P100 Serum inflammatory protein profiling to corroborate selection of therapy naïve moderate-to-severe pediatric Crohn’s disease patients eligible for first-line infliximab treatment

Jongsma , M.M.E.(1);Costes , L.M.M.(2);Tindemans , I.(2);Cozijnsen , M.A.(1);Raatgreep , R.H.C.(2);van Pieterson , M.(1);Li , Y.(3);Escher , J.C.(1);de Ridder , L.(1);Samsom , J.N.(2);

(1)Erasmus University Medical Center/Sophia Children's Hospital, Department of Pediatric Gastroenterology, Rotterdam, The Netherlands;(2)Erasmus University Medical Center, Laboratory of Paediatrics- Division of Gastroenterology and Nutrition, Rotterdam, The Netherlands;(3)Erasmus University Medical Center- Erasmus MC Cancer Institute, Department of Pathology & Clinical Bioinformatics, Rotterdam, The Netherlands;


Early intervention with biologic therapy, especially with anti-Tumor Necrosis Factor-a, is very effective and now more commonly used to prevent disease progression and complications in Crohn’s disease (CD). However, a rational approach to decide which patient should be treated First-line Infliximab (FL-IFX) is lacking. In this study, using serum from a randomized control trial, we assessed how changes in immune profiles prior to therapy and after the first standardized induction treatment with either FL-IFX or conventional treatment (CONV) related to maintaining disease remission at week 52.


Serum samples from therapy naive moderate to severe CD patients from the TISKIDS study1 were used at time of diagnosis and at 10-14 weeks after induction therapy with FL-IFX (n=48) and CONV (n=43). FL-IFX consisted of 5 IFX infusions (5 mg/kg) combined with azathioprine (AZA) maintenance. CONV consisted of induction therapy with Exclusive Enteral Nutrition or oral prednisolone combined with AZA maintenance. Concentrations of 92 inflammatory proteins were determined with Olink Proteomics®. Significant differences in mean protein NPX after FDR adjustment with log2fold > 0.5 were considered meaningful.


After 10-14 weeks of induction therapy, FL-IFX suppressed a larger number inflammatory proteins and induced stronger suppression compared to CONV. Of the 30 significantly modulated FL-IFX target proteins 18 were differentially regulated by FL-IFX only. Hierarchical clustering of patients based on their baseline profiles of the 30 IFX-modulated proteins revealed 2 patient clusters, a CD-hi cluster with significantly higher baseline clinical disease activity, CRP and blood neutrophil concentrations compared to the CD-lo cluster. The CD-hi cluster had an increased abundance of 23/30 proteins amongst which Oncostatin-M, TNFSF14, HGF and TGF-alpha compared to the CD-lo cluster (Figure 1A). Of the CD-lo patients treated with FL-IFX, 58% reached clinical disease remission without treatment escalation at week 52, while this was only 24% for CD-hi patients treated with FL-IFX. Only 20% of the CD-lo patients treated with CONV reached remission while this was only 6% for the CONV treated patients in the CD-hi group (Figure 1B). Strikingly, similar clustering using the immune profiles after 10-14 weeks induction therapy did not predict remission at week 52.


FL-IFX achieves stronger reduction and modulates more inflammatory serum immune protein concentrations than CONV. Stratification of patients according to profiles of IFX-modulated proteins prior to therapy identifies patients with a lower chance of reaching clinical remission without treatment escalation at week 52.

1 Jongsma MME et al. Gut. 2020 Dec; DOI:10.1136/ PMID:33384335