Funding outcome

Background: Crohn disease (CD) is a severely debilitating chronic gut disorder, and is increasing common, especially in Canadian children. There is no permanent cure, only treatments that somewhat control the disease. Despite the improved understanding of CD development, progress in translating these findings into novel and safe therapies is limited. Dietary therapy is the first line therapy for pediatric (but not adult) CD (pCD) patients. Although this treatment is very effective and safe, as it does not suppress the immune system, with no concern for side effects, it is difficult to complete, has a major impact on quality of life, and varies greatly from patient to patient. Therefore, there is a need to distinguish between responders and no-responders. Objective: Use computational models capable of learning patterns from data from a large prospective pediatric national cohort to make predict which pCD patient would respond to dietary therapy. Preliminary results: This work builds on our previous work, using standard statistics, showing that disease severity and microbiota are associated with diet. By expanding our dataset and using more sophisticated mathematical models we expect to build a classifier that would allow to distinguish dietary therapy responders. Significance: This work will provide a systematic analysis to predict therapy response in pCD, which could tailor and personalize therapies. Therefore, this research has the potential to provide better quality of life for children who live with pCD and reduce healthcare burden. Mental Health: We will also study if depression, anxiety, and quality of life are informative for predicting therapy response.