Predicting Endocrine Responsiveness: Novel Biomarkers on the Horizon

February 18, 2009
Kimberly F. Kerstann, PhD

Mark Bouzyk, PhD

Mark Abramovitz, PhD

Brian Leyland-jones, MD, PhD

Oncology, ONCOLOGY Vol 23 No 2, Volume 23, Issue 2

Historically, breast tumor classification and therapeutic decisions have relied on immunohistochemical (IHC) techniques for characterizing biomarkers such as estrogen receptor (ER), progesterone receptor (PR), and the epidermal growth factor receptor 2 (HER2), as described in the review by Ma and colleagues. However, these markers have been found to be inadequate for fully predicting a patient’s response to a given breast cancer treatment such as endocrine therapy.

Historically, breast tumor classification and therapeutic decisions have relied on immunohistochemical (IHC) techniques for characterizing biomarkers such as estrogen receptor (ER), progesterone receptor (PR), and the epidermal growth factor receptor 2 (HER2), as described in the review by Ma and colleagues. However, these markers have been found to be inadequate for fully predicting a patient’s response to a given breast cancer treatment such as endocrine therapy.

Perou and Srlie were two of the first investigators to describe unique gene-expression profiles that characterized breast tumors into distinct subtypes-luminal A, luminal B, and basal-like tumors.[1] These subtypes were found to correlate with clinical outcome and provided an initial glimpse of the molecular drivers potentially responsible for the malignant phenotype of each subtype.[2] Huge efforts are ongoing across the world to determine the molecular drivers for each of these subgroups and to dissect out the molecular signaling pathways responsible. A complete understanding of the key molecular drivers and the appropriate molecular signals should then provide the most comprehensive therapeutic targets and predictive biomarkers.

Available Assays

The review by Ma et al details two recently commercialized multigene panels that have found utility in the clinic-Oncotype DX and MammaPrint. The Oncotype DX assay utilizes RNA prepared from easily obtainable formalin-fixed, paraffin embedded (FFPE) tumor sections. It is a reverse-transcriptase polymerase chain reaction (RT-PCR)–based assay of 16 outcome-related genes and 5 reference genes, initially identified from 250 candidate genes.[3] Oncotype DX is currently included in both the American Society of Clinical Oncology and National Comprehensive Cancer Network clinical guidelines for breast cancer treatment decisions in ER-positive, node-positive and -negative patients. The second multigene assay, MammaPrint, relies on 70-genes selected from microarray data originally validated in only 295 patients.[4] A caveat of the MammaPrint assay is that it requires fresh biopsy tissue that has limited the patient populations in which this test could be validated, and therefore is not as widely used as the Oncotype DX assay.

A recent report by the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group, an initiative established by the Office of Public Health Genomics at the Centers for Disease Control, called into question the clinical validity and utility of both the Oncotype DX and MammaPrint assays based on current data.[5] This issue should be addressed by large clinical trials, which are currently ongoing to validate both assays-Oncotype DX in an 11,000 ER-positive patient Trial Assigning IndividuaLized Options for Treatment (Rx), or TAILORx trial, and MammaPrint in a 6,000 node-negative patient Microarray In Node-Negative Disease May Avoid ChemoTherapy (MINDACT) trial.

A recent complementary review article by Dowsett and Dunbier identified additional multigene panels (not discussed in the review by Ma et al) with either prognostic or predictive value in breast cancer therapy.[6] These are at various stages of validation and utilization in the clinic, and include the breast cancer gene-expression ratio, the genomic grade index, and the Rotterdam signature, which are composed of different multigene panels.[7-12] While there is great potential in the current crop of multigene panels, careful validation is crucial for clinical application.

A Gold Mine of FFPE Samples

Great strides are rapidly being made in both genomic and epigenomic array-based platforms utilizing FFPE tumor samples. These samples represent the largest collection of well-annotated clinical samples readily available for retrospective studies. Indeed, they are a potential gold mine from which important biomarkers are waiting to be discovered once we have the right tools to unlock this invaluable resource. Although considerable RNA degradation occurs as a result of the formalin fixation process, substantial progress has been made for several genomics-based applications that will allow the utilization of this resource in the identification of biomarkers and other genomic data, which, in turn, will improve cancer prognosis, diagnosis, and treatment.

The whole-genome cDNA-mediated Annealing, Selection, extension, and Ligation (WG-DASL) assay, enabling the expression-profiling of over 24,000 protein-coding genes, is now available for use with FFPE specimens. Another recent area of investigation involves micro (mi)RNAs, a distinct class of small noncoding RNA species that control gene expression via translation repression or degradation. The aberrant expression or mutation of these miRNAs has been noted in cancers, suggesting a role for miRNAs in tumorigenesis. Of the 695 currently known human miRNAs,[13] nearly 30 have been implicated in breast cancers, as tumor-suppressor genes or as oncogenes. Moreover, miRNA DASL arrays that enable the interrogation of known miRNA genes are an important tool for analyzing FFPE samples.

Additional Analyses

Ma and coauthors do not discuss the potential utilization of DNA analyses, whether tumor DNA or germline DNA, in the search for predictive or prognostic biomarkers. Genomic instability can give rise to gene amplifications and/or deletions, which is a hallmark of tumor progression and poor prognosis. Comparative genomic hybridization, array-CGH, with ultra–high-density arrays containing 2 million features will be available this year, providing unprecedented resolution across the entire genome. This advance should enable the identification of novel breast cancer–related genes that yield subclass, prognostic, or predictive information.

In addition, single-nucleotide polymorphism (SNP) analysis can now be performed on arrays that assay 1 million SNP loci. Epigenetic modification such as methylation patterns of certain genes may also prove to be important in the development of predictive profiles. Genome-wide–based methylation arrays have become available to interrogate 27,578 CpG loci, covering more than 14,000 genes at single-nucleotide resolution. Ultimately, next-generation sequencing systems may transform our understanding of genome variations, epigenomics, transcriptomics, and the interaction of proteins with DNA and RNA.


As the pursuit of prognostic and predictive biomarkers increases, care must be taken to ensure that patient samples are handled in a consistent manner. It is therefore crucial for clinical trial investigators to insist on and readily implement consistent guidelines for specimen collection, processing, storage, and distribution.[14]

The review by Ma and colleagues provides a synopsis of the current state of knowledge of predictive biomarkers for endocrine responsiveness. While this snapshot is very informative, there are many exciting technologies on the horizon that could lead to the discovery of predictive biomarkers and gene signatures for various treatments in breast cancer. The application of these newly developed technologies will open the door to discovery of a wealth of novel biomarkers as well as more effective predictive panels.

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.


1. Perou CM, Sørlie T, Eisen MB, et al: Molecular portraits of human breast tumours. Nature 406:747-752, 2000.
2. Sørlie 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 98:10869-10874, 2001.
3. 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 351:2817-2826, 2004.
4. 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 347:1999-2009, 2002.
5. Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group: Recommendations for the EGAPP Working Group: Can tumor gene expression profiling improve outcomes in patients with breast cancer? Genet Med 11:66-73, 2009.
6. Dowsett M, Dunbier AK: Emerging biomarkers and new understanding of traditional markers in personalized therapy for breast cancer. Clin Cancer Res 14:8019-8026, 2008.
7. Loi S, Haibe-Kains B, Desmedt C, et al: Definition of clinically distinct molecular subtypes in estrogen receptor-positive breast carcinomas through genomic grade. J Clin Oncol 25:1239-1246, 2007.
8. Ma XJ, Hilsenbeck SG, Wang W, et al: The HOXB13:IL17BR expression index is a prognostic factor in early-stage breast cancer. J Clin Oncol 24:4611-4619, 2006.
9. 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 14:2601-2608, 2008.
10. Sotiriou C, Desmedt C: Gene expression profiling in breast cancer. Ann Oncol 17(suppl 10):x259-x262, 2006.
11. Wang Y, Klijn JG, Zhang Y, et al: Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet 365:671-679, 2005.
12. Wang Z, Dahiya S, Provencher H, et al: The prognostic biomarkers HOXB13, IL17BR, and CHDH are regulated by estrogen in breast cancer. Clin Cancer Res 13:6327-6334, 2007.
13. Griffiths-Jones S, Saini HK, van Dongen S, et al: miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res 34(database issue):D140-D144, 2006.
14. Leyland-Jones BR, Ambrosone CB, Bartlett J, et al: Recommendations for collection and handling of specimens from group breast cancer clinical trials. J Clin Oncol 26:5638-5644, 2008.