Pos0102 Dietary Exposure to Non-Dioxin-Like Polychlorobiphenyls Increases the Risk of Rheumatoid Arthritis in the French E3n Cohort

Background: Rheumatoid arthritis (RA) is an autoimmune disease whose established risk factors are active and passive smoking, obesity and, with less evidence, female hormones and certain dietary profiles. The role of intestinal dysbiosis has also been suggested. Exposure to pollutants, particularly those with endocrine-disrupting effects, has been implicated in the growing incidence of autoimmune diseases, although no specific data are available for RA [1]. Non-dioxin-like polychlorinated biphenyls (PCB-NDL) and dioxins (including dioxins and dioxin-like PCB), whose sources are mainly dietary (cheese, butter, fish, meat), are groups of contaminants stored in adipose tissue and with estrogen-like effects. Objectives: To assess the impact of dietary exposure to dioxins and PCB-NDL on the risk of developing RA in the E3N prospective cohort of French women. Methods: The French E3N cohort included 98,995 women, followed prospectively since 1990. Incident cases of RA in this cohort were validated [2]. A detailed, validated dietary questionnaire was collected in 1993. From this questionnaire, average daily dietary intakes of dioxins and PCB-NDL were estimated on the basis of dietary contamination data from the second French total diet study conducted by ANSES (EAT2). Hazard ratios (HRs) for the risk of incident RA and their 95% confidence intervals [95% CI] were estimated by Cox models adjusted for age (model 1) and for the main confounding factors (model 2: adjustments for age, active smoking, passive smoking, level of education, physical activity, BMI, total energy intake excluding alcohol, total alcohol consumption/day). Analyses were performed on the entire population, and stratified according to active smoking (never/ever) and body mass index (BMI). A sensitivity analysis was performed by further adjusting daily omega3 and calcium intakes, reflecting dietary intakes of fish and dairy products. Results: The overall analysis population comprised 63,022 women, 486 of whom had incident RA occurred after collection of the dietary questionnaire. In the overall population, a statistically significant association was found between the risk of incident RA and dietary exposure to PCB-NDL, for model 1 only (HR=1.28 [1.00-1.63], ptrend=0.04). After stratification and in for both models, high exposure (4th quartile) to PCB-NDL was significantly associated with the risk of RA, compared with low exposure (1st quartile) in non-smoking women (model 2, HR=1.75 [1.20-2.60], ptrend=0.005) and in women with a BMI < 25 kg/m2 (model 2, HR=1.35 [1.02-1.90], ptrend=0.027). Results were similar after further adjustment for omega3 and calcium intakes. With regard to dietary dioxin exposure, the results in the overall population and after stratification were not statistically significant. Conclusion: Dietary exposure to PCB-NDL was significantly associated with an increased risk of developing RA in non-smoking women and in women of normal BMI in the French E3N cohort. These results bring new arguments to the hypothesis of dysbiosis induced by dietary exposure to PCN-NDL and their "estrogen-like" effects [3]. REFERENCES: [1] Merrheim J et al. Estrogen, estrogen-like molecules and autoimmune diseases. Autoimmun Rev. 2020. [2] Nguyen Y et al. Improving accuracy of self-reported diagnoses of rheumatoid arthritis in the French prospective E3N-EPIC cohort: a validation study. BMJ Open. 2019. [3] Popli S et al. Persistent organic pollutants in foods, their interplay with gut microbiota and resultant toxicity. Sci Total Environ. 2022. Acknowledgements: NIL. Disclosure of Interests: Carine Salliot novartis, Roche Chugai, Pfizer, novartis, Océane Mohamed: None declared, Yann Nguyen: None declared, Fanny Artaud: None declared, Xavier Mariette: None declared, Francesca Mancini: None declared, Raphaèle Seror Novartis.