Original Study
Comparison of Adjuvant Radiation Therapy Alone Versus Radiation Therapy and Endocrine Therapy in Elderly Women With Early-Stage, Hormone Receptor-Positive Breast Cancer Treated With Breast-Conserving Surgery Colin T. Murphy,1 Tianyu Li,2 Lora S. Wang,1 Elias I. Obeid,3 Richard J. Bleicher,4 Gary Eastwick,5 Matthew E. Johnson,1 Shelly B. Hayes,1 Stephanie E. Weiss,1 Penny R. Anderson1 Abstract Data are limited regarding the efficacy of adjuvant radiation therapy (RT) alone for elderly women with low-risk breast cancer. The present study of > 500 women identified a group of women for whom adjuvant RT, without endocrine therapy (ET), could result in acceptable outcomes. The study also explored the influence of ET nonadherence and tumor size on outcomes. Background: Randomized data examining adjuvant radiation therapy (RT) alone in elderly women with low-risk, hormone receptor-positive (HRþ) breast cancer is lacking. We investigated the outcomes for elderly women treated with adjuvant RT alone versus RT plus endocrine therapy (ET) after breast-conserving surgery. Patients and Methods: We queried our institutional breast cancer database for the following patients: age > 65 years, stage T1-T2N0, HRþ, and treatment with breast-conserving surgery, including adjuvant RT. The c2 analysis identified significant baseline differences between the groups. Cox proportional hazard methods identified predictors of endpoints on multivariate analysis. Kaplan-Meier estimates of survival were compared using the log-rank test. Results: A total of 504 patients were identified, 311 had undergone RT plus ET (62%) and 193, RT alone (38%). The median follow-up time was 88 months. The RT-alone group versus RT plus ET group had different median age (72 vs.71 years, P < .001), different median tumor size (1 vs. 1.3 cm, P < .001), lower grade (40% vs. 29%, P ¼ .05), and fewer close or positive margins (11% vs. 19%, P ¼ .01). The adherence rate to prescribed ET was 70%. Tumor size predicted an increased risk of distant metastasis (DM) (hazard ratio, 1.96; 95% confidence interval [CI], 1.23-3.13) and worse disease-free survival (DFS) (hazard ratio, 1.86; 95% CI, 1.22-2.86). ET nonadherence versus adherence predicted for risk of DM (hazard ratio, 5.03; 95% CI, 1.98-12.66) and DFS (HR, 4.24; 95% CI, 1.9-10.3). Of the women with DM, 83.8% had tumors > 1 cm in size. Conclusion: ET nonadherence and tumor size > 1 cm predicted an increased risk of DM and worse DFS, favoring the addition of ET in this group. However, RT alone for women with tumors less than or equal to 1 cm may be appropriate. Clinical Breast Cancer, Vol. 15, No. 5, 381-9 ª 2015 Elsevier Inc. All rights reserved. Keywords: Adjuvant radiation alone, Elderly breast cancer, Low risk, Nonadherence, Tumor size less than 1cm
Introduction As the US population continues to age, the incidence of invasive breast cancer in women aged 65 years will increase,1 as will the 1
Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA Department of Biostatistics, Fox Chase Cancer Center, Philadelphia, PA Department of Medical Oncology and Clinical Genetics, Fox Chase Cancer Center, Philadelphia, PA 4 Department of Surgical Oncology, Fox Chase Cancer Center, Philadelphia, PA 5 Department of Radiation Oncology, State University of New York Upstate Medical University, Syracuse, NY 2 3
1526-8209/$ - see frontmatter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clbc.2015.02.005
risk of breast cancer death and death from nonebreast cancer causes.2 Interest exists for de-intensification of treatment for earlystage, low-risk breast cancer in elderly women, given the Submitted: Dec 17, 2014; Revised: Feb 12, 2015; Accepted: Feb 26, 2015; Epub: Mar 2, 2015 Address for correspondence: Colin T. Murphy, MD, Department of Radiation Oncology, Fox Chase Cancer Center, 333 Cottman Avenue, Philadelphia, PA 19111 E-mail contact: [email protected]
Clinical Breast Cancer October 2015
- 381
Elderly Breast RT Alone competing risk of comorbidity and death. Several prospective trials have investigated the role of adjuvant endocrine therapy (ET) alone, adjuvant radiation therapy (RT) alone, and combination therapy (RT plus ET) in the setting of early-stage, low-risk breast cancer.3-6 The subset of elderly women treated with adjuvant RT alone has largely been underrepresented in these trials. Thus, robust prospective data examining the role of adjuvant RT alone for elderly women with early-stage, low-risk, hormone receptor-positive (HRþ) breast cancer are lacking. The risk of toxicity from ET increases with age,7 and the ET noncompliance rate also increases,8 both of which call into question the benefit of adjuvant ET in this population. In the present single-institution retrospective analysis, we compared the outcomes of elderly women who underwent adjuvant RT alone versus RT plus ET for early-stage, low-risk breast cancer and to define a subset of women for whom adjuvant RT alone might be acceptable.
Patients and Methods We reviewed our institutional breast cancer database for patients with histologically confirmed breast cancer who met the following criteria: a minimum of 1 year of follow-up data from RT completion, age greater than or equal to 65 years, estrogen (ER) or progesterone receptor (PR) positive, tumor stage T1 or T2, adjuvant RT as a part of breast-conserving therapy, and pathologically nodenegative disease. The following patients were excluded: male patients, those with ductal carcinoma in situ, those with metastatic disease, and those with HER2þ disease. Patient demographics, tumor characteristics, and treatment-related information were entered prospectively into a database maintained and updated by a data manager. The collection, storage, and retrieval of data were all performed in compliance with the hospital’s institutional review board and the Health Insurance Privacy and Portability Act. The patients’ medical records were reviewed to extract information for patients receiving ET when available because the database does not capture the following elements: ET type (tamoxifen, anastrozole, letrozole, exemestane), ET duration measured in years, side effects attributed to ET, whether a woman switched from one type of ET to another, ET adherence, and the reason for stopping ET. ET adherence was defined as patient-reported use of the prescribed ET medication at the last follow-up visit or documented completion of the prescribed therapy duration by the treating physician (most commonly 5 years). Self-discontinuation was defined as nonadherence in the absence of side effects, toxicity, disease recurrence, or intercurrent medical comorbidity. Extended ET > 5 years was not routinely recommended during the study period. The ET duration was defined as the interval from the date of prescription to either discontinuation or the end of the follow-up period. The medical records of those patients who did not receive ET were also reviewed, and information was extracted regarding whether the patient had been evaluated and recommended to receive adjuvant ET by a medical oncologist and the reason for not receiving ET. We categorized four reasons for not receiving ET: patient refusal, ET not recommended because it was medically contraindicated, ET not recommend because of perceived low risk, and predating the standard of care. Because the study period dates back to 1981, a proportion of women were treated in an era predating the use of adjuvant ET as the standard of care for HRþ breast cancer; the Food and Drug
382
-
Clinical Breast Cancer October 2015
Administration approved the use of adjuvant tamoxifen for nodenegative breast cancer in 1990.9 The patient data were analyzed according to whether the patients had received adjuvant ET in addition to RT. The inclusion of women treated with adjuvant ET alone was not possible because our institutional breast database includes only women treated with RT and does not capture those who do not receive RT as part of their definitive breast cancer management. All patients underwent whole breast RT (range, 46-50 Gy), with a 10- to 16-Gy boost, depending on margin status (a positive margin was defined as microscopic tumor at the inked surface of the lumpectomy specimen), with or without regional nodal RT, or an institutional hypofractionated regimen that has been previously described.10 Whole-breast RT consisted of conventional photon tangents earlier in the study period; however, most patients underwent photon intensitymodulated RT (IMRT) in the later study period. In general, conventional RT consisted of medial and lateral tangential fields covering the clinically palpable breast tissue with a margin. The IMRT technique consisted of a combination of open and segmented tangential fields using volume-based inverse dose planning and step-and-shoot beam delivery using previously described methods.11 The site and date of the first recurrence was measured from the date of RT completion. Local failure was defined as recurrence in the ipsilateral breast, with or without nodal recurrence. Regional failure was defined as an ipsilateral axillary, supraclavicular, or internal mammary nodal recurrence, with or without local recurrence. A contralateral breast event (CBE) was defined as either cancer recurrence or a second breast primary in the contralateral breast. Additional endpoints included freedom from distant metastasis (FFDM), disease-free survival (DFS), and overall survival (OS). Late RT effects were diagnosed clinically by physician breast examinations during routine follow-up visits. The toxicity events routinely recorded in the database included fibrosis, edema, erythema, telangiectasia, and infection.
Statistical Analysis The c2 analysis identified significant baseline differences between the groups. The Kaplan-Meier (K-M) method was used to estimate 7-year outcomes. A Cox proportional hazard model was used to identify independent multivariate (MVA) predictors of each endpoint, adjusted for age, treatment group, tumor size, tumor grade, histologic type, lymphovascular invasion (LVI), surgical margin status, type of lymph node evaluation, RT technique, and treatment year. K-M survival curves were generated using the survival package in R (available at: www.r-cran-project.org). The other analyses were performed using SAS statistical software, version 9.3 (SAS Institute, Cary, NC).
Results Baseline patient information is listed in Table 1 according to treatment group. Those receiving ET were grouped according to whether they were adherent to ET. From 1981 to 2011, 504 women met the inclusion criteria, 197 (38%) had undergone RT alone and 311 (62%), RT plus ET. Of those prescribed ET, 94 (30.2%) were nonadherent. The median follow-up period for the entire cohort was 88 months (range, 13-288 months). Notable
Colin T. Murphy et al Table 1 Baseline Characteristics; Differences Compared Using the c2 Test Variable
RT Alone (%)
RT D ET-adherent (%)
RT D ET-nonadherent (%)
Patients (n)
193
217
94
Median
112
76
74
Range
13-285
13-232
14-288
80
22 (11)
11 (5)
12 (13)
Tumor stage
.510
T1
168 (87)
184 (85)
77 (82)
T2
25 (13)
33 (15)
17 (18)
0.5-1
76 (39)
50 (23)
32 (34)
>1-2
57 (30)
117 (54)
37 (39)
>2-5
19 (10)
33 (15)
17 (18)
Unknown
18 (9)
8 (4)
3 (4)
Invasive ductal
173 (90)
181 (83)
73 (78)
Invasive lobular
17 (9)
34 (16)
20 (21)
Other (invasive)
3 (2)
2 (1)
1 (1)
ERþ/PRþ
143 (74)
176 (81)
74 (79)
ERþ/PR
38 (20)
37 (17)
18 (19)
65 years, ERþ or PRþ, HER2, T1-T2N0). In our cohort, the baseline characteristics of the RT-alone and RT plus ET groups showed that the women in the RT-alone group were older and had more favorable disease characteristics. The RT-alone group had smaller primary tumors with a greater proportion that were well-differentiated and less likely to have close or positive margins. Tumor size independently predicted the risk of DM and DFS, and its strong prognostic association suggested that RT alone could result in acceptable outcomes for appropriately selected patients when the primary tumor is < 1 cm. These differences in patient selection could add to the low rates of overall disease recurrence in the RT-alone group—an estimated 6% at 7 years in the present study. As expected, each group of women in the present study had an estimated 7-year DFS rate that exceeded the estimated 7-year OS, underscoring the competing risk of death from causes other than breast cancer in this elderly, low-risk population. The strengths of the present study included the mature follow-up data, the requirement for known hormone-receptor positivity, a direct
Colin T. Murphy et al Table 2 Treatment Characteristics Stratified by Treatment Group, Compared Using the c2 Test Treatment Characteristic
RT Alone
RT Plus ET-adherent
RT Plus ET-nonadherent
Yes
34 (18)
85 (39)
33 (35)
No
159 (82)
132 (61)
61 (65)
Yes
136 (70)
96 (44)
46 (49)
No
57 (30)
121 (56)
48 (51)
Yes
23 (12)
36 (17)
15 (16)
No
170 (88)
181 (83)
79 (84)
P Value
Lymph node evaluation 80 years. The nonadherence rate in the present study was 30% and was the strongest predictor of increased risk of DM and worse DFS and OS.
Limitations A recently published cohort study aimed at prospectively investigating treatment decisions and outcomes in elderly women with breast cancer found that patient frailty (compared with robust patients) and increasing age predicted the risk of noninitiation and discontinuation of ET among elderly women.8 In the present study, neither performance status nor comorbidity was available for analysis, because these variables were not recorded in our institutional database. It is possible that the improved OS in the
Clinical Breast Cancer October 2015
- 387
Elderly Breast RT Alone Table 4 Seven-Year Kaplan-Meier Estimates of Outcomes, Compared Using the Log-Rank Test for Overall and Pairwise Comparisons Between Groups P Value
Outcome
7-Years Estimate (%) (95% CI)
All Groups
Local control (%) RT alone
98 (94-99)
RT plus ET-nonadherent
98 (85-100)
Regional control RT plus ET-adherent RT plus ET-nonadherent
.54
.3
.9
.361
.06
.2
.1
.012
.18
.97
.06
.176
Comparison of Adjuvant Radiation Therapy Alone Versus Radiation Therapy and Endocrine Therapy in Elderly Women With Early-Stage, Hormone Receptor-Positive Breast Cancer Treated With Breast-Conserving Surgery Colin T. Murphy,1 Tianyu Li,2 Lora S. Wang,1 Elias I. Obeid,3 Richard J. Bleicher,4 Gary Eastwick,5 Matthew E. Johnson,1 Shelly B. Hayes,1 Stephanie E. Weiss,1 Penny R. Anderson1 Abstract Data are limited regarding the efficacy of adjuvant radiation therapy (RT) alone for elderly women with low-risk breast cancer. The present study of > 500 women identified a group of women for whom adjuvant RT, without endocrine therapy (ET), could result in acceptable outcomes. The study also explored the influence of ET nonadherence and tumor size on outcomes. Background: Randomized data examining adjuvant radiation therapy (RT) alone in elderly women with low-risk, hormone receptor-positive (HRþ) breast cancer is lacking. We investigated the outcomes for elderly women treated with adjuvant RT alone versus RT plus endocrine therapy (ET) after breast-conserving surgery. Patients and Methods: We queried our institutional breast cancer database for the following patients: age > 65 years, stage T1-T2N0, HRþ, and treatment with breast-conserving surgery, including adjuvant RT. The c2 analysis identified significant baseline differences between the groups. Cox proportional hazard methods identified predictors of endpoints on multivariate analysis. Kaplan-Meier estimates of survival were compared using the log-rank test. Results: A total of 504 patients were identified, 311 had undergone RT plus ET (62%) and 193, RT alone (38%). The median follow-up time was 88 months. The RT-alone group versus RT plus ET group had different median age (72 vs.71 years, P < .001), different median tumor size (1 vs. 1.3 cm, P < .001), lower grade (40% vs. 29%, P ¼ .05), and fewer close or positive margins (11% vs. 19%, P ¼ .01). The adherence rate to prescribed ET was 70%. Tumor size predicted an increased risk of distant metastasis (DM) (hazard ratio, 1.96; 95% confidence interval [CI], 1.23-3.13) and worse disease-free survival (DFS) (hazard ratio, 1.86; 95% CI, 1.22-2.86). ET nonadherence versus adherence predicted for risk of DM (hazard ratio, 5.03; 95% CI, 1.98-12.66) and DFS (HR, 4.24; 95% CI, 1.9-10.3). Of the women with DM, 83.8% had tumors > 1 cm in size. Conclusion: ET nonadherence and tumor size > 1 cm predicted an increased risk of DM and worse DFS, favoring the addition of ET in this group. However, RT alone for women with tumors less than or equal to 1 cm may be appropriate. Clinical Breast Cancer, Vol. 15, No. 5, 381-9 ª 2015 Elsevier Inc. All rights reserved. Keywords: Adjuvant radiation alone, Elderly breast cancer, Low risk, Nonadherence, Tumor size less than 1cm
Introduction As the US population continues to age, the incidence of invasive breast cancer in women aged 65 years will increase,1 as will the 1
Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA Department of Biostatistics, Fox Chase Cancer Center, Philadelphia, PA Department of Medical Oncology and Clinical Genetics, Fox Chase Cancer Center, Philadelphia, PA 4 Department of Surgical Oncology, Fox Chase Cancer Center, Philadelphia, PA 5 Department of Radiation Oncology, State University of New York Upstate Medical University, Syracuse, NY 2 3
1526-8209/$ - see frontmatter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clbc.2015.02.005
risk of breast cancer death and death from nonebreast cancer causes.2 Interest exists for de-intensification of treatment for earlystage, low-risk breast cancer in elderly women, given the Submitted: Dec 17, 2014; Revised: Feb 12, 2015; Accepted: Feb 26, 2015; Epub: Mar 2, 2015 Address for correspondence: Colin T. Murphy, MD, Department of Radiation Oncology, Fox Chase Cancer Center, 333 Cottman Avenue, Philadelphia, PA 19111 E-mail contact: [email protected]
Clinical Breast Cancer October 2015
- 381
Elderly Breast RT Alone competing risk of comorbidity and death. Several prospective trials have investigated the role of adjuvant endocrine therapy (ET) alone, adjuvant radiation therapy (RT) alone, and combination therapy (RT plus ET) in the setting of early-stage, low-risk breast cancer.3-6 The subset of elderly women treated with adjuvant RT alone has largely been underrepresented in these trials. Thus, robust prospective data examining the role of adjuvant RT alone for elderly women with early-stage, low-risk, hormone receptor-positive (HRþ) breast cancer are lacking. The risk of toxicity from ET increases with age,7 and the ET noncompliance rate also increases,8 both of which call into question the benefit of adjuvant ET in this population. In the present single-institution retrospective analysis, we compared the outcomes of elderly women who underwent adjuvant RT alone versus RT plus ET for early-stage, low-risk breast cancer and to define a subset of women for whom adjuvant RT alone might be acceptable.
Patients and Methods We reviewed our institutional breast cancer database for patients with histologically confirmed breast cancer who met the following criteria: a minimum of 1 year of follow-up data from RT completion, age greater than or equal to 65 years, estrogen (ER) or progesterone receptor (PR) positive, tumor stage T1 or T2, adjuvant RT as a part of breast-conserving therapy, and pathologically nodenegative disease. The following patients were excluded: male patients, those with ductal carcinoma in situ, those with metastatic disease, and those with HER2þ disease. Patient demographics, tumor characteristics, and treatment-related information were entered prospectively into a database maintained and updated by a data manager. The collection, storage, and retrieval of data were all performed in compliance with the hospital’s institutional review board and the Health Insurance Privacy and Portability Act. The patients’ medical records were reviewed to extract information for patients receiving ET when available because the database does not capture the following elements: ET type (tamoxifen, anastrozole, letrozole, exemestane), ET duration measured in years, side effects attributed to ET, whether a woman switched from one type of ET to another, ET adherence, and the reason for stopping ET. ET adherence was defined as patient-reported use of the prescribed ET medication at the last follow-up visit or documented completion of the prescribed therapy duration by the treating physician (most commonly 5 years). Self-discontinuation was defined as nonadherence in the absence of side effects, toxicity, disease recurrence, or intercurrent medical comorbidity. Extended ET > 5 years was not routinely recommended during the study period. The ET duration was defined as the interval from the date of prescription to either discontinuation or the end of the follow-up period. The medical records of those patients who did not receive ET were also reviewed, and information was extracted regarding whether the patient had been evaluated and recommended to receive adjuvant ET by a medical oncologist and the reason for not receiving ET. We categorized four reasons for not receiving ET: patient refusal, ET not recommended because it was medically contraindicated, ET not recommend because of perceived low risk, and predating the standard of care. Because the study period dates back to 1981, a proportion of women were treated in an era predating the use of adjuvant ET as the standard of care for HRþ breast cancer; the Food and Drug
382
-
Clinical Breast Cancer October 2015
Administration approved the use of adjuvant tamoxifen for nodenegative breast cancer in 1990.9 The patient data were analyzed according to whether the patients had received adjuvant ET in addition to RT. The inclusion of women treated with adjuvant ET alone was not possible because our institutional breast database includes only women treated with RT and does not capture those who do not receive RT as part of their definitive breast cancer management. All patients underwent whole breast RT (range, 46-50 Gy), with a 10- to 16-Gy boost, depending on margin status (a positive margin was defined as microscopic tumor at the inked surface of the lumpectomy specimen), with or without regional nodal RT, or an institutional hypofractionated regimen that has been previously described.10 Whole-breast RT consisted of conventional photon tangents earlier in the study period; however, most patients underwent photon intensitymodulated RT (IMRT) in the later study period. In general, conventional RT consisted of medial and lateral tangential fields covering the clinically palpable breast tissue with a margin. The IMRT technique consisted of a combination of open and segmented tangential fields using volume-based inverse dose planning and step-and-shoot beam delivery using previously described methods.11 The site and date of the first recurrence was measured from the date of RT completion. Local failure was defined as recurrence in the ipsilateral breast, with or without nodal recurrence. Regional failure was defined as an ipsilateral axillary, supraclavicular, or internal mammary nodal recurrence, with or without local recurrence. A contralateral breast event (CBE) was defined as either cancer recurrence or a second breast primary in the contralateral breast. Additional endpoints included freedom from distant metastasis (FFDM), disease-free survival (DFS), and overall survival (OS). Late RT effects were diagnosed clinically by physician breast examinations during routine follow-up visits. The toxicity events routinely recorded in the database included fibrosis, edema, erythema, telangiectasia, and infection.
Statistical Analysis The c2 analysis identified significant baseline differences between the groups. The Kaplan-Meier (K-M) method was used to estimate 7-year outcomes. A Cox proportional hazard model was used to identify independent multivariate (MVA) predictors of each endpoint, adjusted for age, treatment group, tumor size, tumor grade, histologic type, lymphovascular invasion (LVI), surgical margin status, type of lymph node evaluation, RT technique, and treatment year. K-M survival curves were generated using the survival package in R (available at: www.r-cran-project.org). The other analyses were performed using SAS statistical software, version 9.3 (SAS Institute, Cary, NC).
Results Baseline patient information is listed in Table 1 according to treatment group. Those receiving ET were grouped according to whether they were adherent to ET. From 1981 to 2011, 504 women met the inclusion criteria, 197 (38%) had undergone RT alone and 311 (62%), RT plus ET. Of those prescribed ET, 94 (30.2%) were nonadherent. The median follow-up period for the entire cohort was 88 months (range, 13-288 months). Notable
Colin T. Murphy et al Table 1 Baseline Characteristics; Differences Compared Using the c2 Test Variable
RT Alone (%)
RT D ET-adherent (%)
RT D ET-nonadherent (%)
Patients (n)
193
217
94
Median
112
76
74
Range
13-285
13-232
14-288
80
22 (11)
11 (5)
12 (13)
Tumor stage
.510
T1
168 (87)
184 (85)
77 (82)
T2
25 (13)
33 (15)
17 (18)
0.5-1
76 (39)
50 (23)
32 (34)
>1-2
57 (30)
117 (54)
37 (39)
>2-5
19 (10)
33 (15)
17 (18)
Unknown
18 (9)
8 (4)
3 (4)
Invasive ductal
173 (90)
181 (83)
73 (78)
Invasive lobular
17 (9)
34 (16)
20 (21)
Other (invasive)
3 (2)
2 (1)
1 (1)
ERþ/PRþ
143 (74)
176 (81)
74 (79)
ERþ/PR
38 (20)
37 (17)
18 (19)
65 years, ERþ or PRþ, HER2, T1-T2N0). In our cohort, the baseline characteristics of the RT-alone and RT plus ET groups showed that the women in the RT-alone group were older and had more favorable disease characteristics. The RT-alone group had smaller primary tumors with a greater proportion that were well-differentiated and less likely to have close or positive margins. Tumor size independently predicted the risk of DM and DFS, and its strong prognostic association suggested that RT alone could result in acceptable outcomes for appropriately selected patients when the primary tumor is < 1 cm. These differences in patient selection could add to the low rates of overall disease recurrence in the RT-alone group—an estimated 6% at 7 years in the present study. As expected, each group of women in the present study had an estimated 7-year DFS rate that exceeded the estimated 7-year OS, underscoring the competing risk of death from causes other than breast cancer in this elderly, low-risk population. The strengths of the present study included the mature follow-up data, the requirement for known hormone-receptor positivity, a direct
Colin T. Murphy et al Table 2 Treatment Characteristics Stratified by Treatment Group, Compared Using the c2 Test Treatment Characteristic
RT Alone
RT Plus ET-adherent
RT Plus ET-nonadherent
Yes
34 (18)
85 (39)
33 (35)
No
159 (82)
132 (61)
61 (65)
Yes
136 (70)
96 (44)
46 (49)
No
57 (30)
121 (56)
48 (51)
Yes
23 (12)
36 (17)
15 (16)
No
170 (88)
181 (83)
79 (84)
P Value
Lymph node evaluation 80 years. The nonadherence rate in the present study was 30% and was the strongest predictor of increased risk of DM and worse DFS and OS.
Limitations A recently published cohort study aimed at prospectively investigating treatment decisions and outcomes in elderly women with breast cancer found that patient frailty (compared with robust patients) and increasing age predicted the risk of noninitiation and discontinuation of ET among elderly women.8 In the present study, neither performance status nor comorbidity was available for analysis, because these variables were not recorded in our institutional database. It is possible that the improved OS in the
Clinical Breast Cancer October 2015
- 387
Elderly Breast RT Alone Table 4 Seven-Year Kaplan-Meier Estimates of Outcomes, Compared Using the Log-Rank Test for Overall and Pairwise Comparisons Between Groups P Value
Outcome
7-Years Estimate (%) (95% CI)
All Groups
Local control (%) RT alone
98 (94-99)
RT plus ET-nonadherent
98 (85-100)
Regional control RT plus ET-adherent RT plus ET-nonadherent
.54
.3
.9
.361
.06
.2
.1
.012
.18
.97
.06
.176
