Computational models to predict disease course and treatment response in multiple sclerosis

Multiple sclerosis (MS) is an autoimmune inflammatory disease of the central nervous system with relapsing or progressive course and potentially leading to severe disability. Although a number of factors, including demographic, clinical, Magnetic Resonance Imaging (MRI), and immunologic, may help to formulate prognostic and therapeutic previsions, the evolution and treatment response remains unpredictable at single patient level. Given the complexity of MS and the emerging need for personalized medicine, computerized methods that can assist clinicians in predicting prognosis, selecting appropriate therapeutic, and timely switching to more efficacious drug may be helpful in tailoring daily patient management. Software of this kind are referred to as clinical decision support systems and their use has already been attempted in MS, although limited to the diagnostic and prognostic field. Based on the biochemical system theory, a set of mathematical and computational tools for simulating biochemical processes, we designed an agent-based model able to replicate the main dynamics of MS. Our Universal Immune System Simulator (UISS) is a computational modeling infrastructure capable to simulate features and dynamics of the immune system, including those underlying MS pathogenesis. The simulation of patients showing relapsing disease were carried out on the basis of age, sex, presence of oligoclonal bands, MRI lesion load at the onset and treatment. For each patient, a set of scenarios was generated, including those simulating the real clinical and MRI course. In addition, the simulator was able to forecast in two patients the timing of subsequent relapses, which really occurred. Further development of our computational modeling infrastructure is underway by integrating additional genetic and immunologic data in order to improve the completeness of patient profiles and to potentiate the accuracy in predicting disease course and treatment response.