Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) is an alternative method for ESR1 and PGR testing. Recently, the Oncotype DX assay (discussed later) started to include in its report the ESR1 and PGR status based on quantitative measurement by RT-PCR and several studies reported good correlation with IHC testing results. Taken in consideration additional advantages of the IHC method such as testing on intact tissue fragments with preserved tumor morphology, lower cost, faster turnaround time, and availability, there is currently no definitive reason to switch to RT-PCR for ESR1 and PGR testing [29, 30].

The erb-b2 receptor tyrosine kinase 2 (ERBB2, which is commonly known as HER2) is a proto-oncogene that belongs to the human epidermal growth factor receptor family of tyrosine kinases together with HER1 (EGFR), HER3, and HER4 [6, 13, 31]. The ERBB2 (HER2) gene is located on chromosome 17.9 and encodes a tyrosine-kinase receptor that is present on the surface of breast epithelium [6, 13, 32].

The epidermal growth factor receptor family are activated by ligand-induced dimerization and can homo- or heterodimerize allowing multiple receptor combinations. The formation of the dimers leads to activation of the intrinsic tyrosine kinase domain with recruitment of a variety of molecules and activation of different downstream signaling pathways, including MAPK proliferation and/or PI3K/Akt pro-survival pathways [25, 33–35].

Amplification of the ERBB2 (HER2) gene induces activation with ERBB2 (HER2) protein overexpression in 15–20 % of all breast cancers. Gene amplification correlates with protein overexpression in 95 % of cases. In the remainder of cases ERBB2 overexpression occurs through a different mechanism [6, 13].

Testing for ERBB2 (HER2) receptor status of invasive breast cancer is primarily performed to identify patients that could benefit from anti-ERBB2 therapy. Several technologies for testing (IHC, fluorescence in situ hybridization (FISH), real-time RT-PCR, and enzyme-linked immunosorbent assay (ELISA)) are available [36]. Several studies have shown a very strong correlation (in approximately 95 % of cases) between IHC ERBB2 (HER2) protein expression and gene amplification by FISH [13, 17, 37]. The current ASCO/CAP guidelines continue to endorse IHC and fluorescence ISH (FISH) techniques for ERBB2 (HER2) testing with the recent addition of “bright-field ISH,” as an alternative to FISH testing [7, 17]. Pathologists must ensure that at least one tumor sample from each invasive breast cancer (primary or metastatic) is tested either by IHC or ISH using a validated testing method.

Testing should be performed on tissue fixed in neutral-buffered 10 % formalin or another validated fixative for 6–72 h. The cold ischemia time should be as short as possible, ideally of no more than 1 h [15, 38–40]. Biopsy tissue samples fixed for less than 1 h or resection specimens fixed for less than 6 h should not be tested [16, 18].

False-positive results for ERBB2 (HER2) may be due to edge artifact, misinterpretation of cytoplasmic staining, overstaining (strong membrane staining of normal cells), or misinterpretation of carcinoma in situ as invasive tumor. Laboratory should follow the standards related to the initial test validation, ongoing internal quality assurance, external proficiency testing, and routine periodic performance monitoring. Quality assurance should include a comparison of ERBB2 IHC-positive results and amplification by FISH. The ASCO/CAP guidelines require a concordance of 95 % between IHC samples and a validated assay. External proficiency testing surveys for ERBB2 are available from CAP and other organizations to ensure that the laboratory assays are working as expected [16, 18].

ERBB2 testing by IHC is a semiquantitative method that evaluates for protein expression on the surface membrane of tumor cells [8]. There are several antibodies of different IHC-based ERBB2 tests available including Pathway ERBB2 (HER2) (clone 4B5; Ventana Medical Systems Inc., Tucson, AZ, USA), HercepTest (polyclonal, Dako Denmark A/S, Glostrup, Denmark), and Oracle ERBB2 (HER2) (clone CB11; Leica Microsystems GmbH, Wetzlar, Germany) [16, 41–43]. For interpretation of ERBB2 IHC results the US Food and Drug Administration (FDA) approved a scoring system based on the level of immunostaining of the tumor cells (Table 9.3). Examples of ERBB2 IHC are illustrated in Figs. 9.3 and 9.4 [16].

Results that scored equivocal as 2+ (11–30 %) should be reflexed to FISH testing (discussed below) for further confirmation [16, 18]. With a negative ERBB2 IHC result on a small biopsy sample, repeat testing on a subsequent specimen or by a different testing method should be taken into consideration, especially if the tumor shows discordant histologic features (high histologic grade, PGR negative or weakly expressed, high proliferation index).

FISH, chromogenic in situ hybridization (CISH), and silver-enhanced in situ hybridization (SISH) are assays that can be used to determine the presence or absence of ERBB2 (HER2) gene amplification [18]. FISH detects specific DNA sequences on a chromosome using fluorescent probes. CISH technique uses a chromogen (diaminobenzidine) to create color signals that can be then identified by light microscopy on tissue section with well-preserved morphologic details. Silver in situ hybridization (SISH) uses an enzyme-linked probe causing silver ions to deposit on the target, producing a dense, high-resolution, black easily identifiable stain [16]. For additional information about ISH techniques and comparison with IHC please see Table 9.4.

FISH for ERBB2 (HER2), the most frequently used technique, detects specific DNA sequences on chromosome 17. Some assays use a single-color probe to measure the ERBB2 gene copy number, while other assays use an additional (dual) probe (chromosome enumeration probe; CEP17) for the centromere of chromosome 17 and determine the ratio of ERBB2 signals to chromosome 17 copies. In the majority of breast carcinoma cases, both methods give the same result with rare discordances that are usually due to variation in the number of CEP17 signals. True polysomy of chromosome 17 is seen in a limited number of breast cancer cases (up to 1–2 %) [18].

Recently updated recommendations for reporting of ERBB2 testing results by ISH are presented in Table 9.5. Examples of ERBB2 FISH testing are illustrated in Figs. 9.5 and 9.6 [17, 18].

If there is discordance between the ERBB2 (HER2) results and the histopathologic features a repeat testing for ERBB2 should be ordered. ERBB2 ISH-negative results should trigger repeat testing if the breast tumor has a high histologic grade, if only a small biopsy with limited amount of tumor was initially available or if the subsequent resection specimen showed a morphologically distinct high-grade carcinoma when compared to the initial biopsy. Retesting should also be ordered for ERBB2-positive cases with discordant histology (histologic grade 1, ESR1 or PGR-positive tumors and tubular, mucinous, cribriform, or adenoid cystic variants of breast carcinomas) [44].

Tumors with ERBB2 (HER2) copy number of <4.0 signals/cell by single-probe ISH analysis should be further tested by dual ISH and reported as negative only if the dual-probe ISH analysis shows an ERBB2/CEP17 ratio of <2.0. Similarly a tumor with a ERBB2/CEP17 ratio of <2.0 should not be considered as negative unless the ERBB2 copy number is <4.0 signals/cell [18]. At the same time, tumors with low-level ERBB2 amplification (ERBB2/CEP17 ratio of ≥2.0 and an average ERBB2 copy number of <4.0 signals/cell) are considered positive for management purposes, although there is not enough evidence of a clinical benefit from targeted treatment for these cases [37, 45, 46].

An issue that can affect the interpretation of this test is the genetic heterogeneity described in 5–30 % of breast cancer tumors. Genetic heterogeneity was defined by the 2007 ASCO/CAP guidelines as having 5–50 % of invasive tumor cells with a ERBB2/CEP17 ratio greater than 2.2 or mean ERBB2 copy number >6 [16]. Chromosome 17 abnormalities can lead to alterations of the ERBB2/CEP17 ratio, potentially leading to equivocal or incorrect ISH results [41]. In these cases, the gene copy number may reflect more accurately the ERBB2 status. The recommendation for addressing ERBB2 interpretation issues resulting from genetic heterogeneity and chromosome 17 aneusomy is that a second aggregate population of >10 % of cells should be counted and reported separately [7]. In the case of a second contiguous population of ≥10 % cells with increased ERBB2 signals/cell, consisting of more than 10 % tumor cells, an additional at least 20 non-overlapping cells within this cell population should be counted separately and reported. An overall random count is not appropriate. In cases with tumor heterogeneity, biopsy samples may not be representative of the whole tumor and repeat testing on the excision specimen should be considered in order to correctly determine the percentage of amplified tumor, if present [7].

There are no clear indications of how to proceed when a population of <10 % is present.

Another 15 % of breast tumors have low to intermediate levels of ERBB2 protein expression with roughly one-third of these cases also showing ERBB2 gene amplification. The benefit of targeted therapy in this last group of patients is still uncertain and needs to be further investigated [5].

Monoclonal antibodies and tyrosine kinase inhibitors are two main types of targeted therapies used in the treatment of ERBB2-positive breast cancer. Additionally, PARP inhibitors are being studied for the treatment of triple-negative breast cancer. Monoclonal antibodies may be used in combination with chemotherapy as adjuvant therapy.

The benefit of a combination of two anti-ERBB2 (HER2)-targeted therapies with chemotherapy in early-stage breast cancer is currently being tested in the ongoing phase III ALTTO (Adjuvant Lapatinib and/or Trastuzumab Treatment Optimization) trial, as well as the Neo ALTTO (Neoadjuvant Lapatinib and/or Trastuzumab Treatment Optimization) [47, 53].

Reverse transcription-polymerase chain reaction (RT-PCR) can be employed to detect ERBB2 gene amplification in breast cancer. Few reports have suggested that RT-PCR testing can be used to confirm FISH test results or to resolve discrepancies between ISH/FISH and IHC (i.e., ISH/FISH negative, IHC 3+). Several studies have shown a concordance of up to 94 % between FISH and RT-PCR results while other authors have reported a high number of false-negative results for ERBB2 with RT-PCR [54, 55]. To date there is insufficient evidence to support qRT-PCR as a substitute for the traditional methods (IHC/FISH) for ERBB2 testing with a possible limitation related to quality assurance. For the small group of patients with equivocal results by both techniques (IHC and FISH), research into novel testing methods (such as qRT-PCR) might be of value [48].

The most recent advances in molecular pathology based on microarray gene expression studies showed that there is a heterogeneous group of breast cancers with distinct molecular features [49, 50, 56]. In their seminal research work, Perou, Sorlie, and colleagues proposed the first innovative molecular classification of breast cancer based on gene cluster subsets [16, 44]. The newly described “intrinsic” gene cluster subsets included genes of epithelial origin (ESR1, GATA–binding protein–3, X–box–binding protein–1, and hepatocyte nuclear factor–3a), ERBB2, basal epithelial cell markers (cytokeratin (CK) 5 and 17, integrin-b4, and laminin), and “normal breast” markers [57]. Initially four breast cancer “intrinsic” subtypes (basal-like, ERBB2 (HER2)-enriched, luminal, and normal breast-like) were described. Luminal breast cancers were subsequently subclassified into luminal A and luminal B and later studies identified additional “intrinsic” molecular subtypes, such as interferon-rich, claudin-low, and molecular apocrine. The major four molecular subtypes of breast cancers are presented in Table 9.6 [49, 50, 58–60].

The “breast–like” subgroup of tumors express non-epithelial cell genes, such as adipose tissue-related genes, that cluster with healthy breast and fibroadenomas. The clinical significance of this group needs further elucidation, some researchers favoring this subtype to be a false category related to misinterpretation of poor tissue sampling [16, 59, 61, 62]. Tumors of the “claudin–low” subtype show low expression of cell-cell junction proteins (claudin-3, 4, 7 and E-cadherin) and overexpression of genes involved in immune responses and linked to mesenchymal differentiation. The majority of these tumors are ESR1, PGR, and ERBB2 (HER2) negative with up to 20 % of the cases positive for hormone receptors. Claudin-low tumors are histologically of high grade and have a poor prognosis [6]. The “molecular apocrine” subtype shows similarities to the ERBB2-enhanced subtype with the addition of androgen receptor (AR) activation while the “interferon–rich” subtype is characterized by STAT1 activation [15].

Based on additional molecular studies (genomic DNA copy number arrays, DNA methylation, exon sequencing, mRNA arrays, microRNA sequencing, and reverse-phase protein arrays) the Cancer Genome Atlas Network reported even more new molecular findings in breast cancer. Up to 10 % of all breast cancers were found to have somatic mutations involving three main genes (TP53, PIK3CA, and GATA3). Specific mutations involving GATA3, PIK3CA, and MAP3K1 were described in the luminal A molecular subtype. HER2-positive tumors were further divided into two ERBB2 (HER2) protein expression-defined subgroups with specific signaling pathways. For the basal-like subtype many molecular similarities with high-grade serous ovarian tumors were reported, suggesting a related etiology and the possibility of similar therapeutic approaches [63]. Other novel mutations reported in breast cancer included those in TBX3, RUNX1, CBFB, AFF2, PIK3R1, PTPN22, PTPRD, NF1, SF3B1, and CCND3. Of note, several overlaps with mutations previously described in other diseases were seen. For example, RUNX1 and CBFB showed the same mutations as previously described in acute myeloid leukemia. PIK3R1 mutations were similar to those previously described in glioma and endometrial cancer. The same SF3B1 mutations previously reported in myelodysplastic syndromes and chronic leukocytic leukemia were identified [63].

Approximately 50 % of breast cancers are associated with cyclin D1 overexpression and only a minority of cases with CCND1 gene amplification [65, 73]. Its prognostic utility remains controversial and needs to be further investigated. Overexpression of cyclin D1 was reported to be associated with a favorable prognosis [74] while gene amplification with poor outcome [75] and both with a poorer response to endocrine therapy [76]. Cyclin E overexpression appears to be inversely related with ESR1 expression, and associated with poor prognosis, less benefit from endocrine therapy, and increased sensitivity to chemotherapy.

Topoisomerase IIα (TOP2A) overexpression is reported in luminal B, ERBB2 (HER2)-enriched, and basal-like breast tumors. Anthracyclines inhibit TOP2A activity and several studies have evaluated the predictive value of TOP2A with inconsistent results. Both overexpression and deletion of TOP2A were seen in breast tumors showing a favorable response to chemotherapeutics [78, 79]. TOP2A FISH pharmDxTM Kit is an FDA-approved FISH-based test designed to detect amplifications and deletions of the TOP2A gene on formalin-fixed, paraffin-embedded tumor samples. Both deletions and amplifications of the TOP2A gene serve as markers of poor prognosis in high-risk breast cancer patients. The results from this test can be used as an adjunct to the clinicopathologic data [78, 79].

Studies have shown that the number of circulating tumor cell (CTC) is associated with lymph node involvement and disease-free survival in early breast cancer and with progression-free and overall survival in metastatic disease. These findings can have implications for assays targeting epithelial antigens that may miss part of the circulating cell populations. The CellSearch CTC detection system (Veridex, LLC) is FDA approved for use in metastatic breast cancer. The test uses an immunomagnetic system to isolate epithelial cells from the peripheral blood and offers a count of CTCs per 7.5 mL blood sample. Greater than five CTCs per 7.5 mL sample have been associated with worse progression-free and overall survivals. SWOG S0500 clinical trial will further determine if the CTC level can be used to assess the benefits from chemotherapy in metastatic breast cancer cases. Until those results are available, this test is not recommended by the ASCO breast cancer tumor marker panel and is not included in the breast cancer NCCN guidelines [80].

A variety of multigene and multiprotein expression assays are available for the identification of tumor subtypes and for the assessment of prognosis and likelihood of response to specific treatments. The majority are proprietary assays developed and performed by a single laboratory. Multigene assays can detect gene expression patterns by several techniques including qRT-PCR; hybridization of labeled tumoral nucleic acids to small, immobilized, synthetic DNA strands (microarrays); FISH; and immunohistochemistry [32, 81, 82]. The aim of gene expression profiling is to further assist clinician in objectively estimating the outcome by providing a better prediction than the traditional clinical and pathologic parameters. Pathologists have the option to include the results of these assays in their reports [83]. Several gene signatures with prognostic and predictive value in breast cancer are available for use and algorithms for inclusion of these assays in clinical therapy decision making have been created [16].