Oncogene amplification in male breast cancer: analysis by multiplex ligation-dependent probe amplification
- First Online:
- Cite this article as:
- Kornegoor, R., Moelans, C.B., Verschuur-Maes, A.H.J. et al. Breast Cancer Res Treat (2012) 135: 49. doi:10.1007/s10549-012-2051-3
- 563 Views
Gene amplification is an important mechanism for oncogene activation, a crucial step in carcinogenesis. Compared to female breast cancer, little is known on the genetic makeup of male breast cancer, because large series are lacking. Copy number changes of 21 breast cancer related genes were studied in 110 male breast cancers using multiplex ligation-dependent probe amplification. A ratio of >1.3 was regarded indicative for gene copy number gain and a ratio >2.0 for gene amplification. Data were correlated with clinicopathological features, prognosis and 17 genes were compared with a group of female breast cancers. Gene copy number gain of CCND1, TRAF4, CDC6 and MTDH was seen in >40 % of the male breast cancer cases, with also frequent amplification. The number of genes with copy number gain and several single genes were associated with high grade, but only CCND1 amplification was an independent predictor of adverse survival in Cox regression (p = 0.015; hazard ratio 3.0). In unsupervised hierarchical clustering a distinctive group of male breast cancer with poor prognosis (p = 0.009; hazard ratio 3.4) was identified, characterized by frequent CCND1, MTDH, CDC6, ADAM9, TRAF4 and MYC copy number gain. Compared to female breast cancers, EGFR (p = 0.005) and CCND1 (p = 0.041) copy number gain was more often seen in male breast cancer, while copy number gain of EMSY (p = 0.004) and CPD (p = 0.001) and amplification in general was less frequent. In conclusion, several female breast cancer genes also seem to be important in male breast carcinogenesis. However, there are also clear differences in copy number changes between male and female breast cancers, pointing toward differences in carcinogenesis between male and female breast cancer and emphasizing the importance of identifying biomarkers and therapeutic agents based on research in male breast cancer. In addition CCND1 amplification seems to be an independent prognosticator in male breast cancer.
KeywordsBreast cancerMaleMLPAAmplificationCopy numberSurvival
Gene amplification is important in the development and progression of cancer and could serve as a potential biomarker for prognosis or as a target for molecular therapy. In female breast cancer, HER2 is the best described oncogene with frequent amplificaion. HER2 amplification is correlated with poor survival and good response to targeted therapy [1, 2]. Other genes, like epidermal growth factor receptor (EGFR), Fibroblast growth factor receptor 1 (FGFR1), topoisomerase IIa (TOP2A) and MYC are also involved in female breast cancer and have prognostic and therapeutic implications [3–6].
Compared to female breast cancer, there is yet little knowledge regarding the genetic makeup of male breast cancer, because male breast cancer is a rare disease and the few available studies are based on small single institutional series . Treatment of male breast cancer has largely been extrapolated from its female counterpart, while there are important differences between male and female breast cancer, with higher ratios of estrogen receptor (ER) and progesterone receptor (PR) positivity in men [8–10]. Also the distribution of molecular subtypes by immunohistochemical analysis shows important differences. Luminal type A and B are by far the most frequently encountered subtypes and HER2 driven, basal-like and triple-negative tumors are very rare in men [11, 12]. The few gene expression studies performed recently in men showed that there might be important differences in molecular profile between male and female breast cancer [13–15]. However, the clinical and prognostic significance of genetic alterations in relevant breast cancer genes still needs to be elucidated in male breast cancer.
Multiplex ligation-dependent probe amplification (MLPA) analysis is a high throughput genomic technique enabling relative quantification of copy number or promoter hypermethylation in a variety of genes in one reaction, based on the simultaneous amplification of specifically hybridized probes on DNA that can be derived from paraffin embedded material [16, 17]. We previously showed in female breast cancer that MLPA analysis with a dedicated “breast cancer kit” allows evaluation of copy numbers in 21 important breast cancer genes, providing an overview of the most common amplifications . In the present study, we used MLPA to investigate copy number changes of 21 (female) breast cancer related genes in a large group of male breast cancer and correlate these genomic anomalies with clinicopathological features, patients’ outcome, and with previously obtained MLPA data from female breast cancers.
Materials and methods
Patients: specimens and clinical information
All consecutive cases of surgical breast specimens of invasive male breast cancer from 1986 to 2010 were collected from four different pathology labs in The Netherlands (St. Antonius Hospital Nieuwegein, Diakonessenhuis Utrecht, University Medical Center Utrecht, Laboratory for Pathology East Netherlands) as described in more detail previously . Hematoxylin and eosin (HE) slides were reviewed by three experienced observers (PJvD, RK, AM) to confirm the diagnosis and to type and grade according to current standards. Pathology reports were used to retrieve information on age, tumor size, and lymph node status. A total of 110 cases from which the paraffin blocks contained enough tumor for DNA isolation were included. The age of these patients ranged from 32 to 89 years (average: 66 years). Tumor size ranged from 0.8 to 5.5 cm (average: 2.2 cm). In 86 % lymph node status was known and 55 % of these patients had lymph node metastases. The majority of cases were diagnosed (according to the WHO) as invasive ductal carcinoma (90 %). The remaining cases were lobular (n = 3), mixed type (ductal/lobular) (n = 2), invasive cribriform (n = 1), papillary (n = 1), mucinous (n = 2), invasive micropapillary (n = 1) or adenoid cystic carcinomas (n = 1). According to the modified Bloom and Richardson score  most tumors were grade 2 (41 %) or grade 3 (36 %). Mitotic activity was assessed as before  with a mean mitotic index of 11 per 2 mm2 (range 0–56). For all cases hormone receptor and HER2 status were re-assessed as described previously . Tissue microarray (TMA) slides were used for immunohistochemical stainings for ER, PR and chromogenic in situ hybridization (CISH) for HER2 assessment, the latter showing HER2 amplification in only 4/110 cases (4 %). TMA slides were also stained for E-cadherin. Most tumors were ER positive (102/110, 93 %) and PR positivity was also common (71/110; 65 %). Only four cases were E-cadherin negative (three lobular carcinomas and one ductal carcinoma).
DNA extraction and MLPA analysis
Contents of the “breast cancer” MLPA kit P078-B1 (MRC Holland)
Function and clinical relevance
Signal transduction; poor survival 
Signal transduction; poor survival, tamoxifen resistance 
Protein metabolism; promotes invasion 
Signal transduction 
Transcription regulatory protein; chemoresistance 
Signal transduction; promoting metastases, chemoresistance, poor survival 
Transcription factor; poor survival 
Signal transduction; ER positivity, poor survival 
Transcription regulatory protein; poor survival 
Cell adhesion 
Signal transduction 
Protein metabolism 
Transcriptional coactivator; ER positivity 
Signal transduction; bad survival; trastuzumab response 
Signal transduction 
Regulation of the topological status of DNA; poor survival, susceptible for certain chemotherapy 
Microtubule stabilization; chemoresistance (taxanes) 
Signal transduction; predict distant recurrence 
Signal transduction; poor survival 
Signal transduction 
Control female breast cancers
A group of female breast cancer described previously was used to study differences in gene copy number change between male and female breast cancer . This group consists of 104 cases with a mean age of 58 years (range 30–86 years). Tumor size ranged from 0.2 to 6.5 cm (average 2.1 cm) and 46 % of the cases had lymph node metastases. Most cases were diagnosed (according to the WHO) as invasive ductal carcinoma (78 %) or invasive lobular carcinoma (11 %). Mean mitotic activity was 21 per 2 mm2 and according to the modified Bloom and Richardson score most tumors were grade 2 (34 %) or grade 3 (45 %). ER positivity was common (69 %, 70/101) and 48 % of the tumors were PR positive (48/101). HER2 amplification defined by immunohistochemistry and CISH was seen in 19 % of cases (19/102). The same “breast cancer kit” (P078-A1 kit; MRC Holland) was used, but because the gene content of the kit had been updated by the manufacturer in the meanwhile, only 17 genes could be compared between the groups. In addition, for two genes (EGFR and HER2) one of the probes was modified and for five genes one probe was deleted. Some reference probes were modified as well (Supplementary Table 1).
Statistical calculations were performed using SPSS for Windows v15.0. Correction for multiple comparisons was applied by resetting the 0.05 threshold according to the Holm–Bonferroni method. Differences between gene copy number and clinicopathological characteristics were calculated with ANOVA for continuous variables and with Pearson Chi-square (or Fisher’s exact test when appropriate) for categorical variables. The following clinicopathological features were dichotomized: age (>50 years), tumor size (>2.0 cm), mitotic activity (>8 mitoses/2 mm2), and histological grade (grade 1/2 vs 3). Correlation between number of gene amplification and clinicopathological features were calculated with Spearman’s rho. Unsupervised hierarchical clustering using the statistical program R (www.r-project.org) was performed to identify relevant clusters and co-amplification. We used the maximum distance and Ward’s clustering method and calculated the stability of the clusters with pvclust. Logistic regression analysis was performed to compare gene amplification in male and female breast cancer, taking significant differences in clinicopathological features between the two groups into account. Information regarding prognosis and therapy was requested from the Integral Cancer registration The Netherlands (IKNL). Survival data were available for 101 cases with a mean follow up of 5.7 years. Therefore, survival analysis was based on 5 years survival rates. For univariate survival analysis Kaplan–Meier curves were plotted and analyzed with the log rank test. Multivariate survival analysis was done with Cox regression including the variables that were significant in univariate analysis.
Copy number analysis by MLPA
Copy number gain was most frequently seen in the genes MTDH (52/106; 49 %) and CCND1 (49/106; 46 %), and these genes were also frequently amplified (13/106; 12 % and 19/106; 18 % respectively). The genes analyzed on chromosome 8 (FGFR1, ADAM9, IKBKB, PRDM14, MTDH and MYC) were also frequently affected with high rates of copy number gain and amplification. Thirteen cases (12 %) showed copy number gain of all genes analyzed on chromosome 8. Also the genes located on chromosome 17 were often affected, particularly TRAF4, CDC6, and BIRC5 with copy number gain in 37, 37 and 26 % of cases, respectively. However, amplification of these genes was rare (<4 %). Amplification of HER2 was also rare (4/106; 4 %). In five cases (5 %), all genes analyzed on chromosome 17 showed copy number gain. In 17 % of cases (18/106) no gene copy number changes were found. Losses were rare and seen in only seven genes of which CDH1 (10/106; 9 %) and AURKA (13/106; 12 %) were most frequently affected.
Correlation with clinicopathological features
Tumors with a copy number gain in one or more genes tended to have a more aggressive phenotype with more mitoses (p = 0.004) and a higher histological grade (p = 0.007) compared to tumors without gene copy number alterations. The number of genes with copy number gain was significantly correlated with a high mitotic count (p = 0.001) and a high histological grade (p < 0.001).
Correlation between gene copy number gain (>1.3) and clinicopathological features
Age (mean) young
Mitoses high (>8)
Mitoses (mean) high
Grade high (3)
LN meta negative
Three out of the four tumors with HER2 amplification (defined by CISH) also showed HER2 amplification using MLPA (p < 0.001). Loss of CDH1 was not correlated with any clinicopathological feature and loss of the CDH1 gene did not correlate with E-cadherin expression.
Comparison with female breast cancer
Reasoning from the cases, two major clusters were found (Fig. 3). These clusters were stable according to the approximately unbiased p values calculated with pvclust (p < 0.001). Cluster A consisted of 55 cases and was characterized by a low rate of gene copy number gain and gene amplification. Cluster B consisted of 51 cases and was characterized by CCND1 (73 %), MTDH (69 %), CDC6 (63 %), ADAM9 (57 %), TRAF4 (57 %) and MYC (53 %) copy number gain. The male breast cancers in cluster B showed significantly more mitosis compared to the tumors in cluster A (8 vs 14 mitosis; p < 0.001). Cluster B tumors were also more often grade 3 (p = 0.020) and were larger (2.4 vs 2.0 cm; p = 0.036) compared to cluster A tumors.
Using a Cox regression analysis, CCND1 amplification appeared to be the only single gene which was a predictor of survival aside from grade, mitotic count, and tumor size (p = 0.015; hazard ratio 3.0). When chemotherapy and hormone therapy were included in Cox regression, CCND1 was retained as an independent prognosticator. However, hormone therapy was an independent prognostic factor as well and was correlated with a favorable prognosis (p = 0.004; hazard ratio 0.225). Cluster B tumors (p = 0.009; hazard ratio 3.4) and tumors with copy number gain of all analyzed genes on chromosome 17 (p = 0.005; hazard ratio 4.8) were also independent predictors of poor survival. The multivariate models are supplied in supplementary format (Supplementary Table 3).
Gene amplification is an important mechanism of oncogene activation and is crucial for the development and progression of cancer. The identification of frequent copy number change in certain chromosomal regions can lead to identification of functional important genes in carcinogenesis, reveal distinctive groups of breast cancer and can be used as prognostic markers. Knowledge of gene profiling in male breast cancer is sparse, because male breast cancer is a rare disease and most studies are based on small single institutional series. In the present study we used the high throughput technique MLPA to study gene copy number alterations of 21 breast cancer related genes in a large multi-institutional cohort of 106 male breast cancer patients.
The average amount of genes that showed copy number gain was four (range 0–12), of which one (range 0–8) was amplified. 18 cases (17 %) did not show any copy number change in the studied genes. These 18 cases tended to be low grade cancers with few mitoses and seem to have favorable prognosis compared to male breast cancers with gene copy number gain. The number of genes with copy number gain was correlated with high grade and a high mitotic count. This is in line with female breast cancers, as the genome in high grade female breast cancers is also more rearranged and these patients have a poor outcome [23, 24]. Simultaneous copy number gain of all analyzed genes on chromosome 8, 11, and 17 was seen in 12, 10, and 5 % of the cases, respectively. This points to polysomy or gain of whole chromosome arms, a finding often seen in male breast cancer . This is interesting, as polysomy of e.g., chromosome 17 has been refuted in female breast cancer [25–30]. In our group of male breast cancer copy number gain of all genes located on chromosome 17 was an independent predictor of adverse prognosis.
Using unsupervised hierarchical cluster analysis a small sub-cluster was formed by FGFR1 and ADAM9. In female breast cancer, co-amplification of these chromosomal regions is also a common finding [24, 31]. In addition, two stable clusters of male breast cancer patients were identified with additional prognostic value to classical clinicopathological prognosticators.
HER2 amplification defined by MLPA in the present study strongly correlated with HER2 amplification status defined by CISH on TMA slides. Small differences found could be due to heterogeneity of tumors which could be missed or overrepresented in TMA slides. We have also previously validated MPLA against CISH and FISH .
CCND1 and MTDH were the genes which most frequently showed copy number gain (49 and 46 %, respectively), and often had amplification, indicating that these genes probably play an important role in male breast carcinogenesis. CCND1 encodes for cyclin D1, which is a cell cycle protein driving cell cycle progression through the G1 phase. It also enhances ER-mediated gene transcription and is especially overexpressed in ER positive female breast cancer . CCND1 amplification has been linked to ER positive tumors as well, although some did not find such a correlation [18, 24, 33]. In the present study, we could not identify a correlation between CCND1 copy number gain or amplification and ER status. A clear cut association between CCND1 amplification and patients’ outcome in female breast cancer is lacking, but CCND1 amplification may be associated with a poor prognosis, particularly in ER positive tumors [23, 24, 33–35]. In the present group of male breast cancers, tumors with CCND1 copy number gain tended to have a higher mean mitotic count compared to tumors without CCND1 amplification, a finding which is in line with the encoding protein function. More importantly, amplification of CCND1 was the only single gene which correlated with poor survival and had additional prognostic value aside from tumor size, mitotic count, and histological grade using a Cox regression analysis.
MTDH is involved in several signaling pathways and amplification of MTDH promotes metastases, enhances chemo-resistance and is associated with poor outcome in female breast cancer patients . In line with these findings in females, we demonstrated that male breast cancer with MTDH copy number gain showed a more aggressive phenotype with a high mitotic count and a high histological grade. However, no correlation with prognosis and MTDH copy number change was found in our group of male breast cancer and no correlation with lymph node metastasis was found either.
The genes located on different amplicons on chromosome 8p11 (FGFR1; 29 %, ADAM9; 39 % and IKBKB; 32 %) were also often gained. These genes have been correlated with ER positive female breast cancers . Since most male breast cancer cases are ER positive (93 % in the present group), frequent copy number gain of these genes can be explained by the high ratio of ER positive tumors [8, 9]. Nevertheless, we could not confirm the correlation between copy number gain or amplification of these genes and ER positive tumors in male breast cancer. However, in view of the low rate of ER negative male breast cancers in the present study, these results need to be interpreted with care. It is important to note that FGFR1 amplification enhances tamoxifen resistance, which is particularly clinically relevant in male breast cancer, as endocrine therapy is often indicated in these patients . Since FGFR1 copy number gain and amplification seems to be common in male breast cancer and is suitable for targeted therapy, this gene could be of further interest in male breast cancer . ADAM9, which is important in cell adhesion and tumor cell invasion, has potential in male breast cancer as well, since this gene is often affected and could be used for targeted therapy [38, 39].
Among the other genes studied, copy number gain of MED1, PRDM14, and BIRC5 were associated with a high grade phenotype, indicating that these genes play a role in the development or progression of aggressive male breast cancer. Indeed MED1 copy number gain tends to correlate with poor survival. We could not confirm the prognostic relevance of the other genes in male breast cancer patients.
Comparison with 103 female breast cancers revealed differences in copy number change between male and female breast cancer in a variety of genes, pointing toward differences in carcinogenesis. In male breast cancer CCND1 and EGFR were more often gained than in the female breast cancer group. In the group of female breast cancers EMSY and CPD copy number gain were seen more often than in males. In line with a previous comparative genomic hybridization study, female breast cancer showed more frequent amplification in a variety of genes, particularly in TRAF4 and EMSY . None of the genes studied were significantly more often amplified in male breast cancer. Alongside gender specific differences between male and female breast cancers, differences in genetic predisposition may also influence the genetic profile of these tumors. Approximately 10 % of men with breast cancer are known to have a genetic predisposition, and especially BRCA2 mutations seem to be important . Differences in BRCA mutations status between male and female breast cancers would have implications for the genetic makeup of these tumors and deserves further investigation.
In conclusion, copy number gain of the genes CCND1 (11q13), TRAF4 (17q11), CDC6 (17q21), and MTDH (8q22) is very common in male breast cancer (>40 %) and these genes probably play a role in male breast carcinogenesis. Tumors with copy number gain of one or more genes showed a highly malignant phenotype. Also MED1, PRDM14, MTDH, and BIRC5 seem to be important in the development or progression of high grade male breast cancer. Amplification of CCND1 was the most important single gene as it correlated with poor survival and had prognostic value in addition to the classical clinicopathological prognostic factors. Using unsupervised hierarchical clustering a distinctive group of male breast cancer tumors was identified with poor survival. Compared to female breast cancer CCND1 and EGFR were found to be more frequently amplified in male breast cancer, while in females EMSY and CPD were more often involved and more frequent amplifications of TRAF4 and EMSY were found. Our results point toward important differences in carcinogenesis between male and female breast cancer, emphasizing the importance in identifying specific biomarkers and therapeutic targets for male breast cancer.
Conflict of interest
The authors declare no conflict of interest.
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.