|Home | About | Journals | Submit | Contact Us | Français|
Decades of research in molecular oncology have brought about promising new therapies that are designed to target specific molecules that promote tumor growth and survival. The epidermal growth factor receptor (EGFR) is one of the first identified important targets of these novel antitumor agents. Approximately half of cases of triple-negative breast cancer (TNBC) and inflammatory breast cancer (IBC) overexpress EGFR. Thus, EGFR inhibitors for treatment of breast cancer have been evaluated in several studies. However, results so far have been disappointing. One of the reasons for these unexpected results is the lack of biomarkers for predicting which patients are most likely to respond to EGFR inhibitors. Recent studies have shown that EGFR and its downstream pathway regulate epithelial-mesenchymal transition, migration, and tumor invasion and that high EGFR expression is an independent predictor of poor prognosis in IBC. Further, recent studies have shown that targeting EGFR enhances the chemosensitivity of TNBC cells by rewiring apoptotic signaling networks in TNBC. These studies indicate that EGFR-targeted therapy might have a promising role in TNBC and IBC. Further studies of the role of EGFR in TNBC and IBC are needed to better understand the best way to use EGFR-targeted therapy—e.g., as a chemosensitizer or to prevent metastases—to treat these aggressive diseases.
In oncology, the search for new molecular predictors of prognosis (prognostic factors) and response to therapy (predictive factors) is an area of intense investigation. Progress in this area will undoubtedly transform cancer drug therapy from use of non-targeted antitumor agents in unselected patients to use of targeted antitumor agents in patients selected on the basis of tumor molecular biology. Among the most notable cancer molecular targets identified to date are the members of the epidermal growth factor receptor (EGFR)/ErbB family: EGFR (also known as ErbB1 and HER1), HER2 (also known as HER2/neu and ErbB2), ErbB3 (also known as HER3), and ErbB4 (also known as HER4) . HER2, which is overexpressed in 20% to 25% of breast cancers, is the most well established therapeutic target in breast cancer.
EGFR overexpression in breast cancer is associated with large tumor size, poor differentiation, and poor clinical outcomes [2, 3]. Though EGFR overexpression is observed in all subtypes of breast cancer, EGFR is more frequently overexpressed in triple-negative breast cancer (TNBC) and inflammatory breast cancer (IBC), which are especially aggressive [4–6]. Treatment of patients with these phenotypes has been challenging not only because of the aggressive behavior of these diseases but also because of the lack of established clinically relevant treatment targets. The role of EGFR in breast cancer has been scrutinized, and several therapies that target EGFR, including gefitinib, cetuximab, lapatinib, and others, have been developed. However, results of clinical studies of EGFR-targeted therapy in breast cancer have been disappointing.
Here, we review the latest studies of EGFR signaling and EGFR-targeted therapies in breast cancer, with a special focus on the relationship between EGFR and TNBC and IBC. We summarize the basic biological characteristics of EGFR and the latest findings from clinical trials of EGFR-targeted therapies for breast cancer.
The human EGFR family comprises 4 closely related receptors that are transmembrane glycoproteins containing an extracellular ligand binding domain and an intracellular receptor tyrosine kinase domain. The major signaling pathways activated by EGFR receptors are mediated by PI3 kinase, Ras-Raf-MAPK, JNK, and PLCγ and result in a plethora of biological functions (Figure 1) [7, 8]. At the cellular level, the ligands not only induce cell proliferation but also alter adhesion and motility and protect against apoptosis; at the physiological level, the ligands promote invasion and angiogenesis . Activation of members of the EGFR family promotes scattering and invasion of breast epithelial cells in 3-dimensional culture, which is associated with loss of cell polarization and other features of epithelial differentiation . In vitroany of these effects may contribute to the malignant phenotype. Dysregulation of EGFR pathways by overexpression or constitutive activation can promote tumor processes including angiogenesis and metastasis and is associated with poor prognosis in many human malignancies [3, 11], . In addition to cross-talk between members of the EGFR family, there is evidence for significant interactions between EGFR family members and other receptor tyrosine kinases, such as c-MET and IGF-1R, and it is possible that such alternative signaling pathways are linked to resistance to EGFR-targeted therapies .
It has been reported that the expression of both EGFR and HER2 is inversely correlated with estrogen receptor (ER) status, and EGFRHER2 heterodimers have been shown to increase the metastatic potential of breast cancer cell lines . The rate of overexpression of EGFR is particularly high in TNBC, and the negative impact of EGFR overexpression is particularly pronounced in TNBC. Thus, EGFR has potential as a therapeutic target in TNBC, for which there are no specific targeted therapies at present.
One of the mechanisms of EGFR overexpression is amplification of the EGFR gene, which has been described in oligodendroglioma,  glioblastoma, lung cancer,  gastric cancer, and breast cancer . EGFR gene amplification is infrequent in breast cancers overall: previous studies showed EGFR gene amplification in 0.8% to 14% of tumors [18, 19]. However, gene amplification has been shown in approximately 25% of cases of metaplastic breast cancer, a specific phenotype of TNBC [20–23].
Another mechanism of EGFR overexpression is through activating mutations of EGFR, which have been demonstrated in central nervous system tumors and lung cancer but is rare in breast cancer. Weber et al. found mutations of EGFR in 7 of 48 sporadic breast carcinomas and 11 of 24 hereditary breast carcinomas . Surprisingly, mutations were found in both stromal and neoplastic epithelium. These authors also showed that EGFR mutations occurred at a significantly higher frequency in hereditary than in sporadic breast cancer (P=0.0079) and that the majority of missense mutations were in the tyrosine kinase domain of EGFR exon 20. These data are in agreement with the fact that the rate of TNBC is higher among patients with hereditary breast cancer than among patients with sporadic breast cancer . Weber et al. suggested that future clinical trials employing molecular targeted therapy should evaluate EGFR mutations not only in neoplastic epithelia but also in the surrounding tumor stroma. This will establish the role of EGFR mutations in response to therapy and their value in predicting individual variation in response. In breast cancer, as has previously been done in lung cancer (with in-frame deletion of exon 19 and point mutations of exon 21) [25, 26], identification of EGFR mutations may be used to select patients most likely to respond to EGFR-targeted therapies.
In breast cancer, EGFR expression level or gene mutation status is increasingly being used to select patients for particular treatments. However, whether EGFR is truly a predictive biomarker remains to be proven.
In several malignancies, EGFR alterations occur at an advanced stage of malignancy characterized by metastatic competence [27–29], and EGFR is thought to promote cancer cell migration and invasion. Recently, EGFR has been shown to promote epithelial-mesenchymal transition (EMT), a process by which cells undergo a morphologic switch from a polarized epithelial phenotype to a mesenchymal fibroblastoid phenotype, in a variety of epithelial cell lines. EMT has been identified as a key process of migration and tumor invasion [30, 31]. In breast cancer, there is some evidence that EMT is involved in development of the normal mammary gland, but EMT is likely to be most important in tumor progression [32, 33].
EMT is characterized by the loss of epithelial markers (E-cadherin and cytokeratins) and the presence of mesenchymal markers (vimentin and fibronectin). Reduction of the E-cadherin level has been associated with metastatic breast cancer, which indicates the importance of EMT in metastasis [34, 35]. EMT can be induced in vitro in several epithelial cell lines by growth factors such as EGFR, scatter factor/hepatocyte growth factor, fibroblast growth factors, and insulin-like growth factors 1 and 2 .
EMT ultimately results in a transcriptional reprogramming of the tumor cell and its transition to a mesenchymal phenotype, promoted by abnormal survival signals through plateletderived growth factor receptor, fibroblast growth factor receptor, cMET, transforming growth factor beta-receptor, insulin-like growth factor 1 receptor, ERKand AKT. These proteins and pathways can be targeted by molecular targeted therapies directed toward EGFR, insulin-like growth factor 1 receptor, mammalian target of rapamycin, vascular endothelial growth factor, and cKIT . We have shown that erlotinib, an EGFR-tyrosine kinase inhibitor (TKI), inhibited cell motility and invasiveness and transformed IBC cells from a mesenchymal phenotype to an epithelial phenotype . The fact that cells treated with erlotinib showed higher expression of E-cadherin and lower expression of vimentin suggested that the antimetastatic effect of erlotinib might be through inhibition of EMT . Thus, EGFR is highly involved in EMT and might be a key target for inhibiting tumor metastasis.
Downstream of EGFR, the Ras-ERK pathway has been shown to also regulate EMT, tumor invasion, and metastasis. Activation of RSK by ERK is known to induce mesenchymal motility and invasion in cancer cells . ERK has also been implicated in transforming growth factor-beta signaling: it was shown that transforming growth factor-beta1 induced EMT through activation of ERK1 . However, recently, Ras-induced EMT that produced a dramatic morphological change in nontransformed human epithelial cell lines was shown to involve ERK2, not ERK1. The ERK2-induced EMT involved Fra1, a transcription factor that regulates expression of ZEB1/2, a marker associated with EMT .
At present, classification of breast cancers on the basis of common molecular features is indispensable for selecting the best treatment strategies. EGFR overexpression is found in at least 50% of cases of TNBC, which is a higher expression rate than the rates seen in other breast cancer subtypes . Because of the high rate of overexpression of EGFR in TNBC, EGFR inhibitors are among the targeted agents being developed for treatment of TNBC.
TNBCs are negative for ER, progesterone receptor (PgR), and HER2 and are generally accepted as a clinical surrogate for basal-like breast cancer, one of the intrinsic subtypes based on microarray analysis . EGFR and cytokeratin 5/6 are readily available positive markers for basal-like breast cancer that are applied to standard pathology specimens in clinics. Though TNBC is generally accepted as a clinical surrogate for basal-like breast cancer, it was recently hypothesized that TNBC is heterogeneous, and 50% to 85% of TNBC tumors were estimated to be true basal-like breast cancer [43, 44]. Recently, Lehmann et al. reported there are 6 subtypes of TNBC, and the EGF pathway is one of the top canonical pathways for Basal-like 2 and Mesenchymal-like subtypes .
Lee et al. recently showed that EGFR-targeted therapy may be used to enhance the initial sensitivity of TNBC cells to cytotoxic therapy , they identified new strategies to enhance the initial chemosensitivity of TNBC cells. They showed that enhanced cell death observed with the use of time-staggered erlotinib-doxorubicin combinations was directly mediated by sustained EGFR inhibition. After sustained EGFR inhibition, oncogene signatures such as RAS and MYC signatures were dramatically decreased in TNBC cells. The authors analyzed multiple types of quantitative data using advanced computation network modeling and found that the most effective strategy for killing aggressive TNBC cells was a time- and orderdependent combination of genotoxic agents with small molecule EGFR inhibitors, such as doxorubicin and erlotinib respectively. They also found that “the enhanced treatment efficacy resulted from dynamic network rewiring of an oncogenic signature maintained by active EGFR signaling to unmask an apoptotic process that involves activation of caspase-8.” They concluded that “phosphorylation of EGFR may constitute a useful biomarker of response to time-staggered inhibition in some tumor types that are EGFR driven, such as TNBC and lung cancer.”[46
IBC, the most clinically aggressive subtype of breast cancer, is also associated with EGFR overexpression: 30% of IBCs express EGFR . Several studies have documented a high frequency of negative ER and PgR status, up to 50%, and a high incidence of HER2 overexpression, up to 40%, in IBC tumors [47, 48]. Lack of expression of ER and PgR is clearly one of the reasons for the poor prognosis of IBC, but whether HER2 overexpression has a prognostic role in IBC has yet to be established. In contrast, EGFR overexpression is clearly correlated with poor prognosis in patients with IBC . Patients with EGFR-positive IBC have a worse 5-year overall survival rate than patients with EGFR-negative tumors, and EGFR expression in IBC is associated with an increased risk of recurrence .
Because conventional chemotherapy regimens are not sufficient for the treatment of IBC, new therapies for IBC are needed. Among the potential candidates are therapies that target the EGFR pathway. An in vitro study showed that gefitinib, an EGFR-TKI, suppressed the growth of SUM149 cells, which overexpress EGFR and lack ER expression and are widely used as a model of aggressive IBC . Another in vitro study showed that treatment with neutralizing antibody against amphiregulin, one of the ligands of EGFR, decreased EGFR activity and reduced cell proliferation in SUM149 cells . These findings led to an ongoing study of panitumumab (an anti-EGFR antibody), albumin-bound paclitaxel (Abraxane), and carboplatin in IBC (NCT01036087). This study will elucidate the biological impact of anti-EGFR therapy on IBC tumors.
Tyrosine kinases are associated with the cytoplasmic domains of growth factor receptors and oncoproteins, and many tyrosine kinases have the potential to cause transformation if they are mutated or overexpressed. Tyrosine kinases therefore represent an excellent target for the development of cancer drugs [52, 53]. The small molecule inhibitors of EGFR are TKIs that bind to the ATP-binding site in the tyrosine kinase domain of EGFR . TKIs act directly on EGFR but also affect the activities of other kinases in the cell; thus, there is some potential for unfavorable side effects. The small molecule inhibitors of EGFR can be classified as pure EGFR TKIs and dual EGFR and HER2 TKIs.
Whereas small molecule EGFR TKIs are not completely specific for EGFR tyrosine kinase receptors, MAbs are completely specific for the EGFR tyrosine kinase, which could be advantageous. MAbs have less capacity to reach normal intestinal epithelium, which appears to be an advantage for MAb-mediated EGFR blockade since diarrhea was a dose-limiting toxicity with the oral kinase inhibitor such as a Lapatinib but was not observed with the MAbs in general. Moreover, MAbs could work through other mechanisms, including activation of immune responses through mediating antibody-dependent cell-mediated cytotoxicity that is not seen with the use of small molecule EGFR inhibitors.
Gefitinib monotherapy did not significantly improve response rates in most studies [55–57]. In a phase II trial of erlotinib monotherapy in patients with metastatic breast cancer (n=69), no patients had a complete response, and only 2 had a partial response . These rather disappointing results may relate to the patient selection criteria: these trials did not select patients on the basis of EGFR expression; rather, they included unselected breast cancer patients who had often been heavily pretreated. Subgroup analysis of results of previous clinical trials may offer clues to optimal patient selection for EGFR-targeted therapy.
Two phase II clinical trials evaluated the efficacy of cetuximab alone or in combination with platinum-based chemotherapy and found that addition of cetuximab to carboplatin did not improve outcome [59, 60]. However in one of these trials , cetuximab alone was associated with a 6% response rate, suggesting that this drug could have some efficacy in selected patients. In that trial, the authors also investigated EGFR pathway activation using gene expression profiling with Agilent DNA microarrays. The findings suggested that cetuximab blocked expression of the EGFR pathway in only a minority of patients, indicating that most patients had alternate mechanisms of pathway activation .
Another phase II trial in patients with advanced breast cancer showed that cetuximab improved the overall response rate only in patients with TNBC: in this subgroup, overall response rates were 38% for patients treated with irinotecan and carboplatin and 49% for patients treated with irinotecan, carboplatin, and cetuximab . EGFR positivity was not an eligibility criterion for this trial, but retrospective assessment showed that EGFR positivity was significantly associated with the TNBC subtype (P<0.0001).
European researchers were the first to prove that anti-EGFR therapy can provide substantial clinical benefit for patients with TNBC . These researchers conducted a phase II randomized trial of cisplatin versus cisplatin plus cetuximab in 173 heavily pretreated patients with TNBC. The overall response rate was twice as high with the cisplatin-cetuximab combination as it was with cisplatin alone (20% vs. 10.3%). Also, the median progression-free survival time was more than twice as long with cetuximab as it was with cisplatin alone (3.7 months vs. 1.5 months).
Panitumumab, an antibody targeting EGFR, has also been investigated for its efficacy in patients with TNBC. In a phase II trial of panitumumab in combination with 5-fluorouracil, epidoxorubicin, and cyclophosphamide followed by docetaxel for neoadjuvant therapy, the overall response rate was 80% . A trial of panitumumab in combination with carboplatin for patients with metastatic TNBC is ongoing (NCT00894504).
Taken together, the findings to date on EGFR-targeted agents suggest that further investigation of these agents in patients with TNBC is warranted .
In a phase I clinical trial, lapatinib monotherapy showed clinical activity in patients with trastuzumab-refractory breast cancer. Four of the 59 evaluable patients with metastatic breast cancer positive for both EGFR and HER2, including 2 with IBC, had a partial response . In a phase II trial of lapatinib monotherapy for heavily pretreated patients with IBC, the response rate was 50% among the 30 patients with HER2-positive tumors but only 7% among the 15 patients with HER2-negative, EGFR-positive tumors . The investigators also evaluated ErbB3 status and found that co-expression of phosphorylated HER2 and phosphorylated ErbB3 was associated with a better response rate. In a phase II trial, BIBW 2992, a novel oral, irreversible EGFR and HER2 inhibitor, had a rate of clinical benefit of 14% in 21 patients with metastatic TNBC . At present, response to dual EGFR and HER2 inhibitors seems to depend on HER2 expression rather than EGFR expression. Further studies of dual inhibitors are required.
Previous trials showed that many patients with EGFR-expressing tumors did not respond to EGFR-targeted therapy, which suggests that EGFR expression alone does not indicate tumor cell dependence on the EGFR pathway. There is now evidence for significant interactions of EGFR with other receptor tyrosine kinases, such as c-MET and IGF-1R, and it is possible that such alternative signaling pathways are linked to resistance to targeted therapies . Thus, we have to consider combining EGFR-targeted therapy with drugs targeting these alternate signaling pathways to improve efficacy. Further, EGFR activation drives migration and invasion through EMT and alters chemosensitivity by rewiring the apoptotic signaling network. Therefore, EGFR-targeted therapy may not produce cancer shrinkage due to suppression of cell proliferation; rather, EGFR-targeted therapy may produce a therapeutic effect by inhibiting metastasis or sensitizing cancer cells to the effects of cytotoxic therapy.
Recent studies confirm that EGFR is a potentially important target in breast cancer, especially TNBC, basal-like breast cancer, and IBC. EGFR-targeted therapy has finally shown some promise in terms of improving outcomes in breast cancer patients, but molecular prognostic and predictive factors need to be identified to optimize selection of patients for EGFR-targeted therapies. Mechanistic, hypothesis-oriented clinical trials are needed rather than trials based on the assumption that EGFR-targeted therapy will be effective against EGFR-overexpressing breast cancer. Further, recent data suggest that expecting EGFR-targeted therapy to produce tumor shrinkage may be unrealistic. EGFR-targeted therapy may be most effective as a chemosensitizer or therapy designed to prevent metastases.
Financial support: This research was supported by NIH grant R01 CA123318 (NT Ueno), MD Anderson’s Cancer Center Support Grant, CA016672, the Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, a State of Texas Rare and Aggressive Breast Cancer Research Program grant (NT Ueno), and Susan G. Komen Postdoctoral Fellowship KG091192 (D Zhang).
Conflicts of interest: The authors have no conflicts to declare.