• Call for Social Sciences Papers
  • Sign-up for PNAS eTOC Alerts

Progress in breast cancer research

  1. Peter K. Vogtc,1
  1. aDepartment of Medical Oncology, Dana–Farber Cancer Institute, Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02215;
  2. bHarvard Stem Cell Institute, Cambridge, MA 02138; and
  3. cThe Scripps Research Institute, La Jolla, CA 92037

Breast cancer is a heterogeneous disease composed of multiple different subtypes with distinct molecular features and clinical behavior (1). Gene expression profiling studies have identified at least four major subtypes classified as luminal A, luminal B, HER2+, and basal-like (2). Genome-wide expression profiling of breast tumors has not entered routine clinical practice, but molecular classification according to immunohistochemical assessment of estrogen and progesterone receptors and of HER2 is the basis of individualized therapy that has been guiding the clinical management of patients with breast cancer for the past two decades. Indeed, the first molecular target for cancer therapy was the estrogen receptor (ER) in breast cancer; inhibitors of estrogen signaling display profound antitumor effects in ER+ tumors that show dependency for this pathway (3). Similarly, breast tumors with amplification and overexpression of HER2 (the ERBB2 oncogene) respond well to anti-HER2 targeted therapy (4). ER and HER2 are two of the best examples for molecules that are therapeutic targets and also identify patients who are likely to show a response to treatment, satisfying two important goals of clinical oncology. As a result, the application of antihormonal and anti-HER2 targeted therapies led to dramatic improvements in the outcomes of patients with ER+ and HER2+ disease. However, one of the inevitable consequences of targeted therapy is the selection for tumor cells with inherent or acquired resistance to treatment. Unfortunately, a significant fraction of patients with ER+ or HER2+ tumors do not show a response to treatment or experience relapse and disease progression. Thus, in addition to continuous improvements in the efficacy of antihormonal and anti-HER2 targeted therapies, investigating and overcoming resistance mechanisms and identifying new therapeutic targets have been one of the most intense areas of breast cancer research. The large number of articles focusing on these topics included in this special feature issue of PNAS is a good reflection of the importance of these efforts.

Recent unbiased cancer genome sequencing studies have identified several recurrently mutated genes in breast cancer that represent putative novel therapeutic targets (5). One of the major findings of these studies is the identification of PI3KCA as one of the most frequently mutated genes in breast and other cancer types. Therapeutic targeting of the PI3K/AKT signaling pathway has been a major focus of several drug companies, leading to the development and clinical testing of several PI3K and AKT inhibitors. Chakrabarty et al., in Arteaga's laboratory, have investigated the consequences of treatment with the XL147 PI3K inhibitor in a panel of HER2+ breast cancer cell lines (6). To their surprise, in a subset of cells, they observed up-regulation of phosphorylated HER3 and partial recovery of phospho-AKT following XL147 treatment, leading to incomplete suppression of tumor cell growth. Based on follow-up experiments, they demonstrated that the combined inhibition of HER2 and PI3K leads to synergistic effects and more efficient eradication of the tumors, a finding that was corroborated by two additional research groups (7, 8). These results are an example for the complexity of signaling pathways in cancer cells complicated by multiple layers of feedback inhibition.

Heiser et al. conducted an unbiased screen in approximately 50 breast cancer cell lines, well characterized for copy number alterations and gene expression profiles, with the aim of finding new subtype-specific therapeutic agents (9). They applied 77 Food and Drug Administration-approved compounds on the cells and assessed their growth following treatment. In addition to confirming the known efficacy of PI3K/AKT pathway-targeting compounds and HDACs in HER2+ and luminal breast cancer cell lines, respectively, they also identified several compounds, including inhibitors of Aurora kinase and FGF receptor, that preferentially inhibited the growth of basal-like and claudin-low cell lines, two subclasses of triple-negative breast cancer (i.e., negative for ER, progesterone receptor, and HER2) that currently lack targeted treatment.

Similarly, Mendes-Pereira et al. performed a genome-wide RNAi screen in the MCF7 ER+ estrogen-dependent breast cancer cell line to identify mediators of tamoxifen resistance (10). They identified several genes, including BAP1, CDK10, NF1, NIPBL, PTEN, RARG, SMC3, and UBA3, whose silencing confers tamoxifen resistance in MCF7 cells and also demonstrated that low expression of these genes in ER+ breast tumors treated with tamoxifen is associated with poor clinical outcome. Two additional manuscripts by Sukumar and coworkers (11) and Haughian et al. (12) describe a role for the HOXB7 homeogene and Notch, respectively, in resistance to hormonal therapies in ER+ breast tumors. A study by Carroll and coworkers, on the contrary, investigated the mechanism by which TLE1 (transducin-like enhancer protein 1) modulates the transcriptional activity of ER (13). By assessing genome-wide TLE1 binding in MCF-7 breast cancer cells by using ChIPseq (ChIP combined with high-throughput sequencing), Carroll and coworkers found a significant overlap with ER targets. Subsequently, they down-regulated TLE1 expression by using siRNAs and demonstrated that TLE1 is required for ER-mediated activation of a subset of genes that play important roles in cell division. These results suggest that the therapeutic modulation of TEL1–ER interaction may represent a new strategy for the treatment of ER+ breast tumors.

Articles from the laboratories of Reed (14) and Maheswaran (15) focus on DNA damage checkpoints and mechanisms of radioresistance and analyze the role of Cks proteins and HOXB9. Oakes et al., on the contrary, demonstrate that combining ABT-737, a BH3 mimetic compound, with chemotherapy leads to increased apoptosis and suppression of tumor growth (16). Overall, all these studies demonstrate that, by combining currently used cancer therapy (e.g., radiation or chemotherapy) with modulators of apoptotic response, one can achieve more effective killing of cancer cells.

Cancer cells with stem cell features (also known as cancer stem cells or tumor-initiating cells) have also been implicated in tumor progression and therapeutic resistance. Several articles in this special feature focus on the characterization of these cells and signaling pathways maintaining their phenotypes. Keller et al. have investigated putative cellular precursors of distinct breast cancer subtypes by transforming cell populations purified according to the expression of cell surface markers such as EpCAM and CD10 (17). They report that EpCAM+ cells may be the precursor to luminal and basal-like breast tumors, whereas metaplastic and claudin-low tumors may originate from CD10+ cells. Proving the cell of origin of breast cancer is an impossible task in human patients, and it is not likely to influence the clinical management of patients with breast cancer, but it would be important for the design of improved risk prediction and cancer prevention strategies.

By using an unbiased comparative oncogenomics approach, Herschkowitz et al. demonstrated that claudin-low breast tumors are enriched in functional cancer stem cells (18), whereas a study by Wicha and coworkers reports that the frequency of cancer stem cells is increasing following antiangiogenic therapy, potentially contributing to the higher incidence of metastatic lesions observed following such treatments (19). The overexpression of transcription factors that may convert breast cancer cells to a less differentiated stem cell-like state have been implicated as a possible mechanism underlying both intertumor and intratumor heterogeneity in breast cancer (20). MYC is one of the key transcription factors critical for reprogramming cells to more embryonic less differentiated state (21), and MYC amplification and overexpression is common in ER? breast cancer with poor outcome. A report by Zhang et al. investigated the mechanisms by which Myc is stabilized in breast cancer cells (22). Interestingly, they found that the protein stability, phosphorylation at threonine 58 and serine 62, and oncogenic activity of Myc were increased in breast cancer cell lines and primary tumors. The phosphorylation of threonine 58/serine 62 is important for Myc degradation and is regulated by the GSK3β/PP2A-B56α/Axin1 complex. The authors found that the mRNA and protein levels of Axin1 were significantly decreased, leading to higher stability and activation of Myc. These findings imply that compounds that stabilize Axin1 may be effective therapeutic agents in breast tumors with increased Myc levels.

The importance of the tumor microenvironment in cancer development and therapeutic responses has been widely recognized (23). Several studies published in this special issue address the role of microenvironmental changes using different approaches. Ruffell et al., in the Coussens laboratory, analyzed the composition of leukocytes in normal breast tissues and in breast tumors before and after neoadjuvant chemotherapy by using an elegant polychromatic FACS and multicolor confocal immunofluorescence (24). Their major findings are that T cells predominate tumors, whereas cells of the myeloid lineage are more prominent in normal breast. Furthermore, the frequency of myeloid cells increases following chemotherapy, accompanied by increased ratio of CD8/CD4 T cells and higher number of granzyme B-expressing cells. The importance of breast tumor leukocyte composition, particularly the relative frequencies of cytotoxic and immunosuppressive T cells, in therapeutic responses and clinical outcome has been demonstrated by several recent studies, igniting interest in immunotherapy in breast cancer (25, 26). In line with this, a collaborative group of B?rresen-Dale and coworkers performed integrated molecular profiling of ductal carcinoma in situ and found that T cell-related signatures correlate with tumor subtype and clinical outcome (27).

Marcotte et al. (28) and Owens et al. in the Moses laboratory (29) used transgenic mouse models to investigate the roles of c-Src and BMPR2, respectively, in mammary tumor development. Both groups performed epithelial-specific targeting, but in the case of BMPR2, expressing a dominant-negative receptor in mammary epithelial cells promoted metastatic progression by cell-autonomous and paracrine mechanisms. Specifically, epithelium-specific inhibition of BMP signaling led to increased CCL9 expression and infiltration of myeloid suppressor cells, which promote metastatic spread.

The role of noncoding RNAs in tumorigenesis has been increasingly recognized and investigated (30). Maruyama and colleagues made the interesting observation that the ratios of antisense-to-sense transcripts is consistently and significantly different between normal breast epithelial and breast cancer cells (31). Specifically, in a significant fraction of protein coding genes, the levels of antisense transcripts are lower in tumor cells than those of sense transcripts transcribed from the same locus. Many of these antisense transcripts appear to correspond to long noncoding RNAs, and many of the genes from which they are transcribed encode proteins with basic metabolic functions. These findings imply global differences in RNA regulation between normal and neoplastic breast epithelial cells that may promote tumorigenesis.

The collection of studies published in this special feature provides an overview of major research areas and of progress in current breast cancer research. Many of the results presented here have immediate translational relevance and may lead to improved treatment outcomes in patients.


We thank members of our laboratories for their critical reading of this manuscript and useful discussions. This work was supported by the National Cancer Institute (K.P.), US Army Congressionally Directed Research (K.P.), Avon Research Foundation (K.P.), V Foundation (K.P.), Breast Cancer Research Foundation (K.P.), Novartis (K.P.), Susan G. Komen Foundation (K.P.), and National Cancer Institute Grants R01 CA078230, R01 CA153124, and R01 CA151574 (to P.K.V.).


  • ?1To whom correspondence may be addressed. E-mail: kornelia_polyak{at}dfci.harvard.edu or pkvogt{at}scripps.edu.
  • Author contributions: K.P. and P.K.V. wrote the paper.

  • Conflict of interest statement: K.P. receives research support from Novartis Oncology and is a consultant to Novartis Oncology. K.P. serves on the Scientific Advisory Board of Theracrine.


  1. ?
  2. ?
  3. ?
  4. ?
  5. ?
  6. ?
  7. ?
  8. ?
  9. ?
  10. ?
  11. ?
  12. ?
  13. ?
  14. ?
  15. ?
  16. ?
  17. ?
  18. ?
  19. ?
  20. ?
  21. ?
  22. ?
  23. ?
  24. ?
  25. ?
  26. ?
  27. ?
  28. ?
  29. ?
  30. ?
  31. ?

Online Impact

    <acronym id="UPyyYwe"></acronym>
    <rt id="UPyyYwe"></rt>
    <rt id="UPyyYwe"><small id="UPyyYwe"></small></rt>
    <rt id="UPyyYwe"><small id="UPyyYwe"></small></rt>
    <acronym id="UPyyYwe"><optgroup id="UPyyYwe"></optgroup></acronym>
    <tr id="UPyyYwe"><optgroup id="UPyyYwe"></optgroup></tr>
    <tr id="UPyyYwe"><optgroup id="UPyyYwe"></optgroup></tr>
    <acronym id="UPyyYwe"></acronym>
    <acronym id="UPyyYwe"></acronym>
    <acronym id="UPyyYwe"><small id="UPyyYwe"></small></acronym>
  • 29548817 2018-01-19
  • 627633816 2018-01-19
  • 515977815 2018-01-19
  • 397754814 2018-01-19
  • 927574813 2018-01-19
  • 122952812 2018-01-19
  • 519337811 2018-01-19
  • 822690810 2018-01-19
  • 475626809 2018-01-19
  • 845873808 2018-01-18
  • 263104807 2018-01-18
  • 787145806 2018-01-18
  • 366736805 2018-01-18
  • 889928804 2018-01-18
  • 251505803 2018-01-18
  • 141443802 2018-01-18
  • 705389801 2018-01-18
  • 449977800 2018-01-18
  • 739705799 2018-01-18
  • 205939798 2018-01-18