A phase II randomized controlled trial of three exercise delivery methods in men with prostate cancer on androgen deprivation therapy

This randomized phase II non-inferiority trial recruited patients from two Canadian academic tertiary care centers – the Princess Margaret Cancer Centre (Toronto) and the Tom Baker Cancer Centre (Calgary). Ethics approval was obtained at both institutions and all participants provided informed consent. The trial was registered at clinicaltrials.gov (Registration #NCT02046837). The detailed trial protocol was published [10] and is summarized below.

Eligible participants were diagnosed with PC of any stage; starting or continuing on ADT for at least 6 months (or who remained biochemically castrate after stopping ADT); able to communicate in English; and living in proximity to a study centre. Each potential participant was screened with the Physical Activity Readiness questionnaire (PAR-Q+ or PARmed-X) [11] and/or received physician clearance to participate. We excluded men already engaging in 150 min of weekly moderate to vigorous physical activity (MVPA) or who had a condition interfering with their ability to participate [10].

Following the baseline assessment, participants were randomized equally (1:1:1) to PT, GROUP, and HOME. Randomization used varying block sizes and was stratified by duration of prior ADT use (< 3 months versus ≥3 months) [12] and site. The randomization sequence was created by a biostatistician using computer-generated random numbers and participants were allocated to treatments via a custom-built website, ensuring allocation concealment.

The 6-month exercise intervention consisted of three exercise delivery arms: PT, GROUP, and HOME, which have been described in detail [10]. All programs were prescribed in accord with the FITT principle: Frequency, Intensity, Time, and Type and were individualized per baseline fitness assessment results. This is detailed in our protocol paper for the phase II trial [10]. Participants were asked to complete 4–5 days per week of mixed modality exercise incorporating aerobic, resistance, and flexibility training. The target time and relative workload (target heart rate 60–70% of heart rate reserve) were consistent across interventions. Exercise intensity was monitored using the 10-point Rating of Perceived Exertion (RPE) scale. Participants maintained their intensity level between an RPE of 3 and 6 during exercise sessions. HR monitors (Polar, NY, USA) were used at 3-week intervals in each intervention arm to ensure that participants reached their target heart rate range, calibrating with the RPE scale. HOME participants were trained to use HR monitors, which were provided to them for the intervention period. If a participant’s HR is outside of his target HR range, exercise intensity will be modified to ensure training within the target HR zone. Participants who need to increase their aerobic exercise workload will first increase exercise duration (e.g. walking minutes), followed by the intensity of exercise (e.g., walking speed). If a participant is able to perform ≥12 repetitions and 3 sets of any given resistance training exercise, the resistance level used for that exercise will be increased (e.g., from a medium to a heavy band). All participants received a study manual outlining exercise techniques and safety principles as well as further details on specific resistance and flexibility exercises they were prescribed with as part of their intervention. These resistance exercises include variations of stability ball wall squat, hamstring curl, push-ups, seated row, biceps curl, triceps extension, upright row, and plank. Participant progression was individualized and monitored by a Certified Exercise Physiologist (CEP) or health coach every 3 weeks. Exercise adherence was documented on standardized forms used at both study sites.

Each program also included an education component that consisted of 12 topics focusing on common issues facing new exercisers (Additional file 1: Table S1). These topics were informally discussed with 1:1 participants during their supervised sessions for PT, and during weekly phone calls for HOME. Similarly these education topics were also discussed with GROUP during and in between exercises. Discussions typically lasted 10–20 min per topic.

Participants in the PT group completed 3 sessions/week in a dedicated gym space and were encouraged to perform 1–2 additional days/week of independent (home-based) exercise. Each session was supervised by a CEP and consisted of aerobic training (15–30 min), resistance training (with a focus on major muscle groups), and flexibility training (5–10 min of static stretching at the end of each session). All participants were provided with resistance bands to support independent exercise.

This protocol mirrored that of the PT protocol described above, and was delivered in a small group format (4–6 individuals per group), supervised by a CEP, in the same gym area but at separate times from the PT group.

The identical protocol to PT and GROUP was implemented in HOME. Participants received resistance bands, stability ball, exercise mat, HR monitor, and smartphone with a 6-month paid plan to connect with a health coach and access specific health software (Connected Wellness Platform, NexJ Systems, Inc.). Weekly health coach (by smartphone) communications via text and voice reviewed exercise sessions, guidance, and helped with smartphone applications. Additional details are described in the protocol [10].

Outcome assessments were completed at baseline, 3, and 6 (end of intervention) months. Additional assessments at 9 and 12 months were completed post-intervention, and will be reported separately. Blinded outcome assessments were conducted by a CEP.

We assessed recruitment rate, retention rate, adherence, outcome capture, and satisfaction/barriers. Definitions and measurement details are as follows:

Safety procedures are described elsewhere [10]. Adverse events were documented using the National Cancer Institute common terminology criteria for adverse events v4.0 [30].

This study investigated the feasibility of completing a companion economic evaluation (as part of a subsequent phase III trial). Although there was no formal cost-utility analysis in this trial, relevant outcomes and health status (utilities) were collected at each time point using the European QOL 5-Dimension 3-Level measure (EQ-5D-3 L) [31]. Costs were also captured with a patient-reported diary [32].

Following standard guidelines for a phase II RCT [33, 34], we determined that a sample size of 30 patients per arm (90 patients in total) would provide sufficiently precise estimates of parameters related to important feasibility information as well as the primary clinical outcomes that would be crucial to planning a phase III study. Assuming a drop-out rate of 10%, our goal was to recruit 100 patients.

Statistical analysis focused on feasibility outcomes, namely estimation of recruitment and retention rates and adherence. In addition, reasons for not participating in the trial were documented and tallied. Descriptive statistics were used for each of these feasibility measures.

QOL, fatigue, and fitness outcomes were analyzed using Bayesian linear mixed effects model with subject-specific random intercepts, fixed effects for time and group, and group-by-time interactions. Models were fitted through rjags in R version 3.4.0. Both the GROUP and HOME arms were compared to PT. Mean differences in baseline to 6-month change and their 95% credible intervals (CrIs) were calculated, along with the posterior probability of inferiority that the mean difference lay outside the pre-specified non-inferiority margin for the specific outcome. Diffuse normal priors were used for all regression coefficients, and uniform prior distributions with a large upper bound were used for the standard deviations of the residual error and between-subject random effects. After a burn-in of 5000 iterations, 10,000 further samples were collected from each of three parallel chains and used for inference. Convergence was assessed with the Gelman diagnostic.

A priori, we defined the following criteria to move forward to a phase III trial: recruitment rate of at least 25%, adherence and retention rates of 70%, moderate or greater participant satisfaction, and at least 80% data capture of clinical outcomes.

Sixty-five men were enrolled (45 in Toronto, 20 in Calgary) between December 19, 2013 and October 31, 2015. Follow-up occurred between June 2013 and October 2016. Six men withdrew before randomization. Baseline characteristics for 59 subjects (mean age 70) are presented in Table 1; groups were similar for sociodemographic, clinical, and outcome measures.