Key Points
Question
What is the breast cancer–specific mortality (BCSM) risk over 20 years in men with stage I to III hormone receptor–positive breast cancer?
Findings
In this cohort study including 2836 men, the cumulative 20-year risk of BCSM was 12.4% for stage I, 26.2% for stage II, and 46.0% for stage III. Smoothed annual hazard estimates for BCSM revealed an increase in late hazard rates with each incremental node category, reaching a bimodal distribution in N3 and stage III.
Meaning
The findings of this study suggest that the risk of BCSM at 20 years is high in men with hormone receptor–positive breast cancer, and the kinetics appear to be different from those in women.
Abstract
Importance
In women with hormone receptor–positive (HR+) breast cancer, the risk of distant recurrence and death persists for at least 20 years from diagnosis. The risk of late mortality in men with HR+ breast cancer has not been reported.
Objective
To report 20-year risks of breast cancer–specific mortality (BCSM) and non-BCSM in men with stage I to III HR+ breast cancer and identify factors associated with late BCSM.
Design, Setting, and Participants
An observational cohort study was conducted of men diagnosed with HR+ breast cancer from 1990 to 2008, using population-based data from the Surveillance, Epidemiology, and End Results program. Men diagnosed with stage I to III HR+ breast cancer were included in the analysis. Cumulative incidence function was used to estimate the outcomes of baseline clinicopathologic variables regarding cumulative risk of BCSM and non-BCSM since diagnosis. Smoothed hazard estimates over time were plotted for BCSM. Fine and Gray multivariable regression evaluated the association of preselected variables with BCSM, conditional on having survived 5 years.
Main Outcome Measure
BCSM.
Results
A total of 2836 men with stage I to III HR+ breast cancer were included, with a median follow-up of 15.41 (IQR, 12.08-18.67) years. Median age at diagnosis was 67 (IQR, 57-76) years. The cumulative 20-year risk of BCSM was 12.4% for stage I, 26.2% for stage II, and 46.0% for stage III. Smoothed annual hazard estimates for BCSM revealed an increase in late hazard rates with each incremental node category, reaching a bimodal distribution in N3 and stage III, with each having peaks in hazard rates at 4 and 11 years. Among patients who survived 5 years from diagnosis, the adjusted BCSM risk was higher for those younger than 50 years vs older than 64 years, those with grade II or III/IV vs grade I tumors, and stage II or III vs stage I disease.
Conclusions and Relevance
The findings of this study suggest that, in men with stage I to III HR+ breast cancer, the risk of BCSM persists for at least 20 years and depends on traditional clinicopathologic factors, such as age, tumor stage, and tumor grade. Among men with higher stages of disease, the kinetics of the BCSM risk appear different from the risk that has been reported in women.
This cohort study examines the risk of mortality in men with vs without breast cancer–specific mortality.
Introduction
Breast cancer (BC) is the leading cause of cancer-related death in women, but is rare in men. In the US, it was estimated that 297 790 women in the US would be diagnosed with BC and 43 170 would die of the disease. In comparison, 2800 men were likely to be diagnosed with BC, and approximately 530 men were anticipated to experience BC-specific mortality (BCSM).
Due to its rarity, invasive BC in men remains understudied. Nevertheless, several studies have identified unique features of male BC (MaBC). More than 90% of patients with MaBC are diagnosed with hormone receptor–positive (HR+) disease, and ERBB2+ (formerly HER2+) disease accounts for 8% to 12% of MaBC cases, and most of these are also HR+. Hormone receptor−negative/ERBB2+ and triple-negative BC are rare in men. A previous population-based study observed that men with stage I to III HR+/ERBB2− BC have significantly better 5-year overall survival outcomes than men with HR+/ERBB2−, HR−/ERBB2+, or triple-negative BC. Invasive ductal carcinoma is the most common histologic subtype, and invasive lobular carcinoma makes up a smaller percentage of cases in men than in women. The median age at diagnosis is substantially higher in men than in women, and patients with MaBC are more likely to have regional nodal involvement.
Most women with BC and patients with MaBC are diagnosed with stage I to III disease. Women with early-stage HR+ BC face a continuous risk of distant recurrence and death for decades. The long-term risk of BCSM in women is influenced by disease features at initial diagnosis, including tumor stage, grade, and nodal status. The impact of these and other factors on the long-term risk of BCSM in patients with MaBC remains uncertain. To date, published studies on BCSM beyond 5 years in men with nonmetastatic HR+ BC have been limited to case series of patients treated at individual institutions.
To investigate this, we conducted a population-based cohort study using Surveillance, Epidemiology, and End Results (SEER) data to evaluate the 20-year risk of BCSM and non-BCSM in men with stage I to III HR+ BC and identify factors associated with late mortality.
Methods
Data Source and Study Design
We conducted a cohort study and followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This study only used deidentified, previously collected, publicly available data. As such, it was exempt from review by the Dana-Farber Cancer Institute Office for Human Research Studies.
We obtained data from the SEER program, using the 18-registry database (November 2020 submission). We extracted all cases of men diagnosed with stage I to III HR+ BC between 1990 and 2008, ending in 2008, to have at least 10 years of potential follow-up (the November 2020 submission contains diagnoses up to December 2018). Additional inclusion criteria were known vital status and cause of death. Variables collected by the SEER program include patient demographic characteristics, primary tumor site, morphologic factors, stage at diagnosis, initial course of therapy, and survival follow-up. Information about estrogen receptor and progesterone receptor status is available from 1990, which was chosen as the starting year for our study. Status of ERBB2 has only been recorded in SEER since 2010, which means it was unavailable for patients in this study. We excluded patients with unknown HR status or HR-negative status, those who did not undergo surgery for their primary tumor, those with no or unknown regional nodal examination, or unknown number of pathologically involved axillary lymph nodes. Patients entered the study at the time of BC diagnosis (year 0). The inclusion and exclusion criteria of the cohort are depicted in eFigure 1 in Supplement 1.
The following variables were obtained from the SEER registries for analysis: age at diagnosis, race and ethnicity, year of diagnosis, histologic characteristics, tumor grade, tumor stage (including tumor size and nodal status), estrogen receptor status, progesterone receptor status, type of surgery, number of lymph nodes examined, number of pathologically involved lymph nodes, marital status, vital status, median income, and population size in area of residence. Tumor stage, size, and nodal status were reported in SEER according to the American Joint Committee on Cancer Cancer Staging Manual, sixth edition. To better understand the contribution of race and ethnicity, information on race and ethnicity was obtained from the SEER variable race and origin recode.
Statistical Analysis
The primary end point for the study was BCSM, defined as the interval from initial BC diagnosis to death from BC or last follow-up for censored patients. Deaths due to causes other than BC were classified as non-BCSM. We used cumulative incidence function to estimate the association between baseline clinical and pathologic variables, including tumor stage, tumor grade, and nodal status, and cumulative risks of BCSM, treating non-BCSM as a competing risk. We estimated the annual rate of BCSM events per 100 person-years. Cumulative incidence function was also used to estimate the risk of non-BCSM over time, treating BCSM as a competing risk.
Among those surviving at least 5 years since diagnosis, we performed Fine and Gray multivariable regression to evaluate the association of preselected variables with BCSM, treating non-BCSM as a competing risk. Prespecified variables included age at diagnosis, race and ethnicity, tumor stage, tumor grade, histologic characteristics, median income, and population size in the area of residence. We preselected these variables due to their relevance in BC. The rationale for excluding men with less than 5 years of survival from diagnosis in the Fine and Gray model was to evaluate factors associated with late BCSM (ie, BCSM beyond 5 years).
Smoothed annual hazard estimates for BCSM over time were calculated as a weighted kernel-density estimate using the estimated hazard contributions stratified by tumor stage or nodal status. All P values reported were 2-sided, and P values <.05 were considered statistically significant. We used Stata, version 12.0 (StataCorp LLC), and SPSS, version 26 (SPSS Inc) for statistical analyses.
Results
Patient and Disease Characteristics
The study included 2836 men who were diagnosed with stage I to III HR+ BC between 1990 and 2008. Baseline patient characteristics are reported in Table 1. Most patients were diagnosed with stage I (978 [34.5%]) or stage II (1304 [46.0%]) disease. The median age at diagnosis was 67 (IQR, 57-76) years, with 1602 (56.5%) patients diagnosed after age 64 years. Most patients identified as non-Hispanic White (2283 [80.5%]) and lived in a metropolitan area with more than 1 million inhabitants (1829 [64.5%]). Invasive ductal carcinoma was the most common histologic subtype (2620 [92.4%]), and most patients underwent mastectomy (2641 [93.1%]) for removal of the primary tumor.
Table 1. Baseline Patient and Disease Characteristics.
| Characteristic | Stage, No. (%) | |||
|---|---|---|---|---|
| I | II | III | Total | |
| All patients | 978 (34.5) | 1304 (46.0) | 554 (19.5) | 2836 (100.0) |
| Age at diagnosis, y | ||||
| <50 | 109 (11.1) | 122 (9.4) | 59 (10.6) | 290 (10.2) |
| 50-64 | 321 (32.8) | 437 (33.5) | 186 (33.6) | 944 (33.3) |
| >64 | 548 (56.0) | 745 (57.1) | 309 (55.8) | 1602 (56.5) |
| Race and ethnicity | ||||
| American Indian/Alaska Native | 1 (0.1) | 4 (0.3) | 1 (0.2) | 6 (0.2) |
| Asian or Pacific Islander | 48 (4.9) | 49 (3.8) | 19 (3.4) | 116 (4.1) |
| Hispanic (all races) | 41 (4.2) | 62 (4.8) | 34 (6.1) | 137 (4.8) |
| Non-Hispanic Black | 75 (7.7) | 148 (11.3) | 64 (11.6) | 287 (10.1) |
| Non-Hispanic White | 812 (83.0) | 1036 (79.4) | 435 (78.5) | 2283 (80.5) |
| Unknown | 1 (0.1) | 5 (0.4) | 1 (0.2) | 7 (0.2) |
| Grade | ||||
| I | 179 (18.3) | 118 (9.0) | 28 (5.1) | 325 (11.5) |
| II | 501 (51.2) | 656 (50.3) | 258 (46.6) | 1415 (49.9) |
| III/IV | 236 (24.1) | 469 (36.0) | 243 (43.9) | 948 (33.4) |
| Unknown | 62 (6.3) | 61 (4.7) | 25 (4.5) | 148 (5.2) |
| Histologic status | ||||
| Ductal | 891 (91.1) | 1209 (92.7) | 520 (93.9) | 2620 (92.4) |
| Other | 87 (8.9) | 95 (7.3) | 34 (6.1) | 216 (7.6) |
| Surgery | ||||
| Partial mastectomy | 93 (9.5) | 81 (6.2) | 18 (3.2) | 192 (6.8) |
| Mastectomy | 885 (90.5) | 1221 (93.6) | 535 (96.6) | 2641 (93.1) |
| Unknown | 0 | 2 (0.2) | 1 (0.2) | 3 (0.1) |
| Median household income, $ | ||||
| ≥75 000 | 375 (38.3) | 532 (40.8) | 185 (33.4) | 1092 (38.5) |
| 65 000-74 999 | 243 (24.8) | 321 (24.6) | 161 (29.1) | 725 (25.6) |
| 55 000-64 999 | 224 (22.9) | 281 (21.5) | 125 (22.6) | 630 (22.2) |
| 45 000-54 999 | 107 (10.9) | 124 (9.5) | 63 (11.4) | 294 (10.4) |
| 35 000-44 999 | 28 (2.9) | 35 (2.7) | 17 (3.1) | 80 (2.8) |
| <35 000 | 1 (0.1) | 10 (0.8) | 3 (0.5) | 14 (0.5) |
| Unknown | 0 | 1 (0.1) | 0 | 1 (0.0) |
| Rural-urban population | ||||
| Metropolitan ≥1 million | 639 (65.3) | 846 (64.9) | 344 (62.1) | 1829 (64.5) |
| Metropolitan 250 000-1 million | 159 (16.3) | 235 (18.0) | 106 (19.1) | 500 (17.6) |
| Metropolitan <250 000 | 62 (6.3) | 75 (5.8) | 41 (7.4) | 178 (6.3) |
| Nonmetropolitan adjacent to metropolitan | 63 (6.4) | 79 (6.1) | 31 (5.6) | 173 (6.1) |
| Nonmetropolitan not adjacent to metropolitan | 43 (4.4) | 59 (4.5) | 31 (5.6) | 133 (4.7) |
| Unknown | 12 (1.2) | 10 (0.8) | 1 (0.2) | 23 (0.8) |
| Vital status and cause of death | ||||
| Alive | 442 (45.2) | 438 (33.6) | 106 (19.1) | 986 (34.8) |
| Died of breast cancer | 102 (10.4) | 298 (22.9) | 234 (42.2) | 634 (22.4) |
| Died of other causes | 434 (44.4) | 568 (43.6) | 214 (38.6) | 1216 (42.9) |
Annual and Cumulative BCSM
After a median follow-up of 15.41 (IQR, 12.08-18.67) years, 986 patients (34.8%) were alive, 634 (22.4%) had died of BC, and 1216 (42.9%) had died of causes not related to BC (Table 1). eTable 1 in Supplement 1 reports the proportion of BCSM events during years 0 to less than 5, years 5 to less than 10, and 10 or more years by stage. Among patients with stage I disease, 22.6% of BCSM events occurred between years 0 to less than 5, and 50% of BCSM events occurred between years 5 to less than 10. Among patients with stage II disease, the proportion of BCSM events was almost 40% within each of the first and second 5-year periods from initial diagnosis and 23.5% after 10 years. In contrast, patients with stage III BC had nearly half of BCSM events within the first 5 years from diagnosis (49.2%).
Among all patients in the study cohort and from initial diagnosis, the annual rate of BCSM varied based on tumor stage, grade, and nodal status. Among stage I MaBC, the annual rate of BCSM ranged between 0.5% and 1.4% from 0 to 20 years after diagnosis, peaking during years 5 to 10. In these patients, the cumulative risk of BCSM by 20 years after diagnosis was 12.4%, compared with a cumulative risk of 52.4% for non-BCSM. For men with stage II disease, the annual rate of BCSM ranged between 1.9% and 2.9%, peaking during years 5 to 10 after diagnosis. This cohort had a cumulative 20-year risk of BCSM of 26.2%, compared with 51.2% for non-BCSM. For men with stage III disease, the annual rate of BCSM ranged between 5.1% and 7.3%, peaking during years 10 to 15. The cumulative 20-year risk of BCSM was 46.0%, compared with 42.8% for non-BCSM (Table 2) (Figure 1A).
Table 2. BCSM Annual Rate and Cumulative Risk by Stage, Nodal Status, and Tumor Grade, in the Entire Population and From Initial Diagnosis.
| Variable | No. | BCSM | Non-BCSM cumulative risk, % 0-20 y | ||||||
|---|---|---|---|---|---|---|---|---|---|
| % Event-free | Annual rate, % | Cumulative risk, % 0-20 y | |||||||
| 5 y | 10 y | 0-<5 y | 5-<10y | 10-<15 y | 15-20 y | ||||
| Stage | |||||||||
| I | 978 | 97.6 | 92.4 | 0.5 | 1.4 | 0.9 | 1.0 | 12.4 | 52.4 |
| II | 1304 | 91.3 | 82.3 | 1.9 | 2.9 | 2.7 | 2.6 | 26.2 | 51.2 |
| III | 554 | 79.1 | 65.6 | 5.1 | 5.8 | 7.3 | 4.8 | 46.0 | 42.8 |
| Nodal status | |||||||||
| N0 | 1538 | 95.8 | 90.4 | 0.9 | 1.6 | 1.4 | 1.1 | 15.2 | 55.5 |
| N1 | 868 | 89.5 | 78.1 | 2.4 | 3.7 | 3.1 | 3.5 | 33.3 | 44.5 |
| N2 | 270 | 82.5 | 69.9 | 4.2 | 5.1 | 5.3 | 4.5 | 40.6 | 46.3 |
| N3 | 160 | 69.8 | 52.6 | 7.6 | 8.5 | 14.3 | 5.2 | 59.8 | 32.9 |
| Tumor grade | |||||||||
| I | 325 | 97.2 | 92.6 | 0.6 | 1.2 | 1.5 | 1.1 | 12.6 | 54.1 |
| II | 1415 | 93.1 | 84.8 | 1.5 | 2.5 | 2.6 | 1.9 | 23.5 | 51.8 |
| III/IV | 948 | 86.5 | 76.0 | 3.1 | 3.8 | 2.7 | 2.4 | 32.4 | 47.3 |
Abbreviation: BCSM, breast cancer–specific mortality.
Figure 1. Cumulative Incidence of Breast Cancer–Specific Mortality (BCSM) by Tumor Stage and Nodal Status.
Similar patterns were observed for nodal status, with more pathologically involved nodes generally corresponding with a higher BCSM risk at all time points. Among men with N0 disease, the annual rate of BCSM ranged between 0.9% and 1.6%, peaking between years 5 and 10. The cumulative 20-year BCSM risk was 15.2%, compared with 55.5% for non-BCSM. In contrast, among men with N3 disease, the annual rate of BCSM ranged between 5.2% and 14.3%, peaking between years 10 and 15. The cumulative 20-year BCSM risk was 59.8%, compared with 32.9% for non-BCSM (Table 2) (Figure 1B). Similarly, higher tumor grade was associated with a higher rate of BCSM at all time points and a higher 20-year cumulative risk of BCSM (Table 2).
Among patients who had survived at least 5 years from diagnosis, Fine and Gray regression analyses revealed that age (<50 vs >64 years), tumor grade (grade II or grade III/IV vs grade I), tumor histologic characteristics (ductal vs other), and tumor stage (stage II or stage III vs stage I) were associated with a greater risk of BCSM (Table 3). A Fine and Gray regression model including the same variables but replacing tumor stage with tumor size and nodal status showed similar results and confirmed the interpretive value of these variables (eTable 2) in Supplement 1.
Table 3. Fine and Gray Regression for BCSM Starting at 5 Years After Diagnosis.
| Variable | P value | Hazard ratio (95% CI) |
|---|---|---|
| Age at diagnosis, y | ||
| <50 | NA | 1 [Reference] |
| 50-64 | .23 | 0.834 (0.622-1.120) |
| >64 | .002 | 0.627 (0.463-0.848) |
| Race and ethnicity | ||
| American Indian/Alaska Native | .91 | 1.127 (0.140-9.079) |
| Asian or Pacific Islander | .87 | 0.957 (0.565-1.622) |
| Hispanic (all races) | .28 | 1.258 (0.829-1.910) |
| Non-Hispanic White | NA | 1 [Reference] |
| Non-Hispanic Black | .20 | 1.236 (0.892-1.712) |
| Grade | ||
| I | NA | 1 [Reference] |
| II | .04 | 1.509 (1.013-2.249) |
| III/IV | .003 | 1.852 (1.228-2.793) |
| Unknown | .10 | 1.651 (0.913-2.988) |
| Histologic status | ||
| Ductal | NA | 1 [Reference] |
| Other | .01 | 0.536 (0.328-0.876) |
| Stage | ||
| I | NA | 1 [Reference] |
| II | <.001 | 2.015 (1.544-2.630) |
| III | <.001 | 3.935 (2.940-5.268) |
| Median income by county, $ | ||
| ≥75 000 | NA | 1 [Reference] |
| 65 000-74 999 | >.99 | 1.001 (0.775-1.292) |
| 55 000-64 999 | .25 | 0.848 (0.638-1.125) |
| 45 000-54 999 | .37 | 0.798 (0.488-1.304) |
| 35 000-44 999 | .58 | 0.814 (0.396-1.674) |
| <35 000 | .72 | 1.348 (0.266-6.824) |
| Rural-urban population | ||
| Metropolitan ≥1 million | NA | 1 [Reference] |
| Metropolitan 250 000-1 million | .21 | 0.835 (0.629-1.110) |
| Metropolitan <250 000 | .37 | 1.226 (0.789-1.907) |
| Nonmetropolitan adjacent to metropolitan | .60 | 0.865 (0.506-1.480) |
| Nonmetropolitan not adjacent to metropolitan | .85 | 0.939 (0.493-1.788) |
Abbreviations: BCSM, breast cancer–specific mortality; NA, not applicable.
Instantaneous Risk of BCSM
Smoothed hazard estimates for BCSM revealed distinct annual mortality patterns based on stage and nodal status. For stage I MaBC, the risk of BCSM peaked at approximately 7 years after diagnosis, and for stage II, the risk peaked at approximately 6 years after diagnosis, with minimal decline thereafter. In contrast, stage III MaBC experienced a bimodal distribution of BCSM hazard, which first peaked at approximately 4 years after diagnosis with a hazard rate of 6.3%, and then decreased slightly before reaching a higher peak at approximately 11 years with a hazard rate of 7.5% (Figure 2A).
Figure 2. Smoothed Annual Hazard Estimates of Breast Cancer–Specific Mortality (BCSM) by Tumor Stage and Nodal Status.
Shaded areas represent 95% CIs.
When analyzing the annual hazard estimates of BCSM by nodal status, the risk for men with N0 tumors peaked around year 7 after diagnosis and remained steady until approximately year 11 before decreasing slightly. For men with N1 tumors, the risk appeared bimodal, with small peaks at years 6 and 13 to 14 after diagnosis, with only a small reduction in between. For men with N2 tumors, the risk initially peaked at year 4 after diagnosis, decreasing slightly before climbing to a larger peak around year 14. The bimodal trend is most apparent in men with N3 tumors, in whom BCSM risk initially peaked at years 4 to 5 after diagnosis, with a hazard rate of 9.9%. The risk then decreased before rising to a higher peak around year 11 after diagnosis, with a hazard rate of 12.7% (Figure 2B). An exploratory ad hoc analysis was conducted within the group of men with N3 tumors to evaluate the annual hazard estimates of BCSM by tumor size and by grade. The results are shown in eFigure 2 in Supplement 1, which illustrate similar bimodal trends in most subgroups to those observed in the overall N3 population.
Discussion
This population-based study revealed that men with nonmetastatic HR+ BC face a long-term risk of BCSM for at least 20 years after diagnosis. Much like in women, the risk varies by tumor size, grade, and nodal status. Stage I HR+ MaBC have a cumulative incidence of BCSM at 20 years that is less than a quarter than the cumulative incidence of non-BCSM (12.4% vs 52.4%, respectively). Stage II MaBC had a cumulative incidence of BCSM that is roughly half the cumulative incidence of non-BCSM at 20 years after diagnosis (26.2% vs 51.2%, respectively). In contrast, Stage III MaBC had a 20-year cumulative incidence of BCSM and non-BCSM that was similarly high (46.0% vs 42.8%, respectively). Since the age distribution at time of diagnosis is approximately the same for patients with stage I, II, or III disease, this reflects the higher risk of BCSM for men with higher stage disease. Similar patterns are observed for tumor grade and nodal status, in which the 20-year cumulative risk of BCSM steadily narrowed in non-BCSM with higher grade or nodal status, but only eclipsed it among men with N3 disease.
Comparing the present results to an earlier study of women with HR+ BC reveals differences in the kinetics of the BCSM risk among patients with higher stage disease. In a previous study of women with stage III HR+ BC, Leone et al observed that the annual hazard rate of BCSM peaked at 6% approximately 5 years after diagnosis and then decreases over the next 20 years. In the present study, we observed a similar peak of approximately 6% around year 4 after diagnosis among men with stage III HR+ BC. However, after a small decline in risk, the annual hazard rate increased to a larger peak of approximately 7.5% at years 11 to 12 after diagnosis, before starting to diminish. The previous study in women with stage III HR+ BC who were diagnosed between 1990 and 2005 showed an annual hazard rate of 3.8% at years 11 to 12 after diagnosis. This annual hazard rate of 3.8% in that study falls below the lower limit of the 95% CI of the present study for the same time point. A similar finding was seen in the present study among men by nodal status, in which we observed that the late peak in the hazard of BCSM occurred across all nodal categories; furthermore, the late peak occurred earlier for each additional nodal category and had a larger hazard magnitude.
The reasons to explain the occurrence of this late peak in HR+ MaBC are unclear. Men have more limited evidence-based endocrine therapy options than women, with tamoxifen being the preferred option for men; this may play a role in the high risks of BCSM reported in our study overall. However, considering the timing of the late peaks and the fact that there is a reduction in the hazard estimates after the peaks are reached raises the issue as to whether endocrine therapy—or lack of adherence to it—would be a reason for the increase in hazard estimates so many years later. In fact, the recommended endocrine therapy duration is similar between men and women, and studies in women with HR+ BC have not shown the occurrence of a late peak in hazard rates, despite known endocrine therapy discontinuation and nonadherence. Other possible explanations may include differences in the tumor biologic factors between female HR+ BC and its male counterpart. Alternatively, differences in tumor dormancy between sexes or host-related factors, such as differences in tumor microenvironment, may play a role. The results of the ad hoc analysis within the N3 tumors are supportive of the hypothesis that indicates an escape from tumor dormancy as a combination of tumor biologic and host factors, with the limitation of small sample size within subgroups.
Among female patients with early-stage HR+/ERBB2− BC, clinicopathologic features, such as tumor size, grade, and nodal status, influence the risk of late BCSM. Using data from SEER, the ESTIMATE tool was developed to calculate the risk of BCSM, non-BSCM, and all-cause mortality for women with stage I to III HR+ BC within any time period up to 20 years from diagnosis. At that time, the paucity of data regarding the factors that influence risk of late BCSM in men with early-stage HR+ BC precluded the development of a similar tool for patients with MaBC, which may be possible now.
Strengths and Limitations
The major strength of the present study is the long-term follow-up. To our knowledge, this is the first population-based study to assess the 20-year risk of BCSM in nonmetastatic HR+ MaBC. The results suggest that patients with MaBC face a long-term risk of recurrence and death, which warrants further investigation into better adjuvant treatment options in men with HR+ disease.
This study has several limitations that are inherent to analyses of SEER data. First, SEER collects very limited data on systemic therapies, which are not publicly available. As such, we do not have data on the use of systemic therapies in this cohort. Prior studies using the National BC Database have suggested that patients with HR+ MaBC have been historically undertreated with adjuvant endocrine therapy. Similarly, an international study revealed concerns about undertreatment with endocrine therapy in men. Part of this undertreatment is due to endocrine therapy intolerance and nonadherence. As such, it is possible that the risks of BCSM for men who complete appropriate adjuvant endocrine therapy may be lower than those reported in our study. The lack of information about disease recurrence is another limitation when using SEER data that also applies to the present study. Overall survival after a diagnosis of metastatic BC has improved with the initiation of CDK4/6 inhibitors, and in addition, these agents are being used in the adjuvant setting in high-risk patients without long-term data available. Therefore, it is possible that the absolute estimates of BCSM may be lower for patients diagnosed with BC today compared with the estimates from our study.
Our study was also limited by lack of ERBB2 status. SEER began collecting information on ERBB2 status in 2010, which means we would need to wait until at least 2030 to assess 20-year survival in ERBB2+ MaBC. However, given that the HR+/ERBB2+ BC subtype represents approximately 12% of BCs in men, it is unlikely that the inclusion of HR+/ERBB2+ cases in our study would have an association with the reported estimates. In addition, our study included relatively small sample sizes of male patients with stage III and N3 disease, which resulted in wide CIs around the smoothed hazard estimates of BCSM over time. Therefore, the results for those subgroups should be interpreted with caution.
Conclusions
Historically, patients with MaBC have been underrepresented in randomized clinical trials, whether through explicit exclusion or lack of enrollment in trials that technically allowed men. As a result, treatment guidelines and counseling for men with BC have largely been derived from clinical trials in female patients, with supporting data generated by observational and population-based studies in men. In this regard, the results from our cohort study show notable findings on the late risk of BCSM in men, which should not be extrapolated from women when counseling patients.
eTable 1. Proportion of BCSM Events by Stage
eTable 2. Fine and Gray Regression for BCSM Starting at 5 Years After Diagnosis, Including Tumor Size and Nodal Status in the Model
eFigure 1. Flow Diagram of the Patient Population
eFigure 2. Smoothed Annual Hazard of BCSM Among Men With N3 Breast Cancer by Tumor Size (eFigure 2A) and Tumor Grade (eFigure 2B)
Data Sharing Statement
References
- 1.Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209-249. doi: 10.3322/caac.21660 [DOI] [PubMed] [Google Scholar]
- 2.Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin. 2023;73(1):17-48. doi: 10.3322/caac.21763 [DOI] [PubMed] [Google Scholar]
- 3.Giordano SH, Cohen DS, Buzdar AU, Perkins G, Hortobagyi GN. Breast carcinoma in men: a population-based study. Cancer. 2004;101(1):51-57. doi: 10.1002/cncr.20312 [DOI] [PubMed] [Google Scholar]
- 4.Leone J, Freedman RA, Lin NU, et al. Tumor subtypes and survival in male breast cancer. Breast Cancer Res Treat. 2021;188(3):695-702. doi: 10.1007/s10549-021-06182-y [DOI] [PubMed] [Google Scholar]
- 5.Vermeulen MA, Slaets L, Cardoso F, et al. Pathological characterisation of male breast cancer: results of the EORTC 10085/TBCRC/BIG/NABCG International Male Breast Cancer Program. Eur J Cancer. 2017;82:219-227. doi: 10.1016/j.ejca.2017.01.034 [DOI] [PubMed] [Google Scholar]
- 6.Cardoso F, Bartlett JMS, Slaets L, et al. Characterization of male breast cancer: results of the EORTC 10085/TBCRC/BIG/NABCG International Male Breast Cancer Program. Ann Oncol. 2018;29(2):405-417. doi: 10.1093/annonc/mdx651 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yadav S, Karam D, Bin Riaz I, et al. Male breast cancer in the United States: treatment patterns and prognostic factors in the 21st century. Cancer. 2020;126(1):26-36. doi: 10.1002/cncr.32472 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Greif JM, Pezzi CM, Klimberg VS, Bailey L, Zuraek M. Gender differences in breast cancer: analysis of 13,000 breast cancers in men from the National Cancer Data Base. Ann Surg Oncol. 2012;19(10):3199-3204. doi: 10.1245/s10434-012-2479-z [DOI] [PubMed] [Google Scholar]
- 9.Pan H, Gray R, Braybrooke J, et al. ; EBCTCG . 20-Year risks of breast-cancer recurrence after stopping endocrine therapy at 5 years. N Engl J Med. 2017;377(19):1836-1846. doi: 10.1056/NEJMoa1701830 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Leone JP, Vallejo CT, Hassett MJ, et al. Factors associated with late risks of breast cancer-specific mortality in the SEER registry. Breast Cancer Res Treat. 2021;189(1):203-212. doi: 10.1007/s10549-021-06233-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.McKinley N, McCain S, Kirk S. Long term follow up of male breast cancer. Ulster Med J. 2017;86(3):177-180. [PMC free article] [PubMed] [Google Scholar]
- 12.Yu XF, Yang HJ, Yu Y, Zou DH, Miao LL. A prognostic analysis of male breast cancer (MBC) compared with post-menopausal female breast cancer (FBC). PLoS One. 2015;10(8):e0136670. doi: 10.1371/journal.pone.0136670 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yu E, Stitt L, Vujovic O, et al. Male breast cancer prognostic factors versus female counterparts with propensity scores and matched-pair analysis. Cureus. 2015;7(10):e355. doi: 10.7759/cureus.355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Surveillance, Epidemiology, and End Results (SEER) Program. SEER*Stat Database: incidence, SEER research data, 18 registries, November 2020 sub (2000-2018), linked to county attributes, time dependent (1990-2018) income/rurality, 1969-2018 counties, National Cancer Institute, DCCPS, Surveillance Research Program, released April 2021, based on the November 2020 submission. Accessed January 25, 2024. http://www.seer.cancer.gov
- 15.AJCC Cancer Staging Manual. 6th ed. AJCC; 2002. [Google Scholar]
- 16.Leone JP, Leone BA, Tayob N, et al. Twenty-year risks of breast cancers–specific mortality for stage III breast cancer in the surveillance, epidemiology, and end results registry. Breast Cancer Res Treat. 2021;187(3):843-852. doi: 10.1007/s10549-021-06121-x [DOI] [PubMed] [Google Scholar]
- 17.Colleoni M, Sun Z, Price KN, et al. Annual hazard rates of recurrence for breast cancer during 24 years of follow-up: results from the International Breast Cancer Study Group trials I to V. J Clin Oncol. 2016;34(9):927-935. doi: 10.1200/JCO.2015.62.3504 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chirgwin JH, Giobbie-Hurder A, Coates AS, et al. Treatment adherence and its impact on disease-free survival in the Breast International Group 1-98 trial of tamoxifen and letrozole, alone and in sequence. J Clin Oncol. 2016;34(21):2452-2459. doi: 10.1200/JCO.2015.63.8619 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hershman DL, Kushi LH, Shao T, et al. Early discontinuation and nonadherence to adjuvant hormonal therapy in a cohort of 8,769 early-stage breast cancer patients. J Clin Oncol. 2010;28(27):4120-4128. doi: 10.1200/JCO.2009.25.9655 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Leone JP, Graham N, Tolaney SM, et al. Estimating long-term mortality in women with hormone receptor-positive breast cancer: The “ESTIMATE” tool. Eur J Cancer. 2022;173:20-29. doi: 10.1016/j.ejca.2022.06.029 [DOI] [PubMed] [Google Scholar]
- 21.Pemmaraju N, Munsell MF, Hortobagyi GN, Giordano SH. Retrospective review of male breast cancer patients: analysis of tamoxifen-related side-effects. Ann Oncol. 2012;23(6):1471-1474. doi: 10.1093/annonc/mdr459 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Oke O, Niu J, Chavez-MacGregor M, Zhao H, Giordano SH. Adjuvant tamoxifen adherence in men with early-stage breast cancer. Cancer. 2022;128(1):59-64. doi: 10.1002/cncr.33899 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Leone JP, Leone J, Zwenger AO, Iturbe J, Vallejo CT, Leone BA. Prognostic significance of tumor subtypes in male breast cancer: a population-based study. Breast Cancer Res Treat. 2015;152(3):601-609. doi: 10.1007/s10549-015-3488-y [DOI] [PubMed] [Google Scholar]
- 24.Corrigan KL, Mainwaring W, Miller AB, et al. Exclusion of men from randomized phase III breast cancer clinical trials. Oncologist. 2020;25(6):e990-e992. doi: 10.1634/theoncologist.2019-0871 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Duma N, Hoversten KP, Ruddy KJ. Exclusion of male patients in breast cancer clinical trials. J Natl Cancer Inst Cancer Spectr. 2018;2(2):pky018. doi: 10.1093/jncics/pky018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hassett MJ, Somerfield MR, Baker ER, et al. Management of male breast cancer: ASCO guideline. J Clin Oncol. 2020;38(16):1849-1863. doi: 10.1200/JCO.19.03120 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Proportion of BCSM Events by Stage
eTable 2. Fine and Gray Regression for BCSM Starting at 5 Years After Diagnosis, Including Tumor Size and Nodal Status in the Model
eFigure 1. Flow Diagram of the Patient Population
eFigure 2. Smoothed Annual Hazard of BCSM Among Men With N3 Breast Cancer by Tumor Size (eFigure 2A) and Tumor Grade (eFigure 2B)
Data Sharing Statement
