Exercise for people with cancer: a systematic review.

BACKGROUND
This systematic review was completed by the Exercise for People with Cancer Guideline Development Group, a group organized by Cancer Care Ontario's Program in Evidence-Based Care (pebc). It provides background and guidance for clinicians with respect to exercise for people living with cancer in active and post treatment. It focuses on the benefits of specific types of exercise, pre-screening requirements for new referrals, safety concerns, and delivery models.


METHODS
Using the pebc's standardized approach, medline and embase were systematically searched for existing guidelines, systematic reviews, and primary literature.


RESULTS
The search identified two guidelines, eighteen systematic reviews, and twenty-nine randomized controlled trials with relevance to the topic. The present review provides conclusions about the duration, frequency, and intensity of exercise appropriate for people living with cancer.


CONCLUSIONS
The evidence shows that exercise is safe and provides benefit in quality of life and in muscular and aerobic fitness for people with cancer both during and after treatment. The evidence is sufficient to support the promotion of exercise for adults with cancer, and some evidence supports the promotion of exercise in group or supervised settings and for a long period of time to improve quality of life and muscular and aerobic fitness. Exercise at moderate intensities could also be sustainable for longer periods and could encourage exercise to be continued over an individual's lifetime. It is important that a pre-screening assessment be conducted to evaluate the effects of disease, treatments, and comorbidities.


INTRODUCTION
The inclusion of exercise into an individual's daily lifestyle is known to promote many health benefits; the same holds true for people with cancer. In addition to improving physical wellbeing, exercise can help in the management of treatment side effects, and its physiologic and psychological changes can drastically affect quality of life (qol).
The present systematic review explores the effects of exercise for people living with cancer with respect to qol, physical fitness, safety, adverse events or injuries, intensity levels, types of exercise, and delivery models. The exercise-specific recommendations are relevant for oncologists, exercise consultants, primary care providers, and other members of health care teams who work with people with cancer. Guidelines, systematic reviews, and primary literature are used as the evidence for the review, which was conducted for the purposes of preparing an evidence-based guideline by Cancer Care Ontario's pebc in 2015.

METHODS
The pebc uses the methods of the practice guidelines development cycle 1 to produce evidence-based and evidence-informed guidance documents. The process consists of conducting a systematic review, conducting a quality appraisal and interpretation of the evidence, drafting recommendations that undergo internal review, and conducting an external review. increased their voluntary exercise, variation in adherence rates or lack of adherence measurements, performance bias, and in some cases, use of questionnaires targeted to patients in active treatment that might not be applicable in a post-treatment population. The next few subsections examine evidence about the safety of exercise and whether, for people with cancer, exercise can be used as an intervention to improve qol as well as physical fitness-and, if so, what types of exercise accomplish that goal the best.

Safety, Adverse Events, or Injuries
The safety of exercise for adults living with cancer is a very important outcome. Safety outcomes include measures such as the frequencies and types of adverse events during exercise sessions or whether treatment delivery or cancerspecific outcomes were negatively affected.
Two guidelines 5,43 concluded that exercise is safe for people with cancer both during active treatment and after treatment. The Belgian Health Care Knowledge Centre 5 developed recommendations concerning the efficacy and safety of exercise treatment during cancer treatment. Based on data about the safety of exercise from a systematic literature review, no harmful effects of exercise during treatment were found. Thus, it was concluded that exercise is safe for patients undergoing treatment for cancer. The American College of Sports Medicine 43 convened an expert panel to create a roundtable consensus statement about exercise for cancer survivors. After reviewing the literature, the panel concluded that exercise training is safe during and after cancer treatments. They recommended that exercises could be specifically adapted based on disease-and treatment-related adverse effects such as lymphedema. They also developed pre-exercise medical assessments to help ensure safety and to guide exercise specialists concerning exercise programs for people living with cancer.
In the systematic reviews and rcts, very few adverse events were attributable to exercise; most studies reported no adverse events at all that were attributable to exercise (Tables v and vi). Of the systematic reviews,
Was an a priori design provided?          Face-to-face group experienced clinically meaningful change over time. Significant differences were found between the face-to-face and telephone groups compared with the usual-care group, p≤0.05 n Strength and endurance Exercise (face-to-face): 7.3; 95% CI: 6.7 to 7.9; 9.2; 95% CI: 8.6 to 9.8 Exercise (telephone): 6.8; 95% CI: 6.1 to 7.5; 8.3; 95% CI: 7.8 to 8.8 Usual care: 6.3; 95% CI: 5.4 to 7.2; 8.0; 95% CI: 7.1 to 9.0 All values statistically significantly different for time and group effect, p<0.05 Adverse Events: No adverse effects, events, or lymphedema were found. Comments: n Scheduled sessions with exercise physiologist were completed by 88% of face-to-face group and 81% of telephone group. n Intervention goal of increasing total physical activity was not met by 25% of face-to-face and telephone groups between measures.     Intervention: Supervised group sessions involving moderate-to high-intensity aerobic (70%-85% maximum heart rate) and resistance exercises involving major muscle groups. Sessions were progressive, and participants were encouraged to supplement with home-based moderate-intensity aerobic exercise for at least 150 minutes. The usual-care group was offered the program after the study was completed.

Frequency and Duration: 1 Hour twice weekly for 3 months plus home-based 150 minutes weekly
Measures: Quality-of-life score (SF-36 mental component summary) and aerobic capacity (VO 2Peak in millilitres per kilogram per minute) and quadriceps strength (leg press, 1 repetition maximum, kilograms) scores at baseline and at 3 months

Comments:
n Retention in the study was 84% (90% in the exercise group and 75% in the control group).
n Median attendance to supervised classes was 84% in the resistance group. n Very active usual-care group; therefore, no difference between groups was observed.
n The exercise group increased physical activity by 3.10 MET-hours weekly. The usual-care group increased physical activity by 3.57 MET-hours weekly (approximately 17%). Increases in both groups were similar (p=0.97). All participants were also very active before the study.  Intervention: Aerobic-based group sessions, plus home exercise program, working up to 75% of heart rate reserve; incremental increases in time for brisk walking at home 3-5 times weekly (usual-care group was offered an exercise program after study completion)

Frequency and Duration: 2 Times weekly plus brisk walking for 8 weeks
Measures: Quality-of-life scores (FACT-G total score) and aerobic capacity (VO 2Peak in millilitres per kilogram per minute) at baseline and at 2 and 3 months

Comments:
n Primary objective was to look at angiogenesis and apoptosis-related molecules. Adverse Events: Six participants in the PACT group developed lymphedema, but continued to follow the progressive resistance training without exacerbation of symptoms.

Comments:
n Adherence to the weekly-supervised exercise training sessions was 66.6%.
n Heart rate during supervised exercise sessions was 77%±7% of the measured heart rate maximum. n Significant improvements in physical activity occurred in the control group.
n High attrition rate: 24% in control group; 32% in exercise group Intervention: Weekly calls, physical acivity counselling, home logs, and a pedometer, then monthly calls for 3 months; start at 10 minutes daily for 2 days weekly, reaching 30 minutes daily for 5 days weekly (brisk walking or use of home exercise equipment at 64%-76% of estimated maximum heart rate) Frequency and Duration: Start: 2 times weekly; end: 5 times weekly for 12 weeks Measures: Quality-of-life scores (FACT-C) and aerobic capacity (VO 2Peak in millilitres per kilogram per minute) at baseline and at 3, 6, and 12 months Comments: n 7-Day physical activity recall showed that, compared with the usual-care group, the exercise group exercised significantly more at 3 months but not at 6 and 12 months.  Intervention: Combined supervised progressive group resistance training of major muscle groups and 20-30 minutes cardiovascular exercises at 70%-85% maximum heart rate, plus two aerobic exercise sessions at home each week; control group received printed materials about physical activity and a pedometer   n Quality-of-life score: No significant differences between groups from baseline to 4 weeks after surgery.
n Quadriceps strength: A significant difference in muscle strength was found between the groups at the 4-week postoperative assessment (p=0.04).
No other significant differences were found.
Adverse Events: There were complications from surgery, but no other adverse events were reported.

Delivery Models and Supervised Settings
Four systematic reviews 8,11,15,49 detected a greater and more consistent benefit of exercise for qol and muscular and aerobic fitness when the intervention was offered in a group or supervised setting compared with a home-based or unsupervised setting ( Table v). Two rcts 32,36 compared various settings for interventions and found that the beneficial effects were greater when exercise was supervised, either in groups or by telephone. One rct 40 found a significant linear trend between an increase in weekly metabolic equivalents of task performed and an improved qol score for all patients in the study.

Intensity Levels and Types of Exercise
Intensity Levels: Three systematic reviews 6,11,18 studied exercise intensity levels and found that studies of longer length (more weeks) and those that included at least moderate-intensity exercise were associated with improved qol and muscular and aerobic fitness ( Table v). Another systematic review 19 that evaluated interventions with positive results for qol found that moderate-intensity aerobic exercise programs used in those interventions resulted in a benefit for qol (Table v). Two rcts 33,39 compared various exercise intensity levels and found improvements in muscular endurance and aerobic capacity for the higher-intensity groups (Table vi). One rct 40 found, for all participants, a significant linear trend between an increase in weekly energy expenditure or metabolic equivalents of task performed and an improved qol score (Table vi).
Resistance Training: Focht et al. 12 analyzed only resistance exercise interventions used in fifteen studies in both active-and post-treatment patients, finding a small and meaningful increase in effect size for qol (Cohen d = 0.25; range: -0.72 to 1.14). In a systematic review, Cramer et al. 10 examined resistance training in three studies with colorectal cancer patients, finding that resistance training improved colorectal cancer-specific qol. Strasser et al. 48 looked at both active-and post-treatment groups in four rcts that compared a resistance training group with a non-exercise group and that measured qol (five other rcts examined muscle strength): two rcts detected a significant effect of resistance training for qol (compared with usual care), and two rcts detected a trend for improved qol in the resistance training group.
Five rcts 22,31,39,42,45 used resistance training alone for their exercise intervention. Winters-Stone et al. 22 and Lonbro et al. 42 both found a significant difference in qol for the exercise group (p < 0.01 and p < 0.05 respectively). Rogers et al. 31 found a positive minimally important difference effect size of d = 0.52 at 6 weeks and d = 0.39 at 12 weeks. Cormie et al. 39,45 did not find a significant difference between groups for qol in both of their rcts (p = 0.195, p = 0.475).
Aerobic Training: Ferrer et al. 11 found that aerobic activity intensity was a significant predictor of qol improvement as a quadratic trend (bivariate moderator analyses: β = 0.25, p = 0.03). Four rcts 26,35,46,47 used only aerobic interventions. Three 26,46,47 showed no significant differences between the intervention and control groups for qol, but in one 35 , paired pre-post t-tests showed a significant difference between groups on the Short Form-36 mental component summary (p ≤ 0.05).
Resistance Versus Aerobic Training: Santa Mina et al. 29 compared aerobic and resistance training programs, both moderate-to-vigorous intensity and home-based. No difference in qol was found between the training groups.

Pre-exercise Assessment for Evaluation of the Effects of Disease Treatments or Comorbidities: The American
College of Sports Medicine's guideline expert panel developed recommendations for pre-exercise medical assessments to help ensure patient safety and to guide exercise specialists with respect to an exercise program for people living with cancer 43 . One systematic review in the panel's literature review 20 found that cardiopulmonary exercise testing is a safe, noninvasive method for measuring the cardiopulmonary fitness of people living with cancer, both during and after treatment. No rct in the literature review reported any adverse events during pre-screening or baseline assessments before initiation of the study intervention [44][45][46][47][50][51][52][53] .

Safety, QOL, and Muscular and Aerobic Capacity
Outcomes of importance to the current guideline included safety, qol, and aerobic and muscular fitness. Numerous studies provide evidence that supports an improvement in qol for patients participating in the interventions. The published guidelines concluded that exercise is safe for people with cancer. Exercise is beneficial for enhancing qol and aerobic and muscular fitness. As with any exercise intervention in an adult population, harm or adverse events can occur, but a cancer diagnosis or its therapy does not exert a negative influence.

Exercising in Group or Supervised Setting
Studies detected a greater and more consistent benefit when the intervention occurred in a group compared with a home (individual) setting. Several systematic reviews assessed the components that were included in successful interventions, concluding that the positive changes in group settings and supervised interventions are substantial.
Almost every intervention started in a supervised setting. A supervised setting can provide not only motivation for an individual to perform exercise, but might also allow for an educational component-especially for individuals receiving resistance-type interventions. Safety and exercise options would then both be optimized. Supervision can also allow individuals who might prefer to exercise outside a group setting to learn about their options and can ensure that exercise professionals have the opportunity to review and instruct people on how to safely perform or use a specific exercise modality.

Exercising at Moderate Intensity and Length of Intervention
No studies directly compared various intensities or lengths of exercise interventions for people with cancer. The systematic reviews detected a benefit for increasing intensities up to a moderate level (3-6 metabolic equivalents of task), but greater amounts of exercise did not necessarily further improve outcomes, including qol.
Longer interventions (18 weeks and ongoing) detected a benefit for qol as well as for aerobic and muscular fitness. Compared with high intensities, moderate intensities of exercise might be sustainable for a longer period and might encourage exercise to be continued throughout a person's life.
The rcts were not conducted for a period adequate to study the long-term effects of exercise. Study lengths were associated with the funding and time available to complete the study rather than with the feasibility or sustainability of an exercise regimen.

Pre-exercise Assessment for Evaluation of Effects of Disease Treatments or Comorbidities
It is a standard recommendation that healthy adults in the general population undergo a fitness assessment before initiating exercise; people living with cancer should therefore also participate in a pre-exercise fitness assessment. The assessment should evaluate comorbidities and any possible latent effects from treatment that might alter a person's ability to safely engage in exercise. Such an assessment also allows the exercise consultant to modify an exercise program, individualizing it for the person and giving consideration to modifications of standard programs based on physical limitations or vulnerabilities.
The time and personnel required to perform a preexercise assessment are acknowledged. However, such assessments can allow clinicians and people living with cancer to feel more safe and secure before an exercise regimen commences. Assessments can also ensure that individuals are aware of potential vulnerabilities connected to their condition.

Implementation Considerations
Some of the issues identified here include concerns about the lack of exercise knowledge, funding, facilities, programs, qualified staff, and exercise specialists in cancer. Clinicians or health care professionals might be inexperienced with exercise regimens, and pre-exercise screening for all cancer survivors could be difficult to achieve, given the additional time and personnel requirements. The type of activity and exercise will depend on the individual patient and his or her preferences, and the choice should be promoted as a part of rehabilitative recovery or the survivorship phase of a treatment program.

Evidence Limitations
The present systematic review was conducted to provide a background and guidance for clinicians with respect to exercise for people living with cancer. It focuses on the benefits of specific types of exercise, pre-screening requirements for new referrals, safety concerns, and delivery models. It covers all cancer types; aerobic and resistance exercise; and qol, muscular, and aerobic outcomes. Other reviews were more specialized in their objectives. Unfortunately, evidence to create specific exercise regimens for specific types of cancer was not available to provide guidance for clinicians. In addition, evidence that met the inclusion criteria was insufficient to produce recommendations based on survival outcomes.
The evidence found in this review showed some weaknesses. A systematic review was not undertaken to inform the guideline from the American College of Sports Medicine; that guideline depended on expert opinion for some topics (such as the pre-screening recommendation). Many of the systematic reviews had issues with heterogeneity in their analysis. Sources of heterogeneity included a population with varying cancer types; varied timing of the exercise intervention (during or after completion of therapy); varying interventions (aerobic compared with resistance training); various lengths of intervention (4-24 weeks); variable exercise intensities; varying frequencies of the intervention (daily to 2, 3, or 5 times per week); multiple measures of qol, aerobic capacity, and strength; varying use of individual or group sessions; and variable timing of the assessments.
The risk of bias in lifestyle trials is an acknowledged issue. Within the rcts reviewed, the following concerns were noted: participants could not be blinded; some assessments (especially qol) were subjective; many trials had performance bias; many trials did not measure exercise activity before entry into the study; adherence during the intervention was variable or not reported; and the exercise level of the control group quite often increased during the intervention, sometimes as much as it did in the exercise group. The lengths of the rcts were not sufficient to fully study a long-term exercise duration. The study lengths were connected to the money and time needed to complete a study rather than to the feasibility or sustainability of the exercise regimen.

CONCLUSIONS
Exercise is safe and can provide qol and fitness benefits for adults living with cancer, whether during active treatment or after treatment. During active treatment, systematic reviews examining patients with all cancers demonstrated a positive influence of exercise on qol. In rcts, benefits within and between groups were found for exercise interventions of moderate intensity. For the post-treatment period, systematic reviews found a positive influence for all exercise interventions. Thus, people living with cancer can be allowed to determine the type of exercise that they would prefer to do for aerobic and resistance training (for example, running, brisk walking, cycling, weight lifting, body weight or elastic band exercises), with similar benefits.
So far, studies have not been designed to determine more exact exercise programs for specific cancer types. Studies with a longer duration are also needed to determine the long-term effects of exercise, as are studies that compare various exercise intensities to determine any difference in benefit. In the present review, evidence that met the inclusion criteria was insufficient to provide recommendations based on survival outcomes. However, sufficient evidence is available to promote exercise to adults with cancer, and some evidence is available to promote exercise in a group or supervised setting and for a long period of time to improve qol and muscular and aerobic fitness. Exercise at a moderate intensity might be sustainable for longer periods and could potentially encourage continuation over a lifetime. It is important that a pre-screening assessment be conducted to evaluate for the effects of disease, treatments, or comorbidities. More research to help create more exact exercise programs for specific cancer types would be beneficial.

REVIEW AND UPDATE
Practice guidelines and literature reviews developed by the pebc are reviewed and updated regularly. For the full guideline and subsequent updates, please visit the Cancer Care Ontario Web site at https://www.cancercare.on.ca/ toolbox/qualityguidelines/clin-program/psychonc/.