Physical activity and survival of postmenopausal, hormone receptor–positive breast cancer patients: Results of the Tamoxifen Exemestane Adjuvant Multicenter Lifestyle study

Authors

  • Nienke A. de Glas MD,

    Corresponding author
    1. Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands
    • Corresponding author: Nienke A. de Glas, MD, Department of Surgery, Leiden University Medical Center, P.O. Box 9600, Postzone K6-R, Room D-06-18, 2300 RC Leiden, The Netherlands; Fax: (011) 31 (0) 71-5266750; n.a.de_glas@lumc.nl

    Search for more papers by this author
    • The first 2 authors contributed equally to this work.

  • Duveken B. Y. Fontein MD,

    1. Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
    • The first 2 authors contributed equally to this work.

  • Esther Bastiaannet PhD,

    1. Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands
    2. Department of Gerontology and Geriatrics, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Anouk Pijpe PhD,

    1. Division of Psychosocial Research and Epidemiology, Netherlands Cancer Institute, Amsterdam, The Netherlands
    Search for more papers by this author
  • Anton J. M. De Craen PhD,

    1. Department of Gerontology and Geriatrics, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Gerrit-Jan Liefers PhD,

    1. Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Hans J. W. R. Nortier PhD,

    1. Department of Medical Oncology, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Hanneke J. C. J. M. de Haes PhD,

    1. Department of Medical Psychology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
    Search for more papers by this author
  • Cornelis J. H. van de Velde PhD,

    1. Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands
    Search for more papers by this author
  • Flora E. van Leeuwen PhD

    1. Division of Psychosocial Research and Epidemiology, Netherlands Cancer Institute, Amsterdam, The Netherlands
    Search for more papers by this author

  • We thank Janneke Willems for study coordination and administrative support, Wieke Heideman for data entry, and the Datacenter Surgery of the Leiden University Medical Center for its help with data management.

Abstract

BACKGROUND

Physical activity has been related to improved breast cancer outcomes. Especially in the older breast cancer population, physical activity may be important because old age is associated with comorbidities and decreased physical function. The purpose of this study was to investigate the relation between physical activity and overall survival, breast cancer–specific survival, and recurrence in several age groups of postmenopausal breast cancer patients.

METHODS

The Tamoxifen Exemestane Adjuvant Multinational Lifestyle study was a side study of the Tamoxifen Exemestane Adjuvant Multinational trial and prospectively investigated lifestyle habits of postmenopausal, hormone receptor–positive breast cancer patients. The relations between prediagnosis and postdiagnosis physical activity and overall survival, breast cancer–specific survival, and recurrence-free survival were assessed with Cox regression and competing risk regression models.

RESULTS

Among 521 patients, high levels of physical activity before and after the diagnosis were associated with better overall survival (the multivariate hazard ratios were 0.50 [95% confidence interval = 0.26-0.98] and 0.57 (95% confidence interval = 0.26-1.40] for patients who were very active before and after the diagnosis, respectively, in comparison with inactive patients). This was most evident in patients who were 65 years old or older. Physical activity was not significantly associated with breast cancer–specific survival or the relapse-free period.

CONCLUSIONS

Overall survival was better for patients who were active before and after the diagnosis. In contrast with previous studies, breast cancer survival and the risk of recurrence were not significantly associated with physical activity. These findings confirmed the need for more studies investigating the use of physical activity to supplement breast cancer treatment in older patients. Cancer 2014;120:2847–2854. © 2014 American Cancer Society.

INTRODUCTION

There is a growing body of evidence for the beneficial effects of physical activity in breast cancer patients. A recent meta-analysis has shown that physical activity improves body composition, physical function, psychological complaints, and quality of life in breast cancer patients.[1] In addition, physical activity has been found to improve fitness and strength and reduce depression and fatigue.[2]

Also, a multitude of studies have demonstrated the positive effects of physical activity in reducing the risk of developing breast cancer.[3] It has been suggested that so-called modifiable components such as obesity and insulin resistance, which are influenced by physical activity, may also affect cancer survival.[4, 5] We have previously reported in a systematic review that physical activity is associated with overall survival and breast cancer–specific survival, with postdiagnosis physical activity being most beneficial for these breast cancer outcomes.[6] However, none of the previous studies reported age-specific outcomes for patients aged 65 years or older, although this older breast cancer population may particularly benefit from physical activity.

Older breast cancer patients deserve extra attention because old age is related to decreased physical function and other comorbidities,[7, 8] and the maintenance of prediagnosis physical activity levels could play an important role in sustaining functional mobility in this more vulnerable population. Older breast cancer patients differ from younger breast cancer patients in many other aspects: they generally present with higher tumor stages[9] and are more frequently affected by comorbid conditions affecting physical functioning in comparison with their younger counterparts.[10]

The aim of this study was to assess the relation between recreational physical activity and overall survival, breast cancer–specific survival, and relapse-free survival in postmenopausal, hormone receptor–positive breast cancer patients of different ages.

MATERIALS AND METHODS

The Tamoxifen Exemestane Adjuvant Multicenter (TEAM) trial is an international, open-label randomized controlled trial comparing 2 regimens of hormonal treatment in postmenopausal, hormone receptor–positive breast cancer patients. The study design and population of the trial have previously been described in detail.[11] The current study is a side study of the TEAM trial, which is also known as the Tamoxifen Exemestane Adjuvant Multicenter Lifestyle (TEAM-L) side study. This side study prospectively investigated lifestyle habits in a subset of patients. The design and study population of TEAM-L have also been described elsewhere.[12] Briefly, TEAM-L consisted of 2 questionnaires taken 1 and 2 years after the diagnosis. In the first questionnaire, patients reported both prediagnosis (T0) and current (T1) physical activity levels, demographic factors, reproductive factors, and other lifestyle habits. The second questionnaire, taken 2 years after the diagnosis (T2), consisted of the same questions, with prediagnosis lifestyle characteristics excluded. Informed consent was obtained from all participating patients, and the study was approved by the medical ethics committee.

Data Collection

Patient, tumor, and treatment characteristics were collected through the TEAM data center.

For this study, patients were included if they completed either one or both questionnaires. If a patient had a recurrence before the questionnaire was filled out, the patient was excluded from the analysis.

Follow-up was defined as the time between filling out the questionnaire (with the T0 and T1 questionnaires taken 1 year after the diagnosis and the T2 questionnaire taken 2 years after the diagnosis) and an event, the last follow-up visit, or loss to follow-up. Follow-up was continued until August 15, 2012.

Physical Activity Measures

Questions assessing physical activity were derived from the validated European Prospective Investigation Into Cancer physical activity questionnaire.[13] Patients were asked to report the amount of time that they spent per week in the summer and winter on the following activities: walking, cycling, gardening, housekeeping, and sports. The mean number of hours spent at these activities per week in the summer and winter was calculated. Metabolic equivalent intensity (MET) values were calculated to include an estimate of the intensity of the reported physical activity.[14]

For this study, we evaluated recreational physical activity only, which was defined as walking, cycling, gardening, and sports. For the analyses, patients were grouped into quartiles based on the distribution of the total number of MET hours of physical activity per week in the whole study population. In addition, we performed all analyses in quartiles on the basis of the total number of absolute hours spent on physical activity per week.

Statistical Analyses

All analyses were 2-sided, and P values of 0.05 were considered statistically significant. If data were missing, patients were not excluded from the analyses, but they were analyzed in the unknown category for the same variable. There was an exception for patients with a missing status for physical activity measurements; these patients were excluded from the analyses.

The effects of the prediagnosis physical activity (T0) and the postdiagnosis physical activity (T2) on overall survival were assessed with Cox regression models. The multivariate models were adjusted for clinically relevant confounders. Breast cancer–specific survival was defined as the time until breast cancer death. The relapse-free period was defined as the time until disease recurrence or breast cancer death. Both breast cancer–specific survival and the relapse-free period were calculated with Fine and Gray analyses, which took the risk of competing mortality into account. Age-stratified analyses (<65 vs ≥65 years) were additionally performed for all outcome measures.

Because of the small numbers of events in some subgroup analyses, we performed sensitivity analyses that compared the 3 highest quartiles of physical activity (any physical activity) with the lowest quartile (very low). Because the number of hours spent on physical activity 1 year after the diagnosis (T1) could be affected by breast cancer treatment during this period, the effect of postdiagnosis physical activity was assessed on the basis of physical activity 2 years after the diagnosis (T2).

To assess potential effect modification by the body mass index (BMI), we performed an analysis in which we assessed the effect of postdiagnosis physical activity on all outcomes in combination with BMI by creating a variable that combined physical activity and BMI. We distinguished 4 groups: low physical activity (defined as the 2 lower quartiles) and BMI ≥ 25 kg/m2 (reference group), low physical activity and BMI < 25 kg/m2, high physical activity (defined as the 2 upper quartiles) and BMI ≥ 25 kg/m2, and high physical activity and BMI < 25 kg/m2.

We performed analyses for breast cancer–specific survival and the relapse-free period with Cox regression models to assess the influence of competing risks on our results. Finally, we presented causes of death by physical activity level.

RESULTS

Patient Characteristics

Overall, 521 patients were included in the analyses of prediagnosis physical activity. For the second analyses investigating postdiagnosis physical activity, 435 patients were included (Fig. 1). The median age of the patient population was 63.6 years (Table 1). Most patients had 1 or more comorbidities (72%), and the majority of the patients (55%) had a BMI of 25 kg/m2 or higher. Patient, tumor, and treatment characteristics did not differ between the 4 quartiles of prediagnosis physical activity with the exception of age and chemotherapy (Supporting Table 1; see online supporting information). The mean prediagnosis physical activity level was 10.9 hours/week, whereas the mean postdiagnosis physical activity level 2 years after the diagnosis was 10.0 hours/week.

Figure 1.

Inclusion of patients in the TEAM-L study. PA indicates physical activity; T1, 1 year after the diagnosis; TEAM, Tamoxifen Exemestane Adjuvant Multicenter; TEAM-L, Tamoxifen Exemestane Adjuvant Multicenter Lifestyle.

Table 1. Patient Characteristics
 All Patients (n = 521)
Patient characteristics 
Age at T1 (years), median (range)63.6 (44-88)
Age at T1, n (%) 
<60 years190 (36.5)
60-64 years110 (21.1)
65-69 years85 (16.3)
70-75 years71 (13.6)
>75 years65 (12.5)
Comorbidities, n (%) 
0146 (28.0)
1158 (30.3)
2117 (22.5)
3 or more100 (19.2)
BMI, n (%) 
<25 kg/m278 (32.4)
25-30 kg/m2107 (44.4)
>30 kg/m242 (17.4)
Unknown14 (5.8)
Tumor characteristics 
Grade, n (%) 
183 (16.6)
2232 (46.5)
3/4172 (34.5)
Unknown12 (2.4)
Tumor stage, n (%) 
T1/in situ255 (48.9)
T2232 (44.5)
T320 (3.8)
T411 (2.1)
Unknown3 (0.6)
Nodal status, n (%) 
N0143 (27.4)
N+376 (72.2)
Unknown2 (0.4)
Estrogen receptor, n (%) 
Positive511 (98.1)
Negative9 (1.7)
Unknown1 (0.2)
Progesterone receptor, n (%) 
Positive349 (67.0)
Negative120 (23.0)
Unknown52 (10.0)
Therapy 
Most extensive surgery, n (%) 
No resection1 (0.2)
Wide local excision256 (49.1)
Mastectomy262 (50.3)
Unknown2 (0.4)
Most extensive axillary surgery, n (%) 
Sentinel node117 (22.5)
Axillary lymph node dissection398 (76.4)
Unknown6 (1.2)
Radiotherapy, n (%) 
No174 (33.4)
Yes340 (65.3)
Unknown7 (1.3)
Chemotherapy, n (%) 
No334 (64.1)
Yes187 (35.9)
Endocrine therapy, n (%) 
Tamoxifen followed by exemestane263 (50.5)
Exemestane alone258 (49.5)

Prediagnosis Physical Activity

Overall survival was better for patients who were physically active before their diagnosis according to both analyses based on MET hours and analyses based on the absolute number of hours. This effect remained similar with additional adjustments for the number of comorbidities, T stage, N stage, mean BMI, and chemotherapy (hazard ratio [HR] = 0.57, 95% confidence interval [CI] = 0.31-1.05 for the second lowest quartile of physical activity [low physical activity], HR = 0.48, 95% CI = 0.25-0.91 for the second highest quartile of physical activity [high physical activity], and HR = 0.50, 95% CI = 0.26-0.98 for the highest quartile of physical activity [very high physical activity] vs the lowest quartile of physical activity [very low physical activity], P = .10; Table 2).

Table 2. Physical Activity With Respect to Survival Outcomes
 Overall SurvivalBreast Cancer–Specific SurvivalRecurrence-Free Period
EventsHRa95% CIPEventsHRb95% CIPEventsHRc95% CIP
  1. a

    Abbreviations: CI, confidence interval; HR, hazard ratio; MET, metabolic equivalent intensity; Ref, reference value; T1, 1 year after the diagnosis.

  2. b

    Cox regression analyses adjusted for age (T1), number of comorbidities, T stage, N stage, body mass index (mean), and chemotherapy.

  3. c

    Fine and Gray analyses adjusted for age (T1), number of comorbidities, T stage, N stage, body mass index (mean), and chemotherapy.

  4. d

    Fine and Gray analyses adjusted for age (T1), T stage, N stage, body mass index (mean), and chemotherapy.

Prediagnosis physical activity (n = 521)MET (h/wk)            
Very low (0-22.3)34Ref 0.1017Ref 0.7521Ref 0.74
Low (22.4-41.5)170.570.31-1.05 130.810.35-1.87 211.070.55-2.11 
High (41.6-70.8)150.480.25-0.91 80.510.22-1.18 140.670.33-1.38 
Very high (>70.8)140.500.26-0.98 120.830.38-1.80 201.040.53-2.02 
h/wk            
Very low (0-5.0)37Ref 0.2019Ref 0.8726Ref 0.60
Low (5.1-8.5)140.500.27-0.95 90.810.35-1.86 151.080.55-2.11 
High (8.6-14.0)130.400.21-0.77 90.510.22-1.16 150.680.33-1.39 
Very high (>14.0)160.650.35-1.21 130.840.39-1.82 201.040.53-2.04 
Postdiagnosis physical activity (n = 435)MET (h/wk)            
Very low (0-21.0)24Ref 0.3414Ref 0.9517Ref 0.79
Low (21.1-40.0)100.430.19-0.94 60.440.15-1.35 90.540.23-1.29 
High (40.1-65.5)130.600.29-1.24 111.000.13-1.32 180.970.44-2.13 
Very high (65.6-258)110.570.26-1.40 80.770.28-2.12 140.900.39-2.10 
h/wk            
Very low (0-4.6)25Ref 0.5115Ref 0.7719Ref 0.69
Low (4.7-6.3)90.400.18-0.89     120.220.02-2.25 
High (6.4-13.5)110.480.23-1.03 72.450.40-14.95 111.370.35-5.38 
Very high (>13.5)130.630.31-1.32 90.890.16-4.94 160.720.22-2.33 

Breast cancer–specific survival and recurrence-free periods were not statistically significant in association with physical activity levels, nor was there a trend toward better outcomes with increasing physical activity levels. The analyses performed with categories based on the absolute hours per week showed similar results (Table 2).

Postdiagnosis Physical Activity

Postdiagnosis survival was assessed for all patients who completed the T2 questionnaire and did not have a recurrence before then (n = 435). Overall survival was statistically significantly better for patients who were relatively active (HR = 0.43, 95% CI = 0.19-0.94 for low physical activity, HR = 0.60, 95% CI = 0.29-1.22 for high physical activity, and HR = 0.57, 95% CI = 0.26-1.40 for very high physical activity versus very low physical activity, P = .34; Table 2). Breast cancer–specific survival and recurrence-free periods did not differ between the 4 quartiles of physical activity (Table 2).

Age-Specific Results

As for prediagnosis physical activity, age-stratified analyses (<65 vs ≥65 years at the diagnosis of breast cancer) demonstrated that overall survival was better for active patients who were 65 years old or older versus relatively inactive patients (multivariate HR = 0.63, 95% CI = 0.30-1.35 for low physical activity, HR = 0.44, 95% CI = 0.19-1.03 for high physical activity, and HR = 0.34, 95% CI = 0.11-1.01 for very high physical activity versus very low physical activity; Table 3). No significant differences between the 4 quartiles were observed for any of the outcomes for patients who were younger than 65 years. Breast cancer–specific survival and relapse-free periods were not significantly associated with physical activity levels in either age group.

Table 3. Prediagnosis Physical Activity With Respect to Survival Outcomes Stratified by the Age at Diagnosis
 Overall SurvivalBreast Cancer–Specific SurvivalRecurrence-Free Period
EventsHRa95% CIPEventsHRb95% CIPEventsHRc95% CIP
  1. a

    Abbreviations: CI, confidence interval; HR, hazard ratio; MET, metabolic equivalent intensity; Ref, reference value; T1, 1 year after the diagnosis.

  2. b

    Cox regression analyses adjusted for age (T1), number of comorbidities, T stage, N stage, body mass index (mean), and chemotherapy.

  3. c

    Fine and Gray analyses adjusted for age (T1), number of comorbidities, T stage, N stage, body mass index (mean), and chemotherapy.

  4. d

    Fine and Gray analyses adjusted for age (T1), T stage, N stage, body mass index (mean), and chemotherapy.

Age < 65 years at diagnosis (n = 300)MET (h/wk)            
Very low (0-22.3)7Ref 0.734Ref 0.827Ref 0.80
Low (22.4-41.5)70.590.21-1.72 50.710.17-2.91 90.740.26-2.13 
High (41.6-70.8)70.600.21-1.72 60.900.27-3.08 70.610.21-1.74 
Very high (>70.8)100.730.28-1.95 81.020.33-3.20 120.920.36-2.30 
MET (h/wk)            
Very low (0-22.3)7Ref 0.324Ref 0.837Ref 0.54
Any physical activity (>22.3)240.650.28-1.52 190.890.30-2.59 280.760.33-1.79 
Age ≥ 65 years at diagnosis (n = 221)MET (h/wk)            
Very low (0-22.3)27Ref 0.0413Ref 0.4014Ref 0.70
Low (22.4-41.5)100.630.30-1.35 80.970.35-2.66 121.600.68-3.74 
High (41.6-70.8)80.440.19-1.03 20.250.06-1.01 70.730.27-2.01 
Very high (>70.8)40.340.11-1.01 40.620.19-2.02 80.980.37-2.59 
MET (h/wk)            
Very low (0-22.3)27Ref 0.0213Ref 0.2614Ref 0.80
Any physical activity (>22.3)220.490.27-0.91 140.610.26-1.45 261.100.53-2.30 

Sensitivity Analyses

Additional sensitivity analyses in which the upper 3 quartiles in the age-specific analyses were combined showed similar results for the 2 age groups (Table 3).

None of the combinations of postdiagnosis physical activity and BMI were significantly different with respect to overall survival, breast cancer–specific survival, or the relapse-free period (Supporting Table 2).

Furthermore, Cox regression analyses for both prediagnosis and postdiagnosis physical activity with respect to breast cancer–specific survival and the relapse-free period showed results similar to those of the Fine and Gray analyses (data not shown). Finally, an analysis of causes of death by physical activity level showed that the majority of patients who died from causes other than breast cancer died from second malignancies or cardiovascular disease (Supporting Table 3).

DISCUSSION

This study showed that physical activity before and after the diagnosis was associated with an improvement in overall survival. Especially in patients aged 65 years and older, a high level of prediagnosis physical activity was related to better overall survival in comparison with lower levels of physical activity. No trends toward better breast cancer–specific survival or a longer recurrence-free period with increasing physical activity levels were observed.

Although we did not observe a significant association between physical activity and breast cancer death or recurrence, many previous studies have demonstrated favorable effects of physical activity on breast cancer–specific survival in breast cancer survivors.[4, 6, 15] Several mechanisms could explain these contradictory results. First, the Dutch population is, in general, more active than other Western societies,[16] especially in comparison with the United States, where most of the previous studies were performed.[4, 15] Furthermore, the TEAM trial included postmenopausal women who generally had a high socioeconomic status and a low comorbidity burden in comparison with the general population.[17] For this reason, it is possible that even patients in the very low physical activity category were still rather active; in this case, differences in outcomes were absent. For comparison, the Nurses' Health Study divided patients into 5 groups according to the number of MET hours of physical activity per week; the lowest category consisted of patients who were active for less than 3 MET hours/week.[4] In contrast, our lowest physical activity category consisted of patients who were active for less than 22 MET hours/week, and this could explain the differences.

Second, the TEAM trial included relatively many older patients.[18] Because the risk of dying from causes other than breast cancer (competing causes of death) strongly increases with age,[18] the absolute benefits of physical activity possibly could be less pronounced in this population. However, competing risk analyses did not differ from regular analyses of our data. Cardiovascular mortality especially could be improved in active patients,[19] but the number of events was too small to investigate this in further detail (Supporting Table 3).

The biological mechanism behind improved breast cancer outcomes for physically active patients, as observed in previous studies, remains to be elucidated. It has been suggested that obesity and insulin resistance play important roles in influencing the risk of breast cancer recurrence and mortality.[20, 21] Obesity and high levels of insulin and diabetes have been associated with increased breast cancer risk,[22, 23] and physical activity is known to lower insulin levels and improve insulin sensitivity. Furthermore, it has been suggested that physical activity could result in a reduction of levels of endogenous estrogens and body fat, which could lead to slower progression of breast cancer.[24] Previous studies have also shown that obesity can lead to poorer breast cancer outcomes.[21] Pierce et al[5] investigated interactions between specific lifestyle factors and mortality in breast cancer patients and found significant improvements in 10-year mortality in patients who were physically active and had a healthy dietary pattern, regardless of BMI. Although we did not take into account dietary patterns, the interaction between BMI and physical activity with respect to outcomes was not statistically significant, and there was no statistically significant association between BMI, physical activity, and survival. Notably, in the older population, a high BMI may even have a protective effect on overall survival,[25] which may reflect the absence of an association in our results.

Interestingly, the results based on the number of MET hours were similar to the results of the analyses based on the absolute number of hours. This suggests that the intensity of physical activity is not related to the association between physical activity and improved overall survival. Also, no trend was observed between increasing levels of physical activity and better overall survival, and this indicates that any activity beyond a certain level improves overall survival. This could be of major importance to developing future physical activity interventions for breast cancer patients.

As for the age-specific results, we found the strongest association between physical activity and overall survival in patients who were 65 years old or older. This is particularly interesting because old age is associated with comorbidities and poor physical function.[26, 27] In developed countries, more than 40% of the breast cancer population is 65 years or older, and this percentage is expected to increase even further in upcoming years.[28] The treatment of breast cancer in older patients is still not evidence-based because older patients are often not included in clinical trials on account of restrictions of age, comorbidities, and physical function.[29] This leads to a large diversity of treatment strategies and poor adherence to guidelines among treating specialists.[30] Consequently, breast cancer–specific mortality increases with age, even amid competing causes of death in older patients.[18] Therefore, new treatment strategies for elderly breast cancer patients are urgently needed.

Although physical activity was not related to breast cancer outcome in our data, it could play an important role in preventing adverse events and noncompliance with hormonal therapy in the older breast cancer population.[31] Breast cancer in older patients is often hormone receptor–positive,[32] and consequently, many patients receive hormonal therapy. Aromatase inhibitors often cause side effects such as arthralgia, musculoskeletal pain, and diminished mineral bone density.[31] Older patients are especially at increased risk for the development of fractures after hormonal treatment.[31] In the TEAM study, older patients were more likely to be noncompliant within the first year of hormonal treatment than younger patients. The most important reasons for nonpersistence were adverse events.[33] Therefore, maintaining physical activity levels might be extra beneficial for the older breast cancer population.

Strengths and Limitations

A major strength of this study is its prospective design. Data were registered at various time points during follow-up with validated questionnaires, and this resulted in a homogeneous study population with well-registered, reliable data. Patients in this side study were a representative subset of the international TEAM trial, as demonstrated in an earlier report by Voskuil et al.[12] To our knowledge, this is the first study to investigate differences in physical activity levels and survival outcomes in older breast cancer patients versus younger patients. Needless to say, our study is limited by the number of events, which was relatively low in comparison with some other studies.[4, 15, 34] Furthermore, patients who exercise regularly may also be more likely to adopt other healthy lifestyles.[5] The more favorable overall survival demonstrated for patients older than 65 years may, therefore, be the result of selection because patients who are capable of participating in physical activity are fit enough to be able to exercise and hence survive longer than unfit patients. To establish a causal relation between physical activity and breast cancer outcomes, randomized clinical trials should be performed, but inevitably, these kinds of trials are not feasible. Therefore, observational data (as from the current study) can still be of value for investigating the association between physical activity levels and breast cancer outcomes.

Finally, the reported number of hours of physical activity was relatively high in comparison with other studies. This may be due to overreporting of prediagnosis physical activity. Nevertheless, the Dutch population is known to engage in high levels of physical activity, especially in walking, cycling, gardening, and sports.[16] Most women in the Netherlands (76%) spend at least 30 minutes on moderate to vigorous physical activity at least 5 days of the week.[35]

In conclusion, overall survival was better for patients who were physically active before and after the diagnosis, whereas breast cancer recurrence and survival were not associated with physical activity. Especially in women aged 65 years or older, high physical activity levels before the diagnosis were associated with improved overall survival. These findings confirm the need for more studies investigating the role of physical activity in supplementing breast cancer treatment in older patients. Physical activity could play an important role in improving survival, maintaining physical function, and improving quality of life both during and after breast cancer treatment, especially in older patients.

FUNDING SUPPORT

The TEAM trial was funded by Pfizer Oncology. The TEAM-L study was funded by the Dutch Pink Ribbon Foundation

CONFLICT OF INTEREST DISCLOSURE

The authors made no disclosure.

Ancillary