Health Risks Related to Presence of Modified Forms of Certain Mycotoxins - EFSA

Following a request from the European Commission, the EFSA Panel on Contaminants in the Food Chain (CONTAM Panel) was asked to deliver a scientific opinion on the risk to human and animal health related to the presence of metabolites and the masked or bound forms of fumonisins, zearalenone, T-2 and HT-2 toxins and nivalenol in food and feed. Modified forms of deoxynivalenol are covered by a separate request. The scientific opinion should comprise the evaluation of the toxicity of these metabolites and masked or bound forms of these mycotoxins for animals and humans compared with the toxicity of the parent mycotoxins. Furthermore, it should contain an assessment of the co-occurrence of the metabolites and masked or bound forms of these mycotoxins in food and feed and an estimation of the human dietary exposure and the animal feed exposure compared with the dietary exposure to the parent mycotoxins. Finally, an assessment of the human and animal health risks should be performed.



The Panel has applied the term modified mycotoxins for masked and bound mycotoxins and mycotoxin metabolites. Modified mycotoxins have up to recently only occasionally been detected when analysing the mycotoxins in their original form, and therefore they have commonly been termed as “masked”. In this opinion, the structurally altered forms of mycotoxins covered by the terms of reference are referred to as “modified mycotoxins”. The mycotoxins in their unchanged forms are referred to as “parent compounds”.



Modified mycotoxins are found in plants resulting from plant defence reactions after fungal infection or are produced by the fungus itself. Modified mycotoxins are metabolites of the parent mycotoxin formed in the plant or fungus, e.g. by conjugation with polar compounds. The most common conjugations are with glucose and modified glucose but also with other groups such as sulphate and glutathione. For fumonisins there is evidence that there are forms that are covalently or non-covalently bound to the matrix. Therefore the term modified fumonisins includes both covalently and non-covalently (i.e. physically entrapped) bound forms for the purpose of this opinion.



Protocols for modified mycotoxin analysis are mainly based on water/acetonitrile extraction followed by analysis with liquid chromatography–tandem mass spectrometry (LC-MS/MS). Both targeted and untargeted methods are used. Immunochemical methods for mycotoxins may not detect modified forms. Concerning fumonisins and their entrapped forms, variability due to different extraction strategies is high. There is a lack of properly validated routine methods owing to the lack of commercially available calibrants and reference materials.



Modified mycotoxins can be released, hydrolysed, biotransformed and absorbed in the gastrointestinal tract, primarily as the parent compound. Although the toxicity of conjugated mycotoxins is not addressed in the present opinion, data in the literature shows that conjugates of xenobiotics can be of toxicological significance. Based on this the  CONTAM Panel assumed, as a pragmatic approach, that all modified forms have the same  toxicity as their parent compounds.



Bran- and fiber-enriched products are more prone to contamination with mycotoxins, including their modified forms. The EFSA occurrence database contains no data on modified mycotoxins covered by the present opinion. Therefore, occurrence is based on limited information reported in the literature. Literature data shows that modified forms of mycotoxins may add substantially to the overall mycotoxin levels, in particular for zearalenone and fumonisins. For fumonisins, the major contribution comes from physically entrapped parent compounds, whereas for the other compounds, metabolites are the main contributors to modified forms. In order to assess occurrence and exposure, the CONTAM Panel added 100 %, 30 %, 10 % and 60 % to the levels of the parent compounds to account for  the modified forms of zearalenone, nivalenol, the sum of T-2 and HT-2 toxins and fumonisins, respectively.



Occurrence data about modified mycotoxins in animal products (i.e. milk and dairy, meat, eggs) are not available, but occurrence in animal products is expected to be very low, since carry-over of Fusarium toxins from feed is considered insignificant for human exposure based on currently available data.



The risk characterization for humans is performed by comparing the estimated combined exposure to parent and modified forms of mycotoxins with the established health based guidance values (HBGVs) for the respective parent compounds.



Using the lower bound (LB) approach, no consumers with mean or 95th percentile exposure have combined exposure to zearalenone and modified zearalenone above the tolerable daily intake (TDI) set for zearalenone, indicating no concern. However, if the concentration of modified and parent zearalenone in cereals would be closer to the upper bound (UB) level, high consumers (95th percentile) may have an exposure up to 2.2-fold the TDI for zearalenone of 0.25 µg/kg body weight (b.w.) per day, which would be of concern.



Exposure to the sum of nivalenol and modified nivalenol is not of concern, as the highest 95th percentile UB exposure across studies is less than 20 % of the TDI for nivalenol of 1.2 µg/kg b.w. per day.



Exposure to the sum of T-2 and HT-2 toxins and their modified forms is considered not to be of concern, since all the LB and UB exposures across studies were lower than the TDI of 0.1 µg/kg b.w. per day for the sum of T-2 and HT-2 toxins, with the exception of the highest UB exposure derived for toddlers' (= 1 year to < 3 years) exposure, which was similar to the TDI.



The exposure to fumonisins and their modified forms could be of concern, especially in children's age groups.  In toddlers, 0.2-14 % (LB) or 43-66 % (UB) and in other children 0-38 % (LB) to 11-59 % (UB) across studies could exceed the provisional maximum TDI (PMTDI) for fumonisins of 2 µg/kg b.w. per day. At high (95th percentile) exposure, the maximal exceedance was 2.5-fold (LB) to 3-fold (UB) the PMTDI.



In order to assess occurrence in feed and exposure of animals, the CONTAM Panel added equal factors for modified mycotoxins in feed as in food.



Levels in feed have been compared with guidance values for feed, if established, and estimated exposure in animals is evaluated in relation to established NOAELs/LOAELs  for different animals, when available. Furthermore, known differences in sensitivities to mycotoxins between different species were taken into account when NOAELs/LOAELs were lacking. For fumonisins, contaminated feed has been used in the studies used for derivation of NOAELs/LOAELs, with the exception of poultry, for which the pure compound has been added to feed. For the other Fusarium toxins, experiments were carried out with application of pure compounds.



For zearalenone and its modified forms, the mean exposure in pigs in general and the 95th percentile exposure in piglets is not of concern. Occurrence data were inadequate to conclude on risks for fattening pigs and sows receiving feed with a higher than average contamination level. Owing to the lack of either occurrence data or NOAELs/LOAELs, the risk for cattle, goats, rabbits and fish cannot be characterised fully. Estimated exposure in sheep, horses, poultry, cats and dogs is not of concern.



For nivalenol and its modified forms, exposure in pigs, ruminants, poultry, horses, rabbits, cats and dogs is not of concern. Risk could not be assessed in fish because of the lack of occurrence data in feed.



For T-2 and HT-2 toxins and their modified forms, the estimated exposure in pigs, ruminants, horses, poultry, and dogs is not of concern. Risk could not be assessed in fish because of the lack of occurrence data in feed. There is no NOAEL/LOAEL for cats, which are known to be sensitive to T-2 and HT-2 toxins. A concern cannot be excluded in cats but is unlikely based on low exposure.