Introduction
Breast cancer is a heterogeneous disease that comprises a range of subgroups with diverse clinical behaviors and responses to treatment [
1]. Many breast-cancer-related genes have been investigated to explore the molecular mechanism of carcinogenesis and diverse clinical outcome of the disease [
2]-[
6]. Among them, some specific genes such as estrogen receptor (ER) [
7], progesterone receptor (PR) [
8] and human epidermal growth factor receptor 2 (HER2) [
9] have been used to identify different subgroups and indicate different prognostic results with different treatment modalities in the clinical. Patients with ER-positive status which account for almost 70% of breast cancer always had a better prognosis compared with those ER-negative types [
10]. However, ER-positive patients also have distinct outcomes and almost 20% might relapse within 10-years after surgery [
11]. Thus, there is an urgent need to identify biomarkers that could predict prognostic outcome in patients with ER-positive breast cancer.
MicroRNAs (miRNAs) are short (approximately 22 nucleotides), single-stranded and highly conserved non-coding RNAs which could regulate almost one-third human genome based on either mRNA degradation or translational repression through base pairing with the 3?-untranslated region of target mRNAs at post-transcriptional level [
12],[
13]. Reportedly, miRNAs play important roles in various biological processes, such as cellular development, differentiation, proliferation, angiogenesis and metabolism [
14]-[
17]. The prognostic value of miRNAs has been explored in several cancer types, such as colon cancer [
18], nasopharyngeal carcinoma [
19], hepatocellular carcinoma [
20] and glioma [
21]. To date, no prognostic miRNA signature for ER-positive breast cancer has been reported. In the present study, we used data retrieved from The Cancer Genome Atlas (TCGA,
http://cancergenome.nih.gov/) and identify a miRNA signature associated with the status of ER which could act as a prognostic predicator for ER-positive patients.
Discussion
Breast cancer is the most common malignancy and the second leading cause of cancer death among women worldwide [
26]. Due to the distinct clinical, pathological and molecular features of the disease, the treatment, response to therapy and corresponding clinical outcome varies greatly [
3]. With the help of molecular profiling and the identification of intrinsic subtypes by specific genes, breast cancer patients could benefit from appropriate treatment [
27]. ER status is one of the strong factors in predicting patients- response to endocrine therapy and its determination has become a standard practice in the management of breast cancer [
28]. The level of ER was positively correlated with the sensitivity of the endocrine therapy and could predict tamoxifen resistance in breast cancer [
29]. However, ER-positive patients are less chemosensitive than ER-negative cases [
30] so that adjuvant chemotherapy might not be beneficial to some ER-positive breast tumors [
11]. And ER-positive patients also have distinct behaviors and outcome due to different molecular features. Thus, a biomarker which could accurately predict clinical outcome in ER-positive patients with breast cancer is needed urgently.
In the present study, we used miRNA expression microarray data from TCGA and divided the data into two cohorts based on the cases from two sequencing platforms. Following the strategy of using the larger cohort as training set, and the smaller one as the validation set [
31], we identified 14 miRNAs which were significantly associated with the status of ER both in training and validation set. The optimal cutoff value for each miRNA to discriminate different status of ER was determined by ROC curve. The risk score calculated from expression of each miRNA weighted by regression coefficient B fitted with status of ER might reflect the tendency from positive to negative status of ER. High score might be more likely related to negative status while low score to positive status. To assess the prognostic value of the miRNA signature, the ER-positive cases were divided into high and low score groups according to the risk score. Twelve miRNAs upregulated in ER-negative breast cancer patients exhibited high expression in high score group and two declined miRNAs showed high expression level in low score group. And the ER-positive patients in high score group suffered poorer survival compared with low score group both in the training stage and validation set.
Lowery et al. [
32] found that a 6-miRNA signature could predict status of ER, of which only miR-135b was consistently included in the 14-miRNA signature in our study. To some extent, difference of target population and/or the entry criteria might be responsible for the phenomenon. However, it was also reported that elevated miR-18a [
33], miR-505 [
34], miR-9 and reduced miR-375 [
35] were correlated to oestrogen receptor negativity. The results were consistent with our findings. In addition, re-expression of miR-375 could reverse tamoxifen resistance and epithelial-mesenchymal transition-like properties in the established tamoxifen-resistant breast cancer cells [
36]. Moreover, high expression of miR-187 in breast cancer could lead to a more aggressive, invasive phenotype and may act as an independent predictor of outcome [
37]. Prognostic value of miR-210 has been explored in many cancer types. Breast cancer patients with elevated miR-210 might have a poor outcome [
38],[
39]. Huang et al. [
40] found that miR-224 might act as an oncogene by directly suppressing the RKIP tumor suppressor resulting in promoting metastasis of breast cancer. A higher expression of miR-9 is associated with lymph node metastasis [
41] and could act as a predictor for local recurrence of breast cancer [
35]. However, the other 5 miRNAs were not explored so widely in breast cancer and further researches are required to investigate their complex molecular mechanisms.
The specificity of biomarkers based on a single miRNA is generally poor [
25]. Thus, we developed a risk score of combination the 14 miRNAs associated with ER status and multiplied their corresponding weight to survival and found that the score could predict overall survival in ER-positive patients. Better insights into the mechanism of the 14-miRNA signature in breast cancer might contribute to an understanding of the genetic aberrations that are involved in tumor genesis, progression and response to treatment.
In conclusion, the ER associated miRNA signature identified in our study might support a potential predictor to indicate clinical outcome for ER-positive patients and serve as potential molecular targets for new therapeutic strategies, subsequently leading to improved outcomes.
Authors- contributions
ZX, WXP, HZB carried out the molecular genetic studies. ZX, XL, ZW participated in the sequence alignment. ZX, LP drafted the manuscript. All authors read and approved the final manuscript.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interest.