Evaluation of responsive gene expression as a sensitive and specific biomarker in patients with ulcerative colitis.

BACKGROUND Clinical trials in ulcerative colitis (UC) rely on certain parameters to evaluate responses that are highly subjective or of low sensitivity. Here, using a select group of genes, we tested the accuracy of gene expression analysis as a biomarker of clinical, endoscopic, and histologic improvements. METHODS Intestinal biopsies were obtained from UC patients included in two cohorts. Cohort 1 was used to select for genes whose expression was modulated in active (vs. inactive) UC. Cohort 2 included patients recruited in a phase II study receiving placebo, mesalazine, or dersalazine sodium for 4 weeks. The expression of 44 genes identified in Cohort 1 was assessed at weeks 0 and 4, and was then correlated with biomarkers, as well as with clinical, endoscopic, and histologic scores. RESULTS Significant changes in the expression of 31 of the 44 genes tested were detected in Cohort 2 at week 4. Gene expression (ΔCt) significantly correlated with the total Mayo score, C-reactive protein (CRP), and fecal calprotectin. The number of genes significantly regulated at week 4 was highly associated with histologic and endoscopic responses. Logistic regression analysis identified four separate genes (IFITM1, ITGB2, IL1R2, IL2RA) whose relative change was independently associated with endoscopic remission with high specificity and sensitivity. CONCLUSIONS Change in the expression of a select set of genes can serve as an early biomarker, one with high specificity and sensitivity to clinical, endoscopic, and histologic responses. This could represent a new tool for identifying early response to treatment in mild to moderately active UC patients.