Exercise for fatigue in people with cancer during and after active treatment

Objectives

This is a protocol for a Cochrane Review (intervention). The objectives are as follows:

Based on this single protocol, we will conduct five separate reviews, each on one of the following types of exercise.

• Aerobic exercise

• Anaerobic exercise

• Combined (aerobic and anaerobic) exercise

• Aerobic versus anaerobic exercise

• Yoga

The primary aims of these reviews are to:

• evaluate the effects of aerobic exercise on CRF in people with cancer;

• evaluate the effects of anaerobic exercise on CRF in people with cancer;

• evaluate the effects of aerobic and anaerobic (combined) exercise on CRF in people with cancer;

• compare the effects of aerobic exercise versus anaerobic exercise on CRF in people with cancer;

• evaluate the effects of yoga on CRF in people with cancer.

Furthermore, our aims for each of these reviews are to evaluate the effects at:

• different periods of treatment in relation to cancer treatment (before, during, or after treatment);

• different periods of assessment (up to 12 weeks follow‐up, more than 12 weeks to less than 6 months follow‐up, or 6 months or longer follow‐up).

Moreover, we will analyse the effects of exercise on HRQoL, adverse events, depression, and anxiety.

Background

Description of the condition

Cancer‐related fatigue (CRF) is defined as a "distressing, persistent, subjective sense of physical, emotional, and/or cognitive tiredness or exhaustion related to cancer and/or cancer treatment that is not proportional to recent activity and interferes with usual functioning" (Bower 2014). In contrast to other types of fatigue, CRF is more severe, persistent, and cannot be reduced by sleep or rest (Fabi 2020).

The proposed diagnosis of CRF is the presence of persisting or recurring diminished energy, increased need to rest that is disproportionate to changes in activity level, and related physical, emotional, and cognitive symptoms, which result in significant distress or impaired social, occupational, or other important areas of functioning, and cannot be explained by the presence of a psychiatric comorbidity or diseases requiring prophylactic medication (Cella 2001).

With an overall prevalence between 50% and 90%, CRF is the most common symptom of cancer and its treatment (Campos 2011). It is observed in people with a wide range of cancer diagnoses (Schmidt 2020). However, CRF remains undiagnosed and untreated in many people with cancer (Passik 2002; Ripamonti 2018).

CRF may be related to cancer, its treatment, or other factors. As summarised in Bower 2014, CRF may occur before active cancer treatment, and usually increases during treatment including radiotherapy (Hickok 2005), chemotherapy (Jacobsen 1999), and hormonal or biological therapies, or both (Phillips 2013). In a substantial proportion of people with cancer, CRF persists even years after treatment (Bower 2006; Cella 2001; Servaes 2007; Wang 2014).

CRF is associated with decreased health‐related quality of life (HRQoL (Curt 2000; Gupta 2007)). Across cancer diagnoses, CRF is reported as the most prevalent and severe core symptom amongst people with cancer (Cleeland 2013), and it is perceived as most distressing, affecting peoples' daily lives even more than other cancer‐related symptoms, such as pain (Vogelzang 1997). CRF interferes with an individual's ability to perform activities of daily living (Mustian 2008). It is correlated with depression and anxiety (Brown 2009), as well as with distress, sleep disturbances, lower physical activity levels, pain, difficulties with coping with cancer, and catastrophising about symptoms (Abrahams 2018). CRF may even be a risk factor contributing to shorter survival (Groenvold 2007; Quinten 2011).

The exact aetiology of CRF is not well understood. However, it is likely to be multifactorial and to involve inter‐related cytokine, muscular, neurotransmitter, and neuroendocrine changes (O’Higgins 2018). According to Bower 2014, the most prominent mechanism linked with CRF is cytokine dysregulation with a focus on pro‐inflammatory cytokines (e.g. Inagaki 2008; Jager 2008; Schubert 2007), which may be produced in response to the cancer itself or as a consequence of common cancer treatments such as radiation or chemotherapy. Links between markers of inflammation and fatigue have been reported before, during, and after treatment. Several factors such as pain, emotional distress (e.g. depression, anxiety), anaemia, sleep disturbance, nutritional deficits, decreased functional status, and comorbidities may contribute to CRF (NCCN 2022). The factors causing CRF may differ amongst individuals, phases of disease, and types of treatment (Ryan 2007).

In summary, CRF is a multifactorial symptom, which can be present at any stage of the cancer experience and results in a significant loss of HRQoL.

Description of the intervention

Besides routine screening for and provision of information on CRF, the European Society for Medical Oncology (ESMO) guideline for the management of CRF recommends physical exercise to improve CRF (Fabi 2020). Similarly, the exercise guideline for cancer survivors considers physical exercise as an effective treatment to improve fatigue (Campbell 2019).

The types of exercise recommended in these guidelines comprise aerobic, anaerobic, combined (i.e. aerobic and anaerobic), or mind‐body exercises (Campbell 2019; Fabi 2020). Aerobic exercises implement the continuous rhythmic use of large muscle groups (Patel 2017). Examples of aerobic exercises include walking, running, cycling, and swimming. Anaerobic exercises focus on increasing muscle strength and functionality (e.g. lifting weights). Yoga is a prominent example of exercises addressing the mind and body (i.e. mind‐body exercises), which may also comprise relaxation techniques or meditation.

Beneficial effects of aerobic, anaerobic, combined exercises (i.e. aerobic and anaerobic), and yoga on CRF as well as on HRQoL in people with cancer have been demonstrated by several Cochrane Reviews (Cramp 2012; Knips 2019; Mishra 2012a; Mishra 2012b).

How the intervention might work

Although the exact mechanism of action of the beneficial effects of physical exercise on CRF, as well as on HRQoL, may not be fully understood, there are some explanatory approaches, as follows.

Exercise has an impact on cancer cells and tumour growth rate (Hojman 2018).

Stimulated by inflammatory cytokines, tryptophan is catalysed to kynurenine (Kim 2015). Kynurenine can cross the blood‐brain barrier and is then degraded in the brain, which may contribute to the pathogenesis of neuropsychiatric symptoms of CRF and depression (Kim 2015; Schlittler 2016). During aerobic endurance training, this neurotransmitter is already metabolised in skeletal muscle and is converted from kynurenine to kynurenic acid (Schlittler 2016). Kynurenic acid cannot pass the blood‐brain barrier, so it has a positive impact on depression, for example (Agudelo 2014).

Another explanation is offered by the bio‐behavioural model of Al‐Majid and Gray, which summarises both biological and psychological effects (Al‐Majid 2009). According to this model, anaemia is associated with CRF, and an increase in haemoglobin levels is associated with a significant improvement in fatigue (Boccia 2006). A study by Drouin demonstrated that aerobic exercise training significantly maintained erythrocyte levels during radiation therapy in women with breast cancer (Drouin 2006). Pro‐inflammatory cytokines are another biological factor that seem to be related to CRF (Inagaki 2008; Jager 2008; Schubert 2007). The concentration of these inflammatory markers can be downregulated by physical exercise (Khosravi 2019). A randomised controlled trial (RCT) examined the effect of physical exercise during radiotherapy in men with prostate cancer on inflammatory markers in the blood, as well as the relationship of these parameters to CRF (Hojan 2016). A significant decrease in pro‐inflammatory cytokine levels, as well as fatigue, was observed. Psycho‐behavioural factors in the model that promote CRF and can be influenced by exercise include psychological distress and sleep disturbances (Al‐Majid 2009).

Why it is important to do this review

The research question of how to manage CRF has been prioritised by people with cancer, clinicians, and caregivers in a James Lind Priority Setting Partnership (Aldiss 2019). Although CRF remains one of the most distressing symptoms associated with cancer, there is no gold standard for treatment, and it is unclear what type of exercise may be most beneficial.

Two Cochrane Reviews have previously been conducted on the effects of different exercise interventions on HRQoL and HRQoL domains including fatigue (Mishra 2012a; Mishra 2012b). These reviews indicated that exercise may have beneficial effects at varying follow‐up periods on fatigue and HRQoL, for both people undergoing active cancer treatment and cancer survivors. The authors also concluded that more research is needed to investigate how to sustain the benefits of exercise over time and to further explore the beneficial attributes of exercise. Beneficial effects of exercise were also observed in another Cochrane Review on the effects of exercise for the management of CRF (Cramp 2012).

Recommendations to engage in physical exercise have recently been implemented in guidelines (Campbell 2019; Fabi 2020), which were partially based on data synthesised in these reviews (Cramp 2012; Mishra 2012a; Mishra 2012b).

Since the publication of the reviews in 2012, numerous relevant RCTs have been identified (approximately over 150 positive results).

Furthermore, the ESMO guideline highlights that the evidence for the efficacy of yoga is limited for people with cancers other than breast cancer (Fabi 2020).

An up‐to‐date and comprehensive evidence synthesis on the trials on exercise for people with cancer is therefore highly relevant to consumers and clinicians.

Based on this single protocol, we will conduct five reviews to synthesise the current evidence on the effects of exercise on CRF, separated by exercise type. In each one of these reviews, we will evaluate the effects on CRF at different periods of treatment in relation to cancer treatment (before, during, or after treatment) and at different periods of assessment (up to 12 weeks follow‐up, more than 12 weeks to less than 6 months follow‐up, or 6 months or longer follow‐up). Moreover, we will explore the beneficial attributes of exercises using subgroup analyses, and we will analyse the effects of exercise on HRQoL, adverse events, depression, and anxiety.

Objectives

Based on this single protocol, we will conduct five separate reviews, each on one of the following types of exercise.

• Aerobic exercise

• Anaerobic exercise

• Combined (aerobic and anaerobic) exercise

• Aerobic versus anaerobic exercise

• Yoga

The primary aims of these reviews are to:

• evaluate the effects of aerobic exercise on CRF in people with cancer;

• evaluate the effects of anaerobic exercise on CRF in people with cancer;

• evaluate the effects of aerobic and anaerobic (combined) exercise on CRF in people with cancer;

• compare the effects of aerobic exercise versus anaerobic exercise on CRF in people with cancer;

• evaluate the effects of yoga on CRF in people with cancer.

Furthermore, our aims for each of these reviews are to evaluate the effects at:

• different periods of treatment in relation to cancer treatment (before, during, or after treatment);

• different periods of assessment (up to 12 weeks follow‐up, more than 12 weeks to less than 6 months follow‐up, or 6 months or longer follow‐up).

Moreover, we will analyse the effects of exercise on HRQoL, adverse events, depression, and anxiety.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised controlled trials (RCTs) evaluating the effects of exercise on cancer‐related fatigue (CRF) in people aged 18 and older with cancer before, during, and after active treatment. We will include online clinical trial results, summaries of otherwise unpublished clinical trials, and abstracts. If there are insufficient data in the abstracts for analysis, we will attempt to locate the full study (e.g. by contacting the study authors). We will exclude non‐randomised studies (including controlled clinical trials), cluster‐RCTs, cross‐over RCTs, case reports, and clinical observations.

Types of participants

We will include studies with a minimum proportion of 80% of people aged 18 years and older with a confirmed diagnosis of cancer. We will include people irrespective of sex, ethnicity, tumour site, tumour type, tumour stage, and type of cancer treatment received. We will include people who receive an intervention beginning either before, during, or after cancer treatment. We will exclude people receiving hospice care and people described as 'terminally ill' by the investigators.

Types of interventions

We will include trials evaluating the effects of exercise on fatigue in people aged 18 and older with cancer before, during, and after active treatment.

We will include interventions that comprise structured exercise and last for at least five sessions. We will include combined interventions only when exercise is the main component of the intervention. Participants need to be instructed in a face‐to‐face fashion. We will exclude interventions that are limited to exercise prescriptions.

We will conduct five separate reviews based on this single protocol, each on one of the following types of exercise.

• Aerobic exercise

• Anaerobic exercise

• Combined (aerobic and anaerobic) exercise

• Aerobic exercise versus anaerobic exercise

• Yoga

We will include trials comparing aerobic exercise, anaerobic exercise, combined (aerobic and anaerobic) exercise, or yoga with a control group (e.g. no intervention, minimally active comparator, or usual care), and trials comparing aerobic exercise with anaerobic exercise.

Timing of exercise intervention in relation to cancer treatment

We will evaluate the effects at different periods of treatment in relation to cancer treatment:

• before treatment;

• during treatment;

• after treatment.

Types of outcome measures

We will only consider studies evaluating the effects of exercise on CRF (i.e. studies that assess CRF as an outcome).

As many different outcomes besides CRF have been assessed in trials on exercise for people with cancer, consumers and consumer representatives have been involved in this project by participating in a discussion on the selection and prioritisation of outcomes in order to ensure that this systematic review produces results of highest consumer relevance. Based on this discussion, we will focus on the outcomes presented below.

Primary outcomes

• Cancer‐related fatigue, measured using validated instruments (e.g. Multidimensional Fatigue Inventory (MFI; Smets 1995), Functional Assessment of Cancer Therapy ‐ Fatigue (FACT‐F; Cella 1997), Brief Fatigue Inventory (BFI; Mendoza 1999)).

• Health‐related quality of life, measured using validated tools (e.g. European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire‐C30 (EORTC QLQ‐C30; Fayers 2002), Functional Assessment of Cancer Therapy – General (FACT‐G; Cella 1993)).

Secondary outcomes

• Adverse events: any harm associated with the intervention (e.g. dizziness, injuries or pain).

• Anxiety, measured using validated tools (e.g. Hospital Anxiety and Depression Scale (HADS; Zigmond 1983)).

• Depression, measured using validated tools (e.g. Beck Depression Inventory II (BDI‐II; Beck 1996 )).

Timing of outcome assessment

We will evaluate the effects at different periods of outcome assessment:

• up to 12 weeks follow‐up;

• more than 12 weeks to less than 6 months follow‐up;

• 6 months or longer follow‐up.

Search methods for identification of studies

Electronic searches

Our search will be based on a search strategy developed by Jo Platt for two previously published Cochrane Reviews (Mishra 2012a; Mishra 2012b; for the search strategy in MEDLINE, please see Appendix 1).

For this review, we will be building on searches run for the previously published Cochrane Reviews in 2012, 2016, and 2019. The updated strategy will search the following databases and trial registries without language restrictions.

• The Cochrane Central Register of Controlled Trials (CENTRAL) via Cochrane Library, latest issue.

• MEDLINE via Ovid (2019 to present)

• Embase via Ovid (2019 to present)

• CINAHL via EBSCO (2019 to present)

• PsycINFO (2019 to present)

• PEDro (2019 to present)

• LILACS (Latin American and Caribbean Health Science Information database; 2019 to present)

• SportDiscus (2019 to present)

The search will be performed by one of the review authors, an experienced Information Specialist (IM).

The results of the updated search will be integrated with the results of the searches run for the previously published reviews (Mishra 2012a; Mishra 2012b).

Searching other resources

We will search the US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov) and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP; apps.who.int/trialsearch/) for ongoing trials.

Moreover, we will check reference lists of the included studies and relevant systematic reviews identified for further literature. We will contact study authors for additional information where necessary.

Data collection and analysis

Selection of studies

At least two review authors from the following list (ME, MA, CW, AO, NC, SM, RS, SIM) will independently determine the eligibility of each study identified by the search. We will eliminate studies that clearly do not satisfy our inclusion criteria and obtain full copies of the remaining studies. At least two review authors from the following list (ME, MA, CW, AO, NC, SM, RS, SIM) will independently evaluate these studies to determine their eligibility for inclusion in the review; in the event of disagreement, a third review author (NS) will adjudicate. We will not anonymise the studies in any way before assessment. We will include a PRISMA flowchart in the full review (Moher 2009). We will include studies in the review irrespective of whether measured outcome data are reported in a 'useable' way. If the data from the full study are unavailable, we will assess the abstract as a study awaiting classification.

Data extraction and management

At least two review authors from the following list (ME, MA, CW, AO, NC, SM) will independently extract data using a standard, piloted form and check for agreement before entry into Review Manager Web (RevMan Web 2022). In the event of disagreement, a third review author will adjudicate (NS). We will collate multiple reports of the same study so that each study, rather than each report, is the unit of interest in the review. We will collect characteristics of the included studies in sufficient detail to populate a 'Characteristics of included studies' table in the full review.

We will extract the following information.

Characteristics of the study

• Funding sources and study authors' declarations of interest

• Trial methods: study design, method of sequence generation, method of allocation concealment, masking (participant, researcher, outcome), exclusions after randomisation, selective outcome reporting, loss to follow‐up and compliance

Characteristics of the study population

• Country

• Trial inclusion and exclusion criteria

• Number randomised in each arm

• Demographic characteristics (e.g. age, sex)

• Type of cancer, including primary site, stage at diagnosis

• Type of treatment regimen (i.e. radiation, surgery, chemotherapy, or combination)

Characteristics of the intervention

• Type of exercise in each arm (aerobic exercise; anaerobic exercise; combined (aerobic and anaerobic) exercise; yoga)

• Details of the intervention(s): frequency, duration, intensity, total number of sessions, duration of follow‐up, format (i.e. individual or group, professionally led or not, home‐ or facility‐based)

• Type of control group (e.g. no intervention, minimally active comparator, or usual care)

• Co‐intervention (e.g. medication use)

Characteristics of the outcome

• Tool for outcome measurement

• Type of analysis (e.g. intention‐to‐treat)

• Length of time between end of intervention and outcome measurement

• Numerical data for outcomes of interest (e.g. means, standard deviations (SDs), standard errors, confidence intervals)

Assessment of risk of bias in included studies

At least two review authors from the following list (ME, MA, CW, AO, NC, SM, RS, SIM) will independently assess risk of bias for each study, using the risk of bias tool outlined in Version 5.1.0 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), with any disagreements resolved by discussion. We will complete a risk of bias table for each included study in Review Manager Web (RevMan Web 2022).

For details on the assessment of risk of bias for each included study, see Appendix 2.

We will judge the overall risk of bias of a study considering the worst judgement in any of the risk of bias domains, excluding the domains for blinding (i.e. blinding of participants and personnel and blinding of outcome assessment) due to the nature of the interventions. We will thus judge studies (overall) as:

• low risk of bias (low risk of bias in all risk of bias domains excluding the domains for blinding);

• unclear risk of bias (unclear risk of bias in at least one risk of bias domain excluding the domains for blinding);

• high risk of bias (high risk of bias in at least one risk of bias domain excluding the domains for blinding).

Measures of treatment effect

We will use intention‐to‐treat data, if available, to calculate treatment effects. We will calculate mean differences (MDs) including 95% confidence intervals (CIs) for continuous outcomes, when assessed with the same scale. Otherwise, we will calculate standardised mean differences (SMDs) including 95% CIs. For dichotomous outcomes, we will extract number of participants and number of events per arm and calculate risk ratios (RRs) with 95% CIs for each trial.

Unit of analysis issues

The unit of randomisation will be the individual.

As recommended in Chapter 23 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021a), for studies with multiple treatment groups, we will combine arms as long as they can be regarded as subtypes of the same intervention.

When arms cannot be pooled this way, we will compare each arm with the common comparator separately. For pair‐wise meta‐analysis, we will split the ‘shared’ group into two or more groups with smaller sample size, and include two or more (reasonably independent) comparisons. For this purpose, for dichotomous outcomes, both the number of events and the total number of participants would be divided up, and for continuous outcomes, the total number of participants would be divided up with unchanged means and SDs.

Dealing with missing data

We will take the following steps to deal with missing data, as suggested in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2021). Whenever possible, we will contact the original investigators to request relevant missing data. If the number of participants evaluated for a given outcome is not reported, we will use the number of participants randomised per treatment arm as the denominator. If only percentages but no absolute number of events are reported for binary outcomes, we will calculate numerators using percentages. If estimates for mean and SDs are missing, we will calculate these statistics from reported data whenever possible, using the approaches described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021b). If SDs are missing and we are not able to calculate them from reported data, we will calculate values according to a validated imputation method (Furukawa 2006). If data are not reported numerically but are reported graphically, we will estimate missing data from figures. We will address the potential impact of missing data on the findings of the review in the Discussion section.

Assessment of heterogeneity

In order to evaluate the presence of clinical heterogeneity, we will generate summary statistics for the important clinical and methodological characteristics across all included studies. Within each pair‐wise comparison, we will assess the presence of clinical heterogeneity by visually inspecting the confidence intervals for the results. Furthermore, we will consider Chi² statistics to identify heterogeneity. We will also use the I² statistic to quantify possible heterogeneity.

Assessment of reporting biases

We will search trial registries for additional studies that were not published, according to Chapter 13 of the Cochrane Handbook for Systematic Reviews of Interventions, which will help to reduce bias in the review process (Page 2021).

In pair‐wise comparisons with at least 10 trials, we will examine the presence of small‐study effects graphically by generating funnel plots. We will use linear regression tests to test for funnel plot asymmetry (Egger 1997). We will consider a P < 0.1 significant for this test (Page 2021). We will additionally consider comparison‐adjusted funnel plots and the accompanying regression test to assess selection bias. We will examine the presence of small‐study effects for the primary outcome only.

Data synthesis

If the clinical and methodological characteristics of individual studies are sufficiently homogeneous, we will pool the data in meta‐analysis. We will perform analyses according to the recommendations in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2021). We will conduct separate meta‐analyses for each comparison.

We will use Review Manager Web for analyses (RevMan Web 2022). One review author will enter the data into the software, and a second review author will check the data for accuracy.

We will use the random‐effects model for all analyses, as we anticipate that true effects will be related, but will not be the same for included studies. For binary outcomes, we will base the estimation of the between‐study variance using the Mantel‐Haenszel method. We will use the inverse variance method for continuous outcomes or outcomes where hazard ratios are available. We plan to explore heterogeneity above 80% with subgroup analyses. If the cause of the heterogeneity cannot be identified, we will not perform a meta‐analysis, but will summarise the results narratively.

We will pool effects within the same:

• period of treatment in relation to cancer treatment (before treatment; during treatment; after treatment);

• period of assessment (up to 12 weeks follow‐up; more than 12 weeks to less than 6 months follow‐up; or 6 months or longer follow‐up).

We will thus potentially conduct nine separate analyses.

Subgroup analysis and investigation of heterogeneity

We will conduct subgroup analysis of treatment effect based on:

• intensity of the intervention based on the rate of perceived exertion or heart rate, or both, or based on the authors' classification (i.e. mild, moderate, vigorous);

• type of cancer of the participants;

• stage of cancer of the participants.

We will also consider conducting further subgroup analysis if data are available based on factors such as format of the intervention (e.g. individual or group, professionally led or not), age of participants (e.g. younger than 65 years or 65 years and older), and type of cancer treatment (e.g. radiation, surgery, chemotherapy, or combination).

Sensitivity analysis

We will conduct sensitivity analysis to assess the effects of including trials with a high overall risk of bias.

We will consider the overall risk of bias as the worst judgement in any of the risk of bias domains, excluding the domains for blinding (i.e. blinding of participants and personnel and blinding of outcome assessment) due to the nature of the interventions.

Summary of findings and assessment of the certainty of the evidence

At least two review authors from the following list (ME, MA, CW, AO, NC, SM, RS, SIM) will independently rate the certainty of the body of evidence for the outcomes. We will use the GRADE approach to rank the certainty of the evidence using GRADEpro GDT (GRADEpro GDT), and the guidelines provided in Chapter 14 of the CochraneHandbook for Systematic Reviews of Interventions (Schünemann 2021).

The GRADE approach uses five considerations (study limitations (risk of bias), unexplained heterogeneity and inconsistency of effect, imprecision, indirectness, and publication bias) to assess the certainty of the body of evidence for each outcome. The GRADE system uses the following criteria for assigning grade of evidence.

• High: we are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of effect, but there is a possibility that it is substantially different.

• Low: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

• Very low: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

The GRADE system considers study design as a marker of quality. Randomised controlled trials are considered to be high certainty evidence, and can be downgraded for important limitations.

Factors that may decrease the certainty level of a body of evidence are as follows.

• Serious or very serious study limitations (risk of bias)

• Important or serious inconsistency of results

• Some or major indirectness of evidence

• Serious or very serious imprecision

• Probability of publication bias

We will create summary of findings tables to present the main findings for the following outcomes:

• cancer‐related fatigue;

• health‐related quality of life;

• adverse events.

We will create separate summary of findings tables for:

• different periods of treatment in relation to cancer treatment (before, during, or after treatment);

• different periods of assessment (up to 12 weeks follow‐up, more than 12 weeks to less than 6 months follow‐up, or 6 months or longer follow‐up).

In particular, we will include key information concerning the certainty of evidence, the magnitude of effect of the interventions examined, and the sum of available data.

For an example of a summary of findings table template (i.e. for aerobic exercise for fatigue in people with cancer during active treatment (measured up to 12 weeks follow‐up)), see Appendix 3.

Notes

Some passages in this review, in particular in the Methods section, are from the standard template of Cochrane Haematology.

Appendices

Appendix 1. MEDLINE search strategy

1. exp exercise/

2. exercise tolerance/

3. exp exertion/

4. Pliability/

5. physical fitness/

6. "Physical Education and Training"/

7. exp physical endurance/

8. exercise therapy/

9. exercising.mp.

10. physical condition$.mp.

11. stamina.mp.

12. motor activity/

13. exercise test/

14. exp Sports/

15. tai chi.mp. or tai ji/

16. yoga/

17. muscle stretching exercises/

18. exp "range of motion, articular"/

19. pilates.mp.

20. qigong.mp.

21. chi kung.mp.

22. resistance training.mp.

23. mind body therap$.mp.

24. exp complementary therapies/

25. Bad Ragaz.mp.

26. Ai Chi.mp.

27. Halliwick.mp.

28. hippotherapy.mp.

29. Hydrotherapy/

30. balance exercise$.mp.

31. aquatic exercise$.mp.

32. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31

33. "quality of life"/

34. exp health status/

35. "activities of daily living"/

36. life qualit$.mp.

37. exp self concept/

38. health level.mp.

39. level of health.mp.

40. wellness.mp.

41. well being.mp.

42. (activities of daily life or daily living activities).mp. [mp=protocol supplementary concept, rare disease supplementary concept, title, original title, abstract, name of substance word, subject heading word, unique identifier]

43. functional ability.mp.

44. good health.mp.

45. healthiness.mp.

46. patient reported outcomes.mp.

47. social adjustment/

48. physical limitations.mp.

49. psychiatric status.mp.

50. pain measurement/

51. functional assessment.mp.

52. fact questionnaire.mp.

53. fact survey.mp.

54. qlc‐c30.mp.

55. facit.mp.

56. toi.mp.

57. (flic or sf‐36 or ces‐d or bdi or sta1 or bfi or hads or lasa or poms or qli or rsci or pais or bpi or msas or mos or ptgi or panas).mp.

58. sense of coherence.mp.

59. randomized.ab.

60. placebo.ab.

61. randomly.ab.

62. trial.ab.

63. randomized controlled trial.pt.

64. controlled clinical trial.pt.

65. random$.ab

66. exp neoplasms/

67. cancer.mp.

68. (neoplasm$ or tumor$ or tumour or malignan$).mp.

69. active treatment.mp.

70. 35 or 33 or 53 or 48 or 42 or 46 or 44 or 55 or 50 or 39 or 57 or 36 or 40 or 51 or 58 or 41 or 47 or 52 or 38 or 34 or 56 or 49 or 37 or 45 or 43 or 54

71. 59 or 60 or 63 or 64 or 61 or 62 or 65

72. 66 or 67 or 68 or 69

73. 32 and 70 and 71 and 72

74. Survivors/

75. survivor.mp.

76. 74 or 75

77. 73 not 76

Appendix 2. Risk of bias assessment

We will assess the following biases for each included study.

• Random sequence generation (checking for possible selection bias). We will assess the method used to generate the allocation sequence as:

  • low risk of bias (any truly random process, e.g. random number table; computer random number generator);

  • unclear risk of bias (insufficient detail about the method of randomisation to permit a judgement of low or high risk of bias);

  • high risk of bias (studies using a non‐random process, e.g. odd or even date of birth; hospital or clinic record number).

• Allocation concealment (checking for possible selection bias). The method used to conceal allocation to interventions prior to assignment determines whether intervention allocation could have been foreseen in advance of, or during, recruitment, or changed after assignment. We will assess the methods as:

  • low risk of bias (e.g. telephone or central randomisation; consecutively numbered, sealed, opaque envelopes);

  • unclear risk of bias (insufficient detail about the method of randomisation to permit a judgement of low or high risk of bias);

  • high risk of bias (studies that do not conceal allocation, e.g. open list).

• Blinding of participants and personnel (checking for possible performance bias). We will assess the methods used to blind study participants and personnel from the knowledge of which intervention a participant received. We will assess the methods as:

  • low risk of bias (study states that it was blinded and describes the method used to achieve blinding, such as identical tablets matched in appearance or smell, or a double‐dummy technique);

  • unclear risk of bias (study states that it was blinded but does not provide an adequate description of how this was achieved);

  • high risk of bias (study states that it was not blinded to the intervention received).

• Blinding of outcome assessment (checking for possible detection bias). We will assess the methods used to blind study participants and outcome assessors from the knowledge of which intervention a participant received. We will assess the methods as:

  • low risk of bias (study has a clear statement that outcome assessors were unaware of treatment allocation, and ideally describes how this was achieved);

  • unclear risk of bias (study states that outcome assessors were blind to treatment allocation but lacks a clear statement on how this was achieved).

  • high risk of bias (study states that outcome assessors were not blinded to treatment allocation).

• Incomplete outcome data (checking for possible attrition bias due to the amount, nature, and handling of incomplete outcome data). We will assess the methods used to deal with incomplete data as:

  • low risk (no missing outcome data; reasons for missing outcome data are unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; missing data have been imputed using 'baseline observation carried forward’ analysis);

  • unclear risk of bias (insufficient reporting of attrition/exclusions to permit a judgement of low risk or high risk (e.g. number randomised not stated, no reasons for missing data provided, or the study did not address this outcome));

  • high risk of bias (reason for missing outcome data is likely to be related to true outcome, with either an imbalance in numbers or reasons for missing data across intervention groups; ‘as‐treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation).

• Selective reporting (checking for reporting bias). We will assess reporting biases due to selective outcome reporting. We will judge studies as:

  • low risk of bias (the study protocol is available, and all of the study’s prespecified (primary and secondary) outcomes that are of interest in the review have been reported in the prespecified way);

  • unclear risk of bias (insufficient information available to permit a judgement of low risk or high risk);

  • high risk of bias (not all of the study’s prespecified primary outcomes have been reported; one or more primary outcomes have been reported using measurements, analysis methods, or subsets of the data (e.g. subscales) that were not prespecified; one or more reported primary outcomes were not prespecified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); one or more outcomes of interest in the review have been reported incompletely so that they cannot be entered in a meta‐analysis; the study report failed to include results for a key outcome that would be expected to have been reported for such a study).

Appendix 3. Summary of findings table

Aerobic exercise for fatigue in people with cancer during active treatment
Patient or population: people who are undergoing active cancer treatment Settings: varied Intervention: aerobic exercise interventions Comparison: usual care
Outcomes	Illustrative comparative risks*		Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Estimated risk or score without aerobic exercise	Risk or relative reduction in score with aerobic exercise
Cancer‐related fatigue (CRF) SD units Investigators measured CRF using different instruments. Lower scores mean less CRF. (follow‐up: up to 12 weeks)
Health‐related quality of life (HRQoL) SD units Investigators measured HRQoL using different instruments. Lower scores mean lower HRQoL. (follow‐up: up to 12 weeks)
Adverse events (follow‐up: up to 12 weeks)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).   CI: confidence interval; RCT: randomised controlled trial; RR: risk ratio; SD: standard deviation
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Footnotes

^aDowngraded by one point: overall risk of bias judged to be high for all studies.

Contributions of authors

Moritz Ernst: methodological expertise, and conception and writing of the review
Marike Andreas: methodological expertise, and conception and writing of the review
Carina Wagner: methodological expertise, and conception and writing of the review
Nora Cryns: methodological expertise and data extraction
Annika Oeser: methodological expertise, and conception and writing of the review
Sarah Messer: methodological expertise, and conception and writing of the review
Paul Bröckelmann: clinical expertise
Ulrike Holtkamp: consumer expertise
Ina Monsef: development of search strategy and search for studies
Roberta W Scherer: search for studies, data extraction, risk of bias assessment, and conception of the review
Shiraz I Mishra: search for studies, data extraction, risk of bias assessment, and conception of the review
Nicole Skoetz: methodological expertise and conception and writing of the review

Sources of support

Internal sources

• University Hospital of Cologne, Germany

  Cochrane Cancer, Department I of Internal Medicine

External sources

• Bundesministerium für Bildung und Forschung, Germany

  This review is funded by the Federal Ministry of Education and Research of Germany (Bundesministerium für Bildung und Forschung (BMBF)), grant number: FKZ 01KG2017.

Declarations of interest

Moritz Ernst: has declared that they have no conflict of interest.
Marike Andreas: reports a grant form the Federal Ministry of Education and Research, Germany (BMBF grant application, No: 01KG2017); paid to institution.
Carina Wagner: reports a grant from the Federal Ministry of Education and Research, Germany (funding number: 01KG2017); paid to institution.
Nora Cryns: has declared that they have no conflict of interest.
Annika Oeser: has declared that they have no conflict of interest.
Sarah Messer: has declared that they have no conflict of interest.
Paul Bröckelmann: reports consultancy fees and travel expenses from BeiGene, Bristol‐Myers Squibb Foudation, Celgene and Takeda Oncology; paid to himself. 
Ulrike Holtkamp: has declared that they have no conflict of interest.
Ina Monsef: has declared that they have no conflict of interest.
Roberta W Scherer: has declared that they have no conflict of interest.
Shiraz I Mishra: has declared that they have no conflict of interest.
Nicole Skoetz: has declared that they have no conflict of interest.

New