12 June 2008 - Following a request from the Commission, the Panel on Food Additives, Flavourings, Processing Aids and Materials in Contact with Food (AFC) was asked to deliver a scientific opinion on the safety in use of rosemary extracts when used as an antioxidant.
Rosemary extracts are derived from Rosmarinus officinalis L. and contain several compounds which have been proven to exert antioxidative functions. These compounds belong mainly to the classes of phenolic acids, flavonoids, diterpenoids and triterpenes.
The present opinion refers to rosemary extracts prepared using several solvent extraction techniques. These will be named by the acronyms provided by the petitioner and are as follows:
F62: rosemary extract produced from dried rosemary leaves by acetone extraction,
D74: rosemary extract prepared by extraction of dried rosemary leaves by means of supercritical carbon dioxide,
AR: rosemary extract prepared from a partially deodorised ethanolic extract of rosemary,
ARD: extract prepared from a deodorised ethanolic extract of rosemary,
RES: extract which is a decolourised and deodorised rosemary extract obtained by a two-step extraction using hexane and ethanol.
The principal antioxidative components of the extracts are the phenolic diterpenes carnosol and carnosic acid.
The Panel suggests to modify the specifications proposed by the petitioner to reflect the specifications of the extracts that were used in the safety testing, with respect to carnosol and carnosic acid and the antioxidant/volatile ratio.
Four of the five rosemary extracts considered in the present opinion, (D74, AR, ARD, and RES) were tested for genotoxicity. Several in vitro genotoxicity studies were performed in both prokaryotic and eukaryotic test systems and an in vivo mouse micronucleus test performed with rosemary extract RES. The Panel concluded that these do not give rise to safety concerns with respect to genotoxicity of the rosemary extracts.
Antioxidant rosemary extracts have low acute and sub-chronic toxicity in the rat. Sub-chronic studies on all five solvent extracts (D74, AR, ARD, F62, RES) reveal that the only effect at high doses of these rosemary extracts is a slight increase in relative liver weight. This effect has been shown to be reversible and may be the result of Phase I and II enzyme induction. The effect was not accompanied by increases in plasma levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (AP). Considering the low magnitude, reversibility and the nature of the hepatic changes, and the absence of increases in plasma ALT, AST and AP, the Panel concludes that the minor increase in the liver weight reported, accompanied by minimal centrilobular hypertrophy and microsomal enzyme induction, represent an adaptive response and are not of toxicological concern.
Overall, the 90-day feeding studies in rats with the different rosemary extracts tested, including AR, ARD, RES and D74, reveal NOAEL values in the range of 180 to 400 mg extract/kg body weight/day equivalent, depending on the carnosol and carnosic acid content of the respective extracts, to 20-60 mg /kg bw /day of carnosol plus carnosic acid.
The toxicological data on the rosemary extracts are insufficient to establish a numerical ADI, because the toxicity data set does not provide reproductive toxicity studies or a long term study. On the other hand, the existing data, including the absence of effects in the 90-day studies on reproductive organs and lack of genotoxicity, do not give reason for concern.
Dietary exposure to carnosol plus carnosic acid has been estimated for adults and pre-school children (aged 1.5 to 4.5 years) and amounts to mean values of respectively 0.04 and 0.11 mg carnosol plus carnosic acid/kg bw/day, 0.10 and 0.20 mg carnosol plus carnosic acid/kg bw/day at the 95th percentiles and 97.5th percentile values of 0.12 and 0.23 mg carnosol plus carnosic acid/kg bw/day.
The Panel notes that the margin between the NOAEL range in the 90-day rat studies with all five extracts of 180 to 400 mg extract/kg bw/day equivalent to 20-60 mg/kg bw/day of carnosol plus carnosic acid, and the dietary exposure estimates for adults would amount between 500-1500 for the mean intake values, between 200-600 for the 95th percentile values and between 167-500 for the 97.5th percentile values. For pre-school children these margins would amount to respectively at least 182-546, 100-300 and 87-261. The Panel notes that these margins of safety are worst case estimates since the NOAELs from the different studies were generally the highest dose levels tested, and that the estimates of dietary exposure were conservative.
Therefore the Panel is of the opinion that the margin of safety is high enough to conclude that dietary exposure resulting from the proposed uses and use levels are not of safety concern.
The Panel notes that to achieve these levels of dietary exposure, high level consumers would need to select a diet that was entirely composed of foods containing rosemary extracts for those food categories in which it was permitted. In reality not all processed foods will contain added antioxidants and it seems unlikely that these extracts would be used at the maximum usage level in all the proposed food in each category or that some consumers would systematically always choose all foods containing rosemary extract.
Based on the margins of safety identified, the Panel concluded that the use of rosemary extracts at the proposed uses and use levels would not be of safety concern.