The use of clinical data in drug development, pharmacovigilance and pharmacoeconomics is well established. Here Real World Data approaches rely on data scale to compensate for the complexity of less controlled data from real practice, and are expected to complement insights from conventional designs that were optimized for efficiency at the cost of the introduction of biases. Realistic scenarios are being worked out for an incremental evolution of analytics solutions from current practice considerate of privacy challenges and financial sustainability. In contrast, reverse translation from the clinic to discovery has suffered from the growing complexities of bringing drugs to the market. It has been substituted with ever more advanced biotechnological solutions, with mixed results. Advances in scalable data mining now enable us to learn much more from historical preclinical data. When extended to Real World Data, these approaches hold the promise to finally start reducing the distance from the clinic. The identification of targets of clinical relevance in much earlier disease stages alone merits the formulation of workable scenarios for discovery, even if these seem more challenging than those at later stages. As reflected by efforts like PatientsLikeMe, the patients are waiting!
Hugo holds the degrees of MD, MSc in Bioinformatics and PhD in Molecular Biology from the University of Leuven. Instilled with a strong believe in the value of multidisciplinary research during his research at the universities of Århus and Leuven and the European Molecular Biology Laboratories in Heidelberg, he joined Janssen in 2008 as a computational biologist. Hugo now heads a multidisciplinary team that applies integrative data analytics for drug discovery. The team secured competitive regional and European level funding for the development with academic and industry partners of an ambitious suite of scalable machine learning solutions for pharmaceutical research.