Article Text

Systematic review and meta-analysis
Exercise improves fatigue during and after breast and prostate cancer treatment, with benefits seen for aerobic exercise
  1. Theresa Pluth Yeo
  1. Department of Surgery, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, USA
  1. Correspondence to : Theresa Pluth Yeo
    Department of Surgery, Thomas Jefferson University Hospital, 1025 Walnut Street, Suite 605B, Philadelphia, PA 19107, USA; Theresa.yeo{at}

Statistics from

Commentary on: OpenUrlPubMed

Implications for practice and research

  • Encouraging cancer patients to follow a structured exercise programme reduces self-reported fatigue.

  • The optimal exercise programme (aerobic, resistance, mind–body or a combined programme) remains undecided; however, evidence supports regularly participating in an exercise programme either at home or in a formal facility.

  • Longitudinal randomised controlled trials (RCTs) of exercise interventions in cancer patients that include biomarkers of the interacting pathways in cancer-related fatigue, such as the inflammatory, nervous system, metabolic and hypothalamic-pituitary-adrenal (HPA)-axis, are critically needed.


Cramp and Byron-Daniel conducted a systematic review of studies addressing the important issue of exercise and its therapeutic benefit in reducing cancer-related fatigue (CRF) both during and after cancer treatment. Fatigue is a commonly reported symptom among newly diagnosed cancer patients (70–100%), those undergoing treatment and even long-term survivors. CRF is influenced by tumour factors, muscle wasting, weight loss, neurological mediators, cancer treatments, decreased sleep, acute anaemia and disease comorbidities. CRF is often described as more distressing than pain and therefore of interest to patients and cancer-care providers.


The current intervention review is an update of the original 2008 review. Included in the analysis are 56 RCTs (4068 total participants); 28 studies from the original review and 28 studies published between 1996 and March 2011 which were included via an extensive literature search. The majority of the studies were conducted in patients with breast cancer (28), followed by prostate cancer (7), haematological malignancies (5), colorectal cancer (2), head and neck cancer (2) and 14 studies included patients with other types of cancers. Thirty studies used an aerobic exercise programme, six used a resistance intervention, eight used mind–body programmes and 12 included multimodal exercise interventions.

The analysis focused on the risk of bias in the studies and improvement in fatigue. Data were extracted based on predefined criteria. When sufficient data were available, meta-analyses were performed using a random-effect model for the outcome of fatigue.


The review found statistically significant evidence that aerobic exercise, but not resistance training or mind–body techniques, was beneficial in reducing CRF, both during treatment as well as after the completion of therapy. Reduction of CRF was evident in the breast and prostate cancer populations, but not in those with haematological malignancies.

One of the weaknesses of all included studies was the lack of long-term patient follow-up regarding exercise maintenance after the completion of the study.

The authors report the potential for selective reporting of positive outcomes in some studies that may lead to overemphasis of the effects of exercise. Additionally, group contamination was found in several studies. This occurs when the usual care group engages in an exercise programme of their own undertaking or when the intervention group does not adhere to their programme. Systematic bias may be introduced into a study when the usual care group receives little or no attention. On the other hand, the intervention group may suffer from the Hawthorne effect and deliberately misrepresent their responses in order to please the researcher. It is interesting to note that, despite many of these common research problems, the majority of the studies found exercise effective in reducing CRF.


Are the results observed in mainly breast and prostate cancer patients in this review applicable to patients with other types of solid tumour? A 2012 RCT of a home walking programme in 102 postoperative pancreas and periampullary cancer patients also demonstrated reduction in fatigue levels, as well as improvement in anxiety and depression, in the walking group as compared to the usual care group.1

Although resistance training and mind-body techniques did not achieve significance, this may have been owing to small sample size or insensitive outcome measures and are not indicative of the individual benefits realised with those programmes. These modalities remain to be prospectively evaluated in large-scale longitudinal investigations. Monitoring of exercise adherence is problematic, particularly in home-based studies. Adding a body monitoring device to capture quantitative measurements of activity would be helpful.

Complex bio-behavioural mechanisms involving the immune system, metabolic pathways, sympathetic nervous system, HPA-axis, genetic polymorphisms and tumour characteristics are all involved in CRF. Correlating these processes and monitoring patients longitudinally will improve our ability to determine who will respond to which therapies, and to monitor their progress. Current and future research will address these interrelated processes to arrive at therapeutic interventions that address not only behavioural targets but also elucidate the underlying mechanisms of CRF. In summary, there seems to be little downside to encouraging cancer patients to perform low-to-moderate intensity exercise provided that comorbidities do not limit them in doing so.

View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.