Introduction

Male breast cancer (MBC) is a rare disease representing less than 1 % of all breast cancers (BCs) and less than 1 % of cancers in men [1–3]. Due to its rarity, information about this disease is still fragmented and clinical approach is mostly based upon data derived from its largely known female counterpart. Indeed, MBC shares some similarities with female breast cancer (FBC) [4, 5].

On the genetic level, MBC susceptibility can be particularly referred to mutations in the two major BC genes, BRCA1 and, mainly, BRCA2 [3]. Besides the two well-known BC susceptibility genes, several genes functionally related to BRCA1 and BRCA2, such as CHEK2 and PALB2, have been associated with MBC susceptibility [6–9]. Overall, only about 10 % of MBCs are hereditary diseases. The largest part of MBCs are so-called sporadic cancers and may derive from genetic alterations at somatic level [10–12].

EMSY (c11orf30) gene maps on chromosome 11q13.5, a locus harboring several potential oncogenic drivers frequently amplified in BC. EMSY encodes for a protein that can bind and inactivate BRCA2 and was shown to localize to sites of repair after DNA damage [13]. Since its first description, EMSY inactivating function over BRCA2 pointed to a major role for EMSY gene into DNA repair mechanism and in cancer pathogenesis. As expected, EMSY involvement in tumorigenesis has been primarily reported in BRCA2-related tumors, mostly breast and ovarian cancers [13, 14].

Studies focusing on sporadic FBC have shown that these tumors may carry amplification of EMSY, suggesting that EMSY amplification could be the somatic counterpart of BRCA2 mutations and may thus explain sporadic BC development [13, 15]. In fact, FBCs with EMSY amplification and BRCA2 deletion seem to display the same pathology, suggesting that overexpression of EMSY and deletion of BRCA2 may insist on a specific pathway [13].

Given the role of EMSY in the DNA repair process, it has been hypothesized that EMSY amplification could represent a putative biomarker to evaluate the sensitivity to treatment with drugs targeting DNA repair mechanisms, such as PARP inhibitors. PARP inhibitors are generally used for treatment of several cancer types characterized by BRCA1/2 mutations [16]. Recent clinical data reported a possible use for these drugs also in patients with sporadic cancers, mostly ovarian cancers (OC) [17]. The contribution of EMSY amplification in affecting PARP inhibitors sensitivity in sporadic BCs and OCs has recently been investigated with contrasting results [18, 19].

Over the recent years, EMSY amplification in FBC has been studied and described with a frequency ranging from 7 to 25 % [13, 15, 20–23]. The parallel view between MBC and FBC and the growing data about an implication of gene copy number variations (CNVs) as a relevant pathogenic mechanism in sporadic MBC [24–28] lead to the attempt to characterize EMSY CNV in MBC. In 2012, Kornegoor et al. reported EMSY amplification in MBC as a less frequent event compared to FBC [27], but information about the real impact of this gene on MBC development, particularly in relation to BRCA1/2 mutations, are still lacking.

In order to deeply characterize EMSY involvement in male breast tumor pathogenesis, we analyzed EMSY CNV in a selected series of MBC cases, characterized for BRCA1 and BRCA2 mutations.

Materials and methods

A total of 75 MBC cases were included in the study. The series included 10 BRCA1/2-positive (2 BRCA1 and 8 BRCA2) and 65 BRCA1/2-negative tumors. Cases were also characterized for the main clinical-pathologic features, including estrogen and progesterone receptor (ER and PR) status, HER2 and MIB1 expression, and tumor grade (G). As shown in Table 1, 86.3 % of the tumors were ER-positive, 78.1 % PR-positive, 71 % HER2-negative, 57.1 % MIB1-negative and 47.1 % were G2.

Table 1 Clinical-pathologic features of the male breast cancer series analyzed
Full size table

Tumor DNA was extracted from FFPE and fresh frozen tissue sections following standard protocols. After extraction, DNA quantity was assessed through NanoDrop 1000 spectrophotometer (Thermo Scientific).

EMSY CNVs were analyzed by quantitative real-time PCR (qRT-PCR) using TaqMan probes (Life Technologies, Carlsbad, California, USA). TaqMan Copy Number assay for EMSY (Hs02953503_cn) and TaqMan Copy Number RNaseP Reference Assay containing primer and probe mix for target and housekeeping gene, respectively, were used (Life Technologies, Carlsbad, California, USA). Real-time PCR reaction was carried out in quadruplicate using a PRISM 7500 Fast platform (Life Technologies, Carlsbad, California, USA). EMSY copy number status, relative to the housekeeping gene RNaseP, was assessed using the \(2^{{ - \Delta \Delta C_{\text{t}} }}\) method and considering a normal male breast tissue sample as a calibrator. For each sample, a RQ value (where RQ = \(2^{{ - \Delta \Delta C_{\text{t}} }}\)) was obtained. Samples with RQ value <1.5, between 1.5 and 2, and ≥2 were defined as having no amplification, gain and amplification, respectively. Based on RQ values, low and high amplification levels were defined as follows: low amplification with 2 ≤ RQ < 3, high amplification with RQ ≥ 3. Associations between EMSY CNV and clinical-pathologic features were evaluated using 2 × 3 Fisher’s Exact test. A p value <0.05 was considered statistically significant.

Informed consent was obtained from all individual participants included in the study. The study was approved by the local ethical committee (Sapienza University of Rome, Protocol 264/12).

Results

Analysis of EMSY CNV was performed in a selected series of 75 MBC cases. As reported in Table 2, among the 75 tumors examined, 30 cases (40 %) showed no EMSY amplification, 19 cases (25.3 %) were characterized by EMSY gain, and 26 cases (34.7 %) showed EMSY amplification. Based on RQ values, two different subgroups were observed among the 26 cases with EMSY amplification: a first subgroup with RQ values between 2 and 3, defined as low amplification, included 17 cases (65.4 %), and a second subgroup with RQ values ≥3, defined as high amplification, included nine cases (34.6 %). Of the eight BRCA2-related tumors, five (62.5 %) showed low EMSY amplification and three (37.5 %) showed no EMSY amplification. Of the two BRCA1-related cases, one showed low EMSY amplification and the other showed high EMSY amplification (Fig. 1).

Table 2 Results from EMSY copy number variation analysis on 75 male breast cancer cases and associations with clinical-pathologic features
Full size table
Fig. 1
figure 1

Histogram showing EMSY copy number variation analysis in 75 male breast cancer cases characterized for BRCA1/2 mutations. Distribution of RQ values is shown. Four different subgroups were identified by RQ values: No amplification (RQ < 1.5; white columns); Gain (1.5 ≤ RQ < 2; light gray columns); Low amplification (2 ≤ RQ < 3; gray columns); High amplification (RQ ≥ 3; black columns). *BRCA2-related cases, °BRCA1-related cases

Full size image

Possible associations between EMSY CNV and BRCA1/2 mutation status, and between EMSY CNV and main pathologic features were evaluated. Considering both BRCA1 and BRCA2 mutations, a significant association between EMSY amplification and BRCA1/2 mutations emerged (p = 0.03). However when considering only BRCA2 mutations, no statistically significant association between EMSY amplification and BRCA2 mutations was observed (p = 0.10). Associations between EMSY CNV and pathologic features, including ER and PR status, HER2 and MIB1 expression and tumor grade were also evaluated but no statistically significant associations emerged (Table 2).

Discussion

Aimed at investigating whether EMSY CNV could be a candidate pathogenic mechanism for MBC, we analyzed EMSY CNV in a selected series of MBC cases characterized for BRCA1/2 mutations.

To date, only one study has investigated the involvement of EMSY alterations in MBC [27]. In this study, by Kornegoor et al. EMSY CNV was evaluated in a series of MBCs, not characterized for BRCA1/2 mutations, by MLPA using a gene panel. Results from this study showed EMSY amplification in 2 % and EMSY gain in 10 % of MBCs. In our study, we performed a targeted analysis of EMSY copy number variations by qRT-PCR with TaqMan probes, a sensitive and specific quantitation method. We found a frequency of 34.7 and 25.3 % for EMSY amplification and EMSY gain, respectively. Compared with the study of Kornegoor et al. we reported a higher frequency of EMSY CNVs in MBC. Our data are in line with the frequency of EMSY amplification in FBC and indicate a similarity among the somatic alterations involved in tumor development in female and male breast tumors.

We also investigated possible differences between BRCA-related and not related MBCs and observed a statistically significant association between EMSY amplification and the presence of BRCA1/2 mutations (p = 0.03). This result may suggest that EMSY amplification and BRCA1/2 mutations may be involved in a specific pathogenic pathway in MBC.

However, the association was not confirmed when considering only BRCA2 mutations. Brown et al. reported that in FBC EMSY gene amplification was a frequent event in BRCA1-related tumors, while was less common in BRCA2-related tumors [21]. Although it is limited by the low number of BRCA1-related tumors analyzed, there was evidence of similar trend in our MBC series. Furthermore, we defined two subgroups within cases with EMSY amplification characterized by low and high amplification levels. We observed that BRCA2-related tumors do not show high EMSY amplification but rather they were characterized by low or no EMSY amplification. Intriguingly, both BRCA1-related tumors showed EMSY amplification, one at high and the other at low level. Thus, our findings would suggest a possible relationship between BRCA1 mutations and a high increase in EMSY gene copy number and between BRCA2 mutations and low or no increase in EMSY gene copy number.

Previous studies on FBCs described an association between EMSY amplification and ER-positive status and poor prognosis [15, 20, 21]. In our study, no significant association between EMSY CNVs and ER status emerged, but this could be affected by the high proportion of ER-positive tumors in our series, as MBCs are more likely to be ER-positive than FBCs [3]. In addition, no significant associations between EMSY amplification status and the other main pathologic features (including PR status, HER2 and MIB1 expression and tumor grade) emerged. These observations deserve further investigation on a larger sample series.

Due to its role in DNA repair mechanism, EMSY has been investigated for its possible role as predictor of response to PARP inhibitors [19]. Indeed, it has been proposed that, as for BRCA mutations, amplification of EMSY may also lead to enhanced sensitivity to PARP inhibitors [13, 19]. Thus, our data might be eventually relevant for identification of MBC cases that would possibly benefit from targeted therapeutic approaches.

Recently, a work by Viré et al. reported a new intriguing EMSY function [29]. In particular, they identified a group of miRNAs whose expression seems to be deregulated by EMSY amplification in BC cells. One of these miRNAs, miR-31, is a known regulator of metastatic mechanism in BC, thus suggesting a role for EMSY also in the regulation of the metastatic process in BC. Based on these findings, it could be hypothesized that the different MBC subgroups identified in our work may be characterized by different miRNA profiles related to different EMSY amplification levels. Further studies are needed to prove this hypothesis.

In conclusion, our study may contribute to define molecular mechanisms underlying MBC pathogenesis. The characterization of MBC cases by EMSY amplification could help in unraveling the heterogeneity of male breast tumors allowing the definition of MBC subgroups with features possibly relevant for clinical management and therapeutic approach.