Courneya 2008.

Study characteristics
Methods	Study design: RCT Number randomized: 55; 26 to the exercise group and 29 to the control group Study start and stop dates: July 2003 to September 2006 Length of intervention: 12 weeks Length of follow‐up: 1 to 2 weeks after intervention
Participants	Type cancer, n (%): breast cancer (primary) or metastatic disease exercise group: primary breast cancer, 15 (57.7%); metastatic disease, 15 (57.7%) control group: primary breast cancer, 18 (62.1%); metastatic disease, 11 (37.9%) Time since cancer diagnosis: not reported Time in active treatment: not reported Inclusion criteria: ≥ 18 years of age histologically confirmed nonmyeloid cancer diagnosis hemoglobin level 80 to 110 g/L ECOG performance status score of 0 to 2 completed definitive surgery expected survival duration of ≥ 3 months English speaking Eligibility criteria related to interest or ability, or both, to exercise: contraindications to maximal exercise testing Exclusion criteria: iron deficiency (ferritin, 12 μg/L) received an erythropoiesis‐stimulating agent within 4 weeks of randomization uncontrolled hypertension cardiac abnormalities psychiatric illness known hematologic disorder causing anemia substantial lung, pleural, or pericardial disease preexisting bone metastases at high risk for fractures Gender, female, n (%): exercise group: 20 (76.9%) control group: 25 (86.2%) Current age, mean (range) years: exercise group: 58 (40 to 77) years control group: 54 (25 to 77) years Age at cancer diagnosis: not reported Ethnicity/race: not reported Education level, completed university, n (%): exercise group: 16 (61.5%) control group: 13 (44.8%) SES, income > USD60,000 per year, n (%) exercise group: 10 (38.5%) control group: 14 (48.3%) Employment status, employed full or part‐time, n (%): exercise group: 8 (30.8%) control group: 6 (20.7%) Comorbidities, n (%): exercise group: lung disease, 6 (23.1%); current heart disease, 4 (15.3%) control group: lung disease, 5 (17.2%); current heart disease, 5 (17.2%) Past exercise history, current exerciser, n (%): exercise group: 3 (11.5%) control group: 3 (10.3%) On hormone therapy: not reported
Interventions	26 participants assigned to the exercise intervention, including: individually tailored exercise program consisting of 3 cycle ergometry per week, aimed at improving cardiorespiratory fitness darbepoetin alfa treatment at a dose of 4.5 μg/kg on weeks 1, 2, 3, 4, 5, 8, and 11 Type exercise (aerobic/anaerobic): aerobic Intensity of the experimental exercise intervention: 60% to 100% of baseline peak power output Frequency: 3 times per week Duration of individual sessions: not reported Duration of exercise program: 12 weeks Total number of exercise sessions: 36 sessions Format: individual Facility: facility Professionally led by an exercise physiologist Adherence: participants attended 84.2% (30.3/36) of scheduled exercise sessions and achieved the prescribed exercise duration and intensity in 94.7% (28.7/30.3) and 94.1% (28.5/30.3) of the sessions 29 participants assigned to control group, including: darbepoetin alfa treatment at a dose of 4.5 μg/kg on weeks 1, 2, 3, 4, 5, 8, and 11 Contamination of control group: mean (SD) 32 (80) minutes of nonprotocol‐related moderate to strenuous exercise per week
Outcomes	Primary outcome: global HRQoL, assessed using the FACT‐An Other outcomes included: fatigue, assessed using the FACT‐An physiologic outcomes, including: cardiorespiratory fitness, assessed by peak VO2 hemoglobin Outcomes were measured at baseline, 12 weeks, and 13 to 14 weeks: exercise group: n = 29 at baseline, n = 29 at 12 weeks, n = 29 at 13 to 14 weeks control group: n = 26 at baseline, n = 25 at 12 weeks, n = 26 at 13 to 14 weeks Subgroup analysis: none reported Adverse events: not reported
Notes	Country: Canada Funding: Canada Research Chairs Program, Research Team Grant from the National Cancer Institute of Canada, CCS, NCIC/CCS Sociobehavioral Cancer Research Network, Health Research Studentships from the Alberta Heritage Foundation for Medical Research
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"computer‐generated program"
Allocation concealment (selection bias)	Low risk	"The allocation sequence was concealed from the project director who assigned participants to groups"
Blinding of participants and personnel (performance bias) All outcomes	High risk	Owing to the nature of the intervention, it was not possible to conceal allocation to the intervention from the participants
Blinding of outcome assessment (detection bias) All outcomes	High risk	Allocation to the intervention was not concealed from the study personnel
Incomplete outcome data (attrition bias) All outcomes	Low risk	"...according to intention‐to‐treat principles using the last observation carried‐forward method"
Selective reporting (reporting bias)	Low risk	There appears to be no selective reporting of outcomes
Other bias	Low risk	The trial appears to be free of other problems that could put it at a high risk of bias