P-180 Gene Expression Profiling Indicates Similar Molecular Pathways Are Active in Adult and Pediatric Ulcerative Colitis

Abstract

Background: Patients with pediatric onset UC manifest more extensive disease at presentation than do adults; however, large genome wide association studies failed to reliably identify pathways that distinguish these 2 populations. It remains unclear whether underlying pathogenesis of pediatric onset UC is distinct or similar to that in adult onset disease. Aside from safety issues, extrapolation of effective medical interventions from adults to children is contingent on similar disease pathogenesis in both age groups. Little is known regarding differential intestinal gene expression implicated in the pathogenesis of adult versus pediatric onset UC. Method(s): RNA was isolated from colonic biopsies obtained from the phase 3 study of golimumab in adults with moderate-severe UC (PURSUIT; n = 87 UC and n = 21 normal). Pediatric colonic gene expression data was obtained from the Gene Expression Omnibus GSE10616 dataset (n = 11 UC and n = 11 normal) and from the phase 1b study of golimumab in pediatric patients with moderate-severe UC (CNTO148UCO1001, n = 20). To evaluate the molecular similarity and differences in adult and pediatric onset UC, the differential gene expression (UC versus normal) profiles of the 2 populations were compared. Differentially expressed genes were defined as those with a fold-change greater than 1.2 and a P-value <0.05 after adjustment for multiple testing. Pathway analysis and inference of upstream regulators of the adult and pediatric disease profiles were performed using a more stringent fold-change cutoff of >2. Genetic profiles by disease extent (extensive versus left-sided involvement) were compared within each of the adult and pediatric UC populations. The PURSUIT (n = 27 extensive and n = 60 left-sided) and CNTO148UCO1001 datasets (n = 13 extensive and 7 left-sided) were used for this analysis. Result(s): A total of 2129 and 2253 probe sets were differentially expressed in pediatric and adult onset UC compared to the normal control in each population. Overall there was substantial overlap in the pediatric and adult disease profiles; only 9.4% (201 of 2129) of the pediatric disease profile was not differentially expressed in the adult population, and only 8.8% (198 of 2253) of the adult disease profile was not differentially expressed in the pediatric population. The set of top-ranked canonical pathways enriched in pediatric UC was very similar to that observed in adult UC; these included pathways in granulocyte adhesion and trafficking, cytokine (IL-1, IL-17) signaling, and dendritic cell maturation. The upstream regulators implied by differential gene expression in UC were also highly parallel in adults and children, including lipopolysaccharide, IL1beta, TNFAalpha, and other inflammatory drivers. Gene expression in inflamed colonic tissues from patients with extensive disease was indistinguishable from that of patients with limited distal involvement in both populations. Conclusion(s): Although children with UC may manifest with more extensive disease at presentation, the genetic landscape of UC is shared by adults and children, and gene signatures for both children and adults are similar regardless of extent of disease. These findings suggest a common molecular profile and disease process in both populations and lend support to the extrapolation of efficacy from adults to children in the development of new medical therapies for UC.