ASSOCIATION OF ULTRA-PROCESSED FOODS WITH GUT MICROBIOME, GUT PERMEABILITY, AND FECAL CALPROTECTIN IN HEALTHY FIRST DEGREE RELATIVE OF CD PATIENTS

Background:
The incidence of inflammatory bowel disease (IBD) is increasing globally. Epidemiological studies suggest an association between consumption of ultra-processed foods (UPFs) and IBD risk. However, the relationship among UPFs, the human gut microbiome and inflammatory markers remains relatively unexplored. We aimed to investigate the association between UPFs, as defined by the NOVA classification system, and previously reported biomarkers of Crohn’s disease (CD) risk.

Methods:
Biospecimens were collected from healthy first-degree relatives (FDRs) enrolled in the Crohn’s Colitis Canada - Genetic, Environment, Microbial (CCC-GEM) Project. Dietary data were collected by food frequency questionnaires; UPF consumption was estimated by the NOVA classification system. Subclinical inflammation was measured using fecal calprotectin (FCP, n=2683). Gut barrier function was estimated by urinary fractional extraction ratio of lactulose to mannitol (LMR). Fecal microbiome composition was determined by 16S rRNA sequencing. Alpha diversity was measured using the Shannon index, and Maaslin2 R package was used to identify significant taxa. General estimating equation and Conditional Logistic Regressions assessed the impact of UPFs on FCP. p<0.05 and q<0.05 when accounting for multiple testing were considered significant.

Results:
This study included 2,683 healthy FDRs of CD recruited between 2008-2017 (n=1,420 female [53%], n=59 [2%] who developed CD). Females and adults had lower mean consumption of UPF than males and children, respectively (p-value <0.001). We found that high UPF intake was associated with decreased alpha diversity (p-value =5.5^-4) and significant enrichment of Ruminococcus torques (β=0.0065, q-value =7.77×10^-4), and Lachnoclostridium (β=0.0081, q-value =8.0×10^-5) which are two taxa that our group identified previously as associated with increased risk of CD onset (PMID: 37263307). Higher UPF consumption was also associated with a decrease of the Eubacterium eligens group (β=-0.019, q-value =1.6×10^-11). We did not observe a significant effect of UPFs on gut permeability (β=-0.003, p-value=0.48) or fecal calprotectin (β=-0.006, p-value=0.37).

Conclusions:
UPF consumption is associated with increased abundance of Ruminococcus torques and Lachnoclostridium, which were previously reported as key contributors of future risk of CD in healthy FDRs. These results suggests that UPF consumption might promote the abundance of CD-associated bacteria rather than a modulation of gut barrier function or fecal calprotectin.