Data and biological knowledge inform mathematical, statistical and computational models to obtain insight and provide solutions to medical challenges. This approach allows to accelerate drug development, identify target groups and get the most out of experimental and approved treatments.
Identifying potential responders is the first step in increasing the success of trials. Mathematical modeling (differential equations, cellular automata, individual-based methods) and advanced data analytic methods (topological data analysis, machine learning, statistical methods, image analysis methods) unveil metrics that summarize the patient individual characteristics and enable the identification of responders.
We capture the most relevant features of the disease’s biology and translate them into mathematical rules. Then, we simulate thousands of virtual patients on computers and analyze their response to treatment. Essaying different protocols in silico is faster than pilot trials and biomedical model studies, thus accelerating the clinical implementation of new treatments.
Rationally designed protocols based on mathematical models go beyond the classical trial-and-error approach. Optimization methods and massive simulations allow the exploration of thousands of different treatment dosing and timings in silico. This way, the optimum protocols in monotherapy and combinations are obtained, getting the most out of the treatment.
Cheaper and faster in silico approaches allow reducing the number of studies in experimental models and pilot clinical trials. This drastically cuts the time and cost needed to get the most out of new treatments. Better identification of the responding patients results in the minimization of drug failures. The selection of optimal schedules increases the drug efficacy, which improves drug efficacy, benefits the patient, and provides higher returns.
