A comparative analysis of radiotherapy use and patient outcome in males and females with breast cancer

G. Macdonald1,*, C. Paltiel2, I. A. Olivotto3 and S. Tyldesley1

1 Radiation Therapy Program, British Columbia Cancer Agency, Vancouver Cancer Centre, Vancouver, BC 2 Population and Preventative Oncology Program, Vancouver, BC 3 Radiation Therapy Program, British Columbia Cancer Agency, Vancouver Island Cancer Centre, Breast Outcomes Unit and University of British Columbia, BC, Canada

* Correspondence to: Dr G. Macdonald, Consultant in Clinical Oncology, ANCHOR Unit, Aberdeen Royal Infirmary, Foresterhill, Aberdeen AB25 2ZN, Scotland, UK. Tel: +44-1224-553498; Fax: +44-1224-554183; Email: graham.macdonald{at}nhs.net


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Background: The aim was to determine whether gender was a significant prognostic factor for post-mastectomy relapse, after accounting for known prognostic factors and delivery of radiotherapy.

Patients and methods: All patients diagnosed with invasive breast cancer between 1 January 1989 and 31 December 1998 who had undergone total mastectomy as primary therapy were identified from the British Columbia Cancer Agency's Breast Cancer Outcomes Unit database. Patients with pT4 or M1 disease were excluded. A comparison of patient, tumour and treatment factors was made between males and females. Outcomes were analysed in terms of locoregional-relapse free survival, breast cancer-specific survival and overall survival.

Results: Sixty males and 4181 females were identified. Multivariable analysis revealed increased tumour size, positive margin status, delivery of chemotherapy, positive nodal status and male gender to be significantly associated with the use of post-mastectomy radiotherapy. Multivariable analysis revealed tumour size, nodal status, tumour grade and presence of vascular space invasion to be significantly associated with locoregional recurrence. Gender was not a prognostic factor for locoregional recurrence, breast cancer-specific survival or overall survival on univariable or multivariable analysis.

Conclusions: These data suggest that gender is not a prognostic factor in patients undergoing mastectomy for early stage breast cancer. Men having mastectomy for breast cancer should receive adjuvant radiotherapy following guidelines similar to those developed for females.

Key words: male breast cancer, radiotherapy, locoregional control


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Breast cancer in males is uncommon, with an annual incidence of <1 compared with >120 per 100 000 in females [1Go]. Age-standardised breast cancer mortality rates are 0.3 and 22.4 per 100 000 for males and females, respectively [1Go]. Overall survival is reportedly worse for males [2Go], but after adjusting for age, T stage and number of positive nodes, survival in men and women has been reported to be similar [3Go–7Go]. An analysis of 3627 pairs of men and women with breast cancer, matched for age and stage, showed that males received significantly more aggressive local treatment (surgery and radiotherapy) than females but had similar locoregional control and survival [4Go].

Many institutions offer adjuvant radiotherapy following mastectomy to all male breast cancer patients because of a perception of a high tumour to breast size ratio [8Go]. In contrast, for women with breast cancer, adjuvant radiotherapy is offered for specific indications such as tumour size, margin status and number of positive nodes [9Go–11Go]. There is continued debate about whether post-mastectomy radiotherapy should be guided by the same principles used in women, or given routinely to all men with breast cancer [12Go].

This is a report of a descriptive analysis of the BC experience of male breast cancer from 1989 to 1998, with particular emphasis on whether gender was a significant prognostic factor for post-mastectomy relapse, after accounting for known prognostic factors and delivery of radiotherapy.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Patient identification
British Columbia (BC) has a population of approximately 4 million with a universal access public health care system. The BC Cancer Agency (BCCA) has the mandate for cancer control in BC including management of the BC Cancer Registry (BCCR), screening programmes for breast and cervical cancer, the budget for all anti-neoplastic drugs and the delivery of all radiation therapy through four regional cancer centres. Males and females diagnosed with breast cancer in British Columbia between 1 January 1989 and 31 December 1998 were identified from the BCCA Breast Cancer Outcomes Unit (BCOU) database, a prospectively maintained resource that includes demographic, pathology, staging, treatment and outcome data on all patients referred to the BCCA who were diagnosed with breast cancer since 1 January 1989. The BCCR contains demographic data and date and cause of death data on all patients diagnosed with cancer in BC whether referred or not to the BCCA.

Ethical approval for this study was obtained from the University of British Columbia BCCA Ethics Review Board.

Descriptive dataset
All cases of breast cancer (invasive and in situ) in males during this 10-year period were abstracted from the database for a descriptive analysis of patient, tumour and treatment factors, as well as recurrence and death rates. A comparison of demographic factors and mortality between referred and non-referred patients was undertaken.

Comparative dataset
The comparative dataset was abstracted from the BCOU database and included all patients (male or female) diagnosed with breast cancer during the 10-year period who had undergone total mastectomy as primary surgery. Patients were excluded if any of the following criteria were met: [1Go] history of prior invasive or in situ breast cancer; [2Go] synchronous contralateral breast cancer; [3Go] age under 40 at diagnosis (as breast cancer is exceptionally rare in males under the age of 40, and in women in this age group, breast cancer frequently manifests a more aggressive biology); [4Go] pre-operative radiotherapy; [5Go] T4 disease at diagnosis (as the requirement for adjuvant radiotherapy in T4 disease is clear); [6Go] M1 disease at diagnosis; [7Go] in situ disease alone; [8Go] out-of-province residence at initial diagnosis; [9Go] referral at recurrence or with residual disease; and [10Go] referral for follow-up of a previously treated cancer.

A comparison of patient-related factors (age, age group), tumour-related factors [histology, grade, margin status, presence of lymphovascular or perineural invasion (LVN status), estrogen receptor (ER) status, pathological T-stage, tumour size, nodal status] and treatment-related factors (delivery of radiotherapy and systemic therapy) was made between males and females in the dataset. The two groups were also compared with regard to locoregional and distant relapse-free survival, breast cancer-specific survival and overall survival.

Definition of end points
The primary end point of the study was locoregional relapse rate in males compared with females. Secondary end points included distant relapse-free, breast cancer-specific and overall survival.

All survival end points were calculated as the time from diagnosis to the time of the event. Patient data were censored at time of death (for relapse-free survival), at time of non-breast cancer death (for breast cancer-specific survival), at time of subsequent contralateral breast cancer (for all survival end points), or on 30 June 2003. Locoregional relapse-free survival times were censored at the time of distant relapse. Where cause of death was unknown and the patient had suffered a regional or distant breast cancer relapse, the cause of death was assumed to be breast cancer. Where the cause of death was documented as breast cancer and the patient had no documented distant relapse, then metastatic disease was assumed and dated, empirically, 4 months prior to the date of death.

Statistical analysis
Statistical analyses were performed using SPSS software, version 11.0. Student's t-tests were used to compare normally distributed data (age and tumour size) and the distribution checked using the P–P plot. The Mann–Whitney U-test was performed for non-parametric continuous data (number of positive lymph nodes, and times from diagnosis to surgery and radiotherapy) and the chi-squared test (with Yates’ correction) was used for categorical data. Binary logistic regression was used to identify independent predictive factors for the delivery of post-mastectomy radiotherapy. Survival data was analysed using the Kaplan–Meier method [13Go] and comparisons made using the log-rank test. Single variable regression analyses for survival end points used the Cox regression method, and the methodology used to check proportional hazards assumptions involved examination of the scaled Schoenfeld residuals [14Go]. Multiple variable models were defined using a stepwise AIC-based model building algorithm [15Go, 16Go]. All statistical tests used two-sided P values. With the numbers of events in our dataset, the study had 80% power (with an alpha level of 0.05) to detect minimum hazard ratios of 3.5 for locoregional relapse, 1.9 for overall survival and 2.4 for breast cancer-specific survival.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Descriptive dataset
The descriptive dataset comprised 159 males, of whom 109 (69%) had been referred to the BCCA. Characteristics of referred and non-referred patients are shown in Table 1. Tumour site was most commonly described as ‘central’ or ‘breast not specified’ (77%), with upper outer quadrant representing only 8% of cases. Median follow-up was 7.9 years (range 4.5–14.3 years). The proportion of males who had died was significantly greater in the non-referred group (P=0.009), and the median actuarial survival time in referred and non-referred patients was 13.8 and 5.9 years, respectively (P <0.001). Sixty of the 109 referred male patients met the aforementioned eligibility criteria for inclusion in the comparative dataset.


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics for the descriptive dataset

 
Comparative dataset
The comparative dataset comprised 60 males and 4181 females, with a median length of follow-up of 8.4 years (4.5–14.3) and 10.0 years (4.5–14.5), respectively. The median time from diagnosis to surgery was 13 days in both sexes (P=0.98); however, the median time from diagnosis to radiotherapy was longer for females than males (118 versus 97 days; P=0.04). Male patients were somewhat older, were more likely to have ER positive tumours and unknown margin status, and were more likely to receive radiation therapy than female patients (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Patient characteristics for the comparative dataset

 
Delivery of radiotherapy
In a multivariable binary logistic regression model including both male and female patients, statistically significant predictors for the delivery of post-mastectomy radiotherapy included tumour size [estimated odds ratio (OR)=1.38, P <0.001], margin status (OR=5.83, P <0.001), delivery of chemotherapy (OR=2.18, P <0.001) and nodal status (1–3 nodes OR=3.67, and 4+ nodes OR=28.58, P <0.001). When gender was added into the regression model, the OR for males to receive RT compared with females was estimated to be 5.82 (P <0.001).

Males who received RT were significantly younger than those who did not (mean age 61.6 versus 68.3, P=0.02). A similar result was seen for females (58.7 versus 62.6 years, P <0.001).

Locoregional recurrence-free survival
As of 30 June 2003, three males had a locoregional relapse (two in the irradiated group and one in the unirradiated), and 349 females had a locoregional relapse (76 in the irradiated group and 273 in the unirradiated). The actuarial survival curves for time to locoregional recurrence in males and females stratified by delivery of radiotherapy are shown in Figure 1. There was no significant difference in locoregional relapse-free survival between males and females who had received RT (P=0.58) and who had not received RT (P=0.58).



View larger version (14K):
[in this window]
[in a new window]
 
Figure 1. Time to locoregional recurrence.

 
Single variable Cox regression revealed positive nodes (P <0.001), tumour size (P <0.001), LVN status (P <0.001) and grade (P <0.001) to be significant prognostic factors for locoregional control (Table 3). After adjusting for these covariates in a multivariable model (Table 4), gender was found not to be prognostic (P=0.69).


View this table:
[in this window]
[in a new window]
 
Table 3. Significance test results (P value) for single variable Cox regression analyses of covariates for locoregional relapse-free, breast cancer-specific and overall survival

 

View this table:
[in this window]
[in a new window]
 
Table 4. Multivariable model for locoregional recurrence

 
Distant relapse-free survival
As of 30 June 2003, 12 males and 927 females had suffered a distant relapse. Actuarial distant relapse-free survival was estimated to be 82% and 77% for males at 5 years and 10 years, respectively, and 83% and 75% for females (P=0.99). The first site of metastatic relapse for males and females was bone in 39% and 33%, lung and/or pleura in 50% and 22%, and liver in 8% and 13%, respectively.

Breast cancer-specific survival
Actuarial breast cancer-specific survival was estimated to be 89% and 81% for males at 5 years and 10 years, respectively, and 88% and 79% for females (P=0.79). When stratified by whether or not radiotherapy had been delivered (Figure 2), no difference in breast cancer-specific survival was seen between males and females (P=0.66 and P=0.11 in the no radiotherapy and radiotherapy strata, respectively).



View larger version (16K):
[in this window]
[in a new window]
 
Figure 2. Time to breast cancer death.

 
Single variable Cox regression revealed use of radiotherapy (P <0.001) and positive nodal status (P <0.001) to be significant negative prognostic factors for breast cancer-specific survival (Table 3). Grade and LVN status were highly significant on log-rank test but did not satisfy the proportional hazards (PH) assumptions. After adjusting for nodal status and age, and stratifying for LVN status and grade, neither delivery of RT nor gender was found to be prognostic for breast cancer-specific survival (P=0.60).

Overall survival
Actuarial overall survival was estimated to be 75% and 53% for males at 5 years and 10 years, respectively, and 82% and 65% for females (P=0.11). When stratified by whether or not radiotherapy had been delivered (Figure 3), no difference was seen between males and females who had received radiotherapy. However, males who received no adjuvant radiotherapy had a poorer survival outcome than females receiving no adjuvant radiotherapy (P=0.03).



View larger version (17K):
[in this window]
[in a new window]
 
Figure 3. Overall survival.

 
Single variable Cox regression (Table 3) revealed delivery of radiotherapy (P <0.001) to be a significant prognostic factor for worse overall survival (OR 1.4). Positive nodes, LVN status, age group and grade were highly significant on log-rank test (P <0.001) but did not satisfy proportional hazards (PH) assumptions. After stratifying for age group and positive nodes (the two categorical covariates with the most ‘significant’ log rank scores), no other covariates satisfied the PH assumptions, and gender was found not to be prognostic for overall survival (P=0.60).


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
This study has demonstrated that gender is not a significant prognostic factor for post-mastectomy relapse, breast cancer-specific or overall survival, when known prognostic factors and delivery of radiotherapy were taken into account. However, the power to detect such an impact was low, in view of the small numbers of male patients and events.

In this cohort of non-metastatic patients with breast cancer treated with mastectomy, males were almost six times more likely to receive radiotherapy than females, even though patients with T4 disease were excluded and other known prognostic factors for locoregional recurrence were taken into account. In the National Cancer Data Base [4Go], a collaborative database reflecting community practice in the United States, males were more likely to undergo mastectomy than females, and more likely to receive adjuvant radiotherapy. In North America, therefore, there has been a bias towards treating males with more extensive local therapy, in view of a perceived increased risk of local relapse, rather than treating males for specific indications such as T stage, number of positive nodes or margin status. In our cohort, males who did not receive radiotherapy had a significantly poorer overall survival than females. The reason for withholding radiotherapy in females was generally related to good prognostic features, whereas in males, the reason for withholding radiotherapy was more likely to relate to poor performance status. The finding that males who do not receive radiotherapy have equivalent breast cancer-specific survival to females is consistent with the hypothesis that it is comorbidity rather than good tumour parameters that lead to males not receiving radiotherapy.

The referral rate was similar for males and females (69% and 72%, respectively; P=0.28; data not shown). Although the reasons for non-referral are unknown, the similar referral rates suggest that similar factors (e.g. patient refusal, poor performance status, referral out of province) may be responsible for non-referral in both groups. While we cannot exclude referral bias, we believe it is unlikely to have a major impact on the results of this study.

In the descriptive dataset, relatively few T3 tumours were seen. For male patients, skin or chest wall invasion probably occurs before tumours exceed 5 cm in diameter. The most common surgical procedure performed was mastectomy and axillary lymph node dissection (in 76% of cases). Only 10% of patients underwent a ‘breast-conserving’ procedure. Most tumours were described as ‘central’ or ‘breast—not otherwise specified’, whereas upper outer quadrant tumours are most common in females [17Go]. This is to be expected from the different anatomy of the male and female breast, with glandular tissue in the male breast being primarily localised to the retroareolar region.

In the comparative dataset, the median time from diagnosis to radiotherapy was longer for females than for males (118 versus 97 days; P=0.04), possibly reflecting increased physician concerns over locoregional recurrence in males, or the delivery of chemotherapy prior to radiotherapy in a greater proportion of females. In support of the former, the median time to radiotherapy in those patients who did not receive chemotherapy was longer in females than males (78 versus 64 days), although this did not reach statistical significance (P=0.40). Stage T4 tumours were excluded from the dataset, because patients with such advanced tumours require radiotherapy irrespective of gender and other factors. No significant difference was seen between the distribution of T1–T3 tumours, the mean tumour size or the numbers of positive nodes between the genders. There was a trend towards older age in the male cohort. Males also had a significantly higher proportion of grade 2 tumours and ER-positive tumours than females. We did not have data on HER-2 status; however, HER-2 over-expression has been shown to be associated with shorter disease-free survival in both males and females [18Go]. With regard to treatment, a statistically higher proportion of males received adjuvant radiotherapy, and there was a trend towards increased delivery of adjuvant hormonal therapy in males consistent with the higher proportion of ER-positive tumours. Approximately one-quarter of both males and females received chemotherapy.

The 5-year male breast cancer survivorship data are generally consistent with other series. The 5-year breast cancer-specific survival was 89% in this selected series, slightly higher than the 5-year breast cancer-specific survival rates of 74%–84% reported from other unselected series [5Go, 8Go]. The 5-year overall survival in the current series was 75%, intermediate between the 62%–90% in previous reports [3Go, 5Go, 8Go, 12Go, 19Go–21Go].


    Conclusions
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
This study found no evidence that gender is a prognostic factor for locoregional relapse-free, breast cancer-specific or overall survival in patients with breast cancer following mastectomy, when known prognostic factors and delivery of adjuvant radiotherapy were accounted for. This suggests that males should receive adjuvant therapy following similar guidelines as developed for females [9Go–11Go].

Received for publication January 24, 2005. Revision received April 22, 2005. Accepted for publication April 25, 2005.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
1. BC Cancer Registry data. http://www.bccancer.bc.ca/HPI/CancerStatistics/default.htm (December 2004, date last accessed).

2. Ciatto S, Iossa A, Bonardi R et al. Male breast carcinoma: review of a multicenter series of 150 cases. Coordinating Center and Writing Committee of FONCAM (National Task Force for Breast Cancer), 76. Italy: Tumori 1990; 555–558.

3. Guinee VF, Olsson H, Moller T et al. The prognosis of breast cancer in males. A report of 335 cases. Cancer 1993; 71: 154–161.

4. Scott-Conner CE, Jochimsen PR, Menck HR et al. An analysis of male and female breast cancer treatment and survival among demographically identical pairs of patients. Surgery 1999; 126: 775–780.[CrossRef][ISI][Medline]

5. Cutuli B, Lacroze M, Dilhuydy JM et al. Male breast cancer: results of the treatments and prognostic factors in 397 cases. Eur J Cancer 1995; 31A: 1960–1964.[CrossRef]

6. Borgen PI, Senie RT, McKinnon WM et al. Carcinoma of the male breast: analysis of prognosis compared with matched female patients. Ann Surg Oncol 1997; 4: 385–388.[Abstract]

7. Willsher PC, Leach IH, Ellis IO et al. A comparison outcome of male breast cancer with female breast cancer. Am J Surg 1997; 173: 185–188.[CrossRef][ISI][Medline]

8. Stranzl H, Mayer R, Quehenberger F et al. Adjuvant radiotherapy in male breast cancer. Radiother Oncol 1999; 53: 29–35.[CrossRef][ISI][Medline]

9. Recht A, Edge SB, Solin LJ et al. Postmastectomy radiotherapy: clinical practice guidelines of the American Society of Clinical Oncology. J Clin Oncol 2001; 19: 1539–1569.[Abstract/Free Full Text]

10. Harris JR, Halpin-Murphy P, McNeese M et al. Consensus statement on postmastectomy radiation therapy. Int J Radiat Oncol Biol Phys 1999; 44: 989–990.[CrossRef][ISI][Medline]

11. Truong PT, Olivotto IA, Whelan TJ et al. Clinical practice guidelines for the care and treatment of breast cancer: 16. Locoregional post-mastectomy radiotherapy. CMAJ 2004; 170: 1263–1273.

12. Chakravarthy A, Kim CR. Post-mastectomy radiation in male breast cancer. Radiother Oncol 2002; 65: 99–103.[CrossRef][ISI][Medline]

13. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

14. Therneau TM, Grambsch PM. Cox PH modeling and testing. In: Statistics for Biology and Health Series. Modelling Survival Data: Extending the Cox Model. New York: Springer-Verlag 2000.

15. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing (http://www.R-project.org); 2004.

16. Venables WN, Ripley BD. Modern Applied Statistics with S, 4th edition. New York: Springer 2002.

17. Kroman N, Wohlfahrt J, Mouridsen HT et al. Influence of tumor location on breast cancer prognosis. Int J Cancer 2003; 105: 542–545.[CrossRef][ISI][Medline]

18. Wang-Rodriguez J, Cross K, Gallagher S et al. Male breast carcinoma: correlation of ER, PR, Ki-67, Her2-Neu, and p53 with treatment and survival, a study of 65 cases. Mod Pathol 2002; 15: 853–861.[CrossRef][ISI][Medline]

19. Atalay C, Kanlioz M, Altinok M. Prognostic factors affecting survival in male breast cancer. J Exp Clin Cancer Res 2003; 22: 29–33.[Medline]

20. Stierer M, Rosen H, Weitensfelder W et al. Male breast cancer: Austrian experience. World J Surg 1995; 19: 687–692.[CrossRef][ISI][Medline]

21. Perkins GH, Middleton LP, Garcia SM et al. Male breast carcinoma: outcomes and predictors of local-regional failure for patients treated without radiation therapy. Breast Cancer Res Treat 2002; 76: 121.[CrossRef]





This Article
Abstract
Full Text (PDF)
All Versions of this Article:
16/9/1442    most recent
mdi274v1
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Macdonald, G.
Articles by Tyldesley, S.
PubMed
PubMed Citation
Articles by Macdonald, G.
Articles by Tyldesley, S.