Genomic Subtypes in Choosing Adjuvant Therapy for Breast Cancer

March 15, 2013
Amelia B. Zelnak, MD, MSc

,
Ruth M. O'Regan, MD

Oncology, ONCOLOGY Vol 27 No 3, Volume 27, Issue 3

Additional insight into the biology of ER-positive breast cancers, particularly the higher risk luminal B cancers, could aid in identifying potential targets and new, effective therapies. And though the majority of triple-negative breast cancers are the “basal-like” subtype, significant proportions are in other subtypes.

The use of gene expression profiling has impacted our understanding of breast cancer biology and increasingly has played a role in guiding clinical decisions. We have used hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) status for years to guide selection of therapy. More recently, gene expression analysis has facilitated the identification of at least five intrinsic subtypes of breast cancer. Potential therapeutic targets have also been identified using genomic profiling. Several tests, such as the 21-gene recurrence score assay (Oncotype DX) and the 70-gene prognosis signature (MammaPrint), have been well validated as prognostic tools for early-stage breast cancer, and have aided in adjuvant therapy decisions for early-stage, HR-positive breast cancer patients. Genomic profiling has the potential to provide additional insight into drug discovery and clinical trial design by identifying appropriate targeted therapies for subtypes of breast cancer.

Introduction

Breast cancer is a heterogeneous disease comprising different subtypes defined on clinical, pathological, and molecular levels. In clinical practice, oncologists have recognized for years that the behavior of breast cancers is variable. Assessment of hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) status is standard of care at the time of breast cancer diagnosis, and currently is used to guide adjuvant therapy recommendations. Estrogen receptor (ER) and HER2 expression are predictive of benefit from endocrine and HER2-targeted therapies, respectively.

Our understanding of breast cancer biology has accelerated with the use of gene expression profiling. Use of gene expression microarrays has facilitated the high-throughput analysis of multiple genes within a single tumor. In 2000, Perou et al first described the use of gene expression arrays in a small cohort of breast cancer patients who were treated with neoadjuvant doxorubicin. They selected a set of approximately 500 genes, which they called the “intrinsic” gene subset since they defined intrinsic properties of the breast cancer. Breast cancers were clustered into groups based upon expression patterns of different genes. They identified clusters of genes associated with proliferation, HER2 signaling, and HR signaling, as well as a group of genes called the “basal” cluster since they shared expression patterns with breast basal epithelial cells.[1]

Subsequent studies in larger cohorts of patients have further defined the intrinsic subtypes of breast cancer.[2-4] The two HR-positive subtypes are called “luminal A” and “luminal B” since they have an expression pattern similar to the luminal epithelial cells of the breast. Luminal A tumors typically have higher levels of ER expression, whereas luminal B tumors typically have higher levels of genes associated with proliferation and HER2. There were several subtypes with low ER expression: the “HER2-enriched” subtype, which is characterized by high expression of HER2 and other genes located in the same region of chromosome 17; the “basal” subtype; and the less common “claudin-low” subtype.[5] The claudin-low subtype, similar to the basal subtype, is characterized by low expression of HR- and HER2-related genes. It remains unclear if the “normal-like” subtype of breast cancer is a real subtype or if it is an artifact related to contamination from tissue surrounding the tumors.

Determination of a breast cancer’s intrinsic subtype using gene expression profiling is not currently performed in routine clinical practice. Often standard immunohistochemistry (IHC) studies of ERs and progesterone receptors (PRs) and HER2 are used as surrogate markers for intrinsic subtypes. The addition of IHC testing for cytokeratin 5/6 facilitated identification of the basal subtype.[6] Carey et al classified luminal A tumors as ER- and/or PR-positive, HER2-negative; luminal B tumors as ER- and/or PR-positive, HER2-positive; HER2-enriched as ER/PR-negative, HER2-positive; and basal-like tumors as ER/PR/HER2-negative (triple-negative), cytokeratin 5/6-positive.[7] The correlation between this IHC-based classification and DNA-based microarray expression profiles was also observed in different studies.

Although one does not obtain the same depth of knowledge regarding tumor biology using IHC instead of DNA microarrays, this information is often readily available in the clinic. One concern in using IHC surrogates is that ER, PR, and HER2 may not accurately identify the intrinsic subtypes. There is the possibility of having false-negative results from the laboratory. In addition, not all basal-like tumors are triple-negative, and some basal-like tumors have ER, PR, or HER2 expression.[6,8] Determining the intrinsic subtype of a breast cancer has significant prognostic value and implications for outcome.[9]

Genomic Profiling in the Clinic

Gene expression profiling by microarray was initially used to identify unique subtypes of breast cancer, but these subtypes also have strong prognostic implications. For example, patients with luminal A tumors have consistently been shown to have a better prognosis than all other subtypes, including the luminal B tumors, which are also ER-positive.[9] There are several assays that clinicians are currently using in their practices to assess the molecular profile of a tumor prior to making recommendations regarding adjuvant systemic therapy.

The 21-gene recurrence score (Oncotype DX)

The 21-gene recurrence score (RS) assay predicts the rate of distant recurrence in patients with early-stage, ER-positive, lymph node–negative breast cancer.[10] The 21-gene RS is performed on fixed tissue from a surgical specimen or core biopsy. Patients receive a score ranging from 0 to 100. The scores are divided into three risk groups: low (scores 0–18), intermediate (scores 19–31), and high (scores > 31). A total of 51% of patients studied in National Surgical Adjuvant Breast and Bowel Project (NSABP) B-20 had a low RS.[11] Patients with high scores, likely due to luminal B tumors, are most likely to benefit from adjuvant chemotherapy, have lower ER expression levels, and have higher expression levels of proliferation genes.[11] Patients with low RS did not have improved long-term outcome with chemotherapy. The potential benefit of adjuvant chemotherapy among patients with intermediate recurrence scores is not well defined, and is being evaluated in TAILORx (Trial Assigning IndividuaLized Options for Treatment [Rx]) (see Figure 1A). Patients with intermediate scores have been randomized to chemotherapy followed by hormonal therapy or to hormonal therapy alone. Based upon results from the prospective validation studies from patients enrolled on NSABP B-20, the 21-gene RS has been incorporated into the National Comprehensive Cancer Network (NCCN), American Society of Clinical Oncology (ASCO), and St. Gallen treatment guidelines for early-stage, ER-positive, lymph node–negative breast cancer.

The use of the 21-gene RS in patients has been better studied in those with lymph node–negative disease compared with node-positive disease. Analysis of node-positive patients from the phase III Southwest Oncology Group (SWOG) 8814 clinical trial, in which patients were randomized between chemotherapy followed by tamoxifen vs tamoxifen alone, showed that the RS was prognostic in this patient population. A high RS predicted for chemotherapy benefit in node-positive patients.[12] Although node-positive patients with low RS derived less benefit from chemotherapy than those with high RS, results of the RxPONDER (Rx for Positive Node, Endocrine Responsive Breast Cancer) trial will be needed prior to recommending routine use of the 21-gene RS in the node-positive population (see Figure 1B).

Although the 21-gene RS costs about $4,000 per patient, it has been shown to be cost-effective in multiple studies.[13-15] Several studies have also shown that results from the 21-gene RS have changed the medical oncologist’s treatment recommendation.[16,17] The largest change was typically from pre-test recommendation for adjuvant chemotherapy followed by hormonal therapy to hormonal therapy alone. So, in addition to being cost-effective, the 21-gene RS also reduces the overall morbidity associated with treating early-stage, ER-positive breast cancer, since fewer patients are exposed to the short- and long-term risks of chemotherapy.

The 21-gene RS has been compared with a combined ER, PR, Ki67, and HER2 IHC score (IHC-4) in a cohort of early-stage breast cancer patients from the ATAC (Anastrozole or Tamoxifen Alone in Combination) trial who did not receive chemotherapy. In this analysis, the IHC-4 score provided prognostic information similar to that of the 21-gene RS, with modest correlation between the two.[18] Although the IHC-4 score provided prognostic information in this study, lack of reproducibility of quantitative IHC assays across laboratories has limited its clinical application.

The 70-gene prognosis signature (MammaPrint)

The 70-gene prognosis signature was initially described by van’t Veer et al in 2002 by performing DNA microarray analysis on primary breast tumors. They were able to identify a gene expression signature that predicted for development of distant metastases. The poor-prognosis signature was characterized by expression of genes associated with proliferation, angiogenesis, and invasion.[19] The prognostic significance of the 70-gene signature was validated in a separate cohort of patients. Patients with a good-prognosis signature had significantly lower rates of distant metastasis compared with patients who had a poor-prognosis signature.[20] A total of 36% to 39% of the patients in evaluated studies have had a good-prognosis signature.[20,21] Additional analysis has shown that nearly all basal-like, HER2-enriched, and luminal B tumors have poor-prognosis signatures.[22]

Clinical use of the 70-gene signature has been limited by the requirement, until recently, of frozen tissue, and by limited data validating the predictive benefit of chemotherapy among good- and poor-prognosis signatures. The 70-gene signature has been well validated in prospective studies of lymph node–negative patients who did not receive adjuvant chemotherapy[21,23-26]; however, results from ongoing clinical trials are needed for prospective validation of predictive benefit from adjuvant chemotherapy. In the MINDACT (Microarray In Node-negative and 1 to 3 positive lymph node Disease may Avoid ChemoTherapy) trial, patients with ER-positive, early-stage breast cancer (node-negative or 1 to 3 positive lymph nodes) receive recommendations for adjuvant chemotherapy based upon the 70-gene signature and an online prognostic tool using clinical and pathologic features (Adjuvant! Online). Patients who are determined to be high risk by both the online assessment and the 70-gene signature will receive adjuvant chemotherapy; those who are good risk by both will receive hormonal therapy alone. Discordant cases will be randomized to adjuvant therapy based upon either the 70-gene signature or the online tool. Recruitment to this multi-institutional randomized phase III trial has been completed (see Figure 1C).

In the past year, it has become possible to perform the 70-gene signature on fixed tissue; this should facilitate using the 70-gene signature in clinical practice, where frozen tissue is not routinely collected. The 70-gene signature has also identified a subgroup of HER2-positive patients with a good prognosis. These tumors were characterized by being ER-positive and low-risk for relapse in absence of adjuvant chemotherapy.[27] The current standard of care for HER2-positive, early-stage breast cancer is to receive adjuvant trastuzumab (Herceptin)-based chemotherapy; however, these results suggest that there may be a subgroup of good-risk HER2-positive patients for whom chemotherapy could be avoided.

Additional prognostic panels

In addition to the 21-gene RS and 70-gene signature, several prognostic predictors have been developed and are commercially available, and others are in development. The Predictor Analysis of Microarray (PAM) 50-gene test has been developed to classify breast cancers into intrinsic subtypes. The PAM-50 assay provides a risk of relapse score and is commercially available.[9,28] A genomic index of sensitivity to endocrine therapy (SET) has also been developed by measuring the level of transcriptional activity related to ER. A high SET index was predictive of lower risk of distant relapse with adjuvant tamoxifen.[29] The genomic grade index (GGI) is a 97-gene measure of histologic grade, and a high GGI is associated with a lower relapse-free survival. High GGI also predicted for increased response to neoadjuvant chemotherapy, and predicted for poor prognosis among ER-positive patients, even in the setting of chemotherapy and endocrine therapy.[30]

The Breast Cancer Index (BCI) is an assay comprising two independently developed biomarkers: a set of five cell-cycle–related genes called the molecular grade index[31] and a two-gene expression ratio of homeobox 13 and interleukin-17B receptor which has been shown to predict recurrence and survival in women receiving adjuvant tamoxifen.[32] The BCI stratifies patients into three risk groups that predict risk of distant recurrence. In a recently presented analysis of a cohort of patients from the ATAC trial, prognostic performance of BCI, the 21-gene recurrence score, and IHC-4 were compared with a clinical treatment score based on size of tumor, nodal status, grade, age, and treatment. All three profiles performed well in predicting recurrence in years 1 through 5; however, only the BCI predicted for late distant recurrence in years 5 through 10 after diagnosis.[33]

Adjuvant Treatment Options

Luminal subtypes

The luminal A and B subtypes are both characterized by HR expression, and 5 years of adjuvant anti-estrogen therapy became the standard of care based upon results from multiple trials.[34] The addition of aromatase inhibitors in the adjuvant setting for postmenopausal women has improved disease-free survival compared with tamoxifen alone. Aromatase inhibitors can be used as upfront continuous treatment for 5 years,[35,36] as sequential therapy after 2 to 3 years of tamoxifen,[37,38] or as extended adjuvant therapy after 5 years of tamoxifen.[39]

Patients with HR-positive breast cancer continue to have relapse rates of 1% to 4% per year between 5 and 15 years from diagnosis, and the optimal duration of adjuvant hormonal therapy remains an important clinical question.[40,41] Long-term results of the
ATLAS trial (Adjuvant Tamoxifen: Longer Against Shorter) were recently presented, indicating that 10 years of adjuvant tamoxifen resulted in a further reduction in recurrence and mortality compared with 5 years of adjuvant tamoxifen, with continued benefit seen beyond 10 years of therapy.[42] These results are most relevant for premenopausal patients, for whom extended adjuvant therapy with an aromatase inhibitor is not an alternative option. Molecular profiling will likely be important in determining which patients are at highest risk of late recurrence, and potentially derive benefit from extended adjuvant hormonal therapy. Patients with luminal B tumors, whose risk of recurrence is greatest in the first 5 years, may not benefit from hormonal therapy beyond 5 years.

Despite the marked success of endocrine agents in the treatment of early-stage, HR-positive breast cancers, many patients will relapse. These tumors have either intrinsic or acquired resistance to anti-estrogen therapy. The mechanisms underlying intrinsic and acquired resistance to endocrine agents are likely similar, and include activation of upstream and downstream pathways resulting in changes in co-regulators of the estrogen receptor.

The 21-gene RS can be used to determine the benefit of tamoxifen in node-negative, ER-positive breast cancers. Breast cancers with recurrence scores greater than 31 appear to derive little benefit from adjuvant tamoxifen compared with cancers that have recurrence scores of 30 or less.[10] Concordance between luminal B and high-recurrence-score cancers has also been shown, suggesting that the poor prognosis seen in these cancers may be due in part to intrinsic resistance to endocrine therapy.[22]

A better understanding of the differential expression of genes and proteins in luminal A and B cancers could shed significant light on the mechanisms underlying resistance to endocrine agents, which could in turn lead to novel therapeutic approaches to circumvent this resistance. There is increasing evidence to suggest that breast cancers that express both HRs and HER2 are somewhat intrinsically resistant to endocrine agents, and that these cancers are, in fact, driven by the HER2 pathway.[43,44] Support for this hypothesis comes from data on patients with metastatic HR-positive, HER2-positive breast cancers, in whom progression-free survival following treatment with single-agent anastrozole is extremely short at just over 2 months.[45] However, there is some evidence to suggest the existence of a subset of HER2-positive cancers that express ER and PR, which may be driven more by ER than HER2.[46] In fact, a subset of HER2-positive breast cancers that are ER-positive have been shown to have a good-prognosis signature based on assessment with the 70-gene signature.[27]

Other growth factor pathways, including the epidermal growth factor receptor (EGFR), insulin growth factor receptor, and vascular endothelial growth factor (VEGF) receptor, have been demonstrated to play a role in resistance to endocrine agents.[44,47-49] Other agents such as the mammalian target of rapamycin (mTOR) inhibitor everolimus (Afinitor) may also play a role, based upon results showing improved response to everolimus in combination with endocrine therapy in the metastatic and neoadjuvant setting.[50-52] Going forward, it is essential that we identify novel therapies for patients with luminal B cancers, given their poor survival when treated using conventional therapies. The use of gene expression profiling can help to identify key genes that can then be exploited therapeutically.

Basal-like subtype

When patients were stratified by breast tumor subtype and analyzed for time to distant metastasis and overall survival, those with the basal subtype had the worst clinical outcome.[3] This likely reflects both the aggressive nature of basal-subtype breast tumors and the lack of targeted therapies, since these tumors do not express the ER and do not overexpress HER2. Conventional anthracycline- and taxane-based regimens are currently used to treat patients with the basal-like subtype of breast cancer.

Although women who carry BRCA1 mutations are predisposed to developing breast cancers of the basal-like subtype, expression levels of BRCA1 have not been well characterized in sporadic triple-negative tumors.[3,53] BRCA1 mediates the cellular response to DNA damage by sensing damage, preventing apoptosis, and participating in DNA repair.[54,55] The loss of BRCA1 expression in basal-like tumors may lead to selective sensitivity to DNA cross-linking chemotherapeutic agents, such as platinum analogues.[55] Cisplatin, a platinum analogue, has demonstrated single-agent activity as neoadjuvant treatment for triple-negative breast cancers.[5] There are multiple ongoing clinical trials investigating the addition of cisplatin or carboplatin to neoadjuvant chemotherapy; however, most of these have enrolled patients with triple-negative breast cancer, not just the basal subtype. Genomic profiling could potentially aid in identifying tumors among BRCA-negative patients with a “BRCA-like” profile for whom platinums or other agents targeting DNA repair, such as poly(ADP-ribose) polymerase (PARP) inhibitors, may be more effective.

Other potential targets for basal-like breast cancer have been identified using genomic profiling. The EGFR is part of the basal cluster, and EGFR-targeting agents have been investigated in the metastatic setting, demonstrating modest clinical activity.[56] In a study combining carboplatin and cetuximab (Erbitux), a monoclonal antibody against EGFR, clinical benefit was seen among patients with EGFR pathway inactivation. Anti-angiogenic agents targeting VEGF have also shown promise in the metastatic setting, and studies in the adjuvant setting are ongoing. Bevacizumab (Avastin), a monoclonal antibody against VEGF, showed improved disease-free survival in the first-line[57,58] and second-line setting,[59] but it did not show an overall survival advantage. Benefit from the addition of bevacizumab was similar among patients with HR-positive vs HR-negative breast cancer, suggesting that intrinsic subtyping might not predict anti-angiogenic benefit.

Additional gene expression analysis of triple-negative breast cancers from multiple data sets has further defined this group of cancers. Cluster analysis identified a second “basal-like” subtype in addition to immunomodulatory, mesenchymal, mesenchymal stem-like, and luminal androgen receptor subtype. Triple-negative breast cancer cell lines corresponding to these subtypes responded differently to therapies such as cisplatin, mTOR inhibitors, Src inhibitors, and an androgen receptor antagonist.[60] This would suggest that gene expression profiling of triple-negative breast cancers should play an important role in future trial design of novel, targeted therapies.

HER2-enriched subtype

The HER2-enriched subtype is characterized by high expression of HER2, most commonly due to amplification of the HER2 gene. Genes such as GRB7 and TOP2A, which are located in close proximity to the HER2 gene on chromosome 17, are often co-amplified.[61] Multiple studies have been performed to correlate TOP2A gene status, topo2a expression levels, and response to anthracyclines.[62-67] The role of TOP2A amplification was examined in the Breast Cancer International Research Group (BCIRG) 006 trial in which early-stage, HER2-positive patients were randomized between three arms: standard anthracycline- and taxane-based chemotherapy with or without trastuzumab, and a third non–anthracycline-containing regimen of docetaxel, carboplatin, and trastuzumab. Patients without co-amplification derived greater benefit from the addition of trastuzumab. In patients with co-amplification of TOP2A and HER2, minimal incremental benefit was seen with the addition of trastuzumab; however, the long-term toxicity profile favored the non–anthracycline-containing regimen.[66]

Conclusion

For ER-positive, early-stage breast cancer, genomic profiling using the 21-gene RS and the 70-gene signature has already impacted clinical decision-making. These tests have aided treating oncologists by differentiating patients with low- vs high-risk ER-positive tumors for whom chemotherapy is indicated. Ongoing clinical trials such as MINDACT and TAILORx have focused primarily on how to best apply these tests in the adjuvant setting, using our current standard treatments. Additional insight into the biology of ER-positive breast cancers, particularly the higher risk luminal B cancers, may also be gained from genomic profiling, and potentially could aid in identifying potential targets and new, effective therapies.

Genomic profiling of triple-negative breast cancers has revealed that this is a heterogeneous group of cancers. Although the majority of triple-negative breast cancers are the “basal-like” subtype, significant proportions are in other subtypes. Incorporation of genomic profiling into future clinical trials will have implications for drug development, where the ability to identify aberrant gene expression will help to inform one’s choice of targeted therapies. Ultimately it is hoped that the ability to better define an individual patient’s breast cancer biology will lead to improvements in therapy selection, discovery of new drug targets, and better long-term outcomes for patients.

Financial Disclosure:The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.

References:

1. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2000;406:747-52.

2. Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A. 2001;98:10869-74.

3. Sorlie T, Tibshirani R, Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A. 2003;100:8418-23.

4. Sotiriou C, Neo SY, McShane LM, et al. Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci U S A. 2003;100:10393-8.

5. Herschkowitz JI, Simin K, Weigman VJ, et al. Identification of conserved gene expression features between murine mammary carcinoma models and human breast tumors. Genome Biol. 2007;8:R76.

6. Cheang MC, Voduc D, Bajdik C, et al. Basal-like breast cancer defined by five biomarkers has superior prognostic value than triple-negative phenotype. Clin Cancer Res. 2008;14:1368-76.

7. Carey LA, Perou CM, Livasy CA, et al. Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA. 2006;295:2492-502.

8. Yamamoto Y, Ibusuki M, Nakano M, et al. Clinical significance of basal-like subtype in triple-negative breast cancer. Breast Cancer. 2009;16:260-7.

9. Parker JS, Mullins M, Cheang MC, et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol. 2009;27:1160-7.

10. Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med. 2004;351:2817-26.

11. Paik S, Tang G, Shak S, et al. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer. J Clin Oncol. 2006;24:3726-34.

12. Albain KS, Barlow WE, Shak S, et al. Prognostic and predictive value of the 21-gene recurrence score assay in postmenopausal women with node-positive, oestrogen-receptor-positive breast cancer on chemotherapy: a retrospective analysis of a randomised trial. Lancet Oncol. 2010;11:55-65.

13. Kondo M, Hoshi SL, Ishiguro H, et al. Economic evaluation of 21-gene reverse transcriptase-polymerase chain reaction assay in lymph-node-negative, estrogen-receptor-positive, early-stage breast cancer in Japan. Breast Cancer Res Treat. 2008;112:175-87.

14. Lyman GH, Cosler LE, Kuderer NM, Hornberger J. Impact of a 21-gene RT-PCR assay on treatment decisions in early-stage breast cancer: an economic analysis based on prognostic and predictive validation studies. Cancer. 2007;109:1011-8.

15. Tsoi DT, Inoue M, Kelly CM, et al. Cost-effectiveness analysis of recurrence score-guided treatment using a 21-gene assay in early breast cancer. Oncologist. 2010;15:457-65.

16. Lo SS, Mumby PB, Norton J, et al. Prospective multicenter study of the impact of the 21-gene recurrence score assay on medical oncologist and patient adjuvant breast cancer treatment selection. J Clin Oncol. 2010;28:1671-6.

17. Eiermann W, Rezai M, Kummel S, et al. The 21-gene recurrence score assay impacts adjuvant therapy recommendations for ER-positive, node-negative and node-positive early breast cancer resulting in a risk-adapted change in chemotherapy use. Ann Oncol. 2012 Nov 7. [Epub ahead of print]

18. Cuzick J, Dowsett M, Pineda S, et al. Prognostic value of a combined estrogen receptor, progesterone receptor, Ki-67, and human epidermal growth factor receptor 2 immunohistochemical score and comparison with the Genomic Health recurrence score in early breast cancer. J Clin Oncol. 2011;29:4273-8.

19. van't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature. 2002;415:530-6.

20. van de Vijver MJ, He YD, van't Veer LJ, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med. 2002;347:1999-2009.

21. Buyse M, Loi S, van't Veer L, et al. Validation and clinical utility of a 70-gene prognostic signature for women with node-negative breast cancer. J Natl Cancer Inst. 2006;98:1183-92.

22. Fan C, Oh DS, Wessels L, et al. Concordance among gene-expression-based predictors for breast cancer. N Engl J Med. 2006;355:560-9.

23. Bueno-de-Mesquita JM, van Harten WH, Retel VP, et al. Use of 70-gene signature to predict prognosis of patients with node-negative breast cancer: a prospective community-based feasibility study (RASTER). Lancet Oncol. 2007;8:1079-87.

24. Bueno-de-Mesquita JM, Linn SC, Keijzer R, et al. Validation of 70-gene prognosis signature in node-negative breast cancer. Breast Cancer Res Treat. 2009;117:483-95.

25. Wittner BS, Sgroi DC, Ryan PD, et al. Analysis of the MammaPrint breast cancer assay in a predominantly postmenopausal cohort. Clin Cancer Res. 2008;14:2988-93.

26. Mook S, Schmidt MK, Viale G, et al. The 70-gene prognosis-signature predicts disease outcome in breast cancer patients with 1-3 positive lymph nodes in an independent validation study. Breast Cancer Res Treat. 2009;116:295-302.

27. Knauer M, Cardoso F, Wesseling J, et al. Identification of a low-risk subgroup of HER-2-positive breast cancer by the 70-gene prognosis signature. Br J Cancer. 2010;103:1788-93.

28. Bastien RR, Rodriguez-Lescure A, Ebbert MT, et al. PAM50 breast cancer subtyping by RT-qPCR and concordance with standard clinical molecular markers. BMC Med Genomics. 2012;5:44.

29. Symmans WF, Hatzis C, Sotiriou C, et al. Genomic index of sensitivity to endocrine therapy for breast cancer. J Clin Oncol. 2010;28:4111-9.

30. Liedtke C, Hatzis C, Symmans WF, et al. Genomic grade index is associated with response to chemotherapy in patients with breast cancer. J Clin Oncol. 2009;27:3185-91.

31. Ma XJ, Salunga R, Dahiya S, et al. A five-gene molecular grade index and HOXB13:IL17BR are complementary prognostic factors in early stage breast cancer. Clin Cancer Res. 2008;14:2601-8.

32. Goetz MP, Suman VJ, Ingle JN, et al. A two-gene expression ratio of homeobox 13 and interleukin-17B receptor for prediction of recurrence and survival in women receiving adjuvant tamoxifen. Clin Cancer Res. 2006;12(7 Pt 1):2080-7.

33. Sgroi D, Sestak I, Zhang Y, et al. Comparative performance of Breast Cancer Index (BCI) vs. OncotypeDx and IHC4 in the prediction of late recurrence in HR-positive, LN-negative breast cancer patients: a TransATAC study. San Antonio Breast Cancer Symposium. 2012:Abstr S1-9.

34. Davies C, Godwin J, Gray R, et al. Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of randomised trials. Lancet. 2011;378:771-84.

35. Cuzick J, Sestak I, Baum M, et al. Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 10-year analysis of the ATAC trial. Lancet Oncol. 2010;11:1135-41.

36. Regan MM, Neven P, Giobbie-Hurder A, et al. Assessment of letrozole and tamoxifen alone and in sequence for postmenopausal women with steroid hormone receptor-positive breast cancer: the BIG 1-98 randomised clinical trial at 8.1 years median follow-up. Lancet Oncol. 2011;12:1101-8.

37. Dubsky PC, Jakesz R, Mlineritsch B, et al. Tamoxifen and anastrozole as a sequencing strategy: a randomized controlled trial in postmenopausal patients with endocrine-responsive early breast cancer from the Austrian Breast and Colorectal Cancer Study Group. J Clin Oncol. 2012;30:722-8.

38. Bliss JM, Kilburn LS, Coleman RE, et al. Disease-related outcomes with long-term follow-up: an updated analysis of the intergroup exemestane study. J Clin Oncol. 2012;30:709-17.

39. Jin H, Tu D, Zhao N, et al. Longer-term outcomes of letrozole versus placebo after 5 years of tamoxifen in the NCIC CTG MA.17 trial: analyses adjusting for treatment crossover. J Clin Oncol. 2012;30:718-21.

40. Early Breast Cancer Trialists Collaborative Group. Tamoxifen for early breast cancer. Cochrane Database Syst Rev. 2001;(1):CD000486.

41. Saphner T, Tormey DC, Gray R. Annual hazard rates of recurrence for breast cancer after primary therapy. J Clin Oncol. 1996;14:2738-46.

42. Davies C, Pan H, Godwin J, et al. Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial. Lancet. 2012 Dec 4. [Epub ahead of print]

43. Benz CC, Scott GK, Sarup JC, et al. Estrogen-dependent, tamoxifen-resistant tumorigenic growth of MCF-7 cells transfected with HER2/neu. Breast Cancer Res Treat. 1992;24:85-95.

44. De Laurentiis M, Arpino G, Massarelli E, et al. A meta-analysis on the interaction between HER-2 expression and response to endocrine treatment in advanced breast cancer. Clin Cancer Res. 2005;11:4741-8.

45. Mackey JR, Kaufman B, Clemens M, et al. Trastuzumab prolongs progression-free survival in hormone-dependent and HER2-positive metastatic breast cancer. San Antonio Breast Cancer Symposium; 2006:Abstr 2.

46. Nahta R, O'Regan RM. Therapeutic implications of estrogen receptor signaling in HER2-positive breast cancers. Breast Cancer Res Treat. 2012;135:39-48.

47. Cristofanilli M, Valero V, Mangalik A, et al. Phase II, randomized trial to compare anastrozole combined with gefitinib or placebo in postmenopausal women with hormone receptor-positive metastatic breast cancer. Clin Cancer Res. 2010;16:1904-14.

48. Traina TA, Rugo HS, Caravelli JF, et al. Feasibility trial of letrozole in combination with bevacizumab in patients with metastatic breast cancer. J Clin Oncol. 2010;28:628-33.

49. Kaklamani VG, Cianfrocca M, Ciccone J, et al. Increased HER2/neu expression in recurrent hormone receptor-positive breast cancer. Biomarkers. 2010;15:191-3.

50. Bachelot T, Bourgier C, Cropet C, et al. Randomized phase II trial of everolimus in combination with tamoxifen in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: a GINECO study. J Clin Oncol. 2012;30:2718-24.

51. Baselga J, Campone M, Piccart M, et al. Everolimus in postmenopausal hormone-receptor-positive advanced breast cancer. N Engl J Med. 2012;
366:520-9.

52. Baselga J, Semiglazov V, van Dam P, et al. Phase II randomized study of neoadjuvant everolimus plus letrozole compared with placebo plus letrozole in patients with estrogen receptor-positive breast cancer. J Clin Oncol. 2009;27:2630-7.

53. Turner N, Tutt A, Ashworth A. Hallmarks of 'BRCAness' in sporadic cancers. Nat Rev Cancer. 2004;4:814-9.

54. MacLachlan TK, Takimoto R, El-Deiry WS. BRCA1 directs a selective p53-dependent transcriptional response towards growth arrest and DNA repair targets. Mol Cell Biol. 2002;22:4280-92.

55. Quinn JE, Kennedy RD, Mullan PB, et al. BRCA1 functions as a differential modulator of chemotherapy-induced apoptosis. Cancer Res. 2003;63:6221-8.

56. Carey LA, Rugo HS, Marcom PK, et al. TBCRC 001: randomized phase II study of cetuximab in combination with carboplatin in stage IV triple-negative breast cancer. J Clin Oncol. 2012;30:2615-23.

57. Miller K, Wang M, Gralow J, et al. Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. New Engl J Med. 2007;357:2666-76.

58. Robert NJ, Dieras V, Glaspy J, et al. RIBBON-1: randomized, double-blind, placebo-controlled, phase III trial of chemotherapy with or without bevacizumab for first-line treatment of human epidermal growth factor receptor 2-negative, locally recurrent or metastatic breast cancer. J Clin Oncol. 2011;29:1252-60.

59. Brufsky AM, Hurvitz S, Perez E, et al. RIBBON-2: a randomized, double-blind, placebo-controlled, phase III trial evaluating the efficacy and safety of bevacizumab in combination with chemotherapy for second-line treatment of human epidermal growth factor receptor 2-negative metastatic breast cancer. J Clin Oncol. 2011;29:4286-93.

60. Lehmann BD, Bauer JA, Chen X, et al. Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Invest. 2011;121:2750-67.

61. Slamon DJ, Press MF. Alterations in the TOP2A and HER2 genes: association with adjuvant anthracycline sensitivity in human breast cancers. J Natl Cancer Inst. 2009;101:615-8.

62. Arriola E, Moreno A, Varela M, et al. Predictive value of HER-2 and topoisomerase II alpha in response to primary doxorubicin in breast cancer. Eur J Cancer. 2006;42:2954-60.

63. Pritchard KI. Are HER2 and TOP2A useful as prognostic or predictive biomarkers for anthracycline-based adjuvant chemotherapy for breast cancer?
J Clin Oncol. 2009;27:3875-6.

64. Gianni L, Valagussa P. Anthracyclines and early breast cancer: the end of an era? J Clin Oncol. 2009;27:1155-7.

65. Buzdar AU. Topoisomerase II alpha gene amplification and response to anthracycline-containing adjuvant chemotherapy in breast cancer. J Clin Oncol. 2006;24:2409-11.

66. Slamon D, Eiermann W, Robert N, et al. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365:1273-83.

67. Martin M, Romero A, Cheang MC, et al. Genomic predictors of response to doxorubicin versus docetaxel in primary breast cancer. Breast Cancer Res Treat. 2011;128:127-36.