DOP047Circulating metabolites unveil a link between ultra-processed food and the development of Inflammatory Bowel Disease: a multi-cohort study among Western and Eastern populations

Abstract Background Increased intake of ultra-processed food (UPF) links to higher risk of inflammatory bowel disease (IBD)1, while its biological mechanism remains unclear. Our previous work implied gene-diet interaction modulates the risk of IBD, especially in those with higher IBD susceptibility2,3. This study is to unveil UPF’s effect on IBD via circulating metabolites, localize key component, and explore its mechanism linking UPF to IBD. Methods We aim to find the mechanism of UPF on incident IBD through key metabolite and related gene. Analyses involved two parts: a. construct UPF metabolic signature in UK Biobank (UKB) discovery cohort (n = 10,229), with internal (n = 91,306) and external validation (Whitehall II WHII, n = 7,893); b. identify and validate key metabolite in UPF-IBD association and its mechanism in Western and Eastern (n = 752) cohorts. UPF metabolic signature was developed by elastic net regression of metabolites (assayed by high-throughput metabolomic platforms) on UPF intake (measured by repeated dietary recalls). We applied Cox regression to assess the metabolic signature on IBD risk. Weighted gene co-expression network analysis (WGCNA) was used to identify metabolite clusters in the signature. We combined linear model, elastic net regression, survival analysis, mediation analysis, and WGCNA to identify key metabolites. Further validation was performed in an Eastern multi-center cohort, ONE-IBD. Mendelian randomisation (MR), colocalization, and gene-environment (G-E) interaction analyses were used to address potential causality and genetic locus (Fig 1). Results A UPF metabolic signature of 73 metabolites was constructed (Fig 2A) and validated in UKB and WHII (Spearman ρ: 0.20-0.25). More pronounced UPF metabolic signature was associated with higher risk of IBD and CD (IBD: HRper SD=1.72, 95% CI 1.32-2.24; CD: HRper SD=2.65, 95% CI 1.57-4.48 Fig 2B). WGCNA revealed a cluster rich in fatty acids inversely associated with CD (HRper SD=0.66, 95% CI 0.45-0.98, Fig 2C), in which docosahexaenoic acid (DHA) was pinpointed as key metabolite (mediated proportion=12.5%, Fig 2D-E). External validation in ONE-IBD supported DHA as the metabolite most strongly associated with UPF and CD (Fig 2F). MR and colocalization revealed rs174546 in FADS1 gene as a shared variant linking DHA with CD (Fig 2G-H). G-E analysis showed UPF’s effect on CD was pronounced in individuals with FADS1 variant (T allele carriers, linking to lower DHA), while metabolic signature’s effect on CD was weakened, stressing the central role of metabolites (Fig 2I). Conclusion Our study reveals the critical role of metabolites linking UPF and IBD risk, especially in CD, with consistent evidence identifying DHA and FADS1 gene as personalised targets in IBD precision prevention. References: 1.Chen J, Wellens J, Kalla R, Fu T, Deng M, Zhang H, Yuan S, Wang X, Theodoratou E, Li X, Satsangi J. Intake of Ultra-processed Foods Is Associated with an Increased Risk of Crohn’s Disease: A Cross-sectional and Prospective Analysis of 187 154 Participants in the UK Biobank. J Crohns Colitis. 2023 Apr 19;17(4):535-552. 2.Sun Y, Yuan S, Chen X, Sun J, Kalla R, Yu L, Wang L, Zhou X, Kong X, Hesketh T, Ho GT, Ding K, Dunlop M, Larsson SC, Satsangi J, Chen J, Wang X, Li X, Theodoratou E, Giovannucci EL. The Contribution of Genetic Risk and Lifestyle Factors in the Development of Adult-Onset Inflammatory Bowel Disease: A Prospective Cohort Study. Am J Gastroenterol. 2023 Mar 1;118(3):511-522. 3.Chen J, Dan L, Yuan S, Fu T, Sun J, Wolk A, Ludvigsson JF, Li X, Wang X, Larsson SC. Dietary Antioxidant Capacity, Genetic Susceptibility and Polymorphism, and Inflammatory Bowel Disease Risk in a Prospective Cohort. Clin Gastroenterol Hepatol. 2025 Aug;23(9):1623-1632.e16. Conflict of interest: Wang, Sidan: No conflict of interest Dan, Lintao: No conflict of interest Ruan, Xixian: No conflict of interest Wellens, Judith: No conflict of interest Sun, Yuhao: No conflict of interest Yao, Jialu: No conflict of interest Tian, Li: No conflict of interest Kalla, Rahul: No conflict of interest Theodoratou, Evropi: No conflict of interest Yuan, Shuai: No conflict of interest Larsson, Susanna C: No conflict of interest Jonas, Ludvigsson: Dr Ludvigsson has received financial support from Merck/MSD for a study on inflammatory bowel disease and fibrosis and for developing a paper reviewing national healthcare registers in China. Dr Ludvigsson also has an ongoing research collaboration on celiac disease with Takeda. Earlier support includes a grant from Janssen for an unrelated study on behalf of the Swedish IBD quality register (SWIBREG). Peyrin-Biroulet, Laurent: No conflict of interest Satsangi, Jack: No conflict of interest Magro, Fernando: Fernando Magro served as speaker and received honoraria from Abbvie, Arena, Biogen, Bristol-Myers Squibb, Falk, Ferring, Hospira, Janssen, Laboratórios Vitoria, Pfizer, Lilly, Merck Sharp & Dohme, Sandoz, Takeda, UCB, Vifor. Li, Xue: No conflict of interest Wang, Xiaoyan: No conflict of interest Dr. Chen, Jie: No conflict of interest