Scientific Opinion on the Safety of smoke flavour Primary Product Zesti Smoke Code Update 1

Transcription

1 EFSA Journal 2011; 9(7):2307 SCIENTIFIC OPINION Scientific Opinion on the Safety of smoke flavour Primary Product 1 ABSTRACT EFSA Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids (CEF) 2,3 European Food Safety Authority (EFSA), Parma, Italy The EFSA CEF Panel updated the safety assessment of the smoke flavouring Primary Product Zesti Smoke Code 10 based on evaluation of a new 90-day dietary rat toxicity study. In 2009, the Panel concluded that the margins of safety (range: 5-32) for this Primary Product were insufficient and that the proposed uses and use levels were of safety concern. In this new application, the petitioner proposed to reduce upper use levels and to avoid use in composite foods. The chemical data provided demonstrate that the material tested in the new 90-day study is representative of the Primary Product evaluated by EFSA in From the new 90-day study the Panel identified a NOAEL for Zesti Smoke of 664 mg/kg bw/day, based on the reduced weight gain and white blood cell changes in female rats elicited by a high dose level. The Panel considers that this NOAEL supersedes that of 134 mg/kg bw/day identified in an earlier 90-day study, since the new study s experimental design permits a more accurate identification of the NOAEL. When use of Zesti Smoke based on upper use levels is considered in all food sources (18 categories), estimated dietary intake ranges from 6.7 to 8.4 mg/kg bw/day, depending on the exposure assessment model used. When use (upper use levels) is assumed in traditionally smoked food only, exposure is estimated to be mg/kg bw/day. The margins of safety for this Primary Product, i.e. ratio between the NOAEL from the new 90-day study and the intake estimates, are either (all food sources) or (traditionally smoked sources only). The Panel considers that these margins of safety are insufficient, since they are based on a 90-day toxicity study, and data are lacking on reproductive, developmental and long-term toxicity. In line with its conclusions of 2009, the Panel concludes that the Primary Product Zesti Smoke Code 10 at the proposed uses and use levels is of safety concern. European Food Safety Authority, On request from the European Commission, Question No EFSA-Q , adopted on 6 July Panel members: Arturo Anadón, Mona-Lise Binderup, Wilfried Bursch, Laurence Castle, Riccardo Crebelli, Karl-Heinz Engel, Roland Franz, Nathalie Gontard, Thomas Haertlé, Trine Husøy, Klaus-Dieter Jany, Catherine Leclercq, Jean-Claude Lhuguenot, Wim C. Mennes, Maria Rosaria Milana, Karla Pfaff, Kettil Svensson, Fidel Toldrá, Rosemary Waring, Detlef Wölfle. Correspondence: cef-unit@efsa.europa.eu 3 Acknowledgement: The Panel wishes to thank the members of the Working Group on Flavourings: Ulla Beckman Sundh, Wilfried Bursch, Angelo Carere, Karl-Heinz Engel, Henrik Frandsen, Rainer Gürtler, Frances Hill, Trine Husøy, Catherine Leclercq, Wim Mennes, Gerard Mulder, Iona Pratt, Gerrit Speijers and Harriet Wallin for the preparatory work on this scientific opinion and EFSA staff: Davide Arcella, Anna F. Castoldi and Anne Theobald for the support provided to this scientific opinion. Suggested citation: EFSA Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids (CEF); Scientific Opinion on the safety of smoke flavour Primary Product. EFSA Journal 2011; 9(7):2307. [17 pp.] doi: /j.efsa Available online: European Food Safety Authority, 2011

2 SUMMARY Following a request from the European Commission (EC), the Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids (CEF) of the European Food Safety Authority (EFSA) reevaluated the safety of the smoke flavouring primary product Zesti Smoke Code 10 based on a new 90-day dietary toxicity study submitted by the petitioner in accordance with Regulation (EC) No 2065/2003 of the European Parliament and the Council on smoke flavourings intended for use in or on foods. In the EFSA opinion of 2009 on Zesti Smoke Code 10, the CEF Panel had concluded that the margins of safety (ranging from 5 to 32, depending on the exposure scenario used) for this Primary Product were insufficient and that the proposed uses and use levels were of safety concern. Such margins of safety were calculated on the basis of a no-observed-adverse-effect-level (NOAEL) of 134 mg/kg bw/day identified from a 90-day dietary toxicity study, and on dietary exposure estimates (normal or upper use levels) from all food sources (18 food categories) as well as from only traditionally smoked foods. Based on these conclusions, the petitioner performed a new 90-day study and submitted the results for evaluation by EFSA. As compared with the previous application, the petitioner also proposed to reduce upper use levels, and to avoid the use in composite foods. The present opinion should be read in conjunction with the EFSA opinion of 2009, since this update only contains the new information provided by the applicant (chemical analysis, 90-day toxicity study and proposed reduced upper use levels) and its evaluation by the CEF Panel. The other (chemical and toxicological) information required for the evaluation of smoke flavouring is available in the previous opinion and is still considered to apply to this product. The new chemical data provided demonstrate that the test material (batch BJW2809) used in the 90- day toxicity study carried out in 2010 is representative of the Primary Product Zesti Smoke Code 10 evaluated in the 2009 EFSA opinion. The levels of the polycyclic aromatic hydrocarbons benzo(a)anthracene and benzo(a)pyrene were below the limits of detection of the employed analytical technique (0.2 µg/kg). The new 90-day dietary toxicity study with the Primary Product was conducted according to OECD TG 408. Zesti Smoke Code 10 was administered to groups of 10 male and 10 female HsdHan :WIST rats at dose levels of 0, 5, 9 or 12 g/kg diet for 13 weeks. These dietary levels were equivalent to a mean intake of Zesti Smoke Code 10 over the 90-day period of 0, 322, 572 or 785 mg/kg bw/day for males and 0, 376, 664 or 910 mg/kg bw/day for females, respectively. The new study used a narrower range of dose levels around the NOAEL identified from the more recent of the two 90-day studies described in the previous opinion on Zesti Smoke Code 10. The CEF Panel considered that a NOAEL of 664 mg/kg bw/day could be established for Zesti Smoke, based on the reduction in weight gain in female rats receiving 910 mg Zesti Smoke/kg bw/day, together with the possible treatment-related effects on white blood cells and lymphocytes in females at this dose level. While the NOAEL for males was higher, at 785 mg Zesti Smoke/kg bw/day, the Panel has conservatively used the lower NOAEL of 664 mg Zesti Smoke/kg bw/day identified in females for the purpose of risk characterisation. For comparison, this NOAEL is about 4 to 5-fold higher than the NOAEL for changes in body weight gain and relative kidney weights identified in the 90-day study performed in 2006 and described in the previous EFSA opinion on Zesti Smoke Code 10. The Panel concluded that the new NOAEL of 664 mg/kg bw/day supersedes that of 134 mg/kg bw/day from the previous study, since the experimental design of the new study permitted a more accurate identification of the NOAEL. In order to estimate dietary exposure to the Primary Product Zesti Smoke Code 10, the CEF Panel used two different methodologies, namely the Smoke Theoretical Added Maximum Daily Intake (SMK-TAMDI) and the smoke flavouring EPIC model (SMK-EPIC), developed by the Panel specifically for smoke flavourings. In both cases, dietary exposure estimates were calculated by assuming that the Primary Product Zesti Smoke Code 10 is present at the upper use levels provided by the applicant for the 18 food categories as outlined in the Commission Regulation of Normal use levels were not provided by the applicant. Depending on the exposure assessment model used, dietary EFSA Journal 2011; 9(7):2307 2

3 exposures from all sources range from 6.7 to 8.4 mg/kg bw/day. When dietary exposure estimates are based on use of the Primary Product Zesti Smoke Code 10 in traditionally smoked foods only (upper use levels), these range from 5.0 to 7.7 mg/kg bw/day. The Panel did not anticipate that smoke flavourings would be used in food specifically designed for infants (0-12 months) and children (12-36 months). Therefore the safety of use of Primary Product Zesti Smoke Code 10 in such products was not assessed. Based on the data above, if the Primary Product Zesti Smoke Code 10 is present at the upper use levels provided by the applicant for the 18 food categories, the margins of safety for the intake estimates, as compared to the NOAEL derived from the 90-day toxicity study in rats of 664 mg/kg bw/day, amount to When assuming the use of Primary Product Zesti Smoke Code 10 in traditionally smoked products only, the margins of safety would amount to for the intake estimates based on the upper use levels. Given i) the fact that these margins of safety are based on a 90-day toxicity study, ii) the absence of data on reproduction and developmental toxicity and iii) the absence of long-term studies, it is concluded that the uses and use levels of Primary Product Zesti Smoke Code 10 would require a larger margin of safety. The Panel concludes that the margins of safety are insufficient and that the use of Primary Product Zesti Smoke Code 10 at the proposed uses and use levels is of safety concern. KEY WORDS Smoke flavouring, Primary Product, Zesti Smoke Code 10. EFSA Journal 2011; 9(7):2307 3

4 TABLE OF CONTENTS Abstract... 1 Table of Contents... 4 Background as provided by the European Commission... 5 Terms of reference as provided by the European Commission... 5 Assessment Introduction Summary of the previous opinion on the safety of the smoke flavouring Primary Product Zesti Smoke Code 10 (EFSA, 2009a) Technical data Identity of the Primary Product Chemical description of the Primary Product Identification and quantification of Primary Product constituents Principal constituents Content of Polycyclic Aromatic Hydrocarbons (PAHs) Proposed uses Dietary exposure assessment Toxicological data Identity of the test material Subacute and subchronic toxicity Discussion Conclusions and recommendations Documentation provided to EFSA References EFSA Journal 2011; 9(7):2307 4

5 BACKGROUND AS PROVIDED BY THE EUROPEAN COMMISSION The European Parliament and Council Regulation (EC) No 2065/ provides the basis for securing a high level of protection for human health and the interests of consumers in relation to smoke flavourings used or intended for use in or on foods. It furthermore lays down a procedure for the evaluation and authorisation of primary smoke condensates and primary tar fractions and for the establishment of a list of primary smoke condensates and tar fractions to the exclusion of all others and their conditions of use. The European Food Safety Authority (EFSA) was asked to provide scientific opinion on the safety of the product Zesti Smoke Code 10. In its opinion adopted on 29 January 2009 EFSA stated that Zesti Smoke Code 10 is genotoxic in vitro, but not in vivo. The NOAEL was derived from a 90-day study and amounts to 134 mg/kg bw/day based on a non reversible increase in kidney weight and a decreased body weight gain at the higher dose levels. EFSA concluded that the margin of safety is insufficient and that the use of Primary Product Zesti Smoke Code 10 at the proposed uses and use levels is of safety concern. Based on the conclusions drawn by EFSA the petitioner has repeated a 90-day study and proposed reduced maximum levels for use. The use in composite foods is not proposed anymore. TERMS OF REFERENCE AS PROVIDED BY THE EUROPEAN COMMISSION The European Commission requests the European Food Safety Authority to carry out a safety assessment of the smoke flavouring primary product Zesti Smoke Code 10 based on the repeated 90- day study submitted by the petitioner in accordance with Regulation (EC) No 2065/ of the European Parliament and the Council on smoke flavourings intended for use in or on foods. 4 Regulation (EC) No 2065/2003 of the European Parliament and of the Council of 10 November 2003 on smoke flavourings used or intended for use in or on foods. OJ L 309, , p EFSA Journal 2011; 9(7):2307 5

6 ASSESSMENT 1. Introduction The European Food Safety Authority (EFSA) has been asked by the European Commission (EC) to reconsider its safety assessment of the smoke flavouring Primary Product Zesti Smoke Code 10 based on evaluation of a new 90-day study submitted by the petitioner in accordance with Regulation (EC) No 2065/ of the European Parliament and the Council on smoke flavourings intended for use in or on foods. In the opinion published by EFSA in 2009 on the safety of smoke flavour Primary Product Zesti Smoke Code 10 (EFSA, 2009a), the CEF Panel had concluded that the margins of safety for this Primary Product were insufficient and that the proposed uses and use levels were of safety concern (for further details also see the opinion summary reported in section 1.1). The margins of safety (ranging from 5 to 32, depending on the exposure scenario used) were calculated on the basis of a noobserved-adverse-effect-level (NOAEL) of 134 mg/kg bw/day identified from a 90-day toxicity study, and on dietary exposure estimates (normal or upper use levels) from all food sources (18 food categories) as well as from only traditionally smoked foods. Based on the 2009 EFSA conclusions, the petitioner has performed a new 90-day study and submitted the results for evaluation by EFSA. As compared with the previous application, the petitioner has also proposed to reduce upper use levels and to avoid the use in composite foods. The present opinion does not replace but complements the previous one published on 11 June 2009 (EFSA, 2009a), and therefore both documents should be read in conjunction. This opinion only refers to the new information received from the applicant and to its impact on the safety assessment of Zesti Smoke Code 10, i.e. chemical data on the material used in the 90-day study, estimation of the NOAEL, dietary exposure and margins of safety. The newly submitted data specifically include (i) the chemical analysis of the principal constituents and polycyclic aromatic hydrocarbons (PAHs) content of the material tested (demonstrating that this material is representative of the commercial Primary Product), (ii) a 90-day oral toxicity study and (iii) new proposed maximum levels for use. With respect to the other information submitted by the applicant prior to 2009 in accordance with the Guidance document (EFSA, 2005), the reader should refer to the opinion of The latter contains complementary chemical and toxicological data that still hold true for this Primary Product Summary of the previous opinion on the safety of the smoke flavouring Primary Product Zesti Smoke Code 10 (EFSA, 2009a) The original summary taken from the opinion published by EFSA in 2009 on the safety of the smoke flavouring Primary Product Zesti Smoke Code 10 (EFSA, 2009a) is given below. The Primary Product Zesti Smoke Code 10 was obtained from a specified mixture of the following woods: hickory (Carya ovata) and oak (Quercus alba). The production of Zesti Smoke Code 10 comprises the following steps: (i) drying and sieving of the hardwood sawdust, (ii) heating of the dried sawdust in a specified reactor, (iii) condensing of the released smoke and (iv) separation of the aqueous part of the smoke condensate from precipitated tar. Essential parameters of the manufacturing process were provided by the applicant. Identification and quantification of Primary Product constituents were performed using different batches than were used for the toxicological studies. In both cases several batches covering a production period from 2004 to 2005 were used. Considering a relative standard deviation for the batch-to-batch reproducibility of 5 to 35% for the content of the individual identified constituents the 5 Regulation (EC) No 2065/2003 of the European Parliament and of the Council of 10 November 2003 on smoke flavourings used or intended for use in or on foods. OJ L 309, , p EFSA Journal 2011; 9(7):2307 6

7 material used in testing can be considered as representative for the Primary Product Zesti Smoke Code 10 in general. The water content of the Primary Product was 64 wt. %. The volatile fraction identified by capillary gas chromatographic analysis (GC) accounts for 15 wt. %. The total amount of the unidentified constituents accounts for 33 wt. %. The concentrations of the polycyclic aromatic hydrocarbons (PAHs) listed in the EFSA guidance document on submission of a dossier on a Smoke Flavouring Primary Product were provided. The concentrations of the individual substances were only above 1 µg/kg for benzo[j]fluoranthene (1.5 μg/kg in batch in 2007) and cyclopenta[c,d]pyrene (1.3 g/kg in batch and 1.5 μg/kg in batch in 2006). The levels of benzo[a]anthracene and benzo[a]pyrene were below the limits as laid down in Regulation (EC) No 2065/2003. Analysis of ten batches revealed no significant batch-to-batch variability. No data were provided on the stability of the Primary Product. Normal use levels of the Primary Product proposed by the applicant ranged between 1 g/kg food (ready-to-eat savouries, soups and broths) and 2.5 g/kg food (meat and meat products). Dietary exposure for the Primary Product, as estimated by the applicant, was 30 mg/kg body weight per day. In order to estimate dietary exposure to the Primary Product Zesti Smoke Code 10, the CEF Panel used two different methodologies, developed by the Panel specifically for smoke flavourings. Dietary exposure estimates were calculated by assuming that the Primary Product Zesti Smoke Code 10 is present at the normal or upper use levels provided by the applicant for the 18 food categories as outlined in Commission Regulation (EC). Dietary exposures from all sources range from 22.0 to 28.3 mg/kg bw/day, when assuming that the Primary Product Zesti Smoke Code 10 is present at the upper use levels, and from 9.3 to 11.7 mg/kg bw/day, when normal use levels are considered. When dietary exposure estimates were based on use in only traditionally smoked foods dietary exposures ranged from 8.3 to 14.0 mg/kg bw/day, when assuming that the Primary Product Zesti Smoke Code 10 was present at the upper use levels, and from 4.2 to 6.5 mg/kg bw/day, when normal use levels were considered. Genotoxicity studies conducted on Primary Product Zesti Smoke Code 10 included three in vitro studies (a bacterial reverse mutation test, a mammalian cell gene mutation assay and a chromosome aberration test) and two in vivo studies (a rat liver unscheduled DNA synthesis assay and a mouse bone marrow micronucleus test). The results obtained in these assays showed positive genotoxic responses in the in vitro assays but negative results in vivo in two well conducted studies. Overall it was concluded that Primary Product Zesti Smoke Code 10 is genotoxic in vitro, whereas two in vivo genotoxicity tests were negative and sufficient to eliminate the concerns over the in vitro genotoxicity. The Primary Product was investigated in two 90-day studies. The first one dated back to 1962 when Good Laboratory Practice (GLP) was not yet developed, the second one was conducted more recently according to GLP and OECD guidelines. The no-observed-adverse-effect-level (NOAEL) was derived from the recent study based on the increased relative kidney weights and reduced body weight gain at the two highest dose groups and amounted to 2.5 g/kg diet equivalent to a minimum dietary intake calculated by the applicant of 134 mg/kg bw/day in male rats and 178 mg/kg bw/day in female rats. The NOAEL from this recent 90-day study appears to be similar to the NOAEL from the older 90-day study. Since for this older study the pathology report stated that somewhat more distinct minor degenerative changes were observed in the liver and kidney at 20 g/kg diet, and because slightly more marrow hyperplasia was also noted in this group, it was concluded that the lowest dose level of 2.5 g/kg diet was the NOAEL. Based on these data it was concluded that when assuming that the Primary Product Zesti Smoke Code 10 was present at the normal or upper use levels provided by the applicant for the 18 food categories, the margins of safety as compared to the NOAEL of 134 mg/kg bw/day derived from the 90-day toxicity study in rats amount to 5-6 for the intake estimates based on the upper use levels and to when normal use levels were considered. EFSA Journal 2011; 9(7):2307 7

8 When assuming the use of Primary Product Zesti Smoke Code 10 in traditionally smoked products only, the margins of safety would amount to for the intake estimates based on the upper use levels and to when normal use levels are considered. Given i) the fact that these margins of safety are based on a 90-day toxicity study, ii) the absence of data on reproduction and developmental toxicity and iii) the absence of long term studies, it was concluded that the uses and use levels of Primary Product Zesti Smoke Code 10 would require a larger margin of safety. The Panel concluded that the margin of safety was insufficient and that the use of Primary Product Zesti Smoke Code 10 at the proposed uses and use levels was of safety concern. To decide whether despite the low margins of safety the use of Primary Product Zesti Smoke Code 10 might be approved for traditionally smoked products, at use levels specified, to replace smoking, was outside the remit of the Panel. 2. Technical data 2.1. Identity of the Primary Product The test material (batch BJW2809) used for the 90-day study performed in 2010 was prepared by blending - in equal quantities - three production batches (lots No , , ) of the Primary Product Zesti Smoke Code 10 manufactured in The material has been stored at less than 4 C in a sealed container for 16 months until analysis Chemical description of the Primary Product The data provided on staining index, acidity, carbonyls, phenols, specific gravity and ph of the test material (batch BJW2809) were in accordance with the descriptions of major chemical parameters provided for batches of the Primary Product Zesti Smoke Code 10 produced in 2003 and 2004 and met the specifications given in the EFSA opinion (Table 1). Table 1: Description of major chemical parameters and specifications of the Primary Product Zesti Smoke Code 10 Parameter Batch BJW (a) 2004 (b) Specifications (d) Mean RSD (c) (%) Mean RSD (c) (%) Staining index Acidity (% acetic acid) Carbonyls (g/100 ml) Phenols (g/100 ml) Specific gravity (at 25 C) ph (a): Means of 119 batches of the Primary Product Zesti Smoke Code 10 (b): Means of 198 batches (c): RSD = relative standard deviation (d): EFSA, 2009a EFSA Journal 2011; 9(7):2307 8

9 2.3. Identification and quantification of Primary Product constituents Principal constituents The volatile fraction of the test material was analysed by capillary gas chromatography (GC) according to the approach previously applied to the Primary Product Zesti Smoke Code 10. Mass spectrometry (MS) was used for identification and flame ionisation detection (FID) for quantification. The applicant provided quantitative data on a total of 127 components. The mean contents and the standard deviations determined for the 20 principal constituents are given in Table 2. Table 2: Comparison of principal constituents in the test material (batch BJW2809), in the Primary Product Zesti Smoke Code 10 and in the Primary Product stored for 6 months Constituent Mean (g/l) Batch BJW2809 SD (a) (g/l) RSD (b) % Primary Product (t = 0) Mean (g/l) SD (a) (g/l) RSD (b) % Primary Product (t = 6 months) Acetic acid Levoglucosan Acetol (1-hydroxy-2-propanone) Formic acid ,6-Dimethoxyphenol Corylone (2-hydroxy-3-methyl-2- cyclopenten-1-one) Glycoaldehyde (hydroxyacetaldehyde) Guaiacol (2-methoxyphenol) Hydroxy-2-butanone Furfural Propanal (5H)-Furanone ,2,3-Trimethoxybenzene Isopropyl hexanoate (Hydroxymethyl)-2-furfural Hydroxy-2-cyclopenten-1-one Butanone Methanol Tetrahydrofurfuryl alcohol ,4:3,6-Dianhydro-α-D-glucopyranose (a): SD: standard deviation (b): RSD: relative standard deviation The test material (batch BJW2809) used for the 90-day study performed in 2010 had been prepared from production batches manufactured in 2009; at the time of analysis the material was 16 months old. The applicant, therefore, compared the contents of the principal constituents of the test material not only to the Primary Product Zesti Smoke Code 10 but also to the Primary Product stored for 6 months. Taking into account the previously reported information on the batch-to-batch variability and the stability of the Primary Product Zesti Smoke Code 10 (EFSA, 2009a), the data shown in Table 2 confirm that the test material (batch BJW2809) is representative of the Primary Product Content of Polycyclic Aromatic Hydrocarbons (PAHs) The levels of benzo[a]pyrene and benz[a]anthracene were shown to be below the detection limits (0.2 µg/kg) of the employed technique (gas chromatography/mass spectrometry in the selected ion Mean (g/l) SD (a) (g/l) RSD (b) % EFSA Journal 2011; 9(7):2307 9

10 monitoring mode) and thus below the maximum permitted values of 10 and 20 µg/kg, respectively, as laid down by Commission Regulation 2065/ No data have been provided on other PAH s. However, taking into account the previously reported limited variability of the contents of these two PAH s and of 13 PAH s listed in Annex 2 of the EFSA guidance document (EFSA, 2005) in 49 batches of the Primary Product Zesti Smoke Code 10 from 2006 (EFSA, 2009a), the Panel considered these data sufficient. 3. Proposed uses The upper use levels as described by the applicant for the Primary Product in each of the 18 food categories as outlined in Commission Regulation (EC) No 1565/ are reported in Table 3. Normal use levels were not provided by the applicant. Table 3: Upper use levels of Primary Product Zesti Smoke Code 10 in food categories as outlined in Commission Regulation (EC) No 1565/ Food categories Upper Use level (g/kg) 1 Dairy products, excluding products of category Fats and oils and fat emulsions (type water-in-oil) 0 3 Edible ices, including sherbet and sorbet Processed fruits Processed vegetables (including mushrooms & fungi, roots & tubers, pulses & legumes) and nuts and seeds 5 Confectionery 0 6 Cereals and cereal products, including flours & starches from roots & tubers, pulses & legumes, excluding bakery 7 Bakery wares 0 8 Meat and meat products, including poultry and game 3 9 Fish and fish products, including molluscs, crustaceans and echinoderms 3 10 Egg and egg products 0 11 Sweeteners, including honey 0 12 Salts, spices, soups, sauces, salads, protein products etc. < 2.75 (a) 13 Foodstuffs intended for particular nutritional uses Non-alcoholic ("soft") beverages, excl. dairy products Alcoholic beverages, incl. alcohol-free and low-alcoholic counterparts 0 15 Ready-to-eat savouries 3 16 Composite foods (e.g. casseroles, meat pies, mincemeat) - foods that could not be placed in categories 1-15 (a): The following upper use levels for the breakdown of the food category 12 Salts, spices, soups, sauces, salads, protein products etc. were provided by the applicant and used to assess the exposure: 2.75 g/kg for herbs, spices, seasonings and condiments (food category 12.2), 0.5 g/kg for soups and broths (food category 12.5) and 0 for the remaining subgroups. Normal use levels were not provided by the applicant Regulation (EC) No 2065/2003 of the European Parliament and of the Council of 10 November 2003 on smoke flavourings used or intended for use in or on foods. OJ L 309, , p Commission Regulation (EC) No 1565/2000 of 18 July 2000 laying down the measures necessary for the adoption of an evaluation programme in application of Regulation (EC) No 2232/96 of the European Parliament and of the Council. OJ L 180, , p EFSA Journal 2011; 9(7):

11 4. Dietary exposure assessment In order to estimate dietary exposure to the Primary Product, the CEF Panel used two different methodologies, namely the Smoke Theoretical Added Maximum Daily Intake (SMK-TAMDI) and the smoke flavouring EPIC model (SMK-EPIC). These methodologies were developed by the Panel specifically for smoke flavourings and are described in detail in the EFSA opinion on dietary exposure assessment methods for smoke flavouring Primary Products (EFSA, 2009b). As reported in Table 4, dietary exposures to Zesti Smoke were estimated by both the SMK-TAMDI and SMK-EPIC methodologies, assuming that the Primary Product is present at the upper use levels provided by the applicant for the 18 food categories as outlined in Commission Regulation (EC). Estimated dietary exposures from all sources were 6.7 and 8.4 mg/kg bw/day, according to the SMK- TAMDI and SMK-EPIC methodologies, respectively. The impact on exposure of using the Primary Product only in traditionally smoked food products was also assessed. Out of the above mentioned 18 food categories, Dairy products, excluding products of category 2, Meat and meat products, including poultry and game and Fish and fish products, including molluscs, crustaceans and echinoderms were considered as Traditionally smoked solid foods. In this case the SMK-EPIC model results in the highest exposure estimates: 7.7 mg/kg bw/day when using upper use levels. Table 4: Summary of the dietary exposure estimates to the Primary Product Zesti Smoke Code 10 calculated on the basis of the upper use levels. Methodologies Dietary exposure (mg/kg bw/day) SMK-TAMDI SMK-EPIC Traditionally smoked food 5.0 Other foods not traditionally smoked 1.7 Beverages (alcoholic or non-alcoholic) 0.0 Total dietary exposure 6.7 Traditionally smoked food 7.7 Other foods not traditionally smoked 0.7 Beverages (alcoholic or non-alcoholic) 0.0 Total dietary exposure Toxicological data 5.1. Identity of the test material The data provided regarding major chemical parameters and principal constituents demonstrate that the test material (batch BJW2809) used for the new 90-day oral (dietary) toxicity study is representative of the Primary Product Zesti Smoke Code 10. The specification of the test material administered in an associated 14-day oral (dietary) range finding study (batch BJW2205) was shown to be in compliance with the specifications given for Zesti Smoke Code 10 in the previous EFSA opinion (EFSA, 2009a). EFSA Journal 2011; 9(7):

12 5.2. Subacute and subchronic toxicity The Primary Product Zesti Smoke Code 10 has been tested in a new 90-day oral (dietary) toxicity study, carried out in addition to the two 90-day studies described in the previous opinion on Zesti Smoke Code 10, adopted on 29 January 2009 (EFSA, 2009a). Prior to the conduct of this study, a 14-day oral (dietary) range finding study was carried out in CRl:WI(Han) rats (n = 5 per sex at each dose level), using dose levels of 0, 2.5, 12.5 or 50 g Zesti Smoke Code 10/kg diet. These dietary levels were equivalent in week 2 to 0, 243, 1188 or 5059 mg/kg bw/day in males and 0, 227, 1228 or 4661 mg/kg bw/day in females. Both sexes given 50 g Zesti Smoke/kg diet showed a reduced weight gain compared with controls over days 1-15, accompanied by a reduction in food consumption at this dose level in week 1 of the study. Increases in aspartate aminotransferase and alanine aminotransferase were seen in some animals in each treatment group, compared with controls. No other treatment-related effects were reported. In the main 90-day study, conducted according to OECD TG 408, Zesti Smoke Code 10 was administered to groups of 10 male and 10 female HsdHan :WIST rats at dose levels of 0, 5, 9 or 12 g/kg diet for 13 weeks. These dietary levels were equivalent to a mean intake of Zesti Smoke Code 10 over the 90-day period of 0, 322, 572 or 785 mg/kg bw/day in males and 0, 376, 664 or 910 mg/kg bw/day for females. The new study used a narrower range of dose levels around the NOAEL of 134 mg/kg bw/day in male rats identified from the more recent of the two 90-day studies described in the previous opinion on Zesti Smoke Code 10 (EFSA, 2009a), and thus permitted a more accurate identification of the NOAEL. Investigations comprised assessment of treatment-related effects on mortality, clinical signs, food conversion efficiency, ophthalmoscopy, functional observation battery investigations, locomotor activity investigations, haematology, clinical chemistry, urinalysis, organ weights, macroscopic and microscopic pathology. An assessment of delayed onset toxicity and/or reversibility of toxicity was made in groups of 5 male and 5 female rats during a 13 week treatmentfree period after the main period of dosing. One male in the low dose group (322 mg Zesti Smoke/kg bw/day) was terminated on day 87 of the study on humane grounds, due to a handling error, and another male in this dose group was found dead on day 88, after blood sampling the previous day. There were otherwise no treatment-related effects on survival, and no clinical findings or findings in the functional observation battery that could be attributed to treatment. Body weight gains in males receiving either 572 mg/kg bw/day or 785 mg/kg bw/day were slightly reduced compared with control group, but the reduction was not statistically significant and did not show a dose-response relationship (91.8% of control at 572 mg/kg bw/day and 92.8% of control at 785 mg/kg bw/day). A similar reduction in body weight gain was seen in treated females (95.6% of control at 376 mg/kg bw/day, 91.2% at 664 mg/kg bw/day and 80.5% of control at 910 mg/kg bw/day, the latter reduction being statistically significant compared with controls, P<0.001). Body weight gain during the reversibility phase was significantly increased in males previously receiving 785 mg/kg bw/day (P<0.001) when compared to concurrent controls, and was increased but not significantly in the other male treatment groups. Overall body weight gain in females receiving 910 mg/kg bw/day was still significantly reduced compared with controls at the end of the reversibility period, although weight gain was similar to controls during the reversibility phase. Food consumption was slightly reduced in treated males (mean for all three treatment groups over weeks 1-13: 95% of control) versus concurrent controls in a non-dose-related manner during the dosing period. A slight dose-related reduction was also seen in females receiving 664 (mean over weeks 1-13: 97% of control) or 910 mg/kg bw/day (mean over weeks 1-13: 95% of control). The Panel considered that overall the effects on body weight gain and food consumption in both males and females were indicative of a palatability effect and not of toxicological significance. However the Panel considered that the reduction in body weight gain in females receiving 910 mg/kg bw/day, at 80.5% of control, in the absence of a marked reduction in food consumption in this group compared with other treatment groups, was indicative of an adverse effect. Haematological examinations showed a significant decrease compared to controls in white blood cells and lymphocytes in females receiving 910 mg Zesti Smoke/kg bw/day, with a trend to lower values in the other female treatment groups, albeit not statistically significant. This effect was reversible on EFSA Journal 2011; 9(7):

13 treatment cessation (examination at week 26). Other than this finding, occasional statistical differences between treated groups and controls were inconsistent between sexes and/or were not dose-related, and are not considered to be treatment-related. Similarly, although clinical chemistry revealed occasional statistical differences between treated groups and controls, these were not dose-related and/or were inconsistent between sexes, and are not considered to be treatment-related. No treatmentrelated effects were seen on urinalysis. Other than an increase in relative prostate weights in males receiving 785 mg Zesti Smoke/kg bw/day (P<0.001) when compared to concurrent controls at the end of the reversibility phase, no consistent effects were seen on absolute or relative organ weights. The effect on prostate weight was unlikely to be related to treatment, since no such effect was seen at the end of the treatment period. Macroscopic and microscopic findings were typical of animals of this age and showed no evidence of treatmentrelated effects. The applicant considered that NOAELs of 785 mg Zesti Smoke/kg bw/day and 910 mg Zesti Smoke/kg bw/day could be established for males and females, respectively, in this study. The applicant argued that the reduction in body weight gain in females receiving 910 mg Zesti Smoke/ kg bw/day did not show an unequivocal trend over the dosing period, was not accompanied by any changes in behaviour or pathological changes, and was reversible on cessation of dosing, and was therefore non-adverse in nature. The Panel did not agree with this argument, and considered the LOAEL for Zesti Smoke to be 910 mg /kg bw/day for female rats, based on the reduced weight gain, together with the possible treatment-related effects on white blood cells and lymphocytes in this group. Considering the higher sensitivity of females, the Panel concluded that overall a NOAEL of 664 mg Zesti Smoke/kg bw/day could be established from this study. While the NOAEL for males was higher, at 785 mg Zesti Smoke/kg bw/day, the Panel has conservatively used the lower NOAEL of 664 mg Zesti Smoke/kg bw/day identified in females for the purpose of risk characterisation. The Panel noted that the increases in aspartate aminotransferase and alanine aminotransferase seen in some animals in the 14-day range-finding study were not seen in the 90-day toxicity study. 6. Discussion The present opinion describes the results of new studies (chemical properties and a 90-day toxicity study) provided by the applicant on Zesti Smoke Code 10. The other information required for the evaluation of Zesti Smoke Code 10, in accordance with the Guidance document (EFSA, 2005), is available in the previous opinion published on 11 June 2009 (EFSA, 2009a). The chemical data provided demonstrate that the test material (batch BJW2809) used in the day toxicity study is representative of the Primary Product Zesti Smoke Code 10 (EFSA, 2009a). Evaluation of the new 90-day toxicity study submitted by the applicant indicated an effect on body weight gains in males receiving either 572 mg Zesti Smoke/kg bw/day or 785 mg/kg bw/day, which were slightly reduced compared with control group. However the reduction was not statistically significant and did not show a dose response relationship (91.8% of control at 572 mg/kg bw/day and 92.8% of control at 785 mg/kg bw/day). A similar reduction in body weight gain was seen in treated females (95.6% of control at 376 mg/kg bw/day, 91.2% at 664 mg/kg bw/day and 80.5% of control at 910 mg/kg bw/day, the latter reduction being statistically significant compared with controls, P<0.001). Although comparable weights between treated males and controls were restored during the reversibility phase of the study, this was not the case for females receiving 910 mg Zesti Smoke/kg bw/day. Overall body weight gain in females in this group was still significantly reduced compared with controls at the end of the reversibility period, although weight gain was similar to controls during the reversibility phase. During the dosing period, food consumption was slightly reduced in treated males versus concurrent controls (mean: 95% of control) in a non-dose-related manner and a slight dose-related reduction was also seen in females receiving 664 mg/kg bw/day (mean: 97% of control) or 910 mg/kg bw/day (mean: 95% of control). Other than these effects on body weight gain and food consumption, findings in this 90-day study were largely unremarkable. The only haematological finding indicative of a treatment-related effect was a significant decrease compared to controls in white blood cells and lymphocytes in females receiving 910 mg Zesti Smoke/kg bw/day, with a trend EFSA Journal 2011; 9(7):

14 to lower values in the other female treatment groups, albeit not statistically significant. These changes were not detectable at the end of the recovery period. There was no evidence of any treatment-related changes on clinical findings, functional observation battery, clinical chemistry, urinalysis, macroscopic or microscopic findings. Considering the higher sensitivity of females, overall the Panel identified a NOAEL for Zesti Smoke of 664 mg /kg bw/day, based on the reduced weight gain and possible treatment-related effects on white blood cells and lymphocytes in female rats at the higher dose of 910 mg /kg bw/day. While the NOAEL for males was higher, at 785 mg Zesti Smoke/kg bw/day, the Panel has conservatively used the lower NOAEL of 664 mg Zesti Smoke/kg bw/day identified in females for the purpose of risk characterisation. For comparison, the NOAEL of 664 mg /kg bw/day is about 4 to 5-fold higher than the NOAEL for changes in body weight gain and relative kidney weights identified in the 90-day study performed in 2006 and described in the previous opinion on Zesti Smoke Code 10 (EFSA, 2009a). The Panel concluded that the NOAEL from the new study supersedes that from the previous study, since the new study used a narrower range of dose levels around the previously established NOAEL and thus permitted a more accurate identification of the NOAEL. In that study adverse effects on body weight gain and relative kidney weights were seen in animals dosed at 25 g Zesti Smoke/kg diet, and the NOAEL was identified as 2.5 g Zesti Smoke/kg diet, equivalent to 134 mg/kg bw/day in male rats and 178 mg/kg bw/day in female rats. In order to estimate dietary exposure to the Primary Product Zesti Smoke Code 10, the CEF Panel used two different methodologies, developed by the Panel specifically for smoke flavourings. Dietary exposure estimates were calculated by assuming that the Primary Product Zesti Smoke Code 10 is present at the upper use levels provided by the applicant for the 18 food categories as outlined in Commission Regulation (EC). Dietary exposure from all sources ranges from 6.7 to 8.4 mg/kg bw/day. Normal use levels were not provided by the applicant. When dietary exposure estimates are based on use in only traditionally smoked foods dietary exposure ranges from 5.0 to 7.7 mg/kg bw/day, when assuming that the Primary Product Zesti Smoke Code 10 is present at the upper use levels. The margins of safety are calculated as the ratio between the NOAEL and the estimated human dietary exposure, both expressed as mg/kg bw/day (Table 5). Based on the data above it is concluded that when assuming that the Primary Product Zesti Smoke Code 10 is present at the upper use levels provided by the applicant for the 18 food categories, the margins of safety for the intake estimates as compared to the NOAEL of 664 mg/kg bw/day, identified in females in the 90-day toxicity study in rats, amount to (Table 5).When assuming the use of Primary Product Zesti Smoke Code 10 in traditionally smoked products only, the margins of safety would amount to for the intake estimates based on the upper use levels (Table 5). Table 5: Margins of safety calculated on the basis of the upper use levels. Dietary exposure (mg/kg bw/day) NOAEL (mg/kg bw/day) Margin of safety Total dietary exposure 6.7 (a) / 8.4 (b) (a) / 79 (b) Traditionally smoked food 5.0 (a) / 7.7 (b) (a) / 86 (b) (a): if dietary exposure is estimated on the basis of the SMOKE-TAMDI model (b): if dietary exposure is estimated on the basis of the SMOKE-EPIC model Given i) the fact that these margins of safety are based on a 90-day toxicity study, ii) the absence of data on reproduction and developmental toxicity and iii) the absence of long term studies, it is concluded that the uses and use levels of Primary Product Zesti Smoke Code 10 would require a larger margin of safety. The Panel concluded that the proposed use of the Primary Product at the uses and use levels specified is of safety concern. EFSA Journal 2011; 9(7):

15 The Panel did not anticipate that smoke flavourings would be used in food specifically designed for infants (0-12 months) and children (12-36 months). Therefore the safety of use of Primary Product Zesti Smoke Code 10 in such products was not assessed. CONCLUSIONS AND RECOMMENDATIONS The European Food Safety Authority (EFSA) has been asked by the European Commission (EC) to reconsider its safety assessment of the smoke flavouring Primary Product Zesti Smoke Code 10 based on a new 90-day dietary toxicity study in rats submitted by the petitioner. The other information required for the evaluation of Zesti Smoke Code 10, in accordance with the Guidance document (EFSA, 2005), is available in the previous opinion published on 11 June 2009 (EFSA, 2009a). The compositional data provided for the test material used in the new 90-day study comply with the provisions given in the EFSA guidance document and in legislation. The Panel considered that a NOAEL of 664 mg Zesti Smoke/kg bw/day could be identified in this new study, based on the reduction in weight gain in females receiving 910 mg Zesti Smoke/kg bw/day, together with possible treatment-related effects on white blood cells and lymphocytes in females at this dose level. The Panel concluded that the NOAEL from the new study supersedes the NOAEL of 134 mg/kg bw/day in male rats and 178 mg/kg bw/day in female rats from the previous 90-day study, since the experimental design of the new study used a narrower range of dose levels around the previously established NOAEL and thus permitted a more accurate identification of the NOAEL. Based on these data it is concluded that when assuming that the Primary Product Zesti Smoke Code 10 is present at the upper use levels provided by the applicant for the 18 food categories, the margins of safety for the intake estimates as compared to the NOAEL of 664 mg/kg bw/day derived from the 90- day toxicity study in rats amount to When assuming the use of Primary Product Zesti Smoke Code 10 in traditionally smoked products only, the margins of safety would amount to for the intake estimates based on the upper use levels. Given i) the fact that these margins of safety are based on a 90-day toxicity study, ii) the absence of data on reproduction and developmental toxicity and iii) the absence of long term studies, it is concluded that the uses and use levels of Primary Product Zesti Smoke Code 10 would require a larger margin of safety. The Panel concludes that the margins of safety are insufficient and that the use of Primary Product Zesti Smoke Code 10 at the proposed uses and use levels is of safety concern. To decide whether despite the low margins of safety the use of Primary Product Zesti Smoke Code 10 might be approved for traditionally smoked products, at use levels specified, to replace smoking, is outside the remit of the Panel. DOCUMENTATION PROVIDED TO EFSA 1. Final Report: Flavouring 37735: 14-Day Oral (Dietary) Range-finding Study in the Rat, from Mastertaste, August Final Report: Flavouring 37735: 13 week Oral (Dietary) Administration Toxicity Study in the Rat followed by a 13 Week Treatment-free Period, from Mastertaste, October Revised maximum levels for use of Primary Product Zesti Code Responses from Mastertaste to request for supplementary information (Chemical Analysis and PAH Analysis on the test material used in 2010 Oral Administration Toxicity Study). EFSA Journal 2011; 9(7):

16 REFERENCES EFSA (European Food Safety Authority), Guidance from the Scientific Panel on Food Additives, Flavourings, Processing aids and Materials in Contact with Food. Guidance on submission of a dossier on a Smoke Flavouring Primary Product for evaluation by EFSA. Adopted on 7 October 2004; Revised on 27 April EFSA (European Food Safety Authority), 2009a. Scientific Opinion of the Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids (CEF) on the safety of smoke flavour Primary Product Zesti Smoke Code 10. The EFSA Journal 982, EFSA (European Food Safety Authority), 2009b. Scientific Opinion of the Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids (CEF) on dietary exposure assessment methods for smoke flavouring Primary Products. The EFSA Journal RN-284, LAB, Covance Laboratories Ltd, England. 13 Week Oral (Dietary) Administration Toxicity Study in the Rat with a 13 Week Treatment-free Period. Study Code No: , October EFSA Journal 2011; 9(7):