Investigating the Causal Relationship Between Gut Microbiota and Crohn’s Disease: A Mendelian Randomization Study

We are writing to commend the recently published study by Garay et al.1Raygoza Garay J.A. et al.Gastroenterology. 2023; 165: 670-681Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar The authors' innovative use of a prospective cohort study design, coupled with a random survival forests (RSF) methodology, is noteworthy for identification of the microbial community characteristics before the onset of Crohn's disease (CD). To our knowledge, this study is the first to establish a correlation between a gut microbiota signature and the future onset of CD, advancing our understanding of the disease's pathogenesis. With the microbiota risk score (MRS), clinicians could identify at-risk individuals earlier to prevent CD. Importantly, these findings offer a compelling argument for a potential causal relationship between gut microbiota characteristics and CD. However, it is essential to note that the findings of Garay et al are correlational and therefore do not directly infer causation. Additional unknown confounders that influence both gut microbiota and the disease could exist. To further investigate the causal relationship between gut microbiota and the onset of CD, we propose the use of mendelian randomization, which uses genetic variants as instrumental variables to test for a causal effect, thereby circumventing the common issues of confounding and reverse causation that can trouble traditional observational studies.2Ke H. et al.Front Med (Lausanne). 2023; 101166683Google Scholar We performed a mendelian randomization study using the publicly available inflammatory bowel disease (IBD) genome-wide association study (GWAS) database by de Lange et al3de Lange K.M. et al.Nat Genet. 2017; 49: 256-261Crossref PubMed Scopus (706) Google Scholar (comprising 12,194 CD case subjects and 28,072 healthy control subjects), and the gut microbiome GWAS database by the MiBioGen consortium4Kurilshikov A. et al.Nat Genet. 2021; 53: 156-165Crossref PubMed Scopus (470) Google Scholar (consisting of 18,040 participants). Of the top 20 most important genera in the RSF model, only the Lachnospiraceae_NK4A136_group (ranked 20th in RSF) was found to have a causal relationship with CD (P = 0.014; β = −0.1778) (Supplementary Table 1). Other genera indicating a causal association with CD included: Hungatella (53rd in RSF; P = 0.019; β = −0.1714), Lachnospiraceae_UCG.001 (85th in RSF; P = 0.002; β = −0.2148), Dialister (89th in RSF; P = 0.0067; β = −0.2490), Eggerthella (101st in RSF; P = 0.017; β = −0.1365), and Akkermansia (131st in RSF; P = 0.05; β = −0.1442) (Supplementary Table 1). Some important genera in the RSF model (like the top 5 in RSF ranking) did not show a causal relationship with CD in our analysis. Similarly, the genera suggested as having a causal relationship with CD in our analysis did not have a relatively high importance in the RSF analysis. We speculate that these results could be attributed to several reasons:1.There might be other confounding factors, such as environmental and genetic factors. These confounders might simultaneously affect the onset of CD and the gut microbiota, but the changing taxa itself does not affect the progression of CD. Nevertheless, the MRS model constructed by the authors remains a useful method for predicting disease in first-degree relatives of patients.2.There might be reverse causality. Even though the authors used a prospective cohort study design, some pathophysiologic changes in CD might exist and affect the gut microbiota before clinical diagnosis of CD.3.In conducting mendelian randomization, we focus on the causal relationship between individual taxa and CD. As mentioned by the authors, the gut microbiota as a whole, rather than individual taxa, might have a greater impact on the onset of CD. This also highlights the value of this study and suggests that future research might need to consider the gut microbiota as a community, rather than focusing solely on individual or a few types of taxa.4.Mendelian randomization is based on genetic variants, which reflect the long-term impact of the gut microbiota on susceptibility to IBD over a lifetime. The influence of the gut microbiota on IBD may vary significantly in the short term. Finally, it should be noted that our analysis has certain limitations. The databases used consist mainly of European populations, limiting the generalizability to other racial groups. The databases we used have not been optimized for healthy first-degree relatives of CD patients, which may differ from the population studied by the authors. Overall, our data provide some guidance for further animal experiments and clinical intervention studies, and further suggest that gut microbiota may play complex roles in the onset and progression of CD. We hope that our findings will encourage further exploration into the complex relationship between gut microbiota and CD and contribute to the ongoing development of effective preventive strategies and interventions. Download .xlsx (.02 MB) Help with xlsx files Supplementary Table 1 Gut Microbiome Composition Is Associated With Future Onset of Crohn's Disease in Healthy First-Degree RelativesGastroenterologyVol. 165Issue 3PreviewGut microbiome composition, function, and stool metabolites can identify those more likely to develop Crohn's disease in a cohort of healthy first-degree relatives of patients with Crohn's disease. Full-Text PDF ReplyGastroenterologyVol. 166Issue 2PreviewWe read with interest the letters from Gao et al1 and Li et al2 about our recent study describing the association of the gut microbiome with the risk of developing Crohn's disease (CD).3 Specifically, we found that the gut microbiome community rather than individual taxa may contribute to the risk of CD. Both letters describe their analyses that are conflicting or partially overlapping with our discovery. Gao et al1 raised the lack of correlation of Escherichia with the future risk of CD onset considering that our group identified that serum antibody response against the cell wall component of Escherichia (anti-OmpC) was associated with CD onset. Full-Text PDF