Advances in Adherence Reporting of Resistance Training in a Clinical Trial during Adjuvant Chemotherapy for Colon Cancer

Abstract

Purpose:

Detailed reporting of individually tailored exercise prescriptions (ExR_x) used in clinical trials is essential to describe feasibility, tolerability, and efficacy of the intervention and to inform translation to clinical care. This paper outlines the methodology used to develop a resistance training (RT) ExR_x for people with colon cancer receiving chemotherapy and reports adherence to the randomized controlled trial testing the impact of RT on relative dose intensity of chemotherapy and patient-reported toxicities.

Methods:

Participants randomized to the exercise arm (n=90) were included. To promote muscle hypertrophy, the ExR_x was twice-weekly, moderate to heavy loads (65-85% 1-RM), high sets (3–5), and intermediate repetitions (6–10) of five large multi-joint movements with adjustable dumbbells. Attendance (achieved frequency) and adherence (achieved volume) were calculated. Group-based trajectory modeling was used to identify clusters of individuals with similar adherence patterns and compared baseline characteristics across adherence groups.

Results:

The median attendance was 69.1%. Adherence was 60.6%, but higher for those receiving 3 versus 6 months of chemotherapy (80.4 vs. 47.4%; p<0.05). Participants engaged in a median of 1.4 days of RT each week, lifting 62% of the 1-RM load, for 3.0 sets and 7.5 repetitions per set. Three distinct adherence groups were identified: 13% “non-starter”, 37% “tapered off”, and 50% “consistent exercisers”. Females were more likely to be in the “non-starter” and “tapered off” groups.

Conclusions:

This paper outlines suggested methods for reporting ExR_x of RT in oncology clinical trials and provides insight into the tolerance of ExR_x of RT during chemotherapy treatment for colon cancer. These findings aim to foster constructive dialogue, and offer a premise for designing future research to elucidate the benefits of exercise during chemotherapy.

INTRODUCTION

Individually tailored exercise prescriptions (ExR_x) maximize the health benefits of exercise and reduce within-person variation in response to structured exercise training. Health and fitness professionals (HFP) utilize the FITT-VP principle to tailor the ExR_x by varying the frequency (how often), intensity (how hard), time (how long), type (what kind), volume (amount), and progression (advancement) of exercise.(1) Applying the FITT-VP principle is essential for developing ExR_x for people with chronic diseases, as their physiological responses to exercise may differ from healthy populations. Detailed reporting of the ExR_x used in clinical trials is necessary to describe feasibility, tolerability, and efficacy of the intervention and inform the translation of research into clinical care. One such trial the FOcus on Reducing dose-limiting toxicities in Colon cancer with resistance Exercise (FORCE) Trial (NCT03291951). This trial aimed to examine the effect of resistance training on preventing relative dose intensity of chemotherapy and dose-limiting toxicities in people with colon cancer undergoing chemotherapy.(2)

People living with and beyond cancer may derive substantial health benefits from tailored ExR_x. Exercise can reduce the frequency of some acute and long-term adverse effects of anticancer treatment and improve overall heath.(3–5) Developing tailored ExR_x for people receiving anticancer treatment is complex. Many anticancer treatments are delivered cyclically, with an interval of time between cycles to allow for physical recovery. Exercise tolerance may vary day-to-day within a treatment cycle due to alterations in autonomic function, cardiovascular function (i.e., volume depletion), or cytokine-related fatigue (6–8). Improving the design and delivery of RT interventions for cancer patients is critical in the exercise oncology field because lower lean mass and sarcopenia predict poor tolerance of anticancer treatments.(9–13) Patients with low lean mass may be less able to receive the full planned dose of anticancer therapy, which is associated with lower survival rates in various cancer types.(14) Therefore, there is interest in interventions aimed at maintaining or increasing muscle mass using best practice, evidence-informed resistance training methods. (15)

The traditional ExR_x often increases exercise volume linearly to accommodate exercise-induced improvements in physiology and promote continued physiological adaption. However, symptoms and side effects from anticancer treatment accumulate with more cycles of treatment(16). A unique challenge in oncology is that the accumulation of side effects from anticancer therapy may impede the traditional linear progression of ExR_x. Nonlinear ExR_x to accommodate cyclic exercise tolerance and accumulating symptoms in the oncology setting have been proposed and tested.(8, 17–19) However, developing a feasible, safe, and efficacious ExR_x during anticancer treatment is hampered by incomplete reporting of ExR_x FITT-VP characteristics in most exercise oncology interventions,(20–25) particularly for resistance training (RT) interventions.(26, 27) (9–14) Therefore, accurately describing ExR_x tolerated during anticancer treatment is of great interest to HFPs and the topic of recent notable publications aiming to develop novel ways of reporting the ExR_x that is prescribed and achieved. (17, 28)

This paper has two aims: 1) to outline the methodology used to develop the resistance training ExR_x for the FORCE Trial; (2) and 2) to report on unique insights gained by systematically collecting adherence to an RT ExR_x using FITT-VP data elements and using this to identify adherence patterns in an exercise oncology clinical trial.

METHODS

Briefly, the FORCE Trial was a multi-institutional, randomized trial of home-based RT vs. usual care (UC) in patients recently diagnosed with stage II and III colon cancer who planned to begin chemotherapy.(29) The institutional review board at all recruitment sites approved the study protocol. Participants provided written informed consent. The study outcomes were relative dose intensity of chemotherapy and patient-reported toxicities (co-primary), body composition (secondary), and inflammation (secondary). Eligible participants included men and women, age 18 years or older, with newly diagnosed stage II or III colon cancer who recently completed curative-intent surgical resection. Participants were recruited before and within the first six weeks of chemotherapy from three sites, including Kaiser Permanente Northern California (KPNC), Dana-Farber Cancer Institute (DFCI), and Penn State Cancer Institute (PSCI). Patients were excluded if they had a concurrent actively treated other cancer (except non-melanoma skin cancer, in situ cervical cancer, or localized prostate cancer treated with surveillance only), untreated hypertension, poor kidney function (creatinine <2 mg/dL), metastatic cancer, were currently engaging in a resistance training regimen; or were enrolled in other behavioral health clinical trials. Following the baseline evaluation, participants were randomized in a 1:1 ratio to RT or UC.

Intervention Design Overview

In this randomized trial, the desired health benefit of exercise was to promote muscle hypertrophy. The tailored resistance training ExR_x was designed to maximize the likelihood of achieving this health benefit, and combined with protein supplementation.(30) The ExR_x was delivered concurrently for the duration of chemotherapy, which for most patients was either 3- or 6-months, as these represent the current standards of care for postoperative colon cancer.(31, 32)

Protein Supplementation

Participants in the RT group were asked to consume 40 g of whey protein in a powdered supplement. They were given 20 g packets (unflavored or vanilla flavored) of BiPro whey protein (Agropur, Eden Praire, MN).(33) Participants were instructed to consume the protein supplement approximately 30 minutes into their meal to prevent the extra protein from interfering with typical whole food intake. Participants who reported lactose intolerance were provided a pea protein supplement (Garden of Life Raw Organic Protein, unflavored or vanilla flavored). Participants were provided with paper logs to record protein supplementation consumption. Logs were physically submitted to the HFPduring in-person visits and reported verbally by participants during telephone conversations. The HPF reviewed if the individual was meeting targets, discussed barriers if targets were not being met using guidance from the staff manual, and worked with the participant to set goals and strategies to improve adherence. All paper logs not already collected at in-person visits were submitted at end of study to the study staff. Adherence was operationalized as number of days per week consumed and amount of protein consumed on days when protein supplement was consumed. Usual protein intake was also measures using 2014 Block Food Frequency Questionnaire (FFQ)(34).

Qualification and Training of the Health & Fitness Professionals

The HFP who implemented the ExR_x had national certification for exercise training, along with prior oncology expertise. The HFP were trained to implement the ExR_x by team leaders (K.L.C. and K.H.S) during a centralized training session held at the start of the trial. HFP hired after the initial centralized training received instruction by the exercise supervisor at each site. Following the initial certification process, all HFP were recertified every two years. The HFP also participated in cross-site conference calls held twice monthly with the exercise team laders to problem-solve challenges that participants had with the ExR_x during chemotherapy. Study staff documented these issues in a frequently asked questions list to standardize how HFP managed challenges across the three study sites.

Intervention Orientation and Baseline Strength Testing

Participants in the RT group completed a 30-minute phone interview with the HFP. This call’s objective was for the HFP to introduce themselves, develop a rapport with the participant, explain the study timeline, outline the goals of the ExR_x, assess exercise self-efficacy, obtain exercise and injury history, and establish expectations for study participation. During this call, the HFP scheduled the first in-person training visit.

The RT group received adjustable dumbbells (Power Blocks, Owatonna, MN) and an aerobic step to implement the home-based ExR_x by mail or at first in-person training visit. Baseline handgrip strength was used to select the appropriate range of adjustable dumbbells needed to complete the ExR_x. For men and women with a handgrip strength <40 kg and <30 kg, respectively, dumbbells that adjusted up to 33 pounds in each hand were provided; for men and women with a handgrip strength ≥40 and ≥30 kg, respectively, or with a self-reported history of engaging in resistance exercise, dumbbells that adjusted up to 60 pounds in each hand were provided; additional dumbbell weights up to 90 pounds were provided, as needed.

At the first in-person training visit, the HFP instructed the participant on using the adjustable dumbbells and how to lay on the aerobic step as a bench. The HFP instructed the participant on how to complete the exercises in the ExR_x. The exercises targeted large muscle groups, including the dumbbell bench press, squat, bent-over row, deadlift, and lunge (see Appendix, Supplemental Digital Content 1). The instruction methods included direct instruction by the HFP and kinesthetic learning by the participant with the professional’s feedback. After acclimating to the selected exercises using lightweight dumbbells (e.g., 2.5 kg), participants completed a submaximal (8-10-repetition maximum) strength test for each exercise to determine the estimated 1-repetition maximum (RM) and initial weight to be lifted according to the specified starting load.(1, 35, 36) If an individual could not complete the exercise with correct form, an alternate exercise was substituted. The alternate exercises included a different position that would allow the participant to use the dumbbells (i.e., reverse lunge or split stance lunge) or bodyweight exercises (i.e., wall sit, wall pushups, wall scapular retraction, or bridge). Developing the prescription for an alternate exercise was based on the HFP’s discretion, namely prescribing a weight, or with body weight only exercises, then number of sets or repetitions, that the participant could complete with good form. At the conclusion of the first in-person visit, participants were alerted about the potential of experiencing delayed onset muscle soreness and instructed on how to complete exercise and protein logs (see Supplemental Digital Content 2, 3, and 4, ExRX for 2- and 3-week cycles and Protein supplemention log).

Development of the Individualized Exercise Prescription.

Participants in the RT group initiated a progressive RT ExR_x designed to promote muscle hypertrophy (Table 1). The progression of resistance exercise was standardized to provide a similar hypertrophic stimulus by accommodating the two most common chemotherapy regime intervals (e.g., one chemotherapy cycle delivered every two- or three weeks), such that progression of intensity could be progressed by the HFP at the in-person visits that were associated with the chemotherapy infusion visits.

Table 1.

Planned exercise prescription and protein recommendations for the two most common chemotherapy regimens for colon cancer

FOLFOX or 5-FU/LV (treatment cycles administered at 2-week intervals over 3 or 6 months)
Study Week	1	2	3	4	5	6	7	8	9	10	11	12	13-24b
Chemotherapy Cycle	1	—	2	—	3	—	4	—	5	—	6	—	—
Days/Week of Resistance Exercise	2	2	2	2	2	2	2	2	2	2	2	2	2
% of 1-RM Loadinga	65	65	65	65	75	75	75	75	80	80	85	85	85
Sets of Each Exercise	3	3	3	3	4	4	4	4	5	5	5	5	5
Repetitions Within Set	6	6	8	8	8	8	10	10	10	10	10	10	10
Resistance Training Exercises	5	5	5	5	5	5	5	5	5	5	5	5	5
Days/Week Protein	7	7	7	7	7	7	7	7	7	7	7	7	7
Protein Dose/Day (g)	40	40	40	40	40	40	40	40	40	40	40	40	40
CAPOX or Capecitabine monotherapy (treatment cycles administered at 3-week intervals over 3 or 6 months)
Study Week	1	2	3	4	5	6	7	8	9	10	11	12	13-24b
Chemotherapy Cycle	1	—	—	2	—	—	3	—	—	4	—	—	—
Days/Week of Resistance Exercise	2	2	2	2	2	2	2	2	2	2	2	2	2
% of 1-RM Loadinga	65	65	70	70	70	70	75	75	75	85	85	85	85
Sets of Each Exercise	3	3	4	4	4	4	4	4	5	5	5	5	5
Repetitions Within Set	6	6	8	8	8	8	10	10	10	10	10	10	10
Resistance Training Exercises	5	5	5	5	5	5	5	5	5	5	5	5	5
Days/Week Protein	7	7	7	7	7	7	7	7	7	7	7	7	7
Protein Dose/Day (g)	40	40	40	40	40	40	40	40	40	40	40	40	40

Abbreviations: FOLFOX: 5-fluorouracil, leucovorin, and oxaliplatin; CAPOX: capecitabine and oxaliplatin; 5-FU/LV: 5-fluorouracil and leucovorin.

^aLoad was individualized based on the results of a submaximal strength test used to predict the one-repetition maximum (% of 1-RM);

^bStudy participants who received 6 months of chemotherapy continued resistance exercise and protein consumption at the prescribed values through the end of chemotherapy

Participants received up to six in-person exercise counseling sessions with theHFP . The in-person visits occurred during a chemotherapy infusion visit or another medical appointment to reduce participant burden. The purpose of the in-person visit with the HFP was to review the ExR_x and progress the % 1-RM load by allowing the HFP to observe the participant and ensure proper form with the increased load. If necessary, adjustments we used to ensure that good form could be maintained. Due to the COVID-19 pandemic, starting in March 2020, participants already on trial and new participants were switched to tele-coaching.

In the weeks that participants did not receive chemotherapy, the HFP contacted the participant once per week via telephone to provide ongoing behavioral support and troubleshoot any challenges adhering to the ExR_x. The HFP reviewed the exercise and protein logs and provided feedback to promote the accurate recording of the completed ExR_x. The HFP documented all participant communications.

Specific to the FITT-VP principle of resistance exercise the ExR_x was:

• frequency, twice per week;

• intensity, initiated at 65% of the 1-RM (as quantified from submaximal testing for each of the five exercises) and progressed to 85% of the 1-RM;

• time was not specified;

• type is the five above-described exercises, with

• volume (e.g., repetitions initiated at six and increased to 10, and sets are initiated at three and expanded to five);

• pattern includes a rest interval of one- to two minutes between each exercise set; and

• progression gradually increased (Table 1).

Protocol for Adjusting the Exercise Prescription

This ExR_x provided a framework to guide clinical decision-making by theHFP . However, the HFP were permitted to deviate from this ExR_x in response to poor participant adherence, musculoskeletal injury, chemotherapy toxicity (e.g., neuropathy), or cases where a participant increased their dumbbell weight, and no concerns were raised following sound clinical judgment. HFP used professional judgement that was clinically sensible and consistent with promoting hypertrophy, including based on collaborative discussions with the study intervention team where any challenges were discussed. Participants who were unable to meet the recommended ExR_x targets were reminded that some, as opposed to no, resistance exercise was considered beneficial.(5) Adverse event in the RT group were captured by self-report by an individual to the HFP , or by review of the electronic medical record by the study coordinator. Each event was categorized by the principal investigator at each site as either an adverse event (AE) or serious adverse event (SAE) and as related or unrelated to the intervention.

Exercise adherence reporting

Using approaches advocated by Fairman et al (28), specific to resistance training, attendance was operationalized as number of sessions completed per week as percent of planned sessions per week. Adherence was operationalized using the completed exercise versus intended exercise volume (a combination of intended load, sets, and repetitions for each exercise session). FITT-VP was operationalized as frequency (mean number of sessions completed per week), intensity (mean overall percent of estimated 1-RM load, capped at 100%), type (mean number of sets completed for an exercise session if attempted and mean number of repetitions completed for sets completed for an exercise session if attempted). If alternate exercises that were body weight only were prescribed, the frequency, repetitions and sets were used to determine adherence.

Statistical Analysis

Baseline participant characteristics are reported as means with standard deviations or medians with interquartile (25-75%) ranges for continuous variables and counts with percentages for categorical variables. Baseline participant characteristics and exercise adherence were compared by planned chemotherapy length using the Student’s t-test (mean values, continuous), non-parametric Kruskal Wallis (median values, continuous), Chi square (categorical) or Fisher’s exact test (categorical if cell sizes less than 5). Standard deviations (SD) or inter-quartile ranges (IQR) are reported. Group-based trajectory modeling was used to empirically identify clusters of individuals with similar adherence patterns over the course of the study. Using weekly adherence reports, models were implemented using the SAS procedure PROC TRAJ,(37) under the assumptions that longitudinal measures of adherence can be modeled as polynomial functions of time (weeks since randomization) and that individuals are clustered into distinct unobserved groups defined by their adherence trajectories. A censored normal distribution that allowed for a minimum (0) and maximum (100) outcome value was used to model adherence. Models with two and three trajectory groups were considered, as well as linear, quadratic, and cubic polynomial trajectories. The final model was determined by considering the Bayesian Information Criterion (a measure of model goodness-of-fit) and scientific judgement. Each participant had their probabilities of membership for each group estimated; we then assigned the participant to the group with the highest probability. Baseline characteristics were compared across adherence trajectory groups. All analyses were conducted using SAS version 9.4 (SAS Institute Inc., Cary, NC, U.S.A.).

RESULTS

Baseline Characteristics

The 90 study participants randomized to the RT group had a mean (SD) age of 56.3 (13) years, a body mass index of 27.0 (5.3) kg/m² and self-reported a median (IQR) of leisure-time physical activity at enrollment of 8.6 (0.0-26.7) MET-hr/wk (Table 2). There was a comparable distribution of male (51%) and female (49%) participants, and the majority of participants were enrolled after the initiation of chemotherapy (57%). The most common planned chemotherapy regimens were six-months of FOLFOX (48%), and three-months of CAPOX (27%).

Table 2.

Baseline characteristics, overall and stratified by planned chemotherapy length

Characteristics	Overall (n = 90)	Planned Chemotherapy Length		P
		3-months (n =29)	6-months (n = 61)
	Mean±SD or n (%)	Mean±SD or n (%)	Mean±SD or n (%)
Age, y	56.3±13.0	58.0±11.9	55.6±13.4	0.412
Sex, n (%)				0.326
Male	46 (51)	17(59)	29(48)
Female	44 (49)	12 (41)	32 (52)
Education				0.510
High school	8 (9)	3 (10)	5 (8)
Some college/technical college	29 (32)	11 (38)	18 (30)
College Graduate	28 (31)	6 (21)	22 (36)
Post Graduate Degree	24 (27)	9 (31)	15 (25)
Unknown	1(1)	0(0)	1 (2)
Race				0.880
White	58 (64)	19 (6)6	39 (64)
African American	7 (8)	1 (3)	6 (10)
Asian	15 (17)	5 (17)	10 (16)
American Indian/Native Alaskan	1 (1)	0 (0)	1 (2)
Native Hawaiian/Pacific Islander	2 (2)	1 (3)	1 (2)
More than one	3 (3)	1 (3)	2 (3)
Other	4 (4)	2 (7)	2 (3)
Ethnicity				0.523
Hispanic or Latino	6 (7)	2 (7)	4 (7)
Not Hispanic or Latino	76 (84)	23 (79)	53 (87)
Unknown	8 (9)	4 (14)	4 (7)
Employment				0.186
Employed, full-time	51 (57)	13 (45)	38 (62)
Employed, part-time	13 (14)	5 (17)	8 (13)
Unemployed	4 (4)	3 (10)	1 (2)
Retired	22 (24)	8 (28)	14 (23)
Body Mass Index, kg/m2	27.0±5.3	26.9±4.3	27.0±5.7	0.917
Physical Activity, MET-hr/wka	8.6 (0-26.7)	6.4 (0-27.0)	8.9 (0-24.1)	0.890
Chemotherapy Regimen, n (%)				<0.001
FOLFOX 6m	43 (48)	0 (0)	43 (70)
FOLFOX 3m	5 (6)	5 (17)	0 (0)
CAPOX 3m	24 (27)	24 (83)	0 (0)
CAPOX 6m	8 (9)	0 (0)	8 (13)
CAP-only 6m	9 (10)	0 (0)	9 (15)
5FU-only 6m	1 (1)	0 (0)	1 (2)
AJCC stage				0.452
Stage 2	12 (13)	5 (17)	7 (11)
Stage 3	78 (87)	24 (83)	54 (89)
Randomized before initiation of chemotherapy				0.514
Yes	39 (43)	14 (48)	25 (41)
No	61(57)	15 (52)	36 (59)
Sarcopenic*
Yes	22 (27)	5 (20)	17 (30)	0.355
No	60 (73)	20 (80)	40 (70)

Abbreviations: FOLFOX: 5-fluorouracil, leucovorin, and oxaliplatin; CAPOX: capecitabine and oxaliplatin; 5-FU/LV: 5-fluorouracil and leucovorin. Values are mean ± standard deviation for continuous variables and n (%) for categorical variables unless otherwise noted.

^aMedian and interquartile (25-75%) range.

^*Sarcopenia = appendicular lean mass <7.25 for men or <5.5 for women on DXA scan.

Exercise and Protein Adherence

The median attendance was 69.1% (36.1-87.5) overall, with higher attendence for participants receiving 3-month (77.8%; 65.0-88.9) versus 6-month regimens (61.1%; 31.0-78.6; p=0.04) (Table 3). The median adherence was 60.6% (36.1-87.5), with higher adherence for participants receiving 3-month (80.4%; 58.7-91.7) versus 6-month regimens (47.4%; 22.4-75.8; p<0.05). For frequency, participants completed a median of 1.4 (0.6-1.7) sessions of RT per week, with no difference by length of planned chemotherapy regimen (p=0.052). For intensity, participants lifted median load of 62% of estimated 1-RM (53-70), as measured at baseline, with no difference by length of planned chemotherapy regimen (p=0.94). For volume, the median number of sets completed was 3.0 (2.6-3.6) with no difference by length of planned chemotherapy regimen (p=0.95); and the median number of repetitions were 7.5 (6.4-8.7), with no difference by length of planned chemotherapy (p=0.85), and average 3.4 (1.7) of the 5 exercises prescribed per session. The pattern of training volume completed compared to prescribed is presented in Figure 1 A–B. With each of the five exercises, there was no clear pattern that one execise required more modification than another (Table 3). After adjusting for intended chemotherapy duration, the total weight lifted by participants in the resistance training group significantly increased by 0.7 kg per week [95% CI: 0.4, 1.0); P<0.0001], demonstrating a statistically significant increase in strength. The majority of participants completed their exercise sessions at home, with only a small number completing sessions at public gyms or other settings. In sensitivity analyses, there was no difference in attendance or adherence metrics by those who received any in-person sessions and those who did not (data not shown). Participants consumed protein supplementation a median of 5.4 (3.0-6.4) days per week at a dose of 20.0 (9.4-30.4) g. Median usual protein intake from usual diet was 63.8 grams per day at baseline and was unchanged over the course of the intervention (p=0.45). There was no difference in protein adherence by length of planned chemotherapy regimen (p=0.62 and p=0.81, respectively). No serious adverse events due to the intervention were reported, and there was one non-serious adverse event that was deemed related to the intervention, namely a sprained wrist.

Table 3.

Overall exercise and protein adherence stratified by planned chemotherapy length

Characteristics	Overall	Three Months	Six Months	P-Value
	(N=90)	(N=29)	(N=61)
Attendance (%)a
Median (IQR)	69.1 (36.1-87.5)	77.8 (65.0-88.9)	61.1 (31.0-78.6)	0.04
Missing, n (%)	4 (4.4)	1 (3.4)	3 (4.9)
Adherence (%)b
Median (IQR)	60.6 (31.0- 87.1)	80.4 (58.7-91.7)	47.4 (22.4-75.8)	<0.05
Missing, n (%)	1 (1.1)	0 (0.0)	1 (1.6)
Frequency: Sessions/week
Median (IQR)	1.4 (0.6-1.7)	1.5 (1.3-1.8)	1.2 (0.6-1.5)	0.052
Missing, n (%)	0 (0.0)	0 (0.0)	0 (0.0)
Intensity: % 1-RM
Median (IQR)	61.5 (53.2-69.9)	60.5 (55.5-69.6)	61.8 (51.5-70.6)	0.94
Missing, n (%)	10 (11.1)	4 (13.8)	6 (9.8)
Type: Sets (number per session)
Median (IQR)	3.0 (2.6-3.6)	2.8 (2.7-3.4)	3.0 (2.3-3.7)	0.95
Missing, n (%)	10 (11.1)	4 (13.8)	6 (9.8)
Type: Repetitions (number per sets completed during a session)
Median (IQR)	7.5 (6.4-8.7)	7.3 (6.3-8.5)	7.5 (6.5-8.7)	0.85
Missing, n (%)	10 (11.1)	4 (13.8)	6 (9.8)
By exercise type (%)c (Median; Min-Max)
Squat	65 (0-113)	NA	NA	0.95
Chest Press	61 (0-126)	NA	NA
Deadlift	64 (0-356)	NA	NA
Row	64 (0-233)	NA	NA
Lunge	61 (0-328)	NA	NA
Protein Supplement: Number of days consumed per week (d)
Median (IQR)	5.4 (3.0-6.4)	4.8 (3.5-6.0)	5.4 (3.0-6.4)	0.62
Missing, n (%)	13 (14.4)	5 (17.2)	8 (13.1)
Protein Supplement: Intake per day on days protein consumed (g)
Median (IQR)	20.0 (9.4-30.4)	20. 3(12.0 - 28.0))	19.7 (9.1-30.9)	0.81
Missing, n (%)	13 (14.4)	5 (17.2)	8 (13.1)
Protein Intake: From usual diet (g)
Median (IQR)	56.3 (42.7-79.9)	63.5 (44.7-77.2)	54.7 (40.3-81.3)	0.45
Missing, n (%)	13 (14.4)	5 (17.2)	8 (13.1)

Abbreviations: IQR, interquartile range; RM, repetition maximum

Legend:

^aNumber of sessions completed per week as percent of planned 2x/week;

^bOverall amount of planned volume achieved; load was individualized based on the results of a submaximal strength test that was used to predict the one-repetition maximum (% of 1-RM); capped at 100%.

^cOverall percent of intended volume completed for each of the 5 exercises.

Figure 1.

Adherence to prescribed and completed exercise volume: A: Overall for duration of intervention; B: For participants receiving CAPOX (3-month regime); C: For participants receiving FOLFOX (6-month regime).

Pattern of Exercise Adherence

There were three distinct patterns of exercise adherence identified based on exercise volume resulting from the group-based trajectory modeling (Figure 2). One group, consisting of 13% participants, was denoted as “non-starter”: these participants had low adherence from the initial week of the intervention and dropped to close to zero early in the intervention. A second group, consisting of 37% participants, was denoted as “tapered off’: these participants started well (i.e., adherence >60%) and then tapered off over time to low adherence (i.e., 10%). A third group, consisting of 50% participants, was denoted as “consistent exercisers”: these participants started well and had adherence greater than 50% of intended volume for the duration of the intervention.

Figure 2.

Group-based trajectory modeling of resistance training exercise adherence

The characteristics of the three groups demonstrated differed by gender, with females more likely to be in the “non-starter” and “tapered off” groups (Table 4). There were no other characteristics that differentiated between groups. Althought pattern of adherence did not differ statistically by duration of chemotherapy (p=0.17), participants receiving 3 months of chemotherapy were represented more in the “consistent exercisers” group (i.e., made up 33% of the overall sample, but 42% of the “consistent exercisers”). Whereas participants who received 6 months of chemotherapy made up 67% of the sample but 73% of the “non-starters” and 78% of the “tapered off” groups. Exploratory analyses stratified by planned chemotherapy duration (3-months or 6-months) yielded three distinct adherence trajectories, consistent with the analysis presented here.

Table 4.

Baseline characteristics of different adherence groups.

Characteristics	Overall (n = 86)	“Non-starter” (n = 11)	“Tapered off” (n = 32)	“Consistent exercisers” (n =43)	P
	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD
Age, years	56.3±13.0	58.6±13.8	55.1±12.3	56.0±13.2	0.72
Body Mass Index, kg/m2	26.8±4.7	28.0±4.5	26.5±5.8	26.7±3.7	0.65
Physical Activity, MET-hr/wka	8.6 (0-26.9)	1.9 (0-23.3)	8.1 (0-29.9)	8.8 (0-23.6)	0.49
	n (%)	n (%)	n (%)	n (%)
Gender					0.02
Male	43 (50)	4 (36)	11 (34)	28 (65)
Female	43 (50)	7 (64)	21 (66)	15 (35)
Education					0.11
High school	7 (8.2%)	3 (27%)	3 (9.4%)	1 (2.4%)
Some college/technical college	28 (33%)	4 (36%)	10 (31%)	14 (33%)
College Graduate	27 (32%)	1 (9.1%)	13 (41%)	13 (31%)
Post Graduate Degree	23 (27%)	3 (27%)	6 (19%)	14 (33%)
Employment					0.88
Employed, full-time	49 (57%)	6 (55%)	21 (66%)	22 (51%)
Employed, part-time	12 (14%)	2 (18%)	3 (9.4%)	7 (16%)
Unemployed	4 (4.7%)	0 (0%)	1 (3.1%)	3 (7.0%)
Retired	21 (24%)	3 (27%)	7 (22%)	11 (26%)
AJCC Stage					0.39
Stage 2	11 (13%)	2 (18%)	2 (6.2%)	7 (16%)
Stage 3	75 (87%)	9 (82%)	30 (94%)	36 (84%)
Randomized before initiation of chemotherapy					0.40
Yes	37 (43)	6 (55)	11 (34)	20 (47)
No	49 (57)	5 (45)	21 (66)	23 (53)
FOLFOX					0.16
Yes	48 (56)	6 (55)	22 (69)	20 (47)
No	38 (44)	5 (45)	10 (31)	23 (53)
Intended chemo duration at screening					0.17
3 month	28 (33)	3 (27)	7 (22)	18 (42)
6 month	58 (67)	8 (73)	25 (78)	25 (58)
Sarcopenicb					>0.99
Yes	31 (38)	4 (36)	11 (38)	16 (39)
No	50 (62)	7 (64)	18 (62)	25 (61)

Abbreviations: MET, metabolic equivalents.

Legend:

^aMedian and interquartile (25-75%) range;

^bSarcopenia = appendicular lean mass <7.25 for men or <5.5 for women on DXA scan;

DISCUSSION

This paper demonstrates the feasibility of detailed reporting of the unsupervised, home-based RT ExR_x during postoperative chemotherapy in patients with colon cancer who participated in a multicenter clinical trial. The FORCE trial hypothesized that RT improves chemotherapy tolerance by increasing muscle hypertrophy.(29) Consequently, the intended ExR_x was designed using the best evidence recommendations to promote muscle hypertrophy.(35, 38) To promote muscle hypertrophy, the RT ExR_x included twice-weekly training with moderate to heavy loads (65-85% 1-RM), high sets (3–5), and intermediate repetitions (6–10) of five large multi-joint movements.(35) Among all participants, attendance was 69% and adherence to the overall prescribed volume was 61%. However, adherence was higher in people who had the 3 month duration of chemotherapy (78%) versus the longer 6-month regimen.

The reporting of adherence to planned ExRx volume is a newer approach to providing readers with more insight into the exercise dose delivered and has been reported more for supervised than home-based interventions like FORCE.(18, 28, 39). Furthermore, this level of adherence is consistent with the attendance reported to exercise during chemotherapy or in the seminal trial of resistance training in women during chemotherapy for breast cancer by Courneya et al. which was 68% for a supervised, in-person intervention that was on average 17 weeks in duration.(40) Furthermore, adherence to unsupervised exercise does tend to have lower adherence to in-person, supervised exercise interventions. (41, 42) The potential of unsupervised, home-based exercise to provide greater access to more people undergoing chemotherapy is highly relevant in light of the limited access currently to in-person, supervised exercise as part of clinical cancer care (43, 44). More complete reporting of attendance and adherence to exercise during chemotherapy and/or in an unsupervised seeting will advance the field of exercise oncology. In men with prostate cancer receiving androgen deprivation therapy, who have a different side effect profile to those receiving chemotherapy, Lopez et al. reported adherence to the planned exercise volume of 88.5% with a three times per week, supervised, in-person combined aerobic and resistance training intervention.(27)

There are several important benefits to explicitly defining the planned ExR_x and describing the degree to which study participants adhered to each program variable of the ExR_x. First, explicitly specifying the intended ExR_x allows HFP to determine if the intended ExR_x is compatible with the study’s underlying hypothesis and objective. Second, describing the degree to which study participants could adhere to each program variable enables HFP to understand what is feasible and tolerated during chemotherapy in this population. Third, a detailed description of the completed ExR_x will provide an informed interpretation of the benefits of RT during chemotherapy for reporting on the primary and secondary study endpoints. Several systematic reviews have requested increased emphasis to report all components of the ExR_x in oncology clinical trials to promote replication and inform the translation of research into clinical care.(20–23, 26, 27) This paper outlines our initial approach to reporting the RT ExR_x during chemotherapy. However, providing standard guidance on how to approach adjustments, such as a flow diagram on when to reduce volume or intensity, could also provide HFP with additional guidance on how best to approach adjustments RT ExR_x. to better standardize decisions that are typically made using professional judgement.

At the time the FORCE trial was designed, the standard of care for postoperative chemotherapy in colon cancer shifted from 6 month regimens for all patients, to a risk-stratified approach wherein patients at low risk for disease recurrence could be offered 3 months of chemotherapy and patients at high risk for disease recurrence could continue to receive 6 months of chemotherapy.(31) We expected that adherence to the overall prescribed RT volume would decline longitudinally, particularly for patients who planned to receive 6 months of chemotherapy. We expected this because adherence to exercise declines over time in various settings,(45) and 6 months of chemotherapy is associated with increased frequency and greater severity of anticancer treatment-related toxicities,(31) which we predicted would negatively influence overall exercise adherence among participants who planned to receive 6 months of chemotherapy. This was observed with higher adherence in participants receiving 3 months of chemotherapy. Participants receiving 6 months of chemotherapy were more likely to be in the “non-starters” or “tapered off” groups, but this was not statistically significant. However, this suggests that behavioral characteristics (e.g., knowing how long you have to engage in exercise) may be an influential determinant, along with accumulation of treatment-related toxicities, resulting in lower overall exercise adherence.

Women were more likely to be in the “non-starter” and “tapered off” groups, which suggests additional behavior strategies could be targeted to these participants in future RT trials during chemotherapy. Demographic and treatment factors that are predictive of exercise adherence during chemotherapy have also been examined by others, including differences by disease stage, education and marital status that we did not observe in FORCE.(46–48) These observations underscore the need for intensive behavioral support to promote high exercise adherence during chemotherapy and that predictors may vary by tumour type, age of individuals and other demographic factors that, if understood, could be leveraged to provide additional behavioral suppport to participants who are particularly vulnerable to poor exercise adherence.

Further, the cyclic nature of chemotherapy could be considered in designing of ExRx for future interventions as a way to improve attendance and adherence. Kirkham et al. have reported on an approach to “chemotherapy-periodize” exercise, which builds the intensity or volume of the ExRx during the weeks when chemotherapy is not received and provides a lighter intensity or volume of the in ExRx in the week immediately following infusion.(8) In FORCE, participants completed a median of 1.4 sessions per week. Anecdotally, the HFP identified that a common pattern was that participants were not able to do any RT sessions in the week following chemotherapy due to treatment side effects, and then were more consistent with 2 sessions per week in the weeks they did not receive chemotherapy.

There are several limitations to this paper. The data used in our analysis were derived from self-reports from participants. Due to the unsupervised, home-based nature of the ExR_x, we do not have objective measures of exercise adherence. It is also consistently reported that adherence to in-person, supervised exercise typically can achieve greater adherence to progressive ExRx, especially for resistance training, so we cannot assume that the results from our unsupervised, home-based program, which although highly translational and easily adaptaive during Covid would have yielded similar results as an in-person exercise program. This study utilized a submaximal evaluation to estimate 1-RM strength for each exercise. It is possible submaximal evaluation overestimated the 1-RM muscle strength, which would explain why the observed % 1-RM was lower than prescribed throughout the trial. The ability to undertake in-person visits with the exercise trainer were impacted COVID-19 pandemic, which required all sessions to be done virtually for individuals who were already enrolled or newly enrolled, which may have negatively impacted adherence to the intervention, particularly around an individual’s willingness to progress the intensity of the load. Lastly, this analysis was not prespecified in the study protocol’s statistical analysis plan and should be viewed as exploratory.

There are several strengths to this paper. The ExR_x utilized an unsupervised, home-based RT program, similar to what has been used in patients with breast cancer.(49, 50) The home-based program reduced barriers to facility-based exercises, such as transportation and travel time, but we acknowledge may induce a less potent physiological stimulus to promote muscle hypertrophy. Reporting of the ExRx followed best practice reporting recommendations to facilitated evaluation and translation of findings into clinical practice.(51, 52) This multicenter trial was conducted at three academic sites, including KPNC (regional integrated managed care consortium), DFCI (urban, comprehensive cancer center), and PSCI (rural cancer center), which improves the generalizability of study findings and demonstrates the feasibility of standardizing complex behavioral procedures across diverse clinical settings.

CONCLUSIONS

In conclusion, this paper outlines suggested methodological advance in providing novel methods for reporting RT ExR_x in oncology clinical trials and highlight the unique complexity of prescribing RT to patients with cancer during chemotherapy. Overall, this paper aims to provide a rationale that will expand our thinking, foster constructive dialogue, and offer a stimulating premise for improved reporting of RT trials in clinical populations and in designing future research to elucidate the benefits of exercise for patients with cancer during chemotherapy.

Supplementary Material

SDC 1

SDC 1: Appendix 1.pdf

SDC 2

SDC 2: Ex Rx for 2 week cycles (FOLFOX and 5FU-LV)_Revised 6.5.19.pdf

SDC 4

SDC 4: Protein Supplementation Log 4.9.18

SDC 3

SDC 3: Ex Rx for 3 week cycles (CAPOX)_Revised 6.5.19.pdf