Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

P212 Serum biomarkers predict disease severity in Ulcerative Colitis

C. Soendergaard*1, O.H. Nielsen1, J.B. Seidelin1, P.H. Kvist2, J.T. Bjerrum1

1Herlev University Hospital, Department of Gastroenterology, Copenhagen, Denmark, 2Novo Nordisk A/S, Histology, Maaloev, Denmark

Background

Clinicians rely on the clinical symptoms and measurements of acute phase proteins like CRP as surrogate markers of inflammation for assessment of the level of disease activity in ulcerative colitis (UC). This is, however, challenging as CRP is often only increased in patients with severe flares of UC, and novel biomarkers are thus needed for the identification of mild and moderate UC. With the use of a commercially available ELISA platform we aimed at identifying a limited panel of potential biomarkers for grading disease severity in UC.

Methods

40 healthy controls along with 65 patients with varying disease activity of UC (Mayo-score 0-12) were enrolled. Using multiplex ELISA the concentration of 78 proteins was measured in the serum. Multivariate statistics using the SIMCA-P+ software were performed to select a limited number of proteins as potential biomarkers for assessment of the disease severity.

Results

For differentiating between patients with mild and moderate disease (i.e. Mayo score of 3-5 or 6-10, respectively), we identified alpha-1 antitrypsin (AAT) (AUC=0.79) with a sensitivity and specificity of 0.90 and 0.70, respectively, thereby outperforming the predictability of CRP (AUC=0.52). The combination of granulocyte colony stimulating factor (G-CSF) and AAT produced predictive models for differentiating mild and moderate disease (AUC=0.72) as well as moderate and severe disease (AUC=0.96) with the latter showing the same predictability as CRP.

Conclusion

Based on the measurements of AAT and G-CSF we were able to differentiate between patient having mild, moderate and severe disease activity. Identification of these potential serum biomarkers measured by a commercially available platform enables clinicians to optimize and individualize the treatment at an early stage while awaiting endoscopic examinations.