OBJECTIVES: Crohn's disease (CD) is a chronic relapsing-remitting gut inflammatory disorder with a heterogeneous unpredictable course. Dysbiosis occurs in CD; however, whether microbial dynamics in quiescent CD are instrumental in increasing the risk of a subsequent flare remains undefined. METHODS: We analyzed the long-term dynamics of microbial composition in a prospective observational cohort of patients with quiescent CD (45 cases, 217 samples) over 2 years or until clinical flare occurred, aiming to identify whether changes in the microbiome precede and predict clinical relapse. Machine learning was used to prioritize microbial and clinical factors that discriminate between relapsers and nonrelapsers in the quiescent phase. RESULTS: Patients with CD in clinical, biomarker, and mucosal remission showed significantly reduced microbial richness and increased dysbiosis index compared with healthy controls. Of the 45 patients with quiescent CD, 12 (27%) flared during follow-up. Samples in quiescent patients preceding flare showed significantly reduced abundance of Christensenellaceae and S24.7, and increased abundance of Gemellaceae compared with those in remission throughout. A composite flare index was associated with a subsequent flare. Notably, higher individualized microbial instability in the quiescent phase was associated with a higher risk of a subsequent flare (hazard ratio 11.32, 95% confidence interval 3–42, P = 0.0035) using two preflare samples. Importantly, machine learning prioritized the flare index and the intrapersonal instability over clinical factors to best discriminate between relapsers and nonrelapsers. DISCUSSION: Individualized microbial variations in quiescent CD significantly increase the risk of future exacerbation and may provide a model to guide personalized preemptive therapy intensification.

Celiac disease is provoked by gluten exposure, but the complete pathogenic process in the duodenum and the loss of tolerance to gluten is not well understood. We aimed to define the core celiac transcriptomic signature and pathologic pathways in pre-treatment formalin-fixed paraffin-embedded (FFPE) duodenum biopsies used for clinical diagnosis. We use mRNAseq to define pre-treatment diagnostic duodenum gene expression in 54 pediatric celiac patients and non-celiac controls, and we validate our key findings in two independent cohorts of 67 adults and pediatric participants that used fresh frozen biopsies. We further define similar and divergent genes and pathways in 177 small bowel Crohn disease patients and controls. We observe a marked suppression of mature epithelial metabolic functions in celiac patients, overlapping substantially with the Crohn disease signature. A marked adaptive immune response was noted for the up-regulated signature including interferon response, alpha-beta, and gamma-delta T-cells that overlapped to some extent with the Crohn disease signature. However, we also identified a celiac disease specific signature linked to increased cell proliferation, nuclear division, and cell cycle activity that was localized primarily to the epithelia as noted by CCNB1 and Ki67 staining. Lastly, we demonstrate the utility of the transcriptomic date to correctly classify disease or healthy states in the discovery and validation cohorts. Our data supplement recently published datasets providing insights into celiac pathogenesis using clinical pathology FFPE samples, and can stimulate new approaches to address this highly prevalent condition.