Characterization and prognosis of estrogen receptor-positive/progesterone receptor-negative male breast cancer: a population-based study

Male breast cancer (MBC) is an uncommon disease, accounting for less than 1% of all breast cancer diagnoses in the USA [1]. However, the annual incidence was reported to increase from 1.0 per 100,000/year in the late 1970s to 1.2 per 100,000/year in 2000–2004 [2]. Due to its rarity, the epidemiology, tumor behavior, treatment, and prognosis remain poorly understood. Current knowledge was mainly based on small series of studies, except for the advancement made by the EORTC 10085/TBCRC/BIG/NABCG International Male Breast Cancer Program. The results of part 1, a retrospective joint central study of 1822 MBC patients, and part 2, a 30-month prospective registry of 557 cases, had been partially released lately [3‐6]; thus, further analysis and prospective trials are still yet to be conducted.

Suffering from lack of clinical trials and knowledge on molecular biology, clinicians have to extrapolate treatment strategies for MBC from female breast cancer (FBC) data, despite differences at the protein, genetic, and epigenetic level [7‐9]. Although several recent studies have assessed the prognostic factors of MBC, the conclusions are controversial and often blighted by the small number of patients [10‐12].

Testing for estrogen receptor (ER) and progesterone receptor (PR) markers has been recommended for all newly diagnosed breast cancer patients by the College of American Pathologists and American Society of Clinical Oncology [13]. Several studies have demonstrated high rates of ER positivity in MBC, for example, Cardoso et al. reported that up to 99.3% of tumors were ER-positive [3, 14, 15]. The range of PR expression is wider than ER among different published reports, from 58.8 to 96% [16]. In FBC, if human epidermal growth factor receptor 2 (HER-2) was negative, ER+/PR− and ER+/PR+ breast cancer would be categorized as luminal B subtype and luminal A subtype, respectively, with different prognosis. Given the fact that HER-2 was dominantly negative in MBC [3, 5, 17], we conducted this population-based study to compare ER+/PR− MBC with ER+/PR+ MBC and further investigate the clinical characterization and prognostic factors of ER+/PR− MBC.

Data were obtained from the US National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) database [18]. We selected patients diagnosed with breast cancer between 1990 and 2010 according to the following criteria: male, pathological diagnosis of invasive carcinoma, unilateral, ER-positive, and breast as the only primary site. Patients with unknown PR status were excluded. Data extraction was performed by SEER*Stat software version 8.3.2 based on the November 2015 data submission [19]. Marital status was divided into three categories: not married, married, and unknown, with the first one consisting of divorced, separated, single (never married), and widowed. The outcome of interests were breast cancer-specific survival (BCSS) and overall survival (OS). The former was calculated from the date of diagnosis to the date of breast cancer death, and OS was defined as the interval from diagnosis to the death from any cause.

Our study was approved by the ethics committee of our hospital, namely Northern Jiangsu People’s Hospital Ethics Committee. No informed patient consent was needed.

MBC is substantially different from FBC, arising with increasing frequency due to BRCA2 mutations with differential effects by gender of single nucleotide polymorphisms (SNPs) [20]. The rarity of MBC resulted in difficulty in operating randomized, controlled clinical trials and limited prognostic information and suboptimal treatment. Only 3 out of the 12 breast cancer trials that included male patients are phase 3 clinical trials, and just 1 trial is actively recruiting [5]. The implications for PR positivity have long been a focus of debate. Some researchers recommended the elimination of PR testing from the routine diagnostic work-up of invasive breast cancer [13]. Other researchers advocated assessment of PR status to distinguish subsets of ER-positive and ER-negative tumors [21]. Also, PR status was suggested as a useful tool for selecting initial therapy, because ER+/PR− tumors might benefit more from initial treatment with an aromatase inhibitor [22]. However, tamoxifen is still the standard endocrine therapy in MBC patients [17]. In our study, we found that the clinical characteristics and prognosis of ER+/PR− MBC were different from ER+/PR+ MBC, the former being more aggressive and experiencing a much shorter OS and BCSS. Given the fact that normally endocrine therapy would be administered to both ER+/PR+ MBC and ER+/PR− MBC patients, the survival difference possibly lay more in tumor behavior than treatments. This verified the importance of PR testing. Besides, according to our study, the prognosis of ER+/PR− MBC patients was significantly related to tumor stage and surgery other than demographic factors like marital status and race; herein, early detection, diagnosis, and intervention were of great importance to improve the outcome of these patients.

The median age at diagnosis of MBC is 65–69 years old [3, 6, 23‐25] in the West counties and a little bit younger in Asia [26] and the Middle East [27]. Most literature has validated the prognostic role of age at diagnosis [12, 24, 26, 28, 29]. Our study was in agreement with the previous research. Nevertheless, we found different effect of age between the PR-positive and PR-negative MBC. OS dropped significantly for PR-positive patients older than 55 years, while patients younger than 70 years old experienced similar OS in PR-negative group. Furthermore, age was not independently related to BCSS in PR-negative MBC patients. The analysis of EORTC 10085/TBCRC/BIG/NABCG International Male Breast Cancer Program partially supported our results [3]. Compared with patients diagnosed ≤ 40 years old, patients diagnosed ≥ 75 years old experienced a 25% higher mortality risk. Nonetheless, the authors did not conduct strata analysis according to breast cancer subtypes, and only deaths following a distant relapse were considered breast cancer mortality event. We assume that older onset of age with a high presence of comorbidities could explain the divergence between OS and BCSS. 32.6% of MBC patients were diagnosed with comorbidities, including hypertension, diabetes, and ischemic heart disease [26]. Nearly 40% of MBC patients died from causes unrelated to their breast cancer [24].

Histological grade is representative of the “aggressive potential” of the tumor and would be expected to predict the survival of MBC. Some literature did report that tumor grade was a predictor of OS and/or BCSS [25, 28, 30], so did our analysis of the PR-positive cohort. On the contrary, Vermeulen et al. [6] found that tumor grade was not independently associated with survival, so did our analysis of the PR-negative patients and some other research [3, 31]. Different “scoring systems” were applied for determining the grade of a breast cancer, including four-tier grading scheme and three-tier grading scheme, which undermined the comparison among different results. Also, the grading system that was initially developed for FBC may not be suitable in the MBC setting. Last, MBC could be a heterogenous disease with different subtypes exhibiting different prognostic patterns, as our work demonstrated.

Interestingly, chemotherapy was confirmed an independent prognostic factor in the multivariate Cox analyses of OS but did not reach significance with this test in BCSS, neither in ER+/PR+ nor ER+/PR− cohort, as shown in Tables 2 and 3. Since few studies analyzed OS and BCSS of MBC at the same time, our finding was not echoed. There might be some possibilities: first, the drugs somehow reduced the risk of dying from causes other than cancer; second, we did not use HER-2 status in the model, which was not available until 2010, so the conclusion might be partial. The application of the 21-gene breast recurrence score (RS) may shed some light on the option of chemotherapy. After testing 38 MBC patients, Turashvili et al. [32] found similar RS distribution in MBC and FBC patients. Besides, RS testing was declared to play a prognostic role in MBC [7]. Larger studies with different cohorts are needed to further identify the risk factors and optimize treatments for MBC patients.

We acknowledge some limitations to our study. We do not have the information regarding HER-2 status, as mentioned above. Also, as a retrospective analysis, our study may have introduced biases. Despite these limitations, our study, to our best knowledge, was the first to expound the characterizations and prognosis of PR-negative MBC. Also, our SEER-based study included the data on systemic treatments of this population, which was recently updated.

In conclusion, ER+/PR− MBC, compared with ER+/PR+ MBC, presented with more aggressive behavior and poorer survival. The prognosis was independently associated with stage and clinical intervention; thus, early diagnosis and individualized treatment were warranted to improve the outcome.