Elsevier

The Lancet

Volume 378, Issue 9805, 19–25 November 2011, Pages 1812-1823
The Lancet

Series
Gene expression profiling in breast cancer: classification, prognostication, and prediction

https://doi.org/10.1016/S0140-6736(11)61539-0Get rights and content

Summary

Microarray-based gene expression profiling has had a major effect on our understanding of breast cancer. Breast cancer is now perceived as a heterogeneous group of different diseases characterised by distinct molecular aberrations, rather than one disease with varying histological features and clinical behaviour. Gene expression profiling studies have shown that oestrogen-receptor (ER)-positive and ER-negative breast cancers are distinct diseases at the transcriptomic level, that additional molecular subtypes might exist within these groups, and that the prognosis of patients with ER-positive disease is largely determined by the expression of proliferation-related genes. On the basis of these principles, a molecular classification system and prognostic multigene classifiers based on microarrays or derivative technologies have been developed and are being tested in randomised clinical trials and incorporated into clinical practice. In this review, we focus on the conceptual effect and potential clinical use of the molecular classification of breast cancer, and discuss prognostic and predictive multigene predictors.

Introduction

Breast cancer was historically perceived as one disease with varying histopathological features and response to systemic treatment. In the 1970s, however, breast cancer began to be divided into two disease subsets on the basis of oestrogen receptor (ER) expression, in view of the distinct clinical characteristics these subgroups display. Treatment decisions were solely based on clinicopathological variables that are prognostic in nature, such as tumour size, presence of lymph-node metastasis, and histological grade, and three predictive markers of response to endocrine therapy and trastuzumab—namely, ER and progesterone receptor (PR) expression for endocrine therapy, and HER2 (also known as ERBB2) for trastuzumab. These factors are combined in the form of algorithms for treatment decision making, such as Adjuvant! Online1 and the Nottingham prognostic index, and form the basis of treatment guidelines, including the National Comprehensive Cancer Network,2 the US National Cancer Institute (NCI), and St Gallen's consensus statements.3, 4 Although these approaches have been successful, in view of the steady reduction in breast cancer mortality during the past three decades, they are not sufficient for implementation of individualised therapy. In fact, with these approaches, about 60% of all patients with early-stage breast cancer still receive adjuvant chemotherapy, of which only a small proportion, 2–15% of patients,5 will ultimately derive benefit, while all remain at risk of toxic side-effects.

The advent of high-throughput platforms for analysis of gene expression, such as microarrays, has led to studies that have challenged the view that breast cancer is a single disease with variations in clinical behaviour and histopathological features.6, 7, 8 Microarray-based gene expression profiling studies have brought to the fore the concept that breast cancer consists of a collection of different diseases that affect the same organ site and originate from the same anatomical structure (ie, the terminal duct lobular unit), but have different risk factors, clinical presentation, histopathological features, outcome, and response to systemic therapies.9, 10, 11, 12 These studies also showed that response to treatment is not determined by anatomical prognostic factors (ie, tumour size or nodal status), but rather by intrinsic molecular characteristics of the tumours that can be probed with molecular methods.9, 10, 11, 12

Key messages

  • Gene expression analysis has changed the way breast cancer is perceived in that it is no longer regarded as a single disease

  • Oestrogen-receptor (ER)-positive and ER-negative cancers represent molecularly and clinically distinct diseases

  • Gene expression profiling has shown the importance of proliferation as a prognostic factor for ER-positive cancers

  • First generation prognostic gene signatures provide information that is complementary to that provided by anatomical prognostic variables such as tumour size and nodal status

  • Current first generation prognostic signatures are clinically useful in patients with ER-positive disease, but of limited clinical value for patients with ER-negative disease

  • The theoretical, experimental, and statistical knowledge acquired from microarray-based gene expression profiling studies will help in the development of the next generation of genomic predictors

The conceptual changes resulting from gene expression profiling studies of breast cancer have led to a new paradigm in the way breast cancer is perceived,9, 10, 11, 12 and, importantly, have provided a rationale for a change in the way clinical trials are undertaken and patients are stratified for treatment decision making.13 Furthermore, some tests that have emerged from these analyses provide an incremental increase in the reproducibility and accuracy of the assessment of variables that are crucial to determine prognosis of disease and tailor therapy.

In this review, we assess the conceptual and practical contribution of gene expression profiling in breast cancer, with special emphasis on assays currently used in clinical practice or under evaluation in the context of randomised phase 3 clinical trials. Additionally, we provide a critical assessment of the limitations of this approach and potential ways forward.

Section snippets

Intrinsic molecular classification

One of the first applications of microarray-based gene expression profiling analysis to the study of breast cancer was in assessment of the diversity of the disease at the molecular level.6, 7, 8 The seminal class-discovery studies (figure 1) undertaken by Perou and colleagues7 and Sorlie and co-workers8 revealed that ER-positive and ER-negative breast cancers are fundamentally distinct diseases in molecular terms. Furthermore, hierarchical cluster analysis of genes that vary more between

Development of microarray-based prognostic signatures

Concurrent with the class-discovery studies that have revealed the heterogeneity of breast cancers, microarray-based gene expression profiling was also used for forecasting of outcomes for individual patients with breast cancer, specifically aiming to identify patients with disease of sufficiently good prognosis to allow the safe omission of adjuvant chemotherapy.10, 11 These supervised class-prediction studies (figure 1), however, did not take into account the molecular heterogeneity of the

Second generation prognostic signatures

The realisation that ER-positive and HER2-negative, HER2-positive, and ER-negative and HER2-negative breast cancers are fundamentally different in terms of their mRNA profiles has led several groups to investigate prognostic signatures for these subgroups separately (figure 3). Analyses of breast tumours from large cohorts stratified in this manner revealed that although many genes, most of which are related to the cell cycle and proliferation, predict the outcome of ER-positive disease, the

Predictive signatures

One of the applications of microarray-based gene expression profiling envisaged at the turn of the 21st century was for the development of multigene classifiers that could predict response to endocrine treatment67, 68, 69 and chemotherapy.70, 71, 72, 73, 74, 75, 76, 77 Since patients assigned as having poor prognosis disease by first generation signatures have tumours with high expression of proliferation-related genes, and conventional chemotherapy agents target the proliferating fraction of

Conclusions

Microarray studies undertaken in the past decade have led to a change in the way breast cancer is perceived and brought forward the concept that breast cancer consists of a collection of different diseases, which need distinct therapeutic approaches. Despite this conceptual advance, we note that at present the clinical value of the intrinsic taxonomy7, 8, 14, 15 above and beyond the subtyping of breast cancers based on ER, PR, HER2, and Ki67 remains unclear.4, 26, 27 First generation prognostic

Search strategy and selection criteria

We undertook a PubMed search with the term “breast cancer” and one of the terms “molecular subtypes”, “intrinsic gene”, “gene expression profiling”, “prognostic signature”, “multigene predictor”, and “predictive signature”. Peer-reviewed articles published between Jan 1, 1995, and June 1, 2011, were included, which resulted in 3295 articles. From this list we only reviewed articles covering the intrinsic molecular classification that were based on microarray-based gene expression profiling, and

References (88)

  • H Bonnefoi et al.

    Retraction—validation of gene signatures that predict the response of breast cancer to neoadjuvant chemotherapy: a substudy of the EORTC 10994/BIG 00-01 clinical trial

    Lancet Oncol

    (2011)
  • PM Ravdin et al.

    Computer program to assist in making decisions about adjuvant therapy for women with early breast cancer

    J Clin Oncol

    (2001)
  • RW Carlson et al.

    Invasive breast cancer

    J Natl Compr Canc Netw

    (2007)
  • Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials

    Lancet

    (2005)
  • S Gruvberger et al.

    Estrogen receptor status in breast cancer is associated with remarkably distinct gene expression patterns

    Cancer Res

    (2001)
  • CM Perou et al.

    Molecular portraits of human breast tumours

    Nature

    (2000)
  • T Sorlie et al.

    Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications

    Proc Natl Acad Sci USA

    (2001)
  • JS Reis-Filho et al.

    Molecular profiling: moving away from tumor philately

    Sci Transl Med

    (2010)
  • C Sotiriou et al.

    Gene-expression signatures in breast cancer

    N Engl J Med

    (2009)
  • B Weigelt et al.

    The contribution of gene expression profiling to breast cancer classification, prognostication and prediction: a retrospective of the last decade

    J Pathol

    (2010)
  • T Iwamoto et al.

    Predicting prognosis of breast cancer with gene signatures: are we lost in a sea of data?

    Genome Med

    (2010)
  • L Pusztai et al.

    Effect of molecular disease subsets on disease-free survival in randomized adjuvant chemotherapy trials for estrogen receptor-positive breast cancer

    J Clin Oncol

    (2008)
  • Z Hu et al.

    The molecular portraits of breast tumors are conserved across microarray platforms

    BMC Genomics

    (2006)
  • JS Parker et al.

    Supervised risk predictor of breast cancer based on intrinsic subtypes

    J Clin Oncol

    (2009)
  • T Sorlie et al.

    Repeated observation of breast tumor subtypes in independent gene expression data sets

    Proc Natl Acad Sci USA

    (2003)
  • LJ van 't Veer et al.

    Gene expression profiling predicts clinical outcome of breast cancer

    Nature

    (2002)
  • C Sotiriou et al.

    Gene expression profiling in breast cancer: understanding the molecular basis of histologic grade to improve prognosis

    J Natl Cancer Inst

    (2006)
  • S Paik et al.

    A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer

    N Engl J Med

    (2004)
  • A Prat et al.

    Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast cancer

    Breast Cancer Res

    (2010)
  • P Farmer et al.

    Identification of molecular apocrine breast tumours by microarray analysis

    Oncogene

    (2005)
  • AS Doane et al.

    An estrogen receptor-negative breast cancer subset characterized by a hormonally regulated transcriptional program and response to androgen

    Oncogene

    (2006)
  • A Mackay et al.

    Microarray-based class discovery for molecular classification of breast cancer: analysis of interobserver agreement

    J Natl Cancer Inst

    (2011)
  • B Weigelt et al.

    Challenges translating breast cancer gene signatures into the clinic

    Nat Rev Clin Oncol

    (2011)
  • B Weigelt et al.

    Molecular profiling currently offers no more than tumour morphology and basic immunohistochemistry

    Breast Cancer Res

    (2010)
  • C Kim et al.

    Gene-expression-based prognostic assays for breast cancer

    Nat Rev Clin Oncol

    (2010)
  • MJ van de Vijver et al.

    A gene-expression signature as a predictor of survival in breast cancer

    N Engl J Med

    (2002)
  • XJ Ma et al.

    A five-gene molecular grade index and HOXB13:IL17BR are complementary prognostic factors in early stage breast cancer

    Clin Cancer Res

    (2008)
  • R Liu et al.

    The prognostic role of a gene signature from tumorigenic breast-cancer cells

    N Engl J Med

    (2007)
  • J Toussaint et al.

    Improvement of the clinical applicability of the Genomic Grade Index through a qRT-PCR test performed on frozen and formalin-fixed paraffin-embedded tissues

    BMC Genomics

    (2009)
  • S Paik et al.

    Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer

    J Clin Oncol

    (2006)
  • S Mook 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)
  • M Knauer et al.

    Identification of a low-risk subgroup of HER-2-positive breast cancer by the 70-gene prognosis signature

    Br J Cancer

    (2010)
  • C Desmedt et al.

    The Gene expression Grade Index: a potential predictor of relapse for endocrine-treated breast cancer patients in the BIG 1-98 trial

    BMC Med Genomics

    (2009)
  • M Dowsett et al.

    Prediction of risk of distant recurrence using the 21-gene recurrence score in node-negative and node-positive postmenopausal patients with breast cancer treated with anastrozole or tamoxifen: a TransATAC study

    J Clin Oncol

    (2010)
  • Cited by (0)

    View full text