You are here
Age and the effect of physical activity on breast cancer survival: A systematic review
Cancer Treatment Reviews, 8, 39, pages 958 - 965
The effect of physical activity (PA) on cancer survival is still the topic of debate in oncology research focusing on survivorship, and has been investigated retrospectively in several large clinical trials. PA has been shown to improve quality of life, fitness and strength, and to reduce depression and fatigue. At present, there is a growing body of evidence on the effects of PA interventions for cancer survivors on health outcomes. PA and functional limitations are interrelated in the elderly. However the relationship between breast cancer survival and PA in older breast cancer patients has not yet been fully investigated. Our systematic review of the existing literature on this topic yielded seventeen studies. Most reports demonstrated an improved overall and breast cancer-specific survival. Furthermore, in studies that compared younger women with older or postmenopausal women, it was suggested that the beneficial effect of PA may be even greater in older women. Understanding the interaction between physical functioning and cancer survival in older breast cancer patients is key, and may contribute to successful treatment and survival. In this population of cancer survivors it is therefore imperative to embark on research focused on improving physical functioning in the context of comorbidities and functional limitations.
Keywords: Breast cancer, Physical activity, Elderly, Survival, Mortality.
There is a growing body of evidence on the effects of physical activity (PA) interventions for cancer survivors on health outcome. PA in breast cancer has shown to improve quality of life, fitness and strength and to reduce depression and fatigue. 1 Also, PA has shown to have an inverse association with postmenopausal breast cancer risk. 2 Previous randomized controlled trials have demonstrated better quality of life and other health outcomes in patients who undertake regular PA. 3 The effect of PA on cancer survival after breast cancer diagnosis, however, is still the topic of debate in oncology research. 4
Breast cancer is increasingly becoming a disease of the elderly and, simultaneously, the number of breast cancer survivors is steadily rising. 5 For example, in the United Kingdom, 45% of breast cancer patients are older than 65 years at diagnosis. 6 In elderly patients, cancer occurs in a background of normal aging and comorbidity, thereby making this specific population heterogeneous in nature. Although older patients are known to have a higher disease-specific mortality, more than 65% of breast cancer patients older than 75 years die of other causes than breast cancer. 7 In contrast to younger patients, survival in the elderly breast cancer population has not improved in the last decade. 8 While breast cancer has a profound effect on psychological functioning and quality of life in younger females, elderly patients are affected by the disease physically more than psychologically. 9 At an older age, the decline in physical functioning may be induced or amplified as a result of local and/or systemic treatments. Regardless of age, decline in physical functioning is associated with higher mortality. 10 Therefore, new strategies to decrease the effect of breast cancer on physical decline and to improve overall and disease-specific survival are needed. The purpose of this study was to review the current literature in relation to the effect of PA on survival in breast cancer patients, with a focus on the elderly breast cancer patient in particular.
Materials and methods
Studies were identified through systematic review of the available literature in the PubMed database, EMBASe, Cochrane and Web of Science up to November 9th, 2012. PubMed, EMBASE, Cochrane and Web of Science were all searched using the following set of search terms which described breast cancer, physical activity, survival and elderly patients according to the following algorithm: (“Breast Neoplasms”[majr] OR “breast cancer”[ti] OR “Breast Neoplasm”[ti] OR “Breast Tumors”[ti] OR “Breast Tumor”[ti] OR “Breast Tumours”[ti] OR “Breast Tumour”[ti] OR “Breast Carcinoma”[ti] OR “Breast Carcinomas”[ti] OR “Cancer of the Breast”[ti] OR “Cancer of Breast”[ti]) AND (“physical activity” OR “Motor activity”[mesh] OR “Sports”[mesh] OR “Exercise”[mesh] OR Exercise[ti] OR Exercises[ti] OR “Resistance Training”[ti] OR Running[ti] OR Jogging[ti] OR Swimming[ti] OR Walking[ti] OR “Sports”[mesh] OR “Physical Exertion”[Mesh] OR exertion[ti]) AND (survival OR surviv∗ OR “Survival”[Mesh] OR “Mortality”[Mesh] OR “mortality”[Subheading] OR “Survival Rate”[Mesh] OR “Survival Analysis”[Mesh]) AND (“Aged”[mesh] OR Elderly OR aged OR Adult OR Adults OR “Adult”[mesh] OR woman[tiab] OR women[tiab]) AND (eng[la]) AND (“Cohort Studies”[mesh] OR prognosis).
Two independent reviewers (D.F., N.G.) selected studies for inclusion in the review. Only reports written in English were eligible. Articles included original studies in which the effect of PA on survival outcomes was investigated in breast cancer patients only. Studies were included if the primary or secondary outcome was related to the influence of PA on overall, breast cancer-specific survival/mortality or recurrence after breast cancer diagnosis. PA was defined as any physical activity relating to aerobic, endurance, or strength training for the purposes of recreation, household, commuting, or work. Where evident, activities including yoga, Pilates, Tai Chi, or stretching were not included in the analyses. There were no restrictions with regard to patient characteristics, breast cancer subtype, treatment, age, condition, comorbidities, and other features. Only studies investigating leisure or total activity (occupational and/or non-occupational) were selected. Studies measuring PA using biological parameters (i.e. cardiorespiratory fitness, energy expenditure (in kJ or kcal), oxygen use (in vO2/vCO2), etc.) were not included in our selected abstracts. In case of an update of previously published data, the most recent publication was included. Reports on pilot data or descriptions of a study design were excluded from this review. Studies were also excluded if the PA intervention was combined with another intervention. To ensure that no other studies were missed for inclusion in the current review, additional articles were identified by a manual search of the reference list of the selected manuscripts.
Data extraction was performed by 3 reviewers (D.F., N.G. and M.D.). Relevant data was determined prior to reading the selected articles. For all included studies, we documented the type of study, inclusion period, year of publication, primary and (where possible) secondary endpoints. Results from each study was documented and independently reviewed and verified by two independent reviewers (D.F. and N.G.). Aspects of PA were the type of activity, PA measurement unit (including frequency, duration and intensity, which were frequently converted into Metabolic Equivalent Task (MET) hours), time and timeframe of measurement(s), method of data collection, number of patients, mean and median age and age range, in- and exclusion criteria, as well as tumor- and patient-related characteristics. Results for specific population subgroups, including age (and/or menopausal status) and Body-Mass Index (BMI) were documented where available. Due to the fact that ages varied greatly between studies, we defined ‘elderly’ as 65 years or older. Primary and/or secondary endpoints included all-cause mortality, breast cancer-specific mortality, and breast cancer recurrence.
A total of 497 unique articles were identified in the PubMed, EMBASe, Cochrane and Web of Science searches, of which seventeen studies were deemed eligible after applying in- and exclusion criteria to the titles and abstracts.( Table 1 and Fig. 1 ) A total of 35,026 breast cancer patients were included in our review. Only one study investigated solely postmenopausal patients. 11 In most studies, PA was measured in the first year after primary diagnosis. Two studies only included patients who were diagnosed with early-stage breast cancer. Fourteen studies also included patients with a more advanced breast cancer stage.
|Study||Patients||Age||Type of study||Measurement of PA||Details of study||Primary outcome||Overall results|
|1||Bertram et al.||2361 BC survivors stage I-III, 1 year post-treatment||18–70, median age 54.3||Side study of randomized controlled trial of dietary change (WHEL-study)||9-item physical activity measure adapted from the WHI. Conversion to MET-h||Questionnaire at baseline and 12-months. Post-treatment and change in PA||Invasive breast-cancer events and all-cause mortality||High baseline (post-treatment) PA (conform guidelines) lead to 35% reduction in all-cause mortality (HR 0.39 (0.21–0.72) for highest quintile total PA versus no PA). Change in PA had no effect on outcomes|
|2||Cleveland et al.||1508 early-stage BC patients||Mean age 58.6||Population-based longitudinal study (Long Island Breast Cancer Study Project)||MET-h/wk||Baseline interview post-diagnosis on pre-diagnosis recreational PA||5-year all-cause mortality and BC-specific mortality||Significant lower risk for BC-specific mortality (HR 0.63 (043–0.92)) (any total lifetime RPA versus no RPA), significantly lower risk for all-cause mortality (HR 0.58 (0.43–0.78))|
|3||Dal Maso et al.||1453 BC patients||23–74, median age 55||Multicentre case-control study||Retrospective questionnaire on PA at various ages||Questionnaire completed once max 1 year post-diagnosis||All-cause mortality and BC-specific mortality||Leisure time PA was not associated with BC mortality: HR = 0.85(0.68–1.07) for PA < 2h/wk versus >2/wk Leisure time PA was not associated with all-cause mortality: HR = 0.92 (0.67–1.01) for PA < 2h/wk versus >2/wk Occupational PA was not associated with BC mortality or all-cause mortality|
|4||Emaus et al.||1364 BC patients||17–79, mean age 57.5||Population-based survival study (Norwegian Counties Study)||Questionnaire on pre-diagnostic PA||Questionnaire on pre-diagnosis PA (for the year preceding each survey)||All-cause mortality and BC-specific mortality||Overall mortality was not significantly better in active patients (HR 0.74 (0.51–1.08)) (highest category versus lowest category of PA) or with BC-specific mortality (HR 0.75 (0.49–1.15))|
|5||Friedenreich et al.||1183 BC patients||Mean age 56||Population-based sample of breast cancer survivors||Assessment of lifetime PA. Conversion to MET-h||Questionnaire on lifetime recreational, household and occupational PA, recorded at baseline only||All-cause mortality, BC-specific mortality, recurrence, progression, new primary||Vigorous intensity was not related to recurrence, BC-specific mortality and all-cause mortality. However, moderate intensity recreational physical activity was beneficial for all the outcomes (HR BC-specific mortality 0.66 (0.48–0.91), HR all-cause mortality 0.76 (0.58–1.06 with p = 0.05), HR recurrence 0.48–0.91)|
|6||Hellman et al.||528 primary BC patients||33–95, median age 66.9||Prospective population-based study (Copenhagen City Heart Study)||Self-administered questionnaire on PA||Self-reported pre-diagnosis PA||All-cause mortality||High PA (>4 h/wk) versus moderate PA (2–4 h/wk) versus no PA was not associated with reduced all-cause mortality after BC diagnosis (Moderate: HR = 1.07, 0.77–1.49; High: HR = 1.00, 0.69–1.45).High versus moderate versus no PA was not associated with reduced BC-specific mortality (Moderate: HR = 0.73, 0.50–1.08; High: HR = 0.70, 0.46–1.09).|
|7||Holick et al.||4482 BC patients, 91% postmenopausal.||20–79, mean age 58.5||Prospective population-based study (Collaborative Women’s Longevity Study (CWLS))||CWLS questionnaire. Conversion to MET-h||Telephone interview within 2 years of diagnosis. Survivors completed a questionnaire for recent post-diagnosis PA.||All-cause mortality and BC-specific mortality||Compared with low recent post-diagnosis recreational PA (<2.8MET-h/wk), higher PA was associated with a decreased risk of BC death. Overall mortality was also improved with increasing levels of PA. When stratified, moderate-intensity recreational PA had a significant beneficial effect on both outcomes, but vigorous-intensity PA did not|
|8||Holmes et al.||2987 nurses with stage I/II/III BC||30–55 on enrolment in NHS||Prospective observational study (Nurses’ Health Study)||Questionnaire on PA, calculated in MET-h/wk (<3/3–8/9–14/15–23/>24)||Questionnaires every 2 years from 1976–2004 for current PA.||BC-specific mortality||The HR for BC mortality was 0.80 (0.60–1.06) for 3–8.9 MET-h/wk; 0.50 (0.31–0.82) for 9–14.9 MET-h/wk; 0.56 (0.38–0.84) for 15 to 23.9 MET-h/wk; and 0.60 (0.40–0.89) for 24 or more MET-h/wk|
|9||Irwin et al.||933 BC patients with local or regional breast cancer||Mean age 55.5 (6 months post-diagnosis)||Population based multicentre prospective cohort study (Health, Eating, Activity and Lifestyle (HEAL) Study)||Modifiable Activity Questionnaire in MET-h/wk (all types of PA) light (<3) moderate (3–6), vigorous (>6)||In-person baseline interview (4–8 months) post-diagnosis. Follow-up interview 3 years post-diagnosis||BC-specific mortality and all-cause mortality||Compared with inactive women (0 MET-h/wk) BC mortality in pre-diagnostic PA: >9MET-h was not significantly better (HR 0.69 (0.45–1.06)), but post-diagnosis PA was beneficial (HR 0.33 (0.15–0.73)). Increase in PA lead to non-significant 45% lower risk of all-cause mortality (HR 0.55,(0.22–1.38)); decrease in PA: HR = 3.95 (1.45–10.5)|
|10||Irwin et al.||4643 post-menopausal BC patients||50–79||Prospective longitudinal study (Women’s Health Initiative (WHI))||Questionnaire on PA. Conversion to MET-h||Questionnaire on recreational and household PA at baseline, year 3 and year 6 post-diagnosis||All-cause mortality and BC-specific mortality||Moderate- to vigorous-intensity post-diagnosis PA lead to a lower risk for all-cause mortality (HR 0.54 (0.38–0.79)) and a lower risk for BC-specific mortality (HR 0.61 (0.35–0.99)). Increase of PA after diagnosis of 9 MET-h/wk or more lead to 33% less risk of all-cause mortality (HR 0.67 (0.46–0.96))|
|11||Pierce et al.||1490 early-stage BC patients||<70, mean age 50||Side study of randomized controlled trial of dietary change (WHEL-study)||9-item PA measure adapted from the WHI. Conversion to MET-h||Baseline PA questionnaire||All-cause mortality||Compared with low PA (0–225 MET-min/wk), high PA was associated with lower all-cause mortality 225–636 MET-min/wk: HR = 0.86, 636–1320 MET-min/wk: HR = 0.76, 1320–6420 MET-min/wk: HR 0.58 (p = 0.02, no 95%-CI reported)|
|12||Rohan et al.||451 BC patients||20–74||Population-based cohort||Questionnaire on recreational PA. Conversion to total kcal expended per wk||Reported pre-diagnosis recreational PA (divided into winter and summer PA)||BC-specific mortality||There was no association between PA and BC-specific mortality in the whole study group (ref. 0 kcal/wk). There was also no association between PA and BC-specific mortality when stratified by PA level (light, moderate, and vigorous PA).|
|13||Sternfeld et al.||1970 early-stage BC patients||18–79, mean 60.6||Prospective, observational study (Life After Cancer Epidemiology (LACE) Study)||Questionnaire based on the Arizona Activity Frequency Questionnaire. Conversion to MET-h||Semi-annual questionnaire about primary outcomes (self-reported)||All-cause mortality, BC-specific mortality, BC recurrence||There was no association between total PA and all-cause mortality, BC-specific mortality and recurrence, but there was a significantly beneficial effect of moderate activity on all-cause mortality (HR 0.66 (0.42–1.03, p for trend = 0.04)), but not on BC-specific mortality and recurrence.|
|14||Keegan et al.||4153 BC patients (320 patients excluded from Australian cases)||18–69||Population-based cancer registries (Breast Cancer Family Registry)||Questionnaires administered post-diagnosis. Conversion to MET-h.||Lifetime history of recreational PA and 3 years pre-diagnosis||All-cause mortality||Any recent moderate/vigorous PA versus no activity showed a significantly lower all-cause mortality (HR = 0.73, (95%CI 0.60–0.89). Stratified by Estrogen Receptor (ER) status, only ER positive patients had a benefit of PA (HR = 0.67 (95%CI 0.4–0.94)) for active versus non-active patients, ER-negative patients: HR = 1.05 (95%CI 0.66–1.68)|
|15||West-Wright et al.||3,539 BC patients with invasive BC||26–94, mean age 58.9||Prospective Cohort Study (California Teacher’s Study)||PA questionnaire on recreational PA. h/wk and months per year recorded.||PA questionnaire for long-term and recent pre-diagnosis recreational PA||BC-specific mortality, all-cause mortality||BC-specific mortality was significantly better for long-term intermediate PA HR 0.65 (95%CI 0.45–0.93) and high PA HR 0.53 (0.35–0.80) versus low PA, (p = 0.003). For recent PA there was no beneficial effect for BC-specific mortality intermediate PA HR 1.17 (0.84–1.65), high PA HR 1.08 (0.73–1.58), p = 0.69. Overall survival was better in long-term PA: intermediate PA HR 0.83 (0.65–1.07) and high PA (0.73 (0.55–0.96) versus low PA (p = 0.03). Recent PA was borderline significantly beneficial for overall survival: intermediate PA HR 0.89 (0.71–1.11), high PA HR 0.78 (0.60–1.02) versus low PA (p = 0.06)|
|16||Enger et al.||717 BC patients with in situ or invasive BC||21–40||Population-based case-control study||Lifetime PA measured in h/wk of recreational exercises.||In-person interview on lifetime PA (from first menses) up to 12 months pre-diagnosis||BC-specific mortality||There was no significant benefit of PA: 0.1–0.7h: HR 0.86 (0.56–1.32); 0.8–1.6 h: HR 0.59 (0.35–1.01); 1.7–3.7h: HR 0.87 (0.57–1.33); 3.8 + h: HR 1.30 (0.81–2.09); p(trend)=0.88|
|17||Abrahamson et al.||1264 BC patients with invasive BC||20–54, median 42||Population-based cohort study (parent study)||Relative MET scores for vigorous (9 METs) and moderate (5 METs) activities. Scores weighted by frequency of the activity per wk.||In-person interview for average PA at 12–13 years, 20 years, and the year before diagnosis.||All-cause mortality||There was no significantly beneficial effect of PA in year before diagnosis for: Q2 HR 0.86 (0.63–1.18); Q3: HR 0.81 (0.6–1.12); Q4: HR 0.78 (0.56–1.08) versus quartile Q1. (p = 0.10)|
BC: breast cancer; HR: Hazard Ratio; PA: physical activity; h: hours; MET-h: Metabolic Equivalent Task hours.
The majority of the studies assessed were prospective observational cohort studies.11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, and 23 Two studies were part of the same side study of a randomized controlled trial investigating dietary change.24 and 25 A final study was a retrospective analysis of PA in patients participating in a multicenter case-control study. 26
In all studies, data collection was through self-administered and/or interviewer-administered questionnaires, performed either in person or by telephone. Questionnaires were completed either once12, 13, 14, 15, 18, 19, 21, 22, 23, 25, and 26 or on two or more occasions11, 16, 17, 20, 24, and 27 Questionnaires included inquiries on occupational, recreational and/or household activities, either combined or separately. Furthermore, data were collected on either current, pre-diagnostic or post-diagnostic PA, or a combination of these. For most studies11, 14, 16, 17, 20, 23, 23, 24, 25, and 27 Metabolic Equivalent Task (MET)-hours were used (or calculated) to assess PA in relation to survival outcomes. MET-hours are a validated measurement of PA, and are frequently used to define intensity and duration of PA. One MET-hour is defined as the equivalent of sitting quietly, or 3.5 ml oxygen*kg−1 body weight*min−1 (1 kcal*kg−1 body weight*h−1). 28 The remaining studies used either the absolute number of hours of PA12, 15, 21, 22, and 26 or the total kcal expended per week. 19 A majority of studies used validated questionnaires, including the Modifiable Activity Questionnaire 17 and the Collaborative Women’s Longevity Study (CWLS) questionnaire. 27 In most studies, tumor and treatment characteristics, sociodemographic characteristics, lifestyle habits (education, occupation, smoking, alcohol use), menstrual and reproductive history, exogenous hormone use, and comorbidities were also documented.
Fourteen studies assessed all-cause mortality,11, 12, 13, 14, 15, 17, 18, 20, 21, 23, 24, 25, 26, and 27 12 studies assessed breast cancer-specific mortality,11, 12, 13, 14, 16, 17, 19, 20, 21, 22, 26, and 27 and three studies investigated breast cancer recurrence as primary endpoint(s).14, 20, and 24 All hazard ratios (HR) reported in this review concern the highest amount of PA versus the lowest amount of PA, the lowest level of PA being the reference value. All reported hazard ratios are multivariable adjusted.
Effects of physical activity on breast cancer-specific and all-cause mortality
Eight out of 12 studies that investigated the effect of PA on breast cancer-specific survival found a significant advantage of PA on survival, ranging from a 36–67% reduction in risk of death for the highest amount of PA versus no PA (HR ranging from 0.53 (95% C.I. 0.35–0.80) to 0.80 (no 95% C.I. reported).11, 12, 13, 14, 16, 17, 18, and 27 Three of the remaining four studies found a non-significant benefit (HR 0.85 (0.68–1.07), HR 0.72 (0.29–1.81), HR 0.90 (0.51–1.58)),19, 20, and 26 while the fourth study found a non-significant worse breast cancer-specific survival (HR 1.30 (0.81–2.09)).
Fourteen studies described the effect of PA on all-cause mortality, of which ten studies found a significant improvement in all-cause mortality with increasing PA.13, 14, 17, 18, 20, 21, 24, 25, and 27 This effect ranged from 14% to 56% (HR ranged from 0.39 (0.21–0.72) to 0.73 (0.60–0.89). The remaining studies reported a non-significant protective effect of PA on all-cause mortality.
Physical activity outcomes in elderly patients
There were no studies that investigated elderly patients (defined as older than 65 years) separately. However, eight studies stratified their analyses by menopausal status or age,12, 13, 15, 16, 17, 18, 19, and 27 and one study included postmenopausal patients only. 11 ( Table 2 ) Five studies grouped their patients by pre- and postmenopausal status,11, 13, 15, 16, 17, and 19 while the remaining studies stratified their analyses by age.12, 17, 18, and 27 Three studies found a significantly lower risk of either breast cancer-specific mortality 16 (HR 0.80 (0.60–1.06)) or overall mortality 17 (HR 0.29 (0.14–0.60)) in the older or postmenopausal active group (highest versus lowest amount of PA) (data from Keegan et al. 18 not shown). The study in which exclusively postmenopausal patients were included demonstrated a significantly lower risk of all-cause mortality (HR 0.67 (0.46–0.96)) and breast cancer-specific mortality (HR 0.61 (0.35–0.99)) in patients who were moderate- to vigorously active after diagnosis. 11 The remaining five studies all reported a non-significant benefit of PA on outcomes.12, 13, 15, 19, and 27
|Study||No. of patients||Age||Primary outcome||Age-specific results|
|2||Cleveland et al.||1508||Mean age 58.6||5-year all-cause mortality and BC-specific mortality||Postmenopausal active women had a non-significant decreased risk of BC-specific mortality (HR 0.79 (0.45–1.39)), and a borderline significant decreased risk of all-cause mortality (HR 0.65 (0.42–1.00))|
|4||Emaus et. Al||1364||17–79, mean age 57.5||All-cause mortality and BC-specific mortality||Non-significant decreased overall mortality in patients >55 (HR 0.60 (0.36–0.99) and non-significant risk of BC death (HR 0.57 (0.31–1.04))|
|6||Hellman et. Al||528||33–95, median age 66.9||All-cause mortality||In active postmenopausal patients, HR for overall death was 1.71 (1.16–2.54) in highest amount of PA versus no PA|
|7||Holick et. Al||4482||20–79, mean age 58.5||All-cause mortality and BC-specific mortality||The most active patients >59 year had a non-significantly lower risk of BC-mortality (HR 0.85 (0.47–1.54))|
|8||Holmes et al.||2987||30–55 on enrolment in NHS||BC-specific mortality||Active postmenopausal women had a significantly lower risk of BC-death (HR 0.73 (0.54–0.98))|
|9||Irwin et al.||933||Mean age 55.5 (6 months post-diagnosis)||All-cause mortality and BC-specific mortality||Active women >55year had a significantly lower risk of all-cause mortality (HR 0.29 (0.14–0.60)|
|10||Irwin et al.||4643||50–79||All-cause mortality and BC-specific mortality||Moderate- to vigorous-intensity post-diagnosis PA lead to a lower risk for all-cause mortality (HR 0.54 (0.38–0.79)) and a lower risk of BC-specific mortality (HR 0.61 (0.35–0.99)). Increase in PA after diagnosis of 9 MET-h/wk or more lead to 33% decreased risk of all-cause mortality (HR 0.67 (0.46–0.96))|
|12||Rohan et.al||451||20–74||BC-specific mortality||Postmenopausal active women did not have a decreased risk of BC mortality (HR 0.72 (0.29–1.81))|
|14||Keegan et al.||4153||18–69||All-cause mortality||No difference between different ages at diagnosis of menopausal stage (no HRs reported)|
BC: breast cancer; HR: Hazard Ratio; PA: physical activity; MET: Metabolic Equivalent Task.
Differences in outcomes between pre- and post-diagnostic physical activity
Out of all studies, eleven investigated the effect of pre-diagnostic PA on overall mortality, breast cancer-specific mortality and/or disease-free survival.12, 13, 14, 15, 18, 19, 21, 22, 23, 25, and 26 Three studies assessed the effect of post-diagnostic PA,16, 20, and 27 one study investigated the effect of post-diagnostic PA at different time points, 24 and the remaining two studies examined both pre- and post-diagnostic PA.11 and 17 These two studies looked at the effect of pre-diagnostic PA as well as the effect of change in PA after diagnosis on at least one of the outcomes.
Pre-diagnostic PA lead to a significantly better overall survival in five out of 11 studies,13, 14, 18, 21, and 25 of which three studies also ascertained a significantly better breast cancer-specific survival.13, 14, and 21 The remaining studies found beneficial effects, although they did not reach statistical significance. All studies investigating post-diagnostic PA found a beneficial effect on overall survival.11, 16, 17, 20, 24, and 27 Breast cancer-specific survival was significantly lower in three out of five studies that had this endpoint.16, 17, and 27
An increase in PA was beneficial in one out of two studies, with a HR of 0.67 (0.46–0.99) (for the greatest increase in physical activity). 11 Although the second study investigating change in physical activity found a non-significant benefit of increased physical activity (HR 0.55 (0.22–1.38)), a much higher risk of all-cause mortality was established in patients with a reduction in PA (HR 3.95 (1.45–10.5)) 17 Table 3 .
|Study||Patients||Age||Breast cancer specific mortality||All-cause mortality||Breast cancer recurrence|
|1||Bertram et al.||2361 BC survivors stage I-III, 1 year post-treatment||18–70, median age 54.3||–||↓||–|
|2||Cleveland et al.||1508 early-stage BC patients||Mean age 58.6||↓||–||–|
|3||Dal Maso et al.||1453 BC patients||23–74, median age 55||No difference||No difference||–|
|4||Emaus et al.||1364 BC patients||17–79, mean age 57.5||No difference||No difference||–|
|5||Friedenreich et al.||1183 BC patients||Mean age 56||↓||↓||↓|
|6||Hellman et al.||528 primary BC patients||33–95, median age 66.9||No difference||No difference||–|
|7||Holick et al.||4482 BC patients, 91% postmenopausal.||20–79, mean age 58.5||↓||↓||–|
|8||Holmes et al.||2987 nurses with stage I/II/III BC||30–55||↓||–||–|
|9||Irwin et al.||933 BC patients with local or regional breast cancer||Mean age 55.5 (6 months post–diagnosis)||↓||–||–|
|10||Irwin et al.||4643 post-menopausal BC patients||50–79||↓||↓||–|
|11||Pierce et al.||1490 early-stage BC patients||<70, mean age 50||–||↓||–|
|12||Rohan et al.||451 BC patients||20–74||No difference||–||–|
|13||Sternfeld et al.||1970 early-stage BC patients||18–79, mean 60.6||No difference||↓||No difference|
|14||Keegan et al.||4153 BC patients (320 patients excluded from Australian cases)||18–69||–||↓||–|
|15||West-Wright et al.||3,539 BC patients with invasive BC||26–94, mean age 58.9||↓||↓||–|
|16||Enger et al.||717 BC patients with in situ or invasive BC||21–40||No difference||–||–|
|17||Abrahamson et al.||1264 BC patients with invasive BC||20–54, median 42||No difference||–||–|
BC: breast cancer.
Physical activity in different BMI-groups
Nine studies stratified their analyses by different BMI-groups.11, 12, 13, 17, 18, 20, 21, 23, and 27 Two studies reported on breast cancer-specific mortality as well as all-cause mortality for the stratified groups, while one study reported on breast cancer-specific mortality only. The remaining six studies reported all-cause mortality only for the different groups. The study by Cleveland et al. 13 showed a non-significant lower risk of breast cancer-specific mortality in active patients with a BMI under 25 (HR 0.57 (0.30–1.09)) and a significantly lower risk in active patients with a BMI over 25 (HR 0.63 (0.40–0.99)). 13 Similarly, the study by Holick et al. 27 found a better breast cancer-specific survival for active patients with a BMI below 25, although non-significant, and a borderline significant effect in patients with a BMI above 25 (HR 0.63 (0.39–1.02)). 27 West-wright et al. 21 did not find a significant benefit for active patients with a BMI under 25 (HR 1.15 (0.32–0.86)), but a significantly better breast cancer-specific survival in patients with a BMI above 25 (HR 0.41 (0.23–0.74) for high PA versus low PA). In patients with a BMI below 25, all-cause mortality was significantly lower in four out of eight studies.12, 13, 20, and 27 One study reported a benefit that was even greater in patients who were 55 years or older. 13 Four studies found a lower all-cause mortality in active patients with a BMI over 25.13, 17, 23, and 27
We present an overview of the available literature on the effect of PA on breast cancer outcomes. Most studies demonstrate that PA has a positive effect on overall survival and breast cancer-specific survival.Post-diagnostic PA was most beneficial for breast cancer outcomes. Most investigations were observational studies with low median ages, with the exception of one study that investigated only postmenopausal breast cancer patients. Although no study specifically addressed patients who were 65 years or older, we observed that older and/or postmenopausal patients acquired the greatest advantage of PA.
Several explanations have been proposed for the age-related benefit. In addition to concurrent comorbid conditions, aging is associated with declines in physical and cognitive functioning. A large cohort study of breast cancer survivors showed that functional limitations were associated with worse all-cause and competing-cause mortality. 10 Cancer patients are at increased risk of reporting limitations in their activities of daily living compared to non-cancer patients.9 and 29 Several exercise programs demonstrated varying positive effects in older people, including improved muscle strength and gait speed, reduction in falls, improved balance, bone mineral density and increased mental health.30, 31, 32, 33, 34, 35, and 36 The Canadian Study of Health and Aging showed that older persons who participated in high levels of PA had a lower risk of death than those who did little or no exercise; the absolute benefits were greatest for those with the highest number of health deficits. 36
Investigators have speculated on the biological mechanism behind these findings. 37 The effect of PA on breast cancer specific survival was stronger for overweight patients (BMI > 25) than normal-weight patients. Although these findings were not paralleled for all-cause mortality, the effect of PA on the reduction in weight and, subsequently, on insulin levels are a likely explanation. It is well-known that postmenopausal patients with obesity and high insulin levels and/or diabetes mellitus have a greater risk of breast cancer.41, 42, and 43 Insulin resistance is thought to influence the risk of breast cancer recurrence and mortality38, 39, and 40 Pasanisi et al. found that higher insulin levels and insulin resistance syndrome are associated with breast cancer mortality. 38 Importantly, PA can significantly lower insulin levels in women with breast cancer. It has also been suggested that PA reduces breast cancer risk by improving metabolic profiles as well as by decreasing levels of endogenous estrogens and body fat. 44
Aging is a strong predictor of functional decline and comorbidities, and literature suggests that older cancer patients are more likely to be affected by cancer in terms of physical than cognitive functioning.9, 45, and 46 Physical domains are especially important in elderly patients, as they can make the difference between independent and assisted living. As breast cancer in elderly patients is predominantly Estrogen Receptor (ER)-positive, systemic endocrine treatment is frequently warranted.Toxicities of endocrine therapy and the larger number of comorbid conditions with increasing age may result in patients being more vulnerable to the clinical consequences of adverse events. Previous reports have demonstrated that PA can improve comorbid conditions and adverse effects of adjuvant treatment,47 and 48 therefore, PA can provide an additional benefit in these patients.
Because of the heterogeneity of the data collected and different definitions of low, moderate and high intensity PA, we could not perform a pooled meta-analysis of the reported HRs. Furthermore, this study is limited to some extent by the lack of evidence that focuses on the effect of PA on breast cancer outcomes in the elderly breast cancer patients specifically. Consequently, the effect of PA on overall and breast cancer-specific mortality in elderly patients may reveal an even greater benefit, given the prevalence of more ER-positive tumors as well as frailer patients who are likely to gain most from PA. To add, it is important to realize that other factors may also influence the possibility for patients to partake in PA at all. One can presume that, generally speaking, patients without functional limitations are more likely to be healthy and capable of exercising, while patients who are restricted due to health deficits are already less physically active, and thus at an increased risk of adverse outcomes. For this reason, further investigations into the cancer-aging interface and the influence of breast cancer on functional decline in elderly patients are indispensable. We have initiated ‘Climb Every Mountain’, a longitudinal, prospective cohort study that measures physical, cognitive, and social health during and after breast cancer treatment. The purpose of this study is to generate new knowledge on the prevalence and impact of functional, cognitive, and social limitations in elderly breast cancer survivors. A second phase will use those domains most affected by cancer diagnosis and treatment, to develop a patient-tailored physical intervention study aimed specifically at the elderly breast cancer patient.
This review covers a very heterogeneous group of patients, consisting of pre-and postmenopausal breast cancer patients with varying tumor-biological and clinical characteristics. In older breast cancer patients specifically, the majority of studies demonstrated a decrease in both all-cause and breast cancer-specific mortality. To add, the decrease in breast cancer-specific mortality was more prominent in overweight patients than in normal-weight patients. Overall, it can be concluded that some degree of PA leads to improved survival outcomes in breast cancer patients, which provides the prospect of employing a practical intervention that can act as a targeted treatment for breast cancer.
Conflicts of interest
None to declare.
-  M.L. McNeely, K.L. Campbell, B.H. Rowe, T.P. Klassen, J.R. Mackey, K.S. Courneya. Effects of exercise on breast cancer patients and survivors: a systematic review and meta-analysis. CMAJ. 2006;175(1):34-41 Crossref
-  E.M. Monninkhof, S.G. Elias, F.A. Vlems, et al. Physical activity and breast cancer – A systematic review. Epidemiology. 2007;18(1):137-157 Crossref
-  A.R. Carmichael, A.J. Daley, D.W. Rea, S.J. Bowden. Physical activity and breast cancer outcome: a brief review of evidence, current practice and future direction. Eur J Surg Oncol. 2010;36(12):1139-1148 Crossref
-  J.A. Meyerhardt, E.L. Giovannucci, M.D. Holmes, et al. Physical activity and survival after colorectal cancer diagnosis. J Clin Oncol. 2006;24(22):3527-3534 Crossref
-  H.D. Klepin, A.M. Geiger, J.A. Tooze, et al. Physical performance and subsequent disability and survival in older adults with malignancy: results from the health, aging and body composition study. J Am Geriatr Soc. 2010;58(1):76-82 Crossref
-  Cancer Research UK. Breast Cancer Incidence Statistics. 2013. 13.3.2013 [Ref Type: Online Source].
-  W. van de Water, C. Markopoulos, C.J. van de Velde, et al. Association between age at diagnosis and disease-specific mortality among postmenopausal women with hormone receptor-positive breast cancer. JAMA. 2012;307(6):590-597
-  E. Bastiaannet, J.E. Portielje, C.J. van de Velde, et al. Lack of survival gain for elderly women with breast cancer. Oncologist. 2011;16(4):415-423 Crossref
-  N.E. Avis, G.T. Deimling. Cancer survivorship and aging. Cancer. 2008;113(12 Suppl):3519-3529 Crossref
-  D. Braithwaite, W.A. Satariano, B. Sternfeld, et al. Long-term prognostic role of functional limitations among women with breast cancer. J Natl Cancer Inst. 2010;102(19):1468-1477 Crossref
-  M.L. Irwin, A. McTiernan, J.E. Manson, et al. Physical activity and survival in postmenopausal women with breast cancer: results from the women’s health initiative. Cancer Prev Res (Phila). 2011;4(4):522-529 Crossref
-  A. Emaus, M.B. Veierod, S. Tretli, et al. Metabolic profile, physical activity, and mortality in breast cancer patients. Breast Cancer Res Treat. 2010;121(3):651-660 Crossref
-  R.J. Cleveland, S.M. Eng, J. Stevens, et al. Influence of prediagnostic recreational physical activity on survival from breast cancer. Eur J Cancer Prev. 2012;21(1):46-54
-  C.M. Friedenreich, J. Gregory, K.A. Kopciuk, J.R. Mackey, K.S. Courneya. Prospective cohort study of lifetime physical activity and breast cancer survival. Int J Cancer. 2009;124(8):1954-1962 Crossref
-  S.S. Hellmann, L.C. Thygesen, J.S. Tolstrup, M. Gronbaek. Modifiable risk factors and survival in women diagnosed with primary breast cancer: results from a prospective cohort study. Eur J Cancer Prev. 2010;19(5):366-373 Crossref
-  M.D. Holmes, W.Y. Chen, D. Feskanich, C.H. Kroenke, G.A. Colditz. Physical activity and survival after breast cancer diagnosis. JAMA. 2005;293(20):2479-2486 Crossref
-  M.L. Irwin, A.W. Smith, A. McTiernan, et al. Influence of pre- and postdiagnosis physical activity on mortality in breast cancer survivors: the health, eating, activity, and lifestyle study. J Clin Oncol. 2008;26(24):3958-3964 Crossref
-  T.H. Keegan, R.L. Milne, I.L. Andrulis, et al. Past recreational physical activity, body size, and all-cause mortality following breast cancer diagnosis: results from the breast cancer family registry. Breast Cancer Res Treat. 2010;123(2):531-542 Crossref
-  T.E. Rohan, W. Fu, J.E. Hiller. Physical activity and survival from breast cancer. Eur J Cancer Prev. 1995;4(5):419-424 Crossref
-  B. Sternfeld, E. Weltzien, C.P. Quesenberry Jr., et al. Physical activity and risk of recurrence and mortality in breast cancer survivors: findings from the LACE study. Cancer Epidemiol Biomarkers Prev. 2009;18(1):87-95 Crossref
-  C.N. West-Wright, K.D. Henderson, J. Sullivan-Halley, et al. Long-term and recent recreational physical activity and survival after breast cancer: the California teachers study. Cancer Epidemiol Biomarkers Prev. 2009;18(11):2851-2859 Crossref
-  S.M. Enger, L. Bernstein. Exercise activity, body size and premenopausal breast cancer survival. Br J Cancer. 2004;90(11):2138-2141
-  P.E. Abrahamson, M.D. Gammon, M.J. Lund, et al. Recreational physical activity and survival among young women with breast cancer. Cancer. 2006;107(8):1777-1785 Crossref
-  L.A. Bertram, M.L. Stefanick, N. Saquib, et al. Physical activity, additional breast cancer events, and mortality among early-stage breast cancer survivors: findings from the WHEL Study. Cancer Causes Control. 2011;22(3):427-435 Crossref
-  J.P. Pierce, M.L. Stefanick, S.W. Flatt, et al. Greater survival after breast cancer in physically active women with high vegetable-fruit intake regardless of obesity. J Clin Oncol. 2007;25(17):2345-2351 Crossref
-  M.L. Dal, A. Zucchetto, R. Talamini, et al. Effect of obesity and other lifestyle factors on mortality in women with breast cancer. Int J Cancer. 2008;123(9):2188-2194
-  C.N. Holick, P.A. Newcomb, A. Trentham-Dietz, et al. Physical activity and survival after diagnosis of invasive breast cancer. Cancer Epidemiol Biomarkers Prev. 2008;17(2):379-386 Crossref
-  B.E. Ainsworth, W.L. Haskell, A.S. Leon, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc. 1993;25(1):71-80 Crossref
-  C. Sweeney, K.H. Schmitz, D. Lazovich, B.A. Virnig, R.B. Wallace, A.R. Folsom. Functional limitations in elderly female cancer survivors. J Natl Cancer Inst. 2006;98(8):521-529 Crossref
-  T. Etgen, D. Sander, U. Huntgeburth, H. Poppert, H. Forstl, H. Bickel. Physical activity and incident cognitive impairment in elderly persons: the INVADE study. Arch Intern Med. 2010;170(2):186-193 Crossref
-  M.E. Nelson, M.A. Fiatarone, C.M. Morganti, I. Trice, R.A. Greenberg, W.J. Evans. Effects of high-intensity strength training on multiple risk factors for osteoporotic fractures. A randomized controlled trial. JAMA. 1994;272(24):1909-1914 Crossref
-  N.K. Latham, D.A. Bennett, C.M. Stretton, C.S. Anderson. Systematic review of progressive resistance strength training in older adults. J Gerontol A Biol Sci Med Sci. 2004;59(1):48-61
-  V. Weerdesteyn, H. Rijken, A.C. Geurts, B.C. Smits-Engelsman, T. Mulder, J. Duysens. A five-week exercise program can reduce falls and improve obstacle avoidance in the elderly. Gerontology. 2006;52(3):131-141 Crossref
-  T. Liu-Ambrose, L.S. Nagamatsu, P. Graf, B.L. Beattie, M.C. Ashe, T.C. Handy. Resistance training and executive functions: a 12-month randomized controlled trial. Arch Intern Med. 2010;170(2):170-178 Crossref
-  T. Bulat, S. Hart-Hughes, S. Ahmed, et al. Effect of a group-based exercise program on balance in elderly. Clin Interv Aging. 2007;2(4):655-660
-  R.E. Hubbard, N. Fallah, S.D. Searle, A. Mitnitski, K. Rockwood. Impact of exercise in community-dwelling older adults. PLoS One. 2009;4(7):e6174 Crossref
-  A.A. Ogunleye, M.D. Holmes. Physical activity and breast cancer survival. Breast Cancer Res. 2009;11(5):106 Crossref
-  P. Pasanisi, F. Berrino, P.M. De, E. Venturelli, A. Mastroianni, S. Panico. Metabolic syndrome as a prognostic factor for breast cancer recurrences. Int J Cancer. 2006;119(1):236-238 Crossref
-  P.J. Goodwin, M. Ennis, K.I. Pritchard, et al. Insulin-like growth factor binding proteins 1 and 3 and breast cancer outcomes. Breast Cancer Res Treat. 2002;74(1):65-76 Crossref
-  A.M. Mulligan, F.P. O’Malley, M. Ennis, I.G. Fantus, P.J. Goodwin. Insulin receptor is an independent predictor of a favorable outcome in early stage breast cancer. Breast Cancer Res Treat. 2007;106(1):39-47 Crossref
-  P.J. Hardefeldt, S. Edirimanne, G.D. Eslick. Diabetes increases the risk of breast cancer: a meta-analysis. Endocr Relat Cancer. 2012;19(6):793-803 Crossref
-  P. Boyle, M. Boniol, A. Koechlin, et al. Diabetes and breast cancer risk: a meta-analysis. Br J Cancer. 2012;107(9):1608-1617 Crossref
-  A.R. Carmichael, T. Bates. Obesity and breast cancer: a review of the literature. Breast. 2004;13(2):85-92 Crossref
-  A. McTiernan. Mechanisms linking physical activity with cancer. Nat Rev Cancer. 2008;8(3):205-211 Crossref
-  W. Demark-Wahnefried, M.C. Morey, R. Sloane, D.C. Snyder, H.J. Cohen. Promoting healthy lifestyles in older cancer survivors to improve health and preserve function. J Am Geriatr Soc. 2009;57(Suppl 2):S262-S264 Crossref
-  M. Hewitt, J.H. Rowland, R. Yancik. Cancer survivors in the United States: age, health, and disability. J Gerontol A Biol Sci Med Sci. 2003;58(1):82-91
-  K.H. Schmitz, R.M. Speck. Risks and benefits of physical activity among breast cancer survivors who have completed treatment. Womens Health (Lond Engl). 2010;6(2):221-238 Crossref
-  A. McTiernan. Physical activity after cancer: physiologic outcomes. Cancer Invest. 2004;22(1):68-81 Crossref
a Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands
b Department of Medical Oncology, Haga Hospital, The Hague, The Netherlands
1 Both authors contributed equally.
© 2013 Elsevier Ltd, All rights reserved.