Lack of expression of androgen receptor may play a critical role in transformation from in situ to invasive basal subtype of high-grade ductal carcinoma of the breast

Summary

Androgen receptor has been implicated in the pathogenesis of breast carcinoma. In this study, we explored the potential role of androgen receptor in breast cancer by analyzing its expression using immunohistochemistry and its relationship with tumor progress (ductal carcinoma in situ [DCIS] versus invasive ductal carcinoma [IDC]); nuclear grades (high grade [HG] versus non–high grade); expression of estrogen receptor (ER), progesterone receptor (PR), HER-2; and 3 molecular classifications: cytokeratin classification, triple (ER/PR/HER-2) negative classification, and ER/HER-2 classification in 184 breast carcinomas. We found that (1) lack of androgen receptor expression was associated with HG-IDC and with basal subtypes of HG-IDC, suggesting androgen receptor may play an important role in preventing the invasive transformation in this subgroup of breast carcinoma. (2) HG-IDC and HG-DCIS more frequently expressed androgen receptor than ER (55%-93% for androgen receptor and 18%-30% for ER) and were frequently androgen receptor+/ER− (63% for HG-DCIS and 39% for HG-IDC), which made androgen receptor a possible therapeutic target. (3) One third of HG-IDC was negative for androgen receptor, ER, PR, and HER-2, suggesting that further studies are needed to identify other key molecules for targeted therapy. We purpose that androgen receptor should be routinely measured for breast cancer.

Introduction

The role of androgen and androgen receptor (AR) in breast cancer is still uncertain. Several lines of evidence support their involvement in breast carcinogenesis. First, epidemiologic studies showed that increased serum androgen level was associated with an increased risk for breast cancer in postmenopausal patients [1]. Also, hormone replacement with combined estrogen and testosterone had a significantly higher risk for breast cancer over estrogen alone [2]. Second, AR is expressed frequently in breast carcinoma, including both invasive and in situ ductal carcinomas [3], lobular carcinoma [4], BRCA-mutated tumors [5], and mammary Paget's disease [6]. Third, expression of AR seems to correlate with better prognosis for breast carcinoma. A univariate analysis showed that AR, along with estrogen receptor (ER), tumor size, tumor grade, lymph node status, and high level of Ki-67, had prognostic power in primary breast cancer [7]. For ER-negative tumors, AR positivity was associated with significantly better disease-free survival compared with AR-negative ones [8]. Patients with AR-positive tumors were shown to have better survival after disease recurrence compared with AR-negative tumor (22 versus 12 months) [9]. Rakha et al [10] reported that in 1944 cases of invasive breast carcinomas, tumor size, lymph node stage, and AR positivity were the most useful prognostic markers, better than tumor grade, and expression of ER, progesterone receptor (PR), HER-2, epidermal growth factor receptor (EGFR), and p53. Fourth, AR may also be served as an effective therapeutic target in certain subgroup of patients. Androgen receptor–negative breast cancer had a significantly poorer response to tamoxifen therapy than AR-positive tumors [11]. Tamoxifen plus fluoxymesterone (androgen) offered some therapeutic advantage over tamoxifen alone in metastatic breast carcinoma [12]. Thus, it seems that AR is not only frequently expressed in breast carcinoma and related to prognosis, but also may serve as a predictive marker for hormone therapy and a potential therapeutic target for breast cancer, especially for ER/PR-negative breast cancer.

Several different but related molecular classifications have been developed recently. Cytokeratin (CK) classification divides tumors based on the expression of several CK markers, with basal/stem cell subtype being CK14-, CK17-, or CK5/6 positive and luminal cell subtype being CK8- or CK18 positive [13], [14], [15]. Breast carcinomas expressing stem/basal markers had worse prognosis [16], [17]. Triple-negative classification classifies tumors that are negative for ER, PR, HER-2 as basal subtype, which lack the benefit of specific target therapies such as tamoxifen, aromatase inhibitors, or Herceptin. Triple-negative phenotype has been shown to be the sole prognostic marker in lymph node–negative tumors [10]. Using ER/HER-2 classification, Matos et al [18] divided 168 cases of invasive breast carcinoma into 4 subtypes: basal (ER and HER-2−, 7.6%), basal-like HER-2 overexpression (ER− and HER-2+, 17.7%), luminal A (ER− and HER-2−, 56.3%), and luminal B (ER+ and HER-2+, 16.5%). Both basal and basal-like types were mostly grade 3 tumors, frequently p-cadherin-, p63-, and CK5-positive. This classification aims to provide better guidances for specific target therapy. Others showed that a combination of CK markers and ER/PR/HER-2 expression may provide a molecular classification that is more related to the classification defined by gene expression profiling data than using either CK markers or ER/PR/HER-2 alone [19], [20].

These previous studies have established a potential role of AR in breast cancer. However, its relationships with tumor progression, tumor grades, and these molecular classifications are not clearly defined. In this study, we explored the potential role of AR in breast cancer by analyzing the relationship of its expression with tumor progression (ductal carcinoma in situ [DCIS] versus invasive ductal carcinoma [IDC]); tumor cell nuclear grades (high grade [HG] versus non–high grade [NHG]); expression of ER, PR, and HER-2; and 3 currently used molecular classifications in 184 cases of DCIS and IDC.