Table 2.

Main characteristics of the studies included in the meta-analysis.

Study	Subjectsa	Age (years)b	Type of study	Intervention	Study design	Duration of intervention	Main outcomes	Other outcomes
Oral supplementation
Carvalho et al. (17)	(I) 29 (4) (C) 24 (7) Oral cavity cancer	(I) 53.3 ± 8.8 (C) 57.3 ± 9.1	Randomized, controlled clinical trial	Hypercaloric and hyperproteic supplement with 2 g EPA/440 mL	135 g/day of hypercaloric and hyperproteic supplement	4 weeks	The supplementation was not able to promote significant changes in the inflammatory profile	The EPA group presented 50% less likelihood of nutritional risk according to the CRP/Albumin ratio, despite not having shown any statistical difference.
Faber et al. (18)	(I) 24 (7) (C) 23 (7) Adenocarcinoma or squamous carcinoma in the esophagus or gastroesophageal junction	(I) 61.1 ± 9.2 (C) 61.6 ± 9.4	Explorative, randomized, controlled, double-blind study	400 mL/day.652 kcal, 39.6 g ptn, 2.4 g EPA, 1.2 g DHA, 4.8 g GOS, 0.8 g FOS and a balanced mix of vitamins, minerals	400 mL per day, 0–5% WL, a non-caloric Placebo product ≥5% WL, isocaloric standard nutritional product	4 weeks	A significant increase in body weight and an improved performance status in patients who received the nutritional intervention with EPA/DHA, which is high in protein and leucine.	There was a significantly higher decrease in the ratio n-6/n-3 compared to the control group.
Liu et al. (19)	(I) 11 (C) 11 Gastric cachexia cancer	(I) 56 (49–75)3 (C) 58 (45–75)3	Randomized study	3.6 g of n-3/day	6 mL of atractylenolideI (ATR)/day	6 weeks	N-3 group showed lower serum values of IL-6. ATR was more effective than FOE in improving appetite and, Karnofsky performance.	ATR decreased the proteolysis- inducing factor in urine.
Persson et al. (20)	(I) 13 (6) (C) 11 (4) Advanced gastrointestinal cancer	(I) 66 ± 9 (C) 69 ± 10	One-center, randomized, non–placebo-controlled, open study	30 mL/day of FO mixture - 4.9 g of EPA and 3.2 g of DHA	18 mg of Melatonin (MLT)/day	4 weeks	FO, MLT did not demonstrate anti-inflammatory effect.	No differences were observed in serum albumin, CRP, TNF-α, IL-1β, soluble IL-2 receptor, IL-6, IL-8 and plasma fibrin.
Enteral nutritional supplementation
Solís-Martínez et al. (21)	(I) 32 (14) (C) 32 (15) HNC squamous cell cancer in cancer treatment	(I) 60 ± 14 (C) 58 ± 14	A randomized single-blind placebo-controlled	A high-protein supplement with 2 g of EPA per day (600 kcal, 40 g of ptn)	A high-protein supplement with 24 g calcium caseinate per day (596 kcal, 40 g of ptn)	6 weeks	EPA supplement was associated with BW and LBM stabilization.	There was a significant increase in IL-8 levels and decreased of fatigue.
Yeh et al. (22)	(I) 31 (1) (C) 37 (0) HNC in cancer treatment	(I) 54.1 ± 9.3 (C) 54.4 ± 9.8	A randomized, prospective, clinical trial	An energy dense oral nutritional supplement with 7.1 g of n-3 and glutamine, probiotics and vitamins.	Isocaloric nutrition formulation	8 weeks	The supplementation improved BW, serum albumin and, prealbumin levels in patients with BMI <19.	Severe diarrhea events reported in the intervention group could be related to the higher osmolarity and different fat content of the formula.

^aFemale gender is in brackets.

^bValues are mean ± SD unless otherwise specified.

I, intervention group; C, Counterpart group; EPA, eicosapentanoic acid; DHA, docosahexaenoic acid; CRP, C-reactive protein; HNC, head and neck cancer; Ptn, protein; GOS, galacto-oligosaccharides; FOS, fructo-oligosaccharides; WL, weight loss; ATR, atractylenolide; FO, fish oil; MLT, melatonin; BW, body weight; LBM, lean body mass; BMI, body mass index.

²Completed patients.

³Point estimation 95%CI.

⁴Median (range).

⁵Median (IQR).