Introduction

Breast cancer is the most common malignancy and the first leading cause of cancer-related death in women worldwide. It is highly heterogeneous at molecular and clinical levels [1]. The main breast cancer subtypes include luminal A, Luminal B, Her-2-enriched and basal-like type, and normal-like type, depending on molecular profiles and several biomarkers [2]. Estrogen receptor (ER) and progesterone receptor (PR) as sex steroid hormone receptors are important biological markers for prognosis prediction in breast cancer. Androgen receptor (AR), another sex steroid hormone receptor family, is expressed in 70%-90% of breast cancer patients. The AR expression varies among breast cancer subtypes, and accounts for a large proportion in ER-positive tumors [3]. AR is a ligand-activated transcription factor. Increasing research shows AR played a dominant role in the development of breast cancer. AR can reportedly accelerate cell proliferation of ER-negative breast cancer and triple-negative breast cancer (TNBC) [4,5,6]. The main reason is that AR competes with ER for binding to androgen-responsive elements, leading to tumor cell growth [2]. The synergy between Her-2 and AR is reinforced by a positive feedback loop mechanism, which promotes Her-2 transcriptional upregulation and then activates related downstream pathways, accelerating AR-positive tumor growth [3, 7]. Additionally, AR promotes tamoxifen resistance probably by regulating cyclin D1 expression and promoting cell cycle progression [8]. However, some studies show serum androgen is positively associated with breast cancer risk whether in premenopausal or postmenopausal women [9,10,11,12]. Moreover, high AR expression is correlated with better disease-free survival (DFS) and overall survival (OS) [13]. Therefore, the existing views concerning the prognostic value of AR in different breast cancer subtypes are controversial.

Although some research shows AR-target drugs are effective for AR-positive breast cancer, the patients who exactly benefit from AR-target therapies remain uncertain. The aim of this study was to investigate the clinical characteristics of AR in breast cancer, and to evaluate the prognostic value and provide a therapeutic tool for breast cancer.

Patients and methods

Totally 432 breast cancer patients receiving surgical resection at the First Affiliated Hospital of Nanjing Medical University between January 2020 and May 2023 were enrolled This study was approved by Ethics Committee of the Hospital (Ethics code 2022-SR-473).

Immunohistochemistry(IHC)

For IHC assay, the breast cancer tissue slides were incubated with the indicated primary antibodies (anti-AR, Clone number: EP120; ZSGB-BIO) overnight at 4℃. And then chromogenic detection was performed through a DBA detection kit (Kit-2031, Maixin, China). Data were obtained and analyzed by two experienced doctors of the pathology department of our hospital. and AR scores were presented as levels (0, none; 1 + , weak; 2 + , moderate; 3 + , strong) and the percentage of stained nuclei and divided into five groups (0: < 5%; 1:6%-25%; 2:26%-50%; 3:51%-75%; and 4–75%). Finally, the score was expressed as (staining intensity × percentage of positively stained cells).

Criteria for pathological results

The ER/PR status criterion is that ≥ 1% stained tumor nucleus is positive. The Her-2 status criteriona is that immunohistochemistry (IHC) confirms 3 + or 2 + (Fish confirm IHC) is positive. Ki67 is expressed in positive cells to calculate the percentage [14]. The 8th edition of Breast cancer TNM Staging Atlas from the American Joint Committee on Cancer (AJCC).

Statistical analysis

Statistical analysis was performed on SPSS 22.0. Associations between AR expression and clinicopathological features were assessed using Chi-square or Student's t-test and analysis of variance (ANOVA). Significant risk factors were used to predict AR expression using a multivariate logistic regression model. p < 0.05 was deemed as significant.

Results

AR expression in breast cancer tissue specimens

Of the 432 cases of ranged from 23 to 81 years (52.15 ± 11.10). Fifty five (12.7%) of the 432 patients were AR-negative and 377 (87.3%) patients were AR-positive (Table 1).

Table 1 Clinical characteristics of breast cancer patients according to AR status
Full size table

Relationship between AR expression and clinical features

Overall, 87 patients received neoadjuvant chemotherapy, among which 14 patients showed pathological complete response (pCR) (10 AR+ patients, 4 AR patients), and 73 patients were non-pCR (58 AR+ patients, 15 AR patients). In our series, AR expression was not significant in pCR (p = 0.5057, Table 1). Furthermore, no significant correlations were observed between AR expression and age, menopausal status, tumor size, lymph node, distant metastasis or tumor stage (Table 1).

In addition, 11 patients relapsed during follow-up. Our series showed a significant difference in AR expression and poor prognosis (p = 2.503e-05, Table 1). According to histological grade, there were significant differences in AR expression (p = 0.001363, Table 1).

Relationship between AR expression and molecular typing of breast cancer tissues

In all patients, 206 (47.7%) patients belonged to Luminal type, 83 (19.2%) patients were HR+/Her-2+, 54 (12.5%) patients were HR-/Her2+, and 82 (19.0%) patients were TNBC (Table 2). AR expression was related to the presence of ER (p=5.357e-15), PR (p=9.594e-11) and Her-2 (p= 8.408e-05). AR expression was also significantly associated with Ki-67 (p= 0.0458) and histological subtype (p<2.2e-16) (Table 2).

Table 2 Relationship between the expression of AR and molecular typing characteristics
Full size table

Expression difference of AR in TNBC and non-TNBC tissues

Among the patients, 82 (18.98%) patients were TNBC with 44 AR-positive cases (53.66%), and 350 (95.14%) patients were non-TNBC with 333 AR-positive cases (95.14%). AR expressions were significantly different between the TNBC group and the non-TNBC group (p < 2.2e-16) (Table 3).

Table 3 Difference analysis of AR expression in triple negative and non-triple negative breast cancer tissues
Full size table

Multivariate analysis with significant factors

Multivariate logistic regression confirmed ER, PR and Her-2 as independent predictors for AR expression (p < 0.05). In addition, AR expression was significantly related to tumor recurrence (p < 0.05) (Table 4).

Table 4 Logistic regression analysis of AR expression and clinico-pathological factors
Full size table

Discussion

Breast cancer is hormone-dependent and routinely examined with ER/PR and AR, ER belong to the nuclear receptor superfamily [15,16,17]. Although AR has the same structure as ER/PR and is more extensively expressed AR is far less understood than ER and PR. We investigated the AR expression in breast cancer patients and correlated it with clinical-pathological characteristics.

We observed AR expression in about 87.3% (n = 377/432) of the cases in our cohort. This large proportion of AR expression is consistent with the literature. Research AR plays an important role in the pathogenesis of breast cancer, and is expressed in more than 70% breast cancer patients [18]. Other studies show the presence of AR in ER-positive breast cancer is correlated with tumor size, and histopathological grading [19, 20]. However, our investigation found AR expression was significantly associated with independent risk factors (e.g., ER, PR, Her-2 and Ki-67), but not with clinical-pathological characteristics (e.g., age, menopausal status, tumor size, lymph node, distant metastasis tumor stage) in breast cancer patients. This result may suggest AR is correlated with the histological subtype, and can be adopted as a potential biomarker.

The AR positive rate in ER-positive breast cancers is 66.9%, which is consistent with the reported rate of 60%-90% [17, 20, 21]. However, some studies indicate AR may act as a tumor suppressor in this sub-type [22, 23]. Meanwhile, AR expression is correlated with ER. One of the most likely mechanisms is that AR can competitively combine estrogen responsive elements to inhibit the transcriptionally active components of ER [22]. In addition, AR can directly bind to p300, a coactivator for competitive binding with ER, and then inhibits the function and downstream signaling pathways of ER [22], thus suppressing tumor growth in luminal breast cancer. Indeed, AR is also regarded as a good prognostic factor. For instance, the presence of AR in ER-positive breast cancer patients shows a better prognostic outcome in terms of DFS and OS [24, 25]. In HR + /Her2-T2N0 breast cancer, AR-positive patients have better DFS and lower risk of recurrence. In addition, AR negativity predicts a worse curative effect for adjuvant chemotherapy and endocrine therapy [13]. We did not discover significant relations between AR expression and time-to-event outcomes, which is because we did not follow up our patients for a long time and we will follow up full life.

AR expression accounts for 60% in Her-2-positive breast cancer patients, but the AR expression rate is low in this sub-type in our work, which may be related to the small sample size. Moreover, AR expression in Her-2-positive breast cancer patients predicts a worse prognosis, which may be involved in mediating Wnt/β catenin and Her-2 signaling pathways [26]. To the delight, we are conducting clinical trials on enzalutamide combined with trastuzumab for Her-2+/AR+ breast cancer patients, and hope to achieve good results.

As we know, TNBC have high heterogeneity and lacks target therapy, such as Her-2 and HR that predict poor prognosis and early metastasis. However, more than 50% of TNBC cases express AR, and more evidences suggest AR to be a solid target. Thike et Al. (2014) found AR-positive TNBC was related to better DFS [27]. Howeverother studies present opposite results. For instance, a clinical study with 559 TNBC cases shows AR negativity is correlated with a better prognostic outcome in terms of OS. For TNBC patients without lymph node metastasis, AR-positive patients also have a higher risk of death and recurrence [20, 24, 28]. Therefore, we are required to confirm the clinical role of AR. A. Di Leone disclosed that AR overexpression in TNBC showed a lower Ki67 rate and was related to a lower rate of pCR in neoadjuvant chemotherapy [29]. Currently, researchers pay more attention to TNBC to ensure long survival. In this aspect, AR-target therapy gives the choice for TNBC, which also need more clinical trials.

In summary, in-depth research on targeted therapy and combination therapy for breast cancer is demanded. Routine assessment of AR may help better personalize treatment for breast cancer, and can be used as a potential therapeutic target and novel biomarker in breast cancer. Furthermore, extensive research containing a large substantiation in patients, and more controlled prospective clinical trials are needed to validate the results.