Introduction

Breast cancer is rare among men, while it is the most common cancer among women, with approximately 91,605 cases (excluding intraepithelial cancer) in 2017 [1]. In 2017, the Japanese Breast Cancer Society’s Breast Cancer Registry reported 591 cases of male breast cancer (MBC), which continues to increase every year [2]. According to the Demographic Survey of the Ministry of Health, Labor and Welfare, the morbidity and mortality rates of both, MBC and female breast cancer (FBC) tend to increase; however, the trend is more moderate in MBC than FBC [3]. On the other hand, the lack of the data and information of MBC is the issue.

In Japan, cancer statistics are evaluated based on regional cancer registries; however, MBC is not documented, and its clinicopathological features have not been examined. In addition, the evaluation of biomarkers such as estrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2) expression in the treatment of breast cancer is crucial yet lacking for MBC cases in Japan. According to previous reports from Western trials, MBC is characterized by older age and more hormone receptor-positive cases than FBC [4].

In 2011, the National Clinical Database (NCD), a nationwide system that links data collection to the first level of surgical specialization in the Japanese Surgical Board Certification System, adopted an annual web-based data collection system. In 2014, data on 1.6 million surgical procedures from more than 4,000 hospitals were collected [5]. Approximately 1.2 million cases are registered annually, representing more than 95% of the surgeries performed in Japan [6]. Many reports using the NCD have been published due to its reliable and abundant data [7,8,9,10,11]. Nevertheless, there are no large-scale reports on the clinicopathological characteristics of MBC in Japan. Understanding such characteristics is expected to be helpful for the treatment of this rare cancer. Thus, this study aimed to clarify the clinicopathological characteristics and identify the unmet needs of MBC cases in Japan.

Methods

Patient selection and data collection

We investigated patients diagnosed with breast cancer registered in the Japanese NCD between January 2012 and December 2018. This study was approved by the Institutional Review Board of the National Center for Global Health and Medicine (NCGM-G-003309-00) on September 6, 2019.

Robust patient variables, including age, sex, body mass index, performance status (PS), surgical procedure, family history, comorbidities, pathological factors, and perioperative systemic therapy, were obtained from the Japanese NCD.

ER and progesterone receptor (PgR) positivity were defined according to the ASCO/CAP 2010 guidelines [12], while HER2 positivity was defined according to the ASCO/CAP 2018 guidelines [13]. Subtypes were defined as follows: luminal type with HER2 negative, ER positive and/or PgR positive; luminal-HER2 type with HER2 positive, ER positive and/or PgR positive; HER2 type with HER2 positive, ER negative and PgR negative; triple negative breast cancer (TNBC) with HER2 negative, ER negative, and PgR negative.

Patients’ backgrounds were analyzed in total populations. TNM classification, histology, family history, and systemic treatment were analyzed according to each subtype. Comorbidity was collected from 2016. Thus it is analyzed data between 2016 and 2018.

Statistical analysis

All statistical analyses were performed with SAS ver.9.4 (SAS Institute, Cary, NC, USA). Statistical comparisons of categorical variables were performed using chi-squared or Fisher’s exact probability tests. Continuous valuables were compared using the Wilcoxon rank-sum test. Continuous valuables are expressed using the median and interquartile range or average. Two-tailed p values < 0.05 were considered statistically significant.

Results

Patients

A total of 594,316 breast cancer cases, including 3780 cases of MBC (0.6%) and 590,536 cases of FBC (99.4%) were diagnosed between January 2012 and December 2018 in Japan (Table 1). The median ages at MBC or FBC diagnosis were 71 (45–86) and 60 years (39–83) (p < 0.001), respectively. Furthermore, 2.0% of the MBC cases and 5.6% of the FBC cases were under 40 years of age (p < 0.001). Bilateral disease was observed in 2.7% of the MBC cases and 9.7% of the FBC cases. Meanwhile, 97.0% of MBC cases and 90.1% of FBC cases were unilateral disease (p < 0.001). The median body mass index was higher in the MBC group than in the FBC group (23.1 vs. 22.3, p < 0.001). In addition, MBC cases tended to have higher clinical stages than FBC cases with resectable and/or locally advanced disease: 7.4 vs. 13.3% stage 0, 37.2 vs. 44.3% stage I, 25.6 vs. 23.9% stage IIA, 8.8 vs. 8.4% stage IIB, 1.9 vs. 2.4% stage IIIA, 10.1 vs. 3.3% stage IIIB, and 1.1 vs. 1.3% stage IIIC (p < 0.001). The frequency of de novo stage IV disease was similar between the two groups (2.1 vs. 2.0%). Over 95% of the cases in both groups underwent surgery (95.2 and 96.1%). Breast-conserving surgery was more frequent in FBC cases (14.6 vs. 46.7%, p = 0.02). Finally, axillary lymph node dissection with or without sentinel lymph node biopsy was more frequent in MBC cases (32.9 vs. 25.2%, p < 0.001).

Table 1 Patients characteristics
Full size table

Pathological feature

ER/PgR and HER2 statuses were available for 3003 (79.4%) MBC cases and 464,346 (78.6%) FBC cases. ER-positive disease was observed in 95.6% of MBC cases and 85.3% of FBC cases (p < 0.001). The PgR group had similar statistics between groups (90.4 vs. 72.9%, p < 0.001). The incidence of HER2-positive disease was 9.5% and 15.7% in MBC and FBC, respectively (p < 0.001). MBC exhibited larger tumors and more lymph node positivity (p < 0.001) but the same rate of M1 disease (1.4 vs. 1.3%, p = 0.37). Furthermore, invasive ductal carcinoma was more frequent in MBC cases (83.7 vs. 77.8%), and invasive lobular carcinoma was more frequent in FBC cases (1.3 vs. 4.6%) (p < 0.001). The distribution of the nuclear grade was similar between MBC and FBC cases (Table 2). Finally, the luminal subtype was more frequent in MBC cases (88 vs. 74%) (Fig. 1).

Table 2 Pathological features
Full size table
Fig. 1
figure 1

Distribution of each subtype in MBC (a) and FBC (b)

Full size image

Family history of cancer and comorbidity

Patients with a family history of cancer were less likely to have MBC than FBC, regardless of the subtype (p < 0.001) (Tables 3). Missing/unknown family history was more frequent in patients with MBC compared with patients with FBC.

Table 3 Family history
Full size table

Comorbidities are recorded in Table 4. The proportion of patients with comorbidities was 57.3% for MBC and 32.8% for FBC (p < 0.001). The most frequent comorbidities of MBC were hypertension (35.1%), diabetes (14.7%), other malignancies (11.7%), neuro/peripheral vascular disease (7.9%), and coronary artery disease (6.9%).

Table 4 Comorbidities
Full size table

Systemic treatment

Neoadjuvant systemic therapy and adjuvant systemic therapy were less frequent in MBC cases than FBC cases (8.9 vs. 11.9%, p < 0.001; 82.7 vs. 86.3%, p < 0.001) (Table 5). Chemotherapy was less commonly administered in MBC cases (22.0 vs. 32.7%, p < 0.001), while endocrine therapy was more frequent in MBC cases (76.0 vs. 67.9%, p < 0.001). Specifically, the administration of chemotherapy in MBC vs. FBC cases according to subtype was 19.2 vs. 23.7% for Luminal, 47.1 vs. 60.4% for Luminal-HER2, 40.0 vs. 62.6% for HER2, 57.3 vs. 69.9% for TNBC. Meanwhile, endocrine therapy administration was similar in Luminal and Luminal-HER2 MBC and FBC cases (84.9 vs. 83.9%, 68.0 vs. 72.5%, respectively) (Table S4). Anthracycline, taxane, and anti-HER2 drug therapy was less frequent in MBC cases compared to FBC cases (13.0 vs. 20.9%, 10.1 vs. 18.0%, and 4.9 vs. 10.2%, respectively, p < 0.001).

Table 5 Systemic treatment
Full size table

Radiation therapy

Perioperative radiation therapy was performed in 14.3% of MBC cases and 44.3% of FBC cases (Table 6).

Table 6 Radiation therapy
Full size table

Discussion

To the best of our knowledge, this is the first study to report the real-world clinicopathological characteristics and treatment trends of Japanese male patients with breast cancer based on a nationwide registry database. Approximately, 600,000 patients with breast cancer were included in the study, and the frequency of male breast cancer was comparable to previous reports from Western countries at 0.6% [14, 15]. Furthermore, similar to previous reports from Western countries, the median age of MBC cases was more than 10 years higher than that of FBC cases [16,17,18,19,20]. The frequency of relatively advanced stage II or III breast cancer was higher in MBC than FBC, and this tendency was similar with pathological stages (Tables S1, S2a). This may be because males have much smaller breast tissue than females; thus, breast cancer in males easily invade the skin and pectoral muscles [19, 21]. In addition, because males rarely visit breast oncologists due to a lack of understanding of breast cancer, they may believe that they do not have a risk of breast cancer. Nevertheless, the frequency of de novo stage IV disease was similar in MBC and FBC.

The breast-conserving surgery rate for MBC was 14.6%, which was lower than that of FBC but higher than that reported in Western countries. This may be due to tumors within 2 cm in size being relatively higher in Japanese MBC [4, 22]. Moreover, axillary lymph node dissection was more common in MBC, which may be due to the higher incidence of node-positive breast cancer in males. Additionally, similar to previous findings, MBC was more likely to be hormone receptor positive and less likely to be HER2 positive than FBC (Table S2b). The luminal subtype was the most common at 87.8%, which is consistent with previous reports [4, 23]. The frequency of the HER2-type and TNBC was slightly higher in our study than that of previous reports, suggesting that the biology may be different in Europe and the United States than in Japan [4, 23,24,25]. A positive family history of breast cancer was more common in FBC, while family history tended to be unknown in MBC cases (Table S3). Thus, clinicians may be taking insufficient family history of breast cancer for males.

Comorbidities were more common in the MBC group. Cardiovascular disease was the most frequent comorbidity, and the frequency of other malignancies was high (11.7%) in the patients with MBC (Table 4). In addition to the higher median age of MBC cases, hereditary tumor syndromes, such as hereditary breast and ovarian cancers, may be a cause of the higher rate of comorbidities and malignancies [25,26,27]. However, there were no data on detailed other malignancies in NCD.

Neoadjuvant systemic treatment was administered in less than 10% of MBC and FBC cases, and no significant difference of frequency was observed (Table S4). The frequency of endocrine therapy as a systemic treatment was similar between MBC and FBC cases; however, the frequency of chemotherapy with anthracyclines, taxanes, and anti-HER2 agents was lower in MBC. This suggests that patients’ general condition, performance status, and/or cardiovascular complications may influence oncologists’ decision-making with patients. The number of MBC cases who received radiotherapy was approximately one-third that of FBC cases, which could be because most MBC cases undergo mastectomy. Nevertheless, the proportion of patients who underwent BCS received radiation was slight in MBC. Furthermore, some patients eligible for post mastectomy radiation therapy, such as pN2 and pN3, also did not receive radiation. This may be related to older age and more comorbidities in Japanese MBC.

Real-world data analyses, such as the current study, boast a high number of patients. Our study included approximately 600,000 patients, which is much higher than of other retrospective studies. This high number of patients aids in the understanding of trends of clinicopathological features and treatment of MBC. Nevertheless, this study had several limitations. First, there was a risk of bias due to the retrospective nature of the study. Second, most databases, including the NCD, have missing data; thus, the true percentage of each value may not be reflected. Further analysis with a large cohort is required to obtain more reliable evidence. Addition to it, the NCD data are registered primarily by breast surgeons. Therefore, de novo stage IV data may be less available than the reality. Third, the NCD does not provide enough data on long-term survival, inhibiting the comparison of survival outcomes of MBC and FBC. Further studies should include additional histopathological and clinical data from the same cohort to obtain stronger conclusions, and such a study would be helpful in conducting clinical trials on MBC.

In conclusion, Japanese MBC had an older age of onset, were more likely to be hormone receptor-positive disease, and received less perioperative chemotherapy than FBC. This is the first comprehensive analysis using real-word data from a nationwide registry database in Japan of clinicopathological features and treatment trends in Japanese MBC cases. Further prospective studies are needed to evaluate the most suitable treatment strategy for MBC in Japan.