A gene-protein assay for human epidermal growth factor receptor 2 (HER2): brightfield tricolor visualization of HER2 protein, the HER2 gene, and chromosome 17 centromere (CEN17) in formalin-fixed, paraffin-embedded breast cancer tissue sections

Abstract

Background: The eligibility of breast cancer patients for human epidermal growth factor receptor 2 (HER2)-directed therapies is determined by the HER2 gene amplification and/or HER2 protein overexpression status of the breast tumor as determined by in situ hybridization (ISH) or immunohistochemistry (IHC), respectively. Our objective was to combine the US Food and Drug Administration (FDA)-approved HER2 & chromosome 17 centromere (CEN17) brightfield ISH (BISH) and HER2 IHC assays into a single automated HER2 gene-protein assay allowing simultaneous detection of all three targets in a single tissue section.

Methods: The HER2 gene-protein assay was optimized using formalin-fixed, paraffin-embedded (FFPE) samples of the xenograft tumors MCF7 [HER2 negative (non-amplified gene, protein negative)] and Calu-3 [HER2 positive (amplified gene, protein positive)]. HER2 IHC was performed using a rabbit monoclonal anti-HER2 antibody (clone 4B5) and a conventional 3,3'-diaminobenzidine IHC detection. The HER2 & CEN17 BISH signals were visualized using horseradish peroxidase-based silver and alkaline phosphatase-based red detection systems, respectively with a cocktail of 2,4-dinitrophenyl-labeled HER2 and digoxigenin-labeled CEN17 probes. The performance of the gene-protein assay on tissue microarray slides containing 189 randomly selected FFPE clinical breast cancer tissue cores was compared to that of the separate HER2 IHC and HER2 & CEN17 BISH assays.

Results: HER2 protein detection was optimal when the HER2 IHC protocol was used before (rather than after) the BISH protocol. The sequential use of HER2 IHC and HER2 & CEN17 BISH detection steps on FFPE xenograft tumor sections appropriately co-localized the HER2 protein, HER2 gene, and CEN17 signals after mitigating the silver background staining by using a naphthol phosphate-containing hybridization buffer for the hybridization step. The HER2 protein and HER2 gene status obtained using the multiplex HER2 gene-protein assay demonstrated high concordance with those obtained using the separate HER2 IHC and HER2 & CEN17 BISH assays, respectively.

Conclusions: We have developed a protocol that allows simultaneous visualization of the HER2 IHC and HER2 & CEN17 BISH targets. This automated protocol facilitated the determination of HER2 protein and HER2 gene status in randomly selected breast cancer samples, particularly in cases that were equivocal or exhibited tumor heterogeneity. The HER2 gene-protein assay produced results virtually equivalent to those of the single FDA-approved HER2 IHC and HER2 & CEN17 BISH assays.

Virtual slides: The virtual slides for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/2041964038705297.

Figure 1

Scheme for simultaneous visualization of human epidermal growth factor receptor 2 (HER2) protein, the HER2 gene, and the chromosome 17 centromere (CEN17) using a novel HER2 gene-protein assay. First, HER2 protein is localized through immunohistochemical staining (IHC) with a rabbit monoclonal anti-HER2 antibody and a conventional 3,3′-diaminobenzidine (DAB)-based detection method. Then, the HER2 gene and CEN17 are localized by brightfield in situ hybridization (BISH) with a cocktail of 2,4-dinitrophenyl (DNP)-labeled HER2 probe and digoxigenin (DIG)-labeled CEN17 probe. The HER2 gene and CEN17 signals are visualized with silver (a silver acetate, hydroquinone, and hydrogen peroxide reaction) and fast red (a fast red and naphthol phosphate reaction), respectively.

Figure 2

Naphthol phosphate blocks silver background staining resulting from the HER2 & CEN17 BISH procedure. Images show HER2 gene-protein staining results obtained without (A, B) or with naphthol phosphate (C, D) on formalin-fixed, paraffin-embedded (FFPE) HER2-negative MCF7 (A, C) and HER2-positive Calu-3 (B, D) xenograft tumors. In the absence of naphthol phosphate, non-specific silver deposition from the HER2 BISH detection procedure obscures the BISH signals for the HER2 gene and CEN17 targets (A, B) whereas the use of a BISH hybridization buffer containing naphthol phosphate eliminates the non-specific silver deposition (C, D). Some silver deposition was also seen in DAB staining (A). In the absence of naphthol phosphate, non-specific silver deposition occurred in mouse cells (yellow asterisks) (A, B) and mouse cells were confirmed without HER2 and CEN17 BISH signals by using naphthol phosphate (C, D). 60×.

Figure 3

The HER2 gene-protein assay yields appropriate staining of HER2 protein, the HER2 gene, and CEN17 in FFPE clinical breast cancer tissues. Tissues with HER2 IHC scores of 3+ (A–C), 2+ (D–F), 1+ (G–I), and 0 (J–L) were subjected to HER2 IHC assay (A, D, G, J), HER2 & CEN17 BISH assay (B, E, H, K), or the HER2 gene-protein assay (C, F, I, L) using tissue microarray slides. HER2 IHC assay yielded the expected HER2 protein staining (A, D, G, J) and the separate HER2 & CEN17 BISH assay yielded the expected staining of the HER2 gene (black dots) and CEN17 (red dots) (B, E, H, K). The combined assay yielded both the appropriate HER2 protein staining and the appropriate HER2 gene and CEN17 staining (C, F, I, L). [All images 60×.].

Figure 4

Differentiation of HER2 gene status in HER2 protein equivocal (IHC 2+) or negative (IHC 0) breast cancer tissues using the HER2 gene-protein assay. The HER2 gene-protein assay carried out on FFPE breast cancer tissue microarray slides was able to differentiate HER2 IHC 2+ cases that were HER2 & CEN17 BISH positive (A) and negative (B). It also distinguished HER2 IHC 0 cases that were HER2 & CEN17 BISH positive (C) and negative (D). [All images 100×.].

Figure 5

Results of HER2 gene-protein staining of FFPE breast cancer tissues exhibiting heterogeneity of HER2 positive tumor cell populations or isolated tumor cell populations. (A) The HER2 gene-protein assay demonstrated the heterogeneity of HER2 positive tumor cell populations in FFPE breast cancer tissues. In the sample shown, cell populations with HER2 IHC scores of 3+, 2+, and 1+ neighbor each other and all tumor populations present amplified HER2 gene. However, the HER2 IHC 3+ tumor cell population contains dispersed HER2 gene copies while the HER2 IHC 2+ and 1+ population contains clustered HER2 gene copies [40×]. (B) The HER2 gene-protein assay clearly visualized small groups of HER2 IHC 3+ breast cancer cells [4×]. The insert shows an isolated individual HER2 IHC positive tumor cell with HER2 gene amplification [100×].