P29 Etanercept and IBD: can we identify at risk patients?

Poster presentation Tuesday 8 October Background We present four cases of patients on etanercept treatment for juvenile idiopathic arthritis (JIA) subsequently diagnosed with inflammatory bowel disease (IBD). We hypothesise that there may be identifiable characteristics of these patients which makes them more identifiable to the physician. Etanercept is a recombinant dimer of human tumour necrosis factor (TNF) receptor proteins bound to human IgG1. It is an effective second line treatment option in JIA for those who have failed, or are intolerant to, methotrexate, and an effective first line treatment option in those who have axial disease at diagnosis. Anti-TNF treatments are effective in IBD, however etanercept in particular does not have therapeutic benefit, and in one trial there was a non-significant trend towards worsening of disease. Methods Please refer to the Results section. Results Four cases, in a 15 year old male, 11 year old female, 10 year old female and 11 year old female, have been identified at our site. The time from etanercept initiation and diagnosis of IBD was between 2 weeks and 2 years 1 month. None of the cases had gastrointestinal symptoms at JIA diagnosis. All four were ANA negative, and in the two with information, both were HLA-B27 negative. Three out of four had hip arthritis involvement. There was a family history of IBD (ulcerative colitis) in only one of the cases. Aside from non-steroidal anti-inflammatory medications, there were no other immunomodulatory medications concurrently being taken in any of the cases while on etanercept. All four were anaemic (Hb < 120g/L) at diagnosis of IBD. Only one had raised ESR (43mm/hr) while all had normal CRP. Albumin was low in two (17g/L and 29g/L). Following gastroenterology review and endoscopy, two cases were diagnosed with Crohn’s disease and two with ulcerative colitis. Conclusion It remains unclear whether etanercept plays a role in the development of IBD in these patients, whether it unmasks subclinical disease, or whether there is no association and it is simply a coincidence. However, there continue to be case reports and series suggesting an association between the two. We hope to identify further cases similar to these from sources such as the UK Biologics Registry and via surveys from teams at paediatric rheumatology centres throughout the UK. Further to the identification of cases, a retrospective review of patient characteristics will be carried out. This may allow for f