1 Introduction

Breast cancer (BC) is the most frequently diagnosed cancer in women worldwide, with 1.7 million new cases diagnosed in 2012 [1]. In the United States alone, 231,840 new cases and 40,290 related deaths are expected to be seen in 2015 [2]. This heterogeneous disease is classified into several molecular subtypes, depending on the presence of specific cell surface receptors, including the estrogen receptor (ER), the progesterone receptor (PR), and the human epidermal growth factor 2 receptor (HER2) [3]. Luminal A BCs are ER+ and/or PR+ but HER2−. These are the most commonly seen BCs and have the best prognosis. Luminal B BCs are ER+ and/or PR+ and sometimes HER2+. These tumors have a higher proliferative index and are more aggressive. HER2+ BCs are ER− and PR− but HER2+. This subtype usually presents at a younger age with a poorer tumor grade and lymph node involvement, but the prognosis has improved dramatically since the clinical implementation of Herceptin, an anti-HER2 antibody. About 20% of BCs are triple negative (TNBC) or basal-like, that is, they are ER−, PR−, and HER2−. These tumors are often aggressive, have poorer prognosis, and lack any targeted therapies.

Research has focused extensively on the role of cell surface receptors like HER2 in the pathobiology of BC. There are numerous families of cell surface receptors, like receptor tyrosine kinases (RTKs), one example being HER2, and G protein-coupled receptors, that sense extracellular cues and transmit them into intracellular messages that regulate cell growth, proliferation, survival, migration, and differentiation. These receptors are often deregulated in BC and lead to tumor growth and metastasis. This chapter will focus on the identification of the receptors most often deregulated in BC, the common signaling pathways they activate, and the crosstalk that links them to one another.

2 RTKs and Their Downstream Signaling Targets

RTKs are cell surface receptors found on a diversity of cell types. All RTKs comprise an N-terminal extracellular ligand-binding domain, a single-pass transmembrane domain, and a C-terminal tyrosine kinase domain. Ligand binding induces a conformational change leading to the receptor homo- or heterodimerization and the consequent autophosphorylation of a series of tyrosine residues in the C-terminal tail. The phosphorylated tyrosines then act as docking sites for the SRC homology 2 (SH2) and phosphotyrosine-binding (PTB) domain-containing proteins, many of which are shared by the different RTKs. The RTK signaling program converges on the two major signaling pathways, namely, the phosphoinositide 3-kinase-protein kinase B/AKT (PI3K-PKB/AKT) and the rat sarcoma-mitogen-activated protein kinase/ERK (Ras-MAPK/ERK), that go on to regulate critical cellular processes like cell growth, proliferation, differentiation, migration, and apoptosis (Fig. 1.1).

Fig. 1.1
figure 1

RTK and GPCR signaling networks

Full size image

One of the SH2 domain-containing effectors of RTKs is the regulatory subunit of the class I PI3K (p85), which when bound to the activated RTK or one of its tyrosine phosphorylated adaptors relieves its inhibition of the p110 catalytic subunit of PI3K, thereby leading to its activation [4]. PI3K p110 then phosphorylates a resident membrane lipid, phosphatidylinositol 4,5-bisphosphate (PIP2), to generate phosphatidylinositol 3,4,5-trisphosphate (PIP3), a major lipid second messenger [5]. The phosphatase and tensin homolog deleted on chromosome 10 (PTEN) counteracts the action of PI3K and converts PIP3 back to PIP2 [6, 7]. PIP3 recruits the pleckstrin homology (PH) domain-containing proteins to the membrane, of which there are more than 250 in the human genome, including AKT and the 3-phosphoinositide-dependent protein kinase 1 (PDK1) [8]. AKT is phosphorylated by PDK1 on threonine 308 (T308) and consequently by the mammalian target of rapamycin complex 2 (mTORC2) on serine 473 (S473), leading to its full activation [9, 10]. AKT then activates the mammalian target of rapamycin complex 1 (mTORC1) through two distinct pathways. AKT phosphorylates and suppresses the GTPase-activating protein (GAP) activity of the tuberous sclerosis complex 2 (TSC2) toward the Ras homolog enriched in the brain (Rheb) [11,12,13]. On the other hand, AKT phosphorylates and inhibits proline-rich AKT substrate of 40 KDa (PRAS40), which is implicated in the regulation of mTORC1 [14, 15]. mTORC1 regulates cell growth by controlling mRNA translation, via direct phosphorylation of the S6 kinase (p70S6K) and the 4E binding proteins (4EBPs) [16, 17]. AKT also phosphorylates the forkhead box O transcription factors (FOXOs), which results in their nuclear exclusion and proteasomal degradation, thus releasing cells from the FOXO-mediated cell cycle arrest [18,19,20]. The deactivation of FOXO, along with another target of AKT, the B-cell lymphoma 2 (BCL2)-associated agonist of cell death (BAD), coordinately represses apoptosis [21]. Finally, the AKT-mediated inhibition of glycogen synthase kinase 3 (GSK3) inhibits nuclear export and proteasomal degradation of cyclin D1, thus leading to its nuclear accumulation and induction of cell proliferation [22]. Thus, the PI3K-AKT pathway mainly regulates the cellular growth, proliferation, and survival programs in cells. Other than RTK deregulation, activating mutations in PI3K or deletion of PTEN is often found in BCs and further drives the oncogenic program in cells [23].

The Ras-ERK pathway is the other major signaling network that is modulated by the RTKs. The Src homology 2 domain-containing (SHC) and the growth factor receptor bound 2 (GRB2) are the main adaptor proteins that link the activated RTKs to the Ras-ERK pathway [24]. The RTKs interact with and activate SHC directly, which then recruits GRB2 to the cell membrane. Alternatively, GRB2 can also interact with RTKs directly or through another adaptor protein, like one of the insulin receptor substrates (IRSs) [25, 26]. GRB2 associates with the son of sevenless (SOS), which then recruits and activates Ras, by acting as a guanine nucleotide exchange factor (GEF), converting the GDP-bound Ras into the active GTP-bound form [27]. In a sequential manner, Ras activates Raf, which phosphorylates and activates MEK, which in turn phosphorylates and activates ERK [28, 29]. ERK is a serine-threonine kinase that has hundreds of cytoplasmic and nuclear targets. For example, ERK translocate to the nucleus and activates transcription factors like Ets-like gene 1 (Elk1) and c-Myc [30, 31]. In the nucleus, ERK also activates the mitogen- and stress-activated protein kinases (MSKs), which activate transcription factors like the cyclic AMP-responsive element-binding protein (CREB) and the activating transcription factor 1 (ATF1) [32]. In the cytoplasm, ERK phosphorylates the p90 ribosomal S6 kinases (RSKs), which inhibit apoptosis by phosphorylating BAD, but also translocate to the nucleus and activate transcription factors like CREB and c-Fos [33,34,35,36]. ERK-mediated phosphorylation of the MAPK-interacting kinases (MNKs) induces mRNA translation through phosphorylation of the eukaryotic initiation factor 4E (eIF4E) [37, 38]. The Ras-ERK pathway thus controls diverse cellular processes including cell growth, proliferation, differentiation, migration, and apoptosis.

In addition to the parallel activation immediately downstream of the receptors, coordination between the PI3K-AKT and the Ras-ERK signaling pathways can also be achieved by the interaction of activated Ras with the PI3K p110 catalytic subunit, independently of p85, leading to the PI3K-AKT pathway activation (Fig. 1.2) [39]. Like AKT, ERK and RSK can also phosphorylate and inhibit TSC2, leading to mTORC1 activation [40, 41]. The two pathways are also subject to the multiple levels of feedback inhibition. MEK promotes the membrane localization of PTEN, which downregulates the PI3K-AKT signaling pathway, while AKT can phosphorylate and inhibit Raf [42,43,44]. Furthermore, mTORC1, S6K, and ERK can downregulate both pathways by phosphorylating RTK substrates, like IRS1, on multiple inhibitory serine residues [45,46,47]. Thus, multiple feedback loops and crosstalk between the PI3K-AKT and the Ras-ERK pathways orchestrate the dynamic and intricate, context-dependent effects of multiple growth factors through their cognate RTKs.

Fig. 1.2
figure 2

Crosstalk between the PI3K-AKT and the Ras-ERK signaling pathways. Green lines indicate activation and red lines indicate inhibition. Solid black lines indicate a direct interaction, while dashed black lines indicate an indirect interaction

Full size image

3 RTKs Often Deregulated in BC

The deregulation of RTK signaling plays an important role in the pathophysiology of many cancers, including BC [48]. Several mechanisms lead to the deregulation of RTK signaling, including RTK gene amplifications, activating mutations, protein overexpression, ligand overexpression or hyperactivation, and crosstalk with other cellular signaling components. Members of the ERBB family, MET, the insulin receptor (INSR), and the insulin-like growth factor receptor (IGF1R) are RTKs that are most often deregulated in BC.

3.1 HER2

The amplification of the HER2 gene, a member of the ERBB family of RTKs, is seen in approximately 20% of BCs, and HER2 overexpression correlates with a worse BC prognosis [49,50,51]. In these patients, HER2 overexpression correlates with tumor size, grade, proliferative index, aneuploidy, lack of steroid hormone receptors, and metastatic disease. HER2 (also named ERBB2 or NEU), belongs to the ERBB family, with three additional members: the epidermal growth factor receptor (EGFR, also named ERBB1), ERBB3, and ERBB4, all of which have been shown to be overexpressed and/or hyperactivated in BC to varying degrees. For example, EGFR is often overexpressed in basal-like TNBC [52]. There are 11 ligands that activate this family of RTKs, and they can be subdivided into three groups [53]. The first includes the epidermal growth factor (EGF), the transforming growth factor α (TGFα), and amphiregulin, which bind specifically to EGFR. The second includes betacellulin, heparin-binding EGF (HB-EGF), and epiregulin, which bind EGFR and ERBB4. Neuregulins (NRGs) make up the third group of ligands and are further subdivided into two subgroups, based on the ability to activate ERBB3 and ERBB4 (NRG1 and NRG2) or ERBB4 alone (NRG3 and NRG4). All ligands exist as membrane-anchored precursors, often co-expressed and even overexpressed with the ERBBs in the same cancer cells. Metalloproteases, mainly of the a disintegrin and metalloprotease (ADAM) family, cleave the precursors, leading to ectodomain shedding and activation of ERBB signaling in an autocrine or paracrine fashion [54].

Like other prototypical RTKs, all ERBBs can form functional homo- or heterodimers, with the exception of HER2, which does not appear to bind a ligand, and ERBB3, which is impaired in the intrinsic kinase activity and thus cannot form functional homodimers [55, 56]. Though HER2 is not self-autonomous, its extracellular domain conformation mimics that of the ligand-bound receptor, thus allowing HER2 to form functional heterodimers with other ERBBs [57]. HER2 is in fact the preferred binding partner of other ERBBs, and intriguingly the HER2-ERBB3 heterodimer is the most mitogenic and transforming of all the receptor combinations [58,59,60,61]. The C terminus of each of the ERBBs is unique (11–25% identity) and is able to bind to a diversity of intracellular targets. All of the ERBB members activate the Ras-ERK signaling pathway by directly interacting with the adaptor proteins SHC and GRB2 [62]. The regulatory subunit of PI3K (p85) directly interacts with ERBB3 and ERBB4. ERBB3 has the most [6] binding sites for p85, while EGFR and HER2 lack them all together, thus the HER2-ERBB3 heterodimer is the most potent activator of the PI3K-AKT signaling pathway, promoting cell growth, proliferation, and survival [63, 64]. Alternatively, ERBBs can activate the PI3K pathway through Ras. Together with the multitude of ligands, the different combinations of receptor dimers, and the unique C-terminal tails, this family of RTKs is capable of regulating diverse cellular processes implicated in cell growth, proliferation, differentiation, migration, and apoptosis.

3.2 MET

The hepatocyte growth factor receptor or MET is another RTK that is overexpressed in about 20% of BCs, particularly in the basal-like TNBCs [65]. Hepatocyte growth factor (HGF) is the only known ligand of MET, and it is often co-expressed with its receptor in the same tumor cells, particularly in the leading edge of the tumor [66, 67]. The expression of both, the receptor and the ligand, correlates with tumor grade, proliferative index, metastatic disease, and poor prognosis [68,69,70,71,72,73,74]. The HGF-mediated activation of MET leads to activation of the Ras-ERK pathway through the direct interaction of SHC and GRB2 with the receptor or through the recruitment of an insulin-like substrate (IRS)-like adaptor, the GRB2-associated-binding protein 1 (GAB1). The p85 regulatory subunit of PI3K also interacts with MET directly or through GAB1 and leads to activation of the PI3K-AKT pathway [75].

3.3 INSR

The INSR is overexpressed in as many as 80% of BCs and is associated with poor survival [76, 77]. The INSR is encoded by a gene composed of 22 exons found on chromosome 19. From this single gene, two receptor isoforms, INSR-A and INSR-B, are expressed as a result of alternative splicing. These two isoforms differ in inclusion/exclusion of exon 11, a 36 bp region encoding a 12 amino acid peptide located at the C-terminal end of the INSR alpha subunit [78]. INSR-B represents the full-length isoform and is expressed in insulin-responsive tissues including the liver, muscle, and adipose tissue. Conversely, INSR-A is expressed from the spliced transcript that lacks exon 11 and plays a significant role in fetal development by regulating cell growth and proliferation [79, 80]. The INSR-A and INSR-B isoforms display unique ligand specificity and downstream signaling potential. INSR-A exhibits an almost twofold higher affinity for insulin as compared to INSR-B and has a much stronger affinity for the insulin-like growth factor II (IGFII) [81,82,83]. INSR-A is the prevailing isoform overexpressed in both BC cells in culture and patient tumors [77, 84]. Therefore, increased INSR-A expression may negatively impact BC development, particularly in the context of hyperinsulinemia, as in the cases of diabetes or obesity. Indeed, hyperinsulinemia is an adverse prognostic factor in BC that is associated with increased risk of recurrence or death [85]. INSR-A expression is also elevated beyond that of the related IGF1R in some BCs suggesting INSR-A plays a role in mediating the growth-promoting effects of IGF-II in breast tumorigenesis [84, 86].

On the cell surface, the INSR exists as a heterotetrameric protein comprised of two extracellular alpha subunits and two transmembrane beta subunits. The beta subunit of the receptor possesses tyrosine kinase activity, which is stimulated upon binding of the ligand to the alpha subunit [87, 88]. Upon activation, INSR phosphorylates a number of substrates including IRS1–4, SHC, and GAB1 [89]. IRSs and GAB1 recruit the p85 regulatory subunit of PI3K, leading to the PI3K-AKT pathway activation. The Ras-ERK pathway is activated by the recruitment of GRB2-SOS complex by SHC or IRSs [26, 90]. INSR-B regulates the metabolic effects of insulin mainly through the PI3K-AKT pathway, while INSR-A activates the mitogenic program through both, the Ras-ERK and the PI3K-AKT pathways [91, 92]. Consequently, inhibition of INSR-A is actively being explored as a therapeutic option in breast and other cancers with clinical trials focusing on testing small molecule inhibitors and monoclonal antibodies directed against key components of these signaling networks [93]. Systemic modification of receptor ligands represents another strategy for targeting INSR signaling in cancer. For example, reduction in circulating insulin levels via administration of the antidiabetic drug metformin is being explored as a treatment option for cancers associated with obesity and hyperinsulinemia, especially BC. Indeed, administration of metformin to early-stage, nondiabetic BC patients led to reductions in circulating insulin and cancer cell proliferation, as well as suppressed INSR activity as indicated by reductions in AKT and ERK signaling [94].

3.4 IGF1R

Close to 50% of human breast tumors express the activated form of IGF1R, and gene expression signatures consistent with IGF1R activation are associated with poor outcome in BC patients [95, 96]. The IGF1R is homologous to the INSR but exhibits preferential binding to IGFI and IGFII over insulin. It is also a heterotetrameric protein complex consisting of two extraceulluar alpha subunits and two transmembrane beta subunits but plays a more significant role in the regulation of mitogenic signaling. The IGF1R shares numerous binding partners and effector proteins with the INSR, including IRSs and SHC adaptors, and is known to stimulate cell growth and proliferation via activation of the PI3K-AKT and Ras-ERK signaling pathways [93, 97]. A second IGF receptor, namely, IGF2R, is also commonly expressed by numerous cells; however, it lacks catalytic activity and is not involved in intracellular signaling [98]. Instead, IGF2R exhibits a high affinity for IGFII and is thought to sequester the growth factor from stimulating IGF1R [97, 99]. As a result, IGF2R may exhibit tumor suppressor properties by decreasing the bioactivity of IGFII and indirectly modulating signaling by IGF1R.

Due to their homology and strong structural similarities, the INSR and IGF1R have the ability to form hybrid receptors composed of one hemireceptor of each type. In addition, the two INSR isoforms can also combine to form hybrids, generating the potential for multiple insulin and IGF-sensitive receptors (INSR-A, INSR-B, INSR-A/B, IGF1R, INSR/IGF1R) to be expressed by a single cell. Hybrid receptors appear to bind IGFI with a higher affinity than insulin, and they exhibit different ligand specificities depending on the INSR isoform present. For example, INSR-A/IGF1R hybrids bind IGFI, IGFII, and insulin, while INSR-B/IGF1R hybrids typically bind IGF1 [91]. Since cancer cells frequently express high levels of both INSR and IGF1R, it is not surprising that they also overexpress hybrid receptors. Indeed, human breast tumors express high levels of hybrid receptors, and most of the effects of IGF1 are believed to be mediated by INSR-A/IGF1R hybrids. Furthermore, BC cells are known to secrete IGFII, creating the potential for autocrine stimulation of tumor cell growth and proliferation via activation of INSR-A, IGF1R, and INSR/IGF1R hybrid receptors [84, 100]. Consequently, human BC cells are highly sensitive to the growth-promoting effects of insulin and IGFs, and INSR/IGF1R expression may be a key event in tumor development and growth.

4 GPCRs and Their Downstream Signaling Targets

The G protein-coupled receptors (GPCRs) are the largest group of cell surface receptors that regulate cell motility, growth, proliferation, differentiation, and survival. The discovery of the Mas oncogene, a GPCR, in 1986 provided the first direct link between GPCRs and their role in cellular transformation [101]. Since then, many GPCRs where shown to be overexpressed or mutated in a diversity of cancers, including BC.

GPCRs are seven-pass transmembrane domain-containing receptors with an intracellular C-terminal tail that interacts with the heterotrimeric G proteins [102]. Upon ligand binding, the receptor undergoes a conformational change that allows it to act as a GEF, converting the GDP-bound G protein α subunit to the GTP-bound, active form. This causes the G protein α subunit to dissociate from the βγ subunits, initiating a multitude of signaling cascades. There are numerous G protein subtypes, each with unique signaling abilities. For example, the Gα12/13 activates several Rho GEFs leading to activation of Rho, a small GTPase that regulates cytoskeletal dynamics and cell motility, largely implicated in cancer metastasis. Gαq/11 activates the phospholipase C beta (PLCβ), which initiates the calcium and diacylglycerol (DAG) signaling cascades that regulate cell motility, proliferation, and gene expression. The GPCR signaling also crosstalks to the PI3K-AKT and the RAS-ERK pathways. The DAG-activated protein kinase C (PKC) phosphorylates and activates Raf, thereby leading to the activation of ERK [103]. The Gβγ subunits bind directly to PI3Kγ and activate the PI3K-AKT signaling pathway [104].

5 GPCRs Often Deregulated in BC

5.1 PAR1

The protease activated receptor 1 (PAR1) is a GPCR that is overexpressed in TNBC and correlates with metastatic disease and poor prognosis [105, 106]. The zinc-dependent matrix metalloprotease 1 (MMP1), thrombin, and other proteases cleave the extracellular domain of PAR1 exposing a new N terminus that binds to and activates the receptor [107, 108]. PAR1 then couples to multiple G proteins (Gαq/11, Gαi/o, Gα12/13) to regulate cell migration and proliferation, in part through the activation of Rho and ERK, respectively. PAR1 has been shown to be required and sufficient for the regulation of growth and invasion of BC cells in a mouse xenograft model [107, 109].

5.2 GPR161

The GPR161 is another GPCR that is overexpressed in TNBC and correlates with cancer relapse [110]. Overexpression of GPR161 in mammary epithelial cells transforms them via a yet unidentified mTORC1-dependent signaling pathway [110].

5.3 Wnt

The Wnt signaling pathway is hyperactivated in basal-like BCs and correlates with poor survival [111, 112]. The canonical Wnt signaling pathway results in nuclear accumulation of β-catenin, where it acts as a transcriptional coactivator for the T-cell factor/lymphoid enhancer factor (TCF/LEF) family of transcription factors [113]. In the absence of the Wnt signal, β-catenin is sequestered in the cytoplasm by a destruction complex, containing GSK3β, which targets β-catenin for proteasomal degradation. Frizzled (FZD) is the GPCR for the Wnt family of ligands. When FZD is activated by the Wnt ligand, it acts together with the co-receptors, the low-density lipoprotein receptor-related protein 5 and 6 (LRP5/6), to disrupt the β-catenin destruction complex. This allows β-catenin to accumulate in the cytoplasm and translocate to the nucleus to activate its transcriptional program. The knockdown of FZD7 in TNBC cell lines reduces expression of β-catenin target genes, the transformation of these cells in vitro, and their ability to form tumors in vivo [114]. In addition, more than 40% of invasive breast tumors have a hypermethylation of the promoter, and therefore a strong downregulation of expression of the secreted frizzled-related proteins (sFRPs), the negative regulators of the Wnt signaling pathway [115, 116].

6 Crosstalk Between RTKs and GPCRs

The RTKs crosstalk with each other through multiple feedback and transactivation mechanisms. For example, MET can interact with and be transactivated by ERBBs, thus synergizing in the regulation of the downstream pathway components [117, 118]. Furthermore, RTK signaling often parallels or synergizes with GPCR signaling. The GPCRs can be upstream or downstream of the RTKs, and GPCRs are under the transcriptional regulation of RTKs and vice versa [119]. Furthermore, GPCRs and RTKs can transactivate each other. For example, GPCR activation regulates ectodomain shedding of the ERBB ligands. The PAR1 and the Wnt pathway have been shown to transactivate EGFR and HER2 in this manner [115,116,117,123]. Thus, GPCRs can activate the PI3K-AKT and the Ras-ERK pathways directly or through transactivation of the RTKs. In addition, EGFR-mediated activation of ERK induces nuclear translocation of the pyruvate kinase (PKM), which regulates β-catenin transcriptional activity [124]. The expression of β-catenin target genes can further be induced through the AKT- or RSK-dependent inhibition of GSK3β or the direct phosphorylation of β-catenin by AKT [120,121,127].

7 Other Receptors Deregulated in BC

The tumor microenvironment is a complex milieu of cell surface and secreted factors that affect BC development and progression. Tumor-associated fibroblasts (TAFs), endothelial cells, and inflammatory cells comprise the majority of the tumor microenvironment and express factors that affect tumor progression. Tumor cells express a number of non-RTK and/or non-GPCR receptors, the discussion of which is beyond the scope of this chapter, that sense signals from the microenvironment and often integrate them into the common pathways described above. For example, plexins, the receptors of semaphorins, originally described for their role in axon guidance, have now been implicated in BC metastasis, in part due to their ability to be transactivated by HER2 and MET and to activate Rho signaling [128, 129]. Tumor cells also express a number of cytokine receptors that interpret the pro-inflammatory signaling from leukocytes, tumor-associated macrophages (TAMs), TAFs, and autocrine loops. Cytokine receptors can activate several pro-survival and proliferation pathways but can also transactivate RTKs [130]. Lastly, integrins and cadherins, the cell adhesion mediators, are often deregulated in metastatic BC and play a central role in the epithelial-to-mesenchymal transition as well as in the activation of oncogenic signaling. Integrins can feed into both the PI3K-AKT and the Ras-ERK signaling pathways [131, 132]. Integrins also regulate ERBB expression at the mRNA translation level, as well as interact directly with ERBBs and regulate their tyrosine kinase activity [133, 134]. Further insight into the complexity of these intracellular crosstalk networks will aid in the identification of effective therapeutic targets and the mechanisms of therapeutic resistance.

8 Outlook

BC is one of the most common cancers worldwide and the second leading cause of cancer-related death in women [135]. It is a heterogeneous disease with a complex molecular etiology. A great deal of research has focused on the mechanisms underlying BC development, growth, and progression. Dysregulated RTK signaling has been identified as a critical event in breast tumorigenesis. For example, HER2 and IR are overexpressed by 20 and 80% of BCs respectively, and mutation of PI3K, a key mediator of RTK signaling, is mutated in 35% of human breast tumors [77, 135, 136]. Identification of such oncogenic proteins has led to a deeper understanding of BC and allowed for the development of targeted therapies for the treatment of this disease. Nevertheless, additional research is required to characterize mechanisms of tumor initiation as well as therapeutic resistance. Indeed, crosstalk between different RTK pathways and the existence of signaling feedback mechanisms are poorly understood processes that play critical roles in BC development and resistance to therapy. In the future, fundamental studies focusing on these issues in vitro should be combined with clinical research and early phase clinical trials to further characterize the role of RTKs in BC and identify new targets for anticancer therapies.