Effects of an exercise-based oncology rehabilitation program and age on strength and physical function in cancer survivors

Abstract

Purpose

Cancer therapy diminishes strength and physical function in cancer survivors. Whether oncology rehabilitation (OR) exercise training following therapy can correct these deficits, and whether its effectiveness differs by age, is not clear. We examine the utility of a clinically based, 12-week, combined aerobic and resistance training intervention on muscle strength and physical function in two age groups of cancer survivors.

Methods

Strength and physical function measures were assessed in middle-aged (45 to 64 years) and older (≥ 65 years) patients following treatment for stage 0–III cancer before and after the OR training program.

Results

Older patients had lower physical function compared to middle-aged patients across a range of subjective and objective measures at baseline, and exercise improved all indices of physical function and strength in both age groups. Compared to the middle-aged individuals, older participants tended to have less improvement leg strength and the 5 time sit to stand (5TSTS) test as a result of OR. In models predicting post-intervention measures, older age contributed to less improvement in walking distance and power as well as the 5TSTS test.

Conclusion

Prior to beginning the OR exercise program, middle-aged patients had higher physical function compared to older patients. However, a 12-week aerobic and resistance training intervention improved physical function across both age groups, although older age did limit responsiveness in some physical function measures. The physical function and strength of middle-aged and older cancer survivors improve in response to an exercise-based OR program after cancer treatment.

Introduction

Exercise rehabilitation following cancer therapy may result in recovery of skeletal muscle strength that is lost due to cancer, treatments, or the combination of these factors [1–5]. Randomized, controlled trials have shown that exercise interventions following treatment improve skeletal muscle strength of cancer survivors [6]. Strength has important implications for maintaining physical function and ultimately physical activity. Limited activity can contribute to weight gain and greater risk for metabolic disease, which is a noted challenge for cancer survivors [7]. Whether the effect of exercise rehabilitation in cancer survivors translates into improved physical function, and whether older survivors benefit from exercise training to the same extent as younger survivors, however, is not clear.

Age is a risk factor for many cancers [8], with the median age of a cancer diagnosis from any site being 65 years [9]. Progress in cancer therapy has improved outcomes, leading to an increase in survivors > 65 years, an age group that comprises the largest proportion of survivors (63%) [10]. Maintaining muscle size and strength and physical function of cancer survivors, in particular those over age 65, is important, as disability impairs quality of life, threatens independence, and increases morbidity and mortality [11–14]. Compared to age matched, non-cancer controls, cancer survivors have greater functional limitations and impaired mobility [15]. In fact, functional losses experienced by elderly cancer survivors are twice that of matched controls without breast cancer [16, 17] and functional limitations are associated with increased risk of death [18, 19].

To address these questions, we examined the impact of an oncology rehabilitation (OR) intervention that combines aerobic and resistance exercise training program on muscle strength and physical function in a large cohort of cancer survivors attending a clinically OR program. We dichotomized the population into middle-aged (45 to 64 years) and older (≥ 65 years) groups to examine whether the intervention differentially affected physical function by age.

Methods

Participants

Patients were selected from cancer survivors participating in an exercise-based OR program at the University of Vermont Medical Center. The study population at baseline consisted of cancer survivors referred to between January 1, 2012 and August 2018. For the analysis, we included men and women with a cancer diagnosis meeting the following criteria: (1) completion of initial therapy at least 2 months prior to participation, (2) receipt of at least two modalities of therapy (surgery, radiation, and/or chemotherapy), (3) cancer stages of 0–III. Individuals who were more than 5 years post-diagnosis at the time of exercise initiation and those with functional limitations that made independent exercise unsafe were excluded. We dichotomized the population based on the median age of cancer diagnosis of 65 years (Miller et al 2016). Middle-aged participants were age 45–64, and the older group had an age of 65 or greater. These age categories represent 91% of cancer survivors (37.3% age 45–64 and 64.0% age ≥ 65 [20]. Informed consent, which was approved by the University of Vermont Institutional Review Board, was obtained from all individual participants included in the study.

Intervention

The OR program described was a clinical exercise intervention available at our cardiac rehabilitation facility offered to all cancer survivors at no cost (supported by a development fund) as a component of care at our institution, as described in detail [21]. Briefly, the program included supervised resistance and aerobic training twice weekly for 12 weeks or 24 total sessions. The resistance training included 3 upper extremity (chest press, overhead press, bent over row, or wall pushups) and 5 lower extremity (squat, lunges, bridges, straight leg raises, side or standing abduction) exercises, with initial weight set at 60–70% of one repetition maximum (1RM). The resistance exercises selected for each muscle group are adjusted if patients have functional deficits that preclude proper performance of the exercise. Two sets of 8–12 repetitions are completed for each exercise, allowing 1–2 min of rest between sets. When 12 repetitions can be completed, the resistance is advanced to a higher load. Participants perform aerobic activity guided by heart rate for individuals who completed an exercise tolerance test (ETT) prior to exercise initiation and rate of perceived exertion if an ETT was not completed. Generally, brisk walking is selected for at least 20 min each session progressing to 40 min by week 8. Three additional days of aerobic training outside of the gym was encouraged. Exercise trainers supervised participants for safety and advanced both aerobic and resistance training prescriptions. To assure safety in a group setting, potential participants are evaluated by a physician or advanced practice provider and a physical therapist prior to initiating the program.

Physical function and muscle strength outcomes

Measurements were conducted prior to starting the intervention and at the completion of the intervention. Lower extremity (LE) strength was measured by one-repetition maximum (1-RM) for double leg press as described in the ACSM Guidelines for Exercise Testing & Prescription [22]. A 6-min walk (6MW) test [23] was performed and power production calculated as: 6-min walk power (W) = body mass (kg) × 9.8 (m/s²) × average gait speed (m/s), where 9.8 represents the acceleration of gravity. Power production takes into account the body mass carried for the distance walked. A five time sit to stand which is a measure of lower extremity functional strength was completed [24]. Finally, self-reported physical function was collected using the Physical Functioning Scale (PFS) of the Medical Outcomes Study, Short Form Health Survey (SF-36) [25]. Self-report of current physical activity was assessed using the Godin Leisure Time Exercise Questionnaire (GLTEQ) [26], and fatigue was determined using the Fatigue Symptom Inventory (FSI) [27, 28].

Statistics

Descriptive statistics for demographic and baseline strength and functional variables were calculated for all eligible participants (n = 299). Results are presented as frequencies or mean ± SD depending on the variable. Comparisons were made between age groups at baseline using Fisher’s exact test for categorical variables and the Kruskal-Wallis test for continuous variables. Paired t tests of pre- and post-strength and functional measures were conducted by age group (i.e., within ≥ 65-year-olds and 45–64-year-olds). Analysis of covariance (ANCOVA) was used to determine which variables predicted the post-intervention values of lower extremity strength, 6MWT, power production, and 5 time sit to stand (5TSTS) over and above their baseline values. Covariates tested in the model include age group, gender, cancer type, time since diagnosis, receipt of chemotherapy, receipt of radiation, exercise attendance, BMI, fatigue, and baseline physical activity. All covariates were modeled one at a time. If a p value ≤ 0.25 was observed, it was included in a multivariable model. After potential model variables were identified, only covariates that contributed at a significance level of p < 0.05 were included in the final model. As a final check, in the model that included significant covariates, covariates that were eliminated were reintroduced one at a time to see if they contributed to a final model. All statistical analysis was performed using SAS 9.4 (SAS Institute, Cary, NC). Across all tests, statistical significance was defined as p < 0.05 (2-tailed).

Results

The study inclusion criteria were met for 481 cancer survivors who received baseline testing (n = 178, age ≥ 65 and n = 303, age 45–65). The program was discontinued by 183 (38%) individuals (n = 56, 31% age ≥ 65 and n = 127, 42% age 45–65). The primary reason for discontinuation was time constraints and work obligations (47%) followed by a new medical issue (28.6%). A total of 299 (62%) participants have pre- and post-test data (n = 122, 41% age ≥ 65 and n = 177, 59% age 45–65). Those not completing the exercise intervention did not differ from those completing in age, type of cancer, BMI, or type of cancer therapy received (data not presented).

Mean age for middle-aged and older groups was 55.6 ± 5.4 years (range 46–64 years) and 70.8 ± 4.9 years (range 65–87), respectively (Table 1), with a median time since diagnosis of 14.5 ± 13.9 and 12.7 ± 14.2 months, respectively. The majority were female (81%) and breast cancer survivors (63%), though there were more men in the older group. The average number of exercise sessions attended out of a total of 24 available was 19.2 ± 3.2 and 19.0 ± 3.3 for middle-aged older groups, respectively.

Table 1.

Demographics for participants of oncology rehabilitation (M ± SD)

	Older (n = 122)	Middle age (n = 177)	p
Age at diagnosis (years)	69.3 ± 4.7	55.3 ± 5.1	< 0.001
Range	(62–86)	(45–64)
Age at intervention (years)	70.6 ± 4.6	56.4 ± 5.1	< 0.001
Range	(65–87)	(46–64)
Time since diagnosis (months)	14.2 ± 16.0	12.7 ± 13.1	0.482
Range	(2–70 months)	(3–70 months)
Gender
Female	83 (68%)	160 (90%)	< 0.001
Male	36 (32%)	17 (10%)
Type of cancer
Breast	57 (47%)	131 (74%)	< 0.001
Other	64 (52%)	46 (26%)
Received chemotherapy	57 (47%)	105 (59%)	0.034
Received surgery	98 (80%)	162 (92%)	0.008
Received radiation	78 (64%)	129 (73%)	0.126
BMI	28.8 ± 6.1	29.0 ± 6.4	0.884
Range	(19.3–48.0)	(17.5–62.8)
Average attendance	19.0 ± 3.3	19.2 ± 3.2	0.649
Range	(9–24)	(8–24)

Categorical variables were tested using Fisher’s exact test. Continuous variables were tested using the Wilcoxon Rank Sum Test

M mean, SD standard deviation

Age differences in muscle strength and physical function at baseline

6MWT distance, gait speed, and power production were all significantly lower in the older versus middle-aged group at baseline (Table 2). No age difference in LE 1RM strength was found. The older group required greater time to perform the 5TSTS test. Self-reports of physical function (PFS) were significantly lower for the older age group, while self-reported physical activity and fatigue did not differ significantly between age groups.

Table 2.

Differences in baseline variables between age groups (M ± SD)

Variable	Older (n = 122)	Middle age (n = 177)	p
Distance walked (m)	475.8 ± 82.0	533.4 ± 110.6	<0.001
Gait speed	1.32 ± 0.23	1.48 ± 0.31	<0.001
Power production	1031.3 ± 240.8	1134.1 ± 290.0	0.002
Leg press (lbs)	137.3 ± 43.1	133.1 ± 43.7	0.390
5 Time sit to stand (s)	12.9 ± 3.7	11.3 ± 2.7	0.002
SF 36 Physical Function Score	69.9 ± 22.2	75.8 ± 21.6	0.014
Self-report of moderate physical activity (min/week) GLTEQ	21.6 ± 34.2	26.7 ± 32.9	0.1891
Fatigue
FSI severity	3.2 ± 1.8	3.5 ± 1.9	0.211
FSI disruption	15.2 ± 13.2	16.3 ± 13.7	0.555

M mean, SD standard deviation

Changes in strength and physical function after training

All strength and physical function measures improved with training in both groups (Table 3). Percent change from baseline was calculated for each individual and compared across age groups (Table 4). These analyses showed that similar improvements were found for all variables. However, there was a strong trend for less improvement in LE 1RM strength and 5TSTS in the older group.

Table 3.

Change in strength and function after the intervention for each age group (M ± SD)

Variable	Older	p	Middle age	p
Distance walked (m)
Pre	475.8 ± 81.9		533.4 ± 110.6
Post	571.1 ± 97.2	< 0.001	631.3 ± 96.0	< 0.001
Power production
Pre	1031.3 ± 240.8		1134.1 ± 290.0
Post	1230.4 ± 296.8	< 0.001	1358.7 ± 294.1	< 0.001
Leg press (lbs)
Pre	137.3 ± 43.1		133.1 ± 43.7
Post	155.7 ± 42.4	< 0.001	159.6 ± 43.1	< 0.001
5 Time sit to stand (s)
Pre	12.4 ± 3.2		11.4 ± 3.1
Post	9.2 ± 2.6	< 0.001	8.1 ± 2.4	< 0.001
Self-report of physical activity (min/week) GLTEQ
Pre	21.6 ± 34.2		26.7 ± 32.9
Post	62.1 ± 43.2	< 0.001	72.7 ± 45.9	< 0.001
Physical Function Score
Pre	69.9 ± 22.2		75.8 ± 20.9
Post	80.6 ± 20.2	< 0.001	87.2 ± 15.9	< 0.001

M mean, SD standard deviation

Table 4.

Percent change in strength and functional measures between age groups (M ± SD)

Variable	Older	Middle age	p
Distance walked	22.3 ± 21.6	25.5 ± 50.1	0.414
Power production	24.3 ± 25.4	25.9 ± 48.8	0.292
Leg press	18.5 ± 30.9	30.1 ± 63.4	0.066
5 Time sit to stand	24.6 ± 15.6	27.6 ± 17.8	0.061
SF 36—Physical Function Score	81.0 ± 579.4	54.0 ± 393.1	0.720

M mean, SD standard deviation

Based on ANCOVA, age influenced intervention response for 6MWT and power production, 5TSTS, and self-report of physical function. For 6MWT and power production improvements, older individuals had less improvement as a result of the intervention, while those who were male had greater improvements. BMI influenced the outcome of the 6MWT and power production but in opposite directions. For distance walked in 6 min, those with a lower BMI walked further post-intervention while those with a larger BMI had greater improvements in power production. Those with an older age had less improvement in 5TSTS, while longer time since cancer diagnosis, larger BMI, and more fatigue disruption contributed to greater improvements in the 5TSTS test after post-intervention. Greater improvements in leg press strength were identified for those who received chemotherapy, attended more sessions, and who had a higher BMI. Those who were older had less improvement in self-reports of physical function after the intervention.

Discussion

Exercise interventions tested in cancer survivors, both during and after therapy, are efficacious at enhancing skeletal muscle function and overall physical function in the research setting of randomized, controlled trials [6, 29]. However, there are hurdles to transferring the benefits of exercise training identified in these strict experimental settings to patients in clinical practice. We provide a clinically relevant exercise–based OR program that is available to all cancer survivors at our institution [21]. Results from this study show that a combined aerobic and resistance exercise training rehabilitation program initiated after the completion of cancer treatment improves physical function and muscle strength in both older and middle-aged cancer survivors. Older cancer survivors experienced comparable gains in some physical function measures but diminished responses in others.

Effects of exercise rehabilitation

The 12-week exercise intervention improved LE muscle strength and functional parameters, including walking performance and the time to complete 5 sit to stand transitions. These indices reflect two different physical function parameters. The 6MW performance reflects sub-maximal aerobic fitness [30], although there is arguably a contribution from LE muscle function [31]. In contrast, a 5TST reflects predominantly LE muscle strength [32]. Thus, our combined program was effective at enhancing both endurance- and strength-based physical function indices. These improvements should have real-world functional significance in cancer survivors for reducing the possibility of disability or disabling events, particularly in older survivors. For instance, prior to the training intervention, 5TSTS was > 12 s in 52.6% of the older group, a cutpoint that is associated with an increased risk for falls [33]. After the intervention, 5TSTS was > 12 s in only 17.8% of the older cohort.

Participants perceived the functional benefits of exercise rehabilitation, as reflected in improvements in self-reported physical functional capacity and fatigue. Fatigue is the most common late and lingering side effect of cancer and its treatment. Evidence suggests a relationship between cancer-related fatigue and inflammatory biomarkers [34]. A potential mechanism for improved fatigue for cancer survivors who exercise is a reduction in pro inflammatory biomarkers, in particular with combined training [35]. Interventions that reduce fatigue have great potential for improving quality of life. In fact, both middle-aged and older participants self-reported improved physical function capacity, an important determinant of quality of life [36]. The combination of improvements in physical function (objective measures), survivors’ perception of that improvement (subjective measures), and reduced fatigue may encourage continued physical activity which has potential for lowering cancer-related mortality [37].

Modifying effects of age

Like older non-cancer populations, older cancer survivors had lower physical functional capacity, and they reported perception of these functional limitations prior to the exercise program (Table 2). Baseline leg strength was low for both age groups. Compared to gender- and age-based normative values [22], the older participants had an average leg strength percentile that was 19% of predicted. For individuals age 45–65, at baseline, the leg strength percentile was 22% of predicted. Collectively, these results suggest that cancer patients are more functionally disabled than the broader population, although we acknowledge that these normative values represent a self-selected segment of the population [38] that may overestimate the strength of the general population. Strength data is not available for participants prior to receiving cancer treatment in the current study, but cancer therapy, and the muscle disuse that often accompanies it [39], contributes to a decline of muscle function. In this context, preventing loss of strength during therapy may represent a potential alternative, rather than rehabilitation following treatment.

Despite greater impairments at baseline among the older participants, exercise training was broadly effective at improving and functional performance and strength across the measures examined. One exception was leg strength and 5TSTS as there was a trend toward reduced responsivity to the training program in older survivors (Table 4). Our results agree with others, who have reported impaired recovery of physical function in older cancer survivors [40]. Less responsiveness in the 5TSTS may be explained by a tendency towards a diminished improvement in leg strength in the older group, as the sit-to-stand transition is heavily dependent on leg strength [32]. Accordingly, strength-based exercises for the lower extremity should be emphasized in older cancer survivors to better enhance function in daily activities. We have previously shown the value of resistance only training programs for improving real-world functionality in healthy elderly [41] and those with cardiac disease [42, 43]. The proper proportion of aerobic and resistive exercises in an oncology rehabilitation program for survivors of different ages, however, will require further study.

After taking into consideration baseline values for each measure, age also influenced the response to the intervention for several functional measures. Distance walked in 6 min and power-produced walking did not improve as much as a result of the intervention for individuals in the older age group. Age also influenced improvement in the 5TST test, with older individuals showing less improvement after the intervention. Finally, those who received chemotherapy as part of their cancer treatment had greater improvements in leg strength, which may be explained by the fact that chemotherapy reduces muscle size and function [4, 5]. It is likely that those who received chemotherapy had less leg strength to start with and therefore had a greater ability to improve strength.

Mechanisms of improvement with exercise

The mechanisms underlying improvements in muscle strength and physical function are unclear. We and others have characterized alterations in skeletal muscle size, oxidative capacity, and contractility in cancer survivors [3, 4] that likely develop during treatment due to the effects of chemotherapy [5] and muscle disuse [39]. Accordingly, the benefit from aerobic and resistance training likely derive, in part, from the prevention of muscle atrophy, weakness, and loss of oxidative capacity [5, 44]. Following therapy, it is unclear if the benefit of exercise training results from correction of these fundamental deficits in muscle size and function or if these improvements derive from alterations in other physiological systems. For instance, we recently demonstrated that the large majority of the beneficial effects of resistance exercise training in physically inactive, older individuals likely derive from improvements in neural activation/innervation [45], rather than intrinsic muscle size and function [46]. This is an important distinction, as functional gains associated with neural improvements may not be as durable as remediation of muscle size and intrinsic functionality. Functional improvements also depend on the intensity of the exercise. Thus, strength gains with the moderate intensity resistance training program (60–70% 1RM) used in the current study may be more dependent on changes in neural activation versus improvements in muscle size or function. Future studies are needed to define the mechanisms underlying these functional improvements and their relationship to training intensity to both insure their durability for prevention of long-term physical disability and better inform exercise prescription.

Strengths and limitations

Several strengths of our study should be noted. First, our results are generalizable to adult cancer survivors across a reasonably broad age range exercising in a clinically implemented OR exercise program similar in design to recent recommendations [47]. Second, we demonstrated the benefits of this OR exercise program in a large sample of patients. Third, improvements were identified across several measures of function including objective measures and self-report for both older and middle-aged cancer survivors. Despite these strengths, we must acknowledge several limitations to our study. First, our cohort self-selected to participate in this exercise program and, accordingly, is likely healthier and more functional than the general cancer population. Indeed, PFS scores were higher than a large cohort of cancer survivors from a Medicare database [48]. Second, as the intervention is a clinically based program available to all cancer survivors at our institution, we did not include a control group that did not exercise. Because of this, we are unable to discern what proportion of the benefit of the exercise training program may be related to simple recovery of strength and function over time. While some have demonstrated functional improvements over time [1], others show no improvement [49]. Third, those completing the program may have less dysfunction than those that dropped out. However, there were no differences in baseline measures of strength and function between those who completed and those who discontinued the program (data not included). Nutrition information and weight loss opportunities are available for survivors, but not all participate in these programs during the supervised exercise program. Therefore, participation or lack of participation in nutrition-related programs may have influenced outcomes. Finally, non-cancer factors, such as comorbidities and their associated symptoms, may modify responsivity to exercise training and these factors were not evaluated.

Conclusion

Our results demonstrate that cancer survivors experience improvements in strength and physical function across a range of objective and self-reported indices in response to a 12-week clinical OR exercise intervention. These improvements have important implications for decreasing morbidity, subsequent disability and disabling events. Additionally, although older and middle-aged cancer survivors showed improvements in all indices with training, older participants showed diminished improvements in walking parameters and sit-to-stand transition, which may relate to an impaired ability to improve muscle strength. Collectively, our findings suggest that a clinically delivered, exercise-focused OR program offers an opportunity to extend the benefits of exercise training to cancer survivors.