Male breast cancer: a 20-year multicenter experience

Male breast cancer (MBC) is a rare clinical entity accounting for approximately 1% of breast cancer in the US and worldwide [1]. However, the incidence of MBC has steadily increased in the past few decades with an estimate of 2620 new diagnosis of MBC in 2020 compared with 900 cases in 1991 [2]. Risk factors for MBC are similar to female breast cancer (FBC) including age, family history, ethnicity and genetic mutations like BRCA1 and BRCA2. However, there are some distinct risk factors for MBC such as testicular dysfunction or liver cirrhosis with hyperestrogenism [3]. Most men present with painless breast lumps and may have a delayed diagnosis when compared with woman [4]. Due to lack of prospective studies, management of MBC models that of FBC. Management options for early-stage disease include surgery such as modified radical mastectomy or breast conserving surgery, adjuvant chemotherapy, adjuvant endocrine therapy such as tamoxifen and adjuvant radiation therapy. Management of advanced stage disease primarily include adjuvant chemotherapy and endocrine therapy [5]. It is reported that MBC carries a higher overall mortality rate than FBC even after adjusting for multiple prognostic factors [6].

Due to the paucity of data in MBC research we aimed to further characterize MBC including risk factors, management and outcomes in the last 20 years across six different hospitals within the New England Steward HealthCare System, and determine the prognostic factors associated with survival outcome.

Materials & methods

Participants & study design

Patients diagnosed with male breast cancer between 1 January 2000 and 1 January 2020 were identified from the database of respective cancer registry across six different hospitals within the New England Steward HealthCare System (St. Elizabeth’s Medical Center, Saint Anne Hospital, Norwood Hospital, Morton Hospital, Good Samaritan Hospital and Holy Family Hospital). The institutional review board of Steward Health System approved this study. After the list of patients were received from respective cancer registry, individual medical records were manually reviewed to include only pathologically confirmed MBC and those with complete treatment records that clearly documented pathological findings, additional work ups and treatments received. Once these patients were identified, their clinical information including age, presenting symptoms, family history, imaging modality, genetic studies, Oncotype DX score (if available) and follow-up duration were recorded in a secure file, as were their pathological data including tumor type, size, grade, nodal involvement, stage and hormonal receptor status. Additionally, their treatment modalities including neoadjuvant treatment, type of surgery, adjuvant radiation, adjuvant endocrine therapy and adjuvant chemotherapy were recorded. Age at diagnosis was classified as <50 years, 50–65 years and >65 years following the same classification as has been previously done in larger MBC studies [7–9].

Statistics

Patients’ clinical, pathological and treatment information were characterized by descriptive statistics. A 5-year, 10-year overall survival (OS) rate, and 5-year relapse free survival (RFS) rate were the primary end points of this study. Kaplan–Meier method was used to determine the 5, 10 year OS rate and RFS rate from date of diagnosis until death or relapse, respectively. Univariable Cox regression test was used to assess the association between patient outcome and age at diagnosis, metastasis status at time of diagnosis, tumor size, nodal involvement and use of chemotherapy. Due to the small sample size in this study, we did not conduct multivariable testing. Data analyses were performed using Stata version 12.

Results

A total of 107 men were identified with pathologically diagnosed BC; however, only 64 men were included in this study due to lack of access to complete medical records as a result of poor documentation for the remaining 43 men. Median age at time of diagnosis was 68 years (range: 30–89). Majority of patients were white (n = 58), four patients were black and the remaining two were Hispanic. A total of 13 patients had family history of breast or ovarian cancer in either first- or second-degree relatives and 19 patients had family history of any type of cancer. One patient had history of liver cirrhosis. None of the patients had other known risk factors of male breast cancer including cryptorchidism, Klinefelter syndrome or exposure to environmental radiation. A total of 57 patients presented with palpable breast lump, five patients had incidental findings of breast mass on chest imaging while undergoing investigations for other intra-thoracic pathologies, and two patients had axillary masses without palpable breast lump. Four of these patients had nipple changes. A total of 27 patients received genetic testing and four patients tested positive for BRCA2, one patient positive PALB2, a gene that encodes BRCA2 interaction protein necessary for DNA repair and one patient positive for AXIN2, a gene that regulates the stability of beta-catenin in the Wnt signaling pathway. Five out of 17 patients (29%) received genetic testing between 2000 and 2010, and 22 out of 47 patients (47%) received genetic testing between 2011 and 2020. Ten patients had Oncotype DX score obtained. Patient characteristics are summarized in Table 1.

A total of 59 patients had invasive ductal carcinoma, which was the predominant histology. Most patients had tumors that were grade 2 or 3. T1 and T2 were the predominant tumor sizes. Approximately half of the patients had nodal involvement at time of diagnosis (n = 31). A total of 62 patients had no distant metastasis at time of diagnosis, one patient had metastasis to lung and bone at time of diagnosis and one patient had metastasis to brain at time of diagnosis. About 97% of patients had ER positive cancer (n = 62), 87.5% of patients had progesterone receptor positive cancer (n = 56) and 7.8% of patients had HER2 positive cancer (n = 5).

Regarding the surgical intervention to treat the primary tumor, 23 patients had simple mastectomy, 33 patients had modified radical mastectomy, five patients received breast-conservative surgery (BCS), two patients did not receive surgery as there were metastatic disease at time of diagnosis and one patient declined surgery. Only five out of 25 patients (20%) who had tumor size less than 2 cm and no nodal involvement at time of diagnosis received BCS. Forty-nine patients underwent lymph node biopsy and 26 of these patients later underwent axillary lymph node dissection. Twenty-two patients underwent axillary lymph node dissection without sentinel lymph node biopsy. Most patients with nodal involvement (30/31) and all patients who underwent BCS (5/5) received adjuvant radiation therapy. A total of 29 patients received chemotherapy with ten patients receiving neoadjuvant chemotherapy and the remaining 19 patients receiving adjuvant chemotherapy. Twenty-four patients received anthracycline-taxane based chemotherapy, and five patients received adriamycin, cyclophosphamide, tratuzumab and pertuzumab (AC-THP). All patients that were ER+, except one that decline treatment, received adjuvant hormonal therapy (n = 61). Fifty-three patients received tamoxifen as adjuvant endocrine therapy; however, four patients were switched to Anastrozole with leuprolide due to intolerable side effects including hot flashes, muscle cramps and deep vein thrombosis. Five patients had disease progression while receiving tamoxifen; of these, four patients went on to receive anastrozole and leuprolide and one patient received fulvestrant and leuprolide. Treatment characteristics are summarized in Table 2.

The median follow-up was 71.5 months (range 6–250 months), 5-year OS rate was 86.9% (95% CI: 74.5–93.6%, Figure 1), 5-year RFS rate was 89.9% (95% CI: 77.4–95.7%, Figure 2) and 10-year OS rate was 58.1% (95% CI: 41.0–71.8%). The presence of distant metastasis at the time of initial diagnosis was the only factor associated with worse survival outcome (hazard ratio: 22.54, p < 0.0001, Figure 3). Age, tumor size, chemotherapy, and positive lymph node at time of diagnosis were not significantly associated with OS (p = 0.5079, p = 0.7504, p = 0.5431 and p = 0.4 respectively, Figure 4).

Discussion

Male breast cancer (MBC) is a rare disease and prospective studies evaluating effective treatments have been limited due to lack of enrollment. As a result, the treatment regimen for MBC mirrors that of female breast cancer (FBC). This multicenter, retrospective study characterizes the clinical, histopathologic factors, management and outcomes of MBC across six hospitals in New England. Most patients presented with painless breast lump, had early-stage disease and tested positive for hormone receptor. Most patients underwent mastectomy and received adjuvant hormonal therapy with tamoxifen. The survival outcome of MBC compared with its female counterpart has been reported in many studies; however, the results vary. Two SEER studies analyzing data from 2005 to 2010 and 2004 to 2014 found that MBC had significantly worse 5-year OS when compared with FBC, even after adjusting multiple risk factors including age, race, disease stage and geography [6,10]. A recent study of 2175 MBC patients from various cancer registries in Italy by Mangone et al. also showed worse 5-year OS (82%) in MBC compared with FBC [11]. Several other studies reported that the 5-year OS is comparable between MBC and FBC after adjusting for risk factors [12–14]. According to the American Cancer Society, the reported 5 and 10 year OS rate for FBC is 91 and 84% respectively [15,16]. In our study, the 5-year OS rate in MBC was 86.9%, similar to the rate reported in the literature for FBC; however, it is difficult to make direct comparison due to the small patient population in our study.

A study examining 2992 MBC patients from SEER between 2003 and 2012, found that age at diagnosis, tumor grade, stage, surgery, radiotherapy, ER and marital status had prognostic value in MBC [9]. In our study only metastasis at time of diagnosis was found to be significantly associated with worse prognosis. The historically cited prognostic factors were not found to have significance in this study, possibly due to small sample size.

Several genetic mutations have been implicated in the pathogenesis of MBC, most notably BRCA2, followed by CHEK2, PALB2 and BRCA1 [17]. Genetic testing is indicated for all newly diagnosed MBC as it provides prognostic value, subject the offspring of these patients for early screen and may identify eligible men for treatment targeting DNA repair pathways such as poly-ADP-ribose polymerase inhibitors [18,19]. In our study, 42% of the patients received genetic testing at time of diagnosis and nearly 50% of them received genetic testing after 2010. This proportion increased compared with the prior decade, reflecting a change in practice over time.

The oncotype DX RS is a 21-gene expression assay that detects the expression of 16 cancer-related genes and five reference genes in BC [20]. It can used to quantify the risk of distant recurrences at 10 years and the benefits of adjuvant chemotherapy in patients with hormone receptor-positive BC and have either no or up to three lymph nodes involvement [21]. Its use has been validated in FBC but not in MBC. However, its routine use is also recommended in MBC as majority of MBCs are hormone receptor-positive [20]. According to Williams et al., significantly fewer MBC patients who were eligible receive oncotype DX score compared with FBC [22]. In our study, about 18.5% of patients who were eligible received oncotype DX score. To better guide the use of chemotherapy, the oncotype DX test needs to be adopted much more frequently in MBC.

BCS shows similar survival outcome when compared with mastectomy in FBC. In a recent retrospective study by Bateni et al. showed BCS followed by radiotherapy in men is associated with improved OS compared with partial mastectomy alone, total mastectomy alone and total mastectomy with radiotherapy (10 year OS 73.8 vs 56.3, 58 and 56.3%, p < 0.0001) [23]. However, most men who are eligible for BCS are still treated with mastectomy. For example, in the study by Cardoso et al., only about 4% of eligible patients received BCS [24]. In the study by Batetni et al. only 18.2% patients received BCS [2]. In our study, only 20% of patients with T1 tumor received BCS; however, all of them received postoperative radiation therapy. While BCS has a number of advantages to mastectomy, the use of postoperative radiation is essential and highlights the importance of shared-decision making and careful treatment planning.

There are several limitations to our study. This is a retrospective analysis and involves a large time span (20 years), in order to capture a reasonable sample size. However, given that MBC is a rare disease, our study sample size is relatively small. Particularly, when considering the outcomes for patients with metastatic disease, which are underrepresented in our study (n = 2). Therefore, caution should be taken when interpreting the poor outcomes of metastatic patients. In addition, the small sample size prevented us from being able to conduct adjusted analyses.

Conclusion

In summary, majority of MBC patients present with palpable breast mass and the tumors are almost exclusively hormone receptor positive. Majority of patients present without distant metastasis and received surgical therapy; however, only a small subset of patients received BCS for T1 tumors. Almost all patients received adjuvant hormonal therapy. OS and RFS in our cohort is similar to historic reports in FBC. However treatment of MBC mirrors that of FBC as no prospective studies of MBC were carried out. Hopefully collective multinational effort will facilitate the conduction of exclusively MBC prospective trials in the near future.