Search tips
Search criteria 


Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS One. 2013; 8(12): e82650.
Published online 2013 December 4. doi:  10.1371/journal.pone.0082650
PMCID: PMC3853592

The Impact of Androgen Receptor Expression on Breast Cancer Survival: A Retrospective Study and Meta-Analysis

Syed A. Aziz, Editor


Recent studies have highlighted the role of androgen receptor (AR) as a prognostic biomarker of breast cancer. However, its predictive role in disease free survival (DFS) and overall survival (OS) still remains inconclusive. The present study aimed to retrospectively investigate the association between AR and survival outcomes in breast cancer and also identify this association by a meta-analysis of published researches. Clinical data from 109 patients with breast cancer, who underwent surgery at Ruijin Hospital, Shanghai, were retrospectively analyzed for immunohistochemical AR expression measured by tissue microarray. For meta-analysis, articles available in Pubmed on the relationship between AR and breast cancer outcomes were included. Data obtained from both were combined and analyzed. Women with AR positive tumors in the retrospective study had a significantly better DFS (HR 0.24, 95% CI 0.07-0.88) and OS (HR 0.19, 95% CI 0.04-0.85) than women with AR negative ones. Meta-analysis showed that AR expression in breast tumors was an indicator of better DFS (HR 0.52, 95% CI 0.43-0.64). In subgroup analysis, AR could predict DFS outcome in estrogen receptor (ER) positive (HR 0.45, 95% CI 0.34-0.59), ER negative (HR 0.42, 95% CI 0.26-0.67), and triple negative breast cancer (HR 0.40, 95% CI 0.23-0.69). Moreover, in ER positive breast cancer patients, the expression of AR could predict better OS (HR 0.39, 95% CI 0.19-0.82). The present analysis indicated that AR expression was associated with lower risk of recurrence in patients with all breast cancer types and better OS in cases with ER positive.


Breast cancer is one of the most common malignancies among women all over the world. It can be classified into five subtypes based on molecular therapy: luminal A , luminal B, normal breast-like, basal-like, and human epidermal growth factor receptor 2 (HER2) overexpressing tumors[1]. Different subtypes require different therapeutic strategies. To date, estrogen receptor (ER), progesterone receptor (PR), and HER2 have been proved to be important prognostic indicators for breast cancer. More importantly, they are also essential in determining the use of hormone therapy, chemotherapy, and targeted therapy in different subtypes. For example, the presence of hormone receptors including ER and PR could suggest the sensitivity of a tumor to endocrine therapy. In early breast cancer, adjuvant endocrine therapy of tamoxifen (TAM) for 5 years could reduce the risk of death and recurrence by 30 to 40% in hormone receptor positive patients[2]. Recently, adjuvant treatment with trastuzumab has been shown to significantly improve outcomes in patients with HER2-positive breast cancer[3]. Determination of HER2 status has been a standard nowadays for every patient with breast cancer to select adjuvant targeted treatment with trastuzumab. Despite tremendous efforts to reduce metastasis and deaths due to breast cancer, the prognosis is still poor. More than 20% of patients with early breast cancer could eventually develop incurable metastatic disease[2,4]. Therefore, it is high time to identify new targets and biomarkers to improve the prognosis of breast cancer.

Like ER and PR, androgen receptor (AR) also belongs to steroid nuclear receptor family. AR has been an important target in prostate cancer, and it has recently been considered as a potential biomarker in breast cancer. AR is commonly expressed in ductal cancer in situ and invasive breast carcinoma[5]. Moreover, AR can be co-expressed with ER and PR, approximately 60% of the time[6].

Roles of AR in breast cancer development and progression have not been very clearly understood. Some researchers have reported that the expression of AR is associated with a better prognosis. But the prognostic significance varies with the different molecular subtypes of breast cancer. To understand the role of AR in disease free survival (DFS) and overall survival (OS), the present study aimed to retrospectively investigate the association between AR and survival outcomes in breast cancer and also to identify this association by a meta-analysis of published researches.


1. Methods of retrospective Study

1.1. Ethics statement

This study was conducted after approval from the institutional review board of Ruijin Hospital, Shanghai, China; which also waived the need for consent, since there was no interaction with patients enrolled. The study was conducted based on their available medical information, which used in a de-identified fashion.

1.2. Patients and methods

Clinical data of 109 Chinese patients with breast cancer were collected from Comprehensive Breast Health Center, Ruijin Hospital, Shanghai Jiaotong University School of Medicine. All patients received surgical treatment for breast cancer between 2003 and 2008, and they were followed up from the date of diagnosis until May, 2013 or death, whichever came first. The median follow up time was 6.2 years. The characteristics of the patients are below (Table 1). TNM disease stage was classified according to the American Joint Committee on Cancer, 7th Edition.

Table 1
Characteristics of patients.

Archived formalin-fixed paraffin-embedded breast cancer blocks were collected for all patients, and a tissue microarray (TMA) was constructed. Immunohistochemical (IHC) staining of AR, ER, PR, and HER2 was performed. Anti-ER antibody, anti-PR antibody, and anti-AR antibody were obtained from DAKO (Carpinteria, USA) at 1: 100 dilutions. Anti-HER2 antibody was from ROCHE (4B5, Ventana). Sections were considered AR, ER, and PR positive, when >1% of tumor cell nuclei stained positive (Figure S1). HER2 was scored as 0, 1+, 2+, and 3+ according to the American Society of Clinical Oncology/College of American Pathologists guidelines. HER2 negativity was considered as HER2 0, 1+, and 2+; HER2 positivity was considered as 3+.

1.3. Statistical analysis

SPSS 16.0 statistical software (SPSS Inc., Chicago, USA) to perform the statistical analysis to identify prognostic significance of AR. Kaplan–Meier plot and Cox regression analysis were used, and P value of <0.05 was considered significant. DFS was considered as the interval (in months) between the date of breast surgery to first recurrence (locoregional recurrence and/or distant metastasis). OS was defined the time (in months) between the date of breast surgery and time of breast cancer-related death.

2. Methods of meta-analysis

2.1. Data selection

PubMed was searched to identify primary research publications on the association between AR and breast cancer prognosis. The following searches terms were used: “breast cancer” and “androgen receptor”. Inclusion criteria of studies searched were: (1) studies which measure AR expression by IHC in breast cancer tissue (2), studies which provide information on DFS or OS comparing AR positive with AR negative patients groups, and (3) articles published in English. The following were considered as exclusion criteria of the search: (1) review or case report, (2) AR was not evaluated by IHC, and (3) lack of key information for hazard ratio (HR) of DFS or OS. When an individual author published several articles obtained from the same patient population, only the newest or most complete article was included in the analysis. After extensive research, a total of 11 original articles were considered for the meta-analysis.

2.2. Data extraction and methodological assessment

Articles included were assessed independently by two reviewers (Qing Qu and Yan Mao). Data retrieved from the reports included author, journal, year of publication, cut-off value, number of patients, hormone receptor status, HR, and 95% confidence interval (CI) of survival.

2.3. Statistical methods

Statistical variables such as HR and corresponding 95% CI were directly taken and used to combine the data, if they were described in articles. LogHR values were used for aggregation of survival results. A meta-analysis on AR expression was performed both for OS and DFS. Forest plots were used to estimate the effect of AR expression on survival outcomes. I2 value was used to evaluate heterogeneity (I2 = 0-50%, no or moderate heterogeneity; I2 > 50%, significant heterogeneity); and fixed-effect model was used, if there was no significant heterogeneity. Otherwise, random-effect model was used. By convention, an observed HR <1 implied better survival for the group with AR expression. All statistical analyses were conducted using STATA version 11.0 (Stata Corporation, College Station, TX, USA).


1. Results of TMA

Of the 109 patients with breast cancer, 52 (47.7%) were AR-positive and 57(52.3%) were AR-negative (Table S1). Overall, there were 13 deaths (2 deaths in AR-positive and 11 in AR-negative groups) and 15 recurrences (3 recurrences in AR-positive and 12 in AR-negative groups) at the end of follow-up period.

Cox regression analysis showed that women with AR positive tumors had a significantly better DFS and OS compared with women with AR negative tumors. In univariate analysis, expression of AR was found to be associated with better DFS (p= 0.026) and OS (p= 0.022) (Table 2). In multivariate analysis, adjusted HRs of AR for DFS were 0.244 (95% CI, 0.068–0.876, p= 0.031), and 0.188 (95% CI, 0.041–0.855, p= 0.031) for OS (Table 3).

Table 2
Univariate analyses of DFS and OS in all population.
Table 3
Multivariate analyses of DFS and OS in all population.

2. Results of meta-analysis

A total of 1376 potentially relevant citations were reviewed. Through extensive research, a total of 12 articles, including 11 published articles and the present retrospective study, were included in the meta-analysis (Table 4)[7-17]. All the included articles were on the association of AR expression with DFS and/or OS and covered a total of 5270 patients with breast cancer. Five articles reported the HR of AR expression in patients with ER positive breast cancer in text or table, while six articles reported the HR in patients with ER negative breast cancer.

Table 4
Main characteristics and results of the eligible studies.

The overall rate of AR expression in these studies was 65.2%. The combined HR of DFS for all 12 eligible studies was 0.52 (95% CI 0.43–0.64) (Figure 1), indicating that AR expression in breast tumors was an indicator of low risk of recurrence. The HR of OS for all studies was 0.66, but it was not statistically significant (Figure 2). In subgroup analysis, the HR of DFS was 0.45 (95% CI 0.34–0.59) for ER positive patients, 0.42 (95% CI 0.26–0.67) for ER negative patients, and 0.40 (95% CI 0.23–0.69) for patients with triple negative breast cancer (TNBC) (Figure 3). The expression of AR in patients with ER positive breast cancer could also predict OS benefit, and the HR was 0.39 (95% CI 0.19–0.82); but the same was not predictable in ER negative and TNBC subgroups (Figure 4).

Figure 1
Forest plot (fixed-effects model) of 11 studies assessing the effect of AR expression on DFS in patients with breast cancer.
Figure 2
Forest plot (random-effects model) of 10 studies assessing the effect of AR expression on OS in patients with breast cancer.
Figure 3
Forest plot (fixed-effects model) of 8 studies assessing the effect of AR expression on DFS in different subtype of breast cancer patients.
Figure 4
Forest plot (random-effects model) of 7 studies assessing the effect of AR expression on OS in different subtype of breast cancer patients.


Various combined-modality therapies, which are in use in the recent days such as surgery, endocrine therapy, chemotherapy, radiation therapy, and targeted therapy, have improved the outcomes in patients with breast cancer. However, metastasis and recurrence are considered major contributors to treatment failure. Current knowledge of etiopathology, biology, and treatment protocols of breast cancer has benefited from the simultaneous analysis of multiple biomarkers, such as ER, PR, HER2, and Ki67. These four markers are essential in identifying a high-risk phenotype and determining the most efficient therapeutic strategies. However, since breast cancer is a complex and heterogenetic disease, these markers could not cover all disease features. Therefore, besides ER, PR, HER2 and Ki67, it is important to find out new markers with predictive value for survival of patients with breast cancer.

In the past decades, androgens have been identified to improve the efficacy of hormonal treatment and have been used to treat advanced breast cancer; however, their use has declined with the advent of tamoxifen[18]. Androgens and AR may have some important roles in breast cancer. Some studies have examined and indicated that androgen acts through AR in carcinoma cells and play important roles in biology and clinical behavior of breast cancer model systems and cell lines[19,20]. AR, commonly expressed in breast cancer tissues, has been reported as a biomarker to understand the prognosis of breast cancer. However, some clinical studies have reported that AR could not improve the survival. Hence, there is no consensus on the association between AR expression detected by IHC and good survival in patients with breast cancer. The role of AR on survival in breast cancer patients is not very clearly known until now.

To observe the correlation between AR expression and DFS or OS in patients with breast cancer, the present study retrospectively examined the role of AR evaluated by IHC in breast cancer outcomes and stratified the published researches in a meta-analysis later. Study results have shown that AR is a good marker both for lower risk of recurrence and longer overall survival. The risk of metastasis and death in AR positive patients was significantly lower than in AR negative patients. However, similar results were not obtained in the subgroup analysis, which might probably be due to the small sample size used.

Meta-analysis is a combination of several studies, and it is less influenced by individual findings from a single study. Since meta-analysis can help to summarize studies on specific topics, the current work also involved a meta-analysis, which compared AR positive versus AR negative expression for DFS and OS in patients with breast cancer. To the best of our knowledge, this is the first comprehensive and detailed meta-analysis, which revealed the prognostic role of AR in breast cancer. The present analysis, combined 12 independent studies, which included 5,270 patients with breast cancer. The results revealed that AR expression could predict lower risk of relapse in patients with breast cancer. In subgroup analysis, AR could also predict better DFS in patients with ER positive, ER negative, and TNBC types. The expression of AR could also predict better OS in patients with ER positive breast cancer.

This systematic review with meta-analysis had to address heterogeneity issues. A moderately significant heterogeneity was observed among the 12 studies. A well-standardized technique is very important to evaluate biological markers. However, there remained some limitations in this meta-analysis. In the studies included, the antibodies used in detecting AR expression were not the same. The definition of cut-off value was also different and varied from 1% to 10%. The heterogeneity also could be explained by the different molecule type of tumors and different disease characteristics.

Some trials had to be excluded from the meta-analysis, because they did not provide sufficient data on survival. Among them, 5 excluded studies, reported significant associations between AR expression and survival, but they did not have the values of HR and 95% CI of survival[21-25].

Several other limitations of these studies could not be ignored. Results of multivariate survival analysis reported in the articles were included in the meta-analysis; if these data were not available, univariate data were extracted instead. Another limitation was from DFS and progression free survival (PFS). Schippinger’s research was focused on the relationship between AR and PFS in the metastatic patients, however the present analysis combined the results of PFS together as DFS[8]. Park’s study separately reported the HR and 95% CI of AR in ER positive and ER negative patients. However, HR and 95% CI was not reported for all the patients studied. This made us to analyze this study as two separate studies for the meta-analysis[13].

Since the review was limited to the published scientific literature, a potential impact of publication bias could not be ruled out. Publication bias is a major concern for meta-analysis, because positive results tend to be accepted by journals, while negative results are often rejected.


Meta-analysis suggested that AR expression was associated with low risk of recurrence of breast cancer. It could be used to identify the low-risk patients earlier and guide clinical decisions. AR expression and breast cancer OS was dependent on ER expression. The influence of AR was significant for the OS among the ER positive patients. Since all these articles are retrospective in nature, well-designed prospective studies are recommended.

Supporting Information

Figure S1

Immunohistochemical analysis of AR (high-power field *400 magnification). A. Negative for AR. B. positive for AR.


Table S1

Patient characteristics based on AR expression.


Checklist S1

PRISMA 2009 checklist.


Flow Diagram S1

PRISMA 2009 Flow Diagram.



We give special thanks for the careful proofreading of this article, provided by Jun Zhang and Zhiwei Wang from Shanghai Key Laboratory of Gastric Neoplasms, Department of Surgery, Shanghai Institute of Digestive Surgery, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine.

Funding Statement

The authors have no support or funding to report.


1. Perou CM, Sørlie T, Eisen MB, van de Rijn M, Jeffrey SS et al. (2000) Molecular portraits of human breast tumours. Nature 406: 747-752.10.1038/35021093 PubMed: 10963602 [PubMed]
2. Early Breast Cancer Trialists' Collaborative Group (EBCTCG) (2005) Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 365: 1687-1717.10.1016/S0140-6736(05)66544-0 PubMed: 15894097 [PubMed]
3. Tuma RS (2005) Trastuzumab trials steal show at ASCO meeting. J Natl Cancer Inst 97: 870-871.10.1093/jnci/97.12.870 PubMed: 15956644 [PubMed]
4. Berry DA, Cronin KA, Plevritis SK, Fryback DG, Clarke L et al. (2005) Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med 353: 1784-1792.10.1056/NEJMoa050518 PubMed: 16251534 [PubMed]
5. Moinfar F, Okcu M, Tsybrovskyy O, Regitnig P, Lax SF et al. (2003) Androgen receptors frequently are expressed in breast carcinomas: potential relevance to new therapeutic strategies. Cancer 98: 703-711.10.1002/cncr.11532 PubMed: 12910513 [PubMed]
6. Kuenen-Boumeester V, Van der Kwast TH, Claassen CC, Look MP, Liem GS et al. (1996) The clinical significance of androgen receptors in breast cancer and their relation to histological and cell biological parameters. Eur J Cancer 32A: 1560-1565 PubMed: 8911118 [PubMed]
7. Agoff SN, Swanson PE, Linden H, Hawes SE, Lawton TJ (2003) Androgen receptor expression in estrogen receptor-negative breast cancer. Immunohistochemical, clinical, and prognostic associations. Am J Clin Pathol 120: 725-731.10.1309/42F00D0DJD0J5EDT PubMed: 14608899 [PubMed]
8. Schippinger W, Regitnig P, Dandachi N, Wernecke KD, Bauernhofer T et al. (2006) Evaluation of the prognostic significance of androgen receptor expression in metastatic breast cancer. Virchows Arch 449: 24-30.10.1007/s00428-006-0213-6 PubMed: 16628414 [PubMed]
9. Gonzalez LO, Corte MD, Vazquez J, Junquera S, Sanchez R et al. (2008) Androgen receptor expresion in breast cancer: relationship with clinicopathological characteristics of the tumors, prognosis, and expression of metalloproteases and their inhibitors. BMC Cancer 8: 149.10.1186/1471-2407-8-149 PubMed: 18507821 [PMC free article] [PubMed]
10. Luo X, Shi YX, Li ZM, Jiang WQ (2010) Expression and clinical significance of androgen receptor in triple negative breast cancer. Chin J Cancer 29: 585-590.10.5732/cjc.009.10673 PubMed: 20507730 [PubMed]
11. Castellano I, Allia E, Accortanzo V, Vandone AM, Chiusa L et al. (2010) Androgen receptor expression is a significant prognostic factor in estrogen receptor positive breast cancers. Breast Cancer Res Treat 124: 607-617.10.1007/s10549-010-0761-y PubMed: 20127405 [PubMed]
12. Yu Q, Niu Y, Liu N, Zhang JZ, Liu TJ et al. (2011) Expression of androgen receptor in breast cancer and its significance as a prognostic factor. Ann Oncol 22: 1288-1294.10.1093/annonc/mdq586 PubMed: 21109569 [PubMed]
13. Park S, Koo JS, Kim MS, Park HS, Lee JS et al. (2011) Androgen receptor expression is significantly associated with better outcomes in estrogen receptor-positive breast cancers. Ann Oncol 22: 1755-1762.10.1093/annonc/mdq678 PubMed: 21310761 [PubMed]
14. Hu R, Dawood S, Holmes MD, Collins LC, Schnitt SJ et al. (2011) Androgen receptor expression and breast cancer survival in postmenopausal women. Clin Cancer Res 17: 1867-1874.10.1158/1078-0432.CCR-10-2021 PubMed: 21325075 [PMC free article] [PubMed]
15. Tang D, Xu S, Zhang Q, Zhao W (2012) The expression and clinical significance of the androgen receptor and E-cadherin in triple-negative breast cancer. Med Oncol 29: 526-533.10.1007/s12032-011-9948-2 PubMed: 21519872 [PubMed]
16. He J, Peng R, Yuan Z, Wang S, Peng J et al. (2012) Prognostic value of androgen receptor expression in operable triple-negative breast cancer: a retrospective analysis based on a tissue microarray. Med Oncol 29: 406-410.10.1007/s12032-011-9832-0 PubMed: 21264529 [PubMed]
17. Park S, Park HS, Koo JS, Yang WI, Kim SI et al. (2012) Higher expression of androgen receptor is a significant predictor for better endocrine-responsiveness in estrogen receptor-positive breast cancers. Breast Cancer Res Treat 133: 311-320.10.1007/s10549-011-1950-z PubMed: 22231421 [PubMed]
18. Ingle JN, Twito DI, Schaid DJ, Cullinan SA, Krook JE et al. (1991) Combination hormonal therapy with tamoxifen plus fluoxymesterone versus tamoxifen alone in postmenopausal women with metastatic breast cancer. An updated analysis. Cancer 67: 886-891.10.1002/1097-0142(19910215)67:4 PubMed: 1991261 [PubMed]
19. Peters AA, Ingman WV, Tilley WD, Butler LM (2011) Differential effects of exogenous androgen and an androgen receptor antagonist in the peri- and postpubertal murine mammary gland. Endocrinology 152: 3728-3737.10.1210/en.2011-1133 PubMed: 21846805 [PubMed]
20. Naderi A, Chia KM, Liu J (2011) Synergy between inhibitors of androgen receptor and MEK has therapeutic implications in estrogen receptor-negative breast cancer. Breast Cancer Res 13: R36.10.1186/bcr2858 PubMed: 21457548 [PMC free article] [PubMed]
21. Søiland H, Kørner H, Skaland I, Janssen EA, Gudlaugsson E et al. (2008) Prognostic relevance of androgen receptor detection in operable breast cancer. J Surg Oncol 98: 551-558.10.1002/jso.21156 PubMed: 18937259 [PubMed]
22. Peters KM, Edwards SL, Nair SS, French JD, Bailey PJ et al. (2012) Androgen receptor expression predicts breast cancer survival: the role of genetic and epigenetic events. BMC Cancer 12: 132.10.1186/1471-2407-12-132 PubMed: 22471922 [PMC free article] [PubMed]
23. Mrklić I, Pogorelić Z, Capkun V, Tomić S (2013) Expression of androgen receptors in triple negative breast carcinomas. Acta Histochem 115: 344-348.10.1016/j.acthis.2012.09.006 PubMed: 23031358 [PubMed]
24. Loibl S, Müller BM, von Minckwitz G, Schwabe M, Roller M et al. (2011) Androgen receptor expression in primary breast cancer and its predictive and prognostic value in patients treated with neoadjuvant chemotherapy. Breast Cancer Res Treat 130: 477-487.10.1007/s10549-011-1715-8 PubMed: 21837479 [PubMed]
25. Agrawal AK, Jeleń M, Grzebieniak Z, Zukrowski P, Rudnicki J et al. (2008) Androgen receptors as a prognostic and predictive factor in breast cancer. Folia Histochem Cytobiol 46: 269-276 PubMed: 19056529 [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science