The Phys-Can randomized controlled trial
This study used data from the Physical Training and Cancer (Phys-Can) intervention study [3, 9]. The primary aim of the Phys-Can RCT was to compare the effects of low-to-moderate (LMI) vs high-intensity (HI) exercise, with or without additional behavior change support (goal-setting, self-monitoring, action planning, review of goal-setting and problem solving), on cancer-related fatigue in patients undergoing cancer treatment. The supervised group-based resistance training was performed at public gyms twice per week over a 6-month period. The LMI groups performed 3 sets, alternating between 12 repetitions (once a week) and 20 repetitions (once a week) with a corresponding training load of 50% of 6 and 10 RM (repetitions maximum), respectively. The HI groups performed 3 sets alternating between 6 RM (once a week) 10 RM (once a week) until failure in the third set. Endurance training was home-based and the LMI groups performed 150 min/week at 40-50% of individual HRR (heart rate reserve), mainly walking or cycling. The HI groups performed 5x2-minute intervals of running or cycling at 80–90% of HRR twice a week, with progression from 5 to 10 intervals over the 6-months training period.
Participants in the Phys-Can RCT were recruited from three Swedish university hospitals between 2015 and 2018. Eligible participants were ≥18 years, literate in Swedish and recently diagnosed with curable breast (BC; women only), prostate (PC) or colorectal cancer (CRC), and scheduled to begin (neo-)adjuvant cancer treatment. Exclusion criteria were: stage IIIb-IV BC, inability to perform basic activities of daily living, cognitive disorders, disabling conditions that might contraindicate HI exercise, treatment for an additional ongoing malignant disease or orthopedic conditions, BMI<18.5 kg/m2 or pregnancy. A research nurse/assistant provided oral and written information to eligible participants prior to start of the treatment.
Participants and data collection
Background data for all patients eligible for the RCT (n=2051) were collected at baseline by questionnaire (age, sex) and medical records (cancer diagnosis and treatment modality: chemotherapy or not). Additional data were collected by an extended questionnaire, and by using a physical activity monitor and fitness testing at baseline for RCT participants (n=577) and a sub-group of decliners who consented to respond (n=436). The extended questionnaire included sociodemographic variables (education, living situation) and a number of variables related to health and exercise as described below. Among the 577 RCT participants, we identified one group who completed the intervention; ‘completers’ (n=410) and one group who withdrew before the end of the intervention; ‘drop-outs’ (n=167) (see Figure 1).
Treatment modality, chemotherapy treatment or not, was assessed by questionnaire for decliners, and by medical records for trial participants.
Comorbidity was assessed with a standard list of conditions including heart disease, respiratory disease, arthritis, diabetes, diseases of the digestive tract, migraine, psychiatric problems, and diseases of the central nervous system.
Anxiety and depression were assessed with the 14-item Hospital Anxiety and Depression Scale (HADS) [42]. It includes two subscales with 7 items on anxiety and 7 items on depression. Responses are given on a 0-3 scale, resulting in a sum score of 0-21 for each subscale, with higher scores indicating more anxiety or depression.
Cancer-related fatigue was assessed with the Multidimensional Fatigue Inventory (MFI) [34]. The 20-item MFI includes five subscales: general fatigue, physical fatigue, reduced activity, reduced motivation, and mental fatigue. Responses are given on a 1-5 scale, resulting in a sum score of 4-20 for each subscale, with higher scores indicating more fatigue.
Exercise stage at the time of invitation to the study was assessed with the Exercise Stage Assessment Instrument (ESAI) [27], based on the Transtheoretical Model Stages of Change [31]. Participants responded to two questions: ‘Which one of the statements below is most accurate to describe your current level of endurance training?’ and ‘Which one of the statements below is most accurate to describe your current level of resistance training?’. Responses are given on a 5-point scale representing different stages of change; 1. ‘Not very physically active and I do not intend to become more physically active during the next 6 months’ (pre-contemplation stage); 2. ‘Not very physically active but I have considered increasing my activity during the next 6 months’ (contemplation stage) 3; ‘Not very physically active but determined to increase my activity during the next 6 months’ (preparation stage); 4. ‘Physically active, but only during the previous 6 months’ (action stage); and 5. ‘Physically active and I have been so for more than 6 months’ (maintenance stage). For analysis, the answers were dichotomized into ‘not physically active’ and ‘physically active’. The ‘not physically active’ category included exercise stages 1, 2 and 3 (‘pre-contemplation’, ‘contemplation’ and ‘preparation’), while the ‘physically active’ category included those in exercise stages 4 and 5 (‘action’ and ‘maintenance’).
Outcome expectations were assessed with three study-specific questions: ‘How certain are you that exercise…’ 1) ‘is healthy for you’, 2) ‘can reduce symptoms from cancer treatment’, 3) ‘can reduce the risk of relapse?’. Responses for each question are given on a 0-10 scale with 0 = ‘not at all certain’ to 10 = ‘extremely certain’. These items have been adapted from a previous exercise intervention study in people with rheumatoid arthritis [28].
Exercise self-efficacy was assessed with the Exercise Barriers Self-efficacy Scale (EBSS) [32], which includes 9 barrier items encompassing lack of discipline, time, interest, inspiration and support for exercise, as well as fatigue, nausea, bad weather and not prioritizing exercise. Responses to each barrier are given on a 0-10 scale with 0 = ‘not at all certain’ to 10 = ‘extremely certain’, resulting in a sum score of 0-90, with higher scores indicating higher exercise self-efficacy.
Kinesiophobia (fear of movement) was assessed using a 14-item Swedish version of the Tampa Scale for Kinesiophobia (TSK-SV-14), which is a shortened and translated version of the original TSK [22]. The questionnaire asks respondents to rate their level of agreement to 14 statements about their attitudes to physical activity. Each statement (e.g., ‘I cannot do the same things as others because there is too much risk of being injured’) is answered with a 4-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree). Four items (3, 6, 9 and 13) are reverse-worded statements. Total scores can range from 14 to 56, with higher scores indicating greater fear of movement. The TSK-SV-14 is a shortened version of the previously validated TSK-SV-17 [20].
Cardiorespiratory fitness was measured as maximal oxygen uptake (VO2max [mL/kg/min]) and assessed using a modified Balke-protocol as previously described in detail [4]. Briefly, participants started walking or running at 4 km/h with an incline of 2% on a treadmill. The inclination was then increased by 1% every minute until reaching 12%, from which the speed increased 0.5 km/h per minute until exhaustion. Tests were accepted if two out of three criteria were fulfilled: (a) tester judged the test as maximal, (b) Borg RPE rating ≥ 17 [5] and (c) respiratory exchange ratio (RER) ≥ 1.1 [4].
Moderate-to-vigorous intensity physical activity was assessed using SenseWear Armband mini (BodyMedia Inc, Pittsburgh, PA, USA) as previously described [9]. Mean time (minutes) in moderate-to-vigorous intensity physical activity (MPVA) per 24 hours was considered that of at least 3 METs according to SenseWear algorithms [11].
Ethical approval
The Phys-Can RCT was approved by the Regional Ethical Review Board in Uppsala, Sweden (Dnr 2014/249) and registered in ClinicalTrials.gov (ClinicalTrials.gov.Id.NCT02473003). The study was conducted in accordance with the Declaration of Helsinki and written informed consent was obtained from the participants before inclusion.
Statistical analyses
Descriptive characteristics are presented as mean and standard deviation (SD) for continuous variables and proportions as number (n) and percentage (%) for categorical variables. Differences in means and proportions were analyzed by independent-samples t-tests and Chi2-tests, respectively. Statistical analyses were performed with SPSS statistical software version 26 (SPSS Inc., Chicago, IL, USA).