For a study, researchers sought to identify juvenile Crohn’s disease (CD) subgroups that responded differently to infliximab treatment using data from electronic health records. They performed a retrospective analysis of 295 pediatric patients with CD treated with infliximab for at least one year at The Children’s Hospital of Philadelphia’s Center for Inflammatory Bowel Disease between January 2010 and December 2017. The progression of the disease was detailed, and patient subgroups were established using C-reactive protein trajectory analysis on longitudinal data (CRP). Across subgroups, they compared patient characteristics, biomarkers for disease activity, and long-term surgical results. The extra benefit of subgroup categorization in predicting long-term surgical outcomes was evaluated using Cox regression models in addition to baseline phenotype and location.

Patients with a greater risk of infliximab non-response, infliximab response but with infrequent illness flares, and patients with long-term response were divided into 3 subgroups with varied relapse-and-remission characteristics (n=33, 65, and 197 from subgroup 1- 3). Patients who responded best to treatment had a lower frequency of complicated disease phenotypes (P=0.01), including perianal involvement (P=0.05), lower baseline CRP (P<0.01) and calprotectin (P=0.01), and the lowest risk of IBD-related gastrointestinal surgery within 10 years of starting treatment (P<0.01). Longitudinal data from electronic health records, which were widely available, could be used to offer a more detailed picture of therapy response in children with CD.

Reference:journals.lww.com/jpgn/Abstract/2022/03000/Learning_Longitudinal_Patterns_and_Subtypes_of.14.aspx