Scientific opinion on the safety of the proposed amendment of the specifications for steviol glycosides (E 960) as a food additive

Transcription

1 SCIENTIFIC OPINION ADOPTED: 17 November 2015 PUBLISHED: 8 December 2015 doi: /j.efsa Scientific opinion on the safety of the proposed amendment of the specifications for steviol glycosides (E 960) as a food additive EFSA Panel on Food Additives and Nutrient Sources added to Food Abstract Following a request from the European Commission, the EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS) provides a scientific opinion regarding the safety of an amendment of the specifications for the food additive steviol glycosides (E 960). It was requested that rebaudioside M be added to the list of steviol glycosides as an authorised component of the mixture and that the minimum amount of 75% of stevioside and/or rebaudioside A be deleted. The ANS Panel received a dossier from the applicant and subsequently requested additional data. The Panel noted that for all stevioside and any rebaudioside following intestinal degradation, only steviol would be available in the plasma. The toxicological conclusions made by the ANS Panel in 2010 were also based on the conversion to steviol in the intestine and that steviol was the only compound systemically available. Further to this, the Panel considered that the toxicological studies performed with stevioside and rebaudioside A were relevant for assessing the safety of any steviol glycoside degraded in the intestine. The ANS Panel therefore concluded that extending the current specifications to include rebaudioside D and M as alternatives to rebaudioside A in the predominant components would be of no safety concern. The Panel also concluded that provided that the total amount of steviol glycosides (stevioside; rebaudioside, A, B, C, D, E, F and M; steviolbioside; rubusoside and dulcoside) were greater than 95% which are all converted to steviol and given that there was no evidence of absorption for intact glycosides at realistic use levels, the specific steviol glycosides (E 960) composition would not be of safety concern. Finally it was considered that the ADI of 4 mg/kg bw/day can also be applied where total steviol glycosides comprise more than 95% of the material. European Food Safety Authority, 2015 Keywords: food additive, sweetener, steviol glycosides, E 960, rebaudioside M, technical specifications Requestor: European Commission Question number: EFSA-Q Correspondence: fip@efsa.europa.eu EFSA Journal 2015;13(12):4316

2 Panel members: Fernando Aguilar, Riccardo Crebelli, Alessandro Di Domenico, Birgit Dusemund, Maria Jose Frutos, Pierre Galtier, David Gott, Ursula Gundert-Remy, Claude Lambré, Jean-Charles Leblanc, Oliver Lindtner, Peter Moldeus, Alicja Mortensen, Pasquale Mosesso, Dominique Parent- Massin, Agneta Oskarsson, Ivan Stankovic, Ine Waalkens-Berendsen, Rudolf Antonius Woutersen, Matthew Wright and Maged Younes. Acknowledgements: The Panel wishes to thank the members of the Working Group on Applications: Maria Jose Frutos, David Gott, Lieve Herman, Claude Lambré, Jean-Charles Leblanc, Peter Moldeus, Alicja Mortensen, Ivan Stankovic, Paul Tobback, Ine Waalkens-Berendsen, Rudolf Antonius Woutersen and Matthew Wright for the preparatory work on this scientific output and EFSA staff members: Paolo Colombo and Camilla Smeraldi for the support provided to this scientific output. Suggested citation: EFSA ANS Panel (EFSA Panel on Food Additives and Nutrient Sources added to food), Scientific opinion on the safety of the proposed amendment of the specifications for steviol glycosides (E 960) as a food additive. EFSA Journal 2015;13(12):4316, 29 pp. doi: /j.efsa ISSN: European Food Safety Authority, 2015 Reproduction is authorised provided the source is acknowledged. The EFSA Journal is a publication of the European Food Safety Authority, an agency of the European Union. 2 EFSA Journal 2015;13(12):4316

3 Summary Following a request from the European Commission, the European Food Safety Authority (EFSA) Panel on Food Additives and Nutrient Sources added to Food (ANS) provides a scientific opinion regarding the safety of an amendment of the specifications for the food additive steviol glycosides (E 960). The current specifications laid down in Commission Regulation (EU) No 231/2012 stipulate that steviol glycosides (E 960) preparations contain not less than 95% of the 10 named steviol glycosides: stevioside; rebaudiosides A, B, C, D, E and F; steviolbioside; rubusoside and dulcoside, on a dried basis. The specifications further define the preparations as comprising mainly (i.e. at least 75%) of stevioside and/or rebaudioside A. Although available commercial preparations of stevia extracts have been shown to contain measurable levels of rebaudioside M, rebaudioside M is presently not listed as one of the steviol glycosides that may be included with other steviol glycosides to make up the set assay limit of at least 95% total steviol glycosides. The use of steviol glycosides (E 960) as a food sweetener is regulated under the European Parliament and Council Regulation (EC) No 1333/2008 on food additives. Several assessments were made for stevioside as a sweetener by the Scientific Committee for Food (SCF in ), the Joint FAO/WHO Expert Committee on Food Additives (JECFA in time frame) and EFSA ( ). JECFA and EFSA have established an Acceptable Daily Intake (ADI) for steviol glycosides (E 960) of 4 mg/kg bw (body weight) per day, expressed as steviol equivalents. Refined exposure assessments for extension of use in period were also performed by EFSA. According to the applicant, steviol glycoside preparations may contain rebaudioside M at a range of concentrations, from 50% up to nearly 100%. While extracts characterised by a 95% content of rebaudioside M contain <5% of rebaudiosides D, A and B combined, extracts with a lower rebaudioside M content (approximately 50%) may comprise close to 40% rebaudioside D and 7% rebaudioside A. Studies regarding the stability of rebaudioside M under normal and exaggerated storage condition as well as a forced degradation studies were also provided by the applicant. As rebaudioside M-rich steviol glycoside products are proposed for the same uses and use levels than previously evaluated steviol glycosides, the expected resulting exposure will be no greater than the one estimated in the last assessment made by EFSA. The Panel considers that the critical step in the assessment of rebaudioside M by read across was an evaluation of its metabolism and a comparison with the metabolism of rebaudioside A and stevioside that are the steviol glycosides covered by the current specification. For rebaudioside A and stevioside there was an essentially complete degradation to steviol in the gastrointestinal tract at realistic doses, the steviol produced was absorbed and then metabolised to steviol glucuronide. The Panel considers that the steviol exposure from an equivalent dose of rebaudioside D and M can be estimated and that considering the common degradation steps their safety could be evaluated on the same basis as stevioside and rebaudioside A. An evaluation of the glycosidic substitutions at the R1 and R2 position in the structure were also taken into consideration in extrapolating the available data for other steviol glycosides. A series of in vitro metabolism studies were conducted to assess individual steviol glycosides at concentrations of 0.2 and 2 mg/ml for up to 48 hours and by looking at the rate and degree of microbial hydrolysis. The Panel has also requested additional data to confirm the plausibility of this hypothesis. Based on the structure activity relationships observed with rebaudiosides A, D and M and the in vitro results, the Panel considered that it was plausible that the metabolism of rebaudioside M to steviol would be no greater than that observed with rebaudioside D and that a significant amount of intact rebaudioside M was unlikely to be absorbed. There is a consistent picture of high break down by the microflora in vitro of steviosides, rebaudiosides and dulcosides with production of steviol which is subsequently absorbed, metabolised and eliminated, the only systemic exposure being to steviol and, as previously concluded, toxicological studies on steviol are considered relevant to the assessment of these compounds. From pooling all the evidence the ANS Panel concluded that extending the current specifications to include rebaudiosides D and M as alternatives to rebaudioside A in the predominant components of steviol glycosides would not be of safety concern. 3 EFSA Journal 2015;13(12):4316

4 The Panel also concluded that provided that the total amount of steviol glycosides (stevioside; rebaudiosides A, B, C, D, E, F and M; steviolbioside; rubusoside and dulcoside) were greater than 95% which are all converted to steviol and given that there was no evidence of absorption for intact glycosides at realistic use levels, the specific steviol glycosides (E 960) composition would not be of safety concern. Finally the Panel concluded that the ADI of 4 mg/kg bw/day can also be applied where total steviol glycosides comprise more than 95% of the material. 4 EFSA Journal 2015;13(12):4316

5 Table of contents Abstract... 1 Summary Introduction Background and Terms of Reference as provided by the European Commission Background Terms of Reference Additional information Authorisations and previous evaluations Data and Methodologies Data Methodologies Assessment Identity of the substance Specifications Manufacturing process Method of analysis in food Reaction and fate in food Degradation products Possible interaction with food Case of need and proposed uses Exposure Biological and toxicological data Absorption, distribution, metabolism and excretion Toxicological data Discussion Conclusion Documentation provided to EFSA References Abbreviations EFSA Journal 2015;13(12):4316

6 1. Introduction 1.1. Background and Terms of Reference as provided by the European Commission Background The use of food additives is regulated under the European Parliament and Council Regulation (EC) No 1333/ on food additives. Only food additives that are included in the Union list, in particular in Annex II to that regulation, may be placed on the market and used under the conditions of use specified therein. Moreover, food additives shall comply with the specifications as referred to in article 14 of that Regulation and laid down in Commission Regulation (EU) No. 231/ The European Food Safety Authority (EFSA) evaluated the safety of steviol glycosides as a food additive (sweetener) and expressed its opinion in Steviol glycosides (E 960) is currently an authorised food additive in the European Union for use in several food categories and specifications have been adopted for it. Presently those specifications stipulate that steviol glycosides (E960) preparations contain not less than 95% of the 10 named steviol glycosides: stevioside; rebaudiosides A, B, C, D, E and F; steviolbioside; rubusoside and dulcoside, on a dried basis. The specifications further define the preparations as comprising mainly (at least 75%) of stevioside and/or rebaudioside A. Although available commercial preparations of stevia extracts have been shown to contain measurable levels of rebaudioside M, rebaudioside M is presently not listed as one of the steviol glycosides that may be included with other steviol glycosides to make up the set assay limit of at least 95% total steviol glycosides. A manufacturer has requested an amendment to the present specifications of steviol glycosides (E 960) in terms of the following: - to add rebaudioside M to the list of steviol glycosides that may comprise the assay value of 95% (total steviol glycosides content) and - to delete the requirement in the definition specifying that steviol glycosides contain at least 75% stevioside and/or rebaudioside A. According to the applicant, the requested changes are consistent with the use of the existing steviol glycoside preparations that is rebaudioside M intended for use in foods and beverages as a highintensity sweetener. Therefore, rebaudioside M-containing preparations are proposed for use in foods and beverages under the same conditions as those adopted for the existing steviol glycoside preparations in accordance with Commission Regulation (EU) No. 1131/ on steviol glycosides. Rebaudioside M, also known as rebaudioside X, is a minor steviol glycoside constituent of the Stevia rebaudiana Bertoni (S.rebaudiana) plant Terms of Reference The European commission requests the European Food Safety Authority to provide a scientific opinion on the safety of the proposed amendment of the specifications of the food additive steviol glycosides (E 960) in accordance with regulation (EC) No 1331/2008 establishing a common authorisation procedure for food additives, food enzymes and food flavourings 5. 1 Regulation (EC) No 1333/2008 of the European Parliament and of the Council of 16 December 2008 on food additives. OJ L 354, , p Regulation (EU) No 231/2012 of 9 March 2012 laying down specifications for food additives listed in Annexes II and III to Regulation (EC) No 1333/2008 of the European Parliament and of the Council. OJ L 83, , p EFSA Journal 2010;8(4):1537 [84 pp.] 4 Regulation (EU) No 1131/2011 of 11 November 2011 amending Annex II to Regulation (EC) No 1333/2008 of the European Parliament and of the Council with regard to steviol glycosides. OJ L 295, , p Regulation (EC) No 1331/2008 of the European Parliament and of the Council of 16 December 2008 establishing a common authorisation procedure for food additives, food enzymes and food flavourings on food additives. OJ L 354, , p OJ L354, , p EFSA Journal 2015;13(12):4316

7 1.2. Additional information Steviol glycosides (E 960) is an approved food additive in the EU for use as a sweetener in food. The composition and purity of the products in commerce should comply with the specifications defined by Commission Regulation (EU) No 231/2012. The final product should contain not less than 95% stevioside; rebaudiosides A, B, C, D, E and F; steviolbioside; rubusoside and dulcoside, on the dried basis and comprising at least 75% of stevioside and/or rebaudioside A Authorisations and previous evaluations Following a request from the European Commission, EFSA performed an evaluation regarding the safety of the proposed amendment of the specifications of the food additive steviol glycosides (E 960), in particular, the application for a change in the steviol glycoside specification, to add rebaudioside M to the list of steviol glycosides and to delete in the definition the requirement specifying that steviol glycoside products contain at least 75% stevioside and/or rebaudioside A. The use of steviol glycosides (E 960) as a food sweetener is regulated under the European Parliament and Council Regulation (EC) No 1333/2008 on food additives. Stevioside as a sweetener was previously evaluated by the Scientific Committee for Food (SCF) in 1984, 1988 and 1999 (SCF, 1984; 1989; 1999). The Joint FAO/WHO Expert Committee on Food Additives (JECFA) reviewed the safety of steviol glycosides in 2000, 2005, 2006, 2007 and 2009 (JECFA, 2000; 2005; 2006a; 2007; 2010) and established an Acceptable Daily Intake (ADI) for steviol glycosides (E 960) of 4 mg/kg bw (body weight) per day, expressed as steviol equivalents. In its opinion of 14 April 2010, the EFSA ANS Panel has established an ADI of 4 mg/kg bw per day for steviol glycosides (E 960), expressed as steviol equivalents. In the same opinion, the ANS Panel evaluated the safety in use of steviol glycosides (E 960) and, on the basis of the intended uses requested by the three applicants, concluded that the established ADI would likely be exceeded at the maximum proposed use levels for both children and adults (EFSA ANS Panel, 2010). In 2011, EFSA carried out a revised exposure assessment of steviol glycosides (E 960), based on revised proposed uses, and concluded that high-level exposure estimates for children are above the ADI for several European countries (EFSA ANS Panel, 2011). Since the last exposure assessment of steviol glycosides performed in 2011, maximum permitted levels (MPLs) of steviol glycosides (E 960) in foods have been defined in Commission Regulation (EU) No 1131/2011 amending Annex II to Regulation (EC) No 1333/2008 (Table 1). In April 2014, EFSA ANS Panel adopted a scientific opinion further to a request for the authorisation of use of steviol glycosides (E 960) as a food additive (sweetener) for the food category Coffee, tea, herbal and fruit infusions, chicory; tea, herbal and fruit infusions and chicory extracts; tea, plant fruit and cereal preparations for infusions, as well as mixes and instant mixes of these products (sub food category Other). In order to ensure an acceptable level of sweetness to the average European consumer, a new request to revise the maximum use level from 10 as assessed in the EFSA 2014 (EFSA ANS Panel, 2014) to 29 mg/l for the tea beverages and instant-coffee and instant-cappuccino products was submitted (EFSA ANS Panel, 2015). 2. Data and Methodologies 2.1. Data The applicant has submitted a dossier in support of its application for the change in specifications for steviol glycosides (E 960) as a food additive in the EU. Additional data were generated by the applicant upon request from the ANS Panel. The request for the additional data was discussed at a technical hearing held during a meeting of the Standing Working Group on Applications on 4 March EFSA Journal 2015;13(12):4316

8 2.2. Methodologies The current Guidance for submission for food additive evaluations (EFSA ANS Panel, 2012) has been followed by the ANS Panel for the evaluation of the proposed change in specifications of the already authorised food additive steviol glycosides (E 960). 3. Assessment 3.1. Identity of the substance Steviol glycosides (E 960) that comply with the specifications defined by Commission Regulation (EU) No 231/2012 are approved in the EU for use in food as sweeteners. According to the EU specifications, the final product should contain, on the dried basis, not less than 95% stevioside; rebaudiosides A, B, C, D, E and F; steviolbioside; rubusoside and dulcoside, and comprising at least 75% of stevioside and/or rebaudioside A. The product described by the applicant consists of not less than 95% total steviol glycosides of which not less than 50% is rebaudioside M, with the remainder of the steviol glycosides composition accounted for by the 10 already recognised steviol glycosides (i.e. stevioside, rebaudioside A, rebaudioside B, rebaudioside C, rebaudioside D, rebaudioside E, rebaudioside F, dulcoside, steviolbioside and rubusoside), in any combination and ratio. The product described does not comply with the current EU specifications. It contains rebaudioside M that is not listed as one of the 10 steviol glycosides that may be present in the product and the content of stevioside and/or rebaudioside A is lower than specified 75% of the final product. A steviol glycoside product characterised by a composition of at least 50% rebaudioside M is produced as a white to off-white powder, with a clean sweet taste. High-purity rebaudioside M preparations are sweeter (approximately 350 times as sweet as sucrose at 5% sucrose equivalency) than the steviol glycoside preparations that are currently available (which are 200 to 300 time as sweet as sucrose) and possess an improved sweetness quality that resembles sucrose more closely. Steviol glycoside rebaudioside M was first time identified and characterised using High-Performance Liquid Chromatography (HPLC), Mass Spectrometry (MS) and Nuclear Magnetic Resonance (NMR) spectrometry techniques by Ohta et al. (2010) in the new cultivar Stevia rebaudiana Morita developed by selective breeding from the plant Stevia rebaudiana Bertoni. However, they have not detected rebaudioside M in the extract of leaves of Stevia rebaudiana Bertoni. Prakash et al. (2013) identified rebaudioside M in most of the commercial extracts of the leaves of Stevia rebaudiana Bertoni by HPLC using conditions described in the JECFA monograph for Steviol glycosides (JECFA 2010) (with appropriate reference standard) as a minor steviol glycoside at levels of up to 0.26%. The compound was isolated as a crystalline material and further characterised using MS and 1H and 13C NMR spectrometry (Prakash et al., 2013). According to the information provided by the applicant, only leaves of the Stevia rebaudiana Bertoni plant comprise the starting material for the production of steviol glycoside extracts containing at least 50% rebaudioside M. Chemical name of the rebaudioside M is 13-[(2-O-β-D-glucopyranosyl-3-O-β-D-glucopyranosyl-β-Dglucopyranosyl)oxy]kaur-16-en-18-oic acid, 2-O-β-D-glucopyranosyl-3-O-β-D-glucopyranosyl-β-Dglucopyranosyl ester. Rebaudioside M is characterised by a molecular weight of g/mol, a chemical formula of C56H90O33 and a CAS No Rebaudioside M, as all other steviol glycoside constituents, is a glycosylated derivate of the aglycone steviol and as such, shares the same backbone structure shown in Figure 1, differing only with respect to the number of glycosyl units attached to steviol at positions R1 and R2. The structural formula of rebaudioside M is presented in Figure EFSA Journal 2015;13(12):4316

9 Figure 1: Backbone structure of steviol glycosides Figure 2: Structural formula of Rebaudioside M For comparison, the molecular weights and formulae, and the R1 and R2 groups for each of the 10 steviol glycosides that are already recognised under the purity specifications for steviol glycosides in the EU, as well as for the aglycone steviol, also are presented in Table 1. The chemical structure of rebaudioside M is similar to that of rebaudioside A, except that rebaudioside M contains 2 additional β- D-glucosyl moieties at the C-2 and C-3 positions which are identical to the R2 group in the rebaudioside A, B and D. 9 EFSA Journal 2015;13(12):4316

10 Table 1: Molecular weight, molecular formula, and the R1 and R2 groups in the backbone structure for steviol, rebaudioside M and each of the 10 already recognised steviol glycosides in the EC specifications (as provided by the applicant) Glc = glucose; Rha = rhamnose; Xyl = xylose According to the applicant, steviol glycoside preparations may contain rebaudioside M at a range of concentrations, from 50% up to nearly 100%. While extracts characterised by a 95% content of rebaudioside M contain <5% of rebaudiosides D, A and B combined, extracts with a lower rebaudioside M content (approximately 50%) may comprise close to 40% rebaudioside D and 7% rebaudioside A. The ranges of steviol glycoside concentrations for extracts containing either approximately 50% rebaudioside M or 95% rebaudioside M are presented in Table EFSA Journal 2015;13(12):4316

11 Table 2: Individual Steviol Glycoside Distribution in Extracts Containing Approximately 50% and 95% Rebaudioside M Steviol Glycoside Range of concentrations (%) Approximately 50% Reb M 1 95% Reb M 1 Rebaudioside M >50 to < 60 ( ) 95.0 ( ) Rebaudioside D* >30 to < 40 ( ) 1 to 2 ( ) Rebaudioside A* >3 to < 7 ( ) <0.1 ( ) Rebaudioside B* < 1 ( ) 1 to 2 ( ) Stevioside* < 1 ( ) ND Rebaudioside F* ND < 0.05 ( ) Rebaudioside C* < 1 (ND 0.16) ND Rebaudioside E* ND ND Dulcoside A* ND ND Rubusoside* ND ND Steviolbioside* ND ND Total Steviol Glycosides > 95 > 95 Reb M = rebaudioside M; * recognised steviol glycoside according to Commission Regulation (EU) No 231/ ranges in parentheses () are the lowest and highest concentration of the particular steviol glycoside identified in the analysed lots The residual levels of materials used in the manufacturing process, that is, the solvents ethanol and methanol, ion-exchange resins may also be present as defined in the EU specifications. Information on ph and solubility of rebaudioside M are provided as part of the batch analysis of the high-purity product with the content of rebaudioside M 96%. According to the applicant, rebaudioside M is slightly soluble in water using the JECFA method (JECFA 2006b). The ph range of rebaudioside M in 1% solution is approximately 5 to Specifications Specifications for steviol glycosides have been defined by Commission Regulation (EU) No 231/2012 and by JECFA (JECFA 2010) (Table 3). Presently, the EU specifications stipulate such preparations as containing not less than 95% of the 10 named steviol glycosides: stevioside; rebaudiosides A, B, C, D, E and F; steviolbioside; rubusoside and dulcoside. The specifications further define the preparations as comprising mainly (at least 75%) of stevioside and/or rebaudioside A EFSA Journal 2015;13(12):4316

12 Table 3: Specifications for steviol glycosides (E 960) according to Commission Regulation (EU) No 231/2012 and to JECFA (2010). Parameter Commission Regulation (EU) No JECFA (2010) 231/2012 Definition The manufacturing process comprises two main phases: the first involving water extraction of the leaves of the Stevia rebaudiana Bertoni plant and preliminary purification of the extract by employing ion exchange chromatography to yield a steviol glycoside primary extract, and the second involving recrystallisation of the steviol glycosides from methanol or aqueous ethanol resulting in a final product comprising mainly (at least 75%) of stevioside and/or rebaudioside A. The additive may contain residues of ionexchange resins used in the manufacturing process. Several other related steviol glycosides that may be generated as a result of the production process, but do not occur naturally in the Stevia rebaudiana plant have been identified in small amounts (0.10 to 0.37% w/w) Assay Not less than 95% stevioside; rebaudiosides A, B, C, D, E and F,; steviolbioside; rubusoside and dulcoside on the dried basis. Description White to light yellow powder, approximately between 200 and 300 times sweeter than sucrose Identification Solubility Freely soluble to slightly soluble in water Freely soluble in water Stevioside and rebaudioside A The main peak in the chromatogram obtained following the procedure in Method of Assay corresponds to either stevioside or rebaudioside A The product is obtained from the leaves of Stevia rebaudiana Bertoni. The leaves are extracted with hot water and the aqueous extract is passed through an adsorption resin to trap and concentrate the component steviol glycosides. The resin is washed with a solvent alcohol to release the glycosides and product is recrystallised from methanol or aqueous ethanol. Ion exchange resins may be used in the purification process. The final product may be spray-dried. Stevioside and rebaudioside A are the component glycosides of principal interest for their sweetening property. Associated glycosides include rebaudioside C, dulcoside A, rubusoside, steviolbioside, rebaudioside B, rebaudioside D, and rebaudioside F generally present in preparations of steviol glycosides at levels lower than stevioside or rebaudioside A. Not less than 95% of the total of the nine named steviol glycosides on the dried basis. White to light yellow powder, odourless or having a slight characteristic odour. About times sweeter than sucrose The main peak in the chromatogram obtained by the following procedure in Method of Assay corresponds to either stevioside or rebaudioside A ph Between 4.5 and 7.0 (1 in 100 solution) Between 4.5 and 7.0 (1 in 100 solution) Purity Total ash Not more than 1% Not more than 1% Loss on drying Not more than 6% (105 C, 2h) Not more than 6% (105, 2h) Residual solvents Not more than 200 mg/kg methanol and not more than 5000 mg/kg ethanol Not more than 200 mg/kg methanol and not more than 5000 mg/kg ethanol Arsenic Not more than 1 mg/kg Not more than 1 mg/kg Lead Not more than 1 mg/kg Not more than 1 mg/kg The current EU specifications for steviol glycosides are mostly consistent with the specifications established for steviol glycosides by JECFA (2010). The Panel noted that while the EU specifications identify 10 steviol glycosides that may contribute to the assay value, the specifications adopted by JECFA for steviol glycosides recognise only nine steviol glycosides (i.e. rebaudioside E is presently not listed in the JECFA specification for steviol glycosides). Although the JECFA specification does identify stevioside and rebaudioside A as the component glycosides of principal interest for their sweetening property, it does not specifically indicate that the final steviol glycoside product must comprise at least 75% stevioside and/or rebaudioside A. In the EFSA opinion (EFSA ANS Panel, 2010), the evaluated toxicological studies were carried out with test materials rich in rebaudioside A and stevioside EFSA Journal 2015;13(12):4316

13 Analysis of commercial stevia extracts, demonstrated the presence of rebaudioside M, at very low levels (up to 0.26%) in addition to the other already recognised steviol glycosides (Prakash et al., 2013). According to information provided by the applicant, a production process has been developed (see section 3.1.2) which allows for the selective isolation of rebaudioside M and resulting in the production of steviol glycoside preparations enriched in rebaudioside M specifically. Analysis on five non-consecutive lots of high-purity rebaudioside M (96.14 to 96.60%), and three lots of a steviol glycoside extract containing approximately 50% rebaudioside M was conducted to determine whether such preparations would meet the current EU specifications for steviol glycosides. The data provided demonstrated that the manufacturing process produces a consistent product and that, apart from the assay value (rebaudioside M is presently not included as one of the glycosides that may comprise the assay value), such preparations do largely meet all other specification criteria for steviol glycosides. Additionally, the same eight sample lots of rebaudioside M-rich steviol glycoside product also were tested for potential environmental contaminants (microorganisms, pesticides and heavy metals), Microorganisms in the preparations were consistently either not detectable or identified at very low levels (10 to 120 colony forming units (cfu)/gram). Individual types of microorganisms, including Escherichia coli, Salmonella and Listeria, were not present in any of the tested samples. Multiresidue pesticide screens were conducted on the samples. Among others, the samples were screened for organophosphate, organochlorine and ethylenebisdithiocarbamate pesticides. In all cases, no pesticide residues were detected (detection limits ranged from 0.01 to 0.05 mg/kg). In addition to testing of the samples for levels of arsenic and lead, for which maximum limits are set as part of the specifications for steviol glycosides in the EU, the samples also were tested for the levels of cadmium and mercury. In all cases levels of <0.005 mg/kg were determined for all samples tested. Ion-exchange and adsorption resins that consists of a styrene and divinyl benzene copolymer are used during the production process to remove various impurities from the final steviol glycoside extract. Use of resins as part of the production process is consistent with the production process that is already recognised for steviol glycosides in general. Samples from three non-consecutive lots of both low- and high-rebaudioside M content were tested for potential resin leachates, including styrene, divinylbenzene (DVB), trimethylamine (TMA) and triethylamine (TEA). In all cases, the results of the analysis demonstrated absence of any resin residues in the final products, with the analytes testing below the respective limits of detection (styrene: 0.1 ppm; divinylbenzene: 0.1 ppm; trimethylamine: 0.05 ppm; and triethylamine: 0.04 ppm). Considering that rebaudioside M is related to the 10 other already recognised steviol glycosides, and can be extracted from the S. rebaudiana Bertoni plant resulting in stevioside M-rich steviol glycoside preparations that comprise 50% rebaudioside M, modification of the current steviol glycoside specification to include rebaudioside M is requested by the applicant. Specifically, it is proposed by the applicant that rebaudioside M be added to the list of the 10 already permitted steviol glycosides as one additional glycoside that may comprise the assay value of not less than 95%. Furthermore, it is also proposed that the statement The manufacturing process comprises two main phases: [...] resulting in a final product comprising mainly (at least 75%) of stevioside and/or rebaudioside A. pursuant to the Definition of steviol glycosides be changed to The manufacturing process comprises two main phases: [...] resulting in a final product comprising not less than 95% of the below identified 11 related steviol glycosides, in any combination and ratio. It is further proposed that the list of chemical names and the list of molecular weights and CAS numbers be expanded to include, in addition to stevioside and rebaudioside A, the other nine steviol glycosides. Also, rebaudioside M should be added to the listing of chemical formulas. Finally, since stevioside and rebaudioside A may not necessarily be the principal steviol glycosides, the identity criterion, The main peak in the chromatogram obtained following the procedure in Method of Assay corresponds to either stevioside or rebaudioside A, should be removed from the specification, as the main peak(s) of a chromatogram of a steviol glycoside preparation that could meet the revised specification could be associated with a steviol glycoside or steviol glycosides other than stevioside or rebaudioside A as showed in chromatograms provided by the applicant EFSA Journal 2015;13(12):4316

14 To account for the greater sweetness potency of rebaudioside M, the statement White to light yellow powder, approximately between 200 and 300 times sweeter than sucrose pursuant to the Description of steviol glycosides, should be modified to...approximately between 200 and 350 times sweeter Manufacturing process The manufacturing process of rebaudioside M is similar to the general method of extracting steviol glycosides from the leaves of S. rebaudiana, which has been previously reviewed by EFSA (EFSA ANS Panel, 2010). Briefly, the crushed stevia leaves are extracted with hot water and the resulting extract is subjected to isolation and purification (by use of ion-exchange chromatography). This initial stage is followed by additional purification steps, including further and repeated recrystallisation and separation steps. Through the manipulation of these purification steps (i.e. specific number of crystallisation steps, solvent concentration, as well as temperature and duration of the process) the manufacturer is able to selectively crystallise a preparation high in rebaudioside M. This process results in a preparation that contains 95% of steviol glycosides, with rebaudioside M representing more than 50% of the finished product and the remainder comprising the following 10 related steviol glycosides in any combination and ratio: stevioside, rebaudioside A, rebaudioside B, rebaudioside C, rebaudioside D, rebaudioside E, rebaudioside F, dulcoside, steviolbioside and rubusoside. The production process utilised to obtain steviol glycoside preparations with a high rebaudioside M content involves use of solvents that are already recognised for use in the manufacture of steviol glycoside preparations. Food-grade ethanol is used as a desorption solvent during the production process of rebaudioside M. High-purity methanol is used to recrystallise rebaudioside M from the refined steviol glycoside powder. Residual levels of methanol and ethanol in the final product are reported in the levels of not more than 70 and 850 mg/kg, respectively, which are well below the acceptable limits for residues of either solvent as per the existing purity criteria for steviol glycosides in the EC specifications. Ion-exchange and adsorption resins that consist of a styrene and divinyl benzene copolymer are used during the production process to remove various impurities from the final steviol glycoside extract. High-purity calcium hydroxide, meeting the specification listed in the Food Chemicals Codex (FCC 8, eighth edition), is used as a flocculant and activated carbon, meeting appropriate food-grade specifications, is used as a decolourising agent in the production of the intermediate steviol glycoside extract Method of analysis in food At the time of the initial EFSA evaluation of steviol glycosides, in-house validated methods employing HPLC for the identifications of stevioside, rebaudioside A and other related steviol glycosides in food and beverage matrices were recognised (EFSA ANS Panel, 2010). Two single laboratory validated methods for the determination of steviol glycosides in foods using HPLC also were noted to have been published (Geuns, 2008; Gardana et al., 2010). While the applicant has not at this time developed any methods for the analysis of rebaudioside M in specific food and beverage matrices, beyond the identification of rebaudioside M in solutions (employed for stability studies and batch analysis), the applicant anticipates that any such methods will be based on the same principles (HPLC-based methods) as the already established methods. The Panel considered that the proposed approach would be sufficient Reaction and fate in food The stability of rebaudioside M was not specifically addressed during any of the previous evaluations. Additional studies of rebaudioside M stability were provided by the applicant. The studies included evaluations of rebaudioside M stability under normal and exaggerated storage conditions (bulk stability) and at a range of temperatures and ph values. In addition, a forced degradation study was conducted to identify possible degradation products in a steviol glycoside preparation rich in rebaudioside M that may be different from degradation products recognised to form in other steviol glycoside preparations EFSA Journal 2015;13(12):4316

15 Stability of dry powder (bulk stability) Samples of a rebaudioside M-rich preparation (96.32% rebaudioside M) were stored in sealed amber glass vials for up to 26 weeks either at (1) 25 C and ambient relative humidity (RH) (50 to 55%) or (2) 40 C and 75% RH. Analyses for steviol glycosides were conducted in accordance with JECFA s assay method (JECFA, 2010) upon study initiation, and after 12, 24 and 26 weeks of storage. Degradation of less than 4% of rebaudioside M was observed when the samples were stored under either condition. Additionally, minimal changes over the study period were observed in levels of the other steviol glycosides detected in the preparation at study initiation (rebaudioside D, A and B). Furthermore, levels of total steviol glycosides were not less than 95% throughout the study. No information on degradation products was provided. Stability in solution Rebaudioside M - preparation 95% The influence of storage time, temperature and ph on the stability of rebaudioside M (96.32% rebaudioside M) was assessed in 0.1 M phosphate buffer at temperatures ranging from 5 to 40 C and at ph values from 2 to 8. Rebaudioside M was added to the solutions at target levels of 500 mg/l and the stability was analysed over 24 weeks under the respective conditions. As with all other steviol glycosides, the extent and rate of rebaudioside M degradation were shown to be dependent on ph, temperature and time. Consistently, rebaudioside M at any ph level was most stable when stored at 5 C. Over the course of the study, minor degradation (<3%) was observed up to 6 weeks at ph 2, 3, 5 and 6 when stored at 5 C. Fluctuations in the rebaudioside M content after 6 weeks of storage were noted. Similar effects were observed at ph values of 3 and 6 when the solution was stored at 25 C and at ph levels of 4 to 6 when stored at 40 C. As these findings are not consistent, the applicant is of the opinion that it is likely that the lower values are due to analytical variation and not a reflection on the overall stability of rebaudioside M. At ph 2 after 2 weeks of storage at both 25 and 40 C, rebaudioside M started to degrade with more than 50% of rebaudioside M degraded after 6 weeks of storage at 40 C, in comparison to only approximately 17 and 3% of rebaudioside M degraded when stored at 25 and 5 C, respectively, at the same time point. After approximately 14 weeks of storage at 40 C, the degradation of rebaudioside M at ph 2 began to plateau, reaching a maximum of approximately 93% of rebaudioside M degraded. Similarly, solutions of rebaudioside M at ph 3 stored at 40 C degraded over the 24 weeks of storage, with approximately 66% of the rebaudioside M sample remaining at study completion. Overall, at higher ph values (3 to 8 at 25 C and 4 to 8 at 40 C), no significant degradation of rebaudioside M was observed over 24 weeks, regardless of the storage temperature of the solution, as previously stated. Similar to other steviol glycosides, the stability of rebaudioside M follows the same degradation pathway and is ph-, temperature- and time-dependent. Rebaudioside M - preparation 50% As had been conducted for extracts containing 95% rebaudioside M, the stability of the lower rebaudioside M content extract was examined under a series of ph and temperature conditions. Specifically, solutions of ph 2 8 were prepared with a steviol glycoside extract containing approximately 50% rebaudioside M (0.1 M phosphate buffer) and stored at 5, 25, 37 or 56 C in amber glass vials for up to 3 weeks. Consistent with results for the high-content rebaudioside M extract, as well as for other steviol glycoside extracts as previously considered by EFSA, the stability of an extract containing rebaudioside M at lower levels (approximately 50%) was shown to be also time-, ph- and temperature-dependent. At the lowest temperature tested (5 C), minimal changes in the concentrations of the steviol glycosides was apparent in solutions of every ph. At the lower ph levels (ph 2 and 3), considerable degradation of the steviol glycosides occurred at the higher temperatures (up to 100%). At ph 2, degradation of some of the steviol glycosides in the extract (rebaudioside M, D, and A) appeared to be accompanied by rebaudioside B and steviolbioside formation, particularly at 25 and 37 C. In solutions of higher ph ( 4) stored at temperatures of 25ºC and 37 C, the total steviol glycoside content remained relatively unchanged over the storage periods (up to 3 weeks). At 56 C, the degree of degradation increased with time, such that decreases of approximately 5 15% in the total steviol glycoside content were observed in the ph 4 and greater solutions following 3 weeks of storage 15 EFSA Journal 2015;13(12):4316

16 Additionally, a further study has been conducted to assess the stability of rebaudioside M-containing solutions (0.1 M sodium phosphate buffer; ph 6.5) heated to 100 C, with the temperature maintained for 10 minutes. The assay was performed with products containing either approximately 50% or 95% rebaudioside M. Following 10 minutes of heating of the solutions at 100 C, no significant differences in the concentrations of individual steviol glycosides or in the total steviol glycoside content were observed. No additional studies have been conducted to assess specifically the stability of steviol glycoside extracts containing rebaudioside M under conditions simulating baking (temperatures >100 C). The applicant is on the opinion that, considering that samples of rebaudioside M-containing product were shown to be stable in neutral ph solutions (ph 6.5 ph representative of baked goods) heated to 100 C and that internal temperatures of food subjected to baking is not expected to exceed 100 C, the results presented above suggest that steviol glycoside extracts containing rebaudioside M also should remain stable upon baking. Therefore, the applicant is of the opinion that conclusions regarding the stability of steviol glycosides made by EFSA and other scientific bodies can be extended to include rebaudioside M. Degradation products Rebaudioside M was subjected to forced degradation conditions, (0.1 M phosphate buffer at ph 2 heated to 80 C for 24 h), to generate and subsequently identify degradation products. Beyond the identification of rebaudioside M in the sample, eight degradation products were identified. Five of these degradation products are: 13-[(2-O-β-D-glucopyranosyl-3-O-β-D-glucopyranosyl-β-D-glucopyranosyl)oxy] ent-kaur-16- hydroxy-19-oic acid-[(2-o -β-d-glucopyranosyl-3-o-β-d-glucopyranosyl-β-d-glucopyranosyl) ester] (CC-00280), 13-[(2-O-β-D-glucopyranosyl-3-O-β-D-glucopyranosyl-β-D-glucopyranosyl)oxy] ent-kaur-15- en-19-oic acid-[(2-o-β-d-xylopyranosyl-3-o-β-d-glucopyranosyl-β-d-glucopyranosyl) ester] (CC-00277), 13-methyl-16-oxo-17-norkauran-19-oic acid-[(2-o-β-d-glucopyranosyl-3-o-β-dglucopyranosyl-β-d-glucopyranosyl) ester] (CC-00281), 13-[(2-O-β-D-glucopyranosyl-3-O-β-D-glucopyranosyl-β-D-glucopyranosyl)oxy]-16-hydroxy ent-kauran-19-oic acid (CC-00209), and 13-[(2-O-β-D-glucopyranosyl-3-O-β-D-glucopyranosyl-β-D-glucopyranosyl)oxy] ent-kaur-15- en-19-oic acid (CC-00210). Degradation products CC-00280, CC and CC (molecular weights of 1,309.3, 1,291.3 and respectively) were identified at study initiation in very low quantities (<0.5%) and increased as a result of the extreme conditions to approximately 7, 22 and 3% respectively. Low levels of rebaudioside B also were detected at study initiation (approximately 1%) and increased to approximately 9% at study completion. Following forced degradation, CC and CC were found in the samples at levels of up to approximately 10 and 23% respectively. These products were not identified at study outset. Other substances identified in the samples following the forced degradation that were not present at study outset included steviol and isosteviol at levels of up to approximately 1% and 7%, respectively, at the end of the study. In this case, the results of the forced degradation study of the rebaudioside M-rich preparations revealed four substances, present in the bulk material, to increase with time (i.e. CC-00280, CC , CC and rebaudioside B). Additionally, four compounds were identified that were not present in the starting material: CC-00209, CC-00210, as well as steviol and its structural isomer, isosteviol. Rebaudioside B is presently included as one of the steviol glycosides that may comprise the assay value of not less than 95%. Like the degradation products previously identified for other steviol glycoside preparations, characterisation of CC-00280, CC and CC revealed these substances to be related to the parent steviol glycosides, but to be slightly differentiated structurally by virtue of the presence of a hydroxyl group at position C-16, an endocyclic double bond and an isosteviol backbone respectively. Likewise, CC and CC also were found to be related to the parent steviol glycoside compounds, the differences being again a hydroxyl group at position C EFSA Journal 2015;13(12):4316

17 and an endocyclic double bond respectively. Furthermore, both CC and CC were identified as degradants of rebaudioside A (Prakash et al., 2012). Given the structural similarities of these substances to other steviol glycosides, the identified degradation products, according to the applicant, are not expected to present any safety concerns. The Panel noted that the identified degradation products are similar to those produced by forced degradation of other rebaudiosides in similar conditions. The Panel considered that these studies demonstrate that the stability of rebaudioside M is similar to other steviol glycosides, as previously evaluated by EFSA. Possible interaction with food No data on the interactions with food for rebaudoside M-rich steviol glycosides products are provided. The applicant is of the opinion that, considering the structural similarities of all steviol glycosides and the existing data, the need for formal stability investigations with rebaudioside M-rich preparations into possible interaction products does not appear to be necessary Case of need and proposed uses Rebaudioside M-rich steviol glycosides products are proposed for the same uses and use levels than previously evaluated steviol glycosides Exposure The Panel considers that if this change in specifications were granted exposure to steviol glycosides will be no greater than the last EFSA statement estimate (EFSA ANS Panel, 2015) Biological and toxicological data Data on the toxicology of steviol glycosides was described in the EFSA opinion of Therefore, the section will mainly describe new data on rebaudioside M or data advanced to allow read across of the existing data to rebaudioside M Absorption, distribution, metabolism and excretion Summary of ADME in the Scientific Opinion on the safety of steviol glycosides for the proposed uses as a food additive (EFSA, 2010): The Panel notes that in vitro studies demonstrated that human digestive enzymes are not capable of hydrolysing β-glycosidic bonds of steviol glycosides. However, the intestinal microflora of humans (and rats) is able to convert steviol glycosides to steviol. In addition, in the Caco-2 cell model the apparent permeability value of steviol was found to be 200 to 300-times higher than that of stevioside or rebaudioside A. Other in vitro studies assessing the metabolic transformation of steviol showed a similar formation of hydroxy-metabolites of steviol in the presence of rat or human liver microsomes. In vivo studies in rats receiving stevioside demonstrated that free steviol was the main metabolite present in plasma and it reached maximum plasma concentration 24 hours after administration. In animal liver, steviol was shown to primarily undergo conjugation with glucuronic acid to form steviol glucuronide, identified as the major metabolite in bile. From the results of biliary and faecal excretion, it can be concluded that in rats, enterohepatic circulation occurs. In rats, steviol has been shown to be primarily excreted in the faeces via the bile, and in smaller amounts in the urine. In human volunteers exposed orally to stevioside or rebaudioside A, no free steviol was detected in the blood but steviol glucuronide was found to be the main metabolite in plasma. No steviol epoxide, which may be mutagenic, was detected in human plasma. The presence of multiple peaks in time of plasma concentrations of steviol glucuronide indicates enterohepatic circulation of steviol in humans as experimentally demonstrated in rats. Steviol glucuronide was also reported to be the main metabolite found in the urine of subjects receiving stevioside or rebaudioside A; this elimination pathway accounted for about 60% of the dose. Steviol was reported to be the main metabolite found in the faeces of humans receiving oral stevioside or rebaudioside A. The Panel considers that these toxicokinetic analyses indicated that rebaudioside A and stevioside underwent similar metabolic and 17 EFSA Journal 2015;13(12):4316