Following a request from the European Commission, the Panel on Dietetic Products, Nutrition and Allergies was asked for an opinion on the safety of synthetic trans-resveratrol as a novel food in the context of Regulation (EC) No 258/97, taking into account the comments and objections of a scientific nature raised by Member States.
The Panel considers that the information provided on the composition and specifications of the Novel food is sufficient.
The novel food with a purity of at least 99% trans-resveratrol is produced by a chemical process, consisting of a Wittig reaction of 3,5-dimethoxybenzylphosphonate ester with 4-methoxy-benzaldehyde leading to trimethoxyresveratrol. This intermediate is purified and then dealkylated with aluminium chloride and diisopropylamine to yield resveratrol which is crystallised from isopropanol/water and recrystallised from ethanol/water. The Panel concludes that the production process is sufficiently described and does not raise safety concerns.
The applicant intends to market the novel food as an ingredient for food supplements in capsule or tablet form at daily doses up to 150 mg/day. The target population is adults. Resveratrol occurs naturally in grapes, grape juice and wine with low amounts found in peanuts, pistachios, and in blueberries. The Panel notes that the intended maximum supplemental intake of 150 mg/day value is about 50 times higher than the high percentile background intake from foods.
The Panel considers that resveratrol does not have a nutritionally relevant role in the human diet and that the consumption of the novel food ingredient is not nutritionally disadvantageous.
A Salmonella Typhimurium reverse mutation assay compliant with good laboratory practice (GLP) and OECD guideline 471 provided negative results. An in vitro chromosome aberration study with trans-resveratrol using human lymphocytes was conducted according to OECD guideline 473 and in compliance with GLP. There were significant increases in structural chromosome aberrations at doses of 10, 20 and 30 μg/ml without activation and at 40 and 50 μg/ml with activation. An in vivo mammalian erythrocyte micronucleus test was conducted in compliance with GLP and according to OECD guideline 474 using Sprague–Dawley rats given 0, 500, 1,000 or 2,000 mg/kg body weight (bw) per day trans-resveratrol for 2 consecutive days by gavage. There was no change in the proportion of immature erythrocytes. There was no increase in micronucleated erythrocytes at any dose of resveratrol thus the test indicates the absence of clastogenic activity in vivo. In accordance with the European Food Safety Authority (EFSA) Scientific opinion on genotoxicity testing strategies (EFSA, 2011), the Panel considers that the negative in vivo genotoxicity assay is sufficient to rule out the concern based on the positive in vitro chromosomal aberration tests considering the endpoints structural and numerical chromosome aberrations (clastogenicity and aneugenicity).
In a 90-day study conducted on synthetic trans-resveratrol produced by the applicant, Wistar rats received diets containing trans-resveratrol at doses of 0, 120, 300 or 750 mg/kg bw per day. The study was performed in compliance with GLP and was based on OECD guideline 408. Body weight and body weight gain of high-dose animals were lower than that of controls throughout the study for males and from week 4 for females, and remained lower during the recovery period. The reductions were approximately 10% but were not always statistically significant. Body weight gain of the intermediate group was slightly lower than controls for both sexes and body weight of females from this group was lower from week 4. The reduced body weight at the mid and highest dose was associated with reduced food consumption. Food consumption returned to control levels during the recovery period. No effect was observed on the body weight and food consumption of the 120 mg/kg bw per day treatment group. A benchmark dose approach provided a BMDL05 of 344 mg/kg bw per day derived from body weight data of female rats.
Chronic toxicity/carcinogenicity studies performed in accordance with OECD Guidance 451 and 452 or 453 were not carried out on trans-resveratrol. Despite the limitations of a provided study with a p53-knockout mouse model for carcinogenicity testing (i.e. short duration, not generally accepted as an alternative to carcinogenicity testing in accordance with OECD Guidance), the Panel notes that this study does not indicate a carcinogenic effect.
In a GLP-compliant embryo–fetal toxicity study with trans-resveratrol, fetal examination for soft tissue and skeletal abnormalities showed no treatment-related effects, thus the no observed adverse effect (NOAEL) for fetal development was at the highest dose (750 mg/kg bw per day) tested. The only maternal effect observed was reduction in body weight, however, this occurred at all doses. The Panel considers 120 mg/kg bw for maternal toxicity as a lowest-observed-effect level (LOEL). The Panel considers that the studies provided do not raise concerns regarding developmental toxicity.
Member States expressed concern that trans-resveratrol has potential for oestrogenic activity, due to its stilbene structure and structural similarity to the potent oestrogen diethylstilbestrol. In an in vitro study, resveratrol was found to bind to oestrogen receptors and to act as a weak oestrogen antagonist in human breast cancer cell lines. In another experiment, resveratrol induced the expression of progesterone receptor in a mouse mammary organ culture model, when administered alone, but expression was suppressed in the presence of estradiol (1 nM). In another in vitro study, it was shown that resveratrol binds to human oestrogen receptor and stimulate MCF-7 cell proliferation in a dose-dependent manner at concentrations ranging from 10 nM to 10 mM. No signs for an oestrogenic effect were found in the histological examination of ovaries and in the organ weight measurements of ovaries, testes and uterus in rats receiving resveratrol at doses up to 1,000 mg/kg per day for 3 months. When administered to female weanling rats, trans-resveratrol at dose levels up to 100 μg did not exhibit oestrogenic effects as indicated by uterine weight and epithelial cell height of the uterus. In a second experiment of this study, 1,000 μg trans-resveratrol administered for 6 days antagonised the E2-induced cholesterol lowering effect in weaning rats. In two uterotrophic assays with trans-resveratrol, no oestrogenic effects were found. The Panel considers that these data do not indicate concerns with respect to oestrogenic activity of resveratrol in vivo.
The dossier of this application contains 23 articles on human studies and one review on resveratrol which covered additional 14 human studies.
The Panel considers that no conclusions with regards to the safety of resveratrol at the intended dose levels can be drawn on fourteen pharmacokinetic studies and clinical studies with a maximum duration of 96 h, on eight studies with dose levels of below 40 mg/day, on three studies with cancer patients and on four studies investigating supposed beneficial effects.
Regarding the other studies, the Panel notes that diarrhoea or other gastrointestinal symptoms were reported in four uncontrolled intervention studies at doses of 1 g resveratrol/day or higher, corresponding to about 14 mg/kg bw per day, which is by a factor of 10 lower than the LOEL of the animal studies. No treatment-related effects were reported for doses of below 1 g/day in studies up to 3 months of duration. Taking into account the limitations of the available human studies, i.e. low number of subjects, short duration, limited safety-relevant endpoints studied, no study reports available, the Panel considers that these studies per se do not provide sufficient evidence of safety. Considering the weight of evidence, the Panel concludes that the intended intake level of 150 mg/day for adults does not raise safety concerns.
The Panel considers that the metabolite trans-resveratrol sulfate could inhibit CYP enzymes in humans and may interact with medicines which are mainly metabolised by CYP2C9.
The Panel concludes that the NF, synthetic trans-resveratrol is safe under the proposed conditions of use.