Strategy support for the Post‐Market Monitoring (PMM) of GM plants: Review of existing PMM strategies developed for the safety assessment of human and animal health

Abstract

Commission Implementing Regulation (EU) No 503/2013 on applications for authorisation of genetically modified (GM) food and feed legalised the requirement for a Post-Market Monitoring (PMM) plan if the outcome of the risk assessment identifies a need to do so. This report provided a systematic review of existing monitoring programmes for food, feed, human and animal health and an inventory of data collection sources that may be useful for PMM of GM food and feed. Under the legislative requirements PMM should confirm that specific recommendations of use are followed by the consumer/animal owner, qualify the predicted consumption of GM food and feed and further characterise any intended or unintended effects highlighted during the pre-market risk assessment. The largest barriers to conducting PMM of GM food and feed in the EU were found to be the ‘unrecognisable’ nature of GM traits further down the food and feed supply chain after processing, and the lack of appropriate consumption data at the branded/product level. EU legislation on traceability and labelling of GMOs requires products to carry a ‘unique identifier’, however, this information is lost once the product is further processed to become a product ‘produced from a GMO’. Consumption data held at a product level by retailers or market research agencies was shown to be more useful for PMM of GM food or feed than that held at the national level by public bodies. The project concludes that several changes would be required in order to conduct comprehensive PMM of GM food and feed in the EU, namely; greater detail on traceability requirements of GMOs, a database of which food and feed products contain which GM traits at specific quantities, consumption data at the branded/product level and a system for reporting the relevance and intensity of effects and unintended effects. Summary Commission Implementing Regulation (EU) No 503/2013 on applications for authorisation of genetically modified (GM) food and feed legalised the requirement for a Post-Market Monitoring (PMM) plan if the outcome of the risk assessment has identified a need to do so. Three specific examples are given in this Regulation: 1) when the GM food or feed has altered nutritional composition; 2) when the nutritional value of the GM food or feed differs from the conventional ones; 3) when there is a likelihood of increased allergenicity due to the genetic modification. The aim of this project was to identify strategies, and existing monitoring or surveillance data that are collected in the EU, and provide recommendations to assist with the development of EFSA guidance for applicants on the design of PMM plans. The project consisted of two main work packages focussed on gathering evidence, and a third work package which consisted of a discussion on how PMM could be conducted in the EU using existing data sources and the methods currently available. The methodology for WP1 and WP2 are detailed below. WP1 – Systematic review of existing monitoring programmes The aim of this work package was to carry out a review of existing PMM programmes used to monitor human and animal health. In this work package the principles of a systematic review methodology were used to; search for, screen and present results of papers for inclusion. The review took into account all PMM programmes, including, those for food, feed, human and animal health. Specifically, programmes identified had a focus on either quantifying consumption of the product being monitored, or identifying adverse/expected effects. The review used structured search criteria, which were developed in ‘Scopus’, the largest abstract and citation database of peer-reviewed literature. The search terms were then used in both ‘Scopus’ and ‘Web of Knowledge’ which returned a total of 3,809 records. The team then applied rigorous selection criteria in two stages, a stage one based on the title and abstract of the paper only, and a stage 2 based on a review of the full paper. Stage 1 reduced the total number of papers to 513 and Stage 2 reduced the total number of papers to 164. Final papers (164) were reviewed in a MS Excel database based on key descriptors relating to the design of programme and summarised in a discussion of results in the final report. WP2 – Inventory of data collection methodologies The objective of Work Package 2 was to identify and review existing survey, monitoring or surveillance programmes for food, feed, human and animal health that may inform the design of a PMM system for GM food or feed. The types of programmes reviewed included existing networks and data sources on food and feed consumption, and animal and human health. Initially, informal consultation was conducted to get an initial appreciation of the types of data collection programmes that may be available, and to direct an internet based search. WP1 SYSTEMATIC REVIEW RESULTS The vast majority of the papers obtained through the systematic review process of existing monitoring programmes were those related to Pharmacovigilance, the name given to the monitoring of adverse effects of medical drugs. Three main stages of Pharmacovigilance were identified through the systematic review; data collection, signal detection/evaluation and the reporting of the signal. In Pharmacovigilance data can be collected both actively and passively. Active surveillance describes methods where the operator instigates work such as conducting questionnaires/interviews with patients or conducting post-clinical trials. Passive surveillance utilises spontaneous reporting systems to pick up on unintended effects of drugs in general, using mediums such as consumer care lines and healthcare professional reporting card schemes e.g. the yellow card system operational in the UK to link adverse effects to specific products. Passive surveillance schemes are easier to conduct than active ones as frameworks for collecting data are already operational in many countries. However, many have criticised this approach due to large scale under reporting of adverse effects by the general public and healthcare professionals and a lack of consistency of information recorded between countries. Active surveillance methods allow for the monitoring of multiple outcomes within the same scheme, and will capture adverse effects even if there is a delay between consumption and the event. The clear disadvantage of active methods are their resource intensive nature. Similar schemes to Pharmacovigilance operate, albeit at a smaller scale for vaccines (Vaccinovigilance) and cosmetics (Cosmetovigilance). These schemes are similar in design to Pharmacovigilance, with country based systems for reporting adverse effects. Monitoring schemes are established in some countries for herbal medicines to understand adverse effects, particularly when they are used in combination with other drugs. Currently, monitoring largely occurs in countries such as Germany where herbal medicines are regulated in the same way as other medicinal drugs (not ‘over the counter’), and thereby their use can be accurately known. Novel foods are the group of products in the literature that have the most applicability to the PMM of GM food and feed. PMM of novel foods is a useful method to clarify expected exposure to the novel food, or novel food ingredient in ‘real world’ populations. PMM of novel foods takes two distinctive steps, 1) clarifying the expected consumption of the food, and 2) identifying adverse effects. Methods of PMM in novel foods have typically employed a combination of active and passive surveillance methods. Active surveillance takes the form of cohort and case-control trials and consumption questionnaires, while passive surveillance utilises spontaneous reporting systems largely via consumer care lines included on the products in question. As the consumer care line is used to report any feedback on the product such as customer complaints, many of the reports received on these lines relate to complaints about the product (such as standards of packaging or organoleptic qualities) rather than reports of adverse health effects from consumption of the product. Our review found that PMM for animal feeds or veterinary products is an unexplored area, with comparatively little focus in the literature. Pharmacovigilance for veterinary medicines operates in a similar way to that for human drugs, however, individual animal owners are encouraged to report adverse effects to a local veterinarian who in turn reports adverse effects. WP2 INVENTORY OF DATA COLLECTION SOURCES RESULTS The overall objective of Work Package 2 was to identify and describe programmes (survey, monitoring or surveillance) that would be relevant for PMM of GM food and feed. The work package majored on food composition and consumption data, as many networks designed to understand adverse effects are covered by the programmes described in WP1. Our analysis started by investigating the extent to which the existing regulatory framework on traceability and labelling may allow a user to understand which GM traits are included in food and feed products. Traceability legislation under Regulation (EC) 178/2002 requires Food Business Operators (FBOs) to understand who products/ingredients are supplied from, and who they are sold to (unless the customer is the end consumer). For GMOs, all products containing, consisting of or produced from GMOs must be positively labelled to indicate their presence. Crucially, these requirements do not require the operator to know exactly which GM trait(s) is included in the product. The system of unique identifiers does require that materials ‘consisting of or containing GMOs’ are labelled with the specific code, but this is not required once those materials are further processed, to constitute a product which is ‘produced from GMOs’. This is for example the case where GM soybeans (the raw material) would be further processed into soybean oil (an ingredient);