Combining mechanistic modeling with machine learning as a strategy to predict inflammatory bowel disease clinical scores

Disease activity scores such as Crohn’s disease activity index (CDAI), Mayo endoscopic score (MES) and Mayo score are efficacy endpoints frequently used in clinical trials of inflammatory bowel disease (IBD) therapies. They rely on either the physician’s observation of the inflammatory state of the patient’s gastrointestinal tissue alone or combined with the patient’s subjective evaluation of general well-being. Given the importance of these scores in evaluating the efficacy of drug treatment and disease severity, there has been interest in developing computational approaches to reliably predict these scores. A novel promising approach is using mechanistic models such as quantitative systems pharmacology (QSP), which simulate the mechanisms of the disease and its modulation by the drug pharmacology. However, extending QSP model simulations to clinical score predictions has been challenging due to the limited availability of gut biopsy measurements and the subjective nature of some of the evaluation criteria for these scores that cannot be described using mechanistic relationships. In this presentation, we will provide a brief overview of IBD disease activity scores and recent progress in building predictive models for these scores. Then, we present a machine-learning approach to leverage simulated markers of inflammation from a QSP model to predict IBD clinical scores. We will demonstrate how this combined approach has been used in IBD drug discovery and development programs to (1) explore mechanistic insights underlying clinical observations; and (2) simulate novel therapeutic strategies that could potentially improve clinical outcomes.