Sex Disparities in Breast Cancer Survival According to Clinical Treatment Score Post-5 Years (CTS5) Risk Stratification

Abstract

Purpose

The role of the Clinical Treatment Score post-5 years (CTS5) in male breast cancer (MBC) remains unclear. This study aimed to investigate the characteristics and prognosis of CTS5 between MBC and female breast cancer (FBC).

Materials and Methods

Patients diagnosed with human epidermal growth factor receptor 2–negative/estrogen receptor–positive breast cancer between 2010 and 2015 were identified from the Surveillance, Epidemiology, and End Results (SEER) database. The chi-square test, Kaplan-Meier analysis, and multivariate Cox proportional hazard model were used for statistical analysis.

Results

This study included 169,869 patients, comprising 168,422 (99.1%) FBC and 1,447 (0.9%) MBC patients. More MBC patients had intermediate risk (IR)/high risk (HR) disease compared to FBC patients (17.7% vs. 8.7%, p < 0.001). MBC patients had an inferior overall survival (OS) but similar breast cancer-specific survival compared to those with FBC. Sensitivity analyses showed that sex was an independent prognostic factor associated with OS but not breast cancer-specific survival (BCSS) in both the low-risk (LR) and IR/HR cohorts. Those with MBC exhibited significantly worse OS (p < 0.001) than FBC patients in both cohorts. In MBC patients, those with IR/HR disease had significantly worse OS (p < 0.001) and BCSS (p < 0.001) compared to those with LR disease. For FBC patients, the IR/HR group had also significantly worse OS (p < 0.001) and BCSS (p < 0.001) compared to the LR group.

Conclusion

Our findings highlight critical differences in clinical characteristics, treatment patterns, and outcomes between MBC and FBC, suggesting the need for sex-tailored approaches in breast cancer management.

Introduction

Historically overlooked, male breast cancer (MBC) has a significantly lower incidence compared to its female counterpart. MBC represents less than 1% of all breast cancers (BC), with approximately 2,000 new cases diagnosed annually in the United States [1-3]. However, heightened awareness should be prompted for a gradual increase in the incidence of MBC. Risk factors such as genetic mutations, chest radiation, and unhealthy habits (obesity, alcohol use) contribute to the development of MBC [4,5]. Regrettably, due to insufficient vigilance, screening, and preventive measures, MBC is often diagnosed at an advanced stage, which may be due to delayed diagnosis and treatment (29% vs. 58%) than their female counterparts within three months of symptom onset [6]. Treatment decisions for MBC often rely on extrapolations from extensive clinical experience with female breast cancer (FBC) patients, due to limited direct evidence. However, these treatments may not be optimal for MBC. For instance, a higher proportion of female patients prefer breast-conserving surgery, but many male patients need to opt for radical mastectomy [7,8]. This preference stems from female patients’ greater concern about cosmetic outcomes post-surgery and early BC screening. Sex, endocrine profiles, and clinicopathological features profoundly impact survival prognosis and treatment outcomes. Stage- and subtype-matched MBC patients generally experience inferior overall survival (OS) compared to their FBC [9,10]. Due to limited data availability, crucial intrinsic biological and clinical characteristics of MBC remain unclear. There is an urgent need for diagnostic indicators to assist clinicians in better assessing the severity and prognosis of MBC.

The Clinical Treatment Score post-5 years (CTS5) is an algorithm designed to predict the risk of late distant recurrence in female patients, utilizing clinicopathological factors such as age, histological grade, tumor size, and positive lymph nodes. Several studies have demonstrated that CTS5 provides excellent prognostic insight and clinical utility in guiding treatment decisions for human epidermal growth factor receptor 2–negative (HER2–)/estrogen receptor–positive (ER+) FBC. FBC patients with low CTS5 scores may potentially avoid unnecessary prolonged endocrine therapy [11]. CTS5 proves valuable in assessing the risk of distant recurrence after 5 years of adjuvant endocrine therapy in FBC [12,13]. Importantly, there is currently no report on the clinical utility of CTS5 in MBC. This study aimed to investigate the characteristics and prognosis of CTS5 between MBC and FBC.

Materials and Methods

1. Patients

We identified patients who were diagnosed between 2010 and 2015 using the data from the Surveillance, Epidemiology, and End Results (SEER) program. Eligible patients met the following criteria: (1) diagnosed with FBC or MBC; (2) ER+ and HER2– disease; (3) available data regarding age, race, tumor (T) category, nodal (N) category, grade, histology, progesterone receptor (PR) status, and treatment information included surgery, chemotherapy, and radiotherapy. Patients with distant metastasis at diagnosis or second primary cancer were excluded. This study is exempt from institutional review board approval as SEER database patient data are de-identified.

2. Measures

We utilized the ATAC (Arimidex, Tamoxifen, Alone or in Combination) and BIG 1-98 datasets to develop a prognostic score model CTS5, for women with BC [14]. This CTS5 model incorporates patients’ age at diagnosis, tumor size, and nodal status (categorized into five groups: 0, negative; 1, one positive node; 2, two to three positive nodes; 3, four to nine positive nodes; and 4, more than nine positive nodes), and tumor grade (categorized into three groups: 1 for grade I; 2 for grade II; and 3 for grade III).

The formula for CTS5 is as follows: CTS5=0.471×nodes+0.980×(0.164×size–0.003×size²+0.312×histology grade+0.03×age at diagnosis) [14].

Patients were classified into three risk groups based on their CTS5 score using an optimized cutoff point: low risk (LR; CTS5 < 3.13), intermediate risk (IR; 3.13-3.86), and high risk (HR; CTS5 > 3.86) [14]. Data analyzed included tumor size, nodal stage, local and systemic treatments, PR status, histology grade, pathological type, age at diagnosis, and race. OS and breast cancer-specific survival (BCSS) were the endpoints of this study.

3. Statistical analysis

Baseline characteristics were compared between MBC and FBC patients using the chi-square test. A similar comparison across CTS risk categories was conducted for MBC. Survival outcomes were calculated using the Kaplan-Meier method and compared using the log-rank test. A multivariate Cox proportional hazards model was employed to assess independent prognostic factors associated with OS and BCSS. Sensitivity analyses were performed by fitting individual models for each independent variable through the application of the multivariate Cox proportional hazards model. Statistical analyses were performed using IBM SPSS ver. 26.0 (IBM Corp.), with a significance set at p < 0.05.

Results

1. Patient characteristics

This study included a total of 169,869 patients, comprising 168,422 (99.1%) FBC and 1,447 (0.9%) MBC patients (Table 1). MBC patients were diagnosed at a significantly older age compared to FBC patients (median age, 68.0 vs. 62.0; p < 0.001). Male patients presented with more advanced T category (p < 0.001), higher tumor grade (p < 0.001), and more advanced N category (p < 0.001). Infiltrating ductal carcinoma was the predominant pathological subtype among MBC patients, with a lower incidence of infiltrating lobular carcinoma compared to FBC patients (p < 0.001). In addition, MBC patients exhibited a higher prevalence of PR positivity (94.4% vs. 88.4%, p < 0.001).

Table 1.

Clinicopathological and CTS5 characteristics of the FBC and MBC patients

Variable	Total	Female	Male	p-value
Age (yr)
1-64	96,424	95,875 (56.9)	549 (37.9)	< 0.001
≥ 65	73,445	72,547 (43.1)	898 (62.1)
Tumor stage
T1	114,254	113,511 (67.4)	743 (51.3)	< 0.001
T2	45,553	44,964 (26.7)	589 (40.7)
T3	7,880	7,845 (4.7)	35 (2.4)
T4	2,182	2,102 (1.2)	80 (5.5)
Tumor grade
Grade I	52,411	52,199 (31.0)	212 (14.7)	< 0.001
Grade II	85,690	84,896 (50.4)	794 (54.9)
Grade III	31,768	31,327 (18.6)	441 (30.5)
Nodal stage
N0	127,558	126,644 (75.2)	914 (63.2)	< 0.001
N1	29,793	29,454 (17.5)	339 (23.4)
N2	8,411	8,279 (4.9)	132 (9.1)
N3	4,107	4,045 (2.4)	62 (4.3)
Race
White	138,931	137,734 (81.8)	1,197 (82.7)	< 0.001
Black	13,780	13,607 (8.1)	173 (12.0)
Other	17,158	17,081 (10.1)	77 (5.3)
Pathology
Infiltrating duct carcinoma	121,175	119,965 (71.2)	1,210 (83.6)	< 0.001
Infiltrating lobular carcinoma	20,119	20,107 (11.9)	12 (0.8)
Other	28,575	28,350 (16.8)	225 (15.5)
PR status
Negative	19,642	19,560 (11.6)	82 (5.7)	< 0.001
Positive	150,227	148,862 (88.4)	1,365 (94.3)
CTS5 risk
LR	154,981	153,790 (91.3)	1,191 (82.3)	< 0.001
IR/HR	14,888	14,621 (8.7)	256 (17.7)
Surgery procedure
Breast conservative surgery	96,350	96,232 (57.1)	118 (8.2)	< 0.001
Mastectomy	73,519	72,190 (42.9)	1,329 (91.8)
Radiotherapy
No	76,262	75,236 (44.7)	1,026 (70.9)	< 0.001
Yes	93,607	93,186 (55.3)	421 (29.1)
Chemotherapy
No	121,196	120,237 (71.4)	959 (66.3)	< 0.001
Yes	48,673	48,185 (28.6)	488 (33.7)

Values are presented as number (%). CTS5, Clinical Treatment Score Post-5 Years; FBC, female breast cancer; IR/HR, intermediate- and high-risk; LR, low-risk; MBC, male breast cancer; N, nodal; PR, progesterone receptor; T, tumor.

Within the FBC cohort (n=168,422), 153,790 (91.3%) were classified as LR, 10,978 (6.5%) as IR, and 3,665 (2.2%) as HR. MBC patients were more likely to undergo mastectomy (p < 0.001) and chemotherapy (p < 0.001), while less likely to receive postoperative radiotherapy (p < 0.001).

In the MBC cohort (n=1,447), 1,191 (82.3%) were classified as LR, 186 (12.9%) as IR, and 70 (4.8%) as HR. A significant difference in the distribution of CTS classifications was observed between FBC and MBC patients (p < 0.001). IR and HR categories were combined for analysis due to the relatively small number of HR cases, particularly among MBC patients. More MBC patients had IR/HR disease compared to FBC patients (17.7% vs. 8.7%, p < 0.001).

2. Survival analysis

The median follow-up for the entire cohort was 75 months (range, 0 to 119 months). Kaplan-Meier analysis was used to assess survival differences between MBC and FBC patients across various CTS5 risk categories. MBC patients exhibited significantly worse 5-year OS (78.1% vs. 92.7%, p < 0.001) (Fig. 1A) and BCSS (93.6% vs. 96.4%, p < 0.001) (Fig. 1B) compared to FBC patients.

Fig. 1.

Comparison of overall survival (A) and breast cancer-specific survival (B) between those with female breast cancer and male breast cancer.

Analyzing the prognostic impact of CTS5, MBC patients with IR/HR categorization had significantly poorer 5-year OS and BCSS compared to those classified as LR (5-year OS, LR 80.5% vs. IR/HR 67.0%; p < 0.001 and 5-year BCSS, LR 95.2% vs. IR/HR 81.1%; p < 0.001) (Fig. 2A and B). Similarly, FBC patients in the IR/HR group had worse OS and BCSS than those in the LR group (5-year overall survival, LR 93.2% vs. IR/HR 72.5%; p < 0.001 and 5-year BCSS, LR 98.5% vs. IR/HR 81.3%; p < 0.001) (Fig. 2C and D).

Fig. 2.

Comparison of overall survival (A, male breast cancer; C, female breast cancer) and breast cancer-specific survival (B, male breast cancer; D, female breast cancer) between low-risk (LR) and intermediate- and high-risk (IR/HR) cohorts after stratification by sex.

After stratification by CTS5 classifications, MBC patients demonstrated significantly worse survival outcomes compared to FBC patients in both the LR and IR/HR groups. For OS, MBC patients in the LR and IR/HR groups had significantly lower survival rates compared to FBC patients (LR, 5-year OS 80.5% vs. 93.1%; p < 0.001 and IR/HR, 5-year OS 67.0% vs. 72.5%; p=0.004) (Fig. 3A and B). In terms of BCSS, MBC patients showed poorer BCSS than FBC patients in the LR group (5-year BCSS, 96.2% vs. 97.8%; p=0.010) (Fig. 3C), while BCSS outcomes were similar between MBC and FBC patients in the IR/HR groups (5-year BCSS, 81.3% vs. 81.1%; p=0.830) (Fig. 3D).

Fig. 3.

Comparison of overall survival (A, low-risk [LR]; C, intermediate- and high-risk [IR/HR]) and breast cancer-specific survival (B, LR; D, IR/HR) between female breast cancer and male breast cancer patients after stratification by risk classifications.

3. Prognostic analysis

A multivariate Cox proportional hazards model was employed to identify independent prognostic factors for OS and BCSS (Table 2). Sex was an independent prognostic factor for OS but not for BCSS. MBC was associated with significantly worse OS (hazard ratio [HaR], 1.729; 95% confidence interval [CI], 1.569 to 1.904; p < 0.001) and similar BCSS (HaR, 1.057; 95% CI, 0.873 to 1.281; p=0.569) compared to FBC. CTS5 classification was also an independent prognostic factor for both OS and BCSS. Patients in the IR/HR cohort had significantly worse OS (HaR, 4.278; 95% CI, 4.134 to 4.428; p < 0.001) and BCSS (HaR, 5.734; 95% CI, 5.451 to 6.032; p < 0.001) compared to the LR cohort. Additional independent prognostic factors included age at diagnosis, histology, race, PR status, surgery procedure, radiotherapy, and chemotherapy.

Table 2.

Multivariate prognostic analysis for independent prognostic factors associated with BCSS and OS in the entire cohort

Variable	BCSS			OS
	HaR	95% CI	p-value	HaR	95% CI	p-value
Sex
Female	1			1
Male	1.057	0.873-1.281	0.569	1.729	1.569-1.904	< 0.001
Race
White	1			1
Black	1.518	1.427-1.615	< 0.001	1.287	1.235-1.340	< 0.001
Other	0.796	0.735-0.861	< 0.001	0.658	0.625-0.692	< 0.001
Pathology
Infiltrating duct carcinoma	1			1
Infiltrating lobular carcinoma	1.052	0.990-1.117	0.099	1.013	0.975-1.052	0.518
Other	0.865	0.814-0.920	< 0.001	0.986	0.953-1.020	0.412
PR status
Negative	1			1
Positive	0.506	0.481-0.532	< 0.001	0.647	0.625-0.669	< 0.001
CTS5 risk
LR	1			1
IR/HR	5.734	5.451-6.032	< 0.001	4.278	4.134-4.428	< 0.001
Surgery
Breast conservative surgery	1			1
Mastectomy	1.456	1.381-1.534	< 0.001	1.035	1.003-1.068	0.030
Radiotherapy
No	1			1
Yes	0.743	0.707-0.780	< 0.001	0.622	0.603-0.641	< 0.001
Chemotherapy
No	1			1
Yes	1.533	1.459-1.610	< 0.001	0.672	0.651-0.693	< 0.001

BCSS, breast cancer-specific survival; CI, confidence interval; CTS5, Clinical Treatment Score Post-5 Years; HaR, hazard ratio; IR/HR, intermediate- and high-risk; LR, low-risk; N, nodal; OS, overall survival; PR, progesterone receptor.

Sensitivity analyses were performed to determine the effect of sex on survival outcomes after stratification by CTS5 classifications (Table 3). The results showed that sex was an independent prognostic factor associated with OS but not BCSS in both the LR and IR/HR cohorts. Those with MBC exhibited significantly worse OS compared to FBC patients both in the LR (HaR, 1.989; 95% CI, 1.775 to 2.228; p < 0.001) and IR/HR (HaR, 1.237; 95% CI, 1.028 to 1.489; p=0.025) cohorts. However, sex was not associated with BCSS in both cohorts.

Table 3.

Sensitivity analyses

Variable	BCSS			OS
	HaR	95% CI	p-value	HaR	95% CI	p-value
LR cohort
Male vs. Female	1.116	0.820-1.464	0.430	1.989	1.775-2.228	< 0.001
IR/HR cohort
Male vs. Female	0.983	0.750-1.289	0.903	1.237	1.028-1.489	0.025
Female cohort
IR/HR vs. LR	5.749	5.464-6.050	< 0.001	4.317	4.169-4.469	< 0.001
Male cohort
IR/HR vs. LR	4.489	2.950-6.831	< 0.001	2.512	1.988-3.174	< 0.001

BCSS, breast cancer-specific survival; CI, confidence interval; HaR, hazard ratio; IR/HR, intermediate- and high-risk; LR, low-risk; OS, overall survival.

Further sensitivity analyses assessed the impact of CTS5 classification on survival outcomes after stratification by sex (Table 3). In MBC patients, those with IR/HR disease had significantly worse OS (HaR, 2.512; 95% CI, 1.988 to 3.174; p < 0.001) and BCSS (HaR, 4.489; 95% CI, 2.950 to 6.831; p < 0.001) compared to those with LR disease. For FBC patients, the IR/HR group also had significantly worse OS (HaR, 4.317; 95% CI, 4.169 to 4.469; p < 0.001) and BCSS (HaR, 5.749; 95% CI, 5.464 to 6.050; p < 0.001) compared to the LR group.

Discussion

This study offers a comprehensive investigation into the characteristics and prognostic implications of the CTS5 score in MBC compared to FBC. Our findings reveal significant differences in patient demographics, tumor characteristics, and survival outcomes between MBC and FBC, highlighting the need for tailored clinical approaches and underscoring the distinct biological and clinical nature of BC in male patients.

Our patient demographics align with those reported in a meta-analysis involving 1,826 MBC and 174,512 FBC patients [15]. They showed that MBC patients had more advanced tumor stage and histologic grade than FBC patients. The analysis indicated that MBC patients tend to present with more advanced tumor stages and histologic grades than FBC patients. The median age in our MBC cohort was 68 years, consistent with previous studies showing that MBC is typically diagnosed 5-10 years later than FBC in the United States. This age difference may be attributed to lower awareness and potential delays in diagnosis due to the absence of screening programs for MBC [16]. Historical data from the SEER database indicate that 92% of 5,494 MBC and 78% of 838,805 FBC were ER-positive [17]. Our study cohort, exclusively HER2-negative and ER-positive, showed a high prevalence of PR positivity (94.4%), reflecting the sample’s representativeness of the broader population. Additionally, MBC patients were more likely to present with advanced T and N categories and higher tumor grades, suggesting a more aggressive clinical course, possibly due to biological differences or delayed diagnosis.

The presentation of MBC is similar to that of postmenopausal. The management of MBC is largely derived from the clinical management experiences of FBC. MBC patients generally have a poorer prognosis compared to FBC patients, largely due to the advanced stage of disease at diagnosis and fewer active treatment options. There are no randomized trials or large-scale data to guide appropriate treatment for MBC specifically. Therefore, developing scientific treatment guidelines for MBC patients in clinical practice is critical. Understanding the prognosis of MBC is essential for guiding clinical decisions and improving treatment outcomes. Research consistently highlights a generally poor prognosis for MBC, largely attributable to late-stage diagnoses and factors such as age, tumor size, stage, metastasis, and hormone receptor status. Longitudinal data demonstrate gradual improvements in the 5-year survival rates of MBC over time, emphasizing the necessity for continuous monitoring and personalized treatment approaches [18,19]. Prognostic tools like nomograms have shown promise in predicting real-time survival probabilities and identifying high-risk individuals who might benefit from aggressive treatments. While sentinel lymph node biopsy has shown promise in small studies, randomized trials evaluating its sensitivity and specificity in MBC are lacking [20].

CTS5 represents a valuable tool for delivering personalized treatment to MBC patients by analyzing clinical and pathological features alongside prognostic factors, thereby enhancing OS prospects. Our study demonstrated that MBC patients experience significantly worse 5-year OS compared to their FBC counterparts. Kaplan-Meier analysis confirmed inferior survival in MBC across all CTS5 risk stratifications, particularly in those with IR/HR status. This highlights the potential for CTS5 to effectively stratify risk but also underscores the need for its validation and potential adjustment in MBC populations. The poorer survival in MBC, despite similar CTS5 stratification to FBC, suggests an inherently more aggressive biological course or different interactions with therapeutic interventions. While our study population was predominantly ER-positive, the inferior OS observed in MBC patients, combined with their higher proportion of IR and HR disease, suggests a more aggressive biological behavior compared to FBC. This is supported by previous evidence indicating that MBC patients often present with more advanced disease characteristics, such as larger tumor sizes and higher-grade tumors [21-23]. However, the ER-positive nature of our cohort may limit the generalizability of these findings, and further research is needed to elucidate the underlying biological mechanisms driving these differences. The independent prognostic value of CTS5 for OS and BCSS was validated in our cohort, supporting its use in clinical settings for risk assessment. However, the divergence in survival outcomes between sexes, even within the same CTS5 categories, indicates that additional sex-specific prognostic markers could enhance its predictive accuracy for MBC.

Notably, in our study, 91.8% of MBC patients underwent mastectomy, whereas only 29.1% received radiotherapy. This discrepancy may be due to the clinical equivalence between breast-conserving surgery plus radiotherapy and radical mastectomy, along with a lesser emphasis on breast retention among MBC patients. Achieving negative surgical margins in MBC breast surgery can be challenging, leading to a high rate of early radical mastectomy [24]. Post-mastectomy radiotherapy (PMRT) should be considered for tumors larger than 5 cm with persistently positive surgical margins, node-positive tumors, lymphovascular invasion, and peripheral nerve invasion [25]. However, assessing the specific benefits of PMRT for MBC patients remains complex. A meta-analysis involving 3,912 MBC patients suggested that PMRT significantly reduces the risk of local recurrence and may lower BC-related mortality, particularly in advanced stages [26]. Previous studies found that 61.3% of MBC patients treated with postoperative radiotherapy experienced improved locoregional recurrence-free survival, although there was no significant impact on OS, potentially due to limited patient numbers [27]. These findings underscore the potential benefit of PMRT in MBC BC and advocate its inclusion in treatment strategies to improve clinical outcomes and mitigate disease recurrence risks.

Unlike the well-established chemotherapy regimens for FBC, specific treatment protocols for MBC patients lack robust validation. A small prospective study of 11 MBC with operable stage II and stage III BC who received adjuvant chemotherapy following local treatment reported recurrence in four patients, while seven remained disease-free. Among those with recurrence, one succumbed to metastatic BC. These limited data suggest that adjuvant chemotherapy reduces recurrence risk and positively impacts survival in this population [28]. Chemotherapy should be considered for MBC patients. In our study, a higher proportion of MBC received chemotherapy, likely due to more advanced T stages and higher tumor grades among MBC patients. Chemotherapy emerged as an independent prognostic factor for OS in both MBC and FBC cohorts, consistent with previous research indicating a 26% reduction in all-cause mortality among MBC receiving chemotherapy [29].

The increased rate of chemotherapy among MBC aligns with their higher classification into high-risk CTS5 categories. Previous studies have emphasized that high-risk CTS5 scores are significantly correlated with poorer survival outcomes in HER2-negative women, highlighting its independent prognostic value [13]. Our study extends this understanding to MBC, providing valuable insights into prognostic assessments. IR/HR cohort was associated with significantly poorer survival outcomes compared to the LR cohort, thus refining predictive tools based on conventional clinical and pathological features for MBC. Previous research has shown that FBC patients at high CTS5 risk should receive extended endocrine therapy for five years to improve survival prognosis. Likewise, MBC with higher CTS5 scores may benefit from more aggressive treatment regimens.

Disparities in OS and BCSS between MBC and FBC patients were primarily observed in those classified under the LR stratification. The CTS5 algorithm incorporates factors such as age, tumor size, number of positive lymph nodes, and grade, which means that patients with similar risk stratifications exhibit comparable basic pathological features across sexes. However, differences in treatment efficacy among LR patients may contribute to these disparities. Applying treatment strategies indiscriminately from female patients to guide MBC patients with LR could compromise prognosis, underscoring the need for tailored approaches.

However, our study has limitations, including its retrospective nature and the comparatively smaller sample size of MBC patients, which could limit its generalizability. Second, details of endocrine therapy, recurrence, and metastasis patterns after treatment were also not included in the SEER database. Future prospective studies with expanded clinical cohorts are essential to validate the prognostic implications of different CTS5 risk categories in MBC patients. Despite these limitations, our study validates the clinical utility of CTS5 and expands its role as a prognostic indicator for MBC patients in clinical practice.

In conclusion, our findings highlight critical differences in clinical characteristics, treatment patterns, and outcomes between MBC and FBC, suggesting the need for sex-tailored approaches in BC management. Developing a deeper understanding of MBC through targeted research will be pivotal in improving prognosis and optimizing therapeutic strategies for MBC patients.