Background
Among the most common distressing symptoms facing patients with advanced cancer is loss of physical function [
1]. Its underlying aetiology is multifactorial, with increasing fatigue, muscle wasting and generalized debility all contributing to this phenomenon [
2]. Loss of physical function impedes the patient’s ability to perform activities of daily living, and increases dependence on caregivers leading to additional emotional and psychological burden [
3]. The importance of keeping mobile is linked to maintaining independence and overall quality of life [QoL] in patients with advanced cancer.
Increasing attention has been given to physical activity as a QoL intervention in cancer patients [
4]. Physical activity interventions can improve cancer-related fatigue and physical functioning in early stage cancer patients [
5,
6]; however, these benefits have not been confirmed for patients at later stages of cancer. There is preliminary evidence that select patients with advanced cancer express willingness to participate in a physical activity intervention, with positive benefit on some supportive care outcomes [
7].
Oldervoll et al. conducted a randomized controlled trial to examine the effects of an eight-week group exercise program versus usual care on 231 patients with advanced cancer with median survival of one year [
8]. Analyses revealed no significant differences between groups in the primary endpoint of physical fatigue, however there were improvements in the intervention group in physical performance as measured by the shuttle walk and handgrip strength tests. Oldervoll et al. earlier acknowledged that patients who declined participation had identified limitations of fatigue, lack of mobility, and the burden of physically getting to the hospital gym where the exercise intervention took place [
9]. Oldervoll et al. concluded that these limitations “might indicate a need for specially tailored interventions…in the form of home-based exercises adjusted for the individual patient” [
10].
No home-based physical activity program has been validated for patients with advanced cancer receiving palliative care. Porock et al. conducted a pilot study of nine home hospice cancer patients who were administered a home-based program based on the Duke Energizing Exercise Plan, with a range of different physical activities prescribed according to the patient’s individual condition and tolerability; despite the trend towards increased QoL scores, the authors concluded that the optimal type of physical activity program for this population is still unknown [
11].
We previously completed a pilot survey of fifty patients with advanced cancer with a median survival of 104 days; 92% of participants reported that they would be interested in and able to participate in a physical activity program [
12]. Moreover, 84% of participants indicated a preference for a home-based individual (i.e. not group) physical activity program. Walking and resistance training were the top two activities endorsed by these participants, with 56% preferring to participate in up to 3 physical activity sessions per week [
13].
Incorporating patient preferences is critical in designing an effective intervention [
14] and may enhance recruitment and adherence, and potential benefits. Based on the preferences identified in our pilot survey and using a similar recruitment strategy, this study examined the initial development and feasibility of a home-based functional walking program in patients with advanced cancer receiving palliative care.
Discussion
The aim of this study was to examine the initial development and pilot testing of a physical activity intervention in patients with advanced cancer receiving palliative care. Based on our pilot survey data, there was a majority preference for home-based, solo interventions, with walking being the most preferred activity [
13]. Therefore a modified home-based functional walking program was designed to incorporate the specific physical activity preferences of this sample, and a similar recruitment strategy was adopted.
There are a number of feasibility issues deserving of attention from this study. From our pilot survey study, we were able to recruit 50 patients over a 7 month period [
12]; using the same eligibility criteria and local recruitment strategy, however, we were only able to recruit 9 patients over a 6 month period. A total of 504 patients were screened through the RPHCP and CCI outpatient radiotherapy units on behalf of all palliative care research studies that were open for accrual during that 6-month period, however only 15% (96/504) consented to being contacted with regards to this particular study (see Figure
1). In both RPHCP and CCI settings, the first contact was such that the patient’s interest in being contacted by the study coordinator took precedence over obtaining physician-estimated survival; those patients who refused, therefore, may not have fulfilled all eligibility criteria at the time of initial screening.
Of the 96 patients who consented to being contacted by the study coordinator, 53% (51/96) fulfilled all eligibility criteria for this study. Therefore of all patients who consented to being contacted by the study coordinator and who met all eligibility criteria for this study, our accrual rate was 18% (9/51). Locally, this accrual rate is comparable to Hutton et al.’s study of dietary intake in 151 patients with advanced cancer, wherein the authors reported an estimated 21% accrual rate from both the CCI and RPHCP [
36]. Elsewhere, Porock et al. reported a recruitment rate of 46% (11/24) in their pilot study of 4-week home-based exercise program in home hospice care patients, with incomplete information as to attrition rates and reasons for withdrawal [
11]. Oldervoll et al. reported a recruitment rate of 58% (231/400) in their recent RCT, however the reasons behind refusal to participate were not reported; 36% of the intervention group, versus 23% of the control usual care group, were lost to follow-up primarily due to disease progression [
8]. Compared to the 104-day median survival of our pilot survey sample [
12], the median survival of the 9 consented participants in this study was 92 days. It is therefore likely that our participants were further along the cancer trajectory than those of Oldervoll et al. [
8]. Untimely attrition over a 6-week period in this population with such limited prognosis is not unexpected [
37].
From our pilot survey, the majority felt willing and able to participate in a physical activity intervention [
13]. The ability to participate in a physical activity program, however, may fluctuate depending on patient-reported symptoms: 69% (35/51) of eligible patients declined consent to the study because of severe symptoms, with fatigue being the most common reported symptom. These findings concur with Mercadante et al.’s prospective study of 400 palliative home care patients with a mean survival of 52 days, wherein there was an increase in fatigue scores over time, with a peak in symptom intensity and frequency at the lowest levels of Karnofsky performance status [
38].
Of the 9 patients who enrolled in our study, 6/9 dropped out with the most common reason being admission to acute care for severe symptoms. This rate of attrition is higher when compared to large palliative care trials; Oldervoll et al’s recent RCT reported that 36% of the intervention group, versus 23% of the control usual care group, were lost to follow-up, primarily due to disease progression [
8]. In contrast, 5/6 dropouts occurred within 4 weeks of starting our physical activity intervention. Given the shorter median survival of our pilot survey sample, consideration was given to maintain the intervention as long as it was feasible and safe for the patients.
Cases #1 and #2 were obese [BMI ≥ 30 kg/m
2], with the former meeting the WHO criteria for morbid obesity [BMI ≥ 40.0 kg/m
2] [
39]. The relative proportion of fat to skeletal muscle mass in these patients, however, was not investigated. In a body composition study utilizing computed tomography images of 111 pancreatic cancer patients undergoing palliative treatment, 40% were overweight or obese, and 16% were both sarcopenic and obese; sarcopenic obesity was shown to independently predict survival, and was associated with poorer self-assessed functional status [
40]. Although one may postulate that obesity contributed to poor mobility and physical functioning in these cases, further studies are required to elucidate the relationship between sarcopenic obesity, physical functioning and physical activity levels in this population.
In all three cases, increasing symptom burden resulted in the delay in progression in both the aerobic and strength exercise components. There were no reported difficulties with use of the activPAL™ or its generation of data; the number of steps and estimated total energy expenditure, however, decreased significantly over the course of six weeks. Although none of the three participants achieved the target daily walking prescription, all 3 participants were able to continue both aerobic and strength components at reduced levels. Future consideration should be given to a maintenance, rather than progressive, target daily walking prescription given the symptom burden of this patient population.
Currently, there is no recommended minimum level of physical activity for palliative cancer patients [
41]; however any amount of physical activity that the patient can tolerate may be better than engaging in no activity at all. Hence one-on-one supervision takes on greater significance in our study, wherein modifications could be made to strength exercises without missing the entire session completely.
On the other hand, one-on-one supervision resulted in the exclusion of potentially eligible participants. Of the 20 eligible patients who were screened from the Department of Symptom Control and Palliative Care and the CCI outpatient radiotherapy units and who consented to being contacted by the study coordinator, 35% (7/20) were unable to participate because they lived out-of-town. While having one-on-one supervision was identified as one of the top advantages by the three presented case reports, the option of a self-directed intervention by means of telehealth approaches, an instructional handbook or video may increase accrual in future pilot trials. Likewise, future consideration should be given to streamlining the number of outcome assessments in view of being less burdensome on this patient population.
Moreover, patients may recognize the difference between a one-time cross-sectional survey on physical activity and a six-week progressive physical activity intervention. Given that the recruitment agencies and processes were identical, one would expect the influence of gatekeeping to be equivalent between this study and our previous pilot survey [
12]. Taken together, our results suggest that patients who expressed interest in the idea of physical activity, may have encountered barriers to participating and carrying through with an actual intervention. Eliciting patient barriers to physical activity would therefore be deserving of future research.
Nevertheless, improvements were noted in total MQOL scores in two of the three cases presented. In contrast, two of the three cases showed a decline in physical functioning, as demonstrated by the total LLFDI scores. All three participants shared an overall trend towards worsening ESAS symptom scores, and worsening total BFI global fatigue scores post-intervention. Because of the small sample size, it is not possible to distinguish whether these effects were secondary to the physical activity program or to progression in the underlying cancer; as shown in Headley et al’s pilot RCT of a seated exercise program in stage IV breast cancer patients [
42], a slowing of the inevitable decline in fatigue and quality of life scores may be a realistic interventional goal which would account for the changes seen in our case series.
Although this small sample precludes drawing conclusions on intervention effects or determining sensitivity of outcome measures, our case series provides rationale for future feasibility studies. With respect to the local recruitment strategy, further characterization of the screened patient population, including exploration of the reasons for declining consent to be contacted for research, would aid in defining which subgroup would most benefit from an intervention. Recruitment and retention may be improved by opening enrolment to advanced cancer patients irrespective of clinician-estimated prognosis. Further modifications, such as shortening the duration of the intervention, examining the effects of aerobic or strength components separately, and including an option for self-directed programming, may also optimize recruitment and retention.
Competing interests
None of the authors have any potential conflicts of interest.
Authors’ contributions
SSL and KSC conceived and designed the study, and drafted the manuscript. SL conducted the study, data collection and data analysis. SW and VB participated in study conception and design, and helped to draft the manuscript. All authors read and approved the final manuscript.