EUROPEAN UROLOGY 69 (2016) 693–703

available at www.sciencedirect.com
journal homepage: www.europeanurology.com

Collaborative [4_TD$IF]Review – [5_TD$IF]Prostatic [6_TD$IF]Disease

Exercise for Men with Prostate Cancer: A Systematic Review

and Meta-analysis

Liam Bourke a,[3_TD$IF] *, Dianna Smith b, Liz Steed c, Richard Hooper c, Anouska Carter a, James Catto d,
Peter C. Albertsen e, Bertrand Tombal f, Heather A. Payne g, Derek J. Rosario h
a b
Health and Wellbeing Research Institute, Centre for Sport and Exercise Science, Sheffield Hallam University, Sheffield, UK; Geography & Environment,
c
University of Southampton, Southampton, UK; Centre for Primary Care and Public Health, Barts and The London School of Medicine and Dentistry, Queen
Mary University of London, London, UK; d Academic Urology Unit and Institute for Cancer Studies, The Medical School, University of Sheffield, Sheffield, UK;
e
Department of Surgery, Division of Urology, University of Connecticut Health Center, Farmington, CT, USA; f University Clinics Saint Luc/Catholic University
of Louvain, Brussels, Belgium; g Department of Oncology, University College London Hospitals, London, UK; h Department of Oncology, University of Sheffield,
Sheffield, UK

Article info Abstract

Article history: Context: Exercise could be beneficial for prostate cancer survivors. However, no sys-
Accepted October 23, 2015 tematic review across cancer stages and treatment types addressing potential benefits
and harms exists to date.
Associate Editor: Objective: To assess the effects of exercise on cancer-specific quality of life and adverse
Giacomo Novara events in prostate cancer trials.
Evidence acquisition: We searched the Cochrane Central Register of Controlled Trials,
MEDLINE, EMBASE, AMED, CINAHL, PsycINFO, SPORTDiscus, and PEDro. We also
Keywords: searched grey literature databases, including trial registers. Searches were from data-
Exercise base inception to March 2015. Standardised mean differences (SMDs) were calculated
for meta-analysis.
Prostate cancer
Evidence synthesis: We included 16 randomised controlled trials (RCTs) involving
Quality of life 1574 men with prostate cancer. Follow-up varied from 8 wk to 12 mo. RCTs involved
Fatigue men with stage I–IV cancers. A high risk of bias was frequently due to problematic
Adverse effects intervention adherence. Seven trials involving 912 men measured cancer-specific
quality of life. Pooling of the data from these seven trials revealed no significant effect
on this outcome (SMD 0.13, 95% confidence interval [CI] –0.08 to 0.34, median follow-up
12 wk). Sensitivity analysis of studies that were judged to be of high quality indicated a
moderate positive effect estimate (SMD 0.33, 95% CI 0.08–0.58; median follow-up
12 wk). Similar beneficial effects were seen for cancer-specific fatigue, submaximal
fitness, and lower body strength. We found no evidence of benefit for disease progres-
sion, cardiovascular health, or sexual function. There were no deaths attributable to
exercise interventions. Other serious adverse events (eg, myocardial infarction) were
equivalent to those seen in controls.
Conclusions: These results support the hypothesis that exercise interventions improve
cancer-specific quality of life, cancer-specific fatigue, submaximal fitness, and lower
body strength.
Patient summary: This review shows that exercise[8_TD$IF]/physical activity interventions can
improve quality of life, fatigue, fitness, and function for men with prostate cancer.
# 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.

* Corresponding author. Health and Wellbeing Research Institute, Centre for Sport and Exercise
Science, Sheffield Hallam University, Sheffield S10 2BP, UK. Tel. +44 114 2252374;
Fax: +44 12345678.
E-mail address: l.bourke@shu.ac.uk (L. Bourke).
http://dx.doi.org/10.1016/j.eururo.2015.10.047
0302-2838/# 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.
694 EUROPEAN UROLOGY 69 (2016) 693–703
1. Introduction
Prostate cancer is the primary cause of years lived with
cancer disability in the Americas, Northwest Europe,
Australia, New Zealand, and much of sub-Saharan Africa
[1]. Management of prostate cancer ranges from no
intervention (active surveillance or watchful waiting)
to radical local treatment (prostatectomy and radiation
therapy) with or without combined androgen deprivation
therapy (ADT), ADT alone, to taxane-based chemotherapy
for progressive castration-resistant disease [2] and second-
line hormone agents [3,4]. First-line radical treatment for
prostate cancer can negatively impact quality of life
(eg, erectile dysfunction, incontinence, radiation proctitis),
as can ADT (eg, loss of muscle mass, fatigue, psychological
morbidity, higher risk of cardiovascular disease and bone
fracture) [5,6]. Direct symptoms for advanced or metastatic
cancer (eg, pain, hypercalcaemia, spinal cord compression,
pathological fractures) can also adversely affect health
[7,8].
Several recent systematic reviews have examined the
effects of exercise in cancer survivors in terms of quality of
life [9,10], exercise behaviour [11], and fatigue [12]. These
reviews cover an amalgamation of heterogeneous primary
cancers. Most evidence comes from trials in breast cancer and
thus cannot be generalised to men with prostate cancer.
Furthermore, exercise therapy appears to be beneficial in the
short term, but little is known about dose, duration, and
longer-term effects of such therapy, including adverse effects
over extended follow-up. Finally, despite the potential health
benefits for men with prostate cancer, few clinicians are
aware of the role of exercise, and in many cases it goes
unprescribed. The aim of this review was primarily to
evaluate the effect of exercise interventions on cancer-
specific quality of life after prostate cancer diagnosis and to
assess adverse effects.
2. Evidence acquisition
Methods for this systematic review have been described in
detail elsewhere [13]. In brief, the primary review outcomes
were quality of life and adverse events. Secondary outcomes
include effects on fatigue, disease progression, cardiovascu-
lar health, physical fitness and function, and sexual function.
We searched the Cochrane Central Register of Controlled
Trials, MEDLINE, EMBASE, AMED, CINAHL, PsycINFO,
SPORTDiscus, and PEDro databases from inception to March
31, 2015. We expanded the database search by attempting
to identify unpublished studies and references in the grey
literature (via the OpenGrey database). We also searched
the World Health Organization (WHO) trials page, the
ISRCTN meta-register of controlled trials (www.isrctn.com),
and ClinicalTrials.gov.
2.1. Inclusion and exclusion criteria
We included only randomised controlled trials (RCTs)
involving adults in which trial participants had been
diagnosed with prostate cancer. Only interventions that
included a component targeted at increasing aerobic
exercise and/or resistance exercise behaviour compared
with a usual care or waiting-list control group with at least
6 wk of follow-up (from trial baseline assessment) were
considered in the review. We excluded trials addressing
recovery of continence only. Only studies that reported the
frequency, duration, and intensity of aerobic exercise
behaviour, or the frequency, intensity, type, sets, and
repetitions of resistance exercise behaviour as prescribed
in the intervention were included in the review.
2.2. Data extraction
After extraction piloting, three review authors (L.B., D.S.,
and A.C.) worked independently to screen all titles and
abstracts to identify records that met the inclusion criteria
or that could not be safely excluded without assessment of
the full text (eg, when no abstract was available).
Disagreements at this stage were resolved by discussion
with another review author (D.J.R.). Full-text articles for
these records were retrieved. After training to ensure a
consistent approach to study assessment and data abstrac-
tion, three review authors (L.B., D.S., and A.C.) worked
independently to assess the full-text articles retrieved. The
selection process is documented in a Preferred Reporting
Items for Systematic Reviews and Meta-analyses (PRISMA)
flow diagram (Figure 1) [14].
The review authors did not conduct data extraction for
any primary studies for which they were listed as an author.
Data were entered into the statistical software of The
Cochrane Collaboration Review Manager (RevMan 5) for
calculation of meta-analyses. Where appropriate, we
contacted study authors to request information that was
missing from reports for the studies included.
The risk of bias was assessed using The Cochrane
Collaboration tool [15]. Two of three review authors (L.B.,
D.S., and A.C.) applied the risk-of-bias tool independently to
each study. Differences were resolved by discussion or by
appeal to a third review author (D.J.R.). Review authors did
not assess the risk of bias for any studies for which they
were an author. The results are summarised in Figure 2.
2.3. Data synthesis
If the data available were sufficient and if it was appropriate
to do so, we performed a meta-analysis using Review
Manager software. I2 calculations were performed in STATA.
If statistical heterogeneity was noted, meta-analysis was
performed using a random-effects model. Fixed-effect
models were used only if no significant statistical hetero-
geneity was present. We noted the time points at which
outcomes were collected and reported. If adverse effects
data were insufficient or if meta-analysis was not appro-
priate, we provide a narrative synthesis.
For continuous outcomes (eg, cancer-specific quality of
life), we extracted the point estimate for the measure
of central tendency for the final value of the outcome
of interest and the number of participants assessed at
stated follow-up in each treatment arm to estimate the
 EUROPEAN UROLOGY 69 (2016) 693–703 695
[(Fig._1)TD$IG]
All database search results Other sources
References from hand searching and
MEDLINE: 2085 snowballing: 22
EMBASE: 728
CENTRAL: 878
PEDRO: 128
CINAHL: 1352
PsychINFO: 215
Sports Discus: 134
AMED: 36
Total: 5556

Duplicates removed
=1200

Results aer deduplicaon: 4356

Excluded = 4120

Abstracts remaining aer screening by tle: 236

Excluded = 145

Full text remaining aer screening by abstract: 91

Excluded = 67

Results aer screening by full text: 24

manuscripts from 16 RCTs.

Fig. 1 – Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flow diagram. RCT = randomised controlled trial.

standardised mean difference (SMD) between treatment 2.5. Assessment of heterogeneity

arms and its 95% confidence interval (CI).
2.4. Unit of analysis
We did not include crossover trials in this review because of
the difficulty involved in washing out interventions to
change behaviour. For trials with multiple intervention
groups, we first eliminated groups for which the interven-
tion did not meet the criteria for inclusion in the review. We
then combined all relevant intervention groups to create a
single pairwise comparison with the control group.
We used clinical expertise to judge whether it was
appropriate to combine trials in a meta-analysis. Consis-
tency of results was assessed using the I2 statistic and its
95% CI. Data were analysed using RevMan and Stata 12.
2.6. Sensitivity analysis
Results of meta-analyses were interpreted in light of the
findings with respect to risk of bias. Risk of bias was
assessed for each follow-up. For sensitivity analysis, we
696
[(Fig._2)TD$IG]
Fig. 2 – Risk of bias for the trials included.
used the longest follow-up for which there was a low risk of
bias. Where appropriate, we contacted study authors for
additional information or for further clarification of study
methods if any doubt arose regarding sources of bias.
2.7. Subgroup analysis
If a sufficient number of studies were identified and if the
reporting resolution was adequate, we performed subgroup
EUROPEAN UROLOGY 69 (2016) 693–703
analyses for anticancer treatment received, cancer stage,
obesity, previous physical activity at baseline, and explan-
atory (efficacy) versus pragmatic (effectiveness) trial
designs. We categorised interventions according to the
theoretical basis, the behaviour change techniques, and
categorisation using the Coventry, Aberdeen & London-
Refined (CALO-RE) taxonomy [14].
3. Evidence synthesis
3.1. Search results
Figure 1 shows the results for the literature searches and
screening process for the review. We identified 4356 unique
records from database searches and 22 manuscripts through
grey literature and hand checking of references for studies
included and related systematic reviews [16,17]. After
reviewing by title and abstract, we evaluated the full text
for 91 records, after which 67 studies were excluded from the
review. All full-text manuscripts were available in English.
We sent 31 emails requesting further information on
published manuscripts and received four responses (only
one of which provided new data).
3.2. Studies included
We included 16 RCTs [18–33] (Table 1) involving 1574 men
with prostate cancer (sample size 20–423). We also
found eight linked manuscripts with secondary analysis
[34–41]. All studies were randomised at the patient level.
Follow-up varied from 8 wk to 12 mo. RCTs involved men
with stage I–IV cancers, but we found no trials for men
exclusively undergoing chemotherapy (two trials included a
small proportion of men who had chemotherapy) [26,33]. Ex-
ercise interventions were supervised [22,27,29–31], home-
based [21,24,32], a mix of supervised and home-based
[18,23,25,33], or supervised with suggested home-based
activity [19,20]; two were unclear [26,28]. Exercise behav-
iour (dose) was monitored using objective [22,29–32],
subjective [24,26], or a mixture of objective or subjective
methods [18,23,25,33], was not monitored [21], or was
unclear [19,20,27,28]. All trials included a usual care
comparator. Two trials supplemented usual care with
standard exercise advice for cancer survivors [23,26]. Our
previous Cochrane review [11] demonstrated that simple
advice is highly unlikely to improve exercise behaviour, so we
judged these studies eligible for inclusion.
The behaviour change techniques used primarily focused
on instruction on how to perform behaviour, with practice
and goals set by trainers. Three trials [18,21,26] reported a
more psychological approach to changing behaviours by
incorporating techniques such as problem solving, social
support, and client-set goals. Of interest in comparison to
our previous review, significantly more studies reported
that they taught generalisation of behaviour; however,
association with outcome was not possible in this review
owing to the small number of studies per outcome. All
studies were conducted in countries categorised as high
income by WHO.
Table 1 – Description of studies included in the review

Study N randomised Follow-up Participants and treatment Intervention Review outcome measures

Bourke et al [18] Exercise n = 50 Baseline, 3 mo, Tumour stage: T3–4 Aerobic frequency: 3 times/wk PCa-specific QoL: FACT-P
Usual care n = 50 6 mo Current PCa treatment: ADT Aerobic intensity: 55–75% of age-predicted HRmax or Adverse events: reported
Metastatic disease: 11 in the exercise and 9 in 11–13 on the Borg rating of perceived exertion scale PCa-specific fatigue: FACT-F
the usual care group Aerobic duration: 30 min CV health: systolic and diastolic BP
Resistance frequency: 3 times/wk Physical function: aerobic exercise
Resistance sets: 2–4 tolerance by submaximal treadmill test
Resistance reps: 8–12
Resistance load: 60% of 1 RM
Cormie et al [19] Exercise n = 10 Baseline, 12 wk Tumour stage: not clear Aerobic frequency: unclear Adverse events: reported
Usual care n = 10 Current PCa treatment: unclear; all men had Aerobic intensity: unclear PCa-specific fatigue: Multidimensional
previous ADT Aerobic duration: 150 min/wk Fatigue Symptom Inventory-Short Form
Metastatic disease: all men Resistance frequency: twice/wk questionnaire
Resistance sets: 2–4 Physical function: (1) one repetition max
Resistance reps: 8–12 in leg extension (2) 400-m walk, (3) usual
Resistance load: 8–12 RM and fast-pace 6-m walk, (4) timed ‘up and

EUROPEAN UROLOGY 69 (2016) 693–703

go’ test
Cormie et al [20] Exercise n = 32 Baseline, 3 mo Tumour stage(s): not clear Aerobic frequency: twice/wk PCa-specific QoL: QLQ-PR25
Usual care n = 31 Current PCa treatment: ADT, RT Aerobic intensity: target intensity 70–85% of Adverse events: reported
Metastatic disease: none estimated HRmax PCa-specific fatigue: FACT-F
Aerobic duration: 20–30 min Disease progression: PSA
Resistance frequency: twice/wk CV health: systolic and diastolic BP
Resistance sets: 1–4 Physical function: 400-m walk, leg press,
Resistance reps: 6–12 chest press, seated row
Resistance load: 60–85% of 1 RM Sexual function: QLQ-PR25 sexual domain
Dieperink et al [21] Intervention n = 79 12 wk before RT, Tumour stage: T1–3 Aerobic frequency: 7 d/wk PCa-specific QoL: EPIC questionnaire
Control n = 82 before intervention Current PCa treatment: RT and ADT Aerobic intensity: lower limit for moderate-intensity Sexual function: EPIC questionnaire
(4 wk after RT), and Metastatic disease: n = 3 in each group physical exercise corresponds to walking at an average sexual sum score
21–22 wk thereafter received pelvic RT for metastatic lymph speed of 4 km/h
nodes Aerobic duration:30 min/d
Galvão et al [22] Exercise n = 29 Baseline, 12 wk Tumour stage: unclear Aerobic frequency: twice/wk PCa-specific QoL: QLQ-C30 and QLQ-PR25
Usual care n = 28 Current PCa treatment: RT + ADT Aerobic intensity: 65–80% HRmax and perceived Adverse events: reported
Nodal metastases: n = 2 in exercise and exertion 11–13 (6- to 20-point Borg scale) PCa-specific fatigue: QLQ-C30 fatigue
n = 3 in usual care group Aerobic duration: 15–20 min of CV exercisesa domain
Resistance frequency: twice/wk Disease progression: PSA
Resistance sets: 2–4 Physical function: 400-m walk, dynamic
Resistance reps: 6–12 strength for upper and lower body
Resistance load: 6–12 RM Sexual function: QLQ-PR25
Galvão et al [23] Exercise n = 50 Baseline, 6 mo, Tumour stage: T2–4 Aerobic frequency: 4 times/wk Adverse events: reported
Usual care n = 50 12 mo Current PCa treatment: none (previous Aerobic intensity: 70–85% HRmax and perceived Disease progression: PSA
ADT + RT + bisphosphonate exertion at 11–13 (6- to 20-point Borg scale) CV health: BP
Metastatic disease: excluded from trial Aerobic duration: 20–30 min of CV exercises Physical function: submaximal exercise
Resistance frequency: twice/wk tolerance, chair rise test
Resistance sets: 2–4
Resistance reps: 6–12
Resistance load: 6–12 RM
Hebert et al [24] Exercise n = 29 Baseline, 3 mo, Tumour stage: unclear Aerobic frequency: 5 times/wk Adverse events: unclear
Control n = 25 6 mo Current PCa treatment: previous surgery, RT, Aerobic intensity: 3.0–6.0 MET min or 4–7 kcal/min Disease progression: PSA
or both Aerobic duration: 30 min
Metastatic disease: unclear

697
 698
Table 1 (Continued )

Study N randomised Follow-up Participants and treatment Intervention Review outcome measures
Jones et al [25] Exercise n = 25 Baseline, 6 mo Tumour stage: T1–2 Aerobic frequency: five sessions/wk PCa-specific QoL: FACT-G and FACT-P
Usual care n = 25 Current PCa treatment: previous bilateral Aerobic intensity: 55–100% of VO2 peak Adverse events: reported
nerve-sparing RP Aerobic duration: 30–45 min/session PCa-specific fatigue: FACT-F
Metastatic disease: none CV health: BP
Physical function: VO2 peak
Sexual function: IIEF questionnaire
McGowan et al [26] Physical activity Baseline, 1 mo, Tumour stage: unclear Aerobic frequency: unclear PCa-specific QoL: FACT-P
guidelines n = 141 3 mo Current PCa treatment: watchful waiting, Aerobic intensity [22]: 500–1000 MET min/wk PCa-specific fatigue: FACT-F
Self-administered surgery, RT, ADT, CTx, PCa recurrence Aerobic duration: 150–300 min/wk, or increase
implementation Metastatic disease: 1.9% of the cohort physical activity by at least 60 min/wk if already
intention n = 141 meeting the guidelines
Telephone-assisted
implementation
intention n = 141
Monga et al [27] Unclear (n = 30 Baseline, 8 wk Tumour stage: unclear Aerobic frequency: 3 times/wk PCa-specific QoL: FACT-P

EUROPEAN UROLOGY 69 (2016) 693–703

randomised in total) Current PCa treatment: RT Aerobic intensity: target HR calculated as [0.65  PCa-specific fatigue: Piper fatigue scale
Metastatic disease: unclear (HRmax – resting HR)] + resting HR Physical function: submaximal fitness
Aerobic duration: 30 min of walking on a treadmill (Bruce treadmill protocol) and timed five
reps of chair sit-to-stand test
Park et al [28] Exercise n = 33 Week before RP, Tumour stage: pT2a–3a Aerobic frequency: 2 times/wk Adverse events: reported
Usual care n = 33 before exercise Current PCa treatment: post RP Aerobic intensity: 45–75% of HR reserve maximum and Physical function: sit-ups, chair stand,
(3 wk after RP), Metastatic disease: unclear 9–13 rated perceived exertion dominant grip strength, adduction ability,
after exercise Aerobic duration: 60 min back lift, and knee lift performed for 2 min
(15 wk after RP)
Segal et al [29] Exercise n = 82 Baseline, 12 wk Tumour stage: I–IV Resistance frequency: 3 times/wk PCa-specific QoL: FACT-P
Control n = 73 Current PCa treatment: ADT Resistance sets: 2 PCa-specific fatigue: FACT-F
Metastatic disease: unclear Resistance reps: 8–12 Disease progression: PSA
Resistance load: 60–70% of 1 RM Physical function: dynamic muscle
endurance for upper and lower body
Segal et al [30] Resistance exercise Baseline, 12 wk, Tumour stage: I–IV Aerobic frequency: 3 times/wk PCa-specific QoL: FACT-P
n = 40 24 wk Current PCa treatment: RT and ADT (61% of Aerobic intensity: up to 60% of predetermined VO2 peak Adverse events: reported
Aerobic exercise cohort on ADT) for weeks 1–4, progressing to 70–75% for weeks 5–24 PCa-specific fatigue: FACT-F
n = 40 Metastatic disease: none (excluded from Aerobic duration: starting at 15 min, increased by Disease progression: PSA
Control n = 41 trial) 5 min every 3 wk to reached 45 min Physical function: VO2 peak, dynamic
Resistance frequency: 3 times/wk strength for upper and lower body
Resistance sets: 2
Resistance reps: 8–12
Resistance load: 60–70% of estimated 1 RM
Uth et al [31] Exercise n = 29 Baseline, 12 wk Tumour stage: unclear Aerobic frequency: 2–3 times/wk Adverse events: reported
Usual care n = 28 Current PCa treatment: ADT (previous RT) Aerobic intensity: 70–100% HRmax Physical function: VO2 peak, knee-
Metastatic disease: nodal metastases, 14% of Aerobic duration: in weeks 1–4, football training extensor maximal strength, chair sit-to-
exercise and 35% of usual care group; bone consisted of two weekly sessions, starting with 15 min stand test
metastases, 24% of exercise and 15% of usual of warm-up exercises (running, dribbling, passing,
care group shooting, balance, and muscle strength) followed by
two 15-min 5–7-a-side games; in weeks 5–8, each
session increased to three 15-min games after warming
up; in weeks 9–12, there were three weekly training
sessions of the same duration
 EUROPEAN UROLOGY 69 (2016) 693–703 699

3.3. Risk of bias and quality

QLQ-PR25 = EORTC prostate-specific module; EPIC = Expanded Prostate Cancer Index composite; IIEF = International Index of Erectile Function; CV = cardiovascular; MET = metabolic equivalent; VO2 peak = peak oxygen
consumption; RM = repetition maximum; HR = heart rate; QoL = quality of life; PCa = prostate cancer; PSA = prostate-specific antigen; RT = radiation therapy; BP = blood pressure; RP = radical prostatectomy;
FACT-P = Functional Assessment of Cancer Therapy-Prostate; FACT-G = FACT-General; FACT-F = FACT-Fatigue; QLQ-C30 = European Organisation for Research and Treatment of Cancer (EORTC) core QoL questionnaire;
Physical function: bench press 1 RM, leg
press 1 RM, chair stand, 4-m fast walk
PCa-specific fatigue: Schwartz fatigue
Figure 2 shows risk-of-bias judgements made for the
PCa-specific fatigue: Brief Fatigue

studies included. Supplementary File 1 describes these

judgements in detail. All trials were judged to have a high
Adverse events: reported
Inventory questionnaire

risk of bias for blinding of participants given that it is not

possible to blind the participant in an exercise intervention.
However, we did not judge that this necessarily compro-
mised study quality. The most common issues around high
risk of bias that would impact on study quality were level of
scale

study attrition during at least one follow-up point, poor

intervention adherence, lack of investigator blinding, and
Resistance load: lower body and displacement 0–15% of

selective reporting bias.

Resistance sets: displacement 1–10, lower body 1–2,

Resistance reps: displacement 10, lower body 8–12,

3.4. Effects of interventions on primary review objectives

Aerobic intensity: 60–70% of calculated HRmax
Aerobic frequency: at least 3 d/wk during RT

Seven trials involving 912 men measured cancer-specific

body weight; upper body 10–15 RM

quality of life using a tool that gave an overall/summary

Resistance frequency: 3 times/wk

score that could used in a meta-analysis [18,22,

25–27,29,30]. No significant effect on this outcome was
Aerobic duration: 30 min

found from pooling the data from these seven trials (SMD
0.13, 95% CI –0.08 to 0.34). No statistical heterogeneity was
observed (I2 46%, 95% CI 0–76%). Sensitivity analysis of
upper body 8–14
upper body 1–2

studies that were judged to be of high quality [18,22,30]

(note that 3-mo follow-up data were used from the study by
Bourke et al [18]) indicated a moderate positive effect
estimate (SMD 0.33, 95% CI 0.08–0.58) with no significant
heterogeneity (I2 0%, 95% CI 0–73%). Figure 3 shows a forest
66, including 10/33 in the control and 9/33 in
hormone therapy for high-risk tumours in19/

RT; 7% of exercise and 14% of control group

Current PCa treatment: RT for all; adjuvant

plot of the results.

exercise and 50% of control group received
Current PCa treatment: ADT for all; 45% of

Metastatic disease: 27.6% of exercise and

Ten studies reported information on adverse events

involving 685 men. Four studies reported no adverse events
[20,22,28,33]. Two studies reported deaths, one due to lung
Tumour stage: T1–2 for 51/65

cancer [23] and one in the control arm of the trial [18]. One
Metastatic disease: unclear

incident of acute myocardial infarction (MI) in a participant

Tumour stage: unclear

13.6% of control group

Participants were encouraged to supplement with exercise at home to reach 150 min/wk.

with no previous cardiac history required hospitalisation

the exercise group

and resuscitation after only the third day of the aerobic

training protocol [30]. In the 2014 trial by Galvão et al [23],
received CTx

one participant in the control group with no previous

history of cardiac disease had a nonfatal MI during the
second half of the study, but made a full recovery. One study
reported adverse electrocardiograph changes during exer-
cise testing in terms of significant ST segment depression in
Baseline, 6 mo,
Baseline, 4 wk,

three patients [25]. One study reported two fibula fractures

in the intervention group [31], one of which revealed
underlying peripheral neuropathy. In one study, one man in
12 mo
8 wk

the exercise group fell while dressing at home and suffered

a fractured rib [19] but was able to complete the final 2 wk
of the intervention with a modified prescription. Three
studies reported mixed musculoskeletal complications
Exercise n = 33

Exercise n = 29
Control n = 33

Control n = 22

from pre-existing back and knee pain [23], training-induced

leg cramps or back pain [25], and three minor tendon/
ligament/quadriceps injuries [31] [9_TD$IF]with exercise.
CTx = chemotherapy.

3.5. Secondary review outcomes

Windsor et al [32]

Winters-Stone

Positive beneficial effects were observed for cancer-specific

et al [33]

fatigue, lower body strength, and aerobic fitness. Supple-

mentary File 2 describes the meta-analysis for secondary
review outcomes. No effects on cardiovascular health or
a
700 EUROPEAN UROLOGY 69 (2016) 693–703
[(Fig._3)TD$IG]
Fig. 3 – Forest plot of (A) quality-of-life outcomes for the trials included and (B) sensitivity analysis. SD = standard [2_TD$IF]deviation; SMD = standardised mean
difference; CI = confidence interval; df = degrees of freedom.
disease progression outcomes were observed. A borderline
positive effect on sexual activity was evident (p = 0.05) but
there was no effect on sexual function.
3.6. Planned subgroup analysis and CALO-RE taxonomy for
behaviour change technique
Supplementary File 3 describes planned subgroup out-
comes and results for the CALO-RE taxonomy data.
3.7. Discussion
Sixteen trials involving 1574 men with prostate cancer were
included in the review. From sensitivity analysis, we found
high quality evidence that exercise interventions can
improve cancer specific quality of life and cancer specific
fatigue in men with prostate cancer at up to 6 mo of follow-
up (with moderate beneficial effect estimate). There were no
deaths attributable to exercise interventions. Other serious
adverse events as a result of exercise (eg, MI) were equivalent
to those seen in controls. In one trial that used a competitive
contact sport as the intervention (football) a high rate of
lower limb fracture was seen in the intervention arm. More
frequently, soft tissue complications such as minor muscu-
loskeletal sprains and strains were reported for intervention
groups in more controlled settings. No effect on cardiovas-
cular health or disease progression outcomes was observed.
Positive beneficial effects were evident for lower body
strength and aerobic fitness. A borderline positive effect for
sexual activity should be viewed with caution as the data
were taken from two small trials.
We specifically selected only trials that reported key
metrics for exercise prescription to support reproducibility.
In doing so, we synthesised data for 11 more RCTs than a
recent systematic review [17]. Furthermore, to the best of
our knowledge, this is the first review to report a quantified
meta-analysis of effect estimates for key patient outcomes
such as cancer-specific quality of life and fatigue. Our
review offers the most up-to-date evidence on adverse
effects systematically gathered from an exhaustive review
of RCTs. Our meta-analysis of improvements in sexual
activity is unique but should be interpreted with caution as
the data are taken from just two available trials.
Much of the uncertainty in judging trial bias came from
poor reporting around randomisation procedures in terms
of both sequence generation and allocation concealment;
however, no trial was judged to be at high risk of bias. As for
other systematic reviews undertaken by our group
[10_TD$IF]evaluating exercise in cancer populations [11], we did not
penalise trials for being at high risk of performance bias for
blinding of participants. Furthermore, bias is not likely
because trials with poor adherence to the exercise
intervention commonly have no effect on clinical outcomes
[24,26]. It is not possible to blind participants to taking part
in an exercise intervention. Some trials have suggested this
should be addressed by sham exercise conditions. However,
given that aerobic exercise recommendation guidelines for
survivors are freely available on the Internet (eg, from the
American Cancer Society) and are often positively promoted
by care providers and cancer support charities (eg,
Macmillan), the legitimacy of any sham condition seems
dubious.
As for any behaviour change intervention, requiring
participants to maintain[1_TD$IF] an increase in exercise behaviour
can be very challenging[12_TD$IF]. [13_TD$IF]Poor [14_TD$IF]adherence [15_TD$IF]to [16_TD$IF]the [17_TD$IF]intervention
[18_TD$IF]prescription [19_TD$IF]means [13_TD$IF]that [20_TD$IF]the [21_TD$IF]intended dose of [2_TD$IF]the
 EUROPEAN UROLOGY 69 (2016) 693–703
[23_TD$IF]intervention was not received. This creates obvious
problems when trying to understand the true effect of
the intervention [24_TD$IF]on trial outcomes. The major reasons why
trials were judged to have a high risk of bias were attrition
and adherence biases, which we judged would have a
substantial impact on the quality of evidence. Selective
reporting bias—particularly with regard to adverse events—
was the other most prevalent issue.
Three studies reported data over up to 12 mo of follow-
up [23,25,33], but all were judged to be at high risk of bias.
Thus, for harmonisation with other high-quality evidence
where possible, we only extracted 6-mo data for use in meta-
analyses. Therefore, the long-term durability of some of the
key findings of this review is uncertain. The studies reported
here involved a mixture of T-stage cancers, with some studies
including men with T1–4 disease [29,30]. This limits the
certainty for recommendations stratified by disease stage
(indeed we were not able to conduct planned subgroup
analyses). This is also true for treatment type, although
several meta-analyses of high-quality studies are largely
representative of men on ADT. High-quality studies are
required for men with earlier disease stages undergoing
radical treatment. No evidence was found for men undergo-
ing chemotherapy (apart from a very small minority of the
cohort in two trials [26,33]). Furthermore, the value of
exercise interventions among men on newer hormone
treatments such as enzalutamide is not clear. In addition,
we did not find any evidence for men undergoing more recent
radical innovations such as high-intensity focused ultra-
sound.
All the studies were taken from peer-reviewed journals as
we were unable to locate any unpublished results, despite
contacting internationally recognised experts in the field.
We found some evidence that exercise might have a
beneficial impact on sexual activity, but in the absence of
concurrent improvements in sexual function, the value of
this finding to patients is uncertain. It should also be noted
that the majority of these interventions took place in a
controlled environment.
All the studies were conducted in countries classified as
high income by WHO. No evidence was derived from
developing countries, and it is uncertain whether the
resources and/or infrastructure required for some of the
interventions included in this review would be available in
these parts of the world. Very few trials reported baseline
ethnicity data, but the data available seem to indicate that
the large majority of studies involved Caucasian men. Given
that prostate cancer disproportionately affects other ethnic
groups (eg, black men), it should be noted that these men
are underrepresented in these trials. We were not able to
identify any trials that satisfied our pragmatic design
criteria, so these data should be considered to address the
efficacy of the interventions rather than the effectiveness of
health services. Our review objective to assess the effect of
exercise interventions on disease progression was difficult
to achieve. We were only able to undertake a synthesis of
prostate-specific antigen data measured as a secondary
outcome in underpowered trials. This finding should be
viewed with much caution. Trials that evaluate the impact
701
of exercise on dichotomous outcomes such as progression-
free survival or overall mortality would be an excellent
addition to the field.
The mechanisms by which exercise interventions
improve cancer-specific quality of life remain speculative.
Any formal analysis of such mechanisms was beyond the
scope of this review. Improvements in fatigue, lower limb
function, and exercise capacity potentially occur because of
well-established adaptations associated with exercise
training, such as improvements in cardiac output, metabolic
adaptations, and recruitment of skeletal muscle motor
units. Exercise has also been linked to improvements in
negative physiological changes associated with advanced
cancer, such as cachexia [42]. To what extent this
contributes to improved physical functioning and quality
of life is uncertain. A substantive psychological benefit
related to empowerment and self-efficacy could be a factor.
Formal mediator and moderator studies would be useful to
address this uncertainty. A number of studies included
dietary interventions as part of a lifestyle intervention.
Although not formally analysed, a minimal impact on
dietary outcomes was reported in most studies, suggesting
that the predominant effector in the intervention was the
exercise component.
The key recommendations from this review are that
treating clinicians and guideline bodies should be aware of
the level 1 evidence that exercise interventions are
efficacious in improving cancer-specific quality of life,
fatigue, and exercise capacity in men with prostate cancer.
Much of the high-quality evidence comes from trials
involving men on ADT. There is very early evidence (that
should be interpreted with caution owing to limited
number of trials) that exercise could also be useful in
improving sexual activity. Trials are ongoing to look at these
outcomes [43]. Any exercise programme should be individ-
ually tailored with consideration of the individual’s physical
capabilities and limitations [11]. The treating clinician
should play a role in directly advocating the benefits of
exercise to men with prostate cancer and leading the
multidisciplinary team in the referral process. Where
possible, men should be directed to relevant exercise
referral schemes, for example in the community. Ideally,
support for behavioural change should also be offered to
maximise adherence and should include periodic re-
evaluation of exercise prescription in terms of either
tapering or progression. Effectiveness and cost-effective-
ness data for these interventions when integrated into
health care services would be informative.
4. Conclusions
There is level 1 evidence that exercise interventions are
efficacious in improving cancer-specific quality of life,
fatigue, and exercise capacity in men with prostate cancer.
The high-quality evidence comes mainly from men with
advanced disease on ADT. Adverse events such as minor
soft-tissue injuries (sprains and strains) can be expected in a
minority of men but can also be mitigated by properly
tailored exercise prescription and progression around
702 EUROPEAN UROLOGY 69 (2016) 693–703
individual capabilities and existing comorbidities. We
found no evidence that exercise improved cardiovascular
health but we were limited to synthesising evidence for
blood pressure only. Effectiveness and cost-effectiveness
data for these interventions when integrated into health
care services would be informative.
Author contributions: Liam Bourke had full access to all the data in the
study and takes responsibility for the integrity of the data and the accuracy
of the data analysis.
Study concept and design: Bourke, Rosario.
Acquisition of data: Bourke, Amed.
Analysis and interpretation of data: Bourke, Smith, Steed, Hooper, Rosario.
Drafting of the manuscript: Bourke.
Critical revision of the manuscript for important intellectual content:
Bourke, Smith, Steed, Hooper, Carter, Catto, Albertsen, Tombal, Payne,
Rosario.
Statistical analysis: Bourke, Hooper.
Obtaining funding: None.
Administrative, technical, or material support: None.
Supervision: None.
Other: None.
Financial disclosures: Liam Bourke certifies that all conflicts of interest,
including specific financial interests and relationships and affiliations
relevant to the subject matter or materials discussed in the manuscript
(eg, employment/affiliation, grants or funding, consultancies, honoraria,
stock ownership or options, expert testimony, royalties, or patents filed,
received, or pending), are the following: Peter C. Albertsen has received
support from the Blue Cross Blue Shield Technology Assessment
Committee for reviewing items related to urology; support from Ferring
Pharmaceuticals to review data; payment for expert testimony for cases
involving prostate cancer and general urology; National Institutes of
Health grant support for studies related to prostate cancer; support for
serving as a critique panel member at Jackson Hole Seminars in 2014; and
support for participating in the 8th Engadin Prostate Cancer Symposium.
Liam Bourke and Derek J. Rosario receive funding from Cancer Research UK
and the National Institute for Health Research to evaluate the role of
exercise in men with prostate cancer. Richard Hooper receives funding
from Cancer Research UK to evaluate the role of exercise in men with
prostate cancer. Heather A. Payne has received consultancy and lecturer
fees from AstraZeneca, Astellas, Janssen, Ferring, Ipsen, Takeda, Sanofi
Aventis, and Sandoz; review preparation fees from AstraZeneca,
Janssen, Astellas, and Ipsen; payment for development of educational
presentations from AstraZeneca, Astellas, Janssen, Ferring, Ipsen,
Sandoz, and Sanofi Aventis; and travel/accommodation/meeting
expenses from Astellas, Ferring, AstraZeneca, Janssen, and Takeda.
Liz Steed has received grant support from the National Institute for
Health Research (RP-PG-0609-10181) and Health Technology Assess-
ments, and book royalties from Open University Press. Bertrand Tombal
has received grant support from Ferring, Astellas, and AstraZeneca, and
consulting fees or honorarium support from Ferring and AstraZeneca;
consultancy fees from Amgen, Sanofi, and Bayer; support for expert
testimony from Dendreon; and investigator or advisor fees from
Medivation. James Catto, Dianna Smith, and Anouska Carter have
nothing to disclose.
Funding/Support and role of the sponsor: None.
Acknowledgments: The team would like to thank Mediah Amed for her
hard work in devising the strategy for the database searches and
adapting the Medline search to other search engines. We would also like
to thank the Cochrane Urology editorial group for their assistance in
developing the protocol.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.
eururo.2015.10.047.
References
[1] Soerjomataram I, Lortet-Tieulent J, Parkin DM, et al. Global burden
of cancer in 2008: a systematic analysis of disability-adjusted life-
years in 12 world regions. Lancet 2012;380:1840–50.
[2] Heidenreich A, Bastian PJ, Bellmunt J, et al., EAU guidelines on
prostate cancer. Part II: treatment of advanced, relapsing, and
castration-resistant prostate cancer. Eur Urol 2014;65:467–79.
[3] de Bono JS, Logothetis CJ, Molina A, et al. Abiraterone and increased
survival in metastatic prostate cancer. N Engl J Med 2011;364:
1995–2005.
[4] Scher HI, Fizazi K, Saad F, et al. Increased survival with enzalutamide
in prostate cancer after chemotherapy. N Engl J Med 2012;367:
1187–97.
[5] Bourke L, Boorjian SA, Briganti A, et al. Survivorship and improving
quality of life in men with prostate cancer. Eur Urol 2015;68:374–83.
[6] Nguyen PL, Alibhai SM, Basaria S, et al. Adverse effects of androgen
deprivation therapy and strategies to mitigate them. Eur Urol
2015;67:825–36.
[7] Berruti A, Dogliotti L, Bitossi R, et al. Incidence of skeletal complica-
tions in patients with bone metastatic prostate cancer and hormone
refractory disease: predictive role of bone resorption and formation
markers evaluated at baseline. J Urol 2000;164:1248–53.
[8] Carlin BI, Andriole GL. The natural history, skeletal complications,
and management of bone metastases in patients with prostate
carcinoma. Cancer 2000;88:2989–94.
[9] Mishra SI, Scherer RW, Geigle PM, et al. Exercise interventions on
health-related quality of life for cancer survivors. Cochrane Data-
base Syst Rev 2012;8:CD007566.
[10] Mishra SI, Scherer RW, Snyder C, Geigle PM, Berlanstein DR,
Topaloglu O. Exercise interventions on health-related quality of
life for people with cancer during active treatment. Cochrane
Database Syst Rev 2012;8:CD008465.
[11] Bourke L, Homer K, Thaha M, et al. Interventions for promoting
habitual exercise in people living with and beyond cancer.
Cochrane Database Syst Rev 2013;9:CD010192.
[12] Cramp F, Byron-Daniel J. Exercise for the management of cancer-
related fatigue in adults. Cochrane Database Syst Rev 2012;11:
CD006145.
[13] Bourke L, Smith D, Steed L, et al. Exercise interventions for men with
prostate cancer. Cochrane Database Syst Rev 2014;8:CD011251.
[14] Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items
for systematic reviews and meta-analyses: the PRISMA statement.
PLoS Med 2009;6:e1000097.
[15] Higgins JPT, Green S, editors. Cochrane handbook for systematic
reviews of interventions version 5. 1.0. The Cochrane Collaboration;
2011.
[16] Baumann FT, Zopf EM, Bloch W. Clinical exercise interventions in
prostate cancer patients–a systematic review of randomized con-
trolled trials. Support Care Cancer 2012;20:221–33.
[17] Gardner JR, Livingston PM, Fraser SF. Effects of exercise on treatment-
related adverse effects for patients with prostate cancer receiving
androgen-deprivation therapy: a systematic review. J Clin Oncol
2014;32:335–46.
[18] Bourke L, Gilbert S, Hooper R, et al. Lifestyle changes for improving
disease-specific quality of life in sedentary men on long-term
androgen-deprivation therapy for advanced prostate cancer: a
randomised controlled trial. Eur Urol 2014;65:865–72.
 EUROPEAN UROLOGY 69 (2016) 693–703
[19] Cormie P, Newton RU, Spry N, Joseph D, Taaffe DR, Galvão DA. Safety
and efficacy of resistance exercise in prostate cancer patients with
bone metastases. Prostate Cancer Prostat Dis 2013;16:328–35.
[20] Cormie P, Galvão DA, Spry N, et al. Can supervised exercise prevent
treatment toxicity in patients with prostate cancer initiating
androgen-deprivation therapy: a randomised controlled trial. BJU
Int 2015;115:256–66.
[21] Dieperink KB, Johansen C, Hansen S, et al. The effects of multidisci-
plinary rehabilitation: RePCa—a randomised study among primary
prostate cancer patients. Br J Cancer 2013;109:3005–13.
[22] Galvão DA, Taaffe DR, Spry N, Joseph D, Newton RU. Combined
resistance and aerobic exercise program reverses muscle loss in
men undergoing androgen suppression therapy for prostate cancer
without bone metastases: a randomized controlled trial. J Clin
Oncol 2010;28:340–7.
[23] Galvão DA, Spry N, Denham J, et al. A multicentre year-long
randomised controlled trial of exercise training targeting physical
functioning in men with prostate cancer previously treated with
androgen suppression and radiation from TROG 03.04 RADAR. Eur
Urol 2014;65:856–64.
[24] Hebert JR, Hurley TG, Harmon BE, Heiney S, Hebert CJ, Steck SE. A
diet, physical activity, and stress reduction intervention in men
with rising prostate-specific antigen after treatment for prostate
cancer. Cancer Epidemiol 2012;36:e128–36.
[25] Jones LW, Hornsby WE, Freedland SJ, et al. Effects of nonlinear
aerobic training on erectile dysfunction and cardiovascular func-
tion following radical prostatectomy for clinically localized pros-
tate cancer. Eur Urol 2014;65:852–5.
[26] McGowan EL, North S, Courneya KS. Randomized controlled trial of a
behavior change intervention to increase physical activity and quali-
ty of life in prostate cancer survivors. Ann Behav Med 2013;46:
382–93.
[27] Monga U, Garber SL, Thornby J, et al. Exercise prevents fatigue and
improves quality of life in prostate cancer patients undergoing
radiotherapy. Arch Phys Med Rehab 2007;88:1416–22.
[28] Park S-W, Kim TN, Nam J-K, et al. Recovery of overall exercise
ability, quality of life, and continence after 12-week combined
exercise intervention in elderly patients who underwent radical
prostatectomy: a randomized controlled study. Urology 2012;80:
299–305.
[29] Segal RJ, Reid RD, Courneya KS, et al. Resistance exercise in men
receiving androgen deprivation therapy for prostate cancer. J Clin
Oncol 2003;21:1653–9.
[30] Segal RJ, Reid RD, Courneya KS, et al. Randomized controlled trial of
resistance or aerobic exercise in men receiving radiation therapy
for prostate cancer. J Clin Oncol 2009;27:344–51.
703
[31] Uth J, Hornstrup T, Schmidt JF, et al. Football training improves lean
body mass in men with prostate cancer undergoing androgen depri-
vation therapy. Scand J Med Sci Sports 2014;24(Suppl 1):105–12.
[32] Windsor PM, Nicol KF, Potter J. A randomized, controlled trial of
aerobic exercise for treatment-related fatigue in men receiving
radical external beam radiotherapy for localized prostate carcino-
ma. Cancer 2004;101:550–7.
[33] Winters-Stone KM, Dobek JC, Bennett JA, Maddalozzo GF, Ryan CW,
Beer TM. Skeletal response to resistance and impact training in
prostate cancer survivors. Med Sci Sport Exerc 2014;46:1482–8.
[34] Bourke L, Doll H, Crank H, Daley A, Rosario D, Saxton JM. Lifestyle
intervention in men with advanced prostate cancer receiving an-
drogen suppression therapy: a feasibility study. Cancer Epidemiol
Biomarkers Prev 2011;20:647–57.
[35] Buffart LM, Galvão DA, Chinapaw MJ, et al. Mediators of the resis-
tance and aerobic exercise intervention effect on physical and
general health in men undergoing androgen deprivation therapy
for prostate cancer. Cancer 2014;120:294–301.
[36] Cormie P, Newton RU, Taaffe DR, et al. Exercise maintains sexual
activity in men undergoing androgen suppression for prostate
cancer: a randomized controlled trial. Prostate Cancer Prostat
Dis 2013;16:170–5.
[37] Galvão DA, Taaffe DR, Spry N, Joseph D, Newton RU. Acute versus
chronic exposure to androgen suppression for prostate cancer:
impact on the exercise response. J Urol 2011;186:1291–7.
[38] Courneya KS, Segal RJ, Reid RD, et al. Three independent factors
predicted adherence in a randomized controlled trial of resistance
exercise training among prostate cancer survivors. J Clin Epidemiol
2004;57:571–9.
[39] Alberga AS, Segal RJ, Reid RD, et al. Age and androgen-deprivation
therapy on exercise outcomes in men with prostate cancer. Support
Care Cancer 2012;20:971–81.
[40] Kapur G, Windsor PM, McCowan C. The effect of aerobic exercise on
treatment-related acute toxicity in men receiving radical external
beam radiotherapy for localised prostate cancer. Eur J Cancer Care
2010;19:643–7.
[41] Winters-Stone KM, Dobek JC, Bennett JA, et al. Resistance training
reduces disability in prostate cancer survivors on androgen depri-
vation therapy: evidence from a randomized controlled trial. Arch
Phys Med Rehab 2015;96:7–14.
[42] Maddocks M, Murton AJ, Wilcock A. Therapeutic exercise in cancer
cachexia. Crit Rev Oncog 2012;17:285–92.
[43] Cormie P, Chambers SK, Newton RU, et al. Improving sexual
health in men with prostate cancer: randomised controlled trial
of exercise and psychosexual therapies. BMC Cancer 2014;14:
199.