Risk assessment of replacing synthetic preservatives with natural ingredients Authors: European Food Safety Authority (EFSA), Tatiane C G Oliveira, F Chamorro, Eliana Pereira, Sara Sousa, Fátima Barroso, Crtistina Delerue‐Matos, Valentina F Domingues, Lillian Barros, Miguel A Prieto Published: 2025 ABSTRACT AbstractThe present work, developed within the EU‐FORA fellowship programme, assessed the risk assessment of vegetable by‐product extracts as natural alternatives to synthetic preservatives in food formulations. Extracts were obtained from onion peel (Allium cepa L.), zucchini (Cucurbita pepo L.) and kale (Brassica oleracea L. var. acephala), which represent abundant residues with limited commercial value but high potential for valorisation. Suitable extraction techniques were applied to each matrix, and the resulting extracts were characterised through phenolic profiling, antioxidant activity, cytocompatibility and pesticide residue analysis. Onion peel extract was rich in quercetin glycosides and showed strong antioxidant activity, while zucchini and kale extracts contained quercetin or kaempferol derivatives with moderate effects. Cytocompatibility assays demonstrated acceptable safety margins at lower concentrations, though dose‐dependent toxicity appeared at higher levels, particularly in hepatic cells. Pesticide residues were mostly absent or present only at trace levels in zucchini and kale extracts. In onion peel extract, hexachlorobenzene and β‐HCH were detected, but remained below EU maximum residue limits (MRLs). Overall, the findings suggest that the benefits of these extracts, including preservative potential and sustainable waste valorisation, outweigh the identified risks.Keywords: bioactive potential, cytocompatibility, pesticide residues, preservative ingredients, vegetable by‐products SUMMARY This report presents the results of a risk assessment of natural ingredients obtained from vegetable by‐products, associated with the benefits that can be incorporated, performed within the framework of the EU‐FORA fellowship programme. The aim was to explore the potential of onion peel, zucchini and kale extracts as natural alternatives to synthetic preservatives, combining food safety considerations with sustainability goals. The study followed the EFSA guidance on risk assessment, integrating data on chemical composition, bioactivity, cytocompatibility and pesticide residues. Extracts were obtained using appropriate extraction techniques adapted to each by‐product, ensuring efficient recovery of phenolic compounds. The phenolic characterisation showed that onion peel and zucchini were rich in quercetin glycosides, while kale contained kaempferol derivatives. Onion peel extract presented the highest phenolic content and exhibited strong antioxidant activity, whereas zucchini and kale extracts showed moderate antioxidant effects. The biological evaluation demonstrated that cytocompatibility was dependent of the concentration of the extract. In gastrointestinal Caco‐2 cells, all extracts maintained ≥ 70% viability up to 2500 μg/mL, suggesting low intestinal toxicity. In hepatic AML12 cells, onion peel extract preserved ≥ 70% activity only up to 1250 μg/mL, zucchini up to 2500 μg/mL and kale up to 5000 μg/mL, indicating greater sensitivity of hepatic cells to high doses. Pesticide occurrence was also assessed. Zucchini and kale extracts contained only trace, non‐quantifiable residues (< method detection limit (MDL) or < method quantification limit (MQL)). In onion peel extract, low levels of hexachlorobenzene (HCB) and β‐hexachlorocyclohexane (β‐HCB) were detected, but these concentrations were below the maximum residue limits (0.01 mg/kg) established by EU legislation. Since results were expressed on a dry extract basis, they represent a worst‐case scenario compared to raw material levels. Overall, the integration of results indicates that the benefits of using vegetable by‐product extracts outweigh the identified risks. These extracts provide significant antioxidant activity, potential preservative function and sustainability benefits by promoting waste valorisation. Risks are limited to dose‐dependent cytotoxicity at high concentrations and low‐level pesticide residues in onion peel, which do not exceed regulatory thresholds. The findings support the safe use of onion peel, zucchini and kale extracts as natural ingredients, provided application levels remain within cytocompatible ranges and pesticide monitoring is maintained. INTRODUCTION Synthetic preservatives are important ingredients in the food industry that provide shelf life by controlling microbial growth and preventing oxidation and browning of products. In the same intensity that synthetic preservatives are important in the food and pharmaceutical industries, their consumption has been related to various adverse effects in the human body, such as neurological, gastrointestinal and dermatological disorders, as well as the development of different cancerous tumours (Caleja et al., 2016; Carocho et al., 2014). As a result, consumers are increasingly concerned about synthetic additives and have been seeking products formulated with natural ingredients. This has driven a growing demand for ingredients of natural origin, which, in addition to their preservative action through antimicrobial and antioxidant properties (Shah & Mir, 2022), may contribute to human healthy by combating free radicals and potentially helping to prevent diseases such as cancer, Alzheimer's, diabetes, allergies and cardiovascular and respiratory conditions (Carocho et al., 2018). This growing demand has led to increased interest in plant‐derived bioactive compounds, particularly phenolic compounds, which are among the most promising candidates for replacing synthetic preservatives (Shah & Mir, 2022). Phenolic compounds are obtained from plant sources and are typically safe for human consumption (Tresserra‐Rimbau et al., 2018). Their hydroxyl groups confer both strong antioxidant potential and antimicrobial activity, which are desirable in food preservation (Alara et al., 2021). However, this same property can also pose toxicity risks to humans at high concentrations, making it essential to determine safe incorporation limits in food formulations (Chen et al., 2024). In the development of natural preservatives, a structured risk–benefit assessment is essential to ensure consumer safety while maximising potential functional advantages. Following the guidance provided by the European Food Safety Authority (EFSA) (EFSA Scientific Committee, 2010), this approach requires clear identification of the ingredient, its intended use and the target population, followed by evaluation of both favourable and adverse effects. For phenolic‐rich extracts obtained from vegetable by‐products, beneficial effects are associated with antimicrobial and antioxidant properties that may enhance food preservation and contribute to human health. Potential adverse effects may arise from contamination with pesticide residues or toxicity at high concentrations. The assessment, therefore, integrates compositional analysis (phenolic profile), determination of potential contaminants and in vitro evaluation of cytocompatibility in relevant human cell models, providing a scientific basis for establishing safe use levels in food applications (Zeng et al., 2023). Vegetable by‐products have therefore been widely studied as potential preservative and functional ingredients. For example, Fernandez et al. (2020) evaluated the stability of beet leaf extract‐enriched smoothies over 21 days of storage at 5°C, showing reduced microbial counts and extended shelf life compared to controls, as well as increased total phenolic content without significant losses during storage. Gong et al. (2022) assessed beet extract in cooked pork for Listeria monocytogenes control, observing significant microbial reductions without compromising product quality. dos Santos Silva et al. (2022) investigated oxidative stability in beef burgers enriched with microencapsulated grape pomace extract, demonstrating better lipid oxidation control than a conventional synthetic antioxidant, without affecting colour during storage. Based on these findings, further studies addressing both the preservative efficacy and the safety profile of natural ingredients are essential. Evaluating their chemical composition, possible contamination with pesticides and cytocompatibility is crucial to define safe application levels and to encourage their adoption by the food industry. Such an approach can support the partial or total replacement of synthetic preservatives, promoting more sustainable practices, foods with greater nutritional and functional value and guaranteed safety for consumers. BACKGROUND AND TERMS OF REFERENCE AS PROVIDED BY THE REQUESTOR Description of work programme As part of this programme, the overall objective was to assess the health risks associated with the obtention and consumption of natural ingredients derived from vegetable by‐products, to support the replacement of synthetic preservatives in food formulations. The study addressed key risk questions, namely the potential benefits of incorporating natural ingredients, the possible risks associated with their consumption, and the likelihood of pesticides being present in vegetable by‐products and their extracts. To achieve this objective, the work programme was divided into four main tasks. First, a literature review was conducted to gather scientific evidence on the use of natural ingredients derived from vegetable by‐products, focusing specifically on their preservative potential and safety aspects. Second, extracts were obtained from vegetable by‐products, such as onion peel, kale, and zucchini, which did not meet commercial quality standards, using an extraction method designed to maximize the recovery of phenolic compounds. Third, the extracts were submitted to in vitro characterization through antioxidant activity assay, cytocompatibility tests on gastrointestinal (Caco‐2) and hepatic (AML12) cell lines, and chromatographic analysis of phenolic compound profiles and pesticide residues. Finally, the data from the compositional analysis, biological activity, pesticide residue testing, and cytocompatibility assessments were integrated into a risk assessment framework, enabling the evaluation of both potential benefits and adverse effects, and providing a scientific basis for the safe use of vegetable by‐products as natural ingredients for food preservation. Additional activities for training and networking All these activities were conducted as part of the European Food Risk Assessment (EU‐FORA) Fellowship Programme for the 2024–2025 cycle. The programme is designed to broaden the expertise available in the field of food safety and to strengthen the European network for food risk assessment. Within the 12‐month EU‐FORA fellowship, the fellow participated in a structured training scheme, which included an initial three‐week introductory course followed by four advanced modules of 1 week each, amounting to 7 weeks of dedicated training. In addition, the fellow completed a 3‐month stay at the Nutrition and Food Group (NuFoG) of the Department of Analytical Chemistry and Food Science, Instituto de Agroecoloxía e Alimentación (IAA), University of Vigo, Spain, under the supervision of Dr. Miguel Ángel Prieto Lage, and also visited the Spanish Agency for Food Safety & Nutrition (AESAN) in Madrid and the laboratory in the National Centre for food (CNA) located in Majadahonda, Spain, that was possible learn more about how they conducted the food safety. The EU‐FORA program also allowed the fellow to participate in a conference, through oral communication: Oliveira, T.C.G., Chamorro, F., Carpena, M., Prieto, M.A., Raimundo, D., Ferreira, I.C.F.R., Oliveira, M.B.P.P., Pereira, E., and Barros, L. Sustainable bioactive recovery: The most efficient quercetin extraction from onion peel. Presented at XXIII European Conference on Food Chemistry, 11‐13 June 2025, Bratislava, Slovakia. Oral presentation. Oliveira, T.C.G., Chamorro, F., Carpena, M., Prieto, M.A., Raimundo, D., Ferreira, I.C.F.R., Oliveira, M.B.P.P., Pereira, E., and Barros, L. Sustainable bioactive recovery: The most efficient quercetin extraction from onion peel. Presented at XXIII European Conference on Food Chemistry, 11‐13 June 2025, Bratislava, Slovakia. Oral presentation. MATERIALS AND METHODS Plant material The risk assessment was conducted on extracts obtained from onion peel (Allium cepa L.), zucchini (Cucurbita pepo L.) and kale (Brassica oleracea L. var. acephala) (Figure 1), used as representative vegetable by‐products. The raw materials were supplied by a company in Portugal and originated from vegetables that did not comply with commercial quality standards for human consumption. Vegetable by‐products used in the study, common and scientific names. Images sourced from Unsplash/Pexels (free licence). Obtention of natural ingredients from vegetables by‐products Extracts from onion peel, zucchini and kale were obtained using different extraction techniques selected for each raw material. Microwave‐assisted extraction (MAE) was applied for onion peel, while pressurised liquid extraction (PLE) was employed for zucchini and kale. The operational conditions were chosen based on prior experimental design studies aimed at maximising the recovery of phenolic compounds. The extracts obtained under these conditions were subsequently used for the risk assessment analyses. Characterisation of natural ingredients from vegetable by‐products Phenolic compounds identification The phenolic profiles of the extracts were analysed using an ultra‐performance liquid chromatography system (UPLC, Dionex Ultimate 3000, Thermo Scientific, San Jose, CA, USA) equipped with a diode array detector and coupled to a mass spectrometer (LTQ XL, Thermo Finnigan, San Jose, CA, USA) operating in negative ion mode. Data acquisition was performed with Xcalibur® software. Identification of the main phenolic compounds was based on retention times, UV–Vis spectra and mass fragmentation patterns, compared with available reference standards and reported literature. Only major compounds corresponding to the most intense chromatographic peaks were considered. In vitro antioxidant activity The antioxidant potential of the extracts was assessed using two complementary colorimetric assays: thiobarbituric acid reactive substances (TBARS) and 1,1‐diphenyl‐2‐picrylhydrazyl (DPPH) radical scavenging. In vitro cytocompatibility The cytocompatibility of the extracts was evaluated in gastrointestinal (Caco‐2) and hepatic (AML12) cell lines, obtained from a commercial supplier. Cells were seeded in 96‐well plates and exposed to different concentrations of the extracts for 24 h. Cell viability was assessed using the resazurin reduction assay, and results were expressed as a percentage of metabolic activity relative to the negative control. Pesticide occurrence The presence of organochlorine pesticides (OCPs) in the extracts was assessed. For this, sample cleanup was performed using solid‐phase extraction as described by Punín Crespo and Lage Yusty (2006). Analysis was conducted with gas chromatography equipped with an electron capture detector, using a capillary column suitable for pesticide detection. Helium served as the carrier gas and nitrogen as the make‐up gas. Chromatographic conditions were optimised to enable the separation and identification of target compounds, including dichlorodiphenyltrichloroethane (DDT), dichlorodiphenyldichloroethane (DDD), aldrin, dieldrin, endrin, endosulfan I and II, hexachlorobenzene (HCB), various isomers of hexachlorocyclohexane (α‐, β‐, γ‐ and δ‐HCH) and methoxychlor. Method validation was conducted in accordance with European Commission guidelines, encompassing assessments of linearity, accuracy and detection and quantification limits. ASSESSMENT The vegetable by‐products chosen for this study – onion peel, zucchini and kale – are important sources of phenolic compounds and show the potential for adding value to raw materials that do not meet commercial quality standards. Besides their bioactive potential, using these by‐products promotes sustainability and circular economy strategies by reducing food waste. Different extraction techniques were applied to each vegetable by‐product based on prior optimisation studies. This approach allowed the recovery of extracts with high phenolic content and ensured that the subsequent characterisation reflected the most suitable conditions for each matrix. While this strategy supports maximising bioactive potential, it also emphasises the importance of assessing potential toxicological risks related to higher phenolic concentrations. Characterisation of natural ingredients Phenolic compounds identification The extracts showed distinct phenolic profiles. Onion peel and zucchini extracts were characterised by quercetin glycosides, while kale extracts contained acylated and glycosylated derivatives of kaempferol. Onion peel presented the highest phenolic content, as reflected by the chromatographic peak intensities. These findings are consistent with the literature and support the potential of these compounds to contribute to antioxidant and preservative effects. Antioxidant activity evaluation The antioxidant potential was strong for onion peel extract and moderate for zucchini and kale extracts. This behaviour is consistent with the phenolic composition identified and suggests that onion peel is particularly promising as a source of natural preservatives with high radical‐scavenging capacity. Cytocompatibility assessment Cytocompatibility was dose‐dependent across the tested concentrations (156–10,000 μg/mL), as shown in Figure 2. In gastrointestinal cells (Caco‐2), all extracts maintained ≥ 70% metabolic activity up to 2500 μg/mL, suggesting low intestinal toxicity in this range. In hepatic cells (AML12), onion peel extract maintained ≥ 70% activity up to 1250 μg/mL, while kale extract was tolerated up to 5000 μg/mL. Zucchini extract preserved high viability up to 2500 μg/mL, but above this concentration, cell viability dropped markedly. These results indicate that hepatic cells are more sensitive than intestinal cells, and toxicity is more evident at higher doses, particularly for onion peel and zucchini. Metabolic activity of AML12 (up) and Caco‐2 (down) cells after 24 h of exposure to onion peel, kale and zucchini extracts. The horizontal line indicates 70% and 100% cell viability. Values expressed as mean ± standard deviation (n = 3). Different letters indicate significant differences (p < 0.05; Tukey's test). Pesticide occurrence The residues of OCPs analysed in vegetable by‐product extracts are presented in Table 1. It showed that onion peel extract had low levels of HCB (0.4 ± 0.1 mg/kg) and β‐HCH (0.5 ± 0.3 mg/kg), while all other compounds were not detected. In zucchini extract, only trace levels of endrin and methoxychlor were observed, below the method detection limit (< MDL). For the kale extract, a trace amount of endrin was detected below the method quantification limit (< MQL). TABLE 1Concentration of organochlorine pesticides (OCPs) tested in onion peel, zucchini and kale extracts. Results expressed in mg OCPs/kg dry extract.CompoundOnion peel extract (mg/kg dry extract)Zucchini extract (mg/kg dry extract)Kale extract (mg/kg dry extract)DDEn.d.n.d.n.d.Aldrinn.d.n.d.n.d.DDT + DDDn.d.n.d.n.d.Dieldrinn.d.n.d.n.d.Endosulfan In.d.n.d.n.d.Endosulfan IIn.d.n.d.n.d.Endrinn.d.< MDL< MQLHCB0.4 ± 0.1n.d.n.d.α‐HCHn.d.n.d.n.d.β‐HCH0.5 ± 0.3n.d.n.d.γ‐HCHn.d.n.d.n.d.δ‐HCHn.d.n.d.n.d.Methoxychlorn.d.< MDLn.d. Note: Methoxychlor − LOD = 0.02 mg/kg and LOQ = 0.0514 mg/kg; Endrin − LOD = 0.02 mg/kg and LOQ = 0.065 mg/kg.Abbreviations: < MDL, below method detection limit; < MQL, below method quantification limit; DDD, dichlorodiphenyldichloroethane; DDE, 2,2‐bis(p‐chlorophenyl)‐1,1‐dichloroethene; DDT, dichlorodiphenyltrichloroethane; HCB, hexachlorobenzene; HCH, hexachlorocyclohexanes; n.d., not detected. Concentration of organochlorine pesticides (OCPs) tested in onion peel, zucchini and kale extracts. Results expressed in mg OCPs/kg dry extract. Note: Methoxychlor − LOD = 0.02 mg/kg and LOQ = 0.0514 mg/kg; Endrin − LOD = 0.02 mg/kg and LOQ = 0.065 mg/kg. Abbreviations: < MDL, below method detection limit; < MQL, below method quantification limit; DDD, dichlorodiphenyldichloroethane; DDE, 2,2‐bis(p‐chlorophenyl)‐1,1‐dichloroethene; DDT, dichlorodiphenyltrichloroethane; HCB, hexachlorobenzene; HCH, hexachlorocyclohexanes; n.d., not detected. According to the EU Pesticides Database (European Commission, 2025, Regulation (EC) No 396/2005), the maximum residue limits (MRLs) for the commodities corresponding to onion, zucchini and kale are set at 0.01 mg/kg for these OCPs. In this study, however, results are expressed on a dry extract basis, which concentrates residues relative to the fresh raw material. Therefore, the values detected in onion peel extract are not directly comparable to the regulatory limits but still indicate that contamination is negligible. The presence of persistent compounds such as HCB and β‐HCH highlights the importance of continued monitoring, although levels remain very low in the context of risk assessment. Risk–benefit integration The integration of chemical composition, bioactivity, cytocompatibility and pesticide occurrence provides an overall picture of the safety and functionality of the studied extracts. From the perspective of benefits, onion peel, zucchini and kale extracts presented relevant phenolic compounds, with onion peel showing the highest content of quercetin glycosides and strong antioxidant activity, while zucchini and kale displayed moderate activity. These properties support their potential as natural preservatives and contribute to the valorisation of vegetable by‐products in line with circular economy principles. In terms of risks, cytocompatibility results indicated acceptable safety margins at lower concentrations, with dose‐dependent cytotoxicity appearing at higher levels, particularly in hepatic cells. Regarding pesticide residues, zucchini and kale extracts contained only non‐quantifiable traces (< MDL or < MQL), while onion peel extract showed low levels of HCB and β‐HCH. Although results are expressed on a dry extract basis, and therefore represent a worst‐case scenario compared with raw vegetables or food formulations, the concentrations detected were consistent with the EU maximum residue limits of 0.01 mg/kg established for the studied commodities (European Commission, EU Pesticides Database). Considering both aspects, the risk–benefit balance suggests that the benefits of using these extracts as natural ingredients outweigh the identified risks, provided that concentration ranges remain within cytocompatibility limits and pesticide levels continue to comply with EU safety thresholds. With this, Table 2 summarises the integration of phenolic composition, antioxidant activity, cytocompatibility, pesticide occurrence and the overall risk–benefit evaluation. TABLE 2Risk–benefit assessment of extracts from vegetable by‐products.Extract sourceMain phenolic compoundsAntioxidant activityPesticides residuesPotential hazardsPotential benefitsRisk–benefit conclusionOnion peel (Allium cepa L.)Quercetin glycosidesStrongLow levels of persistents OCPs detected, below EU safety limitsHepatic cytotoxicity at high doses; presence of persistent OCPsHigh antioxidant effect; preservative potential; by‐product valorisationBenefits outweigh risks, but continued monitoring of pesticide residues is recommendedZucchini (Cucurbita pepo L.)Quercetin glycosidesModerateOnly trace residues (< MDL) negligible for risk assessmentCytotoxicity above 2500 μg/mL in hepatic cells; negligible pesticide residuesModerate antioxidant effect; waste valorisationBenefits outweigh risks, but continued monitoring of pesticide residues is recommendedKale (Brassica oleracea L. var. acephala)Kaempferol acylated/glycosylated derivativesModerateTrace residues (< MQL), negligible for risk assessmentCytotoxicity above 5000 μg/mL; negligible pesticide residuesModerate antioxidant effect; functional ingredient potential; sustainable valorisationBenefits outweigh risks at applicable concentrations Note: EU safety limits: 0.01 mg pesticide/kg vegetable.Abbreviations: < MDL, below method detection limit; < MQL, below method quantification limit; OCPs, organochlorine pesticides. Risk–benefit assessment of extracts from vegetable by‐products. Note: EU safety limits: 0.01 mg pesticide/kg vegetable. Abbreviations: < MDL, below method detection limit; < MQL, below method quantification limit; OCPs, organochlorine pesticides. CONCLUSION The present work demonstrated that extracts obtained from onion peel, zucchini and kale are rich in phenolic compounds with relevant antioxidant potential, supporting their application as natural alternatives to synthetic preservatives. Cytocompatibility assays indicated safe margins at lower concentrations, while toxicity emerged only at higher doses, particularly in hepatic cells. Pesticide residues were mostly absent or present at trace levels, with low detections in onion peel extract remaining below EU maximum residue limits. Overall, the risk–benefit balance supports the safe use of these vegetable by‐product extracts within appropriate concentration ranges, highlighting their potential for food preservation and their contribution to sustainability through by‐product valorisation. RECOMMENDATIONS The EU‐FORA fellowship provided valuable opportunities to build knowledge and practical skills across multiple phases of risk assessment. The interdisciplinary teams established among participating institutions represent a key strength of the programme, laying the groundwork for continued cooperation in risk‐related scientific research. ABBREVIATIONS Spanish Agency for Food Safety & Nutrition laboratory in the National Centre for food dichlorodiphenyldichloroethane dichlorodiphenyltrichloroethane 1,1‐diphenyl‐2‐picrylhydrazyl European Food Risk Assessment Fellowship Programme hexachlorobenzene hexachlorocyclohexane (α‐, β‐, γ‐, δ‐isomers) Instituto de Agroecoloxía e Alimentación limit of detection limit of quantification microwave‐assisted extraction method detection limit method quantification limit maximum residue limit Nutrition and Food Group organochlorine pesticides pressurised liquid extraction thiobarbituric acid reactive substances COPYRIGHT FOR NON‐EFSA CONTENT EFSA may include images or other content for which it does not hold copyright. In such cases, EFSA indicates the copyright holder and users should seek permission to reproduce the content from the original source. ACKNOWLEDGEMENTS The fellow wishes to thank the following for the support provided to this scientific output by acknowledging the participating institutions: IPB affiliation and the Nutrition and Food Group (NuFoG), Department of Analytical Chemistry and Food Science, Instituto de Agroecoloxía e Alimentación (IAA) — Universidade de Vigo. A special thank you is also extended for the support of Cristina Alonso Andicoberry, Francesca Manzoni, Christoph Unger and all the fellows of this EU‐FORA cohort. The Panel [EFSA] wishes to acknowledge all European competent institutions, Member State bodies and other organisations that provided data for this scientific output. This work received financial support from the PT national funds (FCT/MECI, Fundação para a Ciência e Tecnologia and Ministério da Educação, Ciência e Inovação) through the project UID/50006—Laboratório Associado para a Química Verde—Tecnologias e Processos Limpos and the Ibero‐American Program on Science and Technology (CYTED— GENOPSYSEN, 223RT141). EFSA (European Food Safety Authority), Oliveira, T. C. G. , Chamorro, F. , Pereira, E. , Sousa, S. , Barroso, F. , Delerue‐Matos, C. , Domingues, V. F. , Barros, L. , & Prieto, M. A. (2025). Risk assessment of replacing synthetic preservatives with natural ingredients. EFSA Journal, 23(S1), e231117. 10.2903/j.efsa.2025.e231117 Approved: 29 September 2025 The declarations of interest of all scientific experts active in EFSA's work are available at https://open.efsa.europa.eu/experts. REFERENCES