Gut abnormalities persist even in Crohn's remission

Patients with Crohn’s disease in corticosteroid-free clinical remission still show abnormalities in the gut lining, microbiome, metabolism, and diet, despite strong suppression of mucosal immune pathways, according to a multi-omics analysis published in Gastroenterology.

“The remitting-relapsing nature of CD is likely driven by the fact that current advanced therapies only address one side of the disease: the immune response,” the study’s corresponding author, Yael Haberman, MD, PhD, a pediatric gastroenterologist at Sheba Medical Center, Tel HaShomer, Israel, and an adjunct faculty member at the Cincinnati Children's Hospital Medical Center, told GI & Hepatology News. “The underlying perturbations in diet, the microbiome, metabolites, and epithelial barrier function remain largely untouched. To achieve a ‘deeper, longer-lasting remission,’ it may be necessary to broaden the therapeutic focus to include restoring the gut's physical and metabolic health through targeted dietary, microbiome, and epithelial interventions.”

Crohn’s disease therapies have increasingly focused on immune suppression but less is known about how other components of gut biology behave during clinical remission.

In an analysis of 191 subjects recruited for a study known as Capsule & Omics for Predicting Exacerbation of Crohn’s disease (CORE-CD), investigators compared 77 patients with CD in stable remission, 37 patients with active, mostly treatment-naïve CD, and 77 non-IBD controls. Although immune-related gene expression in the ileum was significantly reduced during remission, often to levels lower than those observed in controls, markers of epithelial antimicrobial activity and mucus biology remained elevated. These transcriptomic findings were accompanied by persistent dysbiosis, changes in fecal metabolism, and dietary patterns characterized by higher consumption of ultra-processed foods and lower intake of fiber and micronutrients.

Patients in remission had been clinically stable for at least three months with a Crohn’s Disease Activity Index below 150 and stable medication dosing; most were receiving biologic therapy. Inflammatory markers such as C-reactive protein were comparable to controls and significantly lower than in active CD. Fecal calprotectin levels were reduced relative to active disease but not fully normalized in all patients. A subset undergoing video capsule endoscopy demonstrated small bowel inflammatory scores, some with levels previously associated with increased flare risk.

RNA sequencing of tissue from the ileum showed that, during remission, genes related to adaptive T cells and innate granulocyte pathways were broadly reduced compared with controls, suggesting that immune activity had dropped below normal levels. However, the antimicrobial response in the epithelial cells remained active. Genes including DUOX2, CEACAM6, and REG1B were persistently upregulated during remission, approaching levels seen in active disease. Functional enrichment analysis confirmed sustained activation of antibacterial pathways.

At the same time, genes associated with goblet cells and mucin glycosylation were significantly induced in remission compared with both controls and active CD. Summary of co-expressed genes signals eigengene analysis showed that mucin-related pathways were selectively increased during remission. When the researchers stratified patients by six-month outcomes, they found that those who later had a flare already showed higher levels of genes related to the gut lining’s antimicrobial defenses and mucus production at the start of the study. In contrast, differences in adaptive immune suppression did not help predict who would flare and who would stay in remission.

Analysis of stool samples showed that even when patients were in remission, their gut bacteria had not returned to a healthy balance. Compared with controls, patients in remission had a higher IBD dysbiosis index, lower microbial diversity, and persistent enrichment of taxa previously linked to CD, including Escherichia, Ruminococcus gnavus, and Enterocloster bolteae. Fecal metabolomics demonstrated that most metabolites altered in remission relative to controls were perturbed in the same direction as in active CD, indicating durable metabolic disruption despite clinical quiescence.

Dietary assessment using a validated food frequency questionnaire revealed that patients with CD, regardless of disease activity, consumed more ultra-processed foods and added sugars and fewer vegetables, fiber, folate, and vitamins C and K compared with controls. Principal component analysis demonstrated clear separation between CD and control dietary patterns. Higher exposure to ultra-processed foods correlated with greater microbial dysbiosis, whereas higher intake of olive oil, vegetables, and certain micronutrients was associated with more favorable microbial indices.

Importantly, ultra-processed food exposure was negatively correlated with mucin glycosylation gene signatures, while adherence to a Mediterranean-style diet was positively associated with these epithelial pathways. Neither fecal calprotectin nor the microbial dysbiosis index significantly correlated with mucin-related gene expression.

An outside expert said the findings highlight the complexity of Crohn’s disease even during apparent clinical remission. “This study helps explain why patients with Crohn’s disease frequently experience disease exacerbations,” said Ruben Colman, MD, PhD, director of intestinal ultrasound at Stanford University’s Center for IBD and Celiac Disease in Palo Alto, California. “Using a multiomics approach incorporating gene expression, microbial, metabolomic, and dietary signals, the investigators showed that while biologic therapies effectively suppress immune activity in patients in remission, pathogenic microbial signals and epithelial antibacterial signals — including DUOX2 and CEACAM6 — can persist.” These findings suggest that achieving deep remission may require adjunctive strategies beyond immune-targeting biologics that address the diet–epithelial–microbial axis, he added.

The authors noted certain limitations of the study, including its cross-sectional design, the absence of interventional dietary data, and the inability to capture medication history beyond biologic exposure and certain lifestyle variables.

The study was funded by the Helmsley Charitable Trust, the ERC starting grant, the ERC Proof of Concept grant, Israel Science Foundation, Tel Aviv University’s Colton Center for Autoimmune Diseases, Israel Science, Culture, and Sport, and Litwin IBD Pioneers Awards. The authors reported having no relevant disclosures.

Research implications

GI & Hepatology News invited Dr. Haberman to further discuss the implications of the research.

Why does this study matter?

Dr. Haberman: By comparing patients in remission to those with active disease and healthy controls, the CORE-CD study found that unhealthy eating patterns, specifically the high consumption of ultra-processed foods, persist despite patient education. These dietary choices are linked to a pathogenic microbial composition and metabolic changes that deviate significantly from a healthy state. Intriguingly, while biologic therapies successfully suppress adaptive immune responses (often below the levels of healthy controls), the gut epithelium remains in a state of high alert, showing elevated antimicrobial signals and a novel increase in goblet cell and mucin glycosylation activity. The study establishes a critical link between ultra-processed food intake and the impairment of mucin glycosylation, a key component of gut barrier homeostasis, suggesting that dietary and epithelial-targeted interventions are necessary adjuncts to current immune-centric treatments.

When you had all the data in front of you, was there a finding, or more than one, that surprised you?

Dr. Haberman: Ileal transcriptomics revealed that genes associated with adaptive T-cells and innate granulocytes were suppressed in remission patients to levels lower than those observed in non-IBD healthy controls. This highlights the profound impact of advanced biologic therapies on the immune response, yet it stands in stark contrast to the persistent inflammation markers in the epithelium.

Another interesting finding was the induction of genes linked to goblet cells and to mucin glycosylation specifically in patients in remission. In active CD, these signals are typically reduced. The increase during remission may represent a compensatory mechanism to restore the gut barrier or a specific biological signature of a treated state.

Additionally, the study found a clear negative correlation between the consumption of NOVA Class 4 ultra-processed foods and the expression of genes essential for mucin glycosylation. This provides a potential mechanistic explanation for how modern diets are linked with altering the gut barrier's defense systems, independent of the immune system's activity. This contrasted with higher exposure to the Mediterranean diet, which was associated with higher expression of genes essential for mucin glycosylation.

How might the findings influence clinical practice?

Dr. Haberman: These findings suggest that treating just immune activation may be insufficient for long-term disease clearance. Hence, potential shifts in clinical practice may include three areas of focus:

• Adjunctive dietary therapy. Patients and clinicians should focus on reducing ultra-processed food intake, as it is linked to lower expression of mucus barrier genes (mucin glycosylation).

• Epithelial health monitoring. Measuring epithelial antimicrobial signals like DUOX2 may provide a more accurate prediction of future flares than traditional immune markers.

• Targeting the diet-epithelia-microbial axis. Future management should move beyond simple immune suppression to include strategies that promote a “healthier” CD state through adjunctive dietary modification and metabolite supplementation, along with immune modulation.

What gaps remain, and what research should be done next?

Dr. Haberman: The current study establishes correlations. Interventional trials are needed to determine if removing ultra-processed foods directly restores mucin glycosylation and prevents flares.

Further research is required to understand why some patients maintain higher epithelial antimicrobial signals than others despite receiving similar treatments, and to determine whether these findings can be replicated in more geographically and ethnically diverse populations to ensure that the observed signals are universal rather than specific to the studied cohort.

Finally, discriminating between signals caused by long-term disease duration and those caused by specific medications remains a challenge.