P256 Anti-glycan antibody seropositivity at diagnosis does not predict future disease course in patients with Crohn’s disease

P. Lyons1,2,3, N. Noor2, J.C. Lee1,2,3, E.F. McKinney1,2,3, M. Parkes2, K.G.C. Smith1,2,3

1Cambridge Institute for Therapeutic Immunology and Infectious Disease, Jeffrey Cheah Biomedical Centre- University of Cambridge, Cambridge, UK, 2University of Cambridge, Department of Medicine, Cambridge, UK, 3PredictImmune Ltd, Babraham Research Campus, Cambridge, UK


Biomarkers predictive of a patient’s future disease course are critical for personalised medicine. There are two Crohn’s disease (CD) assays with validated associations to a clinical outcome; seropositivity for anti-glycan antibodies (AGAs, Glycominds) and a transcriptional signature of T-cell exhaustion (PredictSURE-IBD™, PS-IBD). The optimal time to use these tests would be at diagnosis before bowel damage occurs. As AGAs have primarily been measured in patients after therapy it’s unclear if they provide prognostic information at diagnosis. To address this we performed a head-to-head comparison of both tests in newly diagnosed CD patients.


We recruited 74 CD patients with active disease; all patients gave written informed consent and were managed using an accelerated step-up strategy in accordance with international guidelines. Active disease was confirmed by one or more objective marker (raised CRP/calprotectin or endoscopic review) and active symptoms. AGA seropositivity was measured at Addenbrooke’s with the IBDX panel (Glycominds) according to manufacturer’s instructions. Patient stratification was also performed using the PS-IBDtest (PredictImmune Ltd) according to manufacturer’s instructions.


Most (80%–59/74) patients were newly diagnosed; 58% (43/74) were seropositive for at least 1 AGA, 19% (14/74) were positive for ≥2 AGAs. Newly diagnosed patients had fewer AGAs (51%, 30/59) than previously diagnosed (87%, 13/15) patients (p = 0.018). We stratified patients into two groups based on seropositivity for 2 or more AGAs—associated with an increased risk of more severe disease but found no significant difference in time to, or frequency of, treatment escalation between the groups. Stratification of patients into IBDhi and IBDlo groups with PS-IBDrevealed a significant difference in the subsequent disease course with IBDhi patients experiencing more aggressive disease, characterised by a shorter time to treatment escalation compared with IBDlo patients (p = 0.001).


AGA seropositivity at diagnosis (as assessed using the IBDX antibody panel) did not predict the need to escalate treatment due to frequently-relapsing or chronically-active disease. Lower AGA seropositivity in newly diagnosed patients compared with those with previous diagnoses suggested that the extent of antibody positivity is a function of disease duration. Previously reported associations between AGA positivity and disease outcome may, therefore, reflect retrospective aggressive disease course and are unlikely to be useful for informing treatment decisions in newly diagnosed patients. The PS-IBD test result correlated with subsequent disease course indicating suitability for early prediction of clinical outcome.