The HER-2 Receptor and Breast Cancer: Ten Years of Targeted Anti–HER-2 Therapy and Personalized Medicine

Learning Objectives

1. Contrast the current strengths and limitations of the three main slide-based techniques (IHC, FISH, and CISH) currently in clinical use for testing breast cancer tissues for HER-2 status.

2. Compare the efficacy of trastuzumab- and lapatinib-based regimens in the adjuvant and metastatic settings as reported in published clinical trials and regulatory approval databases.

3. Contrast the list of biomarkers that have been associated with clinical resistance to trastuzumab and lapatinib and describe their current level of validation.

This article is available for continuing medical education credit at CME.TheOncologist.com.

Abstract

The human epidermal growth factor receptor (HER-2) oncogene encodes a transmembrane tyrosine kinase receptor that has evolved as a major classifier of invasive breast cancer and target of therapy for the disease. The validation of the general prognostic significance of HER-2 gene amplification and protein overexpression in the absence of anti–HER-2 targeted therapy is discussed in a study of 107 published studies involving 39,730 patients, which produced an overall HER-2–positive rate of 22.2% and a mean relative risk for overall survival (OS) of 2.74. The issue of HER-2 status in primary versus metastatic breast cancer is considered along with a section on the features of metastatic HER-2–positive disease. The major marketed slide-based HER-2 testing approaches, immunohistochemistry, fluorescence in situ hybridization, and chromogenic in situ hybridization, are presented and contrasted in detail against the background of the published American Society of Clinical Oncology–College of American Pathologists guidelines for HER-2 testing. Testing issues, such as the impact of chromosome 17 polysomy and local versus central HER-2 testing, are also discussed. Emerging novel HER-2 testing techniques, including mRNA-based testing by real-time polymerase chain reaction and DNA microarray methods, HER-2 receptor dimerization, phosphorylated HER-2 receptors, and HER-2 status in circulating tumor cells, are also considered. A series of biomarkers potentially associated with resistance to trastuzumab is discussed with emphasis on the phosphatase and tensin homologue deleted on chromosome ten/Akt and insulin-like growth factor receptor pathways. The efficacy results for the more recently approved small molecule HER-1/HER-2 kinase inhibitor lapatinib are also presented along with a more limited review of markers of resistance for this agent. Additional topics in this section include combinations of both anti–HER-2 targeted therapies together as well as with novel agents including bevacizumab, everolimus, and tenespimycin. A series of novel HER-2–targeting agents is also presented, including pertuzumab, ertumaxomab, HER-2 vaccines, and recently discovered tyrosine kinase inhibitors. Biomarkers predictive of HER-2 targeted therapy toxicity are included, and the review concludes with a consideration of HER-2 status in the prediction of response to non–HER-2 targeted treatments including hormonal therapy, anthracyclines, and taxanes.

Section One: Biology, Pathology, Diagnosis, and Clinical Significance of HER-2–Positive Breast Cancer

Introduction and Background Biology

The human epidermal growth factor receptor 2 (HER-2, HER-2/neu, c-erbB-2) gene, first discovered in 1984 by Weinberg and associates [1], is localized to chromosome 17q and encodes a transmembrane tyrosine kinase receptor protein that is a member of the epidermal growth factor receptor (EGFR) or HER family (Fig. 1) [2]. This family of receptors is involved in cell–cell and cell–stroma communication primarily through a process known as signal transduction, in which external growth factors, or ligands, affect the transcription of various genes, by phosphorylating or dephosphorylating a series of transmembrane proteins and intracellular signaling intermediates, many of which possess enzymatic activity. Signal propagation occurs as the enzymatic activity of one protein turns on the enzymatic activity of the next protein in the pathway [3]. Major pathways involved in signal transduction, including the Ras/mitogen-activated protein kinase pathway, the phosphatidylinositol 3′ kinase (PI3K)/Akt pathway, the Janus kinase/signal transducer and activator of transcription pathway, and the phospholipase Cγ pathway, ultimately affect cell proliferation, survival, motility, and adhesion.

Receptor activation requires three variables, a ligand, a receptor, and a dimerization partner [4]. After a ligand binds to a receptor, that receptor must interact with another receptor of identical or related structure in a process known as dimerization in order to trigger phosphorylation and activate signaling cascades. Therefore, after ligand binding to an EGFR family member, the receptor can dimerize with various members of the family (EGFR, HER-2, HER-3, or HER-4). It may dimerize with a like member of the family (homodimerization) or it may dimerize with a different member of the family (heterodimerization). The specific tyrosine residues on the intracellular portion of the HER-2/neu receptor that are phosphorylated, and hence the signaling pathways that are activated, depend on the ligand and dimerization partner. The wide variety of ligands and intracellular crosstalk with other pathways allow for significant diversity in signaling. Although no known ligand for the HER-2 receptor has been identified, it is the preferred dimerization partner of the other family members. HER-2 heterodimers are more stable [5, 6] and their signaling is more potent [7] than receptor combinations without HER-2.

HER-2 gene amplification and/or protein overexpression has been identified in 10%–34% of invasive breast cancers [1]. Unlike a variety of other epithelial malignancies, in breast cancer, HER-2 gene amplification is uniformly associated with HER-2 (p185neu) protein overexpression and the incidence of single copy overexpression is exceedingly rare [8]. HER-2 gene amplification in breast cancer has been associated with increased cell proliferation, cell motility, tumor invasiveness, progressive regional and distant metastases, accelerated angiogenesis, and reduced apoptosis [9]. When classified by routine clinicopathologic parameters and compared with HER-2–negative tumors, HER-2–positive breast cancer is more often of intermediate or high histologic grade, more often lacking estrogen receptors (ERs) and progesterone receptors (PgRs) (ER and PgR negative), and featuring positive lymph node metastases at presentation [1]. In the recent molecular classification of breast cancer, positive HER-2 status does not constitute a unique molecular category and is identified in both the “HER-2” and “luminal” tumor classes [10].

HER-2 Status and Prognosis in Breast Cancer

Both morphology-based and molecular-based techniques have been used to measure HER-2/neu status in breast cancer clinical samples [11–117]. By a substantial majority, abnormalities in HER-2 expression at the gene, message, or protein level have been associated with adverse prognosis in both lymph node–negative and lymph node–positive breast cancer. Of the 107 studies considering 39,730 patients listed in Table 1, 95 (88%) of the studies determined that either HER-2 gene amplification or HER-2 (p185 neu) protein overexpression predicted breast cancer outcome on either univariate or multivariate analysis. In 68 (73%) of the 93 studies that featured multivariate analysis of outcome data, the adverse prognostic significance of HER-2 gene, message, or protein overexpression was independent of all other prognostic variables. In only 13 (12%) of the studies, no correlation between HER-2 status and clinical outcome was identified. Of these 13 noncorrelating studies, eight (62%) used immunohistochemistry (IHC) on paraffin-embedded tissues as the HER-2/protein detection technique, two (15%) used fluorescence in situ hybridization (FISH), two (15%) used Southern analysis, and one (7%) used a real-time polymerase chain reaction (RT-PCR) technique. Of the 15 studies that used the FISH technique, 13 (87%) showed univariate prognostic significance of gene amplification, and 11 of these (85%) showed prognostic significance on multivariate analysis as well. The two studies that used chromogenic in situ hybridization (CISH) HER-2 gene amplification detection techniques both found that HER-2 amplification was an independent predictor of outcome on multivariate analysis [100, 112]. However, interpretation of these studies is complicated by the fact that most studies included patients who received variable types of systemic adjuvant therapy; therefore, the pure prognostic value of HER-2 overexpression in the absence of any systemic adjuvant therapy is incompletely understood.

HER-2 Positivity Rates

The frequency of HER-2 positivity in all of the studies presented in Table 1 was 22.2%, with a range of 9%–74%. The HER-2–positive rate was similar for IHC, at 22% (range, 10%–74%), and FISH, at 23.9% (range, 14.7%–68%). In current practice, HER-2–positive rates have trended below 20%, with most investigators currently reporting that the true positive rate is in the range of 15%–20%. The HER-2–positive rate may be higher when metastatic lesions are tested, and tertiary hospitals and cancer centers report slightly higher rates than community hospitals and national reference laboratories.

Relative Risk and Hazard Ratio

In Table 1, a number of studies provided data as to the relative risk (RR) of untreated HER-2–positive breast cancer being associated with an adverse clinical outcome. For OS, the mean RR was 2.74 (range, 1.39–6.93) and the median was 2.33; for disease-free survival (DFS), the mean RR was 2.04 (range, 1.30–3.01) and the median was 1.8. In several studies, the RR was estimated with a hazard ratio (HR) model. The mean HR was 2.12 (range, 1.6–2.7) and the median was 2.08.

HER-2 Expression and Breast Pathology

The association of HER-2–positive status with specific pathologic conditions of the breast is summarized in Table 2. HER-2 overexpression has been consistently associated with higher grades and extensive forms of ductal carcinoma in situ (DCIS) and DCIS featuring comedo-type necrosis [118–121]. The incidence of HER-2 positivity in DCIS has varied in the range of 24%–38% in the published literature, which appears to be slightly higher than that for invasive breast cancer [118–121]. Routine testing for HER-2 status in DCIS is not widely performed. However, should anti–HER-2 targeted therapies directed at HER-2–positive DCIS result in a reduction in the development of invasive disease, the widespread use of HER-2 testing in DCIS would be adopted. Finally, the invasive carcinoma that develops in association with HER-2–positive DCIS may, on occasion, not feature a HER-2–positive status, a finding that has led investigators to believe that HER-2 gene amplification may not be required for the local progression of breast cancer [122]. Compared with invasive ductal carcinoma (IDC), HER-2 gene amplification occurs at a significantly lower rate in invasive lobular carcinoma (ILC) (<10%), but has also been linked to an adverse outcome [85]. HER-2 positivity is linked exclusively to the pleomorphic variant of ILC and is not encountered in classic ILC [123]. HER-2 amplification is strongly correlated with tumor grade in both IDC and ILC. For example, in one study, only one of 73 grade I IDC cases and one of 67 low-grade classic ILC cases showed HER-2 amplification detected by FISH [86]. HER-2 overexpression and HER-2 amplification have been a consistent feature of both mammary and extramammary Paget’s disease [124, 125] (Fig. 2). HER-2 amplification and HER-2 overexpression have been associated with adverse outcome in some studies of male breast carcinoma [126–129], but not in others [130–132]. The incidence of HER-2 positivity appears to be lower in male breast cancer than in female breast cancer [126–132]. Documented responses in male breast cancer to HER-2–targeting agents have been described, and therefore treatment with trastuzumab is an acceptable option for these patients, but the true activity rate remains uncertain [133]. The rate of HER-2 overexpression in mucinous (colloid) breast cancers is extremely low, although, on occasion, it has been associated with aggressive disease [134–136]. In medullary breast carcinoma, HER-2 testing has consistently found negative results [137]. Similarly, HER-2 positivity is extremely rare in cases of tubular carcinoma [138]. HER-2 status has not been consistently linked to the presence of inflammatory breast cancer [139, 140]. Molecular studies of hereditary breast cancer including cases with either BRCA1 or BRCA2 germline mutations have found a consistently lower incidence of HER-2–positive status for these tumors [141]. Breast sarcomas and phyllodes tumors have consistently been HER-2 negative [142]. Finally, low-level HER-2/neu overexpression has been identified in benign breast disease biopsies and is associated with a greater risk for subsequent invasive breast cancer [143].

HER-2 Status in Primary Versus Metastatic Breast Cancer

The majority of studies that have compared the HER-2 status in paired primary and metastatic tumor tissues have found an overwhelming consistency in the patient’s status regardless of the method of testing (IHC versus FISH) [144–151]. However, several recent studies indicated 20%–30% discordance rates between the HER-2 status of primary and metastatic lesions. Some of these studies have featured relatively high HER-2–positive rates on both paired specimens (>35% positive), which has created concern about the conclusions of these reports [152]. Also, considering that 10%–30% discordance rates have been reported even when the same tumor is tested repeatedly, it remains uncertain if the discordance rates seen between primary and metastatic sites is higher than expected by the less than perfect reproducibility of the various HER-2 assays. Increasingly, emerging data suggest that there are changes in HER-2 expression between primary and metastatic disease. This is particularly true after intervening HER-2–directed therapy, but also happens in the absence of such treatment. In cases where the original primary HER-2 test result is questioned because of technical or interpretive issues and in patients where there has been an unusually long (i.e., >5-year) interval between the primary occurrence and the detection of metastatic disease, retesting of a metastatic lesion may be warranted. Thus, although routine HER-2 testing of metastatic disease is advocated by some investigators, the preponderance of data indicates that the HER-2 status remains stable and that routine retesting of HER-2 may not be needed for most patients with metastatic disease.

Features of Metastatic HER-2–Positive Breast Cancer

Metastatic HER-2–positive breast cancer retains the phenotype of the primary tumor not only in HER-2 status, but also is typically ER/PgR negative, moderate to high tumor grade, DNA aneuploid with high S phase fraction, and featuring ductal rather than lobular histology. In the era prior to the initiation of HER-2–targeted therapy, HER-2–positive breast cancer was more likely to spread early to major visceral sites including the axillary lymph nodes, bone marrow, lungs, liver, adrenal glands, and ovaries [153]. In the post–HER-2 targeted therapy era, the incidence of progressive visceral metastatic disease in HER-2–positive tumors has diminished and has frequently been superseded by the development of clinically significant central nervous system (CNS) metastatic disease [154–157]. It is widely held that the success in the control of visceral disease with trastuzumab has unmasked previously occult CNS disease and, because of the inability of the therapeutic antibody to cross the blood–brain barrier, allowed brain metastases to progress during the extended OS duration of treated patients [154, 155]. The small-molecule drug lapatinib has shown some promise for targeting HER-2–positive CNS metastases that are resistant to trastuzumab-based therapies in initial studies [158].

Interaction of HER-2 Expression with Other Prognosis Variables

HER-2 gene amplification and protein overexpression have been associated consistently with high tumor grade, DNA aneuploidy, high cell proliferation rate, negative assays for nuclear protein receptors for estrogen and progesterone, p53 mutation, topoisomerase IIa amplification, and alterations in a variety of other molecular biomarkers of breast cancer invasiveness and metastasis [159–161].

HER-2 Testing Techniques

A series of morphology-driven, slide-based assays designed to detect HER-2 amplification and HER-2 overexpression and a group of in vitro laboratory HER-2 diagnostics performed on breast tumors and peripheral blood are summarized in Table 3.

Slide-Based Assays

IHC.

IHC was used as the clinical trial assay (CTA) in the phase III trial that led to the U.S. Food and Drug Administration (FDA) approval of trastuzumab for the treatment of HER-2–overexpressing metastatic breast cancer (MBC). IHC staining remains the most frequent initial test for HER-2 status and is performed on approximately 80% of newly diagnosed breast cancers in the U.S. Unlike most IHC assays, the assessment of HER-2 status is quantitative rather than qualitative, because HER-2 is expressed in all breast epithelial cells. In order to provide a meaningful interpretation of a HER-2 immunostain, it was necessary to establish a relationship between the number of HER-2 receptors on a cell’s surface and the distribution and intensity of the immunostain (Fig. 3A). Using cell lines, it was possible to establish a standardized IHC procedure and scoring system in which cells containing <20,000 receptors would show no staining (0), cells containing approximately 100,000 receptors would show partial membrane staining with <10% of the cells showing complete membrane staining (1+), cells containing approximately 500,000 receptors would show light to moderate complete membrane staining in >10% of the cells (2+), and cells containing approximately 2,300,000 receptors would show strong, complete membrane staining in >10% of the cells (3+). Studies have shown that when a standardized IHC assay is performed on specimens that are carefully fixed, processed, and embedded, there is good to excellent correlation between gene copy status and protein expression levels [162–165]. However, the ability to accurately determine HER-2 protein expression status by IHC can be significantly impacted by technical issues accentuated by the tissue fixation in formaldehyde, tissue processing, and embedding procedure in heated paraffin wax. Advantages of IHC testing include its wide availability, relatively low cost, easy preservation of stained slides, and use of a familiar routine microscope. Disadvantages of IHC include the impact of preanalytic issues including storage, duration and type of fixation, intensity of antigen retrieval, type of antibody (polyclonal versus monoclonal), lack of a positive internal control signal, variability in system control samples, and, most importantly, the difficulties in applying a semiquantitative subjective slide-scoring system. The IHC detection rates for HER-2 protein can vary considerably based on the antibody chosen, as shown in a study using a large tissue block containing multiple breast tumors [166]. Problems with IHC standardization in slide scoring have been emphasized in studies of patient response to trastuzumab [167]. Slide scoring can be improved by avoiding overinterpretation of specimen edges, retraction artifacts, under- or overfixation artifacts, cases with significant staining of benign ductal and lobular cells, cytoplasmic tumor cell staining, and membranous tumor cell staining that lacks a complete circumferential staining pattern (the so-called “chicken wire” appearance). Data presented by the National Surgical Adjuvant Breast and Bowel Project (NSABP) initially favored the idea that laboratories performing high-volume HER-2 testing produced a higher concordance between IHC and FISH results (approaching 98% interlaboratory concordance when tumors assessed as 3+ were reanalyzed by both IHC and FISH) when compared with central laboratory testing at the NSABP laboratory [168]. Because most of the submitting laboratories were reference laboratories that cannot control tissue fixation or storage, it has been suggested that preanalytical issues may not be the major cause of interlaboratory variability. In the United Kingdom, it has been recommended that HER-2 testing be restricted to laboratories undertaking an annual minimum of 250 IHC tests (and/or 100 FISH tests) [169]. Results from the United Kingdom National External Quality Assessment Scheme for Immunohistochemistry also suggested that the lack of reproducibility of HER-2 scoring between laboratories was not the result of tumor heterogeneity or differences in fixation or processing but rather the result of how the scoring system was applied [170]. The use of a quantitative image analysis system can reduce slide-scoring variability among pathologists, especially in 2+ cases [171]. When 130 HER-2–immunostained slides were reviewed by 10 pathologists and then were later reviewed with the aid of image analysis, the use of image analysis eliminated most of the interobserver variability that was significant by routine microscopy [172]. Thus, in routine clinical practice, errors in HER-2 testing by the IHC method are caused by both variables associated with antigen retrieval and the reagents and staining protocol and variation in the actual slide scoring [173, 174]. Two commercially available HER-2 IHC kits, the Dako HercepTest™ (Dako Corporation, Glostrup, Denmark) and the Ventana Pathway™ (Ventana Medical Systems, Tucson, AZ), are approved by the FDA for determining the eligibility of patients to receive trastuzumab therapy (Table 3).

FISH.

The FISH technique (Fig. 3B), like IHC, is a morphology-driven slide-based DNA hybridization assay using fluorescent-labeled probes. Both the hybridization steps and the slide scoring can be automated. FISH has the advantages of a more objective scoring system and the presence of a built-in internal control consisting of the two HER-2 gene signals present both in benign cells and in malignant cells that do not feature HER-2 gene amplification. The disadvantages of FISH testing include the higher cost of each test, longer time required for slide scoring, requirement of a fluorescent microscope, inability to preserve the slides for storage and review, and greater difficulty in assessing background morphology such as in distinguishing in situ from invasive tumor. Three versions of the FISH assay are FDA approved. The single-probe Ventana Inform™ test (Ventana Medical Systems) that measures only HER-2 gene copies is approved as a prognostic test. The two dual-probe (HER-2 probe plus chromosome 17 centromere probe) kits, the Abbott-Vysis PathVysion™ test (Abbott Laboratories. Abbott Park, IL) and the Dako Cytomation Her2 PharmDx™ test (Dako Corporation), are approved both as prognostic tests and for the selection of patients for trastuzumab-based therapies. Published studies indicate that the single-probe and dual-probe assays are highly correlated [175]. Although controversial, a group of investigators strongly favors FISH as being more accurate and reliable than IHC in the classification of HER-2 status for breast cancer [176–181].

IHC Versus FISH.

Although the FISH method is more expensive and time-consuming than IHC, numerous studies have concluded that this cost is well borne by the greater accuracy and more precise use of anti–HER-2 targeted therapies [179–180, 182–183]. FISH is considered to be more objective and reproducible in a number of systematic reviews [165, 180, 183–186]. In one study, the concordance rates between IHC and FISH were highest in tumors scored by IHC as 0 and 1+ and lowest for 2+ and 3+ cases [183]. Currently, the majority (approximately 80%) of HER-2 testing in the U.S. commences with a screen by IHC, with results of 0 and 1+ considered negative, 2+ considered equivocal and referred for FISH testing, and 3+ considered positive. In a pharmacoeconomic study of patients being considered for trastuzumab-based treatment for HER-2–positive tumors, FISH was found to be a cost-effective diagnostic approach “from a societal perspective” [187].

Pitfalls in IHC and FISH Test Interpretation.

In addition to the preanalytic variables and issues with the subjective scoring system for IHC, a number of additional pitfalls in IHC test interpretation must be considered. In order to avoid false-positive IHC results, pathologists must learn to avoid scoring specimen edges, areas of tissue thermal injury from cautery, cases with cytoplasmic staining, fibrocystic disease with apocrine metaplasia, and intraductal (DCIS) foci. Although some investigators have favored a normalization approach, most experts hold that, when benign elements stain for HER in IHC procedures, either the antigen retrieval process was overly intense or the anti–HER-2 diagnostic antibody concentration was excessive [188]. A major cause of false-negative IHC staining is either reagent failure or failure of the antibody to be applied to the tissue. The most successful approach to avoiding this problem is to stain the newly diagnosed breast cancer on a slide that has had a 3+ HER-2–positive breast cancer from another patient (positive control) previously placed on the same slide (Fig. 3C).

Given the inability to recognize detailed background morphology during signal counting, a potential cause of false-positive FISH testing is the scoring of HER-2–amplified areas of DCIS in a tumor whose invasive carcinoma areas lack HER-2 amplification (Fig. 3D). In that the technique features a built-in internal control system, false-negative FISH results are rare but may occur when the slide scorer fails to identify the amplified regions in a tumor with heterogeneity in HER-2 gene amplification. HER-2 gene amplification can be heterogeneous in a significant subset of HER-2–positive invasive breast cancers, requiring diligence and care on the part of the slide scorer when scanning the case at low magnification [189].

CISH and Silver In Situ Hybridization.

The CISH method (Fig. 3E) and silver in situ hybridization (SISH) method feature the advantages of both IHC (routine microscope, lower cost, familiarity) and FISH (built-in internal control, subjective scoring, the more robust DNA target) [190, 191]. The CISH technique uses a single HER-2 probe, detects HER-2 gene copy number only, and was recently approved by the FDA to define patient eligibility for trastuzumab treatment. The SISH method employs both HER-2 and chromosome 17 centromere probes hybridized on separate slides and is currently under review by the FDA. Numerous studies have confirmed a very high concordance between CISH and FISH, typically in the 97%–99% range [191–203]. Similar to FISH, CISH has its highest correlation with IHC 0, 1+, and 3+ results and lowest correlation with IHC 2+ staining.

Chromosome 17 Polysomy.

The incidence of chromosome 17 polysomy has varied from as low as 4% to as high as >30% in studies of invasive breast cancer [204–208]. This may reflect differences in the definition of polysomy ranging from a low-level definition of more than two copies per cell to a high of more than four copies per cell of the chromosome. Most studies have linked chromosome 17 polysomy with greater HER-2 protein overexpression [204–207], but some have found that protein overexpression only occurs in the presence of selective HER-2 gene amplification [204]. In one study, 27% of cases featured chromosome 17 polysomy, and 35% of these patients responded to trastuzumab-based treatment [208]. The responding patients were restricted to cases that also had 3+ IHC staining. Another clinical outcome study, in patients with chromosome 17 copy numbers ≥2.2 detected by FISH and HER-2/chromosome 17 ratios <2.0 (HER-2 unamplified), nonetheless featured a significant response rate to a trastuzumab-based regimen [209]. However, this association could not be confirmed for a lapatinib-based clinical trial [210]. Thus, at least for trastuzumab-based treatment of MBC, chromosome 17 polysomy may be a significant cause of the observation that some patients may test negatively for HER-2 gene amplification by ratio-based FISH analysis and still respond to the drug. Finally, although not as yet validated in large prospective trials, it should be noted that positive responses to HER-2–targeted therapy in patients with tumors that are polysomic for chromosome 17 appear to be restricted to tumors that are 3+ by IHC testing.

Central Versus Local Laboratory Testing.

As mentioned above, several studies have considered the impact of local laboratory HER-2 testing in comparison with results obtained at central laboratories associated with major clinical trials [168–170, 184, 211]. In general, these studies have indicated that laboratories performing high-volume testing, that is, >10 tests per week, provide more accurate HER-2 results based on concordance with central laboratory testing results. Although this issue has likely been significantly mitigated by the incorporation of the American Society of Clinical Oncology–College of American Pathologists (ASCO-CAP) HER-2 testing guidelines program (see below), it should be noted that the central laboratory HER-2 test result may not always be the correct one. For example, when the central laboratory exclusively performs ratio-based FISH testing, discrepancies may be caused by HER-2 overexpression associated with chromosome 17 polysomy. Also, the central laboratory may not always receive an appropriate sample to retest and the preparation and shipment of the patient’s tissue, paraffin block, or unstained slides may inadvertently expose the tissue to factors that can degrade the HER-2 signal.

The 2007 ASCO-CAP Guidelines.

In early 2007, a combined task force from ASCO and the CAP issued a series of recommendations designed to improve the accuracy of tissue-based HER-2 testing in breast cancer [212]. A summary of the ASCO-CAP guidelines is provided in Table 4. Highlights of these recommendations include (a) standardizing fixation in neutral-buffered formalin for no less than 6 hours and no more than 48 hours, (b) unlike their respective FDA-approval specifications, defining equivocal zones for the IHC, FISH, and CISH tests, (c) establishing a standardized quality assurance program for testing laboratories, and (d) requiring the participation of these laboratories in a proficiency testing program [212]. The published guidelines were designed to improve the overall precision and reliability of all types of slide-based HER-2 tests and remained neutral as to the relative superiority of one test over the others.

Non–Slide-Based Assays

Southern and Slot Blotting.

These techniques, which measure the relative HER-2 DNA extracted from fresh tumor samples, were the original methods used to confirm that HER-2 amplification was an adverse prognostic factor in breast cancer [11].

RT-PCR.

Relative HER-2 mRNA levels measured by the RT-PCR technique (Fig. 4) have shown better correlation with HER-2 gene amplification results detected by FISH than with HER-2 protein levels determined by IHC [213]. To date, large clinical outcome studies have not been performed to confirm that the RT-PCR method can reliably predict response to HER-2–targeting agents. Nonetheless, there is growing interest in using mRNA levels to measure HER-2 status in breast cancer patients. RT-PCR is a relatively low-cost technique that could be used as a rapid screening method for establishing HER-2 mRNA status in concert with measurements of ER, PgR, and cell proliferation (Ki-67). However, being a non–morphology-driven non–slide-based approach, RT-PCR must be performed carefully on suitable areas of intact invasive cancer guided by examination of slides to confirm sample suitability [214–217]. HER-2 mRNA levels can be readily assessed on formalin-fixed, paraffin-embedded breast cancer samples as evidenced by the Oncotype DX™ (Genomic Health, Redwood City, CA) multigene predictor RT-PCR assay [218, 219]. HER-2 is one of the 21 gene expression measurements in the Oncotype DX™ test, and HER-2 mRNA individual determination has, on occasion, been used separately to assist in the resolution of cases in which initial HER-2 testing by IHC, FISH, and CISH has not been conclusive as to the true HER-2 status of the tumor. However, this approach has not been validated in a prospective trial and the response rate to HER-2–targeted therapies in patients whose HER-2 status is determined in this manner is not currently known.

mRNA by Microarray.

In the original classification of breast cancer by molecular portraits using dense DNA microarray–based relative mRNA measurements, HER-2–positive tumors fell into multiple classes, including the HER-2 and luminal groups, but not the normal or basal (triple-negative) categories [10]. The HER-2 gene is typically amplified as part of an amplicon that includes multiple adjacent genes (Fig. 5). In the various multigene predictor assays that have been commercialized for use in breast cancer management, the Oncotype DX™ test uses a direct measurement of the HER-2 mRNA level using RT-PCR [218, 219]. The recently developed TargetPrint™ assay (Agendia BV, Amsterdam, The Netherlands) measures ER, PgR, and HER-2 mRNA levels on a custom microarray. Other multigene predictors may use the expression of other genes directly related to HER-2 expression (HER-2 pathway genes) to determine breast cancer risk [219]. In a recent microarray-based study using fresh tissues and the U-133 genechip (Affymetrix, Santa Clara, CA), it was concluded that both ER and HER-2 mRNA could be easily and reliably determined by this method [220]. Thus, like the multiplex RT-PCR technique, genomic microarrays hold promise as potential multigene assays that can deliver routine prognostic and complex pharmacogenomic information for individualized patient management.

Dimerization Assays.

In several recent studies, the use of a method for determining the HER-2 receptor dimerization status by quantifying the number of HER-2 homodimers has predicted potential resistance to trastuzumab [221, 222]. This approach in combination with a direct measurement of HER-2 receptor number has recently been commercialized (HERmark™; Monogram Biosciences, South San Francisco, CA).

Phosphorylated HER-2 Receptors.

Activation of the HER-2 receptor by autophosphorylation has not been widely studied in clinical breast cancer samples. Monoclonal antibodies have been developed to detect autophosphorylated HER-2 by IHC [223]. In invasive breast cancer with HER-2 overexpression, the receptor appears to be activated only in a small subset (12%) of patients [223, 224]. Interestingly, the proportion of cases with phosphorylated (phospho)–HER-2 appears to be greater (58%) in DCIS [225]. In one large study of 800 cases of invasive breast cancer with HER-2 overexpression, only cases with phospho–HER-2 displayed an adverse prognosis [224]. Cases with overexpressed but unphosphorylated receptor had a prognosis as favorable as non–HER-2 overexpressing cases, which supports the concept that phospho–HER-2 may be a more powerful prognostic marker than overall HER-2 protein overexpression. Activated HER-2 status has been associated with resistance to taxane-based therapies [226]. When IHC is used as the method of detection of HER-2 receptor phosphorylation, excess antibody concentration or overintense antigen retrieval exposure can cause the antiphospho–HER-2 antibody to lose specificity and begin to detect wild-type HER-2 receptor. The role of phospho–HER-2 as a predictor of trastuzumab therapy response is currently unknown.

Tissue and Serum Enzyme-Linked Immunosorbent Assay.

The tissue enzyme-linked immunosorbent assay (ELISA) technique when performed on tumor cytosols made from fresh tissue samples avoids the potential antigen damage associated with fixation, embedding, and uncontrolled storage. In the six published studies listed in Table 1, ELISA-based measurements of HER-2 protein in tumor cytosols, mostly performed in Europe, are uniformly correlated with disease outcome. However, the small size of breast cancers associated with expanded screening programs in the U.S. generally precludes tumor tissue ELISA methods because insufficient tumor tissue is available to produce a cytosol.

The serum ELISA test measures the concentration of the extracellular domain (ECD) of the HER-2 protein (p185neu) in circulation. This assay has been approved by the FDA for the monitoring of HER-2–positive breast cancers, including the identification of disease relapse and ongoing response to HER-2–targeted therapies [227]. Attempts to use this serum-based test as the sole classifier of HER-2 status for newly diagnosed cases have not been widely accepted, although fairly good correlation exists between serum HER-2 ECD levels and the results of IHC and FISH assessments on primary tumor tissues [228]. It has been recommended that a 37 μg/l serum HER-2 ECD cutoff be used, which can achieve 95% specificity but low sensitivity for HER-2–positive status determined on primary tumors [229]. Studies of breast cancer prognosis based on the serum ECD test have been conflicting, with some finding significant correlation [116] and others finding weak or no correlation [230]. In 22 published studies on 4,088 patients, 16 (73%) studies involving 3,458 (85%) of the patients reported a significant correlation of serum HER-2 protein levels with disease recurrence, metastasis, or shorter survival [231–247]. Two studies involving 379 patients reported no significant association of serum levels with prognosis [248, 249]. Of the 11 studies in which serum HER-2 protein levels were tested for their ability to predict response to therapy, eight (73%) of the studies found that elevated serum HER-2 protein levels predicted therapy resistance [234, 242–244, 247–249], whereas three additional studies did not demonstrate this association [234, 249, 250]. Serum HER-2 levels have been correlated with shorter survival and the absence of clinical response to hormonal therapy in ER-positive tumors in some studies [239, 247], but not in others [249]. Serum HER-2 protein measurements have successfully predicted resistance to high-dose chemotherapy [242–244], bone marrow transplantation [243], and response to trastuzumab single-agent and combination treatment for metastatic HER-2–positive disease [251, 252]. In general, the test is advocated by some oncologists for the continuous monitoring of patients with HER-2–positive disease undergoing anti–HER-2 targeted therapy [231]. Nonetheless, the HER-2 serum ELISA test continues to be regarded as “under investigation” and has not, to date, been validated as a biomarker in large prospective clinical trials.

Circulating Tumor Cells.

The counting of circulating tumor cells (CTCs) as a predictor of response to breast cancer chemotherapy in the metastatic disease setting has been consistently validated in prospective studies [253–256]. The use of captured CTCs for the purpose of determining HER-2 status, however, has been controversial [257, 258]. Some studies have found that CTCs maintain the same HER-2 status, typically assessed by the FISH technique, as the primary tumor assay, whereas other reports have claimed that CTCs may be HER-2 positive in cases where the primary tumor was HER-2 negative [257, 258]. The methodological differences in assessing HER-2 status in the primary tumor versus in CTCs may at least partially account for these discrepant results. The different CTC techniques have influenced the capability of performing HER-2 testing with the immunomagnetic bead cell capture technique, requiring slide-based assays such as the FISH and RT-PCR techniques, with or without immunomagnetic-based cellular enrichment, claiming an enhanced sensitivity based on relative HER-2 mRNA measurements. Novel techniques are being developed to increase the yield of CTCs in a typical blood sample in order to facilitate more accurate biomarker testing and the use of additional assessment techniques including transcriptional profiling [256, 259–261].

Section Two: HER-2–Targeted Therapy and the Treatment of HER-2–Positive Breast Cancer

Trastuzumab: HER-2 Testing and the Prediction of Response to Trastuzumab Therapy

Using recombinant technologies, trastuzumab (Herceptin®; Genentech, South San Francisco, CA), a monoclonal IgG₁ class humanized murine antibody, was developed by the Genentech Corporation to specifically bind the extracellular portion of the HER-2 transmembrane receptor. This antibody therapy was initially targeted specifically for patients with advanced relapsed breast cancer that overexpresses HER-2 protein [262]. Since its launch in 1998, trastuzumab has become an important therapeutic option for patients with HER-2–positive breast cancer and is widely used for its approved indications in both the adjuvant and metastatic settings (Fig. 6) [185, 263–265]. Although trastuzumab is approved as a single-agent regimen, most patients are treated with trastuzumab plus cytotoxic agents. Table 5 summarizes the significant clinical trials that contributed to the regulatory approvals of trastuzumab.

Metastatic Disease Setting

Using a clinical trial IHC assay to select patients for the phase III pivotal trial, the addition of trastuzumab to chemotherapy (either an anthracycline plus cyclophosphamide or a taxane) was associated with a longer time to disease progression (median, 7.4 versus 4.6 months; p < .001), a higher rate of objective response (50% versus 32%; p < .001), a longer duration of response (median, 9.1 versus 6.1 months; p < .001), a lower rate of death at 1 year (22% versus 33%; p = .008), a longer survival duration (median survival time, 25.1 versus 20.3 months; p = .01), and a 20% lower risk for death [266]. Cardiac dysfunction occurred in 27% of the anthracycline and cyclophosphamide plus trastuzumab treated group, compared with 8% of the group given an anthracycline and cyclophosphamide alone [266]. Class III or IV cardiac dysfunction occurred in 16% of patients who received trastuzumab plus an anthracycline, versus 2% of patients treated with trastuzumab plus paclitaxel [267]. In a subsequent, randomized, multicenter trial, the combination of trastuzumab and docetaxel produced additional strong positive results in terms of OS, response rate, duration of response, and time to treatment failure compared with docetaxel treatment alone [268].

The original IHC technique used in the trastuzumab pivotal trial was the CTA, which consisted of two antibodies: (a) 4D5, the monoclonal antibody that is the actual antigen-binding murine component of Herceptin® and is not commercially available, and (b) CB-11, a monoclonal antibody directed toward the internal domain of the p185neu receptor, which is commercially available both as a research reagent and as an FDA-approved diagnostic (Ventana Pathway™). The original CTA was succeeded by the FDA-approved polyclonal HercepTest™. There was moderate concordance between the CTA and HercepTest™, although 58 of the 274 tumors that were scored as positive with the CTA were scored as negative with the HercepTest™ and 59 of the 274 tumors that were scored as negative with the CTA were scored as positive with the HercepTest™ [269]. After its FDA approval and launch, the HercepTest™ assay was initially criticized for yielding false-positive results [270], although better performance was ultimately achieved when the test was performed exactly according to the manufacturer’s instructions. Concern over IHC accuracy using standard formalin-fixed, paraffin-embedded tissue sections has encouraged the use of the FISH assay for its ability to predict trastuzumab response rates. Reports that FISH could outperform IHC in predicting trastuzumab response [271] and well-documented lower response rates of 2+ IHC staining versus 3+ IHC staining tumors [272] have resulted in an approach that either uses IHC as a primary screen with FISH testing of all 2+ cases or primary FISH-based testing. In a recently published study in which trastuzumab was used as a single agent, the response rate in 111 assessable patients with 3+ IHC staining was 35%, and the response rate for 2+ cases was 0%; the response rates in patients with and without HER-2 gene amplification detected by FISH were 34% and 7%, respectively [272]. In another study of breast cancer treated with trastuzumab plus paclitaxel, in patients with HER-2–overexpressing tumors, the overall response rates were in the range of 67%–81%, compared with 41%–46% in patients with normal expression of HER-2 [272]. The CB11 and TAB250 antibodies for IHC and FISH featured the strongest significance [273]. Interestingly, in a recently published review from New York and Italy, it was noted that, although FISH-based testing is more expensive and not as widely available as IHC, the data suggested that FISH was actually the most cost-effective option [274]. In summary, although the superiority of one method over the other continues to be controversial [162, 275–277], approximately 80% of laboratories in the U.S. are screening all new cases with IHC and triaging selected, usually 2+, cases for FISH testing, whereas 20% of testing uses FISH as the only method. It remains to be seen whether the newly approved CISH method will gain market share in the near and long term.

Adjuvant Setting

Table 5 outlines a series of clinical trials demonstrating the efficacy of trastuzumab-based regimens in the adjuvant setting [185, 278, 279]. The major phase III trastuzumab-based adjuvant trials (the NSABP B-31, North Central Cancer Treatment Group N9831, Herceptin® Adjuvant [HERA], and Breast Cancer International Research Group [BCIRG] 006 trials) used a variety of cytotoxic agents in various combinations, doses, and orders of administration [280–283]. When a 12-month course of trastuzumab was added to adjuvant chemotherapy, the DFS time was 33%–52% greater and the OS time was 34%–41% greater [280–283]. The improvements in DFS were independent of age, nodal status, hormonal status, or tumor size in all trials. As in the metastatic disease trials, cardiac toxicity was the most significant adverse event, occurring in 0%–0.9% of patients in the control arms and in 0%–3.8% of patients in the trastuzumab-containing arms, a level considered to be below the safety cutoff points set by the individual studies’ independent data monitoring committees [279]. These adjuvant trastuzumab trials have achieved these notable clinical results despite lacking a standardized approach to HER-2 testing. Of note in the adjuvant treatment trials was the impact of single-agent trastuzumab, which was featured in the treatment of node-negative patients in the HERA trial [281]. In addition, in an early study, trastuzumab monotherapy achieved a >30% overall response rate for IHC 3+ or FISH-positive tumors in the metastatic setting [272]. In current clinical practice, trastuzumab monotherapy is used, on occasion, for patients judged to be at risk for serious adverse events if exposed to a combination with cytotoxic agents. The strategy for using trastuzumab monotherapy, with or without endocrine therapy, for tumors judged to be low risk by routine clinicopathologic or molecular assessment is controversial. Some of the trials have included either IHC 3+ or FISH-positive tumors as entry criteria, whereas others, such as the Finnish Herceptin® trial, have used the CISH method. In addition, some trials have required central HER-2 testing confirmation before entry into the trial, whereas others have performed centralized laboratory testing after trial results were completed. Finally, the adjuvant trials were started before the publication of the ASCO-CAP HER-2 testing guidelines [212] were published and thus frequently used the >2.0 ratio cutoff for a positive FISH result rather than the recommended 2.2 cutoff recommended by the expert task force.

Trastuzumab Combinations.

Since the FDA approval in 1998 of two trastuzumab plus chemotherapy combinations, a number of additional approaches have gained favor in the clinical practice community. The National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines [284] currently recommend the following regimens for the first-line treatment of HER-2–positive MBC: trastuzumab plus single agents—either paclitaxel (every 3 weeks or weekly), docetaxel (every 3 weeks or weekly), or vinorelbine (weekly). For combination therapies, the NCCN recommends trastuzumab plus paclitaxel and carboplatin (every 3 weeks) or docetaxel plus carboplatin. Recently, carboplatin-based trastuzumab combinations have gained interest as a result of both the apparent boost in efficacy as measured by a higher overall response rate and longer progression-free survival time and the cardioprotective benefits of avoiding an anthracycline-containing regimen [285].

Trastuzumab Administration and Pharmacokinetics.

The pharmacokinetics of trastuzumab feature (a) low systemic clearance, (b) a low volume (4 l) of distribution, and (c) a very long, 28-day half-life [286]. Whether trastuzumab is used in the adjuvant setting or for the treatment of metastatic disease, the recommended dosage is the same. The clinical relevance of trastuzumab kinetic variability and elimination routes is unknown [286]. Drug–drug interactions have not been reported. After a loading dose, trastuzumab is typically given by i.v. perfusion at a dose based on body weight, in weekly (adjuvant, neoadjuvant, and metastatic disease protocols) or every-3-week (adjuvant protocols) regimens. For metastatic disease, trastuzumab treatment is typically continued until the time of disease progression. A short-course regimen of trastuzumab (9 weeks) is under investigation and appears promising in terms of activity, tolerance, and cost.

Trastuzumab Benefit in HER-2–Negative Tumors.

In the NSABP B-31 clinical trial of adjuvant trastuzumab plus chemotherapy [280], it was noted that a significant number of patients originally considered HER-2 positive by the local laboratory, and who appeared to benefit from the addition of trastuzumab, were ultimately considered HER-2 negative by the ratio-based FISH method performed at the NSABP central laboratory [287]. There are a number of potential technical explanations for this observation including: (a) because of a variety of factors (wrong tissue block, DNA degradation in specimen shipping, loss of the HER-2–positive focus on deeper sectioning, etc.), the central laboratory negative result may actually be incorrect; (b) some tumors feature chromosome 17 polysomy and overexpression of HER-2 protein that would not be detected when only ratio-based FISH results were evaluated centrally; and (c) because HER-2 status change occurs most commonly for borderline positive (or negative) cases, it is possible that the current threshold for HER-2 amplification (i.e., doubling of DNA copy number) is not the optimal threshold and tumors with lesser average amplification (i.e., ratio <1.0) may also benefit. It is also possible that other biologic pathways that are not linked to HER-2 gene amplification are inhibited by trastuzumab. In the early stages of trastuzumab clinical development in MBC, it was noted that HER-2–negative tumors by IHC rarely responded to the antibody, and that, if all patients were treated, the low response rate in HER-2–negative cases would significantly dilute the enhanced response rate in HER-2–positive cases and mask the overall clinical benefit of the novel therapeutic. One intriguing possible predictor of trastuzumab benefit in HER-2–negative breast cancer is the preliminary observation that HER-2–negative tumors that overexpress neuregulin, an activating ligand for HER-4, are inhibited by HER-2–targeted therapy [288]. Alternatively, it is also possible (although unprecedented) that a drug may have no significant anticancer activity in a particular disease subset in the metastatic stage (i.e., HER-2–normal cancers), but may have antitumor activity against the same subset in the micrometastatic stage. Immunological mechanisms could possibly underlie such an effect.

Neoadjuvant Setting

The results of trastuzumab-based neoadjuvant studies (Table 5) have received significant recent interest in the oncology community [289]. Virtually all completed and in progress clinical trials have demonstrated a significant enhancement in the rate of pathologic complete response (pCR), the primary endpoint in these studies, in cases of patients with HER-2–positive breast cancer that received trastuzumab in the neoadjuvant setting [290–297]. This benefit of the addition of trastuzumab in the neoadjuvant setting appears to be independent of, if not enhanced by, the coexistence of ER positivity [297]. Among the potential explanations for the apparent greater chemosensitivity of HER-2–positive tumors cotreated with trastuzumab in the neoadjuvant setting is the concept that HER-2 gene amplification is in some way related to the growth and survival of breast cancer stem cells [298, 299]. The higher pCR rates in HER-2–positive breast cancers treated with neoadjuvant trastuzumab may conceivably reflect the inhibition of both stem cell and progenitor cell proliferation and invasion by removing or downregulating HER-2–mediated growth signals [299]. Also of interest in the neoadjuvant trial results is the possible observation that HER-2–targeted therapy can convert a HER-2–positive breast cancer into a HER-2–negative tumor [300]. In a recent report, nearly one third of the patients with HER-2–amplified breast cancer treated with a taxane and anthracycline-based chemotherapy with concomitant trastuzumab in the neoadjuvant setting that failed to achieve pCR were found to have converted to HER-2–negative disease [300]. Further validation of these findings awaits additional prospective studies.

Biomarkers of Trastuzumab Resistance

Since trastuzumab was introduced for the treatment of MBC in 1998, there has been growing interest in the discovery and potential clinical utility of biomarkers designed to predict resistance to the drug. Current approaches to HER-2 testing provide a negative predictor of drug response: the test does not predict which patients will respond to trastuzumab, it predicts which patients are unlikely to benefit. The study of resistance biomarkers has been limited to a degree by the lack of a consensus definition of resistance in both the adjuvant and metastatic settings. In the neoadjuvant setting, resistance has been defined as a failure to achieve a pCR or near pCR. A number of biomarkers proposed as predictors of trastuzumab resistance are listed in Table 6. It should also be noted that trastuzumab is typically combined with one or more cytotoxic agents and attempts to determine individual biomarkers predictive of trastuzumab resistance will be significantly impacted by the biologic pathways related to the resistance or sensitivity of the tumor cells to the companion agents as well.

HER-2 Gene Copy Number.

It has been reported that tumors with higher HER-2 gene copy numbers (e.g., >10 HER-2 copies per nucleus) are more sensitive to trastuzumab [301, 302]. Despite this evidence, all patients with gene copy numbers >6.0 per nucleus or gene ratios of HER-2/CEP17 >2.2 are equally considered HER-2 positive by the ASCO-CAP task force, and HER-2 gene copy number is not currently used to determine the intensity or duration of trastuzumab therapy.

Shedding of HER-2 Protein.

Early in the time line of trastuzumab development there was concern that significant shedding of the HER-2 surface receptor (p185neu) protein, as evidenced by a high serum HER-2 protein ELISA test, would be associated with resistance of the tumor as a result of the neutralization of the infused antibody [303]. However, follow-up studies have not confirmed that a high serum HER-2 level at the outset of trastuzumab therapy is predictive of clinical tumor resistance.

Dimerization Status.

Previous studies have suggested that HER-2 dimerization status (HER-2:HER-2 homodimers versus HER-2:HER-3 and HER-2:HER-4 heterodimers) can predict response to trastuzumab-based therapy in MBC [221, 222, 304]. Validation of these initial observations has not been presented in large cohorts of patients and prospective testing of dimerization status as a predictor of trastuzumab resistance has not been published to date.

Fc Receptor Status and Antibody-Dependent Cellular Cytotoxicity Response.

Antibody-dependent cellular cytotoxicity (ADCC) is considered a major aspect of the mechanism of action of trastuzumab [305–309]. Interactions with the Fc receptor may be a critical step in the activation of natural killer lymphocytes and ADCC response. Preliminary studies have linked both germline polymorphisms and post-translational modifications (glycosylation and fucosylation) of the Fcγ receptor with the impaired ADCC response associated with monoclonal antibody therapeutics such as trastuzumab [309]. The clinical development of Fc receptor assays to predict trastuzumab resistance will require validation of these retrospective observations in prospective trials.

Phosphatase and Tensin Homologue Deleted on Chromosome Ten Deficiency/PI3K Pathway Activation.

A number of studies have linked the loss of phosphatase and tensin homologue deleted on chromosome ten (PTEN) tumor suppressor gene expression and activation of the PI3K pathway with resistance to trastuzumab-based therapy [310–314]. Although the potential for PTEN status to predict trastuzumab response appears quite promising, the associations have not been uniformly observed by all investigators and additional validation in prospective studies is clearly required before this biomarker can achieve widespread clinical adoption.

EGFR Expression.

Early studies suggested that amplification of EGFR and overexpression of EGFR would confer clinical resistance to trastuzumab [315]. This observation has not been validated in large-scale follow-up studies.

c-MYC Amplification.

Based on data from the NSABP B-31 trastuzumab adjuvant trial, tumors that feature coamplification of the c-MYC oncogene and HER-2 benefited by the addition of trastuzumab to chemotherapy in terms of both recurrence-free survival and OS, compared with patients whose tumors lacked the c-MYC amplification [316]. In a neoadjuvant study, PgR-negative status and c-MYC amplification were both associated with higher pCR rates after the addition of trastuzumab to chemotherapy [317]. Thus c-MYC gene copy status may be a biomarker of trastuzumab response in the adjuvant or neoadjuvant settings, although this requires large-scale studies for confirmation.

Insulin-Like Growth Factor 1 Receptor Status.

The overexpression of the insulin-like growth factor 1 receptor (IGF-1R) has been associated with resistance to trastuzumab in some studies [318, 319], but not in others [318–321]. Experimental models favor the idea that activation of IGF-1R confers resistance to trastuzumab [322]. Trials examining the potential synergism between trastuzumab and novel anti–IGF-1R therapeutics have been initiated.

Mucin 4.

The mucin 4 glycopeptide may be secreted by some breast cancers and interfere with trastuzumab binding to the HER-2 receptor [323].

p95HER-2.

The accumulation of truncated forms of the HER-2 receptor (p185HER-2) that lack the extracellular trastuzumab-binding domain of the intact receptor have been associated with resistance to trastuzumab in preclinical studies [324]. Amino terminally truncated carboxyl terminal fragments of HER-2 are termed p95HER-2. In one published clinical study involving 46 patients, breast tumors that expressed p95HER-2 showed a lesser or absent response to trastuzumab-based regimens in a retrospective analysis [324].

Phospho–HER-2.

Preliminary studies have linked HER-2 receptor phosphorylation status to response to trastuzumab-based regimens [325]. Large-scale studies of HER-2 phosphorylation status and trastuzumab response have not been performed to date.

Topoisomerase IIα Amplification.

Although topoisomerase IIα amplification has been linked to the benefit of anthracycline chemotherapy in HER-2–positive breast cancers, specific association of topoisomerase IIα status with response to trastuzumab-based regimens in either the adjuvant or metastatic disease settings has not been confirmed [326]. In the BCIRG 006 trastuzumab adjuvant trial, topoisomerase IIα gene copy number detected by FISH was studied as a predictive biomarker, although interim reports have not confirmed its utility [282].

Basal Phenotype.

The basal-like phenotype of breast cancer is associated with IGF-1R overexpression and resistance to inhibition of trastuzumab-mediated blockade of HER-2 tyrosine kinase signaling [327]. In the basal-like phenotype, HER-2–positive status is quite infrequent, occurring in <10% of cases [327].

CD44 Tumor Cell Overexpression.

It has been postulated that CD44 binding at the cell surface may reduce ADCC for trastuzumab [328].

High Tumor/Serum Vascular Endothelial Growth Factor Levels.

Overexpression of vascular endothelial growth factor (VEGF) in breast cancers and high serum levels of VEGF have been postulated as a cause of trastuzumab resistance [329].

MicroRNA.

MicroRNA (miRNA) signatures have not, to date, been linked to HER-2 status. Nonetheless, the association of expression of specific miRNAs with response to hormonal and cytotoxic therapy suggests that miRNA biomarkers of trastuzumab may soon be uncovered [330].

Lapatinib: HER-2 Testing and the Prediction of Response to Lapatinib Therapy

Lapatinib (Tykerb®; Glaxo Smith Kline, Research Triangle Park, NC) is an orally available small-molecule dual inhibitor of the EGFR and HER-2 tyrosine kinases [331].

Metastatic Disease Setting

Lapatinib was approved by the FDA in 2007 for use in combination with capecitabine for the treatment of HER-2–positive MBC that has progressed with standard treatment [332]. In the phase III registration trial involving 399 patients (Table 5), the addition of lapatinib to capecitabine produced a longer median time to progression by 8.5 weeks (27.1 weeks, versus 18.6 weeks for capecitabine alone) with an HR of 0.57 (p = .00013) [331]. The response rates were 23.7% and 13.9%, respectively, with an HR of 1.9 (p = .017) [331]. Patients were eligible for the trial based on either a HER-2 IHC score of 3+ or a FISH ratio >2.0. In an updated efficacy and biomarker report, OS in the lapatinib plus capecitabine treated group trended toward a longer survival duration [333]. In a lapatinib monotherapy trial for patients who had progressed on a trastuzumab regimen, the response rate in the HER-2–positive group was 4.3%, indicating modest clinical activity of the drug as a single agent [334].

Adjuvant Setting

The Adjuvant Lapatinib and/or Trastuzumab Treatment Optimization (ALTTO) and Tykerb® Evaluation After Chemotherapy (TEACH) trials are two of the major phase III trials that are currently evaluating lapatinib in the adjuvant setting (Table 5) [335, 336]. The ALTTO trial plans to enroll 8,000 patients and is sponsored by the National Cancer Institute, part of the U.S. National Institutes of Health, and GlaxoSmithKline, and is being coordinated by The Breast Cancer Intergroup of North America in the U.S. and the Breast International Group in Brussels, Belgium [335]. The TEACH trial is designed to determine whether adjuvant therapy with lapatinib for 1 year will improve DFS in women with early-stage HER-2–positive breast cancer. This trial plans to enroll 3,000 patients [333]. Efficacy data from adjuvant trials featuring lapatinib in combination with cytotoxic agents are not available at this time.

Neoadjuvant Setting

The Neo-ALTTO trial is a randomized, open-label, multicenter, phase III study comparing the efficacy of neoadjuvant lapatinib plus paclitaxel with that of trastuzumab plus paclitaxel and with concomitant lapatinib and trastuzumab plus paclitaxel given as neoadjuvant treatment in HER-2–positive primary breast cancer [337]. The CHERLOB trial is a randomized trial of preoperative chemotherapy plus trastuzumab and lapatinib or the combination of trastuzumab and lapatinib in HER-2–positive operable breast cancer featuring a tumor diameter >2 cm [338, 339]. pCR is the endpoint for both of these lapatinib neoadjuvant trials that compare lapatinib plus cytotoxic agents with lapatinib plus trastuzumab plus cytotoxic agents. Efficacy data have not been published on these trials to date.

HER-2–Positive CNS Metastases

A major goal for the development of lapatinib has been the potential efficacy in cases of CNS involvement in patients with HER-2–positive MBC progressing on trastuzumab-based regimens. The relative success of trastuzumab in the treatment of visceral disease in MBC appears to have unmasked the clinical problem of progressive CNS disease in HER-2–positive patients, a clinical syndrome not frequently encountered in the pretrastuzumab era. In a report of 39 patients heavily pretreated with trastuzumab and taxanes who had progressed despite radiation, two patients achieved a partial response based on the Response Evaluation Criteria In Solid Tumors, and five additional patients were found to have experienced at least a 30% volumetric reduction in their CNS lesions [340]. The potential efficacy of lapatinib in trastuzumab-resistant brain metastases awaits further documentation in larger case cohorts.

Inflammatory Breast Cancer

In a phase II trial, lapatinib treatment has shown early promise in the treatment of HER-2–positive inflammatory breast cancer [331].

Lapatinib Administration and Pharmacokinetics

With oral administration of the FDA-recommended daily dose of 1,250 mg/day, the time of maximum plasma concentration of lapatinib is 3–4 hours [331]. Lapatinib is metabolized primarily by the cytochrome P450 system via the 3A4 isozyme, resulting in a single metabolite active against EGFR but not HER-2. With continuous dosing, the half-life of lapatinib is 24 hours [331].

Biomarkers of Lapatinib Resistance

In that lapatinib was approved 9 years after trastuzumab, considerably less information has been published concerning markers of efficacy or resistance to the drug [331, 341–343].

Serum HER-2 ECD Status.

In the lapatinib plus capecitabine versus capecitabine trial, preliminary study of biomarkers failed to identify tissue EGFR or HER-2 biomarkers predictive of lapatinib resistance [333]. However, a significant reduction in lapatinib response was associated with cases in which the starting serum HER-2 ECD levels were high [329]. In another study, high serum HER-2 ECD levels did not predict benefit from lapatinib-based combination therapy [344].

Tissue HER-2 Status.

All completed and in-progress clinical trials employing lapatinib have required that entering patients have tumors that are HER-2 positive by FISH or IHC. A number of studies have confirmed that HER-2 positivity is required for lapatinib clinical benefit [341]. In one recent report, on a clinical trial that originally found limited lapatinib benefit in cases of HER-2–unamplified tumors tested by FISH, lapatinib efficacy was found to be limited to HER-2–positive cases when tumors were retested by an academic central laboratory and scored by a pathologist rather than a technician [345].

Tissue EGFR Status.

Although lapatinib’s mechanism of action includes the inhibition of the tyrosine kinase activity of both HER-1 (EGFR) and HER-2, a number of studies have failed to link amplification of EGFR or overexpression of EGFR with the efficacy of lapatinib-based therapies [346]. In one study, there was no correlation between EGFR expression (IHC or mRNA) and responsiveness to lapatinib regardless of HER-2 status [346].

Chromosome 17 Polysomy.

Although extra copies of chromosome 17 have been linked to the efficacy of trastuzumab in patients whose HER-2 FISH ratio test is negative for HER amplification, in one recent study, polysomy of chromosome 17 was not associated with lapatinib benefit in HER-2–negative tumors [347].

IGF-1R.

In a preclinical study, lapatinib inhibited IGF-1R signaling and growth-promoting effects in parental and resistant cells [348]. The studies indicating that IGF-1R signaling can cause trastuzumab resistance have encouraged the concept that lapatinib will prove efficacious in breast cancers that have progressed on trastuzumab.

ER Signaling.

In one preclinical/clinical study, it was postulated that signaling through the ER pathways was a significant mechanism of resistance to lapatinib [349].

PTEN.

Although data are limited in comparison with trastuzumab [317–321], in one study, in contrast to the relatively strong supporting data for trastuzumab, loss of PTEN expression was not associated with lapatinib resistance in either cell lines or clinical specimens [350].

Trastuzumab–Lapatinib Combinations and Other Targeted Therapies

Trastuzumab Plus Lapatinib

A number of clinical trials are examining the potential synergy of using both trastuzumab and lapatinib for HER-2–positive breast cancer in the neoadjuvant, adjuvant, and metastatic disease settings. Data from the ALTTO, Neo-ALTTO, and CHERLOB trials are not yet mature, and efficacy data are currently not available to assess the impact of combining HER-2–targeted agents [336, 338, 339]. In a recent interim report, testing the efficacy of the combination of trastuzumab and lapatinib compared with lapatinib alone in a heavily pretreated population of HER-2–positive MBC patients who progressed on trastuzumab-based regimens, significant synergy, as measured by the progression-free survival duration, was shown [351].

Trastuzumab Plus Bevacizumab

The BEvacizumab and Trastuzumab Adjuvant Therapy in HER-2-Positive Breast Cancer trial is a multicenter phase III randomized adjuvant trial comparing chemotherapy plus trastuzumab with chemotherapy plus trastuzumab and the anti-VEGF ligand bevacizumab [352]. No efficacy data are available at this time. In an ongoing neoadjuvant trastuzumab–bevacizumab trial, to date, the addition of bevacizumab has not resulted in a higher rate of pCR [353]. However, in a recent interim report of the combination in a trial of locally advanced disease treated in the neoadjuvant setting, early evidence of synergistic efficacy was noted [354].

Trastuzumab Plus Everolimus

As documented in a preclinical study, one of the strategies for overcoming the resistance of PTEN-deficient breast cancers to trastuzumab is the targeting of the Akt pathway using a mammalian target of rapamycin (mTOR) inhibitor [355]. RAD001 (everolimus) is an inhibitor of mTOR currently in clinical trials for the treatment of HER-2–positive breast cancer in combination with trastuzumab. In an ongoing clinical trial, early efficacy data suggest the possibility of significant synergism from the addition of everolimus to a trastuzumab and taxane regimen in the metastatic disease setting [356].

Trastuzumab Plus Heat Shock Protein 90 Inhibitors

Inhibition of the chaperone protein heat shock protein 90 (HSP90) results in increased degradation of HER-2 ECD [357, 358]. Two anti-HSP90 agents that have been combined with trastuzumab in early-stage clinical trials are geldanamycin and tenespimycin (17-AAG; Kosan Biosciences, Hayword, CA). Reports of a trial of tenespimycin combined with trastuzumab in advanced pretreated MBC have shown good safety and tolerability [359] and early indications of significant clinical activity in HER-2–positive disease [360].

Duration of Anti–HER-2 Targeted Therapy

A number of recent reviews have summarized the lack of standardization of the duration of treatment with anti–HER-2 targeting agents in HER-2–positive breast cancer [361–364]. Although the current recommended duration of trastuzumab treatment is 1 year in the adjuvant setting, different treatment durations, from 9 weeks to 2 years, have been studied with, to date, no optimal duration of treatment achieving consensus among investigators [265]. In the lapatinib plus capecitabine registration trial, oral lapatinib therapy was maintained until the time of disease progression or based on adverse events [331, 332]. In a recent study, a higher efficacy but similar toxicity were found when trastuzumab was continued beyond progression and second-line chemotherapy with capecitabine was initiated [365]. However, there is increasing evidence that continuation of anti–HER-2 therapy after progression on trastuzumab confers clinical benefit. In a recent review by the NCCN, it was noted that 74% of patients with MBC who had progressed after first-line trastuzumab-based therapy continued to receive trastuzumab in a second-line protocol [366]. Currently, no specific biomarkers appear to be capable of preselecting an individual patient for a short-term or long-term treatment regimen. A variety of markers, including serum-based assays and imaging studies, have been proposed to guide the cessation or continuance of treatment with these drugs, but, to date, no clear consensus on what tests should be selected and how they should be used has emerged.

Novel Anti–HER-2 Targeted Therapies

HER-2 Vaccines

A novel approach toward the treatment of HER-2–positive breast cancer has been the use of vaccines and adoptive immunotherapy targeting HER-2 ECD [367–371]. HER-2–specific vaccines have been tested in human clinical trials that have shown that significant levels of durable T-cell HER-2 immunity can be generated with active immunization. No significant autoimmunity directed against normal tissues has been encountered [368]. Moreover, active anti–HER-2 immunization could facilitate the ex vivo expansion of HER-2–specific T cells for use in adoptive immunotherapy for the treatment of established metastatic disease [367]. In addition, early data from trials examining the potential use of HER-2–based vaccines in the adjuvant setting to prevent the relapse of breast cancer in high-risk patients have shown promising results [371]. Future approaches include the development and testing of multiepitope vaccines [370].

Pertuzumab

Pertuzumab (rhuMab 2C4, Omnitarg™; Genentech Corp., South San Francisco, CA) is an anti–HER-1/HER-2 antibody that inhibits HER-1–HER-2 dimerization [372]. Pertuzumab does cause an ADCC reaction, but it does not block HER-2 shedding. Pertuzumab may have efficacy in breast cancers featuring low levels of HER-2 overexpression or in cases in which HER-2 protein levels are normal but HER-1 (EGFR) levels are elevated [372]. Clinical trials evaluating pertuzumab efficacy in MBC have not been successful to date [372, 373]. The observation that pertuzumab can elicit a metabolic response detected by position emission tomography scanning in HER-2–negative MBC has fueled continued interest in the development of the antibody in subsets of breast cancer patients [374]. In a more recent phase II study of trastuzumab and pertuzumab combination therapy in HER-2–positive metastatic disease, a 40% clinical benefit rate with multiple complete and partial responses was described [375].

Ertumaxomab

Ertumaxomab (Fresenius Biotech, Hamburg, Germany) is a trifunctional bispecific antibody targeting HER-2 on tumor cells and CD3 on T cells that has the capability to redirect T cells, macrophages, dendritic cells, and natural killer cells to the sites of tumor metastases [376, 377]. In a phase I trial, ertumaxomab treatment was associated with one complete response and several partial responses in heavily pretreated patients with MBC [376].

MDX-H210

MDX-H210, a bispecific antibody targeting HER-2 combined with G-CSF has been tested in early clinical trials with limited clinical response to date [378].

Trastuzumab Conjugates

Early attempts to conjugate HER-2–targeting antibodies with a toxin involved the use of Pseudomonas aeruginosa exotoxin [379]. More recently, trastuzumab was conjugated with the fungal toxin maytansine (DM-1) [380]. In a recent report of a phase I trial, objective responses to trastuzumab-DM1 (Genentech Corp., South San Francisco, CA) were seen below the maximal tolerated doses of the antibody conjugate [380]. Phase II trials of this agent are currently in progress, with a recent interim report finding a 40% response rate in a heavily pretreated patient cohort including prior trastuzumab and/or lapatinib therapy [381].

Novel Tyrosine Kinase Inhibitors

A number of tyrosine kinase inhibitors (TKIs) are in early-stage clinical development for the treatment of HER-2–positive breast cancer. Similar to lapatinib, HKI-272 (Wyeth Corp., Madison, NJ) is a HER-1/HER-2 dual kinase inhibitor that recently was shown to have efficacy and acceptable toxicity in an early-stage clinical trial for advanced MBC [382, 383]. A number of additional HER-1/HER-2 TKIs, pan-HER TKIs, and dual HER-2/VEGF TKIs are in various stages of preclinical and early clinical development.

Prediction of Toxicity for Anti–HER-2 Targeted Therapies

Trastuzumab

Since its introduction in the MBC setting and continuing throughout its advance into use in both the adjuvant and neoadjuvant settings, trastuzumab has been associated with the development of a variety of toxicities [384]. In the original registration trial for MBC, trastuzumab was associated with a variety of adverse events, including pain, gastrointestinal disturbances, minor hematologic deficiencies, pulmonary symptoms, and congestive heart failure (CHF) [265]. Cardiac toxicity has remained the most significant limiting factor for the use of trastuzumab [384–389]. A major consideration in the development of cardiac toxicity in patients treated with trastuzumab has been their prior or concomitant exposure to anthracycline drugs, also associated with dose-dependent irreversible heart damage [384–389]. Trastuzumab-related cardiac dysfunction is typically reversible and does not appear to increase with cumulative dose or to be associated with ultrastructural changes in the myocardium [389]. In the adjuvant trastuzumab trial NSABP B-31, abnormal left ventricular ejection fraction (LVEF) and advanced patient age were significant predictors of CHF development, with hypertension classified as a near-significant predictor [390]. Although major class III and IV cardiac toxicity has varied in the range of 1%–4% of treated patients in published adjuvant clinical trials, recent evidence suggests that the current severe toxicity incidence is close to 1% [384–390]. This reduction in cardiac dysfunction reflects an increasing familiarity of the signs and symptoms of developing heart toxicity (reduced LVEF, early signs of CHF) by treating oncologists and knowledge of associated risk factors. Currently, there are no validated blood- or tissue-based biomarkers that can reliably predict the development of cardiac toxicity after exposure to trastuzumab. Accepted risk factors include: (a) prior anthracycline exposure, (b) diabetes mellitus, (c) prior coronary artery syndromes, (d) hypertension, and (e) pre-existing CHF. The cardiovascular drugs used to treat anthracycline- and trastuzumab-associated cardiac dysfunction, angiotensin-converting enzyme inhibitors and beta blockers, may be useful as preventative agents when administered immediately prior to the start of trastuzumab treatment, although this approach has not been validated in prospective trials to date. Finally, various strategies for preventing significant drops in LVEF have emerged, including the use of trastuzumab combinations (vinorelbine, taxanes, platinum salts) that avoid anthracyclines [389].

Lapatinib

The most frequent adverse reactions in the lapatinib–capecitabine registration trial for MBC combination were diarrhea (65%), palmar–plantar erythrodysesthesia (53%), nausea (44%), rash (28%), vomiting (26%), and fatigue (23%) [332]. In a comprehensive analysis of the clinical trials featuring lapatinib in combination with various other agents, the overall incidence of LVEF decline was 1.6%, with 0.2% of patients experiencing symptomatic CHF [389]. No validated blood- or tissue-based biomarkers have emerged to predict adverse events associated with lapatinib exposure.

HER-2 Status and the Prediction of Response to Non–HER-2 Targeted Therapy

Hormonal Therapies

The use of HER-2 status to predict responsiveness or resistance to hormonal therapies, advocated by a number of oncologists, remains controversial. It has been reported that ER-positive/HER-2–positive patients are either less responsive or completely resistant to single-agent tamoxifen [391–393]. When measured as continuous variables, the expression of HER-2 appears to be inversely related to the expression of ER and PgR even in hormone receptor–positive tumors [394]. In a number of published studies in both the MBC and adjuvant settings, HER-2–positive tumors were specifically resistant to tamoxifen therapy [62, 239, 395]. However, in other studies, HER-2 status failed to predict tamoxifen resistance in ER-positive cases [396, 397]. In a unique study based in Europe that featured a placebo group in the adjuvant setting, ER-positive/HER-2–positive tumors were not only found to be resistant to tamoxifen, but single-agent tamoxifen treatment actually had an adverse impact when compared with untreated patients [398]. However, this finding was not confirmed by large intergroup studies in the U.S. [399]. Both ER-positive/HER-2–negative and ER-positive/HER-2–positive tumors have been associated with a superior response to aromatase inhibitors [400, 401], a finding that has led some investigators to advocate that aromatase inhibitors be used preferentially in ER-positive/HER-2–positive tumors [400–402]. However, the proposed preferential efficacy of aromatase inhibitors in ER-positive/HER-2–positive breast cancer has not been validated in prospective, randomized trials. Interestingly, HER-2 status did not successfully predict response to fulvestrant-based hormonal therapy [403]. Continued studies of gene expression have revealed considerable crosstalk between the ER and HER-2 pathways in MBC [404, 405]. In summary, reviews of multiple clinical and experimental studies are in consensus that a HER-2–positive status confers resistance of breast cancer tumor cells to hormonal therapy [406–409]. Whether HER-2 status can be used to select individualized approaches to hormonal therapies in ER-positive patients, however, has not been validated.

Anthracyclines

HER-2 overexpression has also been associated with enhanced response rates to anthracycline-containing chemotherapy regimens in most, but not all, studies [42, 410–414]. Because anthracyclines are topoisomerase inhibitors and the topoisomerase IIα gene is coamplified with HER-2 in approximately 35% of HER-2–positive breast cancers, it has been suggested that HER-2 may be serving as a surrogate marker of anthracycline sensitivity. Topoisomerase IIα gene amplification is mostly restricted to HER-2–positive breast cancer and is rarely encountered in tumors that lack HER-2 gene amplification. Although HER-2 protein expression, but not topoisomerase IIα expression, predicted the response of breast cancer to epirubicin in one study [415], it should be noted that topoisomerase IIα expression is regulated according to the cell cycle and, in contrast to HER-2 protein expression, does not directly reflect topoisomerase IIα gene amplification status. A number of studies have linked coamplification of the topoisomerase IIα and HER-2 genes with adverse prognosis and sensitivity to anthracycline drugs [416–419]. However, topoisomerase IIα gene deletion has also been linked to anthracycline sensitivity, which has prevented the development of a consensus as to whether topoisomerase IIα testing should be routinely performed in the management of breast cancer [96]. The BCIRG 006 Trial (Table 5) included the validation of topoisomerase IIα gene amplification as a predictor of anthracycline benefit in a trastuzumab-based adjuvant treatment setting [282]. To date, interim results from that trial have not fully confirmed that topoisomerase IIα amplification testing can be used to select or avoid anthracyclines in the treatment of HER-2–positive breast cancer.

Cyclophosphamide, Methotrexate, and 5-Fluorouracil

Although HER-2 protein overexpression detected by IHC was initially associated with the resistance of tumors in patients treated with cyclophosphamide, methotrexate, and 5-fluorouracil adjuvant chemotherapy [41], larger follow-up studies failed to demonstrate a lack of benefit in HER-2–positive tumors treated with this multidrug regimen [420, 421].

Taxanes

Although initial studies also reported HER-2–positive breast cancer to be resistant to taxane-based regimens [422], subsequent reports suggested that HER-2–positive tumors were selectively sensitive to these agents [423–425]. Most recently, in a large study of 1,300 women with lymph node–positive breast cancer, HER-2–positive status was clearly associated with a benefit from the addition of paclitaxel after adjuvant chemotherapy, regardless of ER status. In addition, patients with HER-negative/ER-positive, node-positive disease appeared to gain little benefit from the addition of paclitaxel after adjuvant chemotherapy with doxorubicin plus cyclophosphamide [426]. Results from neoadjuvant studies also support the notion that HER-2–positive tumors, particularly among the ER-positive patients, represent cancers with greater than average chemotherapy sensitivity. This is indicated by the significantly higher pCR rates in HER-2–negative and ER-positive cancers than in HER-2–normal and ER-positive cancers.

Radiation Therapy

Initially, in the era prior to the introduction of anti–HER-2 targeted therapy, HER-2–positive status was associated with a higher rate of local recurrence in some studies of breast cancer treated with surgery and radiation therapy alone, but not in others [427–429]. However, although large-scale, randomized, prospective studies are lacking, HER-2–positive tumors treated with trastuzumab-based neoadjuvant chemotherapy combined with external-beam radiation have indicated a favorable response in locally advanced breast cancer [430]. In addition, HER-2–positive brain metastases appear to be more sensitive to local radiation than HER-2–negative tumors [431].

Summary

The history of the discovery of the HER-2 oncogene in an animal model in 1984, the translation of this finding to the clinical behavior of human breast cancer, and the introduction of the first anti-HER targeted therapy in 1998 is clearly a triumph of “bench to bedside” medicine. In the 10 years that have now passed since the regulatory approval of the first anti–HER-2 targeted therapy, trastuzumab, thousands of preclinical and clinical studies have considered HER-2 as a prognostic factor, its ability to predict response to hormonal and cytotoxic treatments, the best way to test for it in routine specimens, and the clinical efficacy of targeting it in a wide variety of clinical settings. Given the proven efficacy of trastuzumab and lapatinib for the treatment of MBC, and also in the adjuvant and neoadjuvant settings, the critical issue as to which test (IHC versus FISH versus CISH versus mRNA based) is the most accurate and reliable method to determine HER-2 status in breast cancer has continued to increase in importance. The introduction of the ASCO-CAP guidelines in early 2007 has clearly improved HER-2 testing accuracy in the U.S., but significant work on testing must be undertaken. The anti–HER-2 targeted therapies have significant efficacy, especially when combined with cytotoxic agents. For this reason, whenever possible, HER-2–positive breast cancers must not be misclassified as HER-2 negative, denying those patients the opportunity to benefit from trastuzumab and lapatinib. Similarly, the anti–HER-2 drugs are expensive and can cause serious toxicity. For this reason, whenever possible, HER-2–negative patients must not be misclassified as HER-2 positive and exposed to cost and potential adverse effects of these drugs when they have very little chance of receiving clinical benefit from treatment. In the next 10 years, it is likely that further refinement in HER-2 testing accuracy will be combined with additional testing for validated biomarkers of HER-2–targeted therapy efficacy and resistance, predictors of toxicity, and rational selection of companion cytotoxic drugs in a continuing effort to achieve even greater success towards the ultimate cure and, when necessary, palliation of HER-2–positive breast cancer.

Author Contributions

Conception/design: Jeffrey Ross, Elzbieta Slodkowska, W. Fraser Symmans, Lajos Pusztai, Peter Ravdin, Gabriel Hortobagyi

Financial support: Jeffrey Ross

Collection/assembly of data: Jeffrey Ross, Elzbieta Slodkowska, W. Fraser Symmans, Lajos Pusztai, Peter Ravdin, Gabriel Hortobagyi

Data analysis: Jeffrey Ross, Elzbieta Slodkowska, W. Fraser Symmans, Lajos Pusztai, Peter Ravdin, Gabriel Hortobagyi

Manuscript writing: Jeffrey Ross, Elzbieta Slodkowska, W. Fraser Symmans, Lajos Pusztai, Peter Ravdin, Gabriel Hortobagyi

Final approval of manuscript: Jeffrey Ross, Elzbieta Slodkowska, W. Fraser Symmans, Lajos Pusztai, Peter Ravdin, Gabriel Hortobagyi

References

Author notes