ hybridization. Pathol Res Pract 196(4):227–233 [DOI] [PubMed] [Google Scholar]
- Mendelin J, Grayson M, Wallis T, Visscher DW (1999) Analysis of chromosome aneuploidy in breast carcinoma progression by using fluorescence in situ hybridization. Lab Invest 79(4):387–393 [PubMed] [Google Scholar]
- Pauletti G, Godolphin W, Press MF, Slamon DJ (1996) Detection and quantitation of HER-2/neu gene amplification in human breast cancer archival material using fluorescence in situ hybridization. Oncogene 13(1):63–72 [PubMed] [Google Scholar]
- Pellikainen J, Naukkarinen A, Ropponen K, Rummukainen J, Kataja V, Kellokoski J, Eskelinen M, Kosma V-M (2004) Expression of HER2 and its association with AP-2 in breast cancer. Eur J Cancer 40:1485–1495 [DOI] [PubMed] [Google Scholar]
- Tsuda H, Akiyama F, Terasaki H, Hasegawa T, Kurosumi M, Shimadzu M, Yamamori S, Sakamoto G (2001) Detection of HER-2/neu (c-erb B-2) DNA amplification in primary breast carcinoma. Interobserver reproducibility and correlation with immunohistochemical HER-2 overexpression. Cancer 92(12):2965–2974 [DOI] [PubMed] [Google Scholar]