Benefits of Aerobic Exercise for Breast Cancer Survivors: A Systematic Review of Randomized Controlled Trials

Background: Physical exercise may be beneficial to breast cancer (BC) survivors. Here, we systematically summarized the effects of aerobic exercise in BC survivors. We conducted a systematic review of randomized controlled trials (RCTs). Methods: We searched PubMed, Web of knowledge, Scopus, Cochrane Central, Virtual Health Library and PEDRO databases for relevant RCTs, comparing aerobic exercise with usual care among BC survivors. Data were extracted and evidence was synthesized narratively. Results: Twelve studies were included in this systematic review. Studies reported that aerobic exercise can significantly improve the quality of life in BC survivors. Moreover, aerobic exercise alleviated the symptoms of depression and anxiety. However, current evidence from the included studies showed that there was no significant benefit for aerobic exercise in terms of weight loss. Conclusion: Our study suggests that aerobic exercise is beneficial to BC survivors. Clinical Relevance: Aerobic exercise should be recommended in the therapeutic and rehabilitative regimens of BC survivors.


Introduction
Breast cancer (BC) is the second most common cancer worldwide and the most frequent cancer among women (25% of the overall cancer among women) with an increasing incidence in the last decade. The World Health Organization (WHO) ranked BC as the fifth most common cause of cancer mortality worldwide (522,000 deaths in 2012) (Torre et al., 2015). The significant improvements in BC screening, diagnosis, and treatment over the past few decades led to a decline in BC-related death rate, which in turn caused a substantial increase in the number of BC survivors (Siegel et al., 2017).
Physical activity plays a significant role in improving quality of life (QoL), physical, psychological functions, and other health indicators in cancer patients (Fong et al., 2012). The literature showed several benefits for exercise therapy as a supportive care for BC patients and survivors (Brockow et al., 2016;Kim et al., 2009;McNeely et al., 2006a) These beneficial effects vary according to disease stage, type of primary treatment and the lifestyle of patients (Knols et al., 2005) The survival outcome in BC survivors is likely to have some correlation with exercise; long-term physical activity levels are important for BC prognosis and are associated with improved survival (Bertram et al., 2011). Moreover, previous research showed that exercise improves the QoL and lessens the symptoms of depression and anxiety in BC survivors (Segar et al., 1998). Additionally, it was associated with benefits on muscle strength and body composition (Zhu et al., 2016). Therefore, the American Cancer Society guidelines recommend at least 150 minutes on a weekly basis for exercise practicing. These guidelines also encouraged long-term adherence to daily exercise beyond the initial post-surgical rehabilitation (Rock et al., 2012).
Aerobic exercise is a form of physical activity that depends primarily on the aerobic energy-generating process i.e. it requires free oxygen to meet the demands of aerobic metabolism. It involves running/jogging, swimming, cycling, and walking. Several randomized controlled trials (RCTs) suggested the efficacy of aerobic exercise in reducing cancer-related fatigue among BC survivors (Cantarero-Villanueva et al., 2013;Carter et al., 2016;Vardar Yağlı et al., 2015). Moreover, aerobic physical activity was found effective in improving anthropometric measures as body weight, body composition, and VO2 peak, as well as reducing inflammatory markers Matthews et al., 2007;Tizdast et al.,2016).. have been assessed in several systematic reviews and meta-analyses (Floyd and Moyer, 2010;Meneses-Echávez et al., 2015;Zhu et al., 2016). However, there is a gap in addressing the effects of specific exercise types in BC rehabilitation program as published trials showed conflicting results, especially in terms of physical activity and QoL outcomes. According to our knowledge, no former systematic review or meta-analysis has assessed the benefits of aerobic exercise specifically in BC survivors. The National Comprehensive Cancer Network (NCCN) reported the need for future research work to identify the effects of specific modes of exercise on specific outcomes as fatigue and QoL in cancer survivors (Schmitz et al., 2010). We performed this systematic review of the literature to investigate the efficacy of aerobic exercise intervention on physical activity, QoL (primary outcomes), weight, inflammatory markers and sleep among BC survivors by scrutinizing the published trials.

Materials and Methods
We followed the PRISMA statement guidelines (Table  S1 of Supplementary material) during the preparation of this review (Moher et al., 2009). This systematic review is registered on the PROSPERO international prospective register of systematic reviews (CRD42017060106).

Literature search strategy
We searched PubMed https://www.ncbi.nlm.nih.gov/ pubmed/, Web of knowledge http.apps.webofknowledge. com, Scopus https-www-scopus.com, Cochrane Central https://www.cochranelibrary.com/central, Virtual Health Library (VHL) https://bvsalud.org/en/ and PEDRO https:// www.pedro.org.au/ for RCTs that compared Aerobic Exercise with usual care in BC survivors. We used the following search strategy "(breast cancer survivors* OR breast tumor survivors*) AND (Aerobic exercise OR physical fitness OR physical therapy OR rehabilitation)". No restrictions by publication period were employed.. In addition, we manually scanned the references list of selected articles for relevant studies.

Eligibility criteria and study selection
We included RCTs that met our inclusion criteria, a) Population: BC survivor women, treated with chemotherapy or radiotherapy (stopped at least six months earlier) or patients who have started adjuvant endocrine therapy (antiestrogens, aromatase inhibitors, LHRH agonists, or combinations) not less than 4 months earlier and not scheduled for and not currently undergoing chemo-/radiation therapy, b) Intervention: Studies in which the intervention was aerobic exercise (walking, cycle ergometers, swimming, and stair climbing), not mixed with any other type of exercise, c) Comparator: Studies where the control group received usual care, d) Outcomes: Physical Activity level, QoL, Sleep parameters, weight, as well as inflammatory markers.
We excluded studies with the following criteria: a) patients on active therapy, b) patients with uncontrolled cardiac or vascular disease, c) patients exercising at a regular basis at baseline, and d) non-English articles. The retrieved citations were added to a Microsoft Excel sheet that was distributed to reviewers. Eligibility screening was done through two separate steps: a) titles and abstracts screening, and b) full texts screening. Each study title and/ or full-text was screened by two independent reviewers (therefore, each reviewer screened half the retrieved citations) and disagreements were resolved by discussion between all reviewers.

Data Extraction
Three independent reviewers extracted the relevant data from the included studies, using a preformatted data extraction sheet. Disagreements were resolved by the opinion of a fourth reviewer. The extracted data included: a) baseline characteristics of participants as age and cancer stage, b) general characteristics of included studies as design, sample size, and assessed outcomes, c) characteristics of the intervention as type, intensity, and duration and d) the results of each included study. The summary effect size was difficult to calculate due to the lack of sufficient data for subsequent pooling, as well as lack of the homogeneity needed to make a calculation.

Risk of bias assessment
The risk of bias in included RCTs was assessed according to the Cochrane Handbook for Systematic Reviews of Interventions 5.1.0 (updated March 2011) using the risk of bias assessment table, provided in the same book (part 2, Chapter 8.5). The Cochrane risk of bias assessment tool includes the following domains: Random sequence generation (selection bias), allocation sequence concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias), and other potential sources of bias. The reviewer judgment is categorized as low risk, high risk, or unclear risk of bias. For risk of bias exclusion across the included studies, we compared the reported outcomes between the studies to exclude reporting of selective outcome.

Literature search results
We retrieved 1,800 unique citations. After the initial title and abstract screening, 86 full-text articles were retrieved and screened for eligibility. After full text screening, 12 RCTs (n= 1,120 patients) were included in our systematic review (see PRISMA flow diagram; Figure 1).

Characteristic of included studies
The sample size of included studied ranged from 19 to 357 patients, with an average age of 44 to 61 years. Aerobic exercise was the only intervention of interest in this meta-analysis either supervised or home-based training programs. Most of the included studies reported walking as the preferred type of activity among BC survivors. Exercise sessions were of moderate intensity. Irwin   *PA had borderline significant effect on CRP *PA also had a clinically and statistically significant effect on HRR and clinically but not statistically significant effects on RHR, SBP, DBP, HDL-C, and TG Effect of aerobic exercise on metabolic syndrome prevalence -Waist circumference (cm) -HDL-C (mg/dL) -Triglycerides (mg/dL) -Glucose (mg/dL) -Metabolic syndrome z-score *In overweight or obese, physically inactive BCS the adherence to a moderate intensity aerobic exercise was associated with improvements in metabolic syndrome criteria. Results revealed that the moderate-intensity aerobic exercise intervention did not significantly alter concentrations of IL-6, CRP, or TNF-a. *The highest reduction in weight, body fat percentage and adipose tissue and the highest increase in VO2 peak occurred in the interval exercise group. *The waist-hip ratio reduced equally in both exercise groups *The anthropometric variables, body fat percentage and adipose tissue decreased significantly in both exercise groups but the differencebetween the two exercise groups was not significant Women in the exercise-intervention group increased their average pedometer steps by 1621 steps per day compared with a decrease of 60 steps per day among women in the usual-care group (P <.01).   Roveda et al., (2016) and Tizdast et al., (2016) investigated the role of aerobic exercise on weight reduction among BC survivors. Two studies assessed the effect of aerobic exercise on inflammatory markers (Fairey et al., 2005;. Only one study reported the effect of aerobic exercise on sleep parameters (Roveda et al., 2016). A summary of the design and main findings of included studies is shown in Table 1, and the baseline characteristics of their participants are shown in Table 2.

Risk of Bias assessment results
We used the Cochrane risk of bias assessment tool to assess the bias of included studies. All included studies were of low risk in terms of selective reporting and attrition bias. Only five studies were of low risk of bias in terms of allocation concealment (Carter et al., 2016;Irwin et al., 2008;Murtezani et al., 2014;Roveda et al., 2016;Saarto et al., 2012), while three studies were of low risk of bias in terms of blinding the participants/personnel (Campbell et al., 2017;Fairey et al., 2005;. The summary of risk of bias assessment is shown in Figure 2, and the reasons for authors' judgments are shown in Table S2 of Supplementary material.

Physical Activity
Mathews et al., (2007) reported higher activity levels in the intervention group compared to usual care group (p ≤0.04). Moreover, Irwin et al., (2008) reported that in the 7-day physical activity log, the exercise group increased from moderate to vigorous intensity reactional activity (129 min per week, p <0.001), compared to a smaller increase in the usual care group (45 min per week). However, Saarto et al., (2012) reported an insignificant difference between the exercise and control groups (P=0.97). They detected a significant linear trend between higher physical activity, improved QoL, and recovery       Campbell et al., (2017) used the FACT-COG tool to measure the cognitive symptoms impacting the QoL, but they found no statistically significant improvement. Murtezani et al., (2014) used the FACT B score to measure the overall QoL after 10 weeks of intervention. They reported that the FACT B score was increased in the intervention group and was decreased in controls (P = 0.003).
Weight Tizdast et al., (2016) reported that higher weight reduction was achieved in the interval exercise group (P=0.012). Murtezani et al., (2014) andFairy et al., (2005) did not observe any significant difference between the exercise and control groups in terms of weight loss (95% CI,-4.8 to 2.25 Kg; P=0.47 and 95% CI,-1.6 to 0.6 Kg; P=0.39, respectively). On the contrary, Roveda et al., (2016) found that the control group had a significant weight reduction after 3 months of physical activity (P= 0.01).

Inflammatory markers
Two studies examined the effects of aerobic exercise on inflammatory makers in BC survivors. Fairy et al., (2005) reported that C-reactive protein (CRP) decreased by 139 mg/l in the exercise group, while it increased by 0.1 mg/l in the control group. However, this change was not significant (95% CI,-3.09 to 0.1 mg/l; P=0.66) Moreover,  reported an insignificant difference between randomization groups regarding serum concentrations of interleukin (IL)-6, CRP, and tumor necrosis factor (TNF-α).

Sleep parameters
Only one study reported the effects of exercise on sleep parameters for BC survivors. Roveda et al., (2016) found sleep efficiency (SE) to be significantly decreased in the control group, it but remained unchanged in the exercise group (P=0.001). The mean activity score (MAS) was significantly increased in the control group, but it remained unchanged in the exercise group (p <0.001). The authors reported a significant decrease (P=0.03) in the actual sleep time (AST) and a significant increase (P=0.02) in the movement fragmentation index (MFI) after aerobic exercise intervention, compared to baseline scores.

Discussion
This systematic review summarized the evidence on the effects of aerobic exercise interventions (especially walking as the most frequently investigated approach) in BC survivor women. The results showed that aerobic exercise was superior to usual care in enhancing the physical activity, QoL, and sleep parameters. However, weight and inflammatory markers did not show any remarkable difference between the two groups.
Moderate intensity aerobic exercise is associated with increased short-term physical activity levels among BC survivors, especially in patients who did not follow regular exercise before (Irwin et al., 2008;Matthews et al., 2007).