4.1 Manufacturing process safety
Cell-cultured meat and seafood safety depends on a manufacturing process designed with product safety in mind and an assessment of the final product. This section describes practices and protocols from related fields that can be translated to the cell-cultured manufacturing context as part of creating safe, consistent, high-quality products.
4.1.1 Good Manufacturing Practice
Good Manufacturing Practice (GMP) relates to the overall “good housekeeping” principles intended to prevent a hazard from occurring and is a set of widely applied food production practices that describe appropriate design and construction of facilities, sanitary operations and maintenance, and production and process controls that ensure reliable results and safe production of food (21 C.F.R. §, 117, 2015; Blanchfield, 2005; De Oliviera et al., 2016). GMPs and standard operating protocols from the food, feed, meat and seafood, and pharmaceutical and medical fields (e.g., De Oliviera et al., 2016) can be applied to cell-cultured meat and seafood manufacturing to ensure consistent quality and safety of the product. In addition, Good Hygiene Practices (GHPs) are essential in the food supply and can be audited alongside GMP compliance. GHPs extend beyond industrial food manufacture into the service industry, such as catering, hotels, and restaurants
4.1.2 Good cell culture practice
Also applicable to cell-cultured meats and seafoods are concepts from Good Cell Culture Practice (GCCP). These principles are typically applied to in vitro systems for basic research, medicines, and pharmacology to maximize the reliability of cell and tissue products but some aspects are relevant for handling and management of cell-cultured meat and seafood (Bal-Price & Coecke, 2011; Hartung et al., 2002). GCCP sets minimum standards and provides recommendations for in vitro work to prevent contamination and ensure quality of the final product. Among other relevant recommendations, GCCP suggests working with aseptic techniques and avoiding antibiotic use, developing standard operating procedures, and controlling the quality of media supplements and other inputs. Documentation is emphasized; maintaining a detailed record of all procedures can provide information on what potential contaminants or hazardous inputs may be present in the final product, which can support targeted screening for potentially harmful impurities and contaminants. The standards set by GCCP is likely prohibitive for food; however, development of “food-grade” GCCP based on existing guidelines may be a good next step.
4.1.3 Code of hygienic practice
Until the relationships between source animal health and final food product are understood for cell-cultured products, guidance regarding the health of food animals or recommendations related to the use of animal-derived materials for medical procedures may be useful. A code of hygienic practice already exists for animal food production, which includes procedures for herd management to maintain animal health and prevention of animal disease (WHO, 2008). Animals used as source animals for xenotransplantation (i.e., use of live cells, tissues, or organs from an animal source in a human recipient) are recommended to be healthy and reside in specific pathogen-free closed herds with health screening programs (European Commission, 2006; U.S. Food and Drug Administration, 2010, 2016). These programs track and monitor infectious diseases, and documentation of animal health history is required. This proactive approach can be especially useful where there are no validated screening tests to detect endogenous pathogens. This is the case, for example, for prion-associated diseases, such as BSE where diagnosis can only be made in a postmortem examination of brain tissue (U.S. Food and Drug Administration, 2019). Active herd/flock management and documentation, along with monitoring and screening of source animals for potential infectious disease, will lower the risk of culturing affected cells. Similarly, isolation of animal-derived components of cell media (e.g., BSA, trypsin, collagen, etc.) from low-risk animals reduces the chance of contamination (Jayme & Smith, 2000).
4.1.4 Hazard and risk management systems
Management systems can help prevent or minimize hazards and manage specific risks within a process. This review identifies some potential biological and chemical hazards; as more data are developed and manufacturing processes evolve, other hazards (e.g., physical) and associated risks (i.e., the likelihood that a hazard poses a significant safety issue) can be assessed. Systems to manage food safety risk such as Hazard Analysis and Critical Control Point (HACCP), Hazard Analysis and Risk-Based Preventative Controls (HARPC), Food Safety Plan development, and other risk-based preventative controls programs may be applied to cell-cultured meat and seafood processes. In each approach, a systematic review is performed for each step of the manufacturing process to identify every possible hazard or source of contamination. A control or procedure is introduced to prevent, eliminate, or minimize each hazard based on its risk (e.g., Johnston, 2000). In many countries, a risk prevention system is a regulatory standard or law and is often considered essential to achieving greater market access (e.g., European Commission, 2004a; U.S. Food and Drug Administration, 1997). Detailed documentation of each process step and identification of potential hazards can help identify the impurities and contaminants that warrant examination in the final product. This review delivers a first step toward hazard identification, recognizing that it is premature to conduct a generalized risk assessment.
4.1.5 Input materials and equipment selection
Input material selection and control provides a second example where more general frameworks can be supplemented with practices from specific fields. Cell culture materials can be selected to comply with current GMP and food-grade specifications. The selection and management of equipment, disposables, and cleaning agents made of food-safe materials will limit the amount of toxic extractables and leachables migrating into the product. Standardized tests for such contaminants to ensure quality and safety can be drawn from the biopharmaceutical, medical device, and cosmetics industries, where much has already been established related to testing regimes for process-related contaminants and residue measurement (Gao & Allison, 2016; International Life Sciences Institute, 2015).
4.1.6 Contaminant control
Every new cell line can be cultivated in a quarantine incubator and verified that they are pathogen free. Microbiological controls and testing derived from practices involving stem cells or in vitro practices can be applied throughout the manufacturing process. Many methods exist to evaluate and reduce contamination from infectious agents introduced via equipment, handling, material inputs, or processes where cultures are exposed to the air. A system of daily observation and regular screening of cultures, media, and equipment using standard protocols can be adapted from those provided in regulatory guidance or pharmacopeial standards (Cobo et al., 2005). Investment in rapid microbiological testing and implementation of effective controls and procedures to limit contamination is essential. Viruses and other undesirable agents can be reduced or removed from serum and final products through heat inactivation, irradiation, or filtration (Jayme & Smith, 2000; Laassri et al., 2018). In addition, cells intended for banking may be screened for bacteria, yeast, fungi, prions, and viruses to prevent unintentional propagation in future batches (Cobo et al., 2005; European Commission, 2004b; U.S. Food and Drug Administration, 1998; U.S. Food and Drug Administration, 2010).
4.2 Demonstrating safety of cell-cultured meat and seafood
The extent of toxicity testing required for cell-cultured products or specific inputs in the manufacturing process is yet to be established. To establish the safety of the final product (which includes cell-cultured meat and seafood as an ingredient, additive, or a whole food), a safety assessment of the inputs and then an evaluation of the types and levels of residues, byproducts, and metabolites remaining in the final product will be necessary. If deemed to have significant novel or unique properties, an assessment of the final product itself as a whole may be needed.
Many existing standard toxicity testing methods may be used to assess inputs. Generally, any inputs into food must be of food-grade quality, meeting specifications and criteria specific to that ingredient (e.g., as specified in the Codex Alimentarius). Development of specifications for cell-cultured meat and seafood additives, such as scaffold materials, may be warranted.
Approaches to safety testing of ingredients and food additives are well established, using biochemical, in silico, in vitro, and in vivo methods, as described in the next sections. Globally harmonized testing standards—such as those developed by the Organisation of Economic Co-operation and Development (OECD), World Health Organization (WHO), Food and Agriculture Organization (FAO), or regulatory organizations—may be applied directly or modified for use in the cell-cultured meat and seafood safety testing context. Tests and analyses under these standards are generally carried out following Good Laboratory Practice (GLP), a set of principles designed to assure study quality and integrity.
Products that have compositional, nutritional, and functional equivalency to already accepted foods are in theory as safe as the products to which they are equivalent (European Food Safety Authority, 2008; U.S. Food and Drug Administration, 2008). The comparison then relies on the history of safe use and data supporting the safety of the conventional food. Any identified differences will direct further safety testing (Constable et al., 2007; European Commission, 1997; FAO/WHO, 2000). It is expected that some cell-cultured products will not be exactly the same as their conventional counterparts. For example, cell-cultured products may contain synthetic scaffold materials or other novel inputs, the cells may be genetically modified such that new proteins are expressed or existing proteins are under- or overexpressed, and the biochemistry and composition of those proteins may vary. Accordingly, toxicity testing may be required to demonstrate the safety of inputs and components in the final product.
4.2.1 Microbiological analysis
Typically, microbiological limits are established for conventional livestock or aquaculture products (e.g., Government of Canada, 2020). Guidance has been developed to help identify microbiological hazards in meat, poultry, seafood, and other animal proteins (U.S. Department of Agriculture, 1999). Bacterial and viral contamination may be detected through routine process monitoring; physicochemical changes, pH shifts, changes in turbidity, and compromised cell cultures can signal contamination. Existing standards, guidelines, and specifications for microbiological characterization are likely applicable, employing conventional techniques such as plate counting methods and immunoassays, as well as more efficient techniques including molecular methods (e.g., polymerase chain reaction) and enzyme-linked immunosorbent assays (ELISA). Biosensor technology may also be applied in real-time to screen and detect microbial contamination of meat products (Sionek et al., 2020). Standard methods exist to detect and quantify common microbiological hazards, such as Salmonella, Listeria, and E. coli (U.S. Department of Agriculture, 2020a). Guidance on evaluation of viruses and mycoplasma in products derived from cell lines of animal origin is available for biotechnological products. Infectivity, electron microscopy, reverse transcriptase, antibody production tests, and in vitro assays using susceptible indicator cells may be used to detect viruses (European Medicines Agency, 1997; U.S. Food and Drug Administration, 2010). Mycoplasma can be assessed using nucleic acid amplification technique-based assays, DNA staining, and culture methods (Nübling et al., 2015; U.S. Food and Drug Administration, 2010). Although it is currently unknown whether cell-cultured product manufacturing may pose any unique microbiological hazards, no novel pathogens are expected. An evaluation of whether existing microbiological criteria for conventional meat and seafood products are applicable to cell-cultured products is warranted.
Microbiological challenge testing may be a useful approach to evaluate any potential hazards arising from storage or food processing. Pathogenic organisms are intentionally introduced to food, then products are treated or stored under realistic conditions and analyzed for any physicochemical changes, microbiological growth, or hazardous degradation products (Komitopoulou, 2011). This testing can provide information on product stability and the effectiveness of procedures designed to eliminate pathogens.
4.2.2 Residue, contaminant, and byproduct analysis
The presence of any drugs, additives, processing aids, and contaminants needs to be considered and analyzed. Although some substances are not intentionally included in the final product, residues could carry over from the manufacturing process. Limits and maximum impurity or residue levels (metals, natural toxins, agricultural or veterinary chemicals, environmental contaminants) are established for conventional livestock or aquaculture products in many jurisdictions (e.g., Food Standards Australia & New Zealand, 2021; Government of Canada, 2020). The World Health Organization (WHO) has developed a list of antimicrobials that should not be used in animals due to their critical importance for human medicine (WHO, 2019). Most antibiotic drugs currently approved for use in food animals are also approved for human use (National Research Council, 1999). It remains to be determined whether these existing criteria for conventional products require modifications or if additions are warranted for cell-cultured meats.
Companies that use novel inputs may need to develop and validate their own analytical tests to identify any residues, contaminants, or byproducts in the final product. Many chemical hazards (e.g., dissociation reagents, cryoprotectants) may be screened using conventional analytical methods such as mass spectrometry, chromatography, and immunological techniques (Chiou et al., 2015; Toldrá & Reig, 2006), though sample preparation may require modification for the cell-cultured meat matrices. Bioassays have been developed to detect a wide range of residues in conventional meat products. For example, U.S. Department of Agriculture (2011) endorses a multiple bioassay method designed to screen meat and poultry for common antibiotic groups, after which specific techniques can be used for full identification and quantification. Bioassays may also be used as a screening tool for currently unknown or unexpected hazards (e.g., migrants from equipment). Determining whether the sensitivity and range of the tests are adequate for the various inputs used in cell-cultured products, or whether the techniques will require modification, is an important topic for future research.
4.2.3 Biochemical, molecular, physical, and compositional analyses
Biochemical, molecular, physical, and compositional analyses can be used as part of a comparative approach to assess the similarity to existing products. Analyses developed for safety assessment of GE food and feed and cloned animals intended for food are anticipated to apply to cell-cultured products, whether genetically modified or not (EFSA, 2008; WHO, 2008).
A molecular and biochemical analysis of cell-cultured products can help determine the extent of any differences in the genome and confirm intentional effects or identify unintentional expression of products not normally seen in meat or seafood (Sewalt et al., 2016; Stout et al., 2020). Any expression products may be compared to conventional products to identify any new or increased hazards related to consumption. Safety assessment of a GE fish, AquAdvantage salmon, and a GE pig, GalSafe pig, determined that the introduced DNA was safe for the resulting GE animal and its offspring, and that the animals are safe to eat (U.S. Food and Drug Administration, 2017a). The safety assessments relied on determining the health of the animal, as a healthy animal is likely to be safe to eat. Phenotypic characterization as well as compositional and nutritional analysis of the edible tissues was performed to ensure that there were no biologically relevant differences between the GE animals and comparator conventional animals (U.S. Food and Drug Administration, 2017b, 2020).
Methods already exist to characterize GE animals intended for use as food (WHO, 2008). As part of this analysis, the genome sequence is evaluated to determine whether the inserted genetic material changes essential gene function, to identify whether or not there are new and unintended open reading frames, and to ensure that no genes code for known toxins or antinutrients (WHO, 2008). Similarly, methods exist for biochemical and proteomic analyses to assess expression of new products and identify differences in protein, peptide, amino acid, and metabolite levels as compared to conventional meats (WHO, 2008). Any newly expressed or altered proteins may affect product stability or physical properties and alter their toxic or allergenic potential. The assessment of a novel protein may focus on amino acid sequence similarity to known toxins or allergens (e.g., Ladics et al., 2011); if significant homology is found, then further testing may be performed (e.g., stability or digestibility in the human body, toxicity testing of that protein).
The cells themselves are typically monitored throughout the process for quality control measures, which can provide an indication of cell health. For example, physicochemical properties, proliferation potential, differentiation capacity, karyotype stability, and the expression of specific cell markers to validate identity can provide valuable safety information and identify any unwanted physicochemical transformations.
Compositional analysis is likely to be a key element of a comparative safety assessment. The analysis may include an assessment of macro- and micronutrients, bioactive compounds, toxins, and allergens. This evaluation can provide a baseline to compare cell-cultured meat with a conventional product (U.S. Food and Drug Administration, 2008; Williams, 2007); products that are similar to conventional meat are more likely to be processed and metabolized similarly and may rely on safety assessment of its conventional counterpart. A number of agencies publish specifications for meat and seafood products; these describe protein and fat content, mandatory ingredients (such as specific cuts of meat, bone, blood, organs, or skin), and optional ingredients (such as spices, filler, flavor enhancers, or water). However, even among conventional meats, composition can vary. Fatty acid profiles, for instance, can differ significantly between organic versus nonorganic products or between different muscles of the same species (Ros-Freixedes & Estany, 2014; Średnicka-Tober et al., 2016; U.S. Food and Drug Administration, 2008). Research is needed to evaluate applicability of existing specifications, characterize how cell-cultured products may differ in composition from their conventional counterparts, and determine to what extent those differences may influence safety.
4.2.4 In vitro testing: Cytotoxicity and microbiome assessment
In vitro tests may be a desirable starting point for safety testing as they are more efficient and less resource intensive than in vivo testing and also avoid or reduce animal testing. In vitro testing can be used to screen for and identify potential hazards and is sometimes used to aid in dose selection for conventional animal tests. In vitro testing is typically performed on ingredients rather than whole foods, as test substances must be solubilized in media. To perform these tests on whole foods, samples would need to be freeze dried and homogenized or processed in some manner. This presents a technical challenge and likely does not accurately represent the final product. More research is required to determine whether and how in vitro safety testing can be effectively applied to whole foods. Regardless, in vitro methods may be useful for endpoint testing for any inputs to the process or novel proteins, contaminants, degradation products, metabolites, or byproducts present in the final cell-cultured product.
Beyond genotoxicity and allergenicity testing, few in vitro tests relevant to food safety have been validated for stand-alone use. Nonstandard test methods exist, such as cytotoxicity, digestion, and microbiome tests. However, these types of tests generally lack regulatory acceptance as a complete replacement for animal studies, but may be useful as supporting information as part of a broader safety assessment strategy. Cytotoxicity studies can be used as a screening tool, and may be more sensitive than in vivo tests for demonstrating safety at the cellular level. Primary cells or co-cultures of cells representing the gastrointestinal tract are used to mimic realistic scenarios (EFSA Scientific Committee, 2018; Pradhan et al., 2020). In vitro digestibility testing can analyze the stability and digestibility of foods under processing conditions (e.g., heat, freezing) or with simulated saliva and gastrointestinal fluids; the products of these conditions can then be used in toxicity tests to dose relevant cells, such as stomach cells. To determine their safety (Astwood et al., 1996; EFSA Scientific Committee, 2018; Pradhan et al., 2020).
With increased recognition of the role of the microbiome in maintaining health, in vitro assays to measure positive or negative impacts on the gut microbiome may be important to investigate. The gut microbiome is a complex ecosystem of microorganisms that support physiological, biochemical, and immunological function (McBurney et al., 2019; Roca-Saavedra et al., 2018). The presence of residues (particularly antibiotics), metabolites such as growth factors, and contaminants in food, or changes in micronutrient composition, such as vitamins, iron, and fatty acids, can alter the microbiota composition (Roca-Saavedra et al., 2018). Microbiome communities are highly diverse and individualized, and their relationship to adverse human health outcomes is still not well understood, even for conventional foods (McBurney et al., 2019). In vitro microbiome assessment for cell-cultured meat and seafood would require further research.
4.2.5 Allergenicity testing
Allergenicity is a key focus of food safety assessment. For GE foods, comparative testing has been used to assess allergenicity (EFSA Scientific Committee, 2018; WHO, 2009). GM plants, such as potatoes and soy beans, have generally demonstrated similar allergenicity to their conventional counterparts (Gizzarelli et al., 2006; Lee et al., 2006), which is also expected for cell-cultured meats that are manufactured to mimic existing products; but this has not yet been demonstrated. There is also the potential to reduce the allergenicity of products, for example, through removal of alpha-gal sugars, which has been demonstrated in live pigs (U.S. Food and Drug Administration, 2020).
If any novel proteins are expressed in cell-cultured meat or seafood, in silico assessments can evaluate sequence homology and identify structural similarities to existing proteins (Ladics et al., 2011); this characterization can help identify toxic or allergenic properties (EFSA, 2008). There are a multitude of existing allergenicity tests, including the pepsin resistance test, immunochemical cross-reactivity testing with IgE from serum of individuals known to be allergic to similar proteins, in vitro IgE-binding tests such as the radioallergosorbent test or ELISA, and skin prick testing (EFSA, 2008; U.S. Food and Drug Administration, 2015). The use of animal models to identify human sensitivity to novel allergens may not be reliable (Melo et al., 1994) or necessary unless in silico or in vitro tests indicate a need for further testing.
4.2.6 Genotoxicity testing
A number of validated in vitro genotoxicity tests screen endpoints such as potential mutagenic activity, DNA strand breaks, and cytogenicity (e.g., OECD 476, 2016; OECD 490, 2016; OECD 487, 2016; OECD 473, 2016; OECD 471, 2020) and results of these tests may predict mutagenicity or carcinogenicity. This testing will be useful in identifying potentially genotoxic inputs to the manufacturing process. If it is deemed necessary to apply these tests to whole foods rather than select inputs, some of these techniques may require modification. The most common genotoxicity test, the Ames bacterial mutagenicity test (OECD 471, 2020.), for example, may not be appropriate for meats high in histidine (e.g., pork, beef, lamb, chicken, turkey, fish) because this amino acid interferes with the test (Aeschbacher et al., 1987). If a review of in vitro tests and available toxicokinetic data indicate the possibility of genotoxic effects, in vivo genotoxicity tests may be considered (e.g., OECD 486, 1997; OECD 478, 2016; OECD 489, 2016; OECD 475, 2016; OECD 474, 2016; OECD 488, 2020), though this is not expected for cell-cultured meats.
4.2.7 In vivo testing
Cell-cultured meat and seafoods that are biochemically, genetically, and compositionally similar to existing foods should theoretically be as safe as their conventional counterparts. There is uncertainty as to whether in vivo testing will be required for novel inputs or products with significant differences from existing foods (e.g., because they contain potentially hazardous proteins or metabolites, or lack a conventional counterpart). In some regulatory jurisdictions, testing in rodents remains a required baseline study for novel foods, and may help assuage safety concerns. However, from an industry perspective, where the avoidance of animal use is a key tenet, the performance of in vivo testing to demonstrate safety is not desirable. Some regulatory jurisdictions promote alternative testing strategies (i.e., nonanimal testing) where possible; however, the availability and validation of reliable and representative in vitro tests that represent food consumption and mimic the human gastrointestinal tract that can fully replace in vivo testing remains a barrier and a research need.
A subchronic 90-day feeding trial (OECD 408, 2018), where rodents are fed a test substance daily for 90 days, is typically a fundamental element of ingredient safety testing. This test serves as a basis to demonstrate safety for food, feed, pharmaceuticals, agricultural products, pesticides, contaminants, and industrial chemicals. The study assesses body and organ weight, feed consumption, blood and urine chemistry, histopathology, and animal behavior to determine direct or systemic effects resulting from consuming the food. This test is generally accepted as sufficient to identify adverse effects that could occur after repeat and chronic exposure to a substance (EFSA, 2008). Ingredients or chemically defined substances such as a micronutrients or amino acids can be mixed into specially formulated diets to test in vivo safety. (Roper et al., 2019). However, animal feeding trials may not be appropriate for whole foods such as meat and seafood (EFSA, 2008; Kok et al., 2008; WHO, 2008). To improve the reliability of results obtained from animal-based toxicity testing toward the human safety context, animals are generally fed the test substance at levels that exceed those expected for human consumption of the product. However, animals may not find the food palatable, and feeding abnormally high doses of food may cause nutritional imbalances in the diet (Knudsen & Poulsen, 2007; EFSA, 2008). Therefore, the amount of whole food that can be incorporated into the test animal diet is limited by bulk and nutritional imbalance, and the detection of adverse effects resulting from any toxicants or antinutrients is likely to be missed.
Further studies may be warranted, particularly where biochemical, in silico, or in vitro tests also indicate potential concern. If genotoxicity or subchronic testing of ingredients suggests a need for chronic or carcinogenicity studies, standard tests for chronic testing (OECD 452, 2018), carcinogenicity testing (OECD 451, 2018), or combined chronic/carcinogenicity testing (OECD 453, 2018) exist. If there are any indications from subchronic studies that reproductive organs or systems involved in development may be affected, then in vivo reproductive and developmental toxicity testing may be performed. Tests such as two-generation reproductive toxicity studies (OECD 416, 2001) or prenatal developmental toxicity studies (OECD 414, 2018) may be applicable. If subchronic testing and allergy testing demonstrate possible immunotoxic effects, further investigation of the endpoints assessed in subchronic tests may be warranted. For example, histological assessment of lymphoid organs and tissues, and hematological assessment of various cells and immunoglobulin levels may give further indication of immunotoxic effects. As with all in vivo studies, the development of alternative testing methods to effectively replace these tests is preferred and is a research priority for safety testing of cell-cultured products.
4.2.8 Human studies
For food ingredient safety assessment, a demonstration of similarity to foods that have a history of safe consumption and/or alternative testing studies and animal studies have typically been accepted as sufficient to demonstrate safety, and few foods have required human studies. Human studies may be used for whole foods, but are typically related to tasting/palatability, short-term testing for digestibility and tolerance, allergenicity, testing in support of health claims, or for specialized foods where there is a need to investigate potential negative nutritional effects or adverse health outcomes on specific populations (e.g., food for infants and children, pregnant women, patients at increased risk of a disease, etc.) (Agriculture & Agri-Food Canada, 2013; EFSA Scientific Committee, 2018).
A research gap regarding safety of foods derived from modern technology is whether recombinant DNA (rDNA) in meat is capable of transferring to microbiota in the gastrointestinal tract. In theory, horizontal gene transfer (HGT)—a process by which rDNA can pass from one species to another (e.g., to human gut microorganisms or microorganisms in the environment)—could occur. For example, antibiotic resistance genes are sometimes inserted into plants or microorganisms to distinguish GE cells during cell selection (EFSA, 2008; WHO, 2009); HGT could result in the development of populations of antibiotic-resistant organisms, reducing the effectiveness of current antibiotics (Maghari & Ardekani, 2011). Most research to date has focused on HGT potential from GE prokaryotes and plants, and has generally demonstrated that HGT events are rare, or are eventually lost due to a lack of conferred advantage (Nielsen et al., 2014; Rizzi et al., 2012). Regardless, the insertion of antibiotic resistance genes in food producing GE organisms is discouraged and even prohibited in some jurisdictions. Few studies have researched the risk of gene transfer from GE mammalian cells. Scientists have performed stability studies testing the degradation of DNA in saliva and gastrointestinal fluid, experimental studies using recipient bacteria or cells in vitro, and in vivo studies feeding animals or volunteers the sample then testing for rDNA in the body (Martín-Orúe et al., 2002; Netherwood et al., 2004; EFSA, 2008). If a cell-cultured process introduces a genetic modification, an assessment of whether the modification could introduce a fitness advantage may be warranted; this can help characterize the likelihood that the change would persist in the unlikely event of HGT.
4.2.9 Postmarketing monitoring
Postmarketing monitoring, where large populations of consumers are evaluated over the long term, has been used to complement premarket safety assessment for some novel foods (Hepburn et al., 2008; Wal et al., 2003). A postmarketing program may help detect rare and unintended adverse effects such as allergic responses. Such approaches have already been successfully applied in the medical field (Howlett et al., 2003), but it may be challenging to adapt the approach to cell-cultured products. Pharmaceuticals have well-controlled dosages, and adverse outcomes are relatively easy to track in the medical context. By contrast, it is far more difficult to monitor the adverse health effects resulting from long-term consumption of food (Hepburn et al., 2008; Howlett et al., 2003). However, pre-identification of potential hazards (such as growth factors) and tracking-related adverse outcomes may be merited. Some food manufacturers have developed monitoring systems to obtain feedback from consumers; these systems rely on various strategies such as consumer reporting of adverse effects via hot lines, performing household panels, interacting with market research companies and consumer associations, evaluating supermarket data, and engaging with medical professionals and scientific agencies (Wal et al., 2003).