Postmastectomy Radiation Therapy Is Associated With Improved Survival in Node-Positive Male Breast Cancer: A Population Analysis

Accepted Manuscript Post Mastectomy Radiotherapy Is Associated With Improved Survival in Node Positive Male Breast Cancer, a Population Analysis Matthew J. Abrams, Paul P. Koffer, David E. Wazer, Jaroslaw T. Hepel PII:

S0360-3016(17)30345-0

DOI:

10.1016/j.ijrobp.2017.02.007

Reference:

ROB 24090

To appear in:

International Journal of Radiation Oncology • Biology • Physics

Received Date: 15 September 2016 Revised Date:

17 December 2016

Accepted Date: 7 February 2017

Please cite this article as: Abrams MJ, Koffer PP, Wazer DE, Hepel JT, Post Mastectomy Radiotherapy Is Associated With Improved Survival in Node Positive Male Breast Cancer, a Population Analysis, International Journal of Radiation Oncology • Biology • Physics (2017), doi: 10.1016/ j.ijrobp.2017.02.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

TITLE Post Mastectomy Radiotherapy Is Associated With Improved Survival in Node Positive Male Breast Cancer, a Population Analysis

RUNNING TITLE: PMRT and Male Breast Cancer

CONFLICTS OF INTEREST: • The authors have no conflicts of interest

TE D

ACKNOWLEDGEMENTS: • No acknowledgements

EP

KEYWORDS: • Male breast cancer, post mastectomy radiotherapy, SEER

AC C

SC

M AN U

CORRESPONDING AUTHOR INFORMATION Matthew Abrams, MD [email protected] Tufts University School of Medicine Tufts Medical Center Department of Radiation Oncology 800 Washington Street, Box # 359 Boston, MA 02111 +1.617.636.6161

RI PT

AUTHORS Matthew J. Abrams1, Paul P. Koffer1, David E. Wazer1,2, and Jaroslaw T. Hepel2 1 Department of Radiation Oncology, Tufts University School of Medicine, Tufts Medical Center, Boston, MA 2 Department of Radiation Oncology, The Alpert Medical School of Brown University, Rhode Island Hospital, Providence, Rhode Island

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Because of its rarity, there are no randomized trials providing evidence for or against post mastectomy radiotherapy for male breast cancer. We retrospectively examined its impact via the National Cancer Institute’s Surveillance Epidemiology and End Results database. Post mastectomy radiotherapy and estrogen receptor positive disease were found to be an independent predictors of improved survival. Post mastectomy radiotherapy was associated with improved 5-year overall survival in all node positive patients.

ACCEPTED MANUSCRIPT

ABSTRACT Purpose: Because of its rarity, there are no randomized trials investigating post mastectomy radiotherapy (PMRT) in male breast cancer. This study retrospectively examines the impact of PMRT

RI PT

in male breast cancer patients in the National Cancer Institute’s Surveillance Epidemiology and End Results (SEER) database.

Methods and Materials: The SEER database 8.3.2 was queried for men ages 20+ diagnosed with

SC

localized or regional non-metastatic invasive ductal/lobular carcinoma from 1998-2013. Included patients were treated by modified radical mastectomy (MRM), with or without adjuvant external

M AN U

beam radiation. Univariate and multivariate analyses evaluated predictors for PMRT use after MRM. Kaplan-Meier overall survival (OS) curves of the entire cohort and a case-matched cohort were calculated and compared by the log-rank test. Cox-regression was used for multivariate survival analyses.

TE D

Results: A total of 1933 patients were included in the unmatched cohort. There was no difference in 5 year OS between those who received PMRT and those that did not (78% vs. 77% respectively, p = 0.371), however, in the case-matched analysis, PMRT was associated with improved OS at 5 years (83% vs. 54%, p=<0.001). On subset analysis of the unmatched cohort, PMRT was associated with

EP

improved OS in men with 1-3 nodes positive (5 year OS 79% vs. 72% p=0.05) and those with 4+ nodes positive (5 year OS 73% vs. 53% p<0.001). On multivariate analysis of the unmatched cohort,

AC C

independent predictors for improved OS were use of PMRT HR=0.551 [0.412-0.737] and ER positive disease HR=0.577 [0.339-0.983]. Predictors for a survival detriment were higher grade III/IV HR=1.825 [1.105-3.015], larger tumor T2 HR=1.783 [1.357-2.342] T3/T4 HR=2.683 [1.809-3.978], higher N-stage N1 HR=1.574 [1.184-2.091] N2/N3 HR=2.328 [1.684-3.218], Black race HR=1.689 [1.222-2.336] and older age 81+ HR=4.164 [1.497-11.582].

ACCEPTED MANUSCRIPT

Conclusions: There may be a survival benefit with the addition of PMRT for male breast

cancer with node positive disease.

RI PT

Introduction: Male breast cancer (MBC) comprises less than 1% of breast carcinomas1 with an estimated 2,600

cases and 440 deaths in the United States in 2016.2 The histopathology of MBC is similar to that of estrogen receptor (ER) positive breast cancer in postmenopausal women, and infiltrating ductal

SC

carcinoma comprising the majority of cases.3,4 Due to lack of prospective data and limited

retrospective series, male breast cancer has been treated similar to female breast cancer, however,

M AN U

experts have cautioned that male breast cancer should be evaluated and managed as a distinct entity.5

For example, the incidence of male breast cancer and female breast cancer continue to increase with age, however, male breast cancer lacks “Clemmsen’s hook,” a manifestation of the bimodal age

TE D

distribution of female breast cancer and instead increases continuously with age.5 In addition, there can be a strong genetic component to male breast cancer; approximately 15-20% of men report a family history of breast or ovarian cancer and approximately 10% are found to have a genetic predisposition often with mutations in BRCA1, BRCA2 PTEN, P53, CHEK2 and Klinefelter’s syndrome

EP

(XXY).6-10 Besides genetics, other known risk factors for the development of male breast cancer include age, Black race, radiation exposure (e.g. mantle radiation for Hodgkin lymphoma),

AC C

gynecomastia and increased estradiol levels.

Although clinical and pathological factors (T-stage, extent of lymph node involvement, and ER status) driving prognosis are similar in male and female breast cancer, men tend to present with more advanced stage, larger tumors with higher rates of axillary nodal involvement, likely secondary to lack of screening mammography in men. Historically, survival outcomes have been worse for men than women, thought to be due to their later stage and older age at presentation.11 According to a

ACCEPTED MANUSCRIPT

relatively limited database, after 5 years of relative equivalence, women had improved breast cancerspecific survival compared to men.12 In addition, men with breast cancer tend to have lower grade

RI PT

histology, and with higher rates of ER positivity compared to their female counterparts.4

We need further study in male breast cancer on the molecular pathology, risk factors, genetic

contribution and epidemiology. For these reasons and the rarity of the disease, experts caution

extrapolating from female breast cancer trials when making management decisions on male breast

systemic and radiotherapy for female breast cancer patients

13-16

SC

cancer. Although multiple studies have demonstrated local control and survival benefit of adjuvant

, there are no randomized trials and a

M AN U

paucity of retrospective reviews on the benefit of radiotherapy for males with breast cancer. In this retrospective analysis, we help elucidate the indications for PMRT in men and investigate its association with survival.

Materials and Methods:

TE D

The National Cancer Institute’s SEER (Surveillance, Epidemiology and End Results) database 8.3.2 was queried with a retrospective cohort study. The SEER database includes cancer incidence and survival data from approximately 30% of men and women across the United States.17

EP

Included patients were men aged 20 and over diagnosed with SEER historic stage A localized (invasive disease limited to the breast) or regional (invasive disease limited to surrounding tissue)

AC C

invasive ductal/lobular (ICD 8500-8549) breast cancer that underwent a modified radical mastectomy (surgery codes 50-58) from 1998-2013 with or without post-surgical external beam radiation. The primary outcome was overall survival.

Patients were excluded if they had SEER historic stage A in-situ, distant or un-staged disease. Further exclusion criteria included those who received radiation prior to surgery, radiation before and after surgery, intraoperative radiation, surgery before and after radiation or if the sequence was unknown. Alternative forms of radiation other than external beam (radioactive implants, combination of beam

ACCEPTED MANUSCRIPT

with implants, radiation NOS) were excluded, as were those who were recommended but refused or unknown if they received radiation. To minimize immortal time bias18, patients who survived <2 months from the time of diagnosis were excluded. Data on lymph nodes involved with isolated tumor

RI PT

cells were only available after 2004. Patients with isolated tumor cells only were considered node negative, N0(i+).

Patients were categorized by ER status, PR status, grade, stage, T-stage, N-stage, nodal extension,

SC

race, year diagnosed and age. Grade was categorized as well differentiated (grade I), moderately

differentiated (grade II), poorly differentiated (grade III) or anaplastic (grade IV). These variables

M AN U

were grouped according to PMRT use and compared by the Chi-squared test. Overall survival was calculated by the Kaplan-Meier method and compared by the log-rank test. A case-matched analysis was then performed, matching for ER status, T-stage, N-stage, race grade and age (+/- 5 years) using a randomized IBM SPSS 1:1 case match algorithm to control for confounding variables. Case matching was performed to reduce the confounding bias on the matched variables inherent in a

TE D

population based analysis based on the imbalance of important prognostic factors between the groups. The matched variables above are known to be prognostic for survival for male breast cancer. A total of 315 cases in the no-PMRT group were successfully matched with 315 cases from the PMRT

rank test.

EP

group. Kaplan-Meier curves were created with the case matched cohort and compared by the log-

AC C

Multivariate cox regression analysis was performed with the unmatched overall survival data on potential covariates including ER status, PR Status, grade, T-stage, N-stage, nodal extension, race, year of diagnosis and age. Patients with unknown or borderline ER status, PR status, grade, T-stage, N-stage or race were excluded from the multivariate analysis. A subset analysis was then restricted to patients with 0 nodes positive, 1-3 macrometastatic nodes positive and 4 or more macrometastatic nodes positive. Kaplan-Meier curves were calculated and compared by the log rank test. All statistical analysis was performed using IBM SPSS 22, IBM Corp., Armonk, NY. A p-value ≤0.05 was considered significant.

ACCEPTED MANUSCRIPT

Results: A total of 1933 patients were included in the entire analysis. Most patients had T1/T2 tumors, N0/N1

RI PT

disease and were aged 61-80. The vast majority were ER and PR positive. Most had regional disease rather than local disease. The median number of lymph nodes examined was 13 and the mean

number positive was one. There were significant differences between those who received PMRT and those who did not, as presented in Table 1. Men who received PMRT were more likely to be younger,

SC

have ER/PR positive, higher grade tumors, and have higher T-stage and N-stage disease and rates of regional disease with macrometastases. The use of PMRT has been increasing over time from 25% in

M AN U

1998-2002 to 27% in 2003-2007 and finally 31% in 2008-2013.

In the unmatched cohort there was no difference in 5 year overall survival between those who received PMRT and those that did not (78% vs. 77% respectively, p=0.371, Figure 1A). However, in the case-matched analysis, the benefit of PMRT gained significance and was associated with

TE D

improved overall survival (83% vs. 54%, p<0.001, Figure 1B).

A subgroup analysis based on number of nodes involved was then performed on the unmatched cohort. The unmatched cohort was chosen instead of the matched cohort because the act of

EP

matching removes the matched variable from its association with the outcome. Of the 883 patients with 0 nodes positive, 101 (11%) received PMRT and PMRT was shown to have no statistically

AC C

significant effect on 5 year OS (86% vs. 82% p=0.334, Figure 2). Of the 539 patients with 1-3 nodes positive, 181 (34%) received PMRT and PMRT was associated with improved 5 year OS 79% vs. 72% p=0.05, with an absolute risk reduction of 7% as seen in Figure 3. Furthermore, of the 368 patients with 4 or more nodes positive, 223 (61%) received PMRT and PMRT was associated with improved 5 year OS (73% vs. 53% p<0.001) with an absolute risk reduction of 20% (Figure 4).

On multivariate analysis of the unmatched cohort, independent predictors for improved overall survival were use of PMRT HR=0.551 [0.412-0.737] and ER positive disease HR=0.577 [0.339-0.983].

ACCEPTED MANUSCRIPT

Predictors for an overall survival detriment were higher grade III/IV HR=1.825 [1.105-3.015], larger tumor size T2 HR=1.783 [1.357-2.342] T3/T4 HR=2.683 [1.809-3.978], higher N-stage N1 HR=1.574 [1.184-2.091] N2/N3 HR=2.328 [1.684-3.218], Black race HR=1.689 [1.222-2.336] and older age 81+

RI PT

HR=4.164 [1.497-11.582]. PR status and year of diagnosis were not found to be independent predictors of survival (Table 2).

Discussion:

SC

When deciding on PMRT for male breast cancer patients, clinicians often rely on extrapolated data from the female literature where there is a large body of randomized evidence suggesting a local

M AN U

control as well as an overall survival benefit with the addition of adjuvant radiotherapy in women after breast conserving surgery19 as well as after mastectomy in women with node positive breast cancer.13,20-22 Historically, there has been controversy regarding the benefit of PMRT in women with 1-3 positive nodes versus those with 4 or more positive nodes. The most recent update of the EBCTCG meta-analysis has shown a benefit in both groups, and an increasingly common practice is to

TE D

offer PMRT to women with any number of macroscopically positive nodes.13 Additional risk factors that warrant consideration for PMRT in female breast cancer patients include young age, size of primary disease, lymphovascular invasion, receptor status, grade, response to neoadjuvant therapy

EP

and margin status.

Without any prospective data guiding PMRT decisions, clinicians commonly offer post-operative

AC C

radiation therapy to men with high-risk features. Small retrospective series have attempted to elucidate the differences in benefit from PMRT between male and female breast cancer. A recent series of 81 patients found a statistically significant improvement in local recurrence (24% vs. 4%) with the use of PMRT in male breast cancer patients, but no benefit in overall survival.23 None of the 5 patients with T2 only disease had locoregional relapse without PMRT, however, 4 out of 5 T2 highrisk patients with positive nodes had locoregional relapse without PMRT while 0/6 had locoregional relapse with PMRT. Further analysis showed that the greatest benefit was seen in those with the highest risk. Another series from Chakravarthy et al. showed low rates of locoregional failure with

ACCEPTED MANUSCRIPT

surgery alone in 1-3 nodes positive (6.5%), however much higher rates with 4 or more nodes positive indicating a possible need for further locoregional therapy such as radiation.24

RI PT

While unable to present locoregional failure data (SEER does not code for this), we present one of the largest cohorts evaluating survival after PMRT for male breast cancer23,24. When examining the

entire cohort, PMRT was not associated with improved overall survival. This was expected as there were marked differences in prognostic factors for those selected to receive PMRT (e.g. higher T-

SC

stage, higher N-stage, higher grade etc.)

M AN U

Therefore, a case-matched analysis was performed on 630 patients to reduce the effect of confounding variables. This analysis showed PMRT is associated with improved overall survival for the entire cohort, however, it was unclear whether this was driven mostly by those with positive nodes (one of the strongest prognostic factors).

TE D

Because there has been controversy regarding the benefit of PMRT in the female literature for women with 1-3 versus 4 or more pathologically involved nodes, we performed a subset analysis of the unmatched cohort and showed that a survival benefit persisted in both groups, albeit, the largest benefit 23

EP

was found in those at the highest risk (4 or more nodes positive), in accordance with previous studies.

An ideal group to examine would have been node negative men with tumors > 5cm, however, with only

AC C

29 patients evaluable (of which 17 received PMRT and 12 who did not receive PMRT) for analysis in this group, no statistically meaningful conclusions can be drawn.

Also, we found that current predictors for the use of PMRT are known high risk factors for locoregional recurrence (high grade, increasing nodes positive and larger size). These findings are in accordance with what we have seen with randomized female breast cancer data and provide the strongest support yet for the indications for PMRT for male breast cancer. In addition, the use of PMRT for this cohort has been

ACCEPTED MANUSCRIPT

increasing over time from 1998-2013, likely a result of the increasing amount of data supporting its use for high-risk female breast cancer patients published from 1997-2005.

RI PT

This study (and all studies of the SEER database) has certain limitations. This database can lack completeness for coding data of queried variables, contain miscoded variables and often presents a selection bias for a particular treatment. In regards to coding, there have been instances of

incompleteness of coding of radiation therapy, in particular, breast cancer patients who have

SC

received PMRT as noted in the analysis from Jagsi et al. which showed under-ascertainment of

radiation therapy which was related to geographic location. 25 In another study evaluating breast

M AN U

cancer coding in the California Cancer Registry, quality scores evaluating registry data were found to be 11% lower than those calculated from the medical record. In addition, quality scores decreased with increasing stage, raising the question of the quality registry coded data (such as in the SEER database) in locally advanced breast cancer.26 In order to mitigate some of these concerns with data coding reliability, a sensitivity analysis or analysis using the SEER-medicare linked database could be

TE D

performed. Regarding selection bias, common practice is to use PMRT when in doubt about the effectiveness of surgery alone in preventing a local recurrence. Hence the true benefit of PMRT may be overshadowed by the imbalances in the groups as seen in our study. To reduce confounding, a case-

EP

matched analysis was performed, however, the case-matched analysis can only reduce confounding on matched variables. In this study, the matched variables included were age, T-stage, N-stage, race, tumor

AC C

grade and ER status, as these are known to be prognostic for survival in male breast cancer and can impact the receipt of radiotherapy. Other variable such as grade, year of diagnosis could have been included, however, would limit the matched sample size and therefore would have made a meaningful analysis futile. In addition, there are multiple variable such as socioeconomic status and use of chemotherapy that would have been useful to match for, had that information been provided in the SEER database.

ACCEPTED MANUSCRIPT

Unfortunately there is a large amount of potentially useful data missing or unreliably coded in the SEER database such as chemotherapy, hormone therapy, and margin status and LVI. Chemotherapy is a particularly important missing variable, as the use of chemotherapy in locally advanced breast cancer has

27

RI PT

been shown to have a larger impact on survival than radiotherapy. For instance, the benefit of Herceptin in Her2-neu positive breast cancer reduces the risk of death by 39%. The use of endocrine therapy in 28,29

estrogen receptor positive tumors was similarly reduced by a significant amount.

Margin status may

be a unique issue for male breast cancer, as men have a smaller volume of breast tissue and therefore are 30

SC

more likely to have higher rates of close or positive margins and nipple/chest wall involvement.

M AN U

Another limitation of this study is that it is only applicable to men who undergo a modified radical mastectomy (which by definition, includes a low axillary lymph node dissection). While this is still the most common surgical procedure for a new diagnosis of non-metastatic male breast cancer, recent experience from MSKCC and the EIO showed that sentinel lymph node biopsy (in place of axillary lymph node dissection) has a 100% identification rate in cN0 patients31 and low false

TE D

negative rate (3%)32 which is similar to rates seen in female breast cancer (~10%)33. As we have seen in early stage clinically node negative and limited sentinel node positive female breast cancer, the pendulum in management of the axilla could be swinging from surgical to radiation with less

EP

toxicity and potentially equivalent outcomes.34,35

AC C

One last limitation of this study is the lack of information on a patient’s radiation dose. It is important to evaluate practice patterns in the use of standard fractionation and hypofractionation (especially for node negative patients) over the years. The SEER-medicare linked database may be helpful to investigate this further.

A prospective randomized trial investigating different therapies (systemic therapy, endocrine therapy and radiation therapy) for male breast cancer is desperately needed, however, based on the rarity of the disease, is likely cost prohibitive. A suitable alternative may be inclusion of male breast

ACCEPTED MANUSCRIPT

cancer patients in prospective female breast cancer trials, as historically they have largely been excluded. Currently, large meta-analyses of retrospective data are being analyzed.5

RI PT

In lieu of randomized data, this is one of the largest modern retrospective analyses of male breast cancer patients treated with or without PMRT. This data suggests a potential survival benefit to the use of PMRT in node positive male breast cancer patients. Generally, node negative patients do not benefit from PMRT, however, because of the paucity of T3N0 and T4N0 patients and other risk

SC

factors not reliably evaluable in the SEER database such as LVI, margin status, chemotherapy and

endocrine use, there may be subgroups of high risk node negative patients that benefit and further

M AN U

analysis is required.

Conclusions:

PMRT may be associated with improved survival for node positive male breast cancer, including those with 1-3 positive nodes and those with 4 or more positive nodes. Variables associated with

TE D

improved overall survival include the use of PMRT and ER positive disease while variables associated with an overall survival detriment include higher grade, larger tumor, higher nodal stage, Black race

AC C

EP

and older age.

ACCEPTED MANUSCRIPT

References

AC C

EP

TE D

M AN U

SC

RI PT

1. Borgen PI, Wong GY, Vlamis V, et al: Current management of male breast cancer. A review of 104 cases. Ann Surg 215:451-7; discussion 457-9, 1992 2. ACS: American Cancer Society: Cancer Facts and Figures 2016. Atlanta, GA, ACS, 2016 3. DeVita VT, Lawrence TS, Rosenberg SA: DeVita, Hellman, and Rosenberg's cancer : principles & practice of oncology (ed 9th). Philadelphia, Wolters Kluwer Health/Lippincott Williams & Wilkins, 2011 pp. 1401-46 4. Anderson WF, Jatoi I, Tse J, et al: Male breast cancer: a population-based comparison with female breast cancer. J Clin Oncol 28:232-9, 2010 5. Korde LA, Zujewski JA, Kamin L, et al: Multidisciplinary meeting on male breast cancer: summary and research recommendations. J Clin Oncol 28:2114-22, 2010 6. Wooster R, Bignell G, Lancaster J, et al: Identification of the breast cancer susceptibility gene BRCA2. Nature 378:789-92, 1995 7. Brose MS, Rebbeck TR, Calzone KA, et al: Cancer risk estimates for BRCA1 mutation carriers identified in a risk evaluation program. J Natl Cancer Inst 94:1365-72, 2002 8. Anelli A, Anelli TF, Youngson B, et al: Mutations of the p53 gene in male breast cancer. Cancer 75:2233-8, 1995 9. Fackenthal JD, Marsh DJ, Richardson AL, et al: Male breast cancer in Cowden syndrome patients with germline PTEN mutations. J Med Genet 38:159-64, 2001 10. Evans DB, Crichlow RW: Carcinoma of the male breast and Klinefelter's syndrome: is there an association? CA Cancer J Clin 37:246-51, 1987 11. Giordano SH, Buzdar AU, Hortobagyi GN: Breast cancer in men. Ann Intern Med 137:678-87, 2002 12. Engholm G, Ferlay J, Christensen N: NORDCAN: Cancer Incidence, Mortality and Prevalence in the Nordic Countries, Version 3.4, Association of Nordic Cancer Registries. Copenhagen,Denmark, Danish Cancer Society, 2009. 13. Early Breast Cancer Trialists' Collaborative Group, McGale P, Taylor C, et al: Effect of radiotherapy after mastectomy and axillary surgery on 10-year recurrence and 20-year breast cancer mortality: meta-analysis of individual patient data for 8135 women in 22 randomised trials. Lancet 383:2127-35, 2014 14. Early Breast Cancer Trialists' Collaborative G, Darby S, McGale P, et al: Effect of radiotherapy after breast-conserving surgery on 10-year recurrence and 15-year breast cancer death: meta-analysis of individual patient data for 10,801 women in 17 randomised trials. Lancet 378:1707-16, 2011 15. Early Breast Cancer Trialists' Collaborative G: Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 365:1687-717, 2005 16. Henderson IC, Berry DA, Demetri GD, et al: Improved outcomes from adding sequential Paclitaxel but not from escalating Doxorubicin dose in an adjuvant chemotherapy regimen for patients with node-positive primary breast cancer. J Clin Oncol 21:976-83, 2003 17. Surveillance, Epidemiology, and End Results (SEER) Program 8.3.2 (http://www.seer.cancer.gov) SEER*Stat Database: Incidence - SEER 9 Regs Research Data, Nov 2015 Sub (1973-2013) - Linked To County Attributes Total U.S., 1969-2014 Counties, National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2016, based on the November 2015 submission. 18. Park HS, Gross CP, Makarov DV, et al: Immortal time bias: a frequently unrecognized threat to validity in the evaluation of postoperative radiotherapy. Int J Radiat Oncol Biol Phys 83:1365-73, 2012 19. Early Breast Cancer Trialists' Collaborative Group, Darby S, McGale P, et al: Effect of radiotherapy after breast-conserving surgery on 10-year recurrence and 15-year breast cancer death: meta-analysis of individual patient data for 10,801 women in 17 randomised trials. Lancet 378:1707-16, 2011

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

20. Overgaard M, Hansen PS, Overgaard J, et al: Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 337:949-55, 1997 21. Overgaard M, Jensen MB, Overgaard J, et al: Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82c randomised trial. Lancet 353:1641-8, 1999 22. Ragaz J, Jackson SM, Le N, et al: Adjuvant radiotherapy and chemotherapy in nodepositive premenopausal women with breast cancer. N Engl J Med 337:956-62, 1997 23. Yu E, Suzuki H, Younus J, et al: The impact of post-mastectomy radiation therapy on male breast cancer patients--a case series. Int J Radiat Oncol Biol Phys 82:696-700, 2012 24. Chakravarthy A, Kim CR: Post-mastectomy radiation in male breast cancer. Radiother Oncol 65:99-103, 2002 25. Jagsi R, Abrahamse P, Hawley ST, et al: Underascertainment of radiotherapy receipt in Surveillance, Epidemiology, and End Results registry data. Cancer 118:333-41, 2012 26. Malin JL, Kahn KL, Adams J, et al: Validity of cancer registry data for measuring the quality of breast cancer care. J Natl Cancer Inst 94:835-44, 2002 27. Perez EA, Romond EH, Suman VJ, et al: Four-year follow-up of trastuzumab plus adjuvant chemotherapy for operable human epidermal growth factor receptor 2-positive breast cancer: joint analysis of data from NCCTG N9831 and NSABP B-31. J Clin Oncol 29:3366-73, 2011 28. International Breast Cancer Study G, Colleoni M, Gelber S, et al: Tamoxifen after adjuvant chemotherapy for premenopausal women with lymph node-positive breast cancer: International Breast Cancer Study Group Trial 13-93. J Clin Oncol 24:1332-41, 2006 29. Fisher B, Redmond C, Brown A, et al: Adjuvant chemotherapy with and without tamoxifen in the treatment of primary breast cancer: 5-year results from the National Surgical Adjuvant Breast and Bowel Project Trial. J Clin Oncol 4:459-71, 1986 30. Cutuli B, Lacroze M, Dilhuydy JM, et al: Male breast cancer: results of the treatments and prognostic factors in 397 cases. Eur J Cancer 31A:1960-4, 1995 31. Gentilini O, Chagas E, Zurrida S, et al: Sentinel lymph node biopsy in male patients with early breast cancer. Oncologist 12:512-5, 2007 32. Flynn LW, Park J, Patil SM, et al: Sentinel lymph node biopsy is successful and accurate in male breast carcinoma. J Am Coll Surg 206:616-21, 2008 33. Krag DN, Anderson SJ, Julian TB, et al: Technical outcomes of sentinel-lymph-node resection and conventional axillary-lymph-node dissection in patients with clinically node-negative breast cancer: results from the NSABP B-32 randomised phase III trial. Lancet Oncol 8:881-8, 2007 34. Giuliano AE, McCall L, Beitsch P, et al: Locoregional recurrence after sentinel lymph node dissection with or without axillary dissection in patients with sentinel lymph node metastases: the American College of Surgeons Oncology Group Z0011 randomized trial. Ann Surg 252:426-32; discussion 432-3, 2010 35. Donker M, van Tienhoven G, Straver ME, et al: Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981-22023 AMAROS): a randomised, multicentre, open-label, phase 3 non-inferiority trial. Lancet Oncol 15:1303-10, 2014

ACCEPTED MANUSCRIPT

Table 1: Baseline study characteristics

Characteristic

No PMRT (%)

PMRT (%)

Total

T-stage

p-Value

718 (51)

180 (33)

898

T2

559 (40)

244 (45)

803

T3

27 (2)

32 (6)

59

T4

54 (4)

71 (13)

125

Unknown

37 (3)

11 (2)

48

N-status

<0.001 782 (56)

101 (19)

883

N1

436 (31)

217 (40)

653

N2

93 (7)

143 (27)

N3

48 (3)

70 (13)

Unknown

36 (3)

7 (1)

Age at diagnosis

M AN U

N0

236 118 43

0.032

20 – 40

28 (2)

13 (2)

41 – 60

424 (30)

199 (37)

61– 80

785 (56)

276 (51)

1061

81-100

158 (11)

50 (9)

208

Positive

623

46 (3)

21 (4)

67

1176 (84)

484 (90)

1660

173 (12)

33 (6)

206

AC C

Unknown/borde rline PR status

41

TE D

Negative

EP

ER status

<0.001

<0.001

Negative

172 (12)

64 (12)

236

Positive

1020 (73)

431 (80)

1451

203 (15)

43 (8)

246

Unknown/borde rline Grade

SC

T1

RI PT

<0.001

<0.001

Grade I

155 (11)

31 (6)

186

Grade II

718 (51)

256 (48)

974

Grade III

443 (32)

234 (43)

677

Grade IV

13 (1)

2 (0)

15

ACCEPTED MANUSCRIPT

66 (5)

15 (3)

81

SEER Historical Stage Localized

706 (51)

68 (13)

774

Regional

689 (49)

470 (87)

1159

Nodal Extension No nodes positive Isolated tumor cells

<0.001 797 (57)

102 (19)

899

3 (0)

2 (0)

5

Micrometastases

87 (6)

29 (5)

116

Macrometastases

503 (36)

405 (75)

908

5 (0)

0 (0)

5

Unknown

0.792

White

1144 (82)

445 (83)

Black

186 (13)

66 (12)

65 (5)

27 (5)

Other/Unknown Year Diagnosed

M AN U

Race

RI PT

<0.001

SC

Unknown

1589 252 92

0.049

464 (33)

157 (29)

2003-2007

479 (34)

176 (33)

2008-2013

452 (32)

205 (38)

621 655 657

TE D

1998-2002

PMRT = post mastectomy radiotherapy, SEER = Surveillance Epidemiology End Results Database, ER = estrogen receptor, PR = progesterone receptor, T-stage = tumor stage, N-status = pathologic nodal status (micro and macrometastases), pos=positive

AC C

EP

Note: Percentages may not add up to exactly 100% due to rounding

ACCEPTED MANUSCRIPT

Table 2: Cox regression analysis by covariate for overall survival 95% CI Bounds Characteristic

HR

Lower

Upper

p-Value

0.412

0.737

<0.001

0.339

0.983

0.043

0.613

1.231

0.428

ref

PMRT

0.551

ER status ER negative

ref

ER positive

0.577

PR status PR negative

ref

PR positive

0.868

Grade I

ref

Grade II

1.228

0.746

2.022

Grade III/IV

1.825

1.105

3.015

T-stage T1

ref

T2

1.783

1.357

2.342

T3/T4

2.683

1.809

3.978

N0

ref

0.419 0.019

<0.001

<0.001

TE D

N-Stage

M AN U

Grade

N1

1.574

1.184

2.091

0.002

N2/N3

2.328

1.684

3.218

<0.001

Race ref

Black

1.689

Age

1.222

2.336

0.002

ref

AC C

Age 20-40

EP

White

SC

No PMRT

RI PT

Use of PMRT

Age 41-60

0.798

0.289

2.205

0.663

Age 61-80

1.440

0.530

3.913

0.475

Age 81+

4.164

1.497

11.582

0.006

Year Diagnosed 1998-2002

ref

2003-2007

0.840

0.639

1.104

0.210

2008-2013

0.786

0.572

1.080

0.138

CI = confidence interval; HR = hazards ratio; PMRT = post mastectomy radiotherapy; ref = reference variable

  0   1395  

10   1306  

20   1208  

30   1104  

40   988  

50   886  

60   793  

538  

513  

462  

421  

379  

331  

282  

 

EP

No   PMRT   PMRT  

AC C

Number   at  risk  

Months  

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

  Number   at  risk  

 

Months   No   PMRT   PMRT  

0   315  

10   269  

20   212  

30   159  

40   115  

50   91  

60   75  

315  

301  

274  

258  

232  

208  

174  

 

Months  

No   PMRT   PMRT  

0   782  

10   742  

20   698  

30   647  

40   597  

50   544  

60   490  

101  

97  

90  

85  

78  

71  

63  

AC C

EP

Number   at  risk  

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

 

 

Months  

No   PMRT   PMRT  

0   358  

10   332  

20   299  

30   271  

40   230  

50   203  

60   183  

181  

171  

152  

140  

129  

111  

95  

AC C

EP

Number   at  risk  

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

 

 

Months  

No   PMRT   PMRT  

0   145  

10   132  

20   119  

30   100  

40   79  

50   68  

60   55  

223  

212  

190  

169  

150  

129  

111  

AC C

EP

Number   at  risk  

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT