Inflamm Bowel Dis. 2024 Jun 26:izae107. doi: 10.1093/ibd/izae107. Online ahead of print.

Bernadette White ¹, Vaios Svolos ¹, Lisa Gervais ², Aleksandra Jatkowska ¹, Ben Nichols ¹, Jonathan MacDonald ³, John Paul Seenan ³, Richard Hansen ⁴, Richard K Russell ⁵, Simon Milling ⁶, Konstantinos Gerasimidis ¹

Author information

¹Department of Human Nutrition, College of Medicine, Veterinary and Life Sciences, University of Glasgow, Glasgow Royal Infirmary, Glasgow, United Kingdom.

²Department of Paediatric Gastroenterology, Royal Hospital for Children, Glasgow, United Kingdom.

³Department of Gastroenterology, Queen Elizabeth University Hospital, Glasgow, United Kingdom.

⁴Division of Clinical and Molecular Medicine, School of Medicine, University of Dundee, Dundee, United Kingdom.

⁵Department of Paediatric Gastroenterology, Royal Hospital for Children & Young People, Edinburgh, United Kingdom.

⁶School of Infection & Immunity, College of Medical, Veterinary, and Life Sciences, University of Glasgow, Glasgow, United Kingdom.

Abstract

Background: The immunological effects of treatment with exclusive enteral nutrition (EEN) in Crohn's disease (CD) remain unknown. We characterized the plasma levels of inflammation-related proteins (IRPs) in children with CD and ulcerative colitis (UC) compared with noninflammatory controls (non-IBD) and explored the effect of EEN in CD.

Methods: Ninety-two IRPs were quantified using Olink proteomics in children with CD (n = 53), UC (n = 11), and non-IBD (n = 19). For 18 children with active CD, IRPs were measured before and after 8 weeks of EEN. Relationships with disease phenotype and response to EEN were studied.

Results: Compared with non-IBD, patients with active UC and CD had different levels of 27 (24 raised, 3 decreased) and 29 (26 raised, 3 decreased) IRPs, respectively. Exclusive enteral nutrition modified the levels of 19 IRPs (13 increased, 6 decreased including CCL23, interleukin-24, interleukin-6, and MMP-1). More pronounced changes in IRP profile were observed in patients with ileal involvement and a ≥50% decrease in fecal calprotectin during EEN compared with those with colonic involvement and a <50% decrease in fecal calprotectin, respectively. A machine-learning model utilizing baseline IRP profile predicted response to EEN with a sensitivity of 89%, specificity of 57%, and accuracy of 73%. Thymic stromal lymphopoietin was the most important IRP in the model, this being higher in responders.

Conclusions: Inflammation-related proteins may be useful in the differential diagnosis of IBD. Exclusive enteral nutrition extensively modulated IRPs levels in children with active CD with more pronounced effects observed in patients who showed a reduction in FC and had ileal disease involvement.