Exercise and nutrition interventions in advanced lung cancer : a systematic review

Pain, fatigue, anorexia, and weight loss are some of the most prevalent physical symptoms in advanced cancers1,2. Unintentional weight loss is recognized as an independent predictor of poor health and earlier death in advanced cancer3,4. Nutrition status has also been found to directly affect both tolerance to and effectiveness of palliative chemotherapy treatments for solid tumours3. Although pain control in cancer is continually improving, with standardized guidance for assessment and treatment5, the optimal management of fatigue, anorexia, and weight loss—all recognized components of cancer cachexia syndrome—are still to be determined6,7. Cancer cachexia syndrome is multifactorial and complex, and its causes are still not fully understood. A group of leading international experts in clinical cancer cachexia research and treatment recently defined it thus6: [A] multifactorial syndrome defined by an ongoing loss of skeletal muscle mass (with or without loss of fat mass) that cannot be fully reversed by conventional nutritional support and leads to progressive functional impairment. Its pathophysiology is characterised by a negative protein and energy balance driven by a variable combination of reduced food intake and abnormal metabolism. (p. 490) ABSTRACT


BACKGROUND
Pain, fatigue, anorexia, and weight loss are some of the most prevalent physical symptoms in advanced cancers 1,2 .Unintentional weight loss is recognized as an independent predictor of poor health and earlier death in advanced cancer 3,4 .Nutrition status has also been found to directly affect both tolerance to and effectiveness of palliative chemotherapy treatments for solid tumours 3 .Although pain control in cancer is continually improving, with standardized guidance for assessment and treatment 5 , the optimal management of fatigue, anorexia, and weight loss-all recognized components of cancer cachexia syndrome-are still to be determined 6,7 .Cancer cachexia syndrome is multifactorial and complex, and its causes are still not fully understood.A group of leading international experts in clinical cancer cachexia research and treatment recently defined it thus 6 : [A] multifactorial syndrome defined by an ongoing loss of skeletal muscle mass (with or without loss of fat mass) that cannot be fully reversed by conventional nutritional support and leads to progressive functional impairment.Its pathophysiology is characterised by a negative protein and energy balance driven by a variable combination of reduced food intake and abnormal metabolism.(p.490)

ABSTRACT
In this systematic review, we sought to evaluate the effect of physical activity or nutrition interventions (or both) in adults with advanced non-small-cell lung cancer (nsclc).

Methods
A systematic search for relevant clinical trials was conducted in 6 electronic databases, by hand searching, and by contacting key investigators.No limits were placed on study language.Information about recruitment rates, protocol adherence, patientreported and clinical outcome measures, and study conclusions was extracted.Methodologic quality and risk of bias in each study was assessed using validated tools.

Main Results
Six papers detailing five studies involving 203 participants met the inclusion criteria.Two of the studies were single-cohort physical activity studies (54 participants), and three were controlled nutrition studies (149 participants).All were conducted in an outpatient setting.None of the included studies combined physical activity with nutrition interventions.

Conclusions
Our systematic review suggests that exercise and nutrition interventions are not harmful and may have beneficial effects on unintentional weight loss, physical strength, and functional performance in patients with advanced nsclc.However, the observed improvements must be interpreted with caution, because findings were not consistent across the included studies.Moreover, the included studies were small and at significant risk of bias.
Lung cancer accounts for the highest proportion of cancer deaths in the developed world, with non-small-cell lung cancer (nsclc) accounting for approximately 70%-85% of all lung cancer diagnoses [8][9][10] .The high incidence of cancer cachexia symptoms arising in advanced nsclc 11 has made this patient population a frequent target for cancer cachexia research 12,13 .Specialized multidisciplinary clinics combining individualized nutrition and physical activity interventions, together with optimal psychosocial support and medical management, are being developed worldwide.Within these clinics, dietetic support includes advice on appropriate food selection based on likes, dislikes, and symptoms affecting dietary intake.Dietitians also advise on food fortification with or without macro-and micronutrient supplementation to correct any dietary deficiencies.Physiotherapists provide individualized exercise plans combining resistance and aerobic training for cardiovascular fitness, muscular strength, muscular endurance, flexibility, and lean mass retention.These clinics appear promising in terms of improved physical functioning, better dietary intake, weight stabilization, and fatigue reduction [14][15][16][17] .Optimal program design and timing of interventions has yet to be determined 6,18 .
Our aim was to review trials of interventions in physical activity or nutrition (or both) focusing on the management of any combination of fatigue, anorexia, and unintentional weight loss (symptoms of cancer cachexia) in patients with advanced nsclc.A further aim was to evaluate the effectiveness of the interventions.

Types of Studies
Any type of clinical trial evaluating the effects of physical activity or nutrition interventions for the management of cancer cachexia symptoms in advanced nsclc was eligible for inclusion in the review.

Types of Participants
Participants in the trials had to be adults (≥18 years of age) with stage iiib or iv nsclc.Participants were included regardless of whether they were actively receiving anticancer therapy at the time of the intervention.

Types of Interventions
All included papers were required to have a physical activity or nutrition treatment as the main intervention or to contain independently extractable data on such an intervention.
Physical activity interventions were defined as any one or a combination of flexibility training, resistance training, and cardiovascular training.Interventions could be supervised or unsupervised, be undertaken at any location, and be individualized or group-based in nature.Characteristics of the training program such as the type, intensity, frequency, duration, and extent of supervision and adherence are reported if that information was supplied.
Nutrition interventions included any one or a combination of the provision of dietary counselling, prescribed nutritional supplementation, and use of over-the-counter dietary supplements.Characteristics of the nutrition intervention such as the type, dose, duration, and extent of supervision and adherence are reported if that information was supplied.

Identification of Studies
A search strategy (Appendix A) was designed for identifying studies from the following databases, with no limits imposed on study language: central (Ovid), Cochrane Database of Systematic Reviews (Ovid), medline (Ovid), embase (Ovid), cinahl Plus, and the National Research Trials Register up to October 22, 2012.Hand-searches of relevant journals were also undertaken, and the reference lists of all included studies or relevant systematic reviews were checked for further studies.Investigators known to be carrying out research in this area were also contacted for unpublished data or knowledge of the grey literature.

Data Collection and Analysis
Titles of interest were reviewed by abstract.Potentially significant papers were then obtained in full.Where the relevance of a study was unclear, a consensus was reached by the authors regarding the applicability of the participant group and reported outcome measures.Data were extracted using a pre-designed extraction form.The outcome measures of interest included patient-reported outcomes (provided using validated self-assessment tools) and clinical outcome measures.Information was also extracted on recruitment rates, attrition, adherence to the study protocol, adverse events, survival rates, and key conclusions from each study.

Assessment of Methodologic Quality of Included Reviews
Risk of bias was assessed using the Cochrane Collaboration's tool for assessing risk of bias for randomized controlled trials 19 and the Critical Appraisal Skills Program: Cohort Studies methodology checklist for single-cohort studies 20 .Both of those tools consider potential biases in recruitment, measurement, and reporting of study outcomes.

Fatigue
Fatigue was a reported outcome in three of the five included studies.Using the validated outcome measurement tools Functional Assessment of Cancer Therapy-Lung 37,41 or Functional Assessment of Chronic Illness-Fatigue 37 , Temel et al. 37 and Quist et al. 41 found no statistically significant changes in self-reported fatigue in participants who completed a physical activity intervention.The study by van der Meij et al. 39 also found no significant differences in fatigue as assessed within the European Organisation for Research and Treatment of Cancer 30-item quality-of-life questionnaire (eortc qlq-C30: p = 0.57 at week 3, p = 0.95 at week 5).

3.2.2
Appetite Appetite was a reported outcome in one of the five included studies.No significant differences in appetite as assessed within the eortc qlq-C30 were found in the study by van der Meij 39 .Appetite was not a formal outcome in the study by Tozer et al. 36 , but those authors reported that appetite deteriorated significantly (p < 0.05) in participants shortly before death.

3.2.3
Unintentional Weight Loss Changes in total weight and lean mass were reported in three of the five included studies.
Murphy et al. 40 found that most participants receiving an eicosapentaenoic acid (epa) intervention supplement gained or maintained weight and muscle, and improved the quality of their muscle through loss of intermuscular adipose tissue deposits.The improvement was significantly different ( p < 0.05) from that in the standard-of-care control group.
Tozer et al. 36 found significant mean changes (p < 0.05) in the percentage change of body weight in an intervention group treated with cysteine-rich
In contrast, van der Meij et al. 38 found no significant differences in body weight change between groups receiving an active epa-containing intervention and a control supplement.Fat-free mass as determined by bioelectrical impedance declined in both groups, but a statistically larger loss of muscle was observed at 5 weeks in the control group ( p < 0.05).

3.2.4
Physical Performance Physical performance measures were reported in four of the five included studies.
Temel et al. 37 reported that participants who completed baseline and post-study assessments increased distance walked in 6 minutes and muscle strength, but statistical significance ( p < 0.05) was found only for change in elbow extension, which would indicate increasing power of the triceps brachii.
In an intervention group receiving a cysteine-rich protein supplement, Tozer et al. 36 found a significant difference (p < 0.05) in hand-grip force from baseline to 6 months and at the last measurement taken more than 17 days before death.That improvement was not replicated in a group receiving a casein-based control supplement.
Quist et al. 41 found an increase in 6-minute walk distance and 1-repetition-maximum weight lift tests in their study completers ( p < 0.05), indicating improvements in both exercise capacity and muscle strength.
Van der Meij et al. 39 found no significant differences in the physical performance of their intervention and control groups as assessed by hand-grip dynamometry and an accelerometer worn at the hip.Notably, the group receiving the intervention supplement containing epa tended to be more physically active.

3.2.5
Quality of Life Quality of life (qol) was a reported outcome in three of the five included studies.
Temel et al. 37 and Quist et al. 41 reported no statistically significant changes in qol in study participants from baseline to post-assessment.However, lung cancer symptoms significantly improved ( p < 0.05) in the trial by Tozer et al. 36 over the course of the intervention, as measured by that subscale on the Functional Assessment of Cancer Therapy-Lung.Using the eortc qlq-C30, van der Meij et al. 39 reported significantly higher global qol, better social functioning, less nausea and vomiting, fewer financial concerns (p < 0.05), and better physical and cognitive function (p < 0.01) in their intervention group than in their control group.

Recruitment, Attrition, and Adherence to Study Protocol
Low recruitment rates, attrition, and poor adherence to study protocol were reported as major issues in all five of the included studies (see Table i).An increase in plasma fatty acids was reported by van der Meij et al. 39 in some control participants, indicative of against-protocol fish-oil supplementation.

3.2.7
Adverse Events No serious adverse events were recorded for any of the included studies.Tozer et al. 40 reported incidences of mild gastrointestinal symptoms thought to be related to the increased protein ingestion in both the intervention and the control group.

3.2.8
Survival Survival was a reported outcome in two of the five included studies.The median survival of participants in the Temel et al. 37 study cohort was 12.98 months, which those authors deemed to be consistent with previous estimates of survival for patients with metastatic lung cancer.Tozer et al. 36   statistically nonsignificant, but positive trend for survival in the intervention group ( p = 0.058), with more participants in the intervention group being alive at 6 months, an observation that they suggested might merit further study.

Summary of Main Results
The aim of the present paper was to review trials of physical activity or nutrition interventions (or both) focusing on the management of fatigue, anorexia, and unintentional weight loss (symptoms of cancer cachexia) in patients with advanced nsclc, and also to evaluate the effectiveness of the interventions trialled.Despite an extensive search strategy, only six papers met the inclusion criteria.The included papers detailed five trials with 203 participants.All of the included studies had short intervention and follow-up times, except for the nutrition study undertaken by Tozer et al. 36 .Shorter studies benefited from reduced attrition rates, but they also prevented the drawing of any conclusions about the long-term effects of the intervention 42

Physical Activity Interventions
The physical activity interventions within the present systematic review 37,41 showed that moderateintensity physical activity interventions were not detrimental to qol in advanced nsclc.Also, some indications of improvement in emotional wellbeing 41 and lung cancer symptoms 37 were observed when participants adequately adhered to the intervention guidance.
The beneficial effects of physical activity for cancer survivors have been well established 12,43 .A recent Cochrane systematic review concluded that, compared with usual care or low-intensity activity interventions, moderate-intensity exercise may have physical, psychosocial, and spiritual benefits for cancer patients receiving cancer treatment 44 .
In a cross-sectional study of patients receiving palliative care at a regional cancer centre in Canada from November 2006 to May 2007 45 , higher qol scores were self-reported by physically active patients than by those who were sedentary, even when activity levels were significantly below those recommended for the general population.Cancer patients who are more physically able are less likely to have treatment resistant-disease 46 and to experience increased life expectancy [46][47][48] .
Findings from our systematic review add to the growing body of evidence that promotion of activity is justified, even in the late stages of nsclc 21,49,50 .
A qualitative study of 20 people with advanced nsclc in the United States found that symptoms such as fatigue, nausea, malaise, and intolerance to cold, coupled with a lack of specific activity guidance from health care professionals and a fear of exercising unsupervised were all significant barriers to increasing or maintaining physical activity 51 .It is interesting to note that, regardless of tumour stage and functional ability, patients with advanced nsclc have been found to be more likely to engage with and to tolerate moderate-to highlevel hospital-based prescribed exercise interventions when they are referred earlier in the course of their cancer treatment 52 .

Nutrition
The studies included in the present systematic review provided some evidence of beneficial effects from the provision of nutrition support in advanced nsclc.The nutrition interventions used were a cysteinerich protein supplement 36 , epa 40 , and a high-protein energy-dense supplement containing omega-3 polyunsaturated fatty acids 38 .Reported benefits included maintenance of weight and muscle mass during active cancer treatment 36,53 and improvements in self-reported measures of qol 39 .
Those benefits were not routinely demonstrated across all studies.Ensuring macro-and micronutrient sufficiency is a vital component of the multimodal active management of cancer cachexia 54,55 .Although nutrition assessment and counselling are recommended for all weight-losing cancer patients 56 , those approaches were absent in all of the included studies.Two of the studies used fish-oil supplementation either alone 40 or as part of a more complete nutritional supplement 38,39 .People with advanced cancer are often found to be fatty-acid-deficient, and that deficiency is strongly linked to decreased skeletal muscle mass 57 .Alterations in food preferences and dietary habits are commonly noted in advanced cancer and may exacerbate nutrient insufficiencies 54 .
Obesity before diagnosis can be of prognostic advantage in advanced nsclc 4 , perhaps because of greater lean-mass stores for the body to use 56 .Weight gain through nutritional supplementation 22,58 or appetite stimulation 59 have not been shown to have similar survival benefits.A recent systematic review (13 studies with 1414 participants) compared oral-nutrition interventions against standard care for malnourished patients receiving curative or palliative treatment for any cancer diagnosis 58 .Conclusions were limited because of study heterogeneity, but the authors stated that, although oral-nutrition supplementation increased dietary intake and improved some qol indices such as poor appetite or global qol scores, there was no evidence that nutrition interventions alone can improve survival rates.In the absence of sufficient anabolic drive, additional energy consumed by patients with cancer cachexia syndrome appears to be preferentially stored as fat mass, increasing the metabolic demands imposed on bodily systems and worsening prognosis 3,60 .

Current OnCOlOgy-VOlume 20, number 4, August 2013
Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).

Completeness and Applicability of Evidence
None of the included studies combined advice with respect to both nutrition management and physical activity.That observation is relevant because leantissue anabolism requires sufficiency in both dietary intake and contractile activity 61,62 .
Recruitment into nutrition or physical activity intervention studies in advanced cancer is low and attrition is high.Withdrawal and drop-out rates often leave very small samples from which to determine any significance of findings.Study recruitment is likely to be influenced not only by the issues that affect all palliative care trials, such as participant identification and heterogeneity 63,64 , but also by issues specific to exercise engagement or nutritional supplementation and palliative rehabilitation 65 .The strict criteria for entrance into trials may also be a significant bias.Often, the most unwell people are excluded from studies, making results less applicable to the population as a whole.Interventions that aim to stem weight loss often exclude those for whom the greatest weight loss has already occurred.The new definitions and staging guidance for cancer cachexia 6 have led to calls for researchers to consider more carefully suitability and optimal timing of cachexia interventions for people with cancer 18 .It is hoped that the new criteria proposed by international cancer cachexia experts 6 will better define optimal exclusion and inclusion criteria for active interventions.
Positive psychological effects have been found to occur when patients with cancer feel that something rather than nothing is being done to manage their disease 66,67

Quality of the Evidence
The results of our review must be interpreted with caution because of the high risk of bias across the included studies (Table ii).Studies of interventions relating to physical activity and nutrition pose many inherent risks of bias that are not easily controlled for.It is frequently impossible to blind participants to treatment intent, especially where no placebo is available or when the control intervention is standard care 63 .Advising key stakeholders and potential participants of the study hypothesis, a requirement of research ethics and governance, can also introduce bias through contamination of the control group 42,63 .The timing of research studies for cachexia symptom management has also attracted criticism, because such studies often occur during the window of expected gain from palliative anticancer therapies 68 .It is also possible that benefits observed in non-controlled studies may arise purely as a byproduct of increased monitoring and psychosocial support 69 .

Implications for Practice
The present systematic review suggests that exercise and nutrition interventions are not harmful and may have beneficial effects for unintentional weight loss, physical strength, and functional performance in patients with advanced nsclc.Such improvements must be interpreted with caution, however, because findings were not consistent across the included studies, which were small and at significant risk of bias.The lack of improvement in fatigue scores for all of the interventions is interesting.Improvements in cancer-related fatigue in advanced cancer may be masked through tiredness related to increased exertion.The masking may be particularly pronounced when the outcome measurement is taken immediately after an active physical activity intervention that lacks longer-term follow-up.Pedometers and exercise diaries might be a helpful way of demonstrating gains in function and autonomy where a level of tiredness persists 45 .

Implications for Research
More research is required to ascertain optimal physical activity and nutrition interventions in advanced inoperable nsclc.Specifically, the potential benefits of combining physical activity and nutrition counselling have yet to be adequately explored within this population.Outcome measures for assessing interventions in early-stage cancer or in cancer survivors are often inappropriate in advanced cancer, in which progressive functional decline is inevitable.It is vital that researchers separately report outcome measures in a subgroup analysis for participants with advanced illness, even if the findings are statistically nonsignificant.Adopting uniform reporting mechanisms for outcome measures of fatigue and weight loss would also provide an opportunity for meta-analyses of smaller studies 70 .

ACKNOWLEDGMENTS
We thank the experts who responded to requests for information on their research.The research reported here was funded by the All-Ireland Institute of Hospice and Palliative Care (AIIHPC) and the HSC R&D Division, Public Health Agency, Northern Ireland.AIIHPC is an all-island organization comprising a consortium of hospices and universities, all working to improve the experience of supportive, palliative, and end-of-life care on the island of Ireland by enhancing the capacity to develop knowledge, promote

Figure 1
Figure 1 Search strategy epa = eicosapentaenoic acid; nsclc = non-small-cell lung cancer; ecog = Eastern Cooperative Oncology Group; fact-L = Functional Assessment of Cancer Therapy-Lung; sclc-ed = small-cell lung cancer, extensive disease; who = World Health Organization; dha = docosahexaenoic acid; eortc qlq-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire.Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).
To determine if loss of body weight and body cell mass in advanced frail cancer patients with relatively poor prognosis could be ameliorated and if quality of life and functional performance could be improved with administration of highcysteine whey-derived protein formulation compared with casein Study intervention details Participants instructed to ingest 3 scoops (3×10 g) of cysteine-rich protein daily.The study medication was administered in conjunction with standard of care for cancer type and stage.Compared with the control group, the intervention group reported significantly better quality of life and social functioning, less nausea and vomiting, fewer financial concerns (p<0.05), and better physical and cognitive function (p<0.01) on eortc-qlq-C30 No significant differences between groups in handgrip strength or spontaneous activity Conclusions Key conclusions of study authors Study suggests beneficial effects on quality of life and spontaneous physical activity of a nutritional supplement containing epa Other comments Significant sex discrepancy between intervention and control groups Compared with participants who completed the intervention, those who dropped out early had more weight loss at baseline Planned recruitment numbers not achieved Levels of plasma phospholipids suggestive of epa consumption against protocol figure 1 Search strategy.e324 Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).Percentage change in body weight and body cell mass over 6-month period Secondary outcomes Hand grip strength Karnofsky performance status, McGill Quality of Life questionnaire, Edmonton Symptom Assessment System Mortality, biochemical markers, and disease status e325 Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).iiib nsclc with pleural or pericardial effusions, or stage iv Cancer treatment received Palliative chemotherapy, radiation during or followed by chemotherapy, or radiation alone Inclusion criteria Within 12 weeks of diagnosis of advanced nsclc confirmed by histology or cytology Exclusion criteria Unstable cardiac disease Baseline anemia Untreated bone or brain metastases preventing participation e326 Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).e327 Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).e328 Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).e329 Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).e330 Current OnCOlOgy-VOlume 20, number 4, August 2013 Copyright © 2013 Multimed Inc.Following publication in Current Oncology, the full text of each article is available immediately and archived in PubMed Central (PMC).
found a , but if interventions are too burdensome, then significant attrition and poor adherence are likely.In essence, what is needed are appropriately timed, individually tailored interventions cognizant of individual's enablers and barriers to engagement 51 .