Periodic Reporting for period 2 - ADDITIVES (Exposure to ‘cocktails’ of food additives and chronic disease risk)
Reporting period: 2021-11-01 to 2023-04-30
Diets have been shifting towards a dramatic increase in the consumption of ultra-processed foods (i.e. foods undergoing multiple physical, biological, and/or chemical processes and containing various food additives), while epidemiological evidence linking such consumption to adverse health outcomes is gradually accumulating as highlighted in more than 60 prospective studies worldwide. The NutriNet-Santé cohort in France was among the largest contributors to these results that emphasized the need of public health research on the topic. Food additives, substances added to foodstuffs for technological purposes (e.g. dyes, preservatives, emulsifiers, sweeteners), are a prime hypothesis to explain these results. Their use is widespread in industrially-processed foods, often containing a mixture of additives. As a result, millions of people daily ingest dozens of additives provided by multiple food products. While current evidence suggests that most additives found on the food market are likely to be neutral for health, concerning results of mainly in-vitro/in-vivo experimental studies have emerged regarding several compounds, used in hundreds of commonly consumed foods and beverages, such as TiO2, nitrates/nitrites, artificial sweeteners, phosphates, some emulsifiers, carrageenans, and glutamate, suggesting metabolic, gut microbiota and endocrine perturbations along with carcinogenic, inflammatory or oxidative stress effects. Despite the substantial work and collective expertise performed by health authorities to protect consumers against potential adverse effects of each substance in a given food product, their evaluation is based on the scientific evidence available to-date which is derived mainly from in-vitro and/or in-vivo toxicological studies and simulations of exposure in humans. Meanwhile, information regarding 1) the health impact of chronic and cumulative intake of food additives in humans, and 2) potential ‘cocktail’ effects (interactions) of multi-additive mixtures is still missing. In this context, the aims of ADDITIVES are to 1) estimate individual chronic exposure to food additives (as individual substances and as multi-additive mixtures), 2) investigate the relationships of additive exposure with risk of obesity, cancer, cardiovascular diseases, type 2 diabetes and mortality, 3) study biological mechanisms underlying those relationships, using a multi-disciplinary and multi-faceted approach.
One of the main strength of the project is the unique and highly detailed data collected in the NutriNet-Santé cohort (n=173 000 adults 18+, France, 2009-ongoing), including repeated 24h dietary records with commercial names/brands of industrial products consumed by the participants. This is a key information, given the huge variability of additive composition from one brand to another. In order to evaluate participants’ exposure to food additives, for each food (or beverage, candy, chewing-gum) consumed, the presence of each additive (qualitative composition) and its dose (quantitative composition) were searched by the research team. To determine the qualitative presence/absence of food additives, three databases were used: OQALI, OpenFoodFacts and Mintel GNPD. When several composition data existed for a same product at different dates (reformulations), the date of consumption in the cohort (year) was considered in the matching of composition data (dynamic matching). The quantitative composition of additives has been derived from several sources. Firstly, 2677 ad-hoc laboratory assays in food matrixes were carried out prioritizing the most consumed additives and those with suspected health effects. The second choice was the use of doses by food categories transmitted by EFSA. Last, doses from the Codex General Standard for Food Additives (GSFA) were used. This allowed us to have detailed quantified data on food additives exposure, with a unique level of detail. This first descriptive study allowed us to observed that food additives, including several for which health concerns are currently debated, were widely consumed by to population. Adapted statistical methods also allowed us to identify 6 clusters of participants in the NutriNet-Santé cohort, exposed to different mixtures of food additives. Etiological analyses enabled us to highlight links between nitrites food additives and cancer, type-2 diabetes and hypertension risk, and between artificial sweeteners (especially aspartame, acesulfame potassium, and sucralose) and cancer and cardiovascular diseases risk. Furthermore, associations between trans-fatty acids (ubiquitous in ultra-processed foods) and type-2 diabetes risk have been observed. Several of these scientific publications were selected for press release and were widely disseminated to the public (thousands of media fallouts around the world). Besides, they were also presented in the framework of official auditions at the French National Assembly and Senate. Simultaneously, we have built a case-cohort study nested in the NutriNet-Santé biobank (blood samples from n=6,100 participants) and markers of inflammation, oxidative stress and metabolomics are being measured in order to identify mechanisms and pathways underlying the associations between additives, ultra-processed food and health. Furthermore, biomarkers of exposure to 5 selected food additives have been identified and are currently being tested on urine samples of the participants.
This project has so far allowed the identification of several pioneering results on the associations between some food additives and health, thanks to the detailed assessment provided by the comprehensive NutriNet-Santé questionnaires and a linkage to several databases, and assays. In next steps, analyses are currently ongoing to explore the associations between human health and other classes of food additives, such as emulsifiers and dyes (the preliminary results have been presented in several international congresses). Then, we will investigate the links between the main identified mixtures/cocktails of food additives and the risk of cancer, cardiovascular diseases, diabetes, obesity and mortality. In parallel, these mixtures are currently being tested in experimental models in collaboration with toxicologists, to explain the underlying mechanisms. This will allow to achieve for the first time, a multi-disciplinary approach to study the impact of food additives on human health. Other analyses are currently ongoing to quantify the exposure to food additives in the ESTEBAN representative survey in France, and in the EPIC European cohort, in order to confirm these associations in another largescale European sample. Stool collection is currently ongoing and the analyses between the exposure to food additives and gut microbiota will be shortly starting afterwards. Taken together, these results will create a solid scientific basis to guide food policies towards a better regulation of the use of food additives by the food industry, to further protect the consumers.