Machine Learning–Based Prediction of Pediatric Ulcerative Colitis Treatment Response Using Diagnostic Histopathology

The initial presentation of ulcerative colitis within the pediatric population exhibits a degree of uniformity, the majority characterized by extensive colitis at the time of diagnosis. However, the response to initial therapy demonstrates marked heterogeneity.1 It is challenging to discriminate which patients would successfully improve on corticosteroids followed by mesalamine therapy maintenance therapy and those who would benefit from early introduction of biologic therapy. The Clinical and Biological Predictors of Response to Standardized Pediatric Colitis Therapy (PROTECT) study, a multicenter inception cohort study, aimed to address this.