Potential blood-based markers of celiac disease

BACKGROUND Blood-based diagnostics has the potential to simplify the process of diagnosing celiac disease (CD). Although high levels of autoantibodies against tissue transglutaminase (anti-TG2) are strongly indicative of active CD, several other scenarios involve a need for additional blood-based CD markers. METHODS We investigated the levels of messenger RNA (mRNA) in whole blood (n = 49) and protein in plasma (n = 22) from cases with active CD (n = 20), with confirmed CD and normalized histology (n = 15), and without a CD diagnosis (n = 14). Group differences were analyzed using Kruskal-Wallis one-way analysis of variance by ranks. We also investigated correlations between levels of potential markers, histopathology according to the modified Marsh scale, and CD risk gradient based on HLA type, using Spearman rank correlation. The relation between HLA-DQ2 gene dose effect and the expression levels of selected blood-based markers was investigated using the Mann-Whitney U test. Finally, the diagnostic performance of anti-TG2, potential blood-based CD markers, and logistic regression models of combined markers was evaluated using receiver operating characteristic (ROC) curve analysis. RESULTS CXCL11 protein levels and TNFRSF9 and TNFSF13B mRNA levels were identified as potential CD markers. These are all affected by or involved in the regulation of the NF-κB complex. CXCL11 protein levels and IL21 and IL15 mRNA levels were correlated with histopathology according to the modified Marsh scale, as were the established CD markers. HLA genotype risk and HLA-DQ2 gene dose effect did not show any significant relations with either the potential CD markers or the established CD markers. ROC curve analysis revealed a slight, non-significant increase in the area under the curve for the combined use of anti-TG2 and different constellations of potential blood-based CD markers compared to anti-TG2 alone. CONCLUSIONS The CD markers identified in this study further emphasize the significance of components related to NF-κB regulation in relation to CD. However, the relevance of CXCL11, TNFSF13B, TNFRSF9, and other NF-κB interacting proteins recognized by pathway analysis, needs to be further investigated in relation to diagnosis and monitoring of CD.