Sage Journals: Discover world-class research

Abstract

Breast cancer prognosis and treatment is guided by traditional clinicopathological parameters and individual molecular markers. Despite the remarkable advances in our scientific understanding of breast cancer genetics, the impact of such information on medical care has, to date, been modest. Although the use of simple genetics is already in vogue in clinical practice, the concept of molecular profiling and multiparameter gene classifiers was raised after the introduction of the high-throughput gene expression microarrays. This technology, in addition to highlighting the molecular heterogeneity of breast cancer, has led to the development of prognostic and predictive gene signatures. Studies are underway to assess the clinical validity and clinical utility of these multigene assays and their incorporation into clinical practice. This article reviews the current status and projected future use of genetics and genomics in breast cancer management and their impact on the refinement of risk stratification to permit individualized and patient-tailored therapy. Limitations based on our current scientific understanding and realistic expectations are also explored.

Keywords

breast cancer genetics genomics management outcome

Cancer is a complex genetic disease. The process of carcinogenesis usually results from alteration of one or more key cancer-related genes. Further genetic abnormalities are associated with cancer progression. Initial molecular studies of cancer focused on genes that control cell cycle: mutations of oncogenes typically accelerate the cell cycle, while mutations of tumor suppressor genes typically release cell cycle blocks [1,2]. Further studies demonstrated that many cancer-related genes influence other important cellular functions, such as signal transduction, DNA repair and chromosomal homeostasis, with subsequent impact on the global molecular profile and biological functions of cancer cells [3]. Although phenotypic and biological changes associated with carcinogenesis, tumor progression, response to therapy and development of drug resistance are produced by alterations in the expression of target genes, the underlying defects may be detected at various levels, including the genetic material (i.e., DNA mutation, deletion, translocation or amplification) and/or the gene products (i.e., RNA and proteins). Other important changes that regulate gene expression include methylation of DNA promoter regions and modification of his tones. Tissue interactions, such as angiogenesis and signals from the stroma, may also play a role. The advances in the annotations of the genome, the emergence of high-throughput gene profiling technologies, the discovery of miRNAs and the recognition of epigenetic changes have allowed us to elucidate the complex genetic basis for cancer and to refine our models of prediction of tumor behavior and clinical heterogeneity [4 –6].

This article reviews the current status and projected future use of genetics and genomics in breast cancer (BC) management to: refine BC stratification to permit individualized and patient-tailored therapy; provide prognostic and predictive information to guide management; and provide information about future risk. This article also examines the limitations based on our current scientific understanding and realistic expectations for the future.

BC heterogeneity & clinical management

BC, the most common cancer and the second-leading cause of cancer death in women, represents a heterogeneous group of tumors with varied biological and morphological features, behavior, and response to therapy. A better prediction of these parameters at the individual level should improve patient management and survival. Current routine management of BC relies on the availability of well-validated clinical and pathological prognostic factors and only a few well-defined molecular predictive markers to support treatment decisions. Prognostic and predictive factors in BC include staging prognostic variables, such as tumor size and lymph node (LN) stage, and biological/molecular variables in addition to patient-related variables such as age. Also, there is a strong temporal relationship between biological and molecular prognostic variables and tumor stage [7 –9]. Although advanced stage is a poor prognostic indicator and an indication for systemic therapy, in early-stage disease, therapy is mainly determined based on biological factors. In addition, although tumor stage is an indication for effective ‘proportionally’ aggressive therapy, knowledge about biological and molecular tumor characteristics is needed for determining the nature and type of systemic therapy.

Biological prognostic & predictive variables

Biological prognostic and predictive variables are primary tumor molecular characteristics that are determined by the underlying genetic abnormalities and, thus, invite assessment. However, for a molecular/genetic marker to be of value, its assay results must reflect the associated biological process, although this association does not necessarily imply clinical utility. Clinical utility of a molecular test is measured by its ability to predict a defined clinical outcome with high positive and negative predictive value. If assessment of a molecular marker does not lead to a decision in clinical practice, then its use in routine practice is discouraged [10]. Not only outcome prediction, but also health economic analysis should be considered in the assessment of the clinical utility of potential molecular markers. Furthermore, testing methodologies must be reliable, accessible, interpretable and easily usable.

Biological variables in BC can be assessed using morphological surrogates such as tumor differentiation (e.g., tumor grade and histological type), proliferation status and lymphovascular invasion, or more accurately using molecular parameters individually or in consort (e.g., molecular profiling at DNA, RNA or protein levels). These molecular markers are used in routine practice, in combination with staging parameters, to assign patients into different risk groups to determine the need and likelihood of response to specific systemic therapies. However, tumors with remarkably similar parameters still vary in response to therapy and have distinct outcomes. This is perceived as a weakness of traditional staging parameters and points to the biological and genetic variability inherent in the disease. In addition, using traditional risk classifications, a large proportion of patients are classified as intermediate risk (i.e., grade 2, HER2-negative and with negative/low-volume positive LNs), which is uninformative for choosing optimal treatment strategies [11]. Hence, unsurprisingly, the subtle nuances of individual BC cases have been hard to decode, and personalized therapy remains an unattained target. This emphasizes the need for identification of additional molecular prognostic and predictive variables. For this purpose, several research groups have tried to use genome-wide profiling of chromosomal changes, gene expression, and large-scale screening of genes for mutations/single nucleotide polymorphisms and methylation in an attempt to refine BC prognostication and prediction of response to therapy.

BC & gene expression

Gene expression is a technical term to describe how a particular gene is active, or how many times it is transcribed and translated into its protein. Assessment of gene expression can be performed for an individual gene, a group of genes or for the entire genome (global gene expression profiling [GEP]). Of the individual molecular markers, the ER, the PR and the HER2 genes have proven predictive and prognostic value. Although routine assessment of these markers was validated many years ago using different platforms, such as immunohistochemistry (IHC), in situ hybridization and PCR, the concept and clinical utility of genome-wide profiling and multigene assays, introduced in the last decade, has attracted the most attention. GEP has been proposed as an adjuvant tool to assist in determining outcome probability, choosing treatment options and determining response to therapy. It is proposed that GEP used in conjunction with consensus guidelines and risk assessments can help to identify those women who do not need adjuvant chemotherapy [12].

In 2000, Perou and colleagues introduced the concept of GEP in BC and demonstrated that BC can be classified based on the global molecular profile (molecular taxonomy), and indicated that several genes can be used in combination to identify a distinct subclass that showed association with clinical outcome [13]. Multiple independent studies have refined and validated this concept [14,15]. Subsequently, the concept of prognostic gene signatures that can predict tumor behavior and/or response to therapy (class prediction) was introduced; for example, the recurrence score (Oncotype DX®, Genomic Health, Redwood City, CA, USA) [16], the 70-gene signature (MammaPrint®, Agendia BV, Amsterdam, The Netherlands; also referred to as the ‘Amsterdam signature’) [17] and the Gene Expression Grade Index [18], which have been validated in subsequent independent studies [19 –21]. In addition, GEP can be used to compare different ‘predefined’ classes of BC (class comparison) to determine whether the expression profiles are different between these classes. In addition to transcriptomics, several research groups have tried to determine the molecular profile of BC and relate this to clinical behavior using other high-throughput techniques such as proteomics [22], tissue microarrays and IHC [23], or global DNA profiling [5,6].

Global gene expression & BC classification (molecular taxonomy)

Initial molecular taxonomy has classified BC into three well-defined classes:

A luminal class, which includes tumors that express ER and genes related to activation of the ER pathway;

A HER2 class, which encompasses tumors characterized by overexpression of HER2 and by genes pertaining to the HER2 amplicon;

A basal-like class, which lacks expression of genes characteristic of either luminal or HER2-positive tumors, and is largely characterized by high proliferative activity, and expression of basal cytokeratins and other genes that are characteristic of basal-like cells of the breast [13 –15].

In addition to the aforementioned three classes, GEP has identified a few less well-defined molecular subtypes, such as a normal breast-like class, the subclasses of luminal tumors (luminal A, B and C), and molecular apocrine and claudin-low subtypes. The so-called ‘normal breast-like’ class is characterized by expression profiles similar to those found in normal breast tissue. Unlike other subtypes in which discrepancies exist between IHC and gene expression, some authors have reported a concordance of almost 100% between HER2 molecular subtype and HER2-positive expression by IHC [24].

Although HER2- and ER-positive (luminal) tumors were known before the advent of this GEP molecular taxonomy, the basal-like class attracted attention as a novel class characterized not only by its triple-negative phenotype (ER, PR and HER2 negative), but also by the generally similar molecular profile and poor outcome [25]. Basal-like tumors show the most frequent chromosomal gains and losses but less frequent DNA amplification than other subtypes. These tumors seem to harbor a dysfunctional BRCA1 pathway and tumors arise in BRCA1 mutation carriers [25]. Although the diagnosis of basal-like BC is still not routine, this discussion is by no means an academic exercise, given the prognostic implications and the likely development of specific chemotherapy regimens (discussed later). However, the clinical difference between basal-like BC, as defined by GEP and the triple-negative phenotype on IHC, is disputed, with triple negativity providing a more practical classification preferred by oncologists.

Although molecular taxonomy of BC using GEP generated speculation that this would result in dramatic improvements in management, practical adoption appears limited. Certain critical issues have been raised regarding reproducibility, validation and clinical utility. Most luminal tumors are hormone receptor positive and can be identified in routine practice using IHC. ER expression without the luminal phenotype is recognized as a validated predictor for hormone therapy. The significance of luminal ER-negative tumors is not defined. Subgrouping of luminal tumors is not stable and reflects the rather subjective nature of class assignment based on hierarchical clustering. Lumping of different tumor types with a distinct behavior into a single ‘luminal’ class is not clinically sound. Moreover, the so-called ‘normal breast-like’ class is not well defined and may be an artifact of GEP. In fact, the four main molecular classes frequently reported can be considered as an oversimplification that adds little to our understanding of the biology and behavior of BC. The proportion of some classes varied substantially, and in most of the studies, a proportion of tumors could not be included in any of the well-defined core classes; the so called ‘unclassified’ BC, which varied from 5 to 16% [26,27].

One could argue that the GEP molecular taxonomy is a mere reflection of hormone receptor (HR), HER2 and proliferation status of BC. It has been shown that using simpler approaches based on IHC evaluation of the expression of selected well-defined proteins with relevance to BC can provide a practical surrogate to GEP molecular classification in routine clinical samples [28,29] and is expected to remain as such at least in the near future.

BC & prognostic gene signatures

Several multigene classifiers that predict outcome and response to therapy in BC have been developed; an approach which was pioneered by van de Vijver and colleagues in 2002 [30]. Currently, many classifiers have been generated using different technologies such as cDNA and oligonucleotide arrays, and multiplex PCR. Many of these multigene classifiers have been developed and validated in specific groups of BC, particularly in LN-negative, ER-positive patients [16,31], while some have claimed prognostic significance in BC as a whole [30,32]. The most frequently reported and validated assays are the 21-gene signature (Oncotype DX) [16], the 70-gene MammaPrint [17], the Breast Cancer Gene Expression Ratio (HOXB13:IL17BR; also known as THEROS H/I^SM, formerly Aviara H/I^SM, bioTheranostics, San Diego, CA, USA) [33,34] and the 76-gene ‘Rotterdam signature’ (Veridex LLC, Warren, NJ, USA) [35]. These four commercial assays represent the first introduction of this technology into clinical application. Other prognostic signatures with clinical significance include the invasiveness gene signature (Oncomed Pharmaceuticals, Redwood City, CA, USA) [36], HERmark® Breast Cancer Assay (Monogram Biosciences, San Francisco, CA, USA), wound-response gene-expression signature [37], hypoxia gene signature [38], a 4l-gene signature [39] and a 95-gene signature [40], the genomic grade index [18], and the Breast Cancer Index^SM (bioTheranostics) [41].

From a conceptual viewpoint, there are broadly two types of tests: those that provide results as a continuous variable and those that provide categorical (usually dichotomous) results. The Oncotype DX assay is an example of the former while MammaPrint and intrinsic subtype assays are examples of the latter.

The Oncotype DX has been endorsed by the American Society of Clinical Oncology for clinical use [29] and included in recent National Comprehensive Cancer Network (NCCN) guidelines [201] for use in a specific subgroup of women with BC. The assay was initially developed and validated in HR-positive, LN-negative patients who received tamoxifen therapy. It has been suggested that tamoxifen-treated patients with a low recurrence score (RS) may be spared adjuvant chemotherapy, while patients with a high RS may benefit from adjuvant chemotherapy. Subsequent studies have demonstrated the clinical utility of Oncotype DX in LN-negative patients who did not receive systemic therapy [42], LN-positive postmenopausal women treated with hormone therapy [21] and as a predictor of response to neoadjuvant therapy [43]. MammaPrint is the first prognostic signature to be approved by the US FDA for use in LN-negative BC patients less than 61 years of age with tumors of less than 5 cm in diameter. It was reported that the MammaPrint test can identify groups of patients with very good or very poor prognosis and that it provides prognostic information beyond that of Adjuvant! Online [17].

The Rotterdam Signature is another multigene classifier that consists of a 76-gene set that does not overlap with either the Oncotype DX or MammaPrint assays, and is heavily weighted towards proliferation genes. Unlike the previous two assays, this test is specifically studied in all LN-negative patients, regardless of age, tumor size and grade, or HR status. Similar to MammaPrint, this assay requires frozen tissue.

Another real-time PCR-based assay is the two-gene ratio or H:I (THEROS H/I) that measures the ratio of HOXB13 to IL17BR, and is marketed as a marker of recurrence risk in ER-positive/LN-negative patients [34]. THEROS MGI^SM (Molecular Grade Index, formerly Aviara MGI^SM) is a multigene assay comprised of five genes related to histological grade and tumor progression that can predict clinical outcome [44]. The Molecular Grade Index discriminates between tumor grades 1 and 3, and can reclassify grade 2 tumors into low- and high-risk groups for recurrence. It is also a prognostic indicator for ER-positive patients regardless of nodal status. The Breast Cancer Index is a commercial combination of the THEROS H/I and the bioTHEROS MGI [41]. Results using combined testing are reported as low-, intermediate- or high-risk for recurrence. Another multigene classifier is the Invasiveness gene Signature [36], which consists of 186 genes and is designed for both LN-negative, LN-positive, ER-negative and ER-positive BC. In addition to the previous ‘first-generation’ (i.e., proliferation-driven) signatures, stroma and immune-related signatures have been shown to provide prognostic information in ER-negative [45,46] and HER2-positive [47] disease.

A DNA prognostic signature has recently been assessed in a set of small LN-negative invasive ductal carcinomas using comparative genomic hybridization array [48]. A DNA comparative genomic hybridization classifier, identifying low- and high-risk groups of metastatic recurrence has been reported, and its prognostic significance was claimed to be maintained in multivariate analysis.

Gene expression arrays: comparison & clinical utility

Despite the validation studies and the plethora of publications concerning multiparameter gene assays in BC, their precise clinical utility is still under investigation. The utility of these assays should be based on the clinical context for which they were developed [49]. They are not advocated as standalone tools and are currently recommended as an adjuvant tool to be used with other well-established prognostic indicators.

The added value of these recent molecular classifiers may require more than demonstrating an association with clinical outcome. These novel studies should always assess their additional value, over and above that of known prognostic variables, rather than association with outcome. An example of such an evaluation technique is demonstrated in the ATAC trial, in which RS (Oncotype DX) was shown to add significantly to the prognostic prediction of Adjuvant! Online [21]. The biological roles of the gene sets included in most of these assays are not completely understood, and it is often unclear which characteristics are being measured. Although HR-related genes, and HER2-related genes are essential components in most classifiers and proliferation-related genes are the main denominator for almost all signatures, there is little overlap and instabilities among different gene lists exist. Such signatures have the same prognostic performance [19] and most of them provide prognostic information only in ER-positive disease. Although prognostic signatures are reported to predict early relapses, their predictive value for late recurrences is limited or absent [50]. Moreover, it has been reported that Oncotype DX produces informative results in 60% of cases with the widely used recurrence score thresholds, and this figure drops to 34% with the use of the revised thresholds implemented by the Trial Assigning IndividuaLized Options for Treatment [Rx] (TAILORx) trial while the remaining cases are classified as intermediate risk [31]. Traditional markers such as Ki67, HR, HER2, vascular invasion and tumor grade evaluate the same biological characteristics and, therefore, the significance of its routine application may be doubtful [51]. Cuzick et al. have demonstrated that the prognostic information provided by combined expression of ER, PR, HER2 and Ki67 (IHC4 score) is similar to that provided by RS and using RS with IHC4 adds little additional prognostic value [29].

The clinical utility of multigene assays was assessed in different independent validation studies. The published peer-reviewed literature supports the accuracy and clinical utility of Oncotype DX in its ability to predict the benefits of chemotherapy in women with localized early-stage (stage 1 or 2) ER-positive, HER2-negative and LN-negative BC. Studies reported that patients with a low RS had a 10-year distant recurrence-free survival rate of 93%, which decreased to 86% for patients with an intermediate RS, and 70% for patients with a high RS [16,21,42,52,53]. Although some reports have demonstrated a prognostic value for RS in postmenopausal women with ER-positive, 1–3 LN-positive BC treated with adjuvant tamoxifen [21,50], these studies have limitations. In a systematic review of literature published by the BlueCross BlueShield Association Technology Evaluation Center (TEC) in 2010, it was concluded that the 21-gene recurrence score assay did not meet TEC criteria for GEP to aid in the selection of adjuvant chemotherapy in women with LN-positive BC, and that the available evidence did not allow conclusions for selecting adjuvant chemotherapy in this subpopulation [54].

Marchionni and colleagues conducted a systematic review to summarize studies on Oncotype DX (n = 10), H:I expression ratio (n = 6) and MammaPrint (n = 4). Although the authors concluded that these technologies show great promise, they indicated that more information is needed regarding the extent of improvement in prediction, in whom the tests should be used and how test results are best incorporated into clinical decision-making [32]. Further studies examining the incremental contribution of these assays over conventional predictors, optimal implementation with consideration to cost—benefit analysis and relevance to patients receiving current therapies are required.

In 2008, the BlueCross BlueShield Association TEC published an assessment on the role of GEP in the treatment of BC [55]. It was stated in the report that the evidence for Oncotype DX is sufficient to permit conclusions regarding improved net health outcomes in LN-negative, ER-positive BC patients who met the specific trial enrolment criteria, and it also provides information about the risk of recurrence. In relation to MammaPrint and the two-gene assay, the report stated that the evidence is insufficient for routine use.

The literature review by the ECRI Institute concluded that available data provide evidence of clinical validation for the ability of both the MammaPrint and Oncotype DX assays to predict tumor recurrence and response to chemotherapy [56]. However, the current studies are insufficient to draw strong conclusions regarding these assays' clinical utility for guiding treatment decisions for patients with early-stage invasive BC. In another comprehensive review, the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group (EWG) found adequate evidence regarding the association of the Oncotype DX RS with disease recurrence and response to chemotherapy [57]. Although adequate evidence to characterize the association of MammaPrint with future metastases was found, the evidence to assess the added value to standard risk stratification was inadequate, and it was not possible to determine the population for which the test would best apply. Regarding the clinical utility of these multigene assays, the review concluded that there was no evidence of the MammaPrint and the two-gene ratio tests, and inadequate evidence regarding Oncotype DX. There was insufficient evidence to make a recommendation for or against the use of GEP to improve outcomes in defined populations of BC patients with a significantly positive benefit:harm ratio; despite finding evidence of potential benefit to some women (reduced adverse events due to low-risk women avoiding chemotherapy), it was not possible to rule out the potential for harm for others (BC recurrence that might have been prevented). Thus, these technologies have potential for both benefit and harm and the evidence is insufficient to assess the balance between the two. The EWG has, therefore, encouraged further development and evaluation of these technologies.

The clinical utility of the gene signatures will be further modified by results from prospective randomized controlled trials. Two trials that are currently ongoing are the TAILORx trial and the Microarray in Node Negative Disease May Avoid Chemotherapy (MINDACT) trial trial [58,59]. Table 1 summarizes the main intentions of these trials and the questions they seek to address.

Table 1.

Randomized studies to prove the efficacy of genetic classifiers of risk in breast cancer.

Name of trial	Basis of prognostic signature	Patient cohort	Aims of the trial	Patient stratification and randomization	Primary outcome	Points of note
TAILORx	Clinical utility of Oncotype DX®	Lymph node-negative ER and/or PR-positive patients	Refine the Oncotype DX assay in patients with intermediate risk scores and determine whether hormonal therapy alone is noninferior to chemotherapy plus hormonal therapy in this cohort of patients	Three groups using the recurrence score on Oncotype DX:• Group 1, with a score <10, receive hormonal treatment only• Group 2, with a score>25, receive combination chemotherapy and hormonal therapy• Group 3, with a score ranging from 11 to 25, randomized to receive either combination chemotherapy and hormonal therapy or hormonal therapy alone	Disease-free survival, distant recurrence-free interval, recurrence-free interval and overall survival	The split of the Oncotype DX score into three categories of risk at training and validation of this molecular predictor (<18, 18–30 and >30) is different from the three categories used in this study
MINDACT	Clinical utility of MammaPrint® vs Adjuvant! Online	Lymph node-negative breast cancer patients	To identify if gene-expression profiling can identify women who could be spared chemotherapy without compromising long-term disease outcome	• Patients whose prediction is ‘high risk’ according to both Adjuvant! Online and MammaPrint are given chemotherapy• Patients whose prediction is ‘low risk’ according to both Adjuvant! Online and MammaPrint are not given chemotherapy• Patients who have contrasting predictions (high risk by Adjuvant! Online and low risk by MammaPrint) are randomized to either receive or not receive chemotherapy	Distant metastasis-free survival	Power calculations are not based on the comparison of the two randomized arms. The potential benefit is that an estimated 10–15% of women could be spared toxicities without compromising long-term outcome

MINDACT: Microarray in Node Negative Disease May Avoid Chemotherapy Trial; TAILORx: Trial Assigning Individualized Options for Treatment.

Data taken from [58,59].

Genetics & BC treatment

Genetics & cancer prevention

Regarding BC prevention, genetics has made inroads, in that inherited loss-of-function mutations in the tumor suppressor genes BRCA1 and BRCA2 are known to predispose to high risks of BC and/or ovarian cancer [60,61]. Genetic testing for BRCA1 and BRCA2 mutations has now become an integral part of clinical practice for women with a strong family history of BC or ovarian cancer. BRCA1 or BRCA2 testing is now also being undertaken following/during adjuvant treatment to clarify the risk of second primary cancers and relatives' cancer risks. The role of BRCA testing has evolved from initial genetic counseling and testing to inform adjuvant chemotherapy. BRCA1-associated tumors are characterized by unique features and a high degree of genetic instability, which have implications for the type of systemic therapy used. One of the functions of the BRCA1 gene is DNA damage repair, and in its absence, the cell becomes dependent on error-prone methods of DNA double-strand break repair [62]. These have implications for the response to therapy not only to current agents, but also to novel targeted therapies such as PARP inhibitors that use this deficiency in tumor cells for therapeutic gain [63]. Moreover, it has been reported that tumors with BRCA1 dysfunction may be sensitive to agents that cause DNA damage, such as platinum agents (cisplatin and carboplatin) [64], and to anthracycline-containing chemotherapy [65], but may confer resistance to taxanes [66]. Apart from the screening of BRCA1 for gene mutations, techniques for detecting BRCA1 dysfunction are still evolving. This field became more important after the discovery of the link between basal-like carcinoma and tumors associated with BRCA1 dysfunction [25]. IHC for BRCA1 is still in the research phase, and the correlation between BRCA1 status and BRCA1 IHC expression is not perfect. Similarly, using a set of IHC markers to identify tumors with BRCA1 dysfunction may provide an acceptable degree of sensitivity [67], but the specificity is low [68]. Moreover, despite the success of PARP inhibitors in BRCA1-associated tumors [63,69], predictive markers for response to PARP inhibitors within BRCA1-associated and basal-like tumors are still under investigation.

In an effort to detect mutations in more genes, newer techniques, such as massively parallel, ‘next-generation’ sequencing, have been investigated by researchers to develop an assay to capture, sequence and detect mutations in 21 genes, including BRCA1 and BRCA2, known to predispose to breast or ovarian cancer [70]. Such techniques have the potential to open up future domains of widespread genetic testing and personalized risk assessment for breast and ovarian cancer.

Other germline mutations also exist. Of these, p53 and PTEN mutations are tested in the UK. Other genes that can slightly increase a woman's risk of developing BC include RAD51, CASP8, FGFR2, TNRCP, MAP3K1, rs4973768, LSP, CHEK2, ATM, BRIP1 and PALB2 [71]. No tests are available for these genes yet. In addition to germline mutations and hereditary BC, genetic assays have been applied to identify factors necessary for progression from ductal carcinoma in situ to invasive disease. Genetic signatures can be used to identify cases that are likely to progress to invasive disease before morphological changes in the lesion become apparent [72,73]. Although currently only used as a research tool, it is anticipated that specific genetic markers may guide the management of ductal carcinoma in situ, which comprises approximately 20% of malignant diagnoses in the BC screening setting [72,73].

Genetics & hormonal therapy

The discovery and assessment of the ER has revolutionized BC management in terms of prognostication and prediction of response to hormone and other systemic therapies. BC has traditionally been perceived as a tumor driven by the hormone estrogen. Recent GEP studies have emphasized the role of ER in BC as the key driver molecule together with HER2 control [14,15,17]. More than two thirds of BCs are ER-positive [74] and hormonal manipulation with anti-estrogens forms the backbone of management for such patients. With the availability of robust IHC techniques, complex genetic techniques are rarely used to identify the ER status. However, not all ER-positive patients benefit from hormonal treatment. Here, pharmacokinetics may have some explanations to offer [75]. Tamoxifen, one of the main hormonal therapies in use, is converted to its active metabolite through the CYP pathway, with CYP2D6 being the key enzyme responsible for the conversion of N-desmethyl tamoxifen to endoxifen [75]. Single nucleotide polymorphisms in the CYP2D6 gene are common with some alleles coding for enzymes with reduced, null or increased activity. Various studies suggest that carriers of CYP2D6 alleles encoding for null or reduced activity may have a poorer outcome when treated with tamoxifen in the adjuvant setting compared with women carrying two alleles encoding for normal activity [75]. Unfortunately, the data are not uniformly concordant and, hence, definitive evidence to revise routine clinical practice is not yet available. IHC assessment of some markers, such as progesterone receptors [28,76], HER2 [28], the proliferation marker Ki67 and some ER-related genes such as FOXA1 and GATA3 [77,78], have been suggested as prognostic and predictive markers that can be used in ER-positive BC. Furthermore, while testing ER by IHC is considered a standardized technique, complex genomic techniques still provide important information to differentiate between at least two distinct subtypes of ER-positive tumors with different prognoses (i.e., luminal A and luminal B).

Genetics & targeted therapy

HER2

An area where genetics has everyday use in BC today is the determination of the HER2 status, which has prognostic and treatment implications. HER2 is one of four members of the HER family of receptor tyrosine kinases [79]. The amplification of this gene was first described in DNA prepared from tissue of human mammary carcinoma in 1985 by King et al. [80]. HER2 overexpression is observed in 20–25% of patients with BC [81]. The prognostic value of HER2 amplification was first determined by Slamon et al. [82] and confirmed by subsequent independent studies [83]. In addition to its prognostic significance, HER2 amplification and protein expression has been shown to predict and modulate the response of conventional chemotherapeutic agents and endocrine therapy, and more importantly, it predicts response to anti-HER2-targeted therapies [83]. A recombinant humanized monoclonal antibody termed trastuzumab (Herceptin®) was used to displace ligands to the extracellular portion of HER2, in order to silence the HER2-mediated mitogenic signalling system. Another class of therapeutic agents against HER2 is based on small molecules that suppress ATP-mediated phosphorylation of the HER2 intracellular domain. These molecules are the tyrosine kinase inhibitors (TKIs), the forerunner of which is lapatinib (Tykreb®) [84]. Trastuzumab and TKIs increased the arsenal of therapeutic options for BC in neoadjuvant, adjuvant and metastatic settings. The premise for the use of both these agents, however, is HER2 IHC positivity or HER2 gene amplification.

HER2 gene amplification can be assessed by different genetic assays including southern blot, slot blot, PCR, FISH and chromogenic in situ hybridization, while protein overexpression can be analyzed by western blot, ELISA or IHC. The current recommendations (American Society of Clinical Oncology [ASCO] guidelines [81]) for HER2 positivity in BC is the assessment of HER2 status using IHC, which is scored as follows: negative (0/1+), equivocal (2+) or positive (3+). HER2 IHC is complemented by additional testing for HER2 gene amplification by FISH in cases with equivocal (2+) staining to confirm the HER2 gene status of the tumor [81]. This currently exemplifies a direct and simple method of integration of genetics into clinical practice.

Other targets

The mTOR is a serine-threonine protein kinase that is involved in cell growth and survival. The central component of the pathway is the PI3K heterodimer and mutations in the catalytic domain of PI3K have been identified in 20–25% of BCs [85,86]. The mTOR pathway is pivotal not only in tumorigenesis, but also in chemotherapy and hormonal drug sensitivity. Currently, there are at least three rapamycin-derived mTOR inhibitors in clinical development: viz.CCI-779 (temsirolimus), AP23573 (deforolimus or ridaforolimus), and RADOO1 (everolimus) [87]. The probable requirements for the appropriate selection of BC patients who can successfully be treated with an mTOR inhibitor are the overexpression of HER2 and/or ER positivity. It has also been shown that mTOR activation caused by either PTEN mutation or AKT overexpression is particularly susceptible to mTOR inhibitors [88]. Other molecular biomarkers may be identified in the future.

EGFR is a cell-surface molecule that has been implicated in the pathogenesis of BC. It has been suggested that EGFR may be important in the emergence of resistance to endocrine therapy. Initial clinical studies have found that EGFR inhibitors such as gefitinib may delay the development of resistance to endocrine therapy in patients with BC when given concurrently with hormone therapy [89]. EGFR is expressed at high levels in at least 20% of BCs overall [90], but in 60–70% of patients with the triple-negative phenotype [25]. Furthermore, it is overexpressed in up to 80% of metaplastic carcinomas, a variant of BC, and of these, 34% have EGFR gene amplification [91]. In an overview of studies of the TKIs, erlotinib and gefitinib as targeted treatments for BC, Agrawal et al., point out that efficacy depends on the presence of EGFR protein in the tumor tissue [92]. However, other negative reports indicate that EGFR expression does not indicate benefit from inhibitors [93,94]. Thus, current evidence does not support using routine IHC assessment of EGFR in BC for a therapeutic purpose and research is being carried out to further explore this option.

VEGF is a potent angiogenic factor associated with a poor prognosis in BC [95]. Hence, blocking the action of VEGF appears to be a promising approach to BC treatment. Bevacizumab (Avastin®), a monoclonal antibody targeting VEGF, is effective in a variety of solid tumors, including BC. Other genes that have been involved in the development of targeted therapy include the Src family of tyrosine kinases, which exert a prominent role in phosphorylating key regulators of adhesion and migration, and promoting tyrosine phosphorylation of the receptor tyrosine kinases [96]. Dasatinib is an oral, small-molecule TKI that acts on the Src family of tyrosine kinases. The role of epigenetic gene silencing has also been explored. Inhibitors of the enzyme his tone deacetylase, which suppresses gene transcription by modifying chromatin, have emerged as a potential new treatment option for BC [97,98].

Although these results are promising, further validation, typically in randomized controlled clinical trials, is needed to better understand their efficacy and safety, and to identify and validate target molecules that can be used to predict response to the specific therapy.

Genetics & chemotherapy

Taxanes and anthracyclines are the two most commonly used chemotherapy agents in BC; however, several patients do not respond and are subject to undue toxicity. Studies have looked at molecular markers of prognosis for both treatment options. Triple-negative status was reported as an independent predictor of resistance for doxorubicin while GEP showed that docetaxel was superior to doxorubicin in the basal-like subtype; no significant differences were observed in the other subtypes when comparing these two drugs [99].

TOPO2A encodes topoisomerase IIα, and its amplification is associated with a good response to anthracyclines, whereas deletion may be accompanied by resistance [100]. Some studies have demonstrated a predictive role of TOPO2A particularly in the HER2-amplified cases [101]; however, other studies failed to demonstrate such association and the data so far are inconclusive [102]. This, in addition to the fact that TOP2A amplification is an infrequent event in BC, may limit its clinical utility. Studies on PARP inhibitors and platinum agents are also ongoing in basal-like BC, where traditional chemotherapy performs poorly. Other groups studying material from the EORTC 10994/BIG00–01 clinical trial [103,104] have developed a predictive signature from ER-negative cancers. A 50-gene stromal metagene was found to be predictive of response to chemotherapy. This metagene was significantly associated with the extent of reactive tumor stroma, but not entrapped normal tissue, and stressed the role of the microenvironment in multidrug chemotherapy.

Genetics & radiation response

Radiation has a well-defined application in the locoregional control of BC. Prediction of radiosensitivity, however, has been notoriously difficult, both because of logistical reasons and the lack of appropriate markers. The predictive genetics of radiation response is still in the preclinical phase. Recently, a systems biology approach was developed, and it identified a ten-gene-hub network to model radiation sensitivity [105]; with nine targets demonstrating clinical relevance. The model identified four significant radiosensitivity clusters. Over-represented biological pathways differed between clusters, but included: DNA repair, cell cycle, apoptosis and metabolism. Although still in the nascent stage, simple gene profiles predicting radiosensitivity may be generated from similar studies in the future.

Genetics: integration into routine clinical practice

It is apparent that we currently stand on the verge of integrating individual genetic and genomic information into healthcare provision and maintenance to improve health, increase efficiency and decrease costs. BC management has developed dramatically over time, from surgery alone to systemic cytotoxic chemotherapy. The discovery of the HR, ER and then HER2/neu in BC, and identification of their roles in BC biology and response to therapy, has revolutionized the way BC is managed and ushered in the era of BC genetics and its integration into BC management. As surgery plays a limited role in metastatic and locally advanced BC, ER and HER2 are currently the main determinants of systemic therapy. The discovery of gene expression signatures and multigene assays raised our hope and some assays have now been adopted as standard of care for treating early-stage BC and recommended in BC guidelines by NCCN [201], the St Gallen International Consensus [106] and ASCO. Other area in which BC genetics is of clinical utility is the use of BRCA1/2 mutations status for preventive therapy in high-risk families.

As more genetic tests are introduced in the field of BC, it may be a time to reflect and learn before we move on. Whatever the genetic assay considered, it must be used in the right context, with standard methodology, and intra- and inter-laboratory reproducibility. Currently, most of the recently introduced molecular assays were not tested against a combination of other classical routinely assessed biomarkers to prove or dispute their prognostic superiority. The feasibility and cost of multigene classifiers needs to be considered when compared with surrogate classical tools. It should, however, be acknowledged that these assays in their current stage can potentially provide important prognostic information in clinically indeterminate subgroups, and in such situations, combining this test with conventional predictors yields the most prognostic and predictive information. With the significant increase in systemic treatment options, mammographic screening and the early detection of BC, the proportion of BC that needs additional prognostic and predictive markers increases, which further emphasizes the need for such biological assays to refine treatment decisions and move towards personalized BC management.

Conclusion & future perspective

In the future, we expect a paradigm shift from the traditional concept of using clinicopathological prognostic predictors complemented by certain molecular features to guide BC treatment to the new strategy of using biological and molecular characterization of tumors as the main determinants of management, which is then complemented by clinicopathological variables. This will not only be for early, but also advanced disease.

Research on various genetic tests is still ongoing with emphasis on rigorous clinical validation before routine implementation. Beyond GEP, other genomic technologies such as genomic DNA profiling, epigenomic profiling and miRNA profiling are progressing, and are expected to contribute further towards the discovery of effective biomarkers, novel BC targets and accurate classification of BC. One of the major impediments to integration of genetic/genomic information into healthcare has been the prohibitive high costs. However, as technology improves and the clinical utility becomes proven and widespread, the costs are expected to decrease. Although current evidence supports the prognostic and predictive value for currently available first-generation gene expression signatures, there is still a lot of variation in the outcome of BC that remains unexplained. It is, therefore, hoped that future second-generation gene signatures will provide the fine-tuning needed. Advancements in the understanding of signaling pathways, DNA damage repair mechanisms, tumor microenvironment and host–tumor immune interplay are likely to aid prognostics and therapy prescription. Finally, the most profound gene-based influences in BC clinical practice came from simple tests, such as BRCA testing, HER2 IHC/FISH assessment for targeted therapy and deconstruction of the ER pathway. Genes paved the way to simple proteomic tests and we perceive that this will remain in the future. New gene signatures will help us choose new tests, and the move will be towards the simple, feasible and economical.

Executive summary

Background

Cancer cells suffer from various genetic, epigenetic and postgenetic alterations.

Breast cancer (BC) is a prime example of a heterogeneous genetic disorder.

Global gene expression & BC classification

Molecular taxonomy has classified BC into well-defined classes; however, such classification does not always reflect prognostic behavior.

Global gene expression profiling molecular taxonomy merely portrays hormone receptor status/HER2 status/proliferation status, for which immunohistochemistry remains a clinically simpler approach.

BC & prognostic gene signatures

The most frequently reported and validated prognostic gene assays are the 21-gene signature Oncotype DX®, the 70-gene MammaPrint®, the Breast Cancer Gene Expression Ratio (formerly Aviara H/I^SM) and the 76-gene Rotterdam signature.

Care should be taken in interpretation of results for these assays. The concerns are high-quality material acquisition for testing, lack of definitive incremental contribution over conventional predictors, optimal implementation and relevance to patients receiving current therapies.

No type of assay is definitively superior to the others, and assays should be applied only in the clinical context that they were developed.

Results of randomized studies such as Trial Assigning Individualized Options for Treatment (TAILORx) and Microarray in Node Negative Disease May Avoid Chemotherapy (MINDACT) are awaited.

Genetics & BC treatment

Genetic testing for BRCA1 and BRCA2 mutations are an integral part of clinical practice for women with a strong family history of breast or ovarian cancer. New developments in this category relate to genetic predictors of response to poly(ADP-ribose) polymerase inhibitors and platinum chemotherapy.

For hormonal therapy, complex genetic techniques are rarely used to identify the estrogen receptor status. However, pharmacogenomic markers for response or resistance are being developed.

FISH is regularly used to assay HER2 positivity of tumors. It also indicates where mTOR inhibitors can be applied (in addition to ER positivity).

Topo2α gene amplification has been related to anthracycline response; however, conflicting reports that immunohistochemistry may be a feasible surrogate exist and, therefore, it needs further investigation.

EGFR and VEGF pathway deconstruction, radiation genomics and stromal/immune influences on the tumor microenvironment are future avenues of predictive clinical genomics.

Conclusion & future perspective

Currently, genetic assays help personalized prognosis and therapeutic prediction planning only in a specified context and specified cohorts of BC.

Clinical validation and quality control are imperative before genetic testing becomes the norm.

Costs are currently prohibitive for genetic testing, but will hopefully decline with widespread use.

New gene signatures will bring new simple, feasible and economical tests to the forefront.

Footnotes

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

References

Koeffler

McCormick

Denny

. Molecular mechanisms of cancer. West. J. Med. 155(5), 505–514 (1991).

Canman

Kastan

. Induction of apoptosis by tumor suppressor genes and oncogenes. Semin. Cancer Biol. 6(1), 17–25 (1995).

Sausville

. The challenge of pathway and environment-mediated drug resistance. Cancer Metastasis Rev. 20(1–2), 117–122 (2001).

Sotiriou

Piccart

. Taking gene-expression profiling to the clinic: when will molecular signatures become relevant to patient care? Nat. Rev. Cancer 7(7), 545–553 (2007).

Stokstad

. DNA testing. Genetic screen misses mutations in women at high risk of breast cancer. Science 311 (5769), 1847 (2006).

Pollack

Sorlie

Perou

Microarray analysis reveals a major direct role of DNA copy number alteration in the transcriptional program of human breast tumors. Proc. Natl Acad. Sci. USA 99(20), 12963–12968 (2002).

Reyal

Rouzier

Depont-Hazelzet

The molecular subtype classification is a determinant of sentinel node positivity in early breast carcinoma. PLoaS One 6(5), e20297 (2011).

Rakha

Reis-Filho

Baehner

Breast cancer prognostic classification in the molecular era: the role of histological grade. Breast Cancer Res. 12(4), 207 (2010).

Abba

Sun

Hawkins

Breast cancer molecular signatures as determined by SAGE: correlation with lymph node status. Mol. Cancer Res. 5(9), 881–890 (2007).

10.

Hayes

Bast

Desch

Tumor marker utility grading system: a framework to evaluate clinical utility of tumor markers. J. Natl Cancer Inst. 88(20), 1456–1466 (1996).

11.

van't Veer

Bernards

. Enabling personalized cancer medicine through analysis of gene-expression patterns. Nature 452(7187), 564–570 (2008).

12.

Paik

Tang

Shak

Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer. J. Clin. Oncol. 24(23), 3726–3734 (2006).

13.

Perou

Sorlie

Eisen

Molecular portraits of human breast tumors. Nature 406(6797), 747–752 (2000).

14.

Seminal paper which indicated that gene expression profiling can be applied to breast cancer.

15.

Sorlie

Perou

Tibshirani

Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc. Natl Acad. Sci. USA 98(19), 10869–10874 (2001).

16.

Follow-up paper that validated gene expression profiling for cancers.

17.

Sorlie

Tibshirani

Parker

Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc. Natl Acad. Sci. USA 100(14), 8418–8423 (2003).

18.

Paik

Shak

Tang

A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N. Engl. J. Med. 351(27), 2817–2826 (2004).

19.

Seminal paper for Oncotype DX®.

20.

van't Veer

Dai

van de Vijver

Gene expression profiling predicts clinical outcome of breast cancer. Nature 415 (6871), 530–536 (2002).

21.

Seminal paper for Mammaprint® .

22.

Sotiriou

Wirapati

Loi

Gene expression profiling in breast cancer: understanding the molecular basis of histologic grade to improve prognosis. J. Natl Cancer Inst. 98(4), 262–272 (2006).

23.

Wirapati

Sotiriou

Kunkel

Meta-analysis of gene expression profiles in breast cancer: toward a unified understanding of breast cancer subtyping and prognosis signatures. Breast Cancer Res. 10(4), R65 (2008).

24.

Buyse

Loi

van't Veer

Validation and clinical utility of a 70-gene prognostic signature for women with node-negative breast cancer. J. Natl Cancer Inst. 98(17), 1183–1192 (2006).

25.

Dowsett

Cuzick

Wale

Prediction of risk of distant recurrence using the 21-gene recurrence score in node-negative and node-positive postmenopausal patients with breast cancer treated with anastrozole or tamoxifen: a TransATAC study. J. Clin. Oncol 28(11), 1829–1834 (2010).

26.

Hall

Ptacek

Snyder

. Protein microarray technology. Mech. Ageing Dev. 128(1), 161–167 (2007).

27.

Abd El-Rehim

Ball

Pinder

High-throughput protein expression analysis using tissue microarray technology of a large well-characterized series identifies biologically distinct classes of breast cancer confirming recent cDNA expression analyses. Int. J. Cancer 116(3), 340–350 (2005).

28.

Sotiriou

Pusztai

. Gene-expression signatures in breast cancer. N. Engl. J. Med. 360(8), 790–800 (2009).

29.

Rakha

Reis-Filho

Ellis

. Basal-like breast cancer: a critical review. J. Clin. Oncol. 26(15), 2568–2581 (2008).

30.

Carey

Perou

Livasy

Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 295(21), 2492–2502 (2006).

31.

Kim

Ahn

Kim

Gong

. Clinicopathologic significance of the basal-like subtype of breast cancer: a comparison with hormone receptor and HER2/neu-overexpressing phenotypes. Hum. Pathol. 37(9), 1217–1226 (2006).

32.

Rakha

Reis-Filho

Ellis

. Combinatorial biomarker expression in breast cancer. Breast Cancer Res. Treat. 120(2), 293–308 (2010).

33.

Cuzick

Dowsett

Pineda

Prognostic value of a combined estrogen receptor, progesterone receptor, Ki-67, and human epidermal growth factor receptor 2 immunohistochemical score and comparison with the genomic health recurrence score in early breast cancer. J. Clin. Oncol. 29(32), 4273–4278 (2011).

34.

van de Vijver

van't Veer

A gene-expression signature as a predictor of survival in breast cancer. N. Engl. J. Med. 347(25), 1999–2009 (2002).

35.

Seminal paper of the prognostic value of gene expression signatures .

36.

Kelly

Krishnamurthy

Bianchini

Utility of oncotype DX risk estimates in clinically intermediate risk hormone receptor-positive, HER2-normal, grade II, lymph node-negative breast cancers. Cancer 116(22), 5161–5167 (2010).

37.

Marchionni

Wilson

Wolff

Systematic review: gene expression profiling assays in early-stage breast cancer. Ann. Intern. Med. 148(5), 358–369 (2008).

38.

Goetz

Suman

Ingle

A two-gene expression ratio of homeobox 13 and interleukin 17B receptor for prediction of recurrence and survival in women receiving adjuvant tamoxifen. Clin. Cancer Res. 12(7 Pt 1), 2080–2087 (2006).

39.

Wang

Ryan

A two-gene expression ratio predicts clinical outcome in breast cancer patients treated with tamoxifen. Cancer Cell 5(6), 607–616 (2004).

40.

Wang

Klijn

Zhang

Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet 365(9460), 671–679 (2005).

41.

Liu

Wang

Chen

The prognostic role of a gene signature from tumorigenic breast-cancer cells. N. Engl. J. Med. 356(3), 217–226 (2007).

42.

Chang

Nuyten

Sneddon

Robustness, scalability, and integration of a wound-response gene expression signature in predicting breast cancer survival. Proc. Natl Acad. Sci. USA 102(10), 3738–3743 (2005).

43.

Chi

Wang

Nuyten

Gene expression programs in response to hypoxia: cell type specificity and prognostic significance in human cancers. PLoS Med. 3(3), e47 (2006).

44.

Ahr

Holtrich

Solbach

Molecular classification of breast cancer patients by gene expression profiling. J. Pathol. 195(3), 312–320 (2001).

45.

Naoi

Kishi

Tanei

Development of 95-gene classifier as a powerful predictor of recurrences in node-negative and ER-positive breast cancer patients. Breast Cancer Res. Treat. 128(3), 633–641 (2011).

46.

Jerevall

Prognostic utility of HOXB13:IL17BR and molecular grade index in early-stage breast cancer patients from the Stockholm trial. Br. J. Cancer 104(11), 1762–1769 (2011).

47.

Habel

Shak

Jacobs

A population-based study of tumor gene expression and risk of breast cancer death among lymph node-negative patients. Breast Cancer Res. 8(3), R25 (2006).

48.

Chang

Makris

Gutierrez

Gene expression patterns in formalin-fixed, paraffin-embedded core biopsies predict docetaxel chemosensitivity in breast cancer patients. Breast Cancer Res. Treat. 108(2), 233–240 (2008).

49.

Salunga

Dahiya

A five-gene molecular grade index and HOXB13:IL17BR are complementary prognostic factors in early stage breast cancer. Clin. Cancer Res. 14(9), 2601–2608 (2008).

50.

Teschendorff

Miremadi

Pinder

Ellis

Caldas

. An immune response gene expression module identifies a good prognosis subtype in estrogen receptor negative breast cancer. Genome Biol. 8(8), R157 (2007).

51.

Desmedt

Haibe-Kains

Wirapati

Biological processes associated with breast cancer clinical outcome depend on the molecular subtypes. Clin. Cancer Res. 14(16), 5158–5165 (2008).

52.

Finak

Bertos

Pepin

Stromal gene expression predicts clinical outcome in breast cancer. Nat. Med. 14(5), 518–527 (2008).

53.

Gravier

Pierron

Vincent-Salomon

A prognostic DNA signature for TlT2 node-negative breast cancer patients. Genes Chromosomes Cancer 49(12), 1125–1134 (2010).

54.

Oakman

Santarpia

Di Leo

. Breast cancer assessment tools and optimizing adjuvant therapy. Nat. Rev. Clin. Oncol. 7(12), 725–732 (2010).

55.

Albain

Barlow

Shak

Prognostic and predictive value of the 21-gene recurrence score assay in postmenopausal women with node-positive, oestrogen-receptor-positive breast cancer on chemotherapy: a retrospective analysis of a randomised trial. Lancet Oncol. 11(1), 55–65 (2010).

56.

Jacquemier

Ginestier

Rougemont

Protein expression profiling identifies subclasses of breast cancer and predicts prognosis. Cancer Res. 65(3), 767–779 (2005).

57.

Cobleigh

Tabesh

Bitterman

Tumor gene expression and prognosis in breast cancer patients with 10 or more positive lymph nodes. Clin. Cancer Res. 11(24 Pt 1), 8623–8631 (2005).

58.

Toi

Iwata

Yamanaka

Clinical significance of the 21-gene signature (Oncotype DX) in hormone receptor-positive early stage primary breast cancer in the Japanese population. Cancer 116(13), 3112–3118 (2010).

59.

BlueCross BlueShield Association (BCBSA), Technology Evaluation Center (TEC). Gene Expression Profiling in Women with Lymph-Node-Positive Breast Cancer to Select Adjuvant Chemotherapy Treatment. Blue Cross and Blue Shield Association, Chicago, IL, USA (2010).

60.

BlueCross BlueShield Association (BCBSA), Technology Evaluation Center (TEC). Gene Expression Profiling for Managing Breast Cancer Treatment. Blue Cross and Blue Shield Association, Chicago, IL, USA, 22 (2008).

61.

ECRI Institute. Gene Expression Profiling of Breast Cancer to Predict the Likelihood of Recurrence. Emerging technology evidence report. ECRI Institute, PA, USA (2008).

62.

Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group. Recommendations from the EGAPP Working Group: can tumor gene expression profiling improve outcomes in patients with breast cancer?

Genet. Med. 11(1), 66–73 (2009).

63.

Sparano

Paik

. Development of the 21-gene assay and its application in clinical practice and clinical trials. J. Clin. Oncol. 26(5), 721–728 (2008).

64.

Remits of randomized trial TAILORx for Oncotype DX .

65.

Cardoso

van't Veer

Rutgers

Loi

Mook

Piccart-Gebhart

. Clinical application of the 70-gene profile: the MINDACT trial. J. Clin. Oncol. 26(5), 729–735 (2008).

66.

Remits of randomized trial MINDACT for Mammaprint .

67.

Farmer

McCabe

Lord

Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 434(7035), 917–921 (2005).

68.

Important article for gene expression and cancer prevention .

69.

Verhoog

Brekelmans

Seynaeve

Survival and tumor characteristics of breast-cancer patients with germline mutations of

BRCA 1. Lancet 351(9099), 316–321 (1998).

70.

Important article for gene expression and cancer prevention .

71.

McCabe

Turner

Lord

Deficiency in the repair of DNA damage by homologous recombination and sensitivity to poly(ADP-ribose) polymerase inhibition. Cancer Res. 66(16), 8109–8115 (2006).

72.

Fong

Boss

Yap

Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N. Engl. J. Med. 361(2), 123–134 (2009).

73.

Sirohi

Arnedos

Popat

Platinum-based chemotherapy in triple-negative breast cancer. Ann. Oncol. 19(11), 1847–1852 (2008).

74.

Fourquet

Stoppa-Lyonnet

Kirova

Sigal-Zafrani

Asselain

. Familial breast cancer: clinical response to induction chemotherapy or radiotherapy related to BRCA1/2 mutations status. Am. J. Clin. Oncol. 32(2), 127–131 (2009).

75.

berg

Rotten S

Nygren

Pajic

Selective induction of chemotherapy resistance of mammary tumors in a conditional mouse model for hereditary breast cancer. Proc. Natl Acad. Sci. USA 104 (29), 12117–12122 (2007).

76.

Lakhani

Reis-Filho

Fulford

Prediction of BRCA1 status in patients with breast cancer using estrogen receptor and basal phenotype. Clin. Cancer Res. 11(14), 5175–5180 (2005).

77.

Collins

Martyniak

Kandel

Basal cytokeratin and epidermal growth factor receptor expression are not predictive of BRCA1 mutation status in women with triple-negative breast cancers. Am. J. Surg. Pathol. 33(7), 1093–1097 (2009).

78.

Tutt

Robson

Garber

Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and advanced breast cancer: a proof-of-concept trial. Lancet 376(9737), 235–244 (2010).

79.

Pfeifer

Hainaut

. Next-generation sequencing: emerging lessons on the origins of human cancer. Curr. Opin. Oncol. 23(1), 62–68 (2010).

80.

Easton

Pooley

Dunning

Genome-wide association study identifies novel breast cancer susceptibility loci. Nature 447, 1087–1093 (2007).

81.

Castro

Osorio

Torres

Evidence that molecular changes in cells occur before morphological alterations during the progression of breast ductal carcinoma. Breast Cancer Res. 10(5), R87 (2008).

82.

Schuetz

Bonin

Clare

Progression-specific genes identified by expression profiling of matched ductal carcinomas in situ and invasive breast tumors, combining laser capture microdissection and oligonucleotide microarray analysis. Cancer Res. 66(10), 5278–5286 (2006).

83.

Anderson

Chatterjee

Ershler

Brawley

. Estrogen receptor breast cancer phenotypes in the surveillance, epidemiology, and end results database. Breast Cancer Res. Treat. 76(1), 27–36 (2002).

84.

Higgins

Stearns

. Pharmacogenetics of endocrine therapy for breast cancer. Ann. Rev. Med. 62, 281–293 (2011).

85.

Rakha

El-Sayed

Green

Biologic and clinical characteristics of breast cancer with single hormone receptor positive phenotype. J. Clin. Oncol. 25(30), 4772–4778 (2007).

86.

Habashy

Powe

Glaab

RERG (Ras-like, oestrogen-regulated, growth-inhibitor) expression in breast cancer: a marker of ER-poskive luminal-like subtype. Breast Cancer Res. Treat. 128(2), 315–326 (2010).

87.

Habashy

Powe

Rakha

Forkhead-box Al (EOXA1) expression in breast cancer and its prognostic significance. Eur. J. Cancer 44(11), 1541–1551 (2008).

88.

Witton

Reeves

Going

Cooke

Bartlett

. Expression of the HER1–4 family of receptor tyrosine kinases in breast cancer. J. Pathol. 200(3), 290–297 (2003).

89.

King

Kraus

Aaronson

. Amplification of a novel v-erbB-related gene in a human mammary carcinoma. Science 229(4717), 974–976 (1985).

90.

Wolff

Hammond

Schwartz

American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J. Clin. Oncol. 25(1), 118–145 (2007).

91.

Guidelines for HER2 testing currently in the clinic .

92.

Slamon

Clark

Wong

Levin

Ullrich

McGuire

. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235 (4785), 177–182 (1987).

93.

Shah

Chen

. Testing for HER2 in breast cancer: a continuing evolution. Patholog. Res. Int. 2011, 903202 (2011).

94.

Mukherjee

Dhadda

Shehata

Chan

. Lapatinib: a tyrosine kinase inhibitor with a clinical role in breast cancer. Expert Opin. Pharmacother. 8(13), 2189–2204 (2007).

95.

Saal

Holm

Maurer

. PIK3CA mutations correlate with hormone receptors, node metastasis, and ERBB2, and are mutually exclusive with PTEN loss in human breast carcinoma. Cancer Res. 65(7), 2554–2559 (2005).

96.

Johnston

. Are we missing the mTOR target in breast cancer? Breast Cancer Res. Treat. 128(3), 607–611 (2011).

97.

Margariti

Fox

Bottini

Generali

. ‘Overcoming breast cancer drug resistance with mTOR inhibitors’. Could k be a myth or a real possibility in the short-term future? Breast Cancer Res. Treat. 128(3), 599–606 (2011).

98.

Noh

Mondesire

Peng

Determinants of rapamycin sensitivity in breast cancer cells. Clin. Cancer Res. 10(3), 1013–1023 (2004).

99.

Kurebayashi

Okubo

Yamamoto

Sonoo

. Inhibition of HER1 signaling pathway enhances antitumor effect of endocrine therapy in breast cancer. Breast Cancer 11(1), 38–41 (2004).

100.

Tsutsui

Ohno

Murakami

Hachhanda

Oda

. Prognostic value of epidermal growth factor receptor (EGFR) and its relationship to the estrogen receptor status in 1029 patients with breast cancer. Breast Cancer Res. Treat. 71(1), 67–75 (2002).

101.

Reis-Filho

Pinheiro

Lambros

EGFR amplification and lack of activating mutations in metaplastic breast carcinomas. J. Pathol. 209(4), 445–453 (2006).

102.

Agrawal

Gutteridge

Gee

Nicholson

Robertson

. Overview of tyrosine kinase inhibitors in clinical breast cancer. Endocr. Relat. Cancer 12(Suppl. 1), S135–S144 (2005).

103.

Smith

Walsh

Skene

A Phase II placebo-controlled trial of neoadjuvant anastrozole alone or with gefitinib in early breast cancer. J. Clin. Oncol. 25(25), 3816–3822 (2007).

104.

Finn

Press

Dering

Estrogen receptor, progesterone receptor, human epidermal growth factor receptor 2 (HER2), and epidermal growth factor receptor expression and benefit from lapatinib in a randomized trial of paclitaxel with lapatinib or placebo as first-line treatment in HER2-negative or unknown metastatic breast cancer. J. Clin. Oncol. 27(24), 3908–3915 (2009).

105.

Eppenberger

Kueng

Schlaeppi

Markers of tumor angiogenesis and proteolysis independently define high- and low-risk subsets of node-negative breast cancer patients. J. Clin. Oncol. 16(9), 3129–3136 (1998).

106.

Hochgrafe

Zhang

O'Toole

Tyrosine phosphorylation profiling reveals the signaling network characteristics of basal breast cancer cells. Cancer Res. 70(22), 9391–9401 (2010).

107.

Huang

Gao

Wang

Lee

Ordentlich

Liu

. HDAC inhibitor SNDX-275 induces apoptosis in erbB2- overexpressing breast cancer cells via down-regulation of erbB3 expression. Cancer Res. 69(21), 8403–8411 (2009).

108.

Singh

Shankar

Srivastava

. HDAC inhibitors enhance the apoptosis-inducing potential of TRAIL in breast carcinoma. Oncogene 24(29), 4609–4623 (2005).

109.

Martin

Romero

Cheang

Genomic predictors of response to doxorubicin versus docetaxel in primary breast cancer. Breast Cancer Res. Treat. 128(1), 127–136 (2011).

110.

Burgess

Doles

Zender

Topoisomerase levels determine chemotherapy response in vitro and

in vivo. Proc. Natl Acad. Sci. USA 105(26), 9053–9058 (2008).

111.

Desmedt

Di Leo

de Azambuja

Multifactorial approach to predicting resistance to anthracyclmes. J. Clin. Oncol. 29(12), 1578–1586 (2011).

112.

Oakman

Moretti

Di Leo

. Researching anthracycline therapy. Breast Cancer Res. Treat. 123(1), 171–175 (2010).

113.

Farmer

Bonnefoi

Anderle

A stroma-related gene signature predicts resistance to neoadjuvant chemotherapy in breast cancer. Nat. Med. 15(1), 68–74 (2009).

114.

Bonnefoi

Potti

Delorenzi

Validation of gene signatures that predict the response of breast cancer to neoadjuvant chemotherapy: a substudy of the EORTC 10994/BIG 00–01 clinical trial. Lancet Oncol. 8(12), 1071–1078 (2007).

115.

Eschrich

Zhang

Zhao

Systems biology modeling of the radiation sensitivity network: a biomarker discovery platform. Int. J. Radiat. Oncol. Biol. Phys. 75(2), 497–505 (2009).

116.

Goldhirsch

Ingle

Gelber

Coates

Thurlimann

Senn

. Thresholds for therapies: highlights of the St Gallen International Expert Consensus on the primary therapy of early breast cancer 2009. Ann. Oncol. 20(8), 1319–1329 (2009).

117.

NCCN website. NCCN Updates Breast Cancer Guidelines. www.nccn.org/about/news/newsinfo.asp?NewsID=127

Integrating Breast Cancer Genetics into Clinical Practice

Abstract

Keywords

BC heterogeneity & clinical management

Biological prognostic & predictive variables

BC & gene expression

Global gene expression & BC classification (molecular taxonomy)

BC & prognostic gene signatures

Gene expression arrays: comparison & clinical utility

Genetics & BC treatment

Genetics & cancer prevention

Genetics & hormonal therapy

Genetics & targeted therapy

HER2

Other targets

Genetics & chemotherapy

Genetics & radiation response

Genetics: integration into routine clinical practice

Conclusion & future perspective

Footnotes

References