Erik Andersson1, Daniel Bergemalm2, Robert Kruse1, Gunter Neumann1, Mauro D’Amato3-4, Dirk Repsilber1, and Jonas Halfvarson2
School of Medical Sciences, Faculty of Medicine and Health, Örebro University, Sweden. 2Department of Gastroenterology, Faculty of Medicine and Health, Örebro University, Sweden. 3Clinical Epidemiology Unit, Department of Medicine Solna, Karolinska Institutet, Stockholm, Sweden. 4BioDonostia Health Research Institute San Sebastian, and IKERBASQUE Basque Foundation for Science Bilbao, Spain

There is a need for improved diagnostic biomarkers in inflammatory bowel disease (IBD). We wanted to identify serum proteins to discriminate between Crohn’s disease (CD), ulcerative colitis (UC), and healthy controls (HC).

91 inflammatory serum proteins from a discovery cohort of CD patients (n=54), UC patients (n=54), and healthy controls (n=54) were quantified using a proximity extension assay. We performed univariate analyses by Welsh t-test, and assessed false discovery rates. A sparse partial least-squares (sPLS) approach was used to identify additional discriminative proteins. Cross-validation error rates for discrimination were calculated. The results were validated in a replication cohort.

By univariate analysis, 17 proteins were identified with significantly different abundances in CD and HCs, and 12 when comparing UC and HCs. Additionally, 64 and 45 discriminant candidate proteins, respectively, were identified with the multivariate approach. Correspondingly, significant cross-validation error rates of 0.11 and 0.19 were observed in the discovery cohort. Only FGF-19 was identified from univariate comparisons of CD and UC, but 37 discriminant candidates were also identified using the multivariate approach. Using univariate comparisons, 16 of 17 CD-associated proteins and 8 of 12 UC-associated proteins were validated in the replication cohort. The error rates for discrimination of subgroups increased when the sPLS model from the discovery cohort was applied to the replication cohort.

By investigating a panel of inflammatory proteins, we identified a number of discriminant candidate markers of subphenotypes of IBD, highlighting the potential of inflammatory serum proteins in diagnostic biomarker identification.