Hostname: page-component-6898cfcb55-w5hgh Total loading time: 0 Render date: 2026-11-11T12:23:45.552Z Has data issue: false hasContentIssue false
  • English
  • Français

Targeting tumour-initiating cells to improve the cure rates for triple-negative breast cancer

Published online by Cambridge University Press:  26 July 2010

Anna L. Stratford,
Kristen Reipas,
Christopher Maxwell and
Sandra E. Dunn
Anna L. Stratford
Affiliation:
Department of Pediatrics, Child and Family Research Institute, University of British Columbia, Vancouver, BC, CanadaV5Z 4H4.
Kristen Reipas
Affiliation:
Department of Pediatrics, Child and Family Research Institute, University of British Columbia, Vancouver, BC, CanadaV5Z 4H4.
Christopher Maxwell
Affiliation:
Department of Pediatrics, Child and Family Research Institute, University of British Columbia, Vancouver, BC, CanadaV5Z 4H4.
Sandra E. Dunn*
Affiliation:
Department of Pediatrics, Child and Family Research Institute, University of British Columbia, Vancouver, BC, CanadaV5Z 4H4.
*
*Corresponding author: Sandra E. Dunn, Department of Pediatrics, Child and Family Research Institute, University of British Columbia, Vancouver, BC, CanadaV5Z 4H4. E-mail: sedunn@interchange.ubc.ca
Rights & Permissions [Opens in a new window]

Abstract

Tumour recurrence is one of the biggest challenges in breast cancer management because it affects 25–30% of women with breast cancer and the tumours are often incurable. Women with triple-negative breast cancer (TNBC – lacking expression of the oestrogen receptor, progesterone receptor and the receptor HER2/ERBB2) have the highest rates of early recurrence relative to other breast cancer subtypes. Early recurrence might be due to tumour-initiating cells (TICs), which are resistant to conventional therapies, can remain dormant and can subsequently give rise to secondary tumours. In breast cancer, TICs are identified by the cell-surface markers CD44+/CD24−/EpCAM+ and/or possess ALDH1 enzyme activity. This subpopulation has the ability to self-renew, grow as mammospheres and initiate tumour formation. Fuelling the problem of relapse is the fact that chemotherapy and radiation can induce or select for TICs; this was reported in preclinical models and more recently in women being treated for breast cancer. Thus, new therapeutic agents for TNBC are presently being sought to overcome this problem. Here we review the roles of receptor tyrosine kinases, signalling intermediates and transcription factors in sustaining the TIC subpopulation. Particular emphasis is placed on targeting these molecules in order to eliminate and/or prevent the induction of TICs and ultimately reduce the frequency of TNBC recurrence.

Information

Type
Review Article
Information
Expert Reviews in Molecular Medicine , Volume 12 , January 2010, e22
Copyright
Copyright © Cambridge University Press 2010

Of the five breast cancer subtypes (luminal A, luminal B, normal, HER2 and triple-negative), triple-negative breast cancers (TNBCs) are the most difficult to treat overall. The response of TNBCs to chemotherapy is highly variable, with some tumours responding while others have poor outcome. Although some TNBCs initially respond, this subtype has the highest rates of relapse (Refs Reference Liedtke1, Reference Sorlie2), particularly in the first three years following therapy (Ref. Reference Dent3). TNBCs lack expression of the oestrogen (estrogen) receptor (ER; encoded by ESR1/2), progesterone receptor (PR; encoded by PGR) and human epidermal growth factor receptor 2 (HER2; encoded by ERBB2). The fact that TNBCs are not amenable to specific treatments directed against the ER, PR and HER2 (i.e. the anti-oestrogen tamoxifen, aromatase inhibitors, or trastuzumab and lapatinib, respectively) that provide clinical benefit for the other subtypes of breast cancer may partly explain why a complete therapeutic response is not typically achieved. Given the remarkably high rates of recurrence, there is an immediate need to identify new treatment strategies for TNBC.

The high rates of relapse in TNBC are thought to be due to the tumour-initiating cells (TICs), which have been shown to be resistant to traditional therapies. Phillips et al. demonstrated resistance of TICs to radiation in cell lines (Ref. Reference Phillips, McBride and Pajonk4), and several studies have demonstrated that this cell population does not respond to chemotherapy (Refs Reference Li5, Reference Wicha, Liu and Dontu6). Breast cancer TICs have the expression signature CD44+/CD24−/EpCAM+ and/or possess aldehyde dehydrogenase 1 enzyme activity (ALDH1+). They were originally believed to make up <10% of the total tumour mass (Refs Reference Al-Hajj7, Reference Ponti8). More recently, the proportion of TICs in a given tumour is reported to be quite variable (Ref. Reference Shackleton9). Breast cancer TICs possess the ability to self-renew, grow as mammospheres in culture and initiate tumour formation in mice (Refs Reference Al-Hajj7, Reference Ponti8). TICs are analogous to the mythical phoenix in that, while traditional therapies eliminate the majority of tumour cells, these nests of TICs are immortal to the challenges of therapy (Fig. 1). Like the phoenix, TICs are long-lived and give rise to progeny that are also resistant to cell death.

Figure 1.

Schematic of the analogy between tumour-initiating cells (TICs) and the phoenix. Tumours are heterogeneous, with TICs making up only a small proportion of the total. TICs are analogous to the phoenix in that they are both immortal and long-lived. The phoenix has the ability to self-renew and to give birth to immortalised progeny just as TICs can result in ‘rebirth’ of the tumour. (a) Following chemotherapy, the majority of the cancer cells undergo cell death; however, the TICs remain dormant and poised for cell renewal. These nests of phoenix-like cells then give rise to secondary tumours. (b) Several studies have reported that chemotherapy actually induces TICs (Refs Reference Li5, Reference Chuthapisith10, Reference Creighton11, Reference Dylla12, Reference Fillmore and Kuperwasser13, Reference Marangoni14, Reference To15). To illustrate this, TICs are represented as a minor subpopulation of the entire tumour, but following chemotherapy these resistant cells expand. The emergence of TICs positions the tumour for drug resistance and eventual cancer recurrence.

Confounding our ability to successfully treat breast cancer is the fact that chemotherapy can actually induce the TIC signature through upregulation of CD44 (Refs Reference Li5, Reference Chuthapisith10, Reference Creighton11, Reference Dylla12, Reference Fillmore and Kuperwasser13, Reference Marangoni14, Reference To15), presenting additional challenges. Conversely, CD44 inhibition resensitises cancer cells to chemotherapy and prevents tumour recurrence in preclinical models. This was initially published by John Dick's group in a model of acute myeloid leukaemia (Ref. Reference Jin16). Subsequently, loss of CD44 reportedly inhibited tumour initiation in models of human breast cancer (Ref. Reference Godar17), including one representing basal-like breast cancer specifically (Ref. Reference Marangoni14).

The possibility of targeting TICs as an anticancer strategy in breast cancer has been explored using oncolytic adenoviruses to kill CD44+/CD24− cells. Such an approach was quite promising in that the virus suppressed the growth of CD44+/CD24− primary cells isolated from patients with pleural effusions. Furthermore, exposure of the cells to adenovirus and injection into mice showed that inhibiting TICs reduced tumour reinitiation in mice (Ref. Reference Eriksson18). While the treatment strategies for TNBC have previously been discussed (Refs Reference Stockmans19, Reference Tan and Swain20), the identification of agents that eliminate the TIC population has not. We propose that TIC-targeting agents will improve the overall success of TNBC cancer treatment. If they are given prior to or in conjunction with chemotherapy, these agents may prevent eventual relapse.

Techniques for the isolation and identification of TICs

There are a number of techniques used to identify, isolate and functionally characterise TICs [Table 1, and more extensively reviewed elsewhere (Refs Reference Charafe-Jauffret, Ginestier and Birnbaum21, Reference Nakshatri, Srour and Badve22)]. In brief, the CD44+/CD24− cells are identifiable by flow cytometry, immunohistochemistry, immunofluorescence and/or immunoblotting. There are newer methodologies that rely on magnetic beads to capture CD44+ cells (EasySep, StemCell Technologies, Vancouver, BC, Canada) (Ref. Reference Shipitsin23). Others have relied more extensively on the enzymatic activity of ALDH1 and as such have used the ALDEFLUOR assay for detection (Refs Reference Charafe-Jauffret, Ginestier and Birnbaum21, Reference Ginestier24, Reference Korkaya25). The ALDH isoform ALDH1A1 is thought to be the most common form and it can be readily detected in situ by immunohistochemistry (Refs Reference Ginestier24, Reference Nalwoga26). It is believed that the levels of ALDH1A1 are reflective of its activity within cells; however, this has not been fully demonstrated yet.

Table 1.

Techniques used to characterise breast cancer stem cells

Abbreviations: IHC, immunohistochemistry; NOD/SCID, nonobese diabetic and severe combined immunodeficient; TIC, tumour-initiating cell.

For functional characterisation of breast cancer TICs, self-renewal assays are performed using colony-formation assays (Ref. Reference Fillmore and Kuperwasser13) where cells with the greatest capacity to behave like ‘stem cells’ are those that are able to be serially passaged. In addition, one of the hallmarks of TICs is that they not only have the ability to make more of themselves but can also differentiate into daughter cells. This characteristic can be assessed by measuring single cells for their ability to form colonies that are both CD44+/CD24− and CD44−/CD24+. More commonly, the mammosphere assay is used because it is based on the premise that only cells with stem-like properties can grow in suspension (Refs Reference Ponti8, Reference Charafe-Jauffret, Ginestier and Birnbaum21). It has also been shown that serially passaging mammospheres further enriches TICs (Ref. Reference Fillmore and Kuperwasser13); thus it is an accepted measurement of self-renewal (Ref. Reference Dontu and Wicha27). A limitation of this assay is that some spheres might not be clonal but a result of cells clumping together. However, this can be mostly overcome by seeding the cells at a low density; individual cells can be plated in 96- or 384-well plates and mammosphere formation monitored. Finally, the ability of TICs to form tumours from only a few cells is also used to assess their functionality, as compared with a standard protocol, where 106 cells would typically be injected to initiate tumour formation. Typically, 100–500 cells are injected into the mammary fat pad of NOD/SCID (nonobese diabetic and severe combined immunodeficient) mice and tumour initiation is monitored over time. Tumours that develop can be removed, enzymatically digested and reinjected for serial transplantation studies, which further define the ability of TICs to regenerate cancer formation (Ref. Reference Charafe-Jauffret, Ginestier and Birnbaum21). Each of these techniques is useful for preclinical investigations of putative molecular targets that may alter TIC survival, growth and expansion.

TICs and normal stem cells share common epitopes providing clues to the origins of TNBC

At the present time, the stem cell theory suggests that during the course of cancer progression normal stem cells become altered – for example, by activation of oncogenes or loss of tumour suppressor genes – which ultimately gives rise to the development of cancer. In breast cancer, an example might be the upregulation of an oncogene such as YBX1 (encoding the transcription factor YB-1), which is normally expressed in luminal and myoepithelial primary progenitor cells from the mammary gland (Ref. Reference Finkbeiner28). Another possibility is the loss of the tumour suppressor gene BRCA1, which has been shown to lead to the development of TNBC in mouse models (Ref. Reference Lim29).

It is probably no coincidence then that normal stem cells and cancer stem cells share common surface markers such as CD44 and CD49f (Fig. 2), and also possess ALDH activity. [CD49f is a common denominator that defines human and mouse mammary stem cells (Refs Reference Lim29, Reference Asselin-Labat30, Reference Eirew31, Reference Stingl32, Reference Visvader33). However, it is also classically known as a protein associated with the myoepithelial layer of the mature normal mammary gland along with smooth muscle actin (Fig. 3). Whether or not it plays a role in human breast cancer initiation and the development of TNBC is not clear.] It is also interesting to note that primary mammary stem cells are classically triple negative, as described by Visvader (Ref. Reference Visvader33). Whether these cells directly give rise to TNBC is not yet known. Our current understanding of the hierarchy of mammary development and tumourigenesis is that stem cells give rise to a common progenitor that undergoes differentiation into luminal and myoepithelial progenitors. It is thought that TNBC and more specifically the basal subtype arise from the luminal progenitors (Ref. Reference Lim29), yet the intermediate steps in this process are not yet defined.

Figure 2.

Mammary gland stem cell origins and progression to breast cancer subtypes. Mammary stem cells have been defined by a number of surface markers that have been recently reviewed by several groups (Refs Reference Visvader33, Reference Ginestier, Liu and Wicha122, Reference Kai130). These cells give rise to a common bipotential progenitor that can produce myoepithelial or luminal cell progenitors. At the present time, the evidence is that luminal progenitors give rise to triple-negative breast cancer (TNBC), and through subsequent changes that have not been thoroughly investigated luminal A and luminal B breast cancer subtypes develop as cells differentiate toward the ductal phenotype. The derivation of the ‘normal’ subtype (Ref. Reference Sorlie2) is unclear.

Figure 3.

Organisation of the mammary gland. The mammary gland is depicted schematically (a) and visually from primary mouse mammary tissue via haematoxylin/eosin staining (b). The myoepithelial layer resides below the luminal cell layer and is readily identifiable by staining for either smooth muscle actin (c) or CD49f, similar to findings by Visvader et al. (Ref. Reference Visvader33) (d).

Role of TICs in mediating breast cancer cell survival through their ability to increase metastases

It is commonly known that patients who develop distant metastases have a far greater chance of dying from their breast cancer than those who do not (Ref. Reference Wallgren, Silfverswärd and Eklund34). The combination of stem cell markers ALDHhi/CD44+/CD24− provided the highest odds for breast cancer metastases when cells were injected into the tail vein of NOD/SCID mice (Ref. Reference Croker35). There are many invasion proteases linked to TICs that contribute to the dissemination of breast cancer cells, including urokinase-type plasminogen activators (uPAs) and matrix metalloproteinases, as well as the chemokine CXCL4 (Ref. Reference Sleeman and Cremers131). The expression of CD44 on TICs may facilitate the invasion of breast cancer cells by promoting tumour cell invasion via the induction of uPA (Ref. Reference Sheridan36). Breast cancer cells with the expression signature CD44+/CD24− are associated with increased invasion in vitro (Ref. Reference Sheridan36) and distant metastases in patients (Ref. Reference Abraham37). There is also an overlap in the genes that constitute an invasive gene signature (Ref. Reference Liu38) and those expressed in cells that have been induced to become ‘stem-like’ by silencing E-cadherin (Ref. Reference Gupta39).

As breast cancer metastases typically occur via the lymphatics, it is important to understand whether TICs play a role in this process. In a TNBC model, MDA-MB-468LN cells, a metastatic variant cell line, had a much greater propensity to undergo lymph-node metastases compared with their syngeneic counterpart MDA-MB-468 cells, and this was associated with a very large increase in the CD44+/CD24− population (Ref. Reference Pandit40). These cells also expressed a number of other genes associated with TICs, including ALDH1, and showed a loss of E-cadherin expression (Ref. Reference Pandit40). Thus, one of the challenges to treating TNBC is that the cells metastasise and may remain dormant for a period of time before they trigger tumour regrowth.

The following sections discuss potential targets whose inhibition might allow for modulation of TICs and thereby provide enhanced treatment options.

Targeting receptor tyrosine kinases

EGFR

The epidermal growth factor receptor (EGFR) predominantly signals through the MAP kinase/ERK (mitogen-activated protein kinase/extracellular-signal-regulated kinase) pathway and directly activates p90 ribosomal S6 kinase (RSK) as one of its many downstream targets. To put EGFR in the context of TNBC, EGFR is expressed in >50% of such tumour samples as well as being highly expressed in TNBC cell lines (Ref. Reference Stratford41). Cheang et al. established that the addition of this marker with cytokeratin 5/6 to the established triple-negative markers (lack of expression of ER, PR and HER2) more accurately predicts poor survival (Ref. Reference Cheang42). Furthermore, in an attempt to establish xenografts from a relatively large number of tumours (n = 200 samples), only 12.5% were transplantable into mice, the majority of which were TNBCs (Ref. Reference Marangoni43). Since many TNBCs express EGFR, it would therefore seem likely that this receptor may be important for tumour initiation/reinitiation. In support of this idea, the explanted xenografts were later characterised for a number of proteins known to promote the growth of tumours, and the most consistent finding was that EGFR was activated in 78% of the cases (14/18) (Ref. Reference Marangoni14). Thus, EGFR is commonly activated in TNBC and might be a target for therapy as many small molecules have been developed against this receptor kinase.

Preclinical and clinical studies have examined EGFR inhibition in TNBC. For example, inhibition of EGFR with gefitinib (Iressa) suppressed the growth of the TNBC cell line SUM149 and also prevented the cells from forming colonies in soft agar (Ref. Reference Stratford41). Likewise, inhibition of EGFR attenuated the growth of mammospheres when cells isolated from primary ductal carcinoma in situ were cultured in the presence of gefitinib, yet this was only observed in lesions that were HER2-positive (Ref. Reference Farnie44). The dual EGFR–HER2 inhibitor lapatinib has recently been shown to be beneficial in the treatment of HER2-positive breast cancers, reducing the number of TICs in patients (Ref. Reference Li5). Cells taken from biopsies of these patients following treatment show reduced ability to form mammospheres (Ref. Reference Li5). A more recent study again reported the clinical benefit of treating HER2-amplified cancers with lapatinib in combination with paclitaxel; however, no such response was observed in TNBC (Ref. Reference Di Leo45). There was also no association between EGFR expression in all breast cancers and lapatinib response (Ref. Reference Finn46). Several clinical trials have been conducted with gefitinib in breast cancer, which would have included TNBC; however, the success has been limited (Refs Reference Green47, Reference Polychronis48). No doubt these studies are often confounded by study-design issues such as clearly defining whether EGFR is in fact activated in the tumours prior to patient enrolment. The specifics around the TIC subpopulation would also not have been part of the study as it is such a new and evolving field. As such, carefully controlled clinical trials, which address the inclusion of TICs in a prospective manner, are needed to conclude whether inhibiting EGFR will serve as a good avenue for eliminating TICs.

One of the benefits of considering a target such as EGFR is that it has been the subject of many preclinical and clinical studies. The promise of EGFR as a molecular target for TNBC has however been complicated by drug resistance afforded by many mechanisms that are beyond the scope of this review. However, to highlight just a few examples, a study in 2008 demonstrated that the MAP kinase pathway may become constitutively activated, leading to gefitinib resistance (Ref. Reference Normanno49). The loss of the tumour suppressor PTEN or activation of Ras could also explain drug resistance. In the TNBC models specifically, the activation of components downstream of EGFR can also explain the lack of responsiveness of cell lines such as MDA-MB-468 [PTEN-null leading to hyperactive phosphoinositide 3-kinase (PI3K) activity] (Ref. Reference She50) and MDA-MB-231 cells (which contain an activating mutation of Ras). Therefore, combining EGFR inhibition with therapies designed to target downstream components would provide additional benefit. The combination of gefitinib with small interfering RNAs (siRNAs) against the oncogenic transcription factor YB-1 reduced the ability of HCC1937 cells to form colonies in soft agar (Ref. Reference Stratford41). Similarly, combining the PI3K inhibitor LY294002 with gefitinib in MDA-MB-468 cells inhibits growth (Ref. Reference She50).

Continuing with this combination approach, several studies have looked at the effectiveness of combining EGFR inhibitors with classical chemotherapy. Corkery et al. demonstrated the synergistic effect of treating TNBC cell lines with gefitinib, carboplatin and docetaxel in vitro (Ref. Reference Corkery51). Likewise, the TNBC cell lines SUM102 and SUM149 responded synergistically to cetuximab and carboplatin (Ref. Reference Hoadley52). Similarly, the growth of gefitinib-resistant MDA-MB-468 cells could be inhibited using a combination of cetuximab and the DNA-crosslinking agent cisplatin (Ref. Reference Oliveras-Ferraros53). However, more interestingly, this work identified the importance of understanding how the drugs work to enhance each other's effectiveness and thereby provide optimal treatment regimens and patient selection (Ref. Reference Oliveras-Ferraros53). There are a number of clinical trials under way assessing combinations of platinum-based drugs with either EGFR antibodies (cetuximab) or EGFR kinase inhibitors (erlotinib) in TNBC (http://clinicaltrials.gov). The results of current clinical trials will provide a much clearer picture of the extent to which targeting EGFR in combination with chemotherapy will benefit patients. We may also take some of the lessons learned about resistance to EGFR into the study of TICs to better understand the signalling pathways that prevail in this interesting subpopulation.

IGF1R

The insulin-like growth factor 1 receptor (IGF1R) is commonly overexpressed in breast cancer (Ref. Reference Nielsen54). In the TNBC subtype, it has been reported that this receptor is activated in ~40% of cases (Ref. Reference Law55). Further, BMS-536924, a small-molecule inhibitor of IGF1R, suppresses the growth of TNBC cells in monolayers and in soft agar (Ref. Reference Law55). Subsequently, the IGF1R inhibitor picropodophyllin was used to suppress the growth of MDA-MB-231 cells in mice (Ref. Reference Klinakis56). This model was selected because it is a TNBC cell line that harbours a mutation in KRAS (G13D). Complementary to that finding, expression of KRAS with the same mutation was found to lead to the development of TNBC in mice and these tumours expressed high levels of IGF1R (Ref. Reference Klinakis56). However, it is unclear whether the expression of IGF1R itself leads to the development of tumours with a TNBC signature. This could be determined by examining the tumours that arose from mice that express activated IGF1R in the mammary gland (Ref. Reference Carboni57).

Further, it is of particular interest to explore the possibility that IGF1R may regulate TICs as it has been previously associated with the induction of the transcription factor Snail and the loss of E-cadherin (Ref. Reference Kim58), both of which have ties to TICs (Ref. Reference Moody59). From a therapeutic perspective, it is noteworthy that inhibiting IGF1R with BMS-536924 restores E-cadherin expression and reverses epithelial to mesenchymal transition (EMT) in vitro (Ref. Reference Kim58). In general, EMT is a process whereby cancer cells acquire the ability to invade and metastasise. Although there is evidence accumulating in favour of targeting IGF1R for the treatment of TNBC, there may yet be additional obstacles to contend with such as cells that have undergone EMT in vivo (Ref. Reference Jones60). Based on recent evidence, it was concluded that inhibiting IGF1R caused tumour regression in as many as 93% of the mice, yet cancer recurrence was evident in a proportion of the mice (16%) (Ref. Reference Jones60). The recurrent tumours had undergone EMT based on loss of E-cadherin and the associated induction of Twist and Snail (Ref. Reference Jones60). These seemingly conflicting studies warrant further investigation to definitively understand whether inhibiting IGF1R can be used to kill cells that have undergone EMT and/or possess features of TICs.

MET

MET, the receptor for hepatocyte growth factor (HGF), is a stem cell marker also involved in normal embryonal development. However, it is overexpressed in 20–30% of breast cancers and is associated with a poor outcome (Refs Reference Camp, Rimm and Rimm61, Reference Ghoussoub62, Reference Welm63). More recently, MET was identified, along with CD44, as a transcriptional target of YB-1 (Ref. Reference Finkbeiner28). Critically, MET requires CD44v6 to become fully activated by HGF (Ref. Reference Orian-Rousseau64). This suggests that there is receptor complementation between CD44 variants and MET. Furthermore, the MET receptor is expressed preferentially in TNBC cell lines as compared with cell lines derived from other types of breast cancer (Ref. Reference Finkbeiner28). Two recent studies demonstrated that overexpression of oncogenic MET in mice resulted in breast tumours of varying phenotype (Refs Reference Graveel65, Reference Ponzo66). MET was also frequently amplified in these tumours and the expression was inversely correlated with the expression of ER/PR and HER2 in individual cancer cells (Ref. Reference Graveel65). In two independent data sets of human breast cancer patients, MET was associated with the basal-like subtype and correlated with poor prognosis (Refs Reference Graveel65, Reference Ponzo66). These studies implicate MET as a therapeutic target in TNBC. There are numerous small-molecule inhibitors to this receptor at different stages of development; some, such as XL880 (Exelixis) and ARQ197 (ArQule), are in the Phase I/II clinical trial (Ref. Reference Comoglio, Giordano and Trusolino67) albeit not currently in breast cancer.

KIT

KIT is a cytokine cell-surface receptor that binds stem cell factor (SCF). The expression of KIT and its ligand has been reported in breast cancer – in cell lines as well as tumour specimens (Ref. Reference Hines68). A more recent study demonstrated co-expression of KIT and EGFR in more-undifferentiated breast carcinomas, which have bipotentiality (Ref. Reference Tsuda69). Interestingly, Lim et al. have identified KIT as upregulated in a luminal progenitor signature, which seemed to be more highly associated with basal-like breast cancers (Ref. Reference Lim29). While no KIT-targeting agents have been assessed in breast cancer patients, in lung cancer it has been shown that following chemotherapy the use of imatinib or SCF antibodies will kill the remaining cancer stem cells (Ref. Reference Levina70). In this in vitro study, imatinib alone had no effect on proliferation but when used in combination with cisplatin there was both a decrease in proliferation and a reduction in sphere formation, which was used as a measure of ‘stemness’ (Ref. Reference Levina70). Since cisplatin is currently used for the treatment of TNBC, it is worth considering combining imatinib in the TNBC setting. The expanded luminal progenitor population identified within BRCA1-mutation carriers was associated with elevated expression of KIT (Ref. Reference Lim29). This suggests that it may be an appropriate target in TNBC with mutated BRCA1. The association of KIT in BRCA1-mutation carriers (Ref. Reference Lim29), and sporadic basal-like breast cancer (Ref. Reference Nielsen71), is provocative given the successful targeting of KIT with imatinib in gastrointestinal stromal tumours (reviewed in Ref. Reference Renouf, Wilson and Blanke72). It may actually be important to consider the mutational status of KIT because it predicts response to imatinib (Ref. Reference Heinrich73). Thus, there is an emerging body of evidence that KIT may be an excellent therapeutic target for TNBC.

Nonreceptor kinases as molecular targets for TNBC

Receptor tyrosine kinases elicit cell survival signalling; therefore, targeting the downstream molecules may be an alternative approach to inhibiting TICs. There has been a supportive role for several kinases in sustaining TICs. For example, AKT is found in breast TICs based on side-population sorting (Ref. Reference Zhou74). MAPK has also been implicated in TNBC; for example, it was associated with an increased risk of relapse and poor survival in TNBC in a cohort of 109 ER/PR-negative patient samples. This suggests that components of the MAPK pathway may therefore provide potential therapeutic targets (Ref. Reference Eralp75). Additionally, within the MAPK pathway resides RSK, which is of interest because it is activated by EGFR and fibroblast growth factor receptor (FGFR), both of which sustain the growth of TICs. IGF1R signalling also promotes the activation of RSK (Ref. Reference Law55). The RSKs are a family of AGC serine/threonine kinases (RSK 1–4) that phosphorylate a number of transcription factors, including CREB, FOS (Ref. Reference Chen, Abate and Blenis76), NF-κB (Ref. Reference Schouten77) and ER (Ref. Reference Joel78). YB-1 is a particularly notable RSK substrate, given that it is expressed highly in tumours but not in normal breast cells (Ref. Reference Stratford79).

Blocking RSK is also attractive as a cancer target because it would prevent the inactivation of a number of prodeath factors, such as BAD (Ref. Reference Shimamura80) and death-associated protein kinase (DAPK) (Ref. Reference Anjum81). Additionally RSK inhibits the cell cycle inhibitor protein p27kip (CDKN1B) (Ref. Reference Larrea82). Of the four isoforms, RSK1 and RSK2 are the best characterised, and overexpression has been associated with multiple cancer types such as prostate (Refs Reference Clark83, Reference Cuadrado and Nebreda84) and those of haematopoietic origin (Ref. Reference Cuadrado and Nebreda84). RSK1 and RSK2 kinase activities were also reportly elevated in a small cohort of primary breast tumours (n = 48 tumours) with expression levels elevated in ~50% of tumours compared with controls (48 cancers and 12 controls) (Ref. Reference Smith85). Thus far, the status of RSK1 and RSK2 in the TNBC subgroup is not known although RSK is highly expressed in TNBC cell lines (Ref. Reference Stratford79). The role of RSK in cell transformation has been reported in several settings. For example, a recent study demonstrated the importance of RSK activity in the FGFR1-induced transformation of the normal breast epithelial cell line MCF10A (Ref. Reference Xian86). Inhibition of RSK2 has also been shown to suppress the growth advantage provided by constitutively activated Ras (Ref. Reference Cho87), the latter of which has been shown to lead to TNBC tumours in mice (Ref. Reference Klinakis56). There are a number of agents that inhibit RSK, including BI-D1870 (Ref. Reference Sapkota88), SL0101 (Ref. Reference Smith85) and FMK (Ref. Reference Cohen, Hadjivassiliou and Taunton89); however, preclinical studies have not yet been conducted in models of TNBC. While it seems likely that downstream kinases are tractable targets for TICs, little work has been done in this regard to date.

Transcription factors in sustaining TICs

EZH2

EZH2 is a member of the polycomb group of proteins that are involved in transcriptional repression through promoter methylation. In stem cells, polycomb proteins aid in the maintenance of pluripotency by inhibiting genes that drive differentiation. Overexpression of EZH2 has been demonstrated in tumours of the prostate and breast (Ref. Reference Kleer90) as well as glioblastoma multiforme (Ref. Reference Suvà91). EZH2 is a marker of aggressive, invasive breast cancer (Ref. Reference Kleer90) and is important for ER-negative breast cancer both in vitro and in vivo (Ref. Reference Gonzalez92). Furthermore, inhibition of EZH2 decreased the growth of TNBC cells (Ref. Reference Gonzalez92). A recent study has shown that EZH2 is important for the maintenance of the stem cell population in glioblastoma multiforme. Importantly, EZH2 inhibition perturbed self-renewal and tumour formation (Ref. Reference Suvà91). Taken together, these data identify EZH2 as a potential therapeutic target in TNBC.

YB-1

YB-1 is a bona fide oncogene product that leads to the development of mammary tumours in mice (Ref. Reference Bergmann93), while in breast cancer cell lines it promotes growth and resistance to chemotherapeutic agents (Refs Reference Bargou94, Reference Kuwano95). YB-1 becomes activated by phosphorylation at S102, located in its DNA-binding domain, by the serine/threonine kinases AKT (Ref. Reference Sutherland96) and to a greater extent by RSK (Ref. Reference Stratford79). In turn, P-YB-1S102 induces EGFR (Refs Reference Sakura97, Reference Wu98) and HER2 (Ref. Reference Sakura97), as well as MET (Ref. Reference Finkbeiner28) and the second-messenger PIK3CA (Ref. Reference Astanehe99), leading to enhanced growth. These results indicate that YB-1 may be a signature feature of aggressive forms of breast cancer in women and that tumours may be dependent upon it for survival. Over the past three years, it has been demonstrated that YB-1 is (1) consistently associated with relapse and poor overall survival in all breast cancer subtypes in a cohort of >4049 cases, (2) expressed in 60–70% of the most aggressive subtypes (basal-like and HER2), and (3) a stronger prognostic marker for breast cancer than those currently used in the clinic, including ER and HER2 (Ref. Reference Habibi100). Complementary to these studies, inhibition of YB-1 with siRNAs (1) promotes apoptosis by lifting the constraints of the mTOR–STAT3–MCL1 prosurvival pathway, (2) suppresses the proliferation of TNBC cell lines by 40–80%, and (3) perturbs tumour growth in mice (Ref. Reference Lee101).

Based upon chromatin immunoprecipitation (ChIP) on-chip profiling, YB-1 potentially regulates several cancer stem-cell targets, including CD44, CD49f, MET, KIT, WNT and NOTCH family members (Ref. Reference Finkbeiner28) (Fig. 4). CD44, CD49f and MET were subsequently validated as YB-1 target genes that required P-YB-1S102 (Refs Reference To15, Reference Finkbeiner28). Importantly, activation of YB-1 promotes self-renewal, mammosphere growth and drug resistance (Ref. Reference To15). The studies by To et al. (Ref. Reference To15) ascertain that YB-1 plays a key role in the induction of the TIC signature following chemotherapy exposure; therefore drugs typically used in the clinic to treat TNBC such as adriamycin, cyclophosphamide and paclitaxel induce CD44. Further, paclitaxel induces P-YB-1S102 and promotes YB-1 binding to the CD44 promoter, leading to the induction of CD44 protein (Ref. Reference To15). However, it has been demonstrated that YB-1 imposes ‘stem-like properties’ by inducing CD44, suggesting its value as a potential therapeutic target. Translating this finding into the clinic is a challenge because at present there are no small-molecule inhibitors known to directly inhibit it. However, recent developments in the delivery of siRNAs may pave the way for the use of siRNAs as targeted therapeutic agents (Refs Reference Judge102, Reference Semple103).

Figure 4.

Chromatin immunoprecipitation on-chip can be used to identify novel pathways and potential targets for treating TNBC. (a) Many transcription factors are reportedly induced in TNBC including YB-1. In order to understand how YB-1 promotes TNBC cell growth, invasion and possibly EMT conversion, one can profile targets on a genome-wide basis using chromatin immunoprecipitation on-chip. Briefly, crosslinked and sonicated chromatin is immunoprecipitated with anti-YB-1 antibodies, allowing identification of promoter regions targeted by YB-1 following analysis of the associated DNA. (b) Example of a genome-wide study of YB-1 target genes in the TNBC cell line SUM149 (Ref. Reference Finkbeiner28). Many of the target genes have prior links to tumour initiation, including (1) the cell-surface or receptor molecules CD44, CD49f, MET and several members of the NOTCH family, (2) extracellular signalling molecules of the WNT family, and (3) cell cycle regulators. Solid arrows indicate direct interactions; broken arrows represent indirect relationships; curved arrows indicate ability to self-regulate. Abbreviations: EMT, epithelial to mesenchymal transition; TNBC, triple-negative breast cancer.

BRCA1 and BRCA2

Germline mutations in the BRCA1 and BRCA2 transcription factor genes confer a high risk of breast and ovarian cancer (Ref. Reference Narod and Foulkes104). While BRCA1 and BRCA2 function coordinately during the DNA-damage response, genetic and phenotypic differences distinguish BRCA1-mutated from BRCA2-mutated tumours (Ref. Reference Stefansson105). Tumours with mutated BRCA1 (‘BRCA1 tumours’) are usually high-grade, infiltrating tumours that are TNBCs; they also tend to express cytokeratin intermediate filaments indicative of a basal-like phenotype (Ref. Reference Foulkes106). In striking contrast, ‘BRCA2 tumours’ tend to be invasive ductal, high-grade tumours that, when adjusted for grade, are more often ER-positive and exhibit a luminal phenotype (Ref. Reference Bane107). Recent expression and copy number profiling reveals that about 75% of BRCA1 tumours are basal-like while the same proportion of BRCA2 tumours are luminal A or B (Ref. Reference Waddell108). Even in the absence of BRCA1 mutation, basal-like tumours express twofold lower levels of BRCA1 mRNA in comparison with matched controls (Ref. Reference Turner and Reis-Filho109), consistent with similarities in DNA copy number data between basal-like BRCA1 and basal-like non-BRCA1 tumours (Ref. Reference Waddell108). In mice, the targeted disruption of Brca1 in the mammary gland also leads to the development of tumours that are basal-like and lack ER, PR and HER2 (Ref. Reference McCarthy110). These unique features of BRCA1 tumours suggest that distinct biological pathways are perturbed during BRCA1 and BRCA2 tumourigenesis (Refs Reference Stefansson105, Reference Foulkes106). Therefore, BRCA1 basal-like TNBCs may present therapeutic targets dependent upon the loss of DNA-damage response or other tumourigenic pathways unique to BRCA1 (luminal differentiation; see below).

BRCA1/BRCA2-deficient tumour cells demonstrate genetic and genomic instability due, in part, to impaired DNA-damage repair (Ref. Reference Deng and Wang111). The coordinate function of BRCA1 and BRCA2 in the DNA-damage response led to the landmark discovery of BRCA1 or BRCA2 tumour dependency upon the activity of poly(ADP-ribose) polymerase family member 1 (PARP1) (Refs Reference Bryant112, Reference Farmer113). Against a background of genetic instability, depletion of PARP1 causes synthetic lethality in BRCA1/BRCA2-deficient cells due to a severe defect in homologous recombination (Refs Reference Bryant112, Reference Farmer113). Selective cytotoxicity of the PARP inhibitor AZD2281, alone or in combination with other therapies, has recently been confirmed for BRCA2-deficient mammary tumour cell growth (Ref. Reference Evers114) and BRCA1-deficient triple-negative murine mammary tumours (Ref. Reference Rottenberg115). These exciting findings are the basis for rational design of targeted therapies against BRCA1/BRCA2 tumours. In a recent retrospective analysis, BRCA1-associated breast cancer showed particular sensitivity to cisplatin (Ref. Reference Byrski116), suggesting that chemotherapy regimens with DNA-damaging agents may show the best potential for benefit to these patients (Ref. Reference Carey117). Along a similar path, a high-throughput preclinical screen identified three alkylators with strong and specific toxicity against BRCA2 tumours (Ref. Reference Evers118). Finally, a clinical evaluation of a novel PARP inhibitor (termed olaparib) has recently been reported. In this trial, olaparib had few adverse effects and demonstrated clinical benefit in 12 of 19 (63%) BRCA-mutation carriers, with no antitumour responses in noncarriers (Ref. Reference Fong119). Therefore, PARP inhibitors, and other DNA-damaging agents, may represent promising drugs against BRCA1-associated TNBC in humans.

Foulkes hypothesised that BRCA1 may function to determine lineage choice of breast stem/progenitor cells (Ref. Reference Foulkes106). Recent investigations of mammary differentiation within in vitro and in vivo models support this hypothesis. Knockdown of BRCA1 expression in the nontumourigenic MCF10A cell line impairs differentiation during in vitro culture within reconstituted basement membrane (Ref. Reference Furuta120). Similarly, knockdown of BRCA1 in primary human mammary cells, which were then injected to reconstitute cleared mouse mammary fat pads in vivo, impaired epithelial differentiation, reduced ER expression and resulted in ductal structures composed of a single cell layer that lacked expression of the luminal epithelial marker cytokeratin 18 (Ref. Reference Liu121). Finally, carriers of BRCA1 mutations harbour an expanded luminal progenitor population (Ref. Reference Lim29), consistent with impaired terminal differentiation of luminal epithelia. In humans, the expanded luminal progenitor cells are a probable target population in BRCA1-associated and possibly other basal breast tumours (Ref. Reference Lim29). BRCA1 function alters the differentiation of mammary stem (Ref. Reference Liu121) or progenitor (Ref. Reference Lim29) populations, recently reviewed elsewhere (Ref. Reference Ginestier, Liu and Wicha122). Interestingly, it has been suggested that inhibition of aurora kinase may be a means to inhibit the differentiation block imposed by BRCA1 mutations. Aurora kinase A is a ‘drug-able’ target (Ref. Reference Gautschi123) that has recently been associated with BRCA1 differentiation (Ref. Reference Chang124). In mouse embryonic stem cells, expression of a BRCA1 variant lacking the AURKA phosphorylation site (S308A) resulted in differentiation defects (Ref. Reference Chang124), suggestive of a regulatory role for AURKA on BRCA1 function during differentiation, as shown in G2–M progression and microtubule nucleation (Refs Reference Ouchi125, Reference Sankaran126). Regardless of the ‘cell of origin’, tumours arising in carriers of BRCA1 mutations are strongly predisposed to TNBC features and may present therapeutic opportunities that target differentiation.

Concluding remarks

As we close in on targeting TICs to improve the treatment of TNBC, one should not lose sight of the broader implications of these efforts. While CD44+/CD24− are biomarkers for TICs in TNBC, they are also markers in all of the other breast cancer subtypes (Refs Reference Hennessy127, Reference Honeth128). TICs are also inducible following hormone therapy in ER-positive breast cancers (Refs Reference Li5, Reference Creighton11). Likewise, ALDH enzyme activity is not restricted to TNBC. ALDH1 expression is independent of ER or PR (Ref. Reference Tanei129) and may be considered independent of a particular breast cancer subtype. Thus, while this review has specifically focused on eliminating TICs to improve TNBC treatment, clearly other breast cancer subtypes express these resistant subpopulations. Therefore, targeted agents that eliminate TICs would be broadly applicable to improving breast cancer.

In closing, TNBC treatment is complicated by the high preponderance of TICs, and this is believed to be a major cause for the high rates of recurrence so commonly associated with this disease. Until we are able to target TICs, we will not likely achieve a cure for TNBC. At this time there are several exciting molecular targets for extinguishing the phoenix, including IGF1R, EGFR, KIT, RSK, YB-1 and modifiers of the DNA-damage response, such as PARP inhibitors. We are certainly at the dawn of a new era for treating TNBC by focusing on the elimination of TICs that are believed to be at the root of relapse.

Acknowledgements and funding

This work is supported by a grant through the Canadian Institute for Health Research (S.E.D.) and by the Canadian Breast Cancer Foundation (A.L.S.).

References

References

Publications

1Liedtke, C. et al. (2008) Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. Journal of Clinical Oncology 26, 1275-1281CrossRefGoogle ScholarPubMed
2Sorlie, T. et al. (2001) Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proceedings of the National Academy of Sciences of the United States of America 98, 10869-10874CrossRefGoogle ScholarPubMed
3Dent, R. et al. (2007) Triple-negative breast cancer: clinical features and patterns of recurrence. Clinical Cancer Research 13, 4429-4434CrossRefGoogle ScholarPubMed
4Phillips, T.M., McBride, W.H. and Pajonk, F. (2006) The response of CD24 − /low/CD44+ breast cancer-intiating cells to radiation. Journal of the National Cancer Institute 98, 1777-1785CrossRefGoogle ScholarPubMed
5Li, X. et al. (2008) Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. Journal of the National Cancer Institute 100, 672-679CrossRefGoogle ScholarPubMed
6Wicha, M.S., Liu, S. and Dontu, G. (2006) Cancer stem cells: an old idea – a paradigm shift. Cancer Research 66, 1883-1890CrossRefGoogle ScholarPubMed
7Al-Hajj, M. et al. (2003) Prospective identification of tumorigenic breast cancer cells. Proceedings of the National Academy of Sciences of the United States of America 100, 3983-3988CrossRefGoogle ScholarPubMed
8Ponti, D. et al. (2005) Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Research 65, 5506-5511CrossRefGoogle ScholarPubMed
9Shackleton, M. et al. (2009) Heterogeneity in cancer: cancer stem cells versus clonal evolution. Cell 138, 822-829CrossRefGoogle ScholarPubMed
10Chuthapisith, S. et al. (2010) Breast cancer chemoresistance: emerging importance of cancer stem cells. Surgical Oncology 19, 27-32CrossRefGoogle ScholarPubMed
11Creighton, C.J. et al. (2009) Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proceedings of the National Academy of Sciences of the United States of America 106, 13820-13825CrossRefGoogle ScholarPubMed
12Dylla, S.J. et al. (2008) Colorectal cancer stem cells are enriched in xenogeneic tumors following chemotherapy. PLoS one 3, e2428CrossRefGoogle ScholarPubMed
13Fillmore, C.M. and Kuperwasser, C. (2008) Human breast cancer cell lines contain stem-like cells that self-renew, give rise to phenotypically diverse progeny and survive chemotherapy. Breast Cancer Research 10, R25CrossRefGoogle ScholarPubMed
14Marangoni, E. et al. (2009) CD44 targeting reduces tumour growth and prevents post-chemotherapy relapse of human breast cancer xenographs. British Journal of Cancer 100, 918-922CrossRefGoogle Scholar
15To, K. et al. (2010) YB-1 induces expression of CD44 and CD49f leading to enhanced self-renewal, mammosphere growth, and drug resistance. Cancer Research 70, 2840-2851CrossRefGoogle ScholarPubMed
16Jin, L. et al. (2006) Targeting of CD44 eradicates human acute myeloid leukemic stem cells. Nature Medicine 12, 1167-1174CrossRefGoogle ScholarPubMed
17Godar, S. et al. (2008) Growth-inhibitory and tumor-suppressive functions of p53 depend on its repression of CD44 expression. Cell 134, 62-73CrossRefGoogle ScholarPubMed
18Eriksson, M. et al. (2007) Oncolytic adenoviruses kill breast cancer initiating CD44 + CD24 − /low cells. Molecular Therapeutics 15, 2088-2093Google ScholarPubMed
19Stockmans, G. et al. (2008) Triple-negative breast cancer. Current Opinion in Oncology 20, 614-620CrossRefGoogle ScholarPubMed
20Tan, A.R. and Swain, S.M. (2008) Therapeutic strategies for triple-negative breast cancer. Practice of Oncology: Recent Advances 14, 343-351Google ScholarPubMed
21Charafe-Jauffret, E., Ginestier, C. and Birnbaum, D. (2009) Breast cancer stem cells: tools and models to rely on. BMC Cancer 9, 202-211CrossRefGoogle Scholar
22Nakshatri, H., Srour, E.F. and Badve, S. (2009) Breast cancer stem cells and intrinsic subtypes: controversies rage on. Current Stem Cell Research and Therapies 4, 50-60CrossRefGoogle Scholar
23Shipitsin, M. et al. (2007) Molecular definition of breast tumor heterogeneity. Cancer Cell 11, 259-273CrossRefGoogle ScholarPubMed
24Ginestier, C. et al. (2007) ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 1, 555-567CrossRefGoogle Scholar
25Korkaya, H. et al. (2008) HER2 regulates the mammary stem/progenitor cell population driving tumorigenesis and invasion. Oncogene 27, 6120-6130CrossRefGoogle ScholarPubMed
26Nalwoga, H. et al. (2010) Expression of aldehyde dehydrogenase 1 (ALDH1) is associated with basal-like markers and features of aggressive tumours in African breast cancer. British Journal of Cancer 102, 369-375CrossRefGoogle ScholarPubMed
27Dontu, G. and Wicha, M.S. (2005) Survival of mammary stem cells in suspension culture: implications for stem cell biology and neoplasia. Journal of Mammary Gland Biology and Neoplasia 10, 75-86CrossRefGoogle ScholarPubMed
28Finkbeiner, M.R. et al. (2009) Profiling YB-1 target genes uncovers a new mechanism for MET receptor regulation in normal and malignant human mammary cells. Oncogene 28, 1421-1431CrossRefGoogle ScholarPubMed
29Lim, E. et al. (2009) Aberrant luminal progenitors as the candidate target population for basal tumor development in BRCA1 mutation carriers. Nature Medicine 15, 907-913CrossRefGoogle ScholarPubMed
30Asselin-Labat, M.L. et al. (2006) Steroid hormone receptor status of mouse mammary stem cells. Journal of the National Cancer Institute 98, 1011-1014CrossRefGoogle ScholarPubMed
31Eirew, P. et al. (2008) A method for quantifying normal human mammary epithelial cells with in vivo regenerative ability. Nature Medicine 14, 1384-1389CrossRefGoogle ScholarPubMed
32Stingl, J. et al. (2006) Purification and unique properties of mammary epithelial stem cells. Nature 439, 993-997CrossRefGoogle ScholarPubMed
33Visvader, J.E. (2009) Keeping abreast of the mammary epithelial hierarchy and breast tumorigenesis. Genes and Development 23, 2563-2577CrossRefGoogle ScholarPubMed
34Wallgren, A., Silfverswärd, C. and Eklund, G. (1976) Prognostic factors in mammary carcinoma. Acta Radiologica: Therapy, Physics, Biology 15, 1-16Google ScholarPubMed
35Croker, A.K. et al. (2009) High aldehyde dehydrogenase and expression of cancer stem cell markers selects for breast cancer cells with enhanced malignant and metastatic ability. Journal of Cellular and Molecular Medicine 13, 2236-2252CrossRefGoogle ScholarPubMed
36Sheridan, C. et al. (2006) CD44+/CD24− breast cancer cells exhibit enhanced invasive properties: an early step necessary for metastasis. Breast Cancer Research 8, R59CrossRefGoogle ScholarPubMed
37Abraham, B.K. et al. (2005) Prevelance of CD44+/CD24−/low cells in breast cancer may not be associated with clinical outcome but may favor distant metastasis. Clinical Cancer Research 11, 1154-1159CrossRefGoogle Scholar
38Liu, R. et al. (2007) The prognostic role of a gene signature from tumorigenic breast-cancer cells. New England Journal of Medicine 356, 217-226CrossRefGoogle ScholarPubMed
39Gupta, P.B. et al. (2009) Identification of selective inhibitors of cancer stem cells by high-throughput screening. Cell 138, 1-15CrossRefGoogle ScholarPubMed
40Pandit, T.S. et al. (2009) Lymphatic metastasis of breast cancer cells is associated with differential gene expression profiles that predict cancer stem cell-like properties and the ability to survive, establish and grow in a foreign environment. International Journal of Oncology 35, 297-308Google Scholar
41Stratford, A.L. et al. (2007) Epidermal growth factor receptor (EGFR) is transcriptionally induced by the Y-box binding protein-1 (YB-1) and can be inhibited with Iressa in basal-like breast cancer providing a potential target for therapy. Breast Cancer Research 9, R61CrossRefGoogle ScholarPubMed
42Cheang, M.C.U. et al. (2008) Basal-like breast cancer defined by five biomarkers has superior prognostic value than triple-negative phenotype. Clinical Cancer Research 14, 1368-1376CrossRefGoogle ScholarPubMed
43Marangoni, E. et al. (2007) A new model of patient tumor-derived breast cancer xenografts for preclinical assays. Clinical Cancer Research 13, 3989-3998CrossRefGoogle ScholarPubMed
44Farnie, G. et al. (2007) Novel cell culture technique for primary ductal carcinoma in situ: role of notch and epidermal growth factor receptor signaling pathways. Journal of the National Cancer Institute 99, 616-627CrossRefGoogle ScholarPubMed
45Di Leo, A. et al. (2008) Phase III, double-blind, randomized study comparing lapatinib plus paclitaxel with placebo plus paclitaxel as first-line treatment for metastatic breast cancer. Journal of Clinical Oncology 26, 5544-5552CrossRefGoogle ScholarPubMed
46Finn, R.S. et al. (2009) Estrogen receptor, progesterone receptor, human epidermal growth factor receptor 2 (HER2), and epidermal growth factor receptor expression and benefit from lapatinib in a randomized trial of paclitaxel with lapatinib or placebo as first-line treatment in HER2-negative or unknown metastatic breast cancer. Journal of Clinical Oncology 27, 3908-3915CrossRefGoogle ScholarPubMed
47Green, M.D. et al. (2009) Gefitinib treatment in hormone-resistant and hormone receptor-negative advanced breast cancer. Annals of Oncology 20, 1813-1817CrossRefGoogle ScholarPubMed
48Polychronis, A. et al. (2005) Preoperative gefitinib versus gefitinib and anastrozole in postmenopausal patients with oestrogen-receptor positive and epidermal-growth-factor-receptor-positive primary breast cancer: a double-blind placebo-controlled phase II randomised trial. Lancet Oncology 6, 383-391CrossRefGoogle ScholarPubMed
49Normanno, N. et al. (2008) Breast cancer cells with acquired resistance to EGFR tyrosine kinase inhibitor gefitinib show persistant activation of MAPK signaling. Breast Cancer Research and Treatment 112, 25-33CrossRefGoogle ScholarPubMed
50She, Q.-B. et al. (2003) Resistance to gefitinib in PTEN-null HER-overexpressing tumor cells can be overcome through restoration of PTEN function or pharmacologic modulation of constitutive phosphatidylinositol 3′-kinase/Akt pathway signaling. Clinical Cancer Research 9, 4340-4346Google ScholarPubMed
51Corkery, B. et al. (2009) Epidermal growth factor receptor as a potential therapeutic target in triple-negative breast cancer. Annals of Oncology 20, 862-867CrossRefGoogle ScholarPubMed
52Hoadley, K.A. et al. (2007) EGFR associated expression profiles vary with breast tumor subtype. BMC Genomics 8, 258CrossRefGoogle ScholarPubMed
53Oliveras-Ferraros, C. et al. (2008) Growth and molecular interactions of the anti-EGFR antibody Cetuximab and the DNA cross-linking agent cisplatin in gefitinib-resistant MDA-MB-468 cells: new prospects in the treatment of triple-negative/basal-like breast cancer. International Journal of Oncology 33, 1165-1176Google ScholarPubMed
54Nielsen, T. et al. (2004) Expression of the insulin-like growth factor-1 receptor and urokinase plasminogen activator in breast cancer is associated with poor survival: potential for intervention with 17 allylamino 17-demethoxy geldanamycin. Cancer Research 64, 286-291CrossRefGoogle Scholar
55Law, J.H. et al. (2008) Phosphorylated insulin-like growth factor-1/insulin receptor is present in all breast cancer subtypes and is related to poor survival. Cancer Research 68, 10238-10246CrossRefGoogle ScholarPubMed
56Klinakis, A. et al. (2009) Igf1r as a therapeutic target in a mouse model of basal-like breast cancer. Proceedings of the National Academy of Sciences of the United States of America 106, 2359-2364CrossRefGoogle Scholar
57Carboni, J.M. et al. (2005) Tumor development by transgenic expression of a constitutively active insulin-like growth factor I receptor. Cancer Research 65, 3781-3787CrossRefGoogle ScholarPubMed
58Kim, H.-G. et al. (2007) Constitutively active type I insulin-like growth factor receptor causes transformation and xenograft growth of immortalized mammary epithelial cells and is accompanied by an epithelial-to-mesenchymal transition mediated by NF-ĸB and snail. Molecular and Cellular Biology 27, 3165-3175CrossRefGoogle Scholar
59Moody, S.E. et al. (2005) The transcriptional repressor Snail promotes mammary tumor recurrence. Cancer Cell 8, 197-209CrossRefGoogle ScholarPubMed
60Jones, R.A. et al. (2009) Reversibility and recurrence of IGF-IR-induced mammary tumors. Oncogene 28, 2152-2162CrossRefGoogle ScholarPubMed
61Camp, R.L., Rimm, E.B. and Rimm, D.L. (1999) Met expression is associated with poor outcome in patients with axillary lymph node negative breast carcinoma. Cancer 86, 2259-22653.0.CO;2-2>CrossRefGoogle ScholarPubMed
62Ghoussoub, R.A. et al. (1998) Expression of c-met is a strong independent prognostic factor in breast carcinoma. Cancer 82, 1513-15203.0.CO;2-7>CrossRefGoogle Scholar
63Welm, A.L. et al. (2005) MET and MYC cooperate in mammary tumorigenesis. Proceedings of the National Academy of Sciences of the United States of America 102, 4324-4329CrossRefGoogle ScholarPubMed
64Orian-Rousseau, V. et al. (2002) CD44 is required for two consecutive steps in HGF/c-MET signaling. Genes and Development 16, 3074-3086CrossRefGoogle ScholarPubMed
65Graveel, C.R. et al. (2009) Met induces diverse mammary carcinomas in mice and is associated with human basal breast cancer. Proceedings of the National Academy of Sciences of the United States of America 106, 12909-12914CrossRefGoogle ScholarPubMed
66Ponzo, M.G. et al. (2009) Met induces mammary tumors with diverse histologies and is associated with poor outcome and human basal breast cancer. Proceedings of the National Academy of Sciences of the United States of America 106, 12903-12908CrossRefGoogle ScholarPubMed
67Comoglio, P., Giordano, S. and Trusolino, L. (2008) Drug development of MET inhibitors: targeting oncogene addiction and expedience. Nature Reviews Drug Discovery 7, 504-516CrossRefGoogle ScholarPubMed
68Hines, S.J. et al. (1995) Coexpression of the c-kit and stem cell factor genes in breast carcinomas. Cell Growth & Differentiation 6, 769-779Google ScholarPubMed
69Tsuda, H. et al. (2005) Frequent KIT and epidermal growth factor receptor overexpressions in undifferentiated-type breast carcinomas with ‘stem-cell-like’ features. Cancer Science 96, 333-339CrossRefGoogle ScholarPubMed
70Levina, V. et al. (2010) Elimination of human lung cancer stem cells through targeting of the stem cell factor-c-kit autocrine signaling loop. Cancer Research 70, 338-346CrossRefGoogle ScholarPubMed
71Nielsen, T.O. et al. (2004) Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast cancer. Clinical Cancer Research 10, 5367-5374CrossRefGoogle Scholar
72Renouf, D.J., Wilson, L. and Blanke, C.D. (2009) Successes and challenges in translational research: the development of targeted for gastrointestinal stromal tumours. Clinical Cancer Research 15, 3908-3911CrossRefGoogle ScholarPubMed
73Heinrich, M.C. et al. (2003) Kinase mutations and imatinib response in patients with metastatic gastrointestinal stromal tumor. Journal of Clinical Oncology 21, 4342-4349CrossRefGoogle ScholarPubMed
74Zhou, J. et al. (2007) Activation of the PTEN/mTOR/STAT3 pathway in breast cancer stem-like cells is required for viability and maintenance. Proceedings of the National Academy of Sciences of the United States of America 104, 16158-16163CrossRefGoogle ScholarPubMed
75Eralp, Y. et al. (2008) MAPK overexpression is associated with anthracycline resistance and increased risk for recurrence in patients with triple-negative breast cancer. Annals of Oncology 19, 669-674CrossRefGoogle ScholarPubMed
76Chen, R.-H., Abate, C. and Blenis, J. (1993) Phosphorylation of the c-Fos transrepression domain by mitogen-activated protein kinase and 90-kDa ribosomal S6 kinase. Proceedings of the National Academy of Sciences of the United States of America 90, 10952-10956CrossRefGoogle ScholarPubMed
77Schouten, G.J. et al. (1997) IKappaB alpha is a target for the mitogen-activated 90 kDa ribosomal S6 kinase. EMBO Journal 16, 3133-3144CrossRefGoogle ScholarPubMed
78Joel, P.B. et al. (1998) pp90rsk1 regulates estrogen receptor-mediated transcription through phosphorylation of Ser-167. Molecular and Cellular Biology 18, 1978-1984CrossRefGoogle ScholarPubMed
79Stratford, A.L. et al. (2008) Y-box binding protein-1 serine 102 is a downstream target of p90 ribosomal S6 kinase in basal-like breast cancer cells. Breast Cancer Research 10, R99CrossRefGoogle ScholarPubMed
80Shimamura, A. et al. (2000) Rsk1 mediates a MEK-MAP kinase cell survival signal. Current Biology 10, 127-135CrossRefGoogle ScholarPubMed
81Anjum, R. et al. (2005) The tumor suppressor DAP kinase is a target of RSK-mediated survival signaling. Current Biology 15, 1762-1767CrossRefGoogle ScholarPubMed
82Larrea, M.D. et al. (2009) RSK1 drives p17Kip1 phosphorylation at T198 to promote RhoA inhibition and increase cell motility. Proceedings of the National Academy of Sciences of the United States of America 106, 9268-9273CrossRefGoogle Scholar
83Clark, D.E. et al. (2005) The serine/threonine protein kinase, p90 ribosomal S6 kinase, is an important regulator of prostate cancer cell proliferation. Cancer Research 65, 3108-3116CrossRefGoogle ScholarPubMed
84Cuadrado, A. and Nebreda, A.R. (2007) New insights into RSK activation and hematopoietic cancer. Cancer Cell 12, 187-189CrossRefGoogle ScholarPubMed
85Smith, J.A. et al. (2005) Identification of the first specific inhibitor of p90 ribosomal S6 kinase (RSK) reveals an unexpected role for RSK in cancer cell proliferation. Cancer Research 65, 1027-1034CrossRefGoogle ScholarPubMed
86Xian, W. et al. (2009) Fibroblast growth factor receptor 1-transformed mammary epithelial cells are dependent on RSK activity for growth and survival. Cancer Research 69, 2244-2251CrossRefGoogle ScholarPubMed
87Cho, Y.-Y. et al. (2007) Ribosomal S6 kinase 2 is a key regulator in tumor promoter-induced cell transformation. Cancer Research 67, 8104-8112CrossRefGoogle ScholarPubMed
88Sapkota, G.P. et al. (2007) BI-D1870 is a specific inhibitor of the p90 RSK (ribosomal S6 kinase) isoforms in vitro and in vivo. Journal of Biochemistry 401, 29-38CrossRefGoogle ScholarPubMed
89Cohen, M.S., Hadjivassiliou, H. and Taunton, J. (2007) A clickable inhibitor reveals context-dependent autoactivation of p90 RSK. Nature Chemical Biology 3, 156-160CrossRefGoogle ScholarPubMed
90Kleer, C.G. et al. (2003) EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells. Proceedings of the National Academy of Sciences of the United States of America 100, 11606-11611CrossRefGoogle ScholarPubMed
91Suvà, M.-L. et al. (2009) EZH2 is essential for glioblastoma cancer stem cell maintenance. Cancer Research 69, 9211-9218CrossRefGoogle ScholarPubMed
92Gonzalez, M.E. et al. (2009) Downregulation of EZH2 decreases growth of estrogen receptor-negative invasive breast carcinoma and requires BRCA1. Oncogene 28, 843-853CrossRefGoogle ScholarPubMed
93Bergmann, S. et al. (2005) YB-1 provokes breast cancer through the induction of chromosomal instability that emerges from mitotic failure and centrosome amplification. Cancer Research 65, 4078-4087CrossRefGoogle ScholarPubMed
94Bargou, R.C. et al. (1997) Nuclear localization and increased levels of transcription factor YB-1 in primary human breast cancers are associated with intrinsic MDR1 gene expression. Nature Medicine 3, 447-450CrossRefGoogle ScholarPubMed
95Kuwano, M. et al. (2003) The basic and clinical implications of ABC transporters, Y-box-binding protein-1 (YB-1) and angiogenesis-related factors in human malignancies. Cancer Science 94, 9-14CrossRefGoogle ScholarPubMed
96Sutherland, B.W. et al. (2005) Akt phosphorylates the Y-box binding 1 at Ser102 located in the cold shock domain and affects the anchorage-independent growth of breast cancer cells. Oncogene 24, 4281-4292CrossRefGoogle ScholarPubMed
97Sakura, H. et al. (1988) Two human genes isolated by a novel method encode DNA-binding proteins containing a common region of homology. Gene 73, 499-507Google ScholarPubMed
98Wu, J. et al. (2006) Disruption of the Y-box binding protein-1 results in suppression of the epidermal growth factor receptor and HER-2. Cancer Research 66, 4872-4879CrossRefGoogle ScholarPubMed
99Astanehe, A. et al. (2009) The transcriptional induction of PIK3CA in tumor cells is dependent on the oncoprotein Y-box binding protein-1. Oncogene 28, 2406-2418CrossRefGoogle ScholarPubMed
100Habibi, G. et al. (2008) Re-defining prognostic factors for breast cancer: YB-1 is a stronger predictor of relapse and disease specific survival than estrogen receptor or HER-2 across all tumor subtypes. Breast Cancer Research 10, R86CrossRefGoogle ScholarPubMed
101Lee, C. et al. (2008) Targeting YB-1 in HER-2 over-expressing breast cancer cells induces apoptosis via the mTOR/STAT3 pathway and suppresses tumor growth in mice. Cancer Research 68, 8661-8666CrossRefGoogle Scholar
102Judge, A.D. et al. (2009) Confirming the RNAi-mediated mechanism of action of siRNA-based cancer therapeutics in mice. Journal of Clinical Investigation 119, 661-673CrossRefGoogle ScholarPubMed
103Semple, S.C. et al. (2010) Rational design of cationic lipids for siRNA delivery. Nature Biotechnology 28, 172-176CrossRefGoogle ScholarPubMed
104Narod, S.A. and Foulkes, W.D. (2004) BRCA1 and BRCA2: 1994 and beyond. Nature Reviews Cancer 4, 665-676CrossRefGoogle ScholarPubMed
105Stefansson, O.A. et al. (2009) Genomic profiling of breast tumours in relation to BRCA abnormalities and phenotypes. Breast Cancer Research 11, R47CrossRefGoogle ScholarPubMed
106Foulkes, W.D. (2004) BRCA1 functions as a breast stem cell regulator. Journal of Medical Genetics 41, 1-5CrossRefGoogle ScholarPubMed
107Bane, A.L. et al. (2007) BRCA2 mutation-associated breast cancers exhibit a distinguishing phenotype based on morphology and molecular profiles from tissue microarrays. American Journal of Surgical Pathology 31, 121-128CrossRefGoogle ScholarPubMed
108Waddell, N. et al. (2009) Subtypes of familial breast tumours revealed by expression and copy number profiling. Breast Cancer Research and Treatment Dec 4; [Epub ahead of print]Google Scholar
109Turner, N.C. and Reis-Filho, J.S. (2006) Basal-like breast cancer and BRCA1 phenotype. Oncogene 25, 5846-5853CrossRefGoogle ScholarPubMed
110McCarthy, A. et al. (2007) A mouse model of basal-like breast carcinoma with metaplastic elements. Journal of Pathology 211, 389-398CrossRefGoogle ScholarPubMed
111Deng, C.X. and Wang, R.H. (2003) Roles of BRCA1 in DNA damage repair: a link between development and cancer. Human Molecular Genetics 12 Spec No 1, R113-R123CrossRefGoogle ScholarPubMed
112Bryant, H.E. et al. (2005) Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase. Nature 434, 913-917CrossRefGoogle ScholarPubMed
113Farmer, H. et al. (2005) Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 434, 917-921CrossRefGoogle ScholarPubMed
114Evers, B. et al. (2008) Selective inhibition of BRCA2-deficient mammary tumor cell growth by AZD2281 and cisplatin. Clinical Cancer Research 14, 3916-3925CrossRefGoogle ScholarPubMed
115Rottenberg, S. et al. (2008) High sensitivity of BRCA1-deficient mammary tumors to the PARP inhibitor AZD2281 alone and in combination with platinum drugs. Proceedings of the National Academy of Sciences of the United States of America 105, 17079-17084CrossRefGoogle Scholar
116Byrski, T. et al. (2009) Pathologic complete response rates in young women with BRCA1-positive breast cancers after neoadjuvant chemotherapy. Journal of Clinical Oncology 28, 375-379CrossRefGoogle ScholarPubMed
117Carey, L.A. (2010) Targeted chemotherapy? Platinum in BRCA1-dysfuntional breast cancer. Journal of Clinical Oncology 28, 361-363CrossRefGoogle ScholarPubMed
118Evers, B. et al. (2010) A high-throughput pharmaceutical screen identifies compounds with specific toxicity against BRCA2-deficient tumors. Clinical Cancer Research 16, 99-108CrossRefGoogle ScholarPubMed
119Fong, P.C. et al. (2009) Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. New England Journal of Medicine 361, 123-134CrossRefGoogle ScholarPubMed
120Furuta, S. et al. (2005) Depletion of BRCA1 impairs differentiation but enhances proliferation of mammary epithelial cells. Proceedings of the National Academy of Sciences of the United States of America 102, 9176-9181CrossRefGoogle ScholarPubMed
121Liu, S. et al. (2008) BRCA1 regulates human mammary stem/progenitor cell fate. Proceedings of the National Academy of Sciences of the United States of America 105, 1680-1685CrossRefGoogle ScholarPubMed
122Ginestier, C., Liu, S. and Wicha, M.S. (2009) Getting to the root of BRCA1-deficient breast cancer. Cell Stem Cell 5, 229-230CrossRefGoogle Scholar
123Gautschi, O. et al. (2008) Aurora kinases as anticancer drug targets. Clinical Cancer Research 14, 1639-1648CrossRefGoogle ScholarPubMed
124Chang, S. et al. (2009) Expression of human BRCA1 variants in mouse ES cells allows functional analysis of BRCA1 mutations. Journal of Clinical Investigation 119, 3160-3171CrossRefGoogle ScholarPubMed
125Ouchi, M. et al. (2004) BRCA1 phosphorylation by Aurora-A in the regulation of G2 to M transition. Journal of Biological Chemistry 279, 19643-19648CrossRefGoogle ScholarPubMed
126Sankaran, S. et al. (2007) Aurora-A kinase regulates breast cancer associated gene 1 inhibition of centrosome-dependent microtubule nucleation. Cancer Research 67, 11186-11194CrossRefGoogle ScholarPubMed
127Hennessy, B.T. et al. (2009) Characterization of a naturally occurring breast cancer subset enriched in epithelial-to-mesenchymal transition and stem cell characteristics. Cancer Research 69, 4116-4124CrossRefGoogle ScholarPubMed
128Honeth, G. et al. (2008) The CD44+/CD24− phenotype is enriched in basal-like breast tumors. Breast Cancer Research 10, R53CrossRefGoogle ScholarPubMed
129Tanei, T. et al. (2009) Association of breast cancer stem cells identified by aldehyde dehydrogenase 1 expression with resistance to sequential paclitaxel and epirubicin-based chemotherapy for breast cancer. Clinical Cancer Research 15, 4234-4241CrossRefGoogle Scholar
130Kai, K. et al. (2010) Breast cancer stem cells. Breast Cancer 17, 80-85CrossRefGoogle ScholarPubMed
131Sleeman, J.P. and Cremers, N. (2007) New concepts in breast cancer metastasis: tumour initiating cells and the microenvironment. Clinical and Experimental Metastasis 24, 707-715CrossRefGoogle ScholarPubMed
Stockmans, G. et al. (2008) Triple-negative breast cancer. Current Opinions in Oncology 20, 614-620CrossRefGoogle ScholarPubMed
Tan, A.R. and Swain, S.M. (2008) Therapeutic strategies for triple-negative breast cancer. The Cancer Journal 14, 343-351CrossRefGoogle ScholarPubMed
Visvader, J.E. (2009) Keeping abreast of the mammary epithelial hierarchy and breast tumorigenesis. Genes and Development 23, 2563-2577CrossRefGoogle ScholarPubMed
Shipitsin, M. and Polyak, K. (2008) The cancer stem cell hypothesis: in search of definitions, markers and relevance. Laboratory Investigation 88, 459-463CrossRefGoogle ScholarPubMed
Kai, K., Arima, Y., Kamiya, T. and Saya, H. (2010) Breast cancer stem cells. Breast Cancer 17, 80-85CrossRefGoogle ScholarPubMed
Diamandis, P. et al. (2007) Chemical genetics reveals a complex functional ground state of neural stem cells. Nature Chemical Biology 3, 268-273CrossRefGoogle ScholarPubMed
Stockmans, G. et al. (2008) Triple-negative breast cancer. Current Opinions in Oncology 20, 614-620CrossRefGoogle ScholarPubMed
Tan, A.R. and Swain, S.M. (2008) Therapeutic strategies for triple-negative breast cancer. The Cancer Journal 14, 343-351CrossRefGoogle ScholarPubMed
Visvader, J.E. (2009) Keeping abreast of the mammary epithelial hierarchy and breast tumorigenesis. Genes and Development 23, 2563-2577CrossRefGoogle ScholarPubMed
Shipitsin, M. and Polyak, K. (2008) The cancer stem cell hypothesis: in search of definitions, markers and relevance. Laboratory Investigation 88, 459-463CrossRefGoogle ScholarPubMed
Kai, K., Arima, Y., Kamiya, T. and Saya, H. (2010) Breast cancer stem cells. Breast Cancer 17, 80-85CrossRefGoogle ScholarPubMed
Diamandis, P. et al. (2007) Chemical genetics reveals a complex functional ground state of neural stem cells. Nature Chemical Biology 3, 268-273CrossRefGoogle ScholarPubMed
Figure 0

Figure 1.Schematic of the analogy between tumour-initiating cells (TICs) and the phoenix. Tumours are heterogeneous, with TICs making up only a small proportion of the total. TICs are analogous to the phoenix in that they are both immortal and long-lived. The phoenix has the ability to self-renew and to give birth to immortalised progeny just as TICs can result in ‘rebirth’ of the tumour. (a) Following chemotherapy, the majority of the cancer cells undergo cell death; however, the TICs remain dormant and poised for cell renewal. These nests of phoenix-like cells then give rise to secondary tumours. (b) Several studies have reported that chemotherapy actually induces TICs (Refs 5, 10, 11, 12, 13, 14, 15). To illustrate this, TICs are represented as a minor subpopulation of the entire tumour, but following chemotherapy these resistant cells expand. The emergence of TICs positions the tumour for drug resistance and eventual cancer recurrence.

Figure 1

Table 1.Techniques used to characterise breast cancer stem cells

Figure 2

Figure 2.Mammary gland stem cell origins and progression to breast cancer subtypes. Mammary stem cells have been defined by a number of surface markers that have been recently reviewed by several groups (Refs 33, 122, 130). These cells give rise to a common bipotential progenitor that can produce myoepithelial or luminal cell progenitors. At the present time, the evidence is that luminal progenitors give rise to triple-negative breast cancer (TNBC), and through subsequent changes that have not been thoroughly investigated luminal A and luminal B breast cancer subtypes develop as cells differentiate toward the ductal phenotype. The derivation of the ‘normal’ subtype (Ref. 2) is unclear.

Figure 3

Figure 3.Organisation of the mammary gland. The mammary gland is depicted schematically (a) and visually from primary mouse mammary tissue via haematoxylin/eosin staining (b). The myoepithelial layer resides below the luminal cell layer and is readily identifiable by staining for either smooth muscle actin (c) or CD49f, similar to findings by Visvader et al. (Ref. 33) (d).

Figure 4

Figure 4.Chromatin immunoprecipitation on-chip can be used to identify novel pathways and potential targets for treating TNBC. (a) Many transcription factors are reportedly induced in TNBC including YB-1. In order to understand how YB-1 promotes TNBC cell growth, invasion and possibly EMT conversion, one can profile targets on a genome-wide basis using chromatin immunoprecipitation on-chip. Briefly, crosslinked and sonicated chromatin is immunoprecipitated with anti-YB-1 antibodies, allowing identification of promoter regions targeted by YB-1 following analysis of the associated DNA. (b) Example of a genome-wide study of YB-1 target genes in the TNBC cell line SUM149 (Ref. 28). Many of the target genes have prior links to tumour initiation, including (1) the cell-surface or receptor molecules CD44, CD49f, MET and several members of the NOTCH family, (2) extracellular signalling molecules of the WNT family, and (3) cell cycle regulators. Solid arrows indicate direct interactions; broken arrows represent indirect relationships; curved arrows indicate ability to self-regulate. Abbreviations: EMT, epithelial to mesenchymal transition; TNBC, triple-negative breast cancer.