The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review

Introduction Existing guidelines on nutrition support in patients with cancer cachexia state limited evidence for the beneficial effects of n-3 polyunsaturated fatty acids (PUFAs) on clinical outcome. In order to report on the latest evidence for n-3 PUFAs in cancer cachexia, we conducted a systematic literature review of randomized controlled trials (RCTs), comparing the effects on clinical outcome parameters of oral or enteral supplementation of n-3 PUFAs in cancer patients receiving chemotherapy, radiotherapy, surgery or palliative care. Materials and methods In PubMed, EMBASE and the Cochrane Library, search terms on cancer, n-3 PUFAs and clinical outcome parameters (nutritional status, morbidity, mortality and quality of life) were entered on 1 April 2013, using limits for adults, humans and English language. The quality and evidence of the retrieved publications were appraised by an expert team of Australian and Dutch dieticians and nutritionists, using the ADA grading system. Fifteen RCTs were retrieved. Results Nine RCTs were of positive quality, five of neutral quality and one of negative quality and were performed in patients with various types of cancer. Fair evidence shows that supplementation of n-3 PUFAs appears to be safe and may improve the quality of life and physical activity in patients with cancer. However, supplementation of n-3 PUFAs does not improve energy or protein intake, appetite or survival and does not reduce postoperative complications. The evidence for the effect on body weight, fatfree mass and performance status remains inconclusive. Conclusion Supplementation of n-3 PUFAs may have some positive effects in patients with cancer. Introduction Cancer cachexia, a complex metabolic syndrome associated with underlying illness, characterized by an increased inflammatory status and loss of muscle mass with or without loss of fat mass, is highly prevalent among patients with cancer1–7. This syndrome is a result of complex alterations in carbohydrate, lipid and protein metabolism8,9, caused by inflammatory mediators such as cytokines and tumour-derived catabolic drivers10. Proteolysis-inducing factor (PIF) is produced by the tumour and induces protein catabolism11. As a result of the acute-phase response, the liver shows an increased protein turnover for the production of inflammatory mediators, using muscle mass to release amino acids9,10. Changes in lipid metabolism in cancer include a reduction of lipogenesis, with unchanged whole-body lipolysis and mobilization of fatty acids from fat tissue. Alterations in glucose metabolism are reflected by glucose intolerance and insulin resistance8,9,11. Thus far, conventional nutritional support has been limited in its ability to stabilize body weight and maintain fat-free mass in patients with cachexia. Pharmaceutical interventions sometimes improved appetite, body weight and quality of life, but weight gain mostly consisted of fat mass12–15. n-3 Polyunsaturated fatty acids (PUFAs), especially eicosapentaenoic acid (EPA), seem to be promising agents to treat cancer cachexia. A dose of around 2 g of EPA per day (alone or combined with docosahexaenoic acid, DHA) appears to decrease the production of proinflammatory cytokines16,17 and PIF and is associated with stabilization of body weight and probably fat-free mass13,18. This has been shown in animal studies and in nonrandomized human trials in pancreatic cancer patients19,20. However, randomized controlled trials (RCTs) show contradictory results21–26. This may be due to issues related to study limitations, such as the disease severity, confounding factors and nonadherence with n-3 PUFA supplements. Also, study designs and outcome parameters differ in terms of supplementation dosage, * Corresponding author Email: b.vandermeij@vumc.nl 1 Department of Nutrition and Dietetics, Internal Medicine, VU University Medical Center Amsterdam, the Netherlands 2 Centre for Dietetics Research, School of Human Movement Studies, University of Queensland, Brisbane, Australia 3 Department of Nutrition and Dietetics, Royal Brisbane & Women’s Hospital, Brisbane, Australia 4 Department of Nutrition and Dietetic Services, Princess Alexandra Hospital, Brisbane, Australia 5 Department of Surgery, VU University Medical Center Amsterdam, the Netherlands


Introduction
Cancer cachexia, a complex metabolic syndrome associated with underlying illness, characterized by an increased inflammatory status and loss of muscle mass with or without loss of fat mass, is highly prevalent among patients with cancer [1][2][3][4][5][6][7] .This syndrome is a result of complex alterations in carbohydrate, lipid and protein metabolism 8,9 , caused by inflammatory mediators such as cytokines and tumour-derived catabolic drivers 10 .Proteolysis-inducing factor (PIF) is produced by the tumour and induces protein catabolism 11 .As a result of the acute-phase response, the liver shows an increased protein turnover for the production of inflammatory mediators, using muscle mass to release amino acids 9,10 .
Changes in lipid metabolism in cancer include a reduction of lipogenesis, with unchanged whole-body lipolysis and mobilization of fatty acids from fat tissue.Alterations in glucose metabolism are reflected by glucose intolerance and insulin resistance 8,9,11 .
Thus far, conventional nutritional support has been limited in its ability to stabilize body weight and maintain fat-free mass in patients with cachexia.Pharmaceutical interventions sometimes improved appetite, body weight and quality of life, but weight gain mostly consisted of fat mass [12][13][14][15] .
n-3 Polyunsaturated fatty acids (PUFAs), especially eicosapentaenoic acid (EPA), seem to be promising agents to treat cancer cachexia.A dose of around 2 g of EPA per day (alone or combined with docosahexaenoic acid, DHA) appears to decrease the production of proinflammatory cytokines 16,17 and PIF and is associated with stabilization of body weight and probably fat-free mass 13,18 .This has been shown in animal studies and in nonrandomized human trials in pancreatic cancer patients 19,20 .However, randomized controlled trials (RCTs) show contradictory results [21][22][23][24][25][26] .This may be due to issues related to study limitations, such as the disease severity, confounding factors and nonadherence with n-3 PUFA supplements.Also, study designs and outcome parameters differ in terms of supplementation dosage, scientific soundness, most importantly on potential selection bias, randomization and blinding procedures and the validity and reliability of outcome parameters.Study quality was rated as positive (+), neutral (∅) or negative (-).No authors reviewed their own papers.
BvdM extracted data of individual studies, using a data extraction form, which was composed by using templates from the Cochrane association and the ADA.The data extraction form contained all fields that were considered relevant to address our study aim, such as study eligibility, reason for exclusion, quality rating, details on study design, results per outcome parameter, statistical methods, author's conclusions and reviewer's comments.
Following the quality appraisals, all team members summarized and assessed the strength of the evidence per outcome parameter.The evidence summary was created by adding up the numbers of studies and number of patients reporting a beneficial, negative or no effect on an outcome parameter.The number of good, neutral and negative quality studies was also documented in this summary.Consecutively, the ADA additional levels of evidence and grades for recommendations for developers of guidelines were used to grade the evidence 42 .The ADA grading system for recommendations has been developed to assist guideline developers in assessing the entire body of evidence and indicating the strength of each guideline recommendation.According to the ADA system, RCTs and meta-analyses obtain a class ' A' evidence.The ADA grades of recommendation-I Good, II Fair, III Limited, IV Expert Opinion, V Not Assessable-were used to evaluate the quality of studies and consistency of findings across studies 42 .
Members of the expert team reached consensus on the strength of the evidence and the evidence statements for effects on outcome mass, morbidity, mortality, length of stay and quality of life) were used to select relevant publications (supplementary Table 1).Any oral or enteral administration of n-3 PUFAs was included: (fish oil) capsules, oral nutritional supplements (ONSs) or tube feeding containing n-3 PUFAs.Studies investigating multiple immuneenhancing compounds (e.g.arginine, glutamine, nucleotides and n-3 FAs) or studies with concurrent use of appetite stimulants were excluded.The literature search was limited to RCTs in adult human subjects, which were available in English.Nonplacebo-controlled studies and singlearm studies were excluded.In case a study was reported in more than one publication, the earliest publication reporting at least one relevant outcome variable was included.Two members of the expert committee (BvdM and MvB) assessed the eligibility of publications on n-3 PUFAs in cancer patients by reviewing the title and abstract and included eligible studies.Moreover, reference lists of included articles were reviewed to identify eligible studies, which were not retrieved by the literature search.

Study quality and strength of the evidence
BvdM composed an expert team of Australian and Dutch dieticians and nutritionists during an International Cancer Technology Transfer fellowship from the UICC in Brisbane, Australia.The aim of the expert team was to appraise the included studies and to compose evidence statements and recommendations to be used for clinical practice.
Two independent members of the expert team appraised the quality of individual studies using the Quality Criteria Checklist of the online accessible Evidence Analysis Library of the American Dietetic Association (ADA) 42 .The checklist includes four relevance questions that address applicability to dietetic practice and 10 validity questions that address comparison with control or other agents and outcome parameters.Body weight is the most frequently used primary outcome measure, but may be biased by fluid retention.Another important issue is the heterogeneity in the assessment of cachexia and weight loss in clinical studies.
From clinical and research perspectives, there is a need for an updated literature overview on the effectiveness of n-3 PUFAs in patients with cancer 27,28 .In the past, several reviews and nutrition guidelines have addressed the issue of the prescription of n-3 PUFAs in patients with cancer cachexia [29][30][31][32][33][34] .Overall, these guidelines conclude that there are some indications for the beneficial effects of n-3 PUFAs on body weight and physical function in cancer patients, but no effects on survival.The available reviews only included studies that were published before 2006 29,31,34 .We previously published a systematic review and included studies published until April 2011 35 .Since then, there have been several new studies 23,[36][37][38] and reviews published in this field [39][40][41] .Therefore, the aim of this systematic review is to report on the latest evidence for the effects of n-3 PU-FAs on clinical outcome parameters in patients with cancer and provide recommendations for use in clinical practice.

Patient characteristics
This systematic review of RCTs involved adult (≥ 18 years of age) patients with cancer.

Literature search
A literature search was performed on 1 April 2013, using three databases: PubMed  (start date 1948), EMBASE (start date 1986) and the Cochrane Library (start date 2005).Medical subject headings (MeSH or Emtree) and free text words for n-3 PUFAs, EPA, cancer and clinical outcome parameters (body weight, fat-free

Fish oil capsules
This paragraph describes the design of the seven RCTs supplementing n-3 PUFAs by fish oil capsules, of which the quality was graded to be positive 21,22,37 , neutral 36,47,48 and negative 38 (Table 2).A number of studies have been conducted in than 10% 43 .Two studies excluded either patients who reported ≥10% of unintentional weight loss over the previous 3 months 37 or patients with severe malnutrition, as assessed by the Subjective Global Assessment (SGA) tool for malnutrition 49 .One study did not report details on nutritional status 47 , and the remaining studies only reported on the percentage of patients who had malnutrition at baseline (8-50%) without selecting patients with severe weight loss.However, these studies used various methods to identify malnutrition (see Table 2) 22,23,26,45,46 .parameters.The team resolved conflicting appraisals by discussion and consensus.
Five studies included patients with unintentional, self-reported, weight loss of more than 5% 21,38,44,48 or more  Four small studies during chemotherapy have been described, of which three studies were nonblinded 36,38,47 and one was double blinded 37 .In a Japanese study, 16 patients undergoing chemotherapy and allogeneic bone marrow transplantation were randomized to receive oral EPA three times daily, or no capsules 47 .Another study randomized 23 patients with colorectal cancer.Intervention patients were offered four fish oil capsules per day (total 600 mg EPA + DHA) and the control group did not receive any supplements 36 .Bonatto and colleagues performed an 8-week intervention of 2 g fish oil per day in 38 patients receiving chemotherapy after cancer surgery; control patients did not receive any supplements 38 .Finocchiaro and colleagues studied 33 patients with advanced non-small cell lung cancer undergoing chemotherapy, randomized to receive four capsules (510 mg EPA and 340 mg DHA) or placebo capsules 37 .

Supplementary Table 1 Example of search strategy in
In these seven RCTs, loss to followup of patients varied from 7% to 50% [21, 22, 36, 37, 48] or was not reported 38,47 .Adherence with the fish oil capsules was established by plasma fatty acids or diaries 21,37,48 or not reported 22,36,38,47 .

Oral nutritional supplements containing n-3 PUFAs
We identified five RCTs investigating the effects of ONS containing n-3 PUFAs in patients with cancer, most of positive quality 24,[43][44][45] and one of negative quality 49 .
In a large double-blinded study, 200 weight-losing pancreatic cancer patients were randomized to receive n-3 PUFA-containing ONS or an isonitrogenous control ONS 24 .In a small subgroup (n = 24) of this study, Moses and colleagues investigated resting energy expenditure (REE) by indirect calorimetry as well as total energy expenditure (TEE) by doubly labelled water 22 .Two studies investigated n-3 PUFA-containing ONS day.The duration of this study was 2 weeks, which is too short to reach effects on nutritional parameters and quality of life 48 .In a large double-blinded study, 429 patients with gastrointestinal (GI) or lung cancer without anticancer treatment were randomized to receive 2 g or 4 g EPA by a diester or a placebo during 8 weeks 21 .patients who did not receive anticancer treatment.Gogos and colleagues randomized 64 patients with generalized solid tumour types to receive 18 fish oil or placebo capsules daily until death 22 .Bruera and colleagues randomized 91 advanced cancer patients who showed weight loss and anorexia, to receive 18 fish oil or placebo capsules per          interventions in patients with lung cancer: Guarcello and colleagues compared the effects of n-3 PUFAcontaining ONS with an isocaloric, isonitrogenous control ONS 43 .van der Meij and colleagues compared the effects of an n-3 PUFA-containing ONS with an isocaloric control ONS in a double-blinded RCT 45 .In addition, a very small (n = 13) non-blinded RCT of negative quality compared the effect of n-3 PUFA-containing ONS with dietary counselling in patients with stage IV colorectal cancer, who were not severely malnourished 49 .Again, variable losses to follow-up were reported: 18% 45 , 40.5% 24 and 50% 43 .The adherence with study supplements was >75% of the recommended dose in most studies 24,43,44,49 and about 50% in one study 45 .In some studies, plasma phospholipids EPA were assessed as an objective indicator of compliance with n-3 PUFA supplementation 24,44,45 .Limitations to the quality of two studies included not reporting the betweengroup comparisons 43,49 and not describing blinding and randomization methods 43 .

Enteral nutrition
Three studies investigated enteral nutrition containing n-3 PUFA around cancer surgery (Table 2).These included two large, high-quality double-blinded RCTs in patients undergoing oesophagogastric cancer surgery.The first trial administered n-3 PUFA-containing ONS 5 days preoperatively and n-3 PUFA-containing tube feeding during 21 days postoperatively 26 .In the second large trial, n-3 PUFA-containing enteral nutrition, or an isonitrogenous control enteral nutrition, was administered 7 days before and after surgery.A third group only received standard enteral nutrition during 7 days postoperatively 23 .A smaller study performed a relatively short (7 days) postoperative intervention with n-3 PUFAcontaining enteral nutrition in 50 patients with upper GI cancer 46 .
Dropout rates of 12% 23 , 24% 26 and 30% 46 were reported, and due to tolerance issues, enteral nutrition goals were not always reached in two studies 23,46 .One study did not specify blinding and randomization methods 46 ..One positive quality study found an improvement in appetite over time (30 days, though not after 60 days) in patients with lung cancer receiving n-3 PUFA ONS, but not in control patients (n = 46, Guarcello +) 43 .

Strength of the evidence
Recommendation n-3 PUFA supplementation does not improve appetite (grade II).

Recommendation
There is no evidence that n-3 PUFA supplementation influences energy and protein intake (grade II).

Body weight
Results from 12 studies show equivocal effects on stabilization or improvement of body weight after supplementation of n-3 PU-FAs in cancer patients: five studies did not find significant differences between n-3 and control groups (n = 570, Fearon 2003 + Moses +, Gogos +, Sultan +, Bruera ∅); one study observed a tendency for body weight maintenance in the n-3 PUFA group receiving 2 g EPA by diester emulsion, as compared with groups receiving 4 g EPA or a placebo emulsion (n = 518, Fearon 2006 +) 21 ; three studies observed a significant body weight maintenance in the n-3 PUFA group versus a control intervention (n = 91, van der Meij 2010 +, Trabal ∅, Bonatto -) 38,45,49 ; and three studies demonstrated a within-group weight maintenance over time in the n-3 PUFA group (n = 102, Silva ∅, Guarcello +, Finochiaro ∅) 36,37,43 .

Recommendation
The effects of n-3 PUFA supplementation on body weight maintenance are inconclusive (grade II).

Fat-free mass
Five studies measured the effect of an n-3 PUFA ONS on fat-free mass.Three studies did not observe difference for fat-free mass as compared with a control intervention (n = 742, Fearon 2006 +, Fearon 2003 +, Moses +) 21,24,44 .One study observed a better maintenance of fat-free mass in patients with lung cancer receiving n-3 ONS, as compared with a control ONS (n = 40, van der Meij 2010 +) 45 , and one study observed a maintenance of fat-free mass over time in patients receiving n-3 PUFA-containing ONS and enteral nutrition around oesophageal cancer surgery (n = 70, Ryan +) 26 .

Recommendation
The effects of n-3 PUFA supplementation on fat-free mass are inconclusive (grade II).

Recommendation
The effects of n-3 PUFA supplementation on Karnofsky performance status are inconclusive (grade II).

Quality of life
ONS containing n-3 PUFAs improved global health status, physical, cognitive, and social function compared with a control intervention in a small study (n = 40, van der Meij 2012 +) 50 ; another small study only observed a significant improvement in social function in the n-3 PUFA group, not in other quality-of-life parameters (n = 13, Trabal ∅) 49 ; one large study found a trend for improved physical function in patients receiving 2 g EPA by a diester emulsion, compared with 4 g EPA or a placebo (n = 518, Fearon 2006 +) 21 ; and one small study found reduced tiredness to be correlated with the n-3 PUFA dose (n = 87, Bruera ∅) 48 .A small study observed a within-group improvement of functional status and symptom scores over time in the n-3 PUFA group (n = 46, Guarcello +) 43 .A large positive trial did not find significant differences for quality-of-life parameters (n = 200, Fearon 2003 +) 24 .
Recommendation n-3 PUFA supplementation has modest beneficial effects on some aspects of quality of life (grade II).

Physical activity level
Two small studies investigated the effects of n-3 PUFA-containing ONS on parameters of physical activity (n = 64, Moses +, van der Meij 2012 +) 44,50 .
In one study in pancreatic cancer patients, TEE (measured by doubly labelled water) increased after 8 weeks in the group receiving the n-3 Competing interests: none declared.Conflict of interests: none declared.
All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript.All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.
nutrition were similar in n-3 PUFA and control groups (n = 315, Ryan +, Sultan +, Kenler ∅) 23,26,46 .One study reported minor GI complaints (no serious adverse events) in both n-3 and control groups (n = 70, Ryan +) 26 and another study observed slightly more GI complaints in the control group (P value not reported, n = 50, Kenler ∅) 46 .In one study, only 50% of the overall study population reached the aimed feeding rate owing to problems with tolerance and/ or complications, such as diarrhoea, ileus, nausea, vomiting or bloating (n = 195, Sultan +) 23 .
Recommendations n-3 PUFA supplementation by capsules or ONS appears to be safe to administer in cancer patients receiving chemo(radio)therapy or palliative care (grade II).
Intolerance to enteral nutrition around upper GI cancer surgery does not appear to be related to n-3 PUFAs (grade II).

Discussion
The authors have referenced some of their own studies in this systematic review.These referenced studies have been conducted in accordance with the Declaration of Helsinki (1964), and the protocols of these studies have been approved by the relevant ethics committees related to the institution in which they were performed.All human subjects, in these referenced studies, gave informed consent to participate in these studies.
This literature review gives an upto-date overview on the available evidence for the effects of n-3 PUFAs on clinical outcome parameters in patients with cancer.We conclude that fair evidence shows that supplementation of n-3 PUFA appears to be safe and may improve the quality of life and physical activity.However, supplementation of n-3 PUFA does not influence energy or protein intake, appetite, or survival in cancer (n = 120, Kenler ∅, Ryan +) 21,24 .One study found significantly longer survival in the group receiving n-3 PUFA capsules (n = 64, Gogos +) 22 .Another study demonstrated improved survival in patients who received n-3 PUFA capsules around BMT, but this study had very small numbers (n = 16, Takatsuka ∅) 47 .
Recommendation n-3 PUFA supplementation does not improve survival (grade II).

Safety and GI tolerance
Other important issues include the safety and tolerance of n-3 PUFA supplementation.Of the 15 retrieved studies, 10 reported on tolerance or adverse events during supplementation of n-3 PUFA.
Among the seven studies supplementing n-3 PUFA by capsules (n = 661), two studies observed GI side effects in the n-3 PUFA group slightly more frequently than in the control group (P value not reported) (n = 120, Bruera ∅, Finocchiaro +) 37,48 .Another study reported adverse events in n-3 PUFA and control groups; no serious adverse events were related to the study capsules (n = 518, Fearon 2006 +) 21 .Furthermore, in one small study, no side effects related to fish oil capsules were reported (n = 23, Silva ∅) 36 .
In studies using enteral nutrition in upper GI surgery patients, GI complaints and intolerance to the enteral PUFA-containing ONS, not in the control group 44 .The second study was performed in lung cancer patients.Physical activity assessed by an accelerometer was higher in patients with lung cancer receiving n-3 PUFA ONS during 5 weeks of chemoradiotherapy 50 .
Recommendation n-3 PUFA supplementation appears to have beneficial effects on physical activity in patients with cancer (grade III).

Complications
In one small neutral quality study, n-3 PUFA capsules reduced complication rate in bone marrow transplant (BMT) patients (n = 16, Takatsuka ∅) 47 .Three studies investigated the effects of a single n-3 PUFA intervention in the absence of multiple immune-enhancing compounds around cancer surgery.One study observed a significant reduction in infections in the n-3 PUFA group (n = 50, Kenler ∅) 46 , contrary to a large study that did not observe differences for infectious complications (n = 195, Sultan +) 23 .Furthermore, there were no effects of n-3 PUFAs (by ONSs and/or tube feeding) on major complications (n = 248, Ryan +, Sultan +) 23,26  provided comparable amounts of energy and protein and nutritional counselling in both intervention and control groups.As a result, these studies investigated the pure effect of n-3 PUFAs, provided that both groups consumed a comparable amount of n-3 PUFAs by normal foods.None of the studies described the consumption of fish or other foods containing n-3 PUFAs, such as walnuts containing alpha-linolenic acid.Alternatively, only a few studies checked the plasma phospholipid EPA concentration as a measure of n-3 PUFA intake.Study design is a limiting factor when addressing the evidence of n-3 PUFAs in cancer patients.The adherence to the study intervention was not always monitored; suboptimal adherence to n-3 PUFA supplements could have resulted in a lack of effect.Future trials should consider a choice of supplementation format (capsules or liquid), or the use of n-3 PUFA-containing enteral or parenteral nutrition, in order to improve patients' adherence.
The nutritional parameters also differ between studies.Body weight is widely used but is unreliable in case of ascites or oedema and gives no information on body composition.Even so, bioelectrical impedance analysis to measure fat-free mass might be unreliable in patients with cancer.CT image analysis or dual-energy x-ray absorptiometry is preferred as the method to precisely quantify skeletal muscle 41,58 .Ultimately, positive effects on body composition should be translated to improvements of quality of life.A few RCTs observed positive effects of n-3 PUFAs on both body weight and quality of life; the latter is more relevant to patients with cancer 43,45,49,50 .Still, we need more large RCTs confirming this relationship.
Conventional nutritional interventions for cancer patients have limited effects on clinical endpoints or the quality of life during chemotherapy 59,60 or palliative care 59 .If supplementation of n-3 PUFAs does chemotherapy.Supplements containing pure EPA did not cause chemotherapy resistance 56 .It is yet unclear if these metabolites are abundant in all n-3 PUFA supplements, and if the same resistance occurs in humans.As such, it might be oversimplified to discourage dietary n-3 PUFAs in patients with cancer 57 .When supplementation of n-3 PUFAs during chemotherapy is considered, it is recommended to use purified n-3 PUFA supplements instead of less refined whole fish oil or algae supplements, and to carefully assess patients chemotherapy outcomes in future trials investigating n-3 PUFA supplements 27 .
A limitation of this review is the heterogeneity of the 15 available studies.The trials differed in terms of patient populations, inclusion and exclusion criteria, intervention strategies and study durations.The administered dose ranged from 300 mg to 6 g of EPA, and in some studies, especially the studies with fish oil capsules, patient adherence with study supplements was not documented.In addition, the study populations are heterogeneous in terms of nutritional status and cachexia.Some studies selected patients on the degree of weight loss; others did not apply an inclusion criterion on nutritional status or cachexia.Some studies did not have adequate statistical power: around half of the studies were small (10 studies included 60 or fewer participants).The large studies had a high dropout rate (around 50%): this may also have biased the results.
Apart from immune-modulating effects of n-3 PUFAs in patients with cancer cachexia, additional energy and protein are required for body weight maintenance and synthesis of lean tissue.The effect of n-3 PU-FAs combined with additive energy and protein on nutritional status is expected to be larger than the effect of a single n-3 PUFA intervention.Apart from the study of Trabal and colleagues, all included studies patients.Effects on body weight, fatfree mass and performance status are equivocal: around half of studies find beneficial effects while others do not show differences compared with control interventions.Moreover, fair evidence shows that supplementation does not reduce postoperative complications.
These conclusions were drawn after a systematic qualitative evidence analysis of studies published until April 2013, carried out by an international expert team.In order to objectively evaluate the effects of n-3 PUFAs in cancer, a meta-analysis on this subject is required, but the variability in study designs makes it yet impossible to carry out metaanalyses.Also, some authors report between-group differences, whereas others only report within-group differences.
Thus far, conclusions from metaanalyses and systematic reviews on this topic were equivocal and did not strongly recommend supplementing n-3 PUFAs in patients with cancer 34,40,51  Supplementation of n-3 PUFAs appeared to be safe, but the way in which they interact with anticancer treatment needs to be further unravelled.EPA and DHA can inhibit tumour growth through apoptosis, inhibition of angiogenesis, and alterations to cell signalling 52,53 .An improved response to chemotherapy by n-3 PUFAs has also been observed in clinical studies in patients with lung 54 and breast 55 cancer.In most included RCTs, chemotherapy outcomes in patients who received n-3 PUFAs appeared to be comparable with control patients.On the contrary, findings from a study in mice found that platinum-induced PU-FAs, metabolites of n-3 PUFAs from algae and fish oil, Nutritional interventions are often offered in patients with nutritional issues and weight loss; prophylactic nutritional intervention is not standard.New insights suggest that early nutritional intervention in patients with early-stage cachexia could be more effective than interventions in patients with advanced cancer cachexia, but to our knowledge, few studies on this topic have been published [64][65][66] .Recently, a proposal for the definition of cancer cachexia has been published, and attention has been paid to the identification of patients with 'pre-cachexia' as a probable indication to start nutritional intervention 12,67 .However, definitions of cachexia and pre-cachexia need to be validated and further refined.

Conclusion
This review helps clinicians to decide on the prescription of n-3 PUFAs in patients with cancer cachexia.There is increasing evidence that n-3 PUFAs
Licensee OA Publishing London 2013.Creative Commons Attribution License (CC-BY) For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.
. These reviews included studies published until 2005/2006.The current systematic review includes publications until April 2013, in this way providing a new update.
were involved in the resistance to platinum-based T, Yu D, Reich RR, et al.Cancer cachexia: traditional therapies and novel molecular mechanism-based approaches to treatment.Curr Treat Options Oncol.2010 Dec;11(3-4):107-17.15.Baldwin C. Nutritional support for malnourished patients with cancer.Curr Opin Support Palliat Care.2011 Mar;5(1):29-36.16.Mayer K, Meyer S, Reinholz-Muhly M, Maus U, Merfels M, Lohmeyer J, et al.Short-time infusion of fish oil-based lipid emulsions, approved for parenteral nutrition, reduces monocyte proinflammatory cytokine generation and adhesive interaction with endothelium in humans.J Immunol.2003 Nov;71(9):4837-43.17. Barber MD, Ross JA, Preston T, Shenkin A, Fearon KC.Fish oil-enriched nutritional supplement attenuates progression of the acute-phase response in weightlosing patients with advanced pancreatic cancer.J Nutr. 1999 Jun;129(6):1120-5.18. Barber MD, Ross JA, Voss AC, Tisdale MJ, Fearon KC.The effect of an oral are safe to administer to cancer patients and can improve quality of life and physical activity.However the evidence for beneficial effects on body weight and fat-free mass remains inconsistent.Research should focus on early intervention and on multimodal treatment strategies, with attention to the adherence and feasibility of n-3 PUFA supplementation methods.improve cancer cachexia parameters, such as body weight and quality of life, the effect is relatively small.Therefore, more benefits are expected from combination treatments including anticatabolic and orexigenic agents and n-3 PUFAs 61 .In a five-arm RCT in patients with cachexia, the combination regimen that included the administration of an appetite stimulant, ONS with EPA and DHA, L-carnitine, and thalidomide was shown to be the most effective treatment in terms of lean body mass, REE and fatigue 62 .Another promising intervention to improve fat-free mass and physical function is physical exercise training with resistance training.Accumulating evidence suggests beneficial effects of physical exercise training during chemotherapy, but a combination intervention of training and nutritional counselling or nutritional supplementation has yet to be studied.Rogers and colleagues are currently conducting an open-label, prospective RCT on EPA, the COX-2 inhibitor celecoxib, combined with leucine supplementation and resistance training 63 .
Oral or enteral nutrition enteral* OR supplement* OR sip OR feed OR formula* OR liquid OR tube OR nasogastric OR nasojejunal OR nasoduodenal OR gastrostomy OR jejunostomy OR Enteral Nutrition" [MeSH] For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.

Figure 1: Flow chart of systematic literature search
For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.

Table 2 Summary of randomized controlled trials on the role of EPA in patients with cancer Author, year study design quality Aim Study population Intervention Outcomes Conclusions limitations
E: n = 30 18 g/ day fish oil capsules (max-EPA, 3.06 g EPA + 2.07 g DHA) C: n = 30 18 g/ day placebo capsules Four patients (6.7%) dropped out of the study because of poor adherence Energy intake-NA Protein intake-NA Weight-ns improvement FFM-NA Functional capacity-KPS after 40 days in malnourished patients receiving fish oil only Quality of life-NA Survival-increased in E group only (P < 0.025 compared with C) Other-no effect of fish oil on albumin or transferrin.No toxicity of fish oil except for mild abdominal discomfort and transient diarrhoea C: n = 9 no capsules Zero dropouts reported Energy intake-NA Protein intake-NA Weight-NA FFM-NA Functional capacity-NA Quality of life-NA Survival-higher survival rate in E group (E: n = 0 died vs. C: n = 5 died) (P < 0.01) Other-reduced complications in E group (E: n = 3 graft-vs.-hostdisease,n= 4 no complications vs. C: n = 6 graft-vs.-hostdisease,n= 4 thrombotic microangiopathy, n = 4 CMV disease) Licensee OA Publishing London 2013.Creative Commons Attribution License (CC-BY) For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.Competing interests: none declared.Conflict of interests: none declared.All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript.All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.

Table 2 (Continued)
Other-appetite, tiredness, nausea, well-being (ns).Same number of patients with gastrointestinal complaints in both groups (E: n = 5, C: n = 5) EPA: -0.1 vs. C (ns) Functional capacity-KPS, weakness (ns) Physical function and KPS-ns Quality of life-2 g EPA: ↑ 4.3 physical functioning; 4 g EPA: -3.4 (P = 0.04) Survival-2 g EPA: 155 days; 4 g EPA: 142 days; C: 140 days (ns) Other-appetite, albumin, CRP, nausea, vomiting, diarrhoea (ns), supplements were well tolerated, adverse events (ns), Licensee OA Publishing London 2013.Creative Commons Attribution License (CC-BY) For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.Competing interests: none declared.Conflict of interests: none declared.All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript.All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.

Table 2 (
Continued) Licensee OA Publishing London 2013.Creative Commons Attribution License (CC-BY) For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.Competing interests: none declared.Conflict of interests: none declared.All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript.All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.

Table 2 (
Continued) Five patients did not complete the study (E: n = 3, C: n = 2).Adherence: mean intake E: 1.9 cans/day, C: 1.5 cans/day (ns) Other-REE, TEE and PAL of C patients did not change significantly; TEE and PAL of E patients increased significantly, REE did not change; no significant differences between groups For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.Competing interests: none declared.Conflict of interests: none declared.All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript.All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.

Table 2 (
Continued) Licensee OA Publishing London 2013.Creative Commons Attribution License (CC-BY) For citation purposes: van der Meij BS, Bauer JD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.Competing interests: none declared.Conflict of interests: none declared.All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript.All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.
For citation purposes: van der Meij BS,BauerJD, Isenring EA, Brown T, Davidson WL, van Bokhorst MAE, et al.The effects of supplementation of n-3 polyunsaturated fatty acids on clinical outcome parameters in patients with cancer: a systematic review.OA Epidemiology 2013 Apr 19;1(1):2.Competing interests: none declared.Conflict of interests: none declared.All authors contributed to conception and design, manuscript preparation, read and approved the final manuscript.All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.