Cancer Letters

Cancer Letters

Volume 355, Issue 1, 1 December 2014, Pages 70-75
Cancer Letters

Original Articles
EndoPredict predicts for the response to neoadjuvant chemotherapy in ER-positive, HER2-negative breast cancer

https://doi.org/10.1016/j.canlet.2014.09.014Get rights and content

Abstract

The EndoPredict (EP) signature is a prognostic 11-gene expression signature specifically developed in ER+/HER2– node-negative/positive breast cancer. It is associated with relapse-free survival in patients treated with adjuvant hormone therapy, suggesting that EP low-risk patients could be treated with adjuvant hormone therapy alone whereas high-risk patients would deserve addition of adjuvant chemotherapy. Thus, it is important to determine whether EP high-risk patients are or are not more sensitive to chemotherapy than low-risk patients. Here, we have assessed the EP predictive value for pathological complete response to neoadjuvant chemotherapy in ER+/HER2– breast cancer. We gathered gene expression and histoclinical data of 553 pre-treatment ER+/HER2– breast carcinomas treated with anthracycline-based neoadjuvant chemotherapy. We searched for correlation between the pathological complete response (pCR) and the EP score-based classification. The overall pCR rate was 12%. Fifty-one percent of samples were classified as low-risk according to the EP score and 49% as high-risk. EP classification was associated with a pCR rate of 7% in the low-risk group and 17% in the high-risk group (p < 0.001). In multivariate analysis, the EP score remained significantly associated with pCR. Many genes upregulated in the high-risk tumours were involved in cell proliferation, whereas many genes upregulated in the low-risk tumours were involved in ER-signalling and stroma. Despite higher chemosensitivity, the high-risk group was associated with worse disease-free survival. In conclusion, EP high-risk ER+/HER2– breast cancers are more likely to respond to anthracycline-based chemotherapy.

Introduction

During the last decade, the survival of patients with breast cancer improved thanks to mass screening and adjuvant systemic therapy. A crucial issue regarding adjuvant chemotherapy is to improve the prognostic features to avoid overtreatment of patients – notably those without or less than four pathologically involved axillary lymph nodes – who receive post-operative chemotherapy but would be cured by surgery and radiotherapy alone. That is particularly critical for patients with oestrogen receptor-positive (ER+)/HER2– tumour, who represent the largest subset of early breast cancer patients. Indeed, if adjuvant chemotherapy is currently recommended for most women with HER2+ tumours or with triple-negative tumours [1], its indications are much more challenging in patients with ER+/HER2– tumours that are either candidate for adjuvant hormone therapy alone or for both hormone therapy and chemotherapy.

During the last decade, gene expression profiling identified promising prognostic multigene signatures in breast cancer [2]. In the setting of ER+/HER2– tumours, several signatures including Mammaprint [3], Recurrence Score [4], or EndoPredict [5] have been proposed to improve the selection of patients candidate or not to adjuvant chemotherapy [1]. They were developed in heterogeneous cohorts of patients with respect to ER and HER2 status, except EndoPredict (EP), which was specifically developed in ER+/HER2– node-negative/positive tumours [5]. EP is based on mRNA expression of 8 genes of interest and 3 reference genes measured using quantitative RT-PCR on formalin-fixed paraffin-embedded (FFPE) samples. In contrast with Mammaprint and Recurrence Score, EP is suitable for decentralised testing in molecular pathology laboratories instead of a single reference laboratory [6]. EP independent prognostic value was first validated in a series of 1702 independent tumours from two prospective trials of adjuvant hormone therapy. EP segregated patients into two groups according to their distant relapse risk: for example, in the ABCSG-8 prospective trial, the distant recurrence rate was 6% in the low-risk group and 15% in the high-risk group, suggesting that low-risk patients could be treated with adjuvant hormone therapy without any chemotherapy whereas high-risk patients would deserve addition of adjuvant chemotherapy. Thus, it is important to determine whether EP high-risk patients are or are not more sensitive to chemotherapy than low-risk patients. This has already been shown with other signatures such as Recurrence Score [7], Mammaprint [8] and Genomic Grade Index [9]. We thus hypothesised that patients with breast cancer classified as high-risk of relapse according to EP would respond better to chemotherapy.

To prove this issue, we retrospectively analysed the predictive value of EP signature in terms of pathological response to chemotherapy in a large pooled series of breast cancer patients treated with anthracycline-based neo-adjuvant chemotherapy and whose tumour had been previously profiled using DNA microarrays. We show that the EP signature is associated with response to chemotherapy.

Section snippets

Patients' selection

We collected seven retrospective gene expression data sets of breast cancer samples profiled using DNA microarrays (Supplementary Table S1), including our own set [10] and six public sets [11], [12], [13], [14], [15], [16]. Gene expression and histoclinical data were retrieved from the NCBI GEO database and authors' website. Selection criteria of data sets were the presence of at least 10 profiled samples representing pre-treatment invasive breast carcinoma, treated with anthracycline-based

Patients' characteristics

We collected personal and public gene expression and histoclinical data of 553 pre-treatment ER+/HER2– invasive breast carcinomas with available pathological response assessed on the surgical specimen after anthracycline-based neoadjuvant chemotherapy. Their characteristics are summarised in Table 1. Median age of patients was 49 years. Clinically, the tumours were classified T2–T4 in 93%, and the axillary lymph node status was positive in 65% of patients. Most tumours were ductal type and SBR

Discussion

We searched for correlations between the EP classification and the response to chemotherapy in ER+/HER2– breast cancer. Through a retrospective analysis of gene expression data of 553 pre-treatment samples, we show that tumours with an EP high-risk score are associated with higher pCR rate to chemotherapy than tumours with a low-risk score. To our knowledge, this is the first demonstration of correlation between the EP signature and the response to chemotherapy.

The EP signature was developed as

Conflict of interest

The authors declare that they have no conflict of interest.

Acknowledgments

Our work is supported by Institut Paoli-Calmettes, Inserm, Institut National du Cancer (AOPL2010 IVOIRES), Ligue Nationale contre le Cancer (label DB) and SIRIC INCa-DGOS-Inserm 6038.

References (32)

  • R. Kronenwett et al.

    Decentral gene expression analysis: analytical validation of the Endopredict genomic multianalyte breast cancer prognosis test

    BMC Cancer

    (2012)
  • L. Gianni et al.

    Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer

    J. Clin. Oncol

    (2005)
  • M.E. Straver et al.

    The 70-gene signature as a response predictor for neoadjuvant chemotherapy in breast cancer

    Breast Cancer Res. Treat

    (2010)
  • C. Liedtke et al.

    Genomic grade index is associated with response to chemotherapy in patients with breast cancer

    J. Clin. Oncol

    (2009)
  • R. Sabatier et al.

    Down-regulation of ECRG4, a candidate tumor suppressor gene, in human breast cancer

    PLoS ONE

    (2011)
  • C. Hatzis et al.

    A genomic predictor of response and survival following taxane-anthracycline chemotherapy for invasive breast cancer

    JAMA

    (2011)
  • Cited by (40)

    • Tailoring neoadjuvant treatment of HR-positive/HER2-negative breast cancers: Which role for gene expression assays?

      2022, Cancer Treatment Reviews
      Citation Excerpt :

      Of note, this study also tested the correlation between the EP classification with those based on the RS and the MammaPrint scores. Compared with low-risk EP, high-risk tumors were more often categorized as high-risk tumors according to these two gene expression signatures [81], providing an agreement between the results of the different gene expression signatures. Soliman et al. evaluated the ability of the EP test to predict response to NACT, in addition to RS already mentioned, by comparing their predictive power.

    • Gene signatures in the breast cancer

      2022, Revista de Senologia y Patologia Mamaria
    • Gene-expression signatures to inform neoadjuvant treatment decision in HR+/HER2− breast cancer: Available evidence and clinical implications

      2022, Cancer Treatment Reviews
      Citation Excerpt :

      An association between EndoPredict and pCR was first described in 553 ER+/HER2− BCs treated with NCT. EndoPredict signature, both as categorical and continuous variable, was significantly correlated with pCR, with a lower pCR rate in the low-risk group compared to the high-risk group (7% vs 17%, respectively, p < 0.001) [50]. More recently, similar results were observed by Dubsky et al. in 134 HR+ patients treated with NCT in the ABCSG-34 trial.

    • Neoadjuvant chemotherapy in hormone receptor-positive/HER2-negative early breast cancer: When, why and what?

      2021, Critical Reviews in Oncology/Hematology
      Citation Excerpt :

      The Endopredict score was the only independent predictive factor in multivariate analysis (OR = 1.13; 95 % CI: 1.04–1.24). A supervised analysis showed that many genes upregulated in the high-risk group were involved in cell cycle regulation and proliferation, features generally associated with greater chemo-sensitivity (Bertucci et al., 2014). Soliman et al. compared Oncotype DX and Endopredict as predictors of NACT in six microarray data sets including 764 patients with HR+/HER2- tumors (Soliman et al., 2020).

    • The EndoPredict score predicts response to neoadjuvant chemotherapy and neoendocrine therapy in hormone receptor-positive, human epidermal growth factor receptor 2-negative breast cancer patients from the ABCSG-34 trial

      2020, European Journal of Cancer
      Citation Excerpt :

      Recently, the 12-gene MS has also been validated to predict response to adjuvant chemotherapy [14]. A previous publication [19] and a recent modelling study [20] have shown that EndoPredict may also predict response to NaCT, although this has yet to be confirmed. This prospectively defined translational study within the Austrian Breast and Colorectal Cancer Study Group (ABCSG)-34 protocol assessed the ability of the 12-gene MS to predict tumour response in patients with HR-positive, HER2-negative early-stage breast cancer treated with NaCT or NET.

    View all citing articles on Scopus
    View full text